Handbook of Clinical Child Neuropsychology
Critical Issues in Neuropsychology Series Editors:
Cecil R. Reynolds
Antonio E. Puente
Texas A&M University
University of North Carolina, Wilmington
Editorial Advisory Board: Erin Bigler, University of Texas, Austin Raymond S. Dean, Ball State University Hans J. Eysenck, University of London Charles J. Golden, Drexel University John Gruzelier, University of London Lawrence C. Hartlage, Fairfax University Merril Hiscock, University of Saskatchewan Lawrence Majovski, Huntington Memorial Hospital Francis J. Pirozzolo, Baylor University School of Medicine Karl Pribram, Stanford University
ASSESSMENT ISSUES IN CHILD NEUROPSYCHOLOGY Edited by Michael G. Tramontana and Stephen R. Hooper HANDBOOK OF CLINICAL CHILD NEUROPSYCHOLOGY Edited by Cecil R. Reynolds and Elaine Fletcher-Janzen MEDICAL NEUROPSYCHOLOGY: The Impact of Disease on Behavior Edited by Ralph E. Tarter, David H. Van Thiel, and Kathleen L. Edwards NEUROPSYCHOLOGICAL FUNCTION AND BRAIN IMAGING Edited by Erin D. Bigler, Ronald A. Yeo, and Eric Turkheimer NEUROPSYCHOLOGY, NEUROPSYCHIATRY, AND BEHAVIORAL NEUROLOGY R. Joseph RELIABILITY AND VALIDITY IN NEUROPSYCHOLOGICAL ASSESSMENT Michael D. Franzen
A Continuation Order Plan is available for this series. A continuation order will bring delivery of each new volume immediately upon publication. Volumes are billed only upon actual shipment. For further information please contact the publisher.
Handbo ok of Clinical Child Neuropsychology Edited by
CECIL R. REYNOLDS and
ELAINE FLETCHER-JANZEN Texas A&M University College Station, Texas
Springer Science+Business Media, LLC
Library of Congress Cataloging in Publication Data Handbook of clinica! child neuropsychology 1 edited by Ceci! R. Reynolds and Elaine Fletcher-Janzen. p. cm-(Critical issues in neuropsychology) Includes bibliographies and index. ISBN 978-1-4899-6809-8 ISBN 978-1-4899-6807-4 (eBook) DOI 10.1007/978-1-4899-6807-4 1. Pediatric neuropsychology. I. Reynolds, Ceci! R., 1952. II. Fletcher-Janzen, Eiaine. III. Series. [DNLM: 1. Brain-growth & development. 2. Child Behavior. 3. Child Development Disorders-diagnosis. 4. Child Development Disorders-therapy. 5. Neuropsychological Tests-in infancy & childhood. 6. Neuropsychology. WS 350.6 H2364] RJ486.5.H26 1989 155.4-dc19 DNLM!DLC 88-39536 for Library of Congress OP
© 1989 Springer Science+Business Media New York Originally published by Plenum Press, New York in 1989 Softcover reprint of the hardcover 1st edition 1989 AII rights reserved No part of this book may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, microfilming, recording, or otherwise, without written permission from the Publisher
Contributors Eileen B. FenneU •
RusseU M. Bauer •
Department of Clinical and Health Psychology, University of Florida, Gainesville, Florida 32610
Department of Clinical and Health Psychology, University of Florida, Gainesville, Florida 32610
Thomas L. Bennett •
Charles J. Golden • Departments of Psychology, Sociology, and Anthropology, Drexel University, Philadelphia, Pennsylvania 19104
Department of Psychology, Colorado State University, Fort Collins, Colorado 80523
Richard A. Berg •
Department of Behavioral Medicine and Psychiatry, West Virginia University Medical Center-Charleston Division, Charleston, West Virginia 25326
Erin D. Bigler • Department of Psychology, University of Texas at Austin, Austin, Texas 78712; and Austin Neurological Clinic, Austin, Texas 78705
Kathi A. Borden • Department of Psychology, Pepperdine University, Los Angeles, California 90034 Ronald T. Brown • Division of Child and Adolescent Psychiatry, Emory University School of Medicine, Atlanta, Georgia 30322 Manuel L. Cepeda •
Department of Psychiatry, University of South Alabama College of Medicine, Mobile, Alabama 36617
Raymond S. Dean •
Neuropsychology Laboratory, Ball State University, Muncie, Indiana 47306; and Indiana University School of Medicine, Indianapolis, Indiana 46282
Robert W. Elliott •
Department of Special Education, South Bay Union High School District, Redondo Beach, California 90277
BryanFantie •
DepartmentofPsychology, University of Lethbridge, Lethbridge, Alberta TlK 3M4, Canada
Jeffrey W. Gray • Neuropsychology Laboratory, Ball State University, Muncie, Indiana 47306 Frank M. Gresham • Department of Psychology, Louisiana State University, Baton Rouge, Louisiana 70803 Ruth Adlof Haak • Balcones Special Services Cooperative, Austin, Texas 78746
Thalia Harmony •
Neurosciences Research Program, Iztacala School, National Autonomous University of Mexico, 54030 Tlalnepantla, Mexico City, Mexico
Lawrence C. Hartlage • Department of Psychology, University of Arkansas, Fayetteville, Arkansas 72701 Patricia L. Hartlage •
Department of Pediatrics and Neurology, Medical College of Georgia, Augusta, Georgia 30902
Robert L. Hodes •
Department of Neurology, University of Wisconsin, Madison, Wisconsin 53792
Stephen R. Hooper • Clinical Center for the Study of Development and Learning, The University of North Carolina, and Department of Psychiatry, The University of North Carolina School of Medicine, Chapel Hill, North Carolina 27599 v
vi
CONTRIBUTORS
Arthur MacNeUI Horton, Jr. • Veterans Administration Medical Center, Baltimore, Maryland 21218; Department of Psychiatry, University of Maryland Medical School, Baltimore, Maryland 21201; and Psych Associates, Towson, Maryland 21214
Cecil R. Reynolds • Department of Educational Psychology, Texas A&M University, College Station, Texas 77843
Randy W. Kamphaus • Department of Educational Psychology, University of Georgia, Athens, Georgia 30602
Becky L. Rosenthal • Department of Educational Psychology, University of Georgia, Athens, Georgia 30602
Marcel Kinsbourne • Department of Behavioral Neurology, Eunice Kennedy Shriver Center, Waltham, Massachusetts 02254
MarionJ. Selz • Rehabilitation Psychology, St. Mary's Hospital, Tucson, Arizona 85703
Bryan Kolb • Department of Psychology, University of Lethbridge, Lethbridge, Alberta T1K 3M4, Canada Linda K. Krein • Department of Psychology, Colorado State University, Fort Collins, Colorado, 80523 Che Kan Leong • Department for the Education of Exceptional Children, University of Saskatchewan, Saskatoon, Saskatchewan S7N OWO, Canada John C. Linton • Department of Behavioral Medicine and Psychiatry, West Virginia University Medical Center-Charleston Division, Charleston, West Virginia 25326 Charles J. Long • Psychology Department, Memphis State University, Memphis, Tennessee 38152 Lawrence V. Majovski • Huntington Medical Research Institutes, Advanced Neurosurgical Laboratories, Pasadena, California 91105 Nancy L. Nussbaum • Learning Diagnostic Center/ Austin Neurological Clinic, Austin, Texas 78705 Antonio E. Puente • Department of Psychology, The University of North Carolina, Wilmington, North Carolina 28403-3297
Daniel J. Reschly • Department of Psychology, Iowa State University, Ames, Iowa 50011-3180
Phyllis Anne Teeter • Department of Educational Psychology, University of Wisconsin, Milwaukee, Wisconsin 53201 Cathy F. Telzrow • Cuyahoga Special Education Service Center, Maple Heights, Ohio 44137 Michael G. Tramontana • Department of Psychology, Bradley Hospital, East Providence, Rhode Island 02915; and Department of Psychiatry and Human Behavior, Brown University, Providence, Rhode Island 02912 Marie L. Walker • Department of Educational Psychology, University of Texas, Austin, Texas 78712 Timothy B. Whelan • Department of Psychiatry, Bay State Medical Center, Springfield, Massachusetts 0 ll99
Greta N. Wilkening • Department of Neurology, The Children's Hospital, Denver, Colorado 80218 Sheryl L. Wilson • Department of Psychology, University of Arizona, Tucson, Arizona 85721 Robert Henley Woody • Department of Psychology, University of Nebraska at Omaha, Omaha, Nebraska 68182
Preface Alexandra Luria sometimes would use one or two tests with a child and other times ten or fifteen tests. His diagnostic ability was unpredictable, irretrievable, and sometimes appeared magical. Attempts to copy his genius in assessment instruments or clinical methods have varied in success, and years later he still stands out as a unique and imaginative pioneer in clinical neuropsychology. Today we have come much further in our ability to assess children uniformly. We have a plethora of informal assessment instruments that can be given in clinics and in the schools to screen children, saving countless hours and money for parents, children, and the medical community. We have the latest sophisticated equipment that can scan the brain for pathology, giving precise localization of problems in a moment. This is certainly a time to celebrate in the development of the field of child neuropsychology. Although a relatively young field, clinical neuropsychology has moved far and fast and is now recognized as a necessity in most medical schools and in professional preparation programs in clinical, pediatric, and school psychology. Neuropsychology is also clearly present in the public schools and, although most prominent in the area of learning disabilities, contributes to all areas of handicapping conditions. Children with severe head injuries are no longer relegated to special schools or institutional settings. The child neuropsychologist, then, works in liaison with schools, not just in purely clinical settings. Never before has science worked in the schools. Thousands of school personnel actively assess soft neurological signs (that may or may not affect learning) and they communicate directly with child neurologists and neuropsychologists who evaluate the hard neuropsychological signs. This is definitely a field at work. The practice of clinical child neuropsychology remains an enigma in many ways. Children are growing and developing physically, neurologically, behaviorally, and emotionally, albeit in lawful if not well-understood ways, in uneven spurts, and not at all in concert across dimensions. Difficult problems
such as localization, prognostication, and differential diagnosis of emotional versus brain-behavior problems with adults become geometrically more abstruse with the interactions present in the course of the child's development. This volume thus tries to take a developmental view in examining neuropsychological function and development that is both normal and abnormal. We also try to take a stand on the necessity of principles of scientific inquiry, psychometric methods, and clinical acumen as central to clinical aspects of child neuropsychology. None can stand alone; the volume presents diverse approaches ranging from the heavily clinical to the near actuarial. Ideally, this work will foster a melding of such ideas, practices, and concepts into stronger clinical practice. The work is eclectic in the theoretical approaches touted as well. There is no more complex problem in all of the sciences than understanding developing brain-behavior relations, and we have far to go. We also have much to offer, a good deal of which is reviewed and discussed-along with its limitations-in this Handbook. There is another side of the coin, however. There are many quasi-professionals with perhaps a weekend of training using neuropsychological assessment, overinterpreting the results, and scaring parents with pseudoscientific neurological interpretations. Clinical neuropsychology is young and popular and too many people fail to recognize appropriate educational standards or even the need for years of training and supervision. This Handbook also addresses these highly controversial issues. There are also heavy criticisms of neuropsychological assessment instruments, rehabilitative therapies, and the inferential level of neuropsychological research studies; there are even some who negate any objective involvement at all. This book is intended to serve the children who need neuropsychological services in several ways: ( 1) as a text for undergraduate and graduate courses in clinical child neuropsychology; (2) in a more general way, as a text that crosses all aspects of this rapidly growing field; (3) as a reference source to vii
viii
PREFACE
practitioners and professors in the discipline; and (4) as a potential "idea book" for researchers in the field. We feel that it is important to include varied aspects of opinion on proper clinical practice to represent the field faithfully. Clinical neuropsychology is too young for a consolidation of opinion that would give it a singular approach to practice. Knowledge is developing too rapidly as well to allow us such luxury. Unfortunately, this has led to a polemic-at time vitriolic-division of the field into divergent theory-driven camps. It is too easy, with a subject this size, to use reductionist tactics to forward a cause or ideology that may not be in the best interests of children. The radical behaviorist position, which seems to deny us the convenience of metaphor and the use of inferential constructs we see as so essential to the scientific process, is represented in a chapter by Reschly and Gresham. Although we find the views in the chapter myopic at best, it is an accurate portrayal of the behaviorist view of clinical neuropsychology and its application to disorders of learning. We have presented this disagreeable view to help provide balance and to show the contrasting stance of other theoretical positions. They are particularly harsh on the use of neuropsychology in educational settings, which seems appropriate because-in the remainder of the volume-we have included information not only on clinical applications in the schools but also on developing neuropsychology services in the public schools. For those who are in the field, we hope to have provided some exciting arguments from chapter to chapter. For those new to the field, we hope to have provided a grand but no less than geode understanding of the entire scope of the field. Part !-Foundations and Current Issues-gives an overview of the history and development of the field and introduces basic foundations such as mea-
surement theory, neuropsychological frameworks such as hemispheric specialization, and bases for psychopathology. Part II addresses neuropsychological diagnosis. Part III focuses on techniques of intervention, and Part IV speaks to new aspects such as the neuropsychologist in private practice and establishing neuropsychological assessment and principles in the schools. We hope that the inclusion of so many applications to the public schools will not offend those in the field: After all, that is where the children are. We would like to express our appreciation to several people who aided us in the completion of this work. Angela Bailey, who.performs superbly as our administrative editor on a variety of projects, was instrumental in coordinating our own work and keeping track of many of the details of the project. Mike Ash and Victor Willson of the Department of Educational Psychology at Texas A&M University deserve appreciation for their moral support and the precious places they always fmd to house our special projects. To each of our families, we express our thanks for your patience in our hours away. C.R.R. would like to extend a special note of gratitude to Julia; E. F. J. expresses her appreciation to David, Emma, and C.R.R. We also extend appreciation to Lawrence Hartlage for the demonstration of his remarkable clinical acumen, which continues to interest us in this field. Eliot Werner, Senior Editor at Plenum, as always proved a valuable ally and supporter in the entire development of the work. Last, we are grateful to our authors, for without their hard work this Handbook would not exist. Cecil R. Reynolds Elaine Fletcher-Janzen
Contents I. Foundations and Current Issues 1. Historical Perspectives in the Development of Neuropsychology as a Professional Psychological Specialty . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ANTONIO
3
E. PUENTE
2. Development ofthe Child's Brain and Behavior . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
17
BRYAN KoLB AND BRYAN PANTIE
3. Higher Cortical Functions in Children: A Developmental Perspective . . . . . . . . . . . . . . LAWRENCE
41
v. MAJOVSKI
4. Mechanisms and Development of Hemisphere Specialization in Children . . . . . . . . . .
69
MARCEL I
5. Neuropsychology of Child Psychopathology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MICHAEL
G.
TRAMONTANA AND STEPHEN
R.
87
HooPER
6. Neuropsychological Sequelae of Chronic Medical Disorders. . .................... .
107
RICHARD A. BERG AND jOHN C. LINTON
7. Neuropsychological Bases of Common Learning and Behavior Problems in Children . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MARION
J. SELZ AND SHERYL L. WILSON
8. Measurement and Statistical Problems in NE:uropsychological Assessment of Children . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CECIL
129
147
R. REYNOLDS
ix
CONTENTS
X
9. Models of Inference in Evaluating Brain-Behavior Relationships in Children EILEEN B. FENNELL AND RUSSELL
M.
167
BAUER
II. Neuropsychological Diagnosis 10. Halstead-Reitan Neuropsychological Test Batteries for Children NANcY
L.
NussBAUM AND ERIN
D.
BIGLER
11. The Nebraska Neuropsychological Children's Battery . . . . . . . . . . . . . . . . . . . . . . . . . . . . CHARLES
J.
193
GOLDEN
12. Applications of the Kaufman Assessment Battery for Children (K-ABC) in Neuropsychological Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CECIL
181
R. REYNOLDS, RANDY
w. KAMPHAUS
I
AND BECKY
L.
205
ROSENTHAL
13. Neuropsychological Applications of Common Educational and Psychological Tests
227
CATHY F. TELZROW
14. Radiological Techniques in Neuropsychological Assessment . . . . . . . . . . . . . . . . . . . . . . ERIN
D.
247
BIGLER
15. Psychophysiological Evaluation of Children's Neuropsychological Disorders
265
THALIA HARMONY
16. Techniques of Localization in Child Neuropsychology
291
GRETA N. WILKENING
17. Neuropsychological Sequelae of Substance Abuse by Youths
311
ROBERT W. ELLIOTT
III. Techniques of Intervention 18. Neuropsychological Models of Learning Disabilities: Contribution to Remediation CHE
I
335
CONTENTS
19. Neuropsychological Approaches to the Remediation of Educational Deficits . . . . . . . .
xi
357
PHYLLIS ANNE TEETER
20. The Biofeedback Treatment of Neurological and Neuropsychological Disorders of Childhood and Adolescence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ROBERT
L. HODES
21. Approaches to the Cognitive Rehabilitation of Children with Neuropsychological Impairment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . JEFFREY
377
w.
GRAY AND RAYMOND
s. DEAN
22. Neuropsychological Aspects of Epilepsy: Introduction and Overview . . . . . . . . . . . . . . LAWRENCE
C.
HARTLAGE AND PATRICIA
23. The Neuropsychology of Epilepsy: THOMAS
409
L. HARTLAGE
~sychological
and Social Impact
419
L. BENNETT AND LINDA K. KREIN
24. Neuropsychological Effects of Stimulant Medication on Children's Learning and Behavior . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RoNALD T. BROWN AND
l
443
BoRDEN
25. Nonstimulant Psychotropic Medication: Side Effects on Children's Cognition and Behavior . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MANUEL
397
475
L. CEPEDA
IV. New Aspects of Neuropsychology 26. Establishing Neuropsychology in a School Setting: Organization, Problems, and Benefits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
489
RUTH ADLOF HAAK
27. Current Neuropsychological Diagnosis of Learning Problems: A Leap of Faith . . . . . .
503
DANIEL}. RESCHLY AND FRANK M. GRESHAM
28. Child Behavioral Neuropsychology ARTHUR MAcNEILL HoRTON, JR.
521
xii
CONTENTS
29. Coping and Adjustment of Children with Neurological Disorder . . . . . . . . . . . . . . . . . . TIMOTHY B. WHELAN AND MARIE
L.
WALKER
30. Child Neuropsychology in the Private Medical Practice ERIN
535
557
D. BIGLER AND NANCY L. NussBAUM
31. Public Policy and Legal Issues for Clinical Child Neuropsychology
573
ROBERT HENLEY WOODY
32. Training and Credentialing in Child Neuropsychology . . . . . . . . . . . . . . . . . . . . . . . . . . . LAWRENCE
C.
585
HARTLAGE AND CHARLES J. LONG
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
595
I Foundations and Current Issues
1 Historical Perspectives in the Development of Neuropsychology as a Professional Psychological Specialty ANTONIO E. PUENTE
The growth of neuropsychology, and clinical neuropsychology in particular, has been rapid though poorly documented. Although clinical neuropsychology texts provide overviews on theories of brain function, only a few review how the field developed. This lack of information is not typical of related disciplines (e.g. , neurology) or of other specialties within psychology (e.g., clinical psychology). Clinical psychology, for example, has experienced rapid growth over the past 25-40 years and its development is well documented (Fox, 1982; Fox, Barclay, & Rogers, 1982). Documentation is helpful for a variety of reasons. First, students must be provided with a comprehensive analysis of the discipline's development. Historical perspectives should serve as foundation for a more comprehensive appreciation of current trends and limitations. Similarly, health professionals not directly involved in the field should have a clearer understanding of our techniques and trends, if not for the professional welfare of clinical neuropsychology, at least for the welfare of consumers serviced by the discipline. Finally, and becoming increasingly important, documentation must be available to individuals outside of health care who are in a position to affect the discipline through funding and legislation. This chapter chronicles and critiques the development of clinical neuropsychology as a professional or practitioner specialty in psychology. A brief history of research and clinical developments precedes a ANTONIO E. PUENTE • Department of Psychology, University of North Carolina, Wilmington, North Carolina 28403-3297.
discussion of the growth of publications, organizations, and continuing education activities in clinical neuropsychology. Recent trends in professional practice, certification, and credentialing are also addressed. The chapter concludes with suggestions for maximizing the growth and efficacy of the field.
Historical Perspectives in the Development of Neuropsychology Localization of brain function has been the focus of philosophers, physiologists, and psychologists for many centuries. Around 400 BC, Hippocrates attempted to correlate his behavioral observations with what he knew about anatomical localization; this was conjecture because he was legally and socially prohibited from dissecting the human body, especially the cranium. Later, Aristotle (not Plato) surmised that the heart was the seat of the mind. Almost 600 years after Hippocrates, Galen shifted the site of the mind to the brain. Clarification of overall mind function was later offered by Descartes who suggested that the soul was localized in the pineal gland. In 1810, Gall described cortical localization of function through the concept of phrenology. Flourens and Broca introduced more accurate observations of brain function during the mid 19th century, forcing physiologtsts to research localization of function more systematically and with more precise measurement tools. This research was informed by early recordings of brain function or dysfunction which can be traced to at least the 17th century (Gibson, 1962) when several cases describing 3
4
CHAPTERl
traumatic brain injury were documented. Precise experimental, rather than observational, analysis began with the electrical stimulation work of Fritsch and Hietzig in 1790. New scientific techniques were introduced to the study of brain function by one of Catell's students, Sheperd Franz, during the early part of this century. While in Washington, D.C., Franz taught KarlS. Lashley who, in turn, advanced the understanding of brain-behavior relationships as well as the theory of equipotentiality. During the mid 20th century, Nobel laureate Roger Sperry and colleagues extended this earlier work by developing procedures for experimentally examining disconnection syndromes. As noncultural variables have traditionally been attributed little value in neuropsychological information, it is surprising to note that different approaches to the application of neuropsychological knowledge have developed across three major cultures, i.e., North America, Russia, and Great Britain. The approach to clinical neuropsychological understanding in Russia grew from the classical psychophysiological reflexive studies of Pavlov and other Russian physiologists (Bechtereva, 1978). Clearly the best recognized individual to apply this orientation to clinical assessment of neuropsychological dysfunction was A. R. Luria (1902-1977). According to Luria (1970), there are two basic principles that guide assessment of brain dysfunction: localization of brain lesions and analysis of psychological activities associated with brain function. The Russian approach to assessment is based on a qualitative, rather than quantitative or psychometric method. Specifically, this approach attempts to provide a "clinical description using flexible but systematic sets of tests" (Luria & Majovski, 1977, p. 962). The foundation for this flexible approach is based on the concept that strong individual differences preclude development of accurate norms. Empirically derived analyses cannot replace a comprehensive understanding of brain or individual patient functioning. Each client presents with an individual set of symptoms; thus, individual hypotheses and experiments must be performed. Observation of form and content, replication, and flexibility of thinking are central to this approach. Whereas the methods of Luria represent the historical foundations for the clinical application of neuropsychological principles in Russia, Henry Head and Hughlings Jackson represent the foundation for British approaches to clinical neuropsychology. According to Beaumont (1983), British clinical neuropsychologists favor the "individual-centered
normative approach." Such an approach builds on the uniqueness of the individual and on the complexities of syndromes by tailoring the assessment. However, ·unlike the Russian methods, the British approach does rely on psychometric tests. An evaluation may begin with the Wechsler Adult Intelligence Scale and proceed to the Wisconsin Card Sorting Test, Halstead Category Test, or Trail Making Test, depending on the functions that are to be examined. Gaps in the assessment are filled with more experimental (i.e., poorly standardized) and individual tasks. A final yet important aspect of the British approach is the shift from strict localization (which is central to Luria's approach) to an understanding of behavioral and psychological deficits. Canadian and American or North American approaches to clinical neuropsychology have historical roots in the work of Franz and Lashley in Washington, D.C. However, the clinical or applied study of brain dysfunction in the United States could be traced back to Kurt Goldstein. Goldstein's (1939) approach to the study of brain dysfunction was similar to that of Luria's in the sense that he did not use psychometric tests and that an extensive clinical case study was favored over short, structured contacts (Hanfmann, Rickers-Ovsiankina, & Goldstein, 1944). Early psychometric approaches to brain assessment can be traced to Babcock (1930). However, it was Ralph Reitan who launched clinical neuropsychology in North America toward the now-accepted psychometric tradition. In his seminal paper in 1955, he indicated that the purpose of a neuropsychological evaluation was to measure deficits accurately in a standardized psychometric fashion. An interesting comparison of Reitan's and Luria's approach to brain assessment is found in Diamant ( 1981). An extension of Luria's approach into the psychometric realm of brain assessment served as the foundation for the work of Golden, Hammeke, and Purisch ( 1980) with the Luria-Nebraska Neuropsychological Battery. Although numerous criticisms have been leveled at this battery and approach (Adams, 1980), the battery continues to be used and with increasing regularity (Seretny, Dean, Gray, & Hartlage, 1986; Lubin, Larsen, Matarazzo, & Seeven, 1986). Although neuropsychology as a field of investigation has a long past, formal efforts in clinical neuropsychology have a more recent onset and more of a divergent geographical origin. Nevertheless, the discipline has recently made significant strides toward the understanding of brain function from both research and clinical perspectives.
HISTORICAL PERSPECTIVES
Journal and Book Publications
5
Georgemiller, and Hymen ( 1982) analyzed the affiliation, geographic region, and context of manuThe proliferation of head-injured World War II scripts published between 1979 and 1983 in these veterans into Veterans Administration domiciliary journals. Whereas the University of Nebraska represettings occurred with the rapid growth of clinical sented 11.3% of articles in CN, a wide variety of psychology. Such growth was steady though not nec- universities (e.g., City College of New York) were essarily remarkable through the 1950s and 1960s. represented in JCN. Southern and north central states This growth is well chronicled in research and were the geographic origin of articles in CN, whereas clinical studies published in various journals. For northeastern states and Canadian locations were betexample, Reitan's (1955) classical psychometric ter represented in JCN. Furthermore, institutional study of head-injured adults appeared in the Journal contribution did not overlap from one journal to the of Comparative and Physiological Psychology other. CN was found to be more assessment oriented (JCPP). However, by about 1950 well over 70% of (e.g., Luria-Nebraska Neuropsychological Battery) the studies in JCPP used the Norway rat and close to whereas JCN focused more on methodological and 80% of the studies dealt either with conditioning and theoretical articles, 17.1 and 30.2% of articles, relearning or with reflexes and simple reaction patterns spectively. In summary, Ryan et al. (1982) sug(Beach, 1950). In many respects, Reitan's article gested that ''one journal (CN) will become more was not mainstream physiological psychology. identified with practical issues while the other (JCN) To document publication trends, all clinical will deal more with academic interest.'' neuropsychological citations in Psychological AbGeorgemiller, Ryan, and Setley (1986) later stracts, Biological Abstracts, and Index Medicus as surveyed 115 sites offering neuropsychological trainwell as in three separate computer searches were ing and asked the directors to rate the value of neurochronicled. The trend of published articles approxi- psychology-related journals. In order of perceived mates two articles per year until about 1960. From importance were Journal of Clinical and Experimen1960 until 1975, a sharp rise in publication rate oc- tal Neuropsychology, Journal of Consulting and curred with an average of 28 articles per year. Be- ·Clinical Neuropsychology, Clinical Neuropsycholtween 1974 and 1985 the rate continued to rise sharp- ogy, Cortex, Archives of Neurology, Brain, Brain ly to about 66 articles per year. These articles were and lAnguage, Journal of Clinical Psychology, and found in a wide variety of esoteric and interdisciplin- Archives of General Psychology. ary journals; to date, 161 different journals have pubIf one examines book publishing, similar lished articles on neuropsychology. The journals growth patterns emerge. Prior to the 1970s, applicapublishing the most articles include (in alphabetical tion of neuropsychological principles was rarely covorder) American Journal of Psychiatry, Clinical ered in clinical psychology or related texts. HowevNeuropsychology/International Journal of Clinical er, such books as Lezak's Neuropsychological Neuropsychology, Cortex, International Journal of Assessment (1976) and Golden's Diagnosis andReNeuroscience, Journal of Clinical Neuropsychol- habilitation in Clinical Neuropsychology (1978) durogy/Journal ofClinical and Experimental Neuropsy- ing the 1970s and more recently Filskov and Boll's chology, Journal of Clinical Psychology, Journal of Handbook of Clinical Neuropsychology (1981) and Consulting and Clinical Psychology, Neuropsycho- Wedding, Horton, and Webster's The Neuropsychollogia, andPerceptualandMotorSkills. TheArchives ogy Handbook (1986) have introduced this "new" of Clinical Neuropsychology, The Clinical Neuro- field to clinical and nonclinical psychologists. Severpsychologist and Neuropsychology are new to the al publishing companies have mounted intensive field, exclusively publishing clinical neuropsycholo- efforts in the field and one, Plenum, has developed a gical studies, and have not been in existence long book series entitled Critical Issues in Clinical Neuroenough to have been listed but should certainly psychology. This Handbook is one of the ftrst books achieve such status quickly. to be published in the series. Until recently, the two major neuropsycholWedding, Franzen, and Hartlage (1987) reogical journals addressing clinical issues were cently reported the number of clinical neuropsycholClinical Neuropsychology (CN; now International ogy books published yearly from 1960 to 1986. BeJournal ofClinicalNeuropsychology) and Journal of tween 1960 and 1967 an average of less than one Clinical Neuropsychology (JCN; now the Journal of book was published per year. Twenty years later the Experimental and Clinical Neuropsychology). In an average number published per year was well over 10, interesting study of publication trends, Ryan, with close to 25 books published in 1986. This pro-
6
CHAPTER!
liferation of books and journals shows no indication of slowing.
Professional Organizations Along with the proliferation of published findings has been the development of three major organizations representing clinical neuropsychology. The International Neuropsychological Society (INS) was established in 1970 by individuals of varying disciplines interested in neuropsychological issues. In 1973 the flfSt meeting of INS was convened in New Orleans. Such well-known clinicians as Benton, Butters, Goldstein, Hartlage, Kinsbourne, Mirsky, Pribram, Rourke, and Satz dotted the small yet robust program. Others, such as Fred King, now director for the Yerkes Primate Research Center in Atlanta, represented more academically oriented disciplines such as physiological psychology and the neurosciences. From the 1970 membership of 175 (mostly from North America), INS has grown to 2000 members worldwide representing a variety of disciplines, including speech pathology, clinical neurology, and neuropsychology. Of these, over 600 reside outside of the United States. To accommodate the large number of non-North American members, INS now holds two meetings per year-one in North America and the other in Europe. Over 800 attended the 1987 USA meeting (Washington, D.C.) with a smaller number attending the European meeting (Barcelona, Spain). According to Hartlage (1987), the National Academy of Neuropsychologists (NAN) evolved from a group of INS and APA members interested in developing a separate organization with national representation as well as focusing on the professional aspects of neuropsychology. The first formal meeting was held in August 1976 at the Washington School of Psychiatry. Robert Woody, the flfSt president of NAN, was instrumental in initiating a newsletter, Gram-ma. W. Lynn Smith chaired the next meeting held in cooperation with the annual APA meeting. In 1981, NAN met independent of APA in Orlando, Florida, with approximately 220 individuals registered. Since Orlando, NAN has met in Atlanta, Houston, San Diego, Philadelphia, Las Vegas, and Chicago ( 1987). Registration for the Chicago meeting exceeded more than 350 and membership in the organization is currently nearing 1000. Although members reside in many different countries, including Australia and several countries in Europe, membership is largely composed of individuals from the United States and Canada. Most members
are practicing, rather than academic, professionals and a large majority are involved in direct service, typically in private practice settings. NAN membership requirements include specific training and experiential components rather than the simple interest criteria of INS. Division 40 of the APA (Division of Clinical Neuropsychology) was formed in 1980 to serve the growing interest of APA members in clinical neuropsychology. Initially, APA members from other divisions (e.g., 6 and 12) joined to petition for this Division. As of January 1986, the Division had 49 fellows, 1829 members, and 153 associates with rapid growth expected to continue. Ancillary divisions within APA as well as numerous non-APA groups have also experienced growth indirectly associated with clinical neuropsychology. The Association for the Advancement of Behavior Therapy has members interested in clinical neuropsychology as does Divisions 38 (Health Psychology) and 6 (Physiological and Comparative Psychology). In nonpsychologically oriented organizations, similar growth has been observed in groups such as the Society for Neuroscience. Finally, it is worth noting that although not associated with specific national or international organizations, geographically limited groups have surfaced throughout the United States and abroad. Groups in New York, Philadelphia, California, and Puerto Rico (to name a few) have been formed to serve more local needs. One particular group, the Philadelphia Neuropsychological Society, has recently launched its own journal. Thus, strong evidence exists that organizations in clinical neuropsychology are and, most likely, will continue flourishing.
Continuing Education Though a relative newcomer to the discipline, another area of growth in clinical neuropsychology has been continuing education and the freestanding workshop. These activities have served as a central focus in providing training for clinical neuropsychologists. Recent examples of these workshops include: mild head injury in New York; Luria-Nebraska in Chicago; traumatic head injury in Braintree, Massachusetts; head trauma in Kansas City; dementia in Baltimore; behavioral neurology and neuropsychology in Lake Buena Vista, Florida; and head injury rehabilitation in Williamsburg. Virginia. Many of these freestanding workshops now also have free communications or poster sessions as part of the program.
HISTORICAL PERSPECTIVES
Although these freestanding workshops often sharpen the skills of professionals, they potentially pose significant complications (e.g., retraining involves more than workshops). To avoid these pitfalls, Meier (1987) outlined the concept of "Learning and Assessment Center'' for clinical neuropsychology practice to define personal insufficiencies and to provide the necessary knowledge, scientific or professional, to respecialize in clinical neuropsychology. Educational materials would be based on identified knowledge, skills, and attention needed to practice and provide an advanced level of proficiency. The information to be disseminated could take one of several forms including workshops, conventions, and published materials. However, of the specific modalities of professional information, some forms of dissemination appear to be more efficient than others. For example, Allen, Nelson, and Sheckley ( 1987) reported on continuing education activities of Connecticut psychologists. Books and contacts with other professional psychologists were the most favored continuing education activities, with books rated as the most valuable. Surprisingly, the average respondent read 9.9 books, 3.8 journals, and attended 2.5 workshops and 2.2 conventions per year. Continuing education activities appear to be critical in the development of professional practice and to date, numerous opportunities have been available for those interested in furthering their training in clinical neuropsychology.
Psychological Health Care Personnel and Practice During the 1970s there occurred an influx of psychology personnel into the workplace. This influx has continued unabated and has affected clinical neuropsychology. According to Stapp, Tucker, and VandenBos (1985), the estimated number of psychology personnel in the United States was 102,100 in mid 1983. Of these, 61.6% were primarily providing health services, 49.2% were involved in research and 63.7% in education. Approximately two thirds of these were doctoral level psychologists, and most (both master's and doctorate level) identified themselves with clinical psychology (followed by counseling and educational psychology). University settings were the largest single category of employment for doctorates; approximately 44% of the respondents were employed primarily in direct service through independent practice, hospitals, clinics, or
7
counseling centers. Interestingly, approximately 50% of the respondents have secondary employment, engaging in independent group or individual private practice. Of those providing health services, over half were involved in clinical activities in independent practices, clinics, hospitals, or counseling/other service settings (in order of prominence). Although not as current as the Stapp et al. (1985) data, Dorken and Webb (1981) reported large increases in the number of clinical psychologists during the mid to late 1970s, supporting the trend that more and more individuals are providing health care, regardless of their primary employment. In a recent analysis of doctorate productions by subfield, the APA Committee on Employment and Human Resources (Howard et al., 1986) indicated that whereas 50 years ago 70% of all new PhD recipients were in experimental psychology, in 1984 53.2% were in health services specialties. Furthermore, "the trend was for new doctorate recipients in clinical, counseling, and school psychology to increasingly assume positions in organized human service settings'' (Howard et al., 1986, p. 1322). In a review of the 1982 APA's Human Resources Survey, VandenBos and Stapp (1983) provided a detailed analysis of the characteristics of practice settings of service providers, profiles of professional practice, and other aspects of independent practice. Whereas the first two issues were addressed in the Stapp et al. (1985) report, issues of professional practice were more comprehensively described by VandenBos and Stapp (1983). Health problems, substance abuse, mental retardation, and schizophrenia combined represented close to 50% of the types of client problems seen by health providers. In an earlier study by VandenBos and colleagues (VandenBos, Stapp, & Kilberg, 1981), about40% of the respondents reported performing complete assessments regularly or often. These results strongly suggest that the number of health providers is rapidly increasing and they are quickly becoming the majority of psychology personnel in the United States. Although independent practice groups or individuals appear to be enjoying the most significant growth, similar trends are seen in all health service settings (e.g., hospitals). Assuming that a significant percentage of all health service clients have organic disorders and that few health providers limit their practice to one type of service or client (VandenBos and Stapp, 1983), one may conclude that a large percentage of psychologists are involved either in therapy Ol' in assessment of individuals with neuropsychologically based problems. According to VandenBos and Stapp (1983), "it is
8
CHAPTER!
interesting to note that psychologists tend not to specialize with specific problems or with specific age populations" (p. 1346). Thus, more psychologists will either eventually be involved in or specialize in clinical neuropsychological services. Using Social Security Administration data as a specific example, organic clients do comprise a significant segment of general clinical practice. In fiscal years 1984 and 1985, 22.5% (or approximately 250,000 people) of all those applying for Social Security disability were mental impairment cases (Dapper, 1987). Of these, 6% were classified as organic mental disorder and 35% as mentally retarded. Retardation could be ''caused'' by head trauma, for example, as IQs and not etiologies ofbehavioral disruption are important in a disability evaluation. This accounts for between 200,000 and 300,000 potential neuropsychological clients per year. In summary, it appears that not only are more psychologists dealing with neuropsychologically impaired clients, but that a large number of mental health consumers have organic disorders.
Professional Practice The practice of clinical neuropsychology has become increasingly popular in the last few years. Several surveys over the last 5 years have outlined the current practice of clinical neuropsychology in the United States. In 1982, Hartlage and Telzrow completed a mail survey of all members of the National Academy ofNeuropsychologists. Four major content areas were covered: professional practice, tests, practice preparation, and most important figure in the history of clinical neuropsychology in the United States. From these data, one can develop a basic sense of a "typical" clinical neuropsychological practice. The mean neuropsychological evaluation time was 8 hours. Approximately 59% of the respondents used technicians. Only three tests were used by at least 50% of the respondents and these included the Wechsler Intelligence Scales (89%), portions of the Halstead-Reitan Neuropsychological Battery (56%), and the Wide Range Achievement Test (52%). The remaining tests used were (in order of descending popularity): Bender Gestalt Test, Halstead-Reitan Neuropsychological Battery, Benton Visual Retention Test, Luria Tests (Christensen or Golden versions), Wechsler Memory Scale, Memory for Designs, and the Minnesota Multiphasic Personality Inventory (MMPI). With regard to practice preparation, 78% of the respondents indicated that
clinical psychology was the best preparation for delivery of clinical neuropsychology services. Finally, Benton and Golden were cited for their unique contributions to the development of clinical neuropsychology as a professional specialty. In a more recent study, Seretny et al. (1986) surveyed members from APA's Division 40 (N = 314) and the National Academy of Neuropsychologists (N = 300). The purpose of the survey was essentially to follow-up earlier surveys and to expand the information available for specific aspects of professional practice settings. Private practice was reported as the primary work. setting of the respondents, followed by (in decreasing order of occurrence) hospitals, medical schools, and academic settings. According to the authors, there has been a shift to private practice settings over the last 5 years. Whether this is due to academicians going into practice or new doctorate recipients choosing professional rather than academic settings, or both, is uncertain. Little has changed in terms of the average number of evaluations per month ( 11.13) or the average amount of time required to complete a full evaluation (7 .30 hours). About half of the respondents employed technicians. The Wechsler Intelligence Scales remained the most frequently used instruments followed by the Halstead-Reitan and the Luria-Nebraska Neuropsychological batteries. Other single tests frequently used included the WRAT, Bender, and Benton, as well as the MMPI and the Wechsler Memory Scale. A wide variety of referral sources was cited. Referrals were primarily from neurologists, although neurosurgeons, psychologists, general physicians, physical rehabilitation specialists, and attorneys also referred regularly. The mean dollar amount for a complete neuropsychological evaluation was $479.30 or about $65.65 per hour. Most respondents indicated that they were involved in some nonneuropsychological activities as well as cognitive rehabilitation and forensic evaluations. These results support the fact that specific trends are surfacing in terms of tests used, time used for an evaluation, and the use of technicians. Additional longitudinal information would be useful with regard to such issues as cost of service, place of employment, and referral sources. Ryan, Parage, and Lips (1983) identified psychological health providers in the 1981 version of the
National Register ofHealth Service Providers in Psychology and the winter 1981-1982 supplement, who
offered neuropsychological services in an effort to understand the geographical distribution of persons providing such service. The range noted was relatively large with the District of Columbia having one
HISTORICAL PERSPECTIVES
neuropsychology provider per 42,510 persons, and South Dakota with one provider per 690, 178. The top ten states in terms of per capita providers were (in rank order): California, New York, Texas, Pennsylvania, llinois, Ohio, Florida, Michigan, New Jersey, and North Carolina. Alaska and several midwestern states had the fewest number of practicing neuropsychologists. It is interesting to note that in Indiana only 21 individuals offered neuropsychological services and only one did so in South Dakota. These states represent the geographical origins of the HalsteadReitan (Indiana) and Luria-Nebraska Neuropsychological batteries (South Dakota). Sladen, Mozdzierz, and Greenblatt (1986) also examined the geographical distributions of neuropsychological service providers using the same subject selection criteria as Ryan et al. (1983). Sladen et al. reported ''marked disparities'' in the distribution of professional psychologists providing clinical neuropsychological services. These disparities existed both among states as well as between rural and metropolitan areas. Generally, these services were offered more frequently in densely populated states and in metropolitan centers. In a related survey of 316 randomly chosen community hospitals, Anchor (1983) reported on the availability and awareness of neuropsychological services. The average hospital in the survey had 143 beds and all had emergency rooms. Only 8.5% of the hospitals offered any type of neuropsychological, neuropsychiatric, or neurological testing services. A recent article by Molloy ( 1987) indicated that conditions facing neuropsychologists appear more difficult in countries outside of North America. According to Molloy, insufficient understanding, prejudicial distrust, and limited reimbursement have hampered the development of neuropsychology as a clinical specialty in Australia. However, lawyers and medical specialists appear to constitute the primary source of referrals there. Such patterns appear prevalent in other countries as well. For example, in Spain, pockets of practitioners exist only in larger cities (Madrid and Barcelona). In other countries such as Argentina, clinical neuropsychological services are essentially nonexistent.
Certification and Credentialing Although a separate chapter on certification, training, and credentialing is found in this Handbook, the current models of training and credentialing have their roots in and have an impact on historical trends and, thus, warrant historical analysis.
9
Until the I 980s, clinical neuropsychology was not a formally recognized subspecialty in health care. Behavioral neurologists, speech pathologists, and clinical psychologists (among others) with an interest in brain dysfunction worked using informal titles, and in many cases, constructs. Realizing the need for specific professional identity in the applied field of brain dysfunction, several psychologists within several Divisions of APA, especially 6 (Physiological and Comparative) and 12 (Clinical Psychology), as well as within INS and NAN, initiated discussion for the development of guidelines that would define how a psychologist (and not an individual of a related discipline) would be identified as a clinical neuropsychologist. Prior to the development of Division 40, informal discussions were centered within INS circles. By the early 1980s, APA Division 40 had been formed, and the original group of individuals developing these guidelines split into two major factions. One group, who remained entrenched within INS, developed the American Board of Clinical Neuropsychology, Inc. (ABCN), by late 1982 for the purpose of awarding specialty diplomas in clinical neuropsychology. The initial eligibility criteria required the following: A. Doctoral degree in psychology from aregionally accredited university B. Licensed or certified at the level of independent practice in some state or province C. Areas of training and experience included: I. Basic neurosciences 2. Neuroanatomy 3. Neuropathology 4. Clinical neurology 5. Psychological assessment 6. Oinical neuropsychological assessment 7. Psychopathology 8. Psychological intervention D. Five years of postdoctoral professional experience in psychology which could include: Clinical Research Teaching Administration E. Three or more years of clinical neuropsychological experience defined as follows: I. Equivalent of at least I year of full time supervised clinical neuropsychology experience at the postdoctoral level (6 months may be credited for documented predoctoral
10
CHAPTERl
specialty internship in neuropsychology) 2. Equivalentofat]east 1 yearofadditional experience as a clinical neuropsychologist 3. In the absence of any supervised clinical experience, the equivalent of 3 years of unsupervised postdoctoral experience as a clinical neuropsychologist The application fonn also required the submission of: 1. A copy of current state license or certificate, 2. Names of two professionals who could attest to the extent, nature, and quality of your experience and competence in clinical neuropsychology.
2. Licensure by a state board of psychology 3. At least 5 years' postdoctoral experience in professional neuropsychology 4. Combination of coursework; additional training such as continuing education workshops, supervised pre- or postdoctoral training; and relevant work experience to provide evidence of high degree of competence in professional neuropsychology 5. Recommendation by at least two supervisors or professional colleagues attesting to high degree of competencies in professional neuropsychology
To date, ABCN and ABPN have not made strides to merge and appear to be taking independent courses. However, regardless of the apparent split, agreement has been reached on specific guidelines In April 1985, Manfred Meier, the president of for identifying requirements for clinical neuropsyABCN, announced affiliation of this group with the chology education. To alleviate potential complications for indiAmerican Board of Professional Psychology viduals wishing to be trained in the field of clinical (ABPP). According to Meier, ABPP voted on March neuropsychology, recent guidelines have been pub4-5, 1985, to add clinical neuropsychology to the lished by INS and Division 40 (August 1986). The existing fields of applied competency of clinical, counseling, school, and industrial/organizational guidelines, in their entirety, are as follows: Doctoral training in Clinical Neuropsychology should psychology. As a function of its ABPP affiliation, ordinarily result in the awarding of a Ph.D. degree from ABCN adopted ABPP's defmition of a psychology a regionally accredited university. It may be accomgraduate program as well as adding several related plished through a Ph.D. program in Clinical Neuropsyrequirements (e.g., APA membership). At the curchology offered by a psychology department or medirent time, the ABPP/ ABCN examination includes a cal faculty or through the completion of a Ph.D. "work sample" (e.g., neuropsychological evaluaprogram in a related specialty area (e.g., Clinical Psytion or treatment summary) as well as an oral examchology) which offers sufficient specialization in ination. The oral examination involves analyses of Clinical Neuropsychology. the work sample, ethics, and a sample case. The Training programs in Clinical Neuropsychology written multiple-choice examination is being stanprepare students for health service delivery, basic clinical research, teaching and consultation. As such, dardized as of this writing. The new ABCN-ABPP they must contain (a) a generic psychology core, (b) a examination has become more refmed and extensive generic clinical core, (c) specialized training in the neurelative to the original criteria published in 1982. rosciences and basic human and animal neuropsycholThe American Board of Professional Neuropsyogy, (d) specific training in clinical neuropsychology. chology (ABPN) was developed in 1982 under the This should include a 1800 hour internship which direction of Lawrence Hartlage and several other should be preceded by an appropriate practicum ABPP members, most of whom were associated with experience. NAN. The guidelines for the original diplomate staA. Generic Psychology Core tus, which are shown below, essentially focused on 1. Statistics and Methodology relevant training and education, work with relevant 2. Learning, Cognition and Perception populations, and supervision with a qualified practi3. Social Psychology and Personality tioner. At present, the ABPN is being restructured 4. Physiology Psychology and will probably include actual testing along with an 5. Life Span Development extensive application fonn. The basic requirements 6. History have been as follows: B. Generic Clinical Core 1. Minimum educational requirement is the Ph.D. (or similar doctoral degree, e.g., Psy.D., Ed.D.)
I. Psychopathology 2. Psychometric Theory 3. Interview and Assessment Techniques a. Interviewing
HISTORICAL PERSPECTIVES
b. Intelligence Assessment c. Personality Assessment 4. Intervention Techniques a. Counseling and Psychotherapy b. Behavior Therapy/Modification c. Consultation 5. Professional Ethics C. Neuroscience and Basic Human and Animal Neuropsychology I. Basic Neurosciences 2. Advanced Physiological Psychology and Pharmacology 3. Neuropsychology of Perceptual, Cognitive and Executive Processes 4. Research Design and Research Practicum in Neuropsychology D. Specific Clinical Neuropsychological Training I. Clinical Neurology and Neuropathology 2. Specialized Neuropsychological Assessment Techniques 3. Specialized Neuropsychological Intervention Techniques 4. Assessment Practicum "Children and/or Adults in University Supervised Assessment Facility 5. Clinical Neuropsychological Internship of 1800 hours preferably in a university facility. (As per INS-Div. 40 task force guidelines). Ordinarily this internship will be completed in a single year, but in exceptional circumstances may be completed in a two-year period. E. Doctoral Dissertation It is recognized that the completion of a Ph.D. in Clinical Neuropsychology prepares the person to begin work as a clinical neuropsychologist. In most jurisdictions, an additional year of supervised clinical practice will be required in order to qualify for licensure. Furtherrnore, training at the post-doctoral level to increase both general and sub-speciality competencies, is viewed as desirable. Post-doctoral training, as described herein, is designed to provide clinical training, in order to produce an advanced level of competence in the speciality of clinical neuropsychology. It is recognized that clinical neuropsychology is a scientifically-based evolving discipline and that such training should also provide a significant research component. Thus, this report is concerned with post-doctoral training in clinical neuropsychology which is specifically geared toward producing independent practitioner level competence. which includes both necessary clinical and research skills. This report does not address training in neuropsychology which is focused solely on research. Entry into a clinical neuropsychology post-doctoral training program ordinarily should be based on completion of a regionally accredited Ph.D. graduate
training program in one of the health service delivery areas of psychology or a Ph.D. in psychology with additional completion of a "respecialization" program designed to meet equivalent criteria as a health services delivery program in psychology. In all cases, candidacy for post-doctoral training in clinical neuropsychology must be based on demonstration of training and research methodology designed to meet equivalent criteria as a health services delivery professional in the scientist-practitioner model. Ordinarily. a clinical internship, listed by the Association of Psychology Internship Centers, must also have been completed. A post-doctoral training program in clinical neuropsychology should be directed by a board certified clinical neuropsychologist. In most cases, the program should extend over at least a two-year period. The only exception would be for individuals who have completed a specific clinical neuropsychology specialization in their graduate programs and/ or a clinical neuropsychology internship provided the exit criteria are met (see below). As a general guideline, the post-doctoral training program should provide at least 50% of time in clinical service and at least 25% of time in clinical research. Variance within these guidelines should be tailored to the needs of the individual. Specific neuropsychology training must be provided, including any areas where the individual is deemed to be deficient (testing, consultation, intervention, neurosciences, neurology, etc.). Such a post-doctoral training program should be associated with hospital settings which have neurological and/ or neurosurgical services to offer and the training should be provided in both a didactic and experiential format and should include the following: A. Training in neurological and psychiatric diagnosis. B. Training in consultation to neurological and neurosurgical services. C. Training in direct consultation to psychiatric, pediatric, or general medical services. D. Exposure to methods and practices of neurological and neurosurgical consultation (Grand Rounds, Bed Rounds, Seminars, etc.). E. Observation of neurosurgical procedures and biomedical tests (revascularization procedures, cerebral blood flow, W ADA testing, etc.). F. Participation in seminars offered to neurology and neurosurgery residents (neuropharmacology, EEG, brain cutting, etc.). G. Training in neuropsychological techniques, examinations, interpretation of test results, report writing. H. Training in consultation to patients and referral sources. I. Training in methods of intervention specific to clinical neuropsychology.
11
12
CHAPTER1 1. Seminars, readings, etc., in neuropsychology (cases conferences, journal discussions, topic-specific seminars). K. Didactic training in neuroanatomy, neurosciences. Additional experiential training should be offered as follows: A. Neuropsychological examination and evaluation of patients with actual and suspected neurological diseases and disorders. B. Neuropsychological examination and evaluation of patients with psychiatric disorders and/ or pediatric or general medical patients with neurobehavioral disorders. C. Participation in clinical activities with neurologists and neurosurgeons (Bed Rounds, Grand Rounds, etc.). D. Experience at a specialty clinic, such as a dementia clinic or epilepsy clinic, which emphasizes multidisciplinary approaches to diagnosis and treatment. E. Direct consultation to patients involving neuropsychological assessment. F. Direct intervention with patients, specific to neuropsychological issues, and to include psychotherapy and/or family therapy where indicated. G. Research in neuropsychology, e.g., collaboration on a research project or other scholarly academic activity, initiation of an independent research project or other scholarly academic activity, and presentation or publication of research data where appropriate.
At the conclusion of the post-doctoral training program, the individual should be able to undertake consultation to patients and professionals on an independent basis. Accomplishment in research should also be demonstrated. The program is designed to produce a competent practitioner in the areas designated in Section B of the Task Force Report and to provide eligibility for certification in Clinical Neuropsychology by the American Board of Professional Psychology. (1986,
pp. 4-5)
As strict as these guidelines may be, they presumably are more prescriptive of how programs may develop neuropsychology tracks than descriptive of existing neuropsychology programs. For example, Golden and Kuperman (1980) surveyed all APA-approved clinical psychology graduate programs in North America in 1977. Approximately 60% of these programs offered clinical neuropsychology courses including lectures, practicums, and work placements. Interestingly, however, fewer schools were involved in neuropsychological research. Furthermore, most schools offered training by one or two staff members and relatively few had specific neuropsychology tracks.
Scheer and Lubin ( 1980) also published an independent survey of ''training" programs in clinical neuropsychology. In this survey, the authors sampled the 627 members of INS in 1977. The results indicated that, at most, training included "minimal neuropsychological training activities associated with primary service function and allied disciplines'' (e.g. , neurology). Several internship programs existed at both the pre- and the postdoctoral level. Scheer and Lubin noted that a typical pattern of training involved obtaining a standard Ph.D. with 1- or 2-year postdoctoral specialization in clinical neuropsychology .. Another pattern of training was to specialize in a spectfic Ph.D. program (e.g., clinical or neuroscience) with neuropsychological concentration. The authors reported that "notable pioneers" such as Benton, Satz, Milner, and Reitan followed this mode of training. One particularly interesting observation by Scheer and Lubin was that "notable pioneers" actually "individually designed" their curriculum. With the recent guidelines or recommendations, it would appear that such an approach would be increasingly difficult, if not impossible, to accomplish. Although Meier ( 1982) cogently argued for different models of education in clinical neuropsychology, it appears that such a variety of models might actually be replaced with one or two specific ones as pressure from licensing or credentialing groups becomes more defined and intensified. The major reason for certification and credentialing the practice of clinical neuropsychology is to ensure that our clients/patients in particular, and society in general, are not harmed by incompetent or unethical practitioners (Hogan, 1983a). However, unexpected, even undesired by-products of this current trend are generally not being considered. The first step in regulating a particular discipline is not to regulate the practitioners, but to regulate educational institutions or formal internships (Hogan, 1983b). This approach was first used in 1803 in the state of Massachusetts to regulate the medical profession (Shryock, 1967). However, Hogan (1983b) argued that despite the fact that sophistication in the licensing and certification process grows, "little evidence suggests that the quality of professional services has improved" (p. 121) (see alsoKessel, 1970,andGross, 1978). Specifically, he contended that such an approach is aimed to "eliminate competition, rather than incompetence." Second, such restrictions tend to increase the cost of services and limit the services to disadvantaged groups. According to Dorsey (1983), restrictions tend to decrease the "lower-quality and price" service that low-income individuals would be able to
HISTORICAL PERSPECTIVES
use. The possibility exists, furthermore, that individuals unable to be credentialed by formal procedures would be relegated to less prestigious jobs. Another issue is that having failed to be accepted at the highest level, an individual, due to interest or necessity, would still practice out of the mainstream of clinical neuropsychology without adequate direction, information, or technique (which would beespecially critical in clinical neuropsychology due to its inherent complexity and novelty). Furthermore, this trend would probably occur more frequently with minority populations. Indeed, recent statistics published by APA (Howard et al., 1986) indicate that although women are becoming increasingly represented, other minorities, including blacks and Hispanics, are not. Analysis of name lists available from both ABPP/ABCN and ABPN appears to confirm this. An alternative to such efforts is the more recent peer review system, first enacted by Congress in 1972 to monitor federally aided health care programs such as Medicare (Young, 1982). Such an approach is based on the assumption that if professional work is not acceptable, professionals would learn from their peers through defined interactive methods. Nevertheless, if testing is to be continued as a means to define clinical neuropsychology or neuropsychologists, a more accurate analysis of an individual's capabilities would be to have a certification process which is based on empirically validated situations. As Milton Friedman so aptly indicated in 1962, •'conforming to 'prevailing orthodoxy,' is certain to reduce the amount of experimentation that goes in [a discipline] and hence to reduce the rates of growth of knowledge" (p. 157). The current trend in clinical neuropsychology appears to be toward greater sophistication, efficacy, recognition, and certification. Sophistication, efficacy, and recognition are needed for the development of a healthy subspeciality in psychology. However, more "precise" certification may be, as Friedman (1962) argued, incompatible with experimentation. Such experimentation, after all, was the catalyst for our present growth and status. One way to assure such continued growth and status is to facilitate a system, in whatever way possible, that meets the needs of the public instead of protecting the public (Hogan, 1983a). This is especially true in a field such as clinical neuropsychology where the field of practice and appropriate standards of practice are being developed. Diversity and experimentation with appropriate empirical analysis and validation of what is professional or acceptable in clinical neuropsychology must be encouraged. If certification is to be used, several methods
13
may be applied to ensure continued and appropriate growth of our discipline (based on Hogan, 1983b). First, clinical neuropsychology practice should be narrowly defined. For example, how is clinical neuropsychology different from clinical psychology, behavioral neurology, or speech therapy? Second, standards used in defining qualified clinical neuropsychologist must be based on empirical assessment, competence, and related to actual (versus hypothetical) performance. Shimberg (1981) provided specific suggestions as to how these concerns may apply to psychology, in general, with regard to content, criterion, and construct validity of tests for psychological certification. Third, alternative paths to certification should be kept open (see also Meier, 1987, for specific suggestions on continuing education). This might include internships, postdoctoral fellowships, supervision, peer and client review, workshops, and at home/office study. Fourth, regulatory policies should be based on the representation of appropriate constituencies. This would involve clinical neuropsychologists with different approaches (even geographical locations) as well as government and possibly health care/insurance agency officials. Most of all, our clients or their representatives should also be included. Next, our goal should not be to restrict the right of a competent person to practice clinical neuropsychology, but to restrict the title clinical neuropsychologist. Finally, the consumers of our product should be educated. Psychologists, neurologists, attorneys, allied disciplines, agencies (to name but a few) who refer to clinical neuropsychologists should be e.ducated. According to Olson (1983), the unempirical exclusion of the competent as a legal protection of special interest ultimately has negative effects on the discipline and on society. Mahoney (1985) argued for the importance of •'open and ongoing exchange'' as part of the epistemological processes and that it is necessary for scientific progress in psychology. As Lakatos (1970) suggested, superseding the present by exploration and novelty (rather than assuring adherence to current orthodoxy) is critical to the development of any scientific discipline. Clinical neuropsychology has too much to offer; the need for our services, expertise, and knowledge is too critical to focus on limitations. If such theoretical arguments do not provide enough support for the continued development of clinical neuropsychology, the work of McCaffrey and colleagues (McCaffrey, 1985; McCaffrey & Isaac, 1984) provides additional reasons to be careful about our recent trend for excJusivity. In a survey of internship instructors of clinical neuropsychology
14
CHAPTER 1
(with low response rate), instructors fared surprisingly well on how they compared to INS/Division 40 guidelines. However, in a separate survey (with better response rate), McCaffrey and Isaac ( 1984) found that few instructors at the pre- or postdoctoral levels met the educational requirements of the INS/Division 40 guidelines. Possibly many of those currently providing training would themselves be excluded.
Future of Clinical Neuropsychology As Fishman and Neigher (1982) aptly noted, the 1980s (and presumably the 1990s) have been and will be an age of increasing accountability. We have taken psychology to the public, to the referral sources, to the government with an overwhelming degree of success. Indeed by the early 1980s, we were spending over $2 billion per year in the support of psychological endeavors (Fishman & Neigher, 1982). The public now wants to account for the $2 billion. One way clinical neuropsychology has provided accountability has been to provide the health marketplace with a plethora of assessment and rehabilitation techniques. The assumption has been that nonresearch or service activities have been built on a solid foundation of "scientifically derived knowledge.'' As altruistic and as interesting as these techniques may be, they must first be subject to the same rigorous scientific tests that other psychological and health practices (e.g., psychotherapy) have undergone. In a review of Fishman and Neigher's (1982) activity by mission matrix of psychology's goals, it appears that clinical neuropsychology has focused over the recent years mostly on service delivery. However, much effort needs to be placed on other aspects of the activity by mission matrix of these authors. These would include, in no specific order, basic and applied research, social policy consultation, education of the general (and health care) public, training of new clinical neuropsychologist&, continuing education of both general clinical and clinical neuropsychologists, and political advocacy for the discipline. To illustrate each of these areas, examples are presented. I. In terms of research, specific efforts must be placed on replicating existing studies (including those historically cited). Furthermore, studies that yield negative results must be considered, especially those focusing on rehabilitation. 2. In publishing, review and editorial process
must become more reliable and less biased. Creative imagination must have its place next to methodological rigor. 3. Meier ( 1987) recognized the importance of education in the general mission of our discipline. Great care must be taken to recruit and train new neuropsychologists. Of special significance here is the alarmingly low number of minorities (including women, blacks, and Hispanics) entering the discipline. Additionally, Meier ( 1987) also aptly considered continuing education as critical to the continued development of those practicing in the field. This becomes especially important as larger segments of individuals in clinical, counseling, school, and physiological psychology become interested in clinical neuropsychology. 4. With regard to social policy consultation, clinical neuropsychologists must leave their clinics, hospitals, and laboratories to go to their capitols, both in the United States and abroad. One critical test of clinical neuropsychology will undoubtedly be federal legislation recognizing our discipline (see DeLeon, VandenBos, & Kraut, 1984). The public, whether it be lay and uninformed or professional and in health care settings, must be educated to the unique contribution of clinical neuropsychology. Clinical neuropsychologists have typically been sheltered from outside-institution political issues. Although establishing political ties within institutional settings is clearly an important frrst step, national (and possibly international) political advocacy is needed. An excellent example of this was the Home versus Goodson case in North Carolina. In this workmen's compensation case, the testimony of a clinical neuropsychologist (the author) was considered by the initial Industrial Commission judge and later by the full Industrial Commission Board of the state as being incompetent and not credible because the injury involved physical ''brain damage.'' Despite repeated testimony and reports describing the role of clinical neuropsychology as defining behavioral (and not anatomical) dysfunction, the pleas went unneeded. With the assistance of both the North Carolina Psychological Association and the American Psychological Association, a comprehensive amicus brief was submitted on behalf of the claimant when the case was appealed to the North Carolina Court of Appeals. While the decision (North Carolina Court of Appeals, 1986) and amicus (American Psychological Association, 1986) are available elsewhere, the decision was overruled. In the dissenting opinion, Judge Phillips stated that it was erroneous to assume .. that only doctors of medicine can make more reliable deductions as to conditions in the brain." According
HISTORICAL PERSPECTIVES
to Judge Phillips, "psychology is the study of the human mind and how it works and . . . the brain controls conduct, thought, speech, feeling and judgment" (p. 2). Such decisions are clearly necessary as clinical neuropsychology will be increasingly tested in the courtroom and beyond. Clinical neuropsychology has made significant strides in recent times, both in terms of contribution, as well as recognition in the area of brain dysfunction. However, extreme cases must be taken to ensure continued growth and development. This chapter chronicles some of our successes and pitfalls. We must bear in mind, as Smith (1979) and Costa (1983) have, that clinical neuropsychology is indeed ''a discipline in evolution." Our history remains in our future.
References Adams, K. (1980). In search of Luria's battery: A false start. Journal of Consulting and Clinical Psychology, 48, 511516. Allen, G. J., Nelson, W. J., & Sheckley, B. G. (1987). Continuing education activities of Connecticut psychologists. Professional Psychology: Research and Practice, 18, 78-80. American Psychological Association and North Carolina Psychological Association. (1986). Amici curiae; Horne versus Goodson. (Unpublished manuscript available from APA, NCPA, or Antonio E. Puente) Anchor, K. N. (1983). Availability and awareness of neuropsychological assessment in the community hospital: A survey. International Journal of Clinical Neuropsychology, 5, 7-8. Author. ( 1986). INS/Division 40 guidelines approved. Division of Clinical Neuropsychology Newsletter, 4, 4-5. Babcock, A. (1930). An experiment in the measurement of mental deterioration. Archives of Psychology, 18, 5-105. Beach, F. D. (1950). The snark was a boojum.AmericanPsychol. ogist, 5, 115-124. Beaumont, J. G. (1983). Introduction to neuropsychology. New York: Guilford. Bechtereva, N. P. (1978). The neurophysiological aspects of human mental activity. New York: Oxford University Press. Costa, L. (1983). Clinical neuropsychology: A discipline in evolution. Journal of Clinical Neuropsychology. 5. 1-11. Dapper, N. J. ( 1987). Selected Social Security disability determination data concerning mental impairments, FY 84-FY 86. Baltimore: Social Security Administration. DeLeon, P. H., VandenBos, G. R., & Kraut, A. G. (1984). Federal legislation recognizing psychology. American Psychologist, 39, 933-946. Diamant, J. J. (1981). Similarities and differences in the approach of R. M. Reitan and A. R. Luria. Acta Psychiatrica Scandinavica, 63, 431-443. Dorken, H., & Webb, J. T. (1981). Licensed psychologists on the increase, 1974-1979. American Psychologist, 36, 14191429.
15
Dorsey, S. (1983). Occupational licensing and minorities. Law and Human Behavior, 7, 171-181. Filskov, S., & Boll, T. (1981). Handbook of clinical neuropsy· chology. New York: Wiley. Fishman, D. B., & Neigher, W. D. (1982). American psychology in the eighties. American Psychologist, 37, 533-546. Fox, R. E. (1982). The need fora reorientation of clinical psychology. American Psychologist, 37, 1051-1057. Fox, R. E., Barclay, A. G., & Rogers, D. A. (1982). The foundation of clinical psychology. American Psychologist, 37, 306316. Friedman, M. (1962). Capitalism and freedom. Chicago: University of Chicago Press. Georgemiller, R. J., Ryan, J. J., & Setley, K. N. (1986). Clinical utility rankings of neuropsychologically related journals. Professional Psychology: Research and Practice, 17, 278279. Gibson, W. C. (1962). Pioneers of localization of function in the brain. Journal of the American Medical Association, 180, 944-951. Golden, C. J. (1978). Diagnosis and rehabilitation in clinical neuropsychology. Springfield, IL: Thomas. Golden, C. J., Hammeke, T. A., & Purisch, A. D. (1980). Diagnostic validity of a standardized neuropsychological battery derived from Luria's neuropsychological tests. Journal of Consulting and Clinical Psychology, 46, 1258-1268. Golden, C. J., & Kuperman, S. K. (1980). Graduate training in clinical neuropsychology. Professional Psychology, 11, 55-63. Goldstein, K. (1939). The organism. New York: American. Gross, S. J. (1978). The myth of professional licensing. American Psychologist, 33, 1009-1016. Hanfmann, E., Rickers-Ovsiankina, M., & Goldstein, K. (1944). Lanuti: Extreme concretizations of behavior due to damage of the brain cortex. Psychological Monographs, 57. Hartlage, L. C. (1987). Brief history of the National Academy of Neuropsychologists. Bulletin of the National Academy of Neuropsychologists, 4, 5-6. Harlage, L. C., & Telzrow, C. F. (1982). The practice of clinical neuropsychology in the U.S. Clinical Neuropsychology, 2, 200-202. Hogan, D. B. (1983a). Professional regulation. Law and lfuman Behavior, 7, 99-101. Hogan, D. B. (l983b). The effectiveness of licensing. Law and Human Behavior, 7, 117-138. Howard, A., Pion, G. M., Gottfredson, G. D., Flattau, P. E., Oskamp, S., Pfafflin, S.M., Bray,D. W.,&Burstein, A. G. (1986). The changing face of American psychology. American Psychologist, 41, 1311-1327. Kessel, R. (1970). The A. M. A. and the supply of physicians. Law and Contemporary Problems, 35, 267-283. Lakatos, I. (1970). Falsification and the methodology of scientific research programmes. In I. Lakatos & A. Musgrave (Eds.). Criticism and the growth of knowledge. London: Cambridge University Press. Lezak, M. (1976). Neuropsychological assessment. London: Oxford University Press. Lubin, B., Larsen, R. M., Matarazzo, J. P., & Seeven, M. F. (1986). Selected characteristics of psychologists and psycho-
16
CHAPTER 1
logical assessment in five settings: 1959-1982. Professional Ryan, J. J., Georgemiller, R. J., & Hymen, S. P. (1982). ComPsychology: Research and Practice, 17, 155-157. parison of article sources and context in Clinical NeuropsyLuria, A. R. (1970). Traumatic aphasia. The Hague: Mouton. chology and Journal of Clinical Neuropsychology. UnLuria, A. R., & Majovski, L. V. (1977). Basic approaches used in published manuscript available from J. J. Ryan. American and Soviet clinical neuropsychology. American Scheer, D. E., &Lubin, B. (1980). Surveyoftrainingprogramsin Psychologist, 32, 959-968. clinical neuropsychology. Journal of Clinical Psychology, Mahoney, M. J. (1985). Open exchange and epistemic progress. 36, 1035-1040. American Psychologist, 40, 29-39. Seretny, M. C., Dean, R. S., Gray, J. W., & Hartlage, L. C. McCaffrey, R. J. (1985). Educational backgrounds of clinical neu(1986). The practice of clinical neuropsychology in the ropsychologists in APA-approved internship sites. ProfesUnited States. Archives of Clinical Neuropsychology, 1, sional Psychology: Research and Practice, 16, 773-780. 5-12. McCaffrey, R. J., & Isaac, W. L. (1984). Survey of the educa- Shimberg, B. (1981). Testing for licensure and certification. tional backgrounds and specialty of instructors of clinical American Psychologist, 36, 1138-1146. neuropsychology in APA approved graduate training proShryock, R. H. (1967). Medical licensing in America, 1650grams. Professional Psychology: Research and Practice, 15, 1965. Baltimore: Johns Hopkins Press. Sladen, B. J., Mozdzierz, G. J., & Greenblatt, R. C. (1986). 26-33. Geographic distribution of neuropsychological service proMeier, M. J. (1982). Education for competency assurance in human neuropsychology: Antecedents, models and direcviders and its correlates. Professional Psychology: Research and Practice, 17, 256-259. tions. In S. B. Filskov & T. I. Boll (Eels.), Handbook of clinical neuropsychology. New York: Wiley. Smith, A. (1979). Practices and principles of clinical neuropsyMeier, M. I. (1987). Continuing education: An alternative torechology. International Journal of Neuroscience, 9, 233238. specialization in clinical neuropsychology. The Clinical Stapp, J., Tucker, A.M., & VandenBos, G. R. (1985). Census of Neuropsychologist, 1, 9-20. psychological personnel: 1983. American Psychologist, 40, Molloy, M. (1987). On starting a private practice as a clinical neuropsychologist. The Clinical Neuropsychologist, 1, 1317-1351. VandenBos, G. R., & Stapp, J. (1983). Service providers in psy61-66. chology. American Psychologist, 38, 1346-1352. North Carolina Court of Appeals. (1986). Home versus Goodson VandenBos, G., Stapp, I., & Kilberg, R. R. (1981). Health serdecision. (Unpublished manuscript available from APA, vice providers in psychology. American Psychologist, 36, NCPA, or Antonio E. Puente) 1395-1418. Olson, P. A. (1983). Credentialism as monopoly, class war, and Wedding, D., Franzen, M. D., & Hartlage, L. C. (1987). socialization scheme. Law and Human Behavior, 7, 291Milestones, assessment models, and emerging issues in 299. clinical neuropsychology. Bulletin of the National Academy Reitan, R. M. (1955). Certain differential effects of left and right ofNeuropsychologists, 4, 6-12. cerebral lesions in human adults. Journal ofComparative and Wedding, D., Horton, A.M., & Webster, J. S. (1986). The neuroPhysiological Psychology, 48, 474-477. psychology handbook. Berlin: Springer. Ryan, I. J., Farage, C. M., & Lips, 0. J. (1983). National disYoung, H. H. (1982). A brief history of quality assurance and peer tribution of neuropsychological service providers. American review. Professional Psychology, 13, 9-13. Psychologist, 38, 859-861.
2 Development of the Child's Brain and Behavior BRYAN KOLB
AND
BRYAN FANTIE
Introduction Perhaps the central issue in neuropsychology over the past l 00 years has been the question of how psychological functions are represented in the brain. At the tum of the century, the debate was largely whether or not functions were actually localized in the cortex. Although today this is no longer a subject of major discussion, the general problem of determining what is localized in the cortex remains. One way in which to examine this issue is to look at the way in which function and structure emerge in the developing child. This will be the goal of this chapter. The development of structure-function relationships can be examined in three basic ways. First, we can look at the structural development of the nervous system and correlate it with the emergence of specific behaviors. At first blush this approach seems ideal, as the development of both the nervous system and behavior is orderly and consistent across individuals. Unfortunately, it is not as simple as it appears. The nervous system matures in a relatively unremitting way, unfolding to the dictates of time. Behavioral change, on the other hand, is often more highly dependent upon environmental factors. Thus, the degree of damage caused by sensory deprivation is largely determined by when it occurs during an animal's life (Hubel & Wiesel, 1970). In contrast, whether or not someone can ice-skate will be more easily predicted when one knows if the person was raised in Canada or Brazil. In addition, age-related neural changes are seldom immediately observable in
BRYAN KOLB AND BRYAN FANTIE • Department of Psychology, University of Lethbridge, Lethbridge, Albena TIK 3M4, Canada.
vivo so it is extraordinarily difficult to directly correlate structural and functional variables. Further, hypotheses regarding brain development are hard to verify, especially because the human nervous system cannot be manipulated during development. Nevertheless, despite these impediments, this approach is still possible. The second way to examine morphological and psychological development is to scrutinize behavior and then make inferences about neural maturation. For example, we might carefully study the emergence of distinct cognitive stages, as Piaget (1952) and his followers have done, and then predict what alterations must have occurred in the nervous system to account for the behavioral change. This approach has not been widely used, largely because psychologists most interested in human development have not been very interested in brain function and many behaviors considered important to child development may not be related directly to neural growth. Nevertheless, this approach is promising and has been pursued actively by Gibson (1977). There is a tendency to emphasize school-related skills as the most important for study in child neuropsychology. This is reasonable because many types of childhood learning disorders are likely related to abnormalities in neural development. It must be remembered, however, that the human brain did not evolve in a schoolroom. Thus, the basic functions that are related to neural development may not be found easily by studying scholastic behaviors such as reading.' Rather, the neural mechanisms underlying reading ability may best be understood by examining fundamental visuospatial or visuomotor skills. The third way to study structure-function relationships is to relate brain malfunction to behavioral disorders. This method, which is prevalent in research dealing with adults, is difficult to apply to the 17
18
CHAPTER2
developing brain. The major problem is that the function of a specific neural area may change over time. For instance, Goldman (1974) found that although juvenile rhesus monkeys that had sustained frontal cortex lesions in infancy could solve tasks sensitive to frontal-lobe damage in adults, they subsequently lost this ability as they matured. This result can be interpreted as showing that some other structure, probably the striatum, initially controlled the behaviors necessary for the successful performance of the tasks. Through the natural course of development, this function is eventually transferred to the frontal cortex as the original structure takes some other role in the production of behavior. Because, in this case, the frontal cortex was damaged, it was unable to assume the function when required and the task could not be fulfilled. Therefore, because the association of functions and brain sites that are applicable at one age may be inappropriate at other ages, there is not just one form of the immature brain. The plasticity of the immature brain poses another problem to inferring structure-function relations from malfunction in the developing nervous system. Brain damage occurring in infants may produce very different behavioral effects than in adults because early injury has also altered fundamental brain organization. For example, Rasmussen and Milner (1977) showed that if neonatal speech zones, usually found in the left cerebral hemisphere, are damaged, language may develop in the right cerebral hemisphere. Similar damage at 5 years of age may cause the speech zones to move within the left hemisphere. In both cases, language would then occupy space normally serving other functions. The chronic behavioral loss would manifest itself in some other cognitive function, such as spatial orientation, even though the damage can be shown to have been in the cortical site that normally subserves language functions. Identical lesions could result in very different deficits depending upon the age at which the damage occurred. Such effects do not occur in the adult. We point out the pitfalls in developmental neuropsychology not to discourage the study of the child's brain, but to caution that what follows in this chapter must be considered in the light of these problems. We shall summarize research on neocortical development using each of the three approaches outlined above. We begin by considering the anatomical development of the cerebral cortex. We then consider functional development and try to draw correlations between the emergence of particular behaviors and neural development. Finally, we examine factors affecting brain development.
Anatomical Development of the Child's Brain The process of brain growth can be understood by considering the composition of the nervous system. The cortex is a laminated structure of approximately six layers made up of neurons and glial cells. Some of the glial cells that, in the brain, are called oligodendrocytes, insulate certain portions of many neurons by wrapping around them. Other glial cells, mainly astrocytes and microcytes, are believed to perform basic maintenance and support functions for neighboring neurons. Neurons receive input from other neurons across tiny spaces known as synaptic gaps through processes called dendrites while sending output to other neurons via processes called axons. Cortical neurons exchange information with other cortical neurons as well as neurons located in subcortical structures. Additionally, the many projections each neuron usually receives from other neurons often use different chemical substances to transmit information. Basically, these chemicals excite or inhibit the activity of the target cell and it is the net total of these influences that determines whether or not the neuron fires. The successful development of the brain into a properly functioning, integrated organ requires that each component first be formed and then be correctly interrelated with the others. We will consider each of these developmental processes in tum.
Neural Generation The human brain follows a general pattern of development, beginning as a neural tube and gradually acquiring the features of the adult brain (illustrated in Figure 1), that is typical of all mammals. The basic neural tube surrounds a single ventricle where cells are generated along the ventricular wall and then migrate out to their proper location. In humans, approximately 109 cells are required to eventually form the mature neocortex of a single cerebral hemisphere (Rakic, 1975). During development, the cortex is composed of four embryonic regions: the ventricular, marginal, intermediate, and subventricular zones (as illustrated in Figure 2). These zones are transient features uniquely related to early development for each either disappears or becomes transformed so that they are no longer identifiable in the adult nervous system. Sidman and Rakic (1973) combined the extensive studies of Poliakov (1949, 1961, 1965) with
.-- _. -~--- -
/ ·-
;{---F -, ·'-f .. ---- . ( · ·--
Q
DEVELOPMENT OF THE CHILD'S BRAIN AND BEHAVIOR
:;..-----.,,,
_r
19
·· ;--:co:-:."\,
~~) ~l~~~ /
J J " '· ~ \
25 DAYS
35 DAYS
,.
. .
.!
· . ··~_/
40 DAYS
.·
50 DAYS
.
:,
100 DAYS
.
.
.
FIVE MONTHS
EIGHT MONTHS
FIGURE 1. Prenatal development of the human brain showing a series of embryonic and fetal stages (adapted from Cowan, 1979).
their own observations to produce a summary of the timing and phases of cortical development in humans. They suggested that there are roughly five developmental stages in neuronal development, which are summarized in Figure 2. There is some disagreement over how long cells destined for the
cortex divide and migrate in the human, but cortical ceU proliferation appears to be complete by the middle of gestation; although, at this stage, the cortex by no means appears like that of an adult. Cell migration may still proceed for some ·months after this time, possibly continuing postnatally, and the cortical lam-
20
CHAPTER2
STAGE 1Z:
STAGElSl:
STAGE
m
STAGE II
Mg
v <6wks
6·8wks
8-IOwks
10-11 wks
ll-13wks
13-15wkl
>16wks
FIGURE 2. Schematic illustration of events occurring sequentially during development of the cerebral cortex (from Sidman & Rakic, 1973). Stage 1: initial formation of the cortical plate. Stage II: primary condensation of the cortical plate. Stage III: bilaminate cortical plate. Stage IV: secondary condensation. StageY: Conical maturation. CP, cortical plate; Im, intermediate zone; l.lm and O.lm, inner and outer intermediate zones, respectively; Mg, marginal zone; SGL, subpial granular layer; SY, subventricular zone; V, ventricular zone. Age is in fetal weeks. (See Sidman & Rakic, 1973, for more details.)
ination continues to develop and ditlerentiate until after birth. Marin-Padilla (1970) studied the sequential lamination of the human motor cortex in ontogenesis and found that by the fifth embryonic month, cortical layers V and VI are visible, although not yet completely mature. Over the ensuring months the remaining layers develop (as summarized in Table l). Marin-Padilla's results illustrate two principles of cortical development. First, cortical neurons develop in an inside-out pattern. The neurons of layer VI, that stratum ultimately destined to lie farthest from the external cortical surface, migrate to their locations first, foiJowed by those destined for layer V, and so on. Thus, we see that successive waves of neurons pass earlier-arriving neurons to assume progressively
more superficial positions. Second, the cortex overproduces neurons, which are later lost through normal cell death. Layer IV in the motor cortex is a particularly clear example of this because cells that are visible there in the seventh month and at birth, later degenerate, leaving an agranular layer. As might be predicted, the precise timing of the development and migration of cells to different cytoarchitectonic regions varies with the particular area in question. For example, Rakic (1976) showed that while the ventricular zone is producing layer IV cells for area 17, the neighboring ventricular zone is generating layer III cells that will migrate to area 18. Thus, at any given moment during cortical ontogenesis, cells migrating from the ventricular zone are destined for different regions and layers of the
DEVELOPMENT OF THE CHILD'S BRAIN AND BEHAVIOR
21
TABLE 1. Sequential Lamination of the Human Motor Cortex in Ontogenesisa.b Cortical layers
III Case 5-Month fetus 7-Month fetus 7.5-Month fetus Newborn infant 2.5-Month-old infant 8-Month-old infant
II
++ ++-+++ +++-++++ ++++ ++++ ++++
0
+ ++ +++ +++ +++
Upper 0
+ ++ +++ ++++ ++++
Lower
IV
0
0
+-++ ++-+++ ++++ ++++ ++++
+ ++ ++++ Very thin Agranular
v
VI
+ ++ +++ ++++ ++++ ++++
+ ++ +++ ++++ ++++ ++++
"From Marin-Padilla (1970). •Key: 0, unrecognizable; +, immature; + +, developing; + + +, established; + + + +, fully developed.
cortex. One implication of this phenomenon is that events that might affect the fetus during cortical development, like the presence of a toxic agent such as alcohol, will affect different cytoarchitectonic zones differently. Furthermore, because specific populations of cells are migrating at different times to any given cortical laminae, it implies that toxic agents, or other environmental events, could perturb the development of a specific population of cells to a particular cytoarchitectonic area.
Cell Migration Because cortical cells are born distal to the cortical plate and must migrate there, one can ask how this occurs, particularly as cells traveling to the outer layers must traverse the cells and fibers of the inner layers. In a series of elegant studies, Rakic (1972, 1975, 1981, 1984) showed that neurons migrate to the appropriate laminae within the cortex along specialized filaments, known as radial glial fibers, which span the fetal cerebral wall at early ages. These radial glial cells originate in the ventricular zone and extend outward to the cortical plate. As the cortex develops, thickens, and sulci begin to appear, the radial glial fibers stretch and curve, guiding the migrating neurons to their correct location (see Figure 3).
Axonal Development As cells migrate along the radial glial fibers, they begin to develop axons that run to subcortical areas, other cortical areas, or across the midline as commissural fibers. The rate of axon development is extremely rapid, apparently on the order of I
mm/day. In addition to axons of cortical cells growing out, axons from the thalamus enter the cortex after the principal cortical target cells complete their migrations and assume the appropriate positions within the developing cortical plate (Rakic, 1976).
Dendritic Development Two processes occur during development of the dendrite: dendritic arborization and spine growth. The dendrites begin as individual processes protruding from the cell body. Later, they develop increasingly complex extensions, looking much like the branches of trees in winter. Spines are little appendages, resembling thorns on a rose stem, that begin to appear in the seventh intrauterine month (Poliakov, 1961). Before birth they are observed only on the biggest neurons (mainly those found in layer V). After birth, they can also be found on other neurons where they spread and densely cover the dendritic surface. Although dendritic development begins prenatally in the human, it continues for a long time postnatally. In laboratory animals, the development of both dendritic branches and spines has been shown to be influenced dramatically by environmental stimulation (Greenough, 1976), a phenomenon that is probably very important in relation to the human child's development. In contrast to the development of axons, dendritic growth usually commences after the cell reaches its final position in the cortex and proceeds at a relatively slow rate, on the order of micrometers per day. The disparate developmental rates of axons and dendrites are important because the faster-growing axon can contact its target cell before the dendritic processes of that cell are elaborated, suggesting that the axon may play a role in dendritic differentiation
22
CHAPTER2
p
CP
CP ON
A
B
FIGURE 3. Schematic representations of the pattern of neural migration along the radial glial cells (from Rakic, 1981 ). (A) Migrating neurons leave the ventricular (V) and subventricular (SV) zones and travel to more superficial layers of the cortical plate (CP). En route they pass through the deeper neurons (DN), which are already in place. (B) Portion of the fetal cortex showing the radial glial fibers. (C) Section of the corresponding region in an older fetal cortex showing the changes in the radial glial fibers that allow the formation of gyri.
(Berry, 1982). The morphological changes associated with dendritic growth in the frontal cortex are illustrated in Figure 4.
Synaptic Development The mechanism that controls synapse formation is one of the major mysteries of developmental neurobiology, largely because synapses are perceptible only by electron microscopy, which does not allow direct observation of their sequence of development in living tissue. The onset of synaptogenesis is abrupt and the appearance of synapses in any particular area is remarkably rapid although neurons may be juxtaposed for days before they actually make synaptic connections. Synapses usually form between the axon of one neuron and the dendrites, cell body, axons, or established synapses of other cells. Because synaptogenesis begins before neurogenesis is complete, neurons migrating to the superficial layers
of the cortex must bypass cortical neurons upon which synapses have already formed or are in the process of forming. Little is known of the details of synapse formation in humans but simple synaptic contacts have been observed during the fifth gestational month and there is extensive postnatal synaptic proliferation. Indeed, in the frontal lobes, synaptic density continues to increase until about 2 years of age, where it is approximately 50% above the adult mean, then decreases untill6 years of age (Huttenlocher, 1979). It is interesting that the synaptic density of infants appears to exceed that of adults as it has generally been assumed that a larger number, or a greater density, of synapses implies a higher functional capacity. Evidence of decreasing synaptic density coincident with increasing cognitive skill is thus intriguing, especially because high numbers of synapses have been found in certain cases of mental retardation (Cragg, 1975). It is not surprising that intellectual ability can-
DEVELOPMENT OF THE CHILD'S BRAIN AND BEHAVIOR
23
Myelin Development
Ill
IV
v
Myelination is the process by which the glial cells of the nervous system begin to surround axons and provide them with insulation. Although nerves can become functional before they are myelinated, it was assumed in the 1920s and 1930s that they only reach adult functional levels after myelination is complete (Flechsig, 1920). This notion now appears to be an oversimplification but is, nonetheless, useful as a rough index of cerebral maturation. In contrast to other aspects of cortical development, myelin appears late, at a time when cellular proliferation and migration are virtually complete. The primary sensory and motor areas begin to myelinate just before term whereas the frontal and parietal association areas, the last to myelinate, begin postnatally and continue until about age 15 years or, sometimes, even later. Because different regions of the cortex myelinate at different times, and myelination begins in the lower layers of each cortical area and gradually spreads upward, the upper layers of the motor and primary sensory areas are myelinating at the same time that the lower areas of some association areas are just beginning to myelinate.
Neurochemical Development
FIGURE 4. Postnatal development of human cerebral cortex around Broca's area as taken from camera Iucida drawings of Golgi-Cox preparations (from Conel , 1939-1967). (A) Newborn; (8) I month; (C) 3 months; (D) 6 months; (E) 15 months; (F) 24 months.
not be predicted merely by its relation to the quantity of some anatomical feature, such as synapses, and, perhaps, the process involved in reducing synaptic density represents some sort of qualitative refinement.
Glial Development The differentiation and growth of neurons, which are generaiJy produced before their associated glia, appear to play some role in stimulating the growth and proliferation of glial cells, but the mechanisms are unknown (Jacobsen, 1978). In contrast to neurons, glial cells continue to proliferate after birth and may continue to do so throughout life.
Chemical neurotransmitters serve as the primary means of interneuronal communication, yet virtually nothing is known about the neurochemical development of the human cortex. Although there are numerous studies of neurotransmitter development in the rat, knowledge about the relationships among transmitters in the adult neocortex is still limited and the most completely described neurochemical systems make only a modest contribution to the overall synaptic activity of the neocortex (see Table 2). There are, however, some recent developmental studies using nonhuman primates that are worth reviewing as the human brain is likely to be similar. Goldman-Rakic and Brown ( 1981, 1982) investigated the regional distribution of catecholamines in rhesus monkeys ranging in age from newborns to young adults. Their overall findings were that, although monoaminergic systems are present in the cortex at birth, these networks continue to develop for years. Catecholamine development varies greatly between different cortical regions and the most striking postnatal increases in content were observed in the frontal and parietal association areas. Perhaps most interesting was their observation that catecholamine development (especialJy that of the mono-
24
CHAPTER2
TABLE 2. Neocortical Neurotransmitters" Transmitter type Afferents Norepinephrine Dopamine Serotonin Acetylcholine Intrinsic GABA Vasoactive intestinal peptide Cholecystokinin Efferents Glutamate
Cell location
Locus coeruleus Substantia nigra A 10 Raphe Globus pallidus magnocellular Aspinous stellate (all layers) Aspinous bipolar (II, III, IV) Aspinous bipolar and aspinous stellates Pyramidal cells (layer V corticostriatal)
"After Coyle (1982).
amines) parallels functional development in the prefrontal cortex over the first 2-3 years of life. These data support the suggestion that catecholamines may play an important role in the development of functional activity in the frontal cortex, and likely affect the morphological development of various neuronal processes such as dendritic fields.
Postnatal Brain Development After birth, the brain does not grow uniformly but rather tends to increase its mass during irregular periods commonly called growth spurts. In his analysis of brain/body weight ratios, Epstein (1978, 1979) found consistent spurts in brain growth at 3-10 months, accounting for an increase of 30% in brain weight by the age of I i years, as well as between ages 2-4, 6-8, 10-12, and 14-16+ years. The increments in brain weight were about 5-10% over each 2-year period. This expansion takes place without a concurrent increase in neuronal proliferation and is unlikely to be accounted for by increases in the number of glial cells. Rather, it is most likely due to the growth of dendritic processes and myelination. Such an increase in cortical complexity would be expected to correlate with increased complexity in behavioral functions, and it could be predicted that there would be significant, and perhaps qualitative, changes in cognitive function during each growth spurt. It may be significant that the first four brain growth stages coincide with the classically given ages of onset of the foqr main stages of intelligence development described by Piaget. We return to this later.
Cortical Function at Birth The extreme paucity of behavioral skill in the newborn leads to the notion that, shortly after birth, the cortex has not yet begun to function. Thus, the cortically injured infant was once believed to be indistinguishable from the normal child at birth (Peiper, 1963). Several lines of evidence suggest that the cortex is indeed functioning, if only at a rudimentary level. It is now known that cortically hemiplegic infants can be distinguished from normal babies on the basis of muscle tone (Gibson, 1977) and cortically damaged infants may also have abnormal sleep-waking cycles and abnormal cries (Robinson, 1966). There are also several measures of electrical activity that imply cortical activity is present at birth. EEG activity can be recorded from the fetal brain (Bergstrom, 1969) and epileptic seizures of cortical origin can occur in the neonate (Caveness, 1969). Perhaps the most compelling evidence of early cortical activity comes from the extensive work of Purpura (Purpura, 1976, 1982). In his study of cortical activity in premature human infants, Purpura took advantage of the fact that, between 26 and 34 weeks of gestation, cortical pyramidal cells in primary visual cortex undergo significant growth and branching. These changes are associated with corresponding maturational changes in the electrophysiological characteristics of the visual evoked potentials (VEPs) in preterm infants. Although, even at birth, the VEPs are not identical to those of adults, they are present and indicate that at least primary visual cortex is functioning in some capacity. Chugani and Phelps ( 1986) studied glucose utilization in the brain of infants using positron emission
DEVELOPMENT OF THE CHILD'S BRAIN AND BEHAVIOR
tomography. Their results showed that, in infants 5 weeks of age or younger, glucose utilization, which can be taken as a crude measure of neural activity, was highest in the sensorimotor cortex, a result that is in accordance with anatomical evidence that this is the most mature cortical region at birth. By 3 months of age, glucose metabolism had increased in most other cortical regions, with subsequent increases in frontal and posterior association cortex occurring by 8 months. Thus, by about 8-9 months there is evidence of activity throughout the cerebral cortex, although it probably changes in the years to come.
Abnormal Development of the Child's Brain We have seen that the anatomical development of the child's brain consists of the proliferation and migration of cells, the growth of axons and dendrites, synapse formation and loss, myelin growth, and so on. These processes begin early in embryonic development and continue until late adolescence. In view of the complexity of the cortex and its prolonged development, it is reasonable to expect that normal cortical development could be disrupted by any number of events. These include abnormalities in the normal genetic program of neural growth, the influences of exogenous factors such as toxic substances or brain trauma, and nutritional or other environmental circumstances. We do not propose to discuss all of these possibilities, but will confine our discussion to
those events that are most likely to be important to the neuropsychologist: abnormal neural differentiation and early brain damage.
Abnormal Neural Structure In the event that either neurogenesis or neural migration is abnormal, one would expect gross abnormalities in cortical development. Clinically, a variety of conditions are recognized (Table 3) but little is known about the details of cell differentiation in these disorders. The major experimental study of disturbed migration in the cerebral cortex involves the reeler mouse mutant. Caviness (Caviness, 1982; Caviness & Rakic, 1978; Caviness & Sidman, 1973) showed that, in this animal, the cortex is inverted compared to that of a normal mouse: the cells generated first lie nearest the cortical surface and those generated last lie deepest. In addition, many of the pyramidal cells are abnormally oriented, in some cases with their major dendrites (the apical dendrites) oriented downwards rather than upwards as in the normal mouse. Despite their aberrant position, the cells develop connections as they would have had they been normally situated. Caviness and his colleagues studied the cortex of humans with various similar abnormalities, finding some of the same aberrant features (Caviness & Williams, 1979). Thus, in lissencephalic cortex, Williams et al. (1975) found that cells failed to migrate into the appropriate layers and some cells were abnormally oriented, much as in
TABLE 3. Types of Abnormal Development Type
Symptom
Anencephaly Holoprosencephaly Lissencephaly
Absence of cerbral hemispheres, diencephalon, and midbrain Cortex fonns as a single undifferentiated hemisphere The brain fails to fonn sulci and gyri and corresponds to a 12-week embryo Gyri are more numerous, smaller, and more poorly developed than nonnal Gyri are broader and less numerous than nonnal Development of the brain is rudimentary and the person has lowgrade intelligence Symmetrical cavities in the cortex, where cortex and white matter should be Displaced islands of gray matter appear in the ventricular walls or white matter, caused by aborted cell migrdtion Complete or partial absence of the corpus callosum
Micropolygyria Macrogyria Microencephaly Porencephaly Heterotopia Agenesis of the corpus callosum Cerebellar agenesis
25
Portions of the cerebellum, basal ganglia, or spinal cord are absent or malfonned
26
CHAPTER2
the reeler mouse. The cause of these anomalies remains unknown.
Injury and Brain Development If the brain is damaged during development, it is reasonable to suppose that its development might be fundamentally altered. We are unaware of any anatomical studies of human brains with early lesions but there is a considerable literature from work with laboratory animals. In an extensive examination of monkeys with prenatal or perinatal frontal cortex injuries, Goldman-Rakic has shown a variety of changes in cortical development including abnormal gyral formation and abnormal corticostriatal connections (Goldman & Galkin, 1978; Goldman-Rakic, Isseroff, Schwartz, & Bugbee, 1983). Similarly, Kolb and his colleagues have found abnormal corticostriatal and subcorticocortical connections, abnormal myelination, altered cortical catecholamine distribution, thalamic shrinkage, reduced gliosis relative to adult operates, and markedly thinner cortex following early frontal lesions in rats (Kolb, 1987; Kolb & Nonneman, 1978; Kolb & van der Kooy, 1985; Kolb & Whishaw, 1981). The thin cortex appears to result not from a loss in the number of cortical cells, which seems to be normal, but from a loss in dendritic arborization, a result also described by Jones and Thomas (1962). In sum, there is good reason to presume that early damage to the human brain produces significant changes in cortical morphology that extend far beyond the boundaries of the tissue directly traumatized. Also, there is no evidence that the brain-damaged infant (or adult) can show any additional neuronal proliferation to compensate for lost neurons.
Behavioral Correlates of Brain Development Two types of behavior have been extensively studied and correlated with anatomical development, namely motor behavior and language. We shall consider each separately, and then consider the development of their asymmetrical representation in the cortex. Finally, we will discuss the behavior of children on standardized tests typically used by clinical neuropsychologists. We shall not attempt to be exhaustive in our coverage of each, but rather try to give a flavor of the findings to date.
Motor Systems The development of locomotion in human infants is quite familiar to most of us. Infants are, at first, unable to move about independently, but eventually they learn to crawl and then to walk. The waY. in which other motor patterns develop is less obvious but one has been described in an elegant study by Twitchell (1965) who documented the st~g_es an infant passes through while acquiring the ab1hty ~o reach out wi!h one limb and bring objects toward Itself. Before b1rth the fetus's movements are ~ssentially of the whole body. Shortly after birth the mfant can flex all the joints of an arm in such a way !hat it could scoop something toward its body, but it IS not clear that this movement is executed independent of other body movements. Between 1 and 3 m~nths it orients its hand toward, and gropes for, objects that have contacted it. Between 8 and 11 ~onths it develops the "pincer grasp," using the mdex finger and thumb in opposition to each other. J?e. development of. the pincer grasp is extremely s1gmficant, because It allows the infant to make a very precise grasping movement that enables the manipulation of small objects. In summary, there is a sequential development of the grasping reaction: first scooping, then reaching and grasping with all fingers, then independent finger movements. The fact that motor cortex lesions in adults abolish the grasp reaction with independent finger movements implies that there could be anatomical changes within the motor strip that correlate with the original development of the behavior. Although there are probably multiple changes occurring, especially in the development of dendritic arborizations a correlation has been noted between myelin fo~ation and the ability to grasp. In particular, the small motor fibers become myelinated at about the same time that reaching and grasping with the whole hand develop while the giant Betz cells of the motor cortex become myelinated at about the time the pincer grasp develops. These different motor fibers are thought to control arm and finger movements, respectively (Kolb & Whishaw, l985b). The correlation between myelin development and motor behaviors can also be found in many other activities. Table 4 summarizes the development of a variety of behavioral patterns and myelin formation. It is of course difficult to be certain which correlations are meaningful and, as we have noted, there are obviously many other anatomical changes occurring concurrently. Careful study of these data, however, does show some intriguing correlations that warrant more detailed study.
DEVELOPMENT OF THE CHILD'S BRAIN AND BEHAVIOR
27
TABLE 4. Summary of Postnatal Human Developmenta
Age
Visual and motor function
Birth
Reflex sucking, rooting, swallowing, and Moro reflexes; infantile grasping; blinks to light
6 weeks
Extends and turns neck when prone; regards mother's face, follows objects Infantile grasp and suck modified by volition; keeps head above horizontal for long periods; turns to objects presented in visual field; may respond to sound
3 months
6 months
9 months
Grasps objects with both hands, will place weight on forearms or hands when prone; rolls supine to prone; supports almost all weight on legs for very brief periods; sits briefly Sits well and pulls self to sitting position; thumb-forefinger grasp; crawls
12 months
Able to release objects; cruises and walks with one hand held; plantar reflex flexor in 50% of children
24 months
Walks up and down stairs (two feet a step); bends over and picks up objects without falling; turns knob; can partially dress self; plantar reflex flexor in I00% Goes up stairs (one foot a step); pedals tricycle; dresses fully except for shoelaces, belt, and buttons; visual acuity 20120/0U
36 months
Social and intellectual function
Average brain weight (grams a) 350
Smiles when played with
410
Watches own hands
515
Laughs aloud and shows pleasure; primitive articulated sounds, "ga-goo"; smiles at self in mirror
660
Waves bye-bye, plays pany cake, uses "dada," "baba," imitates sounds 2-4 words with meaning; understands several proper nouns; may kiss on request
750
Two- or three- word sen-
925
1065
tences; uses ''I,''
"me," and "you" correctly; plays simple games; points to 4-5 body parts; obeys simpie commands Numerous questions; knows nursery rhymes; copies circle; plays with others
1140
Degree of myelinationh Motor roots + + +; sensory roots + +; medial lemniscus + +; superior cerebellar peduncle + +; optic tract + +; optic radiation ± Optic tract + +; optic radiation +; middle cerebral peduncle ±; pyramidal tract + Sensory roots + + +; optic tract and radiation + + +; pyramidal tract + +; cingulum +; frontopontine tract + ; middle cerebellar peduncle +; corpus callosum ±; reticular formation ± Medial lemniscus + + +; superior cerebellar peduncle + + +; middle cerebellar peduncle + +; pyramidal tract + +; corpus callosum +; reticular formation + ; associational areas ±; acoustic radiation + Cingulum + + +; fornix + + ; others as previously given Medial leniscus + + +; pyramidal tracts + + +; frontopontine tract +++;fornix+++; corpus callosum +; intracortical neuropil ± ; associational areas ± ; acoustic radiation + + Acoustic radiation + + +; corpus callosum + +; associational areas +; nonspecific thalmic radiation ++
Middle cerebellar peduncle +++
(continued)
28
CHAPTER2
TABLE 4. (Continued)
Age
Visual and motor function
5 years
Skips; ties shoelaces; copies triangle; gives age correctly
Social and intellectual function Repeats 4 colors
digi~s;
Adult
names 4
Average brain weight (grams 0 ) 1240
1400
Degree of myelinationb Nonspecific thalmic radiation + + +; reticular formation + +; corpus callosum + + +; intracortical neuropil and associational areas + + Intracortical neuropil and associational areas + + to+++
Source: Spreen, Tupper, Risser, Tuokko, & Edgell (1984). bfrom Yakovlev and Lecours (1967). Estimates are made from their graphic data(::!:, minimal amounts; +,mild;++, moderate; +++,heavy).
0
Language Development Tbe onset of speech consists of a gradual appearance of generally well-circumscribed events that take place between the second and third years of life. According to Lenneberg (1967), certain important speech milestones are reached in a fixed sequence and at relatively constant chronological ages (as summarized in Table 5). Although there is a general parallel between language development and the development of motor capacities, Lenneberg emphasized that language development is independent of motor coordination. In particular, it appears that the precise movements of the Eps and tongue needed for speech are fully developed well before the acquisition of finger and hand control. Furthermore, once children are capable of correctly pronouncing a few words, it can be presumed that there is sufficient motor skill to articulate many more words and yet expansion of the child's vocabulary is a very slow process. Thus, language development appears to depend upon factors other than simple mechanical skill. This argument is further underscored by the observation that for a small proportion of children (less than I%), who have normal intelligence and normal skeletal and motor development, speech acquisition is markedly delayed. For example, such children may not begin to speak in phrases until after age 4, in spite of an apparently normal environment and the absence of any obvious neurological signs that might suggest brain damage. It could be argued that language development is not dependent upon the maturation of some neural structure but upon some other factor such as environmental stimulation. Although this is possible, the likelihood that speech development is constrained
exclusively by some environmental event is unlikely because the timing of the onset of speech appears to be universal in humans (Lenneberg, 1967). In addition, it can be shown that an enormous variety of environmental conditions fail to affect the age of onset of many of the developmental speech milestones (summarized in Table 5). Thus, the emergence of speech and language_ habits is most easily accounted for by assuming that there are maturational changes within the brain. The difftculty is in specifying what these changes might be. Indeed, in view of the complexity of the neural control of language, it is futile to look for any specific growth process that might explain language acquisition. Nonetheless, it is likely to be instructive to know in what ways the cortex is different before the onset of language (age 2) and after the majority of language acquisition is completed (about age 12). It will be recalled from our discussion of neural maturation that by 2 years of age there is no longer any cell division and most cells have migrated to their final location in the cortical laminae. The major changes that occur between the ages of 2 and 12 years are in the interconnection of neurons, largely through a decrease in the number of synapses as well as increase in the complexity of their dendritic arborizations. If one assumes that language acquisition requires the development of functional connections between neurons, much as hypothesized by Hebb (1949) in his concept of cell assemblies, then these changes in synaptic density and dendritic detail may be logical candidates as constraints of speech development. The postnatal changes in dendritic complexity within the speech areas are among the most impressive in the brain. As illustrated in Figure 4, the dendrites are simple at birth and develop slowly until
DEVELOPMENT OF THE CHILD'S BRAIN AND BEHAVIOR
TABLE 5. Developmental Milestones in Motor and Language Development At the completion of:
12 weeks
16 weeks
20 weeks
6 months
8 months
10 months
12 months
18 months
24 months
30 months
3 years
4 years
VocaJi~tion and language Markedly less crying than at 8 weeks; when talked to and nodded at, smiles, followed by squealing-gurgling sounds usually called cooing, which is vowel-like in character and pitch-modulated; sustains cooing for 15-20 seconds Responds to human sounds more definitely; turns head; eyes seem to search for speaker; occasionally some chuckling sounds The vowel-like cooing sounds begin to be interspersed with more consonantal sounds; labial fricatives, spirants, and nasals are common; acoustically, all vocalizations are very different from the sounds of the mature language of the environment Cooing changing into babbling resembling one-syllable utterances; neither vowels nor consonants have very fixed recurrences; most common utterances sound somewhat like rna, mu, da, or di Reduplication (or more continuous repetitions) becomes frequent; intonation patterns become distinct; utterances can signal emphasis and emotions Vocalizations are mixed with sound-play such as gurgling or bubble-blowing; appears to wish to imitate sounds, but the imitations are never quite successful; beginning to differentiate between words heard by making differential adjustment Identical sound sequences are replicated with higher relative frequency of occurrence and words (mamma or dadda) are emerging; definite signs of understanding some words and simple commands (show me your eyes) Has a definite repertoire of words-more than 3 but less than 50; still much babbling but now of several syllables with intricate intonation pattern; no attempt at communicating information and no frustration for not being understood; words may include items such as thank you or come here, but there is little ability to join any of the lexical items into spontaneous two-item phrases; understanding is progressing rapidly Vocabulary of more than 50 items (some children seem to be able to name everything in environment); begins spontaneously to join vocabulary items into two-word phrases; all phrases appear to be own creations; definite increase in communicative behavior and interest in language Fastest increase in vocabulary with many new additions every day; no babbling at all; utterances have communicative intent; frustrated if not understood by adults; utterances consist of at least two words, many have three or even five words; sentences and phrases have characteristic child grammar, that is, they are rarely verbatim repetitions of an adult utterance; intelligibility is not very good yet, though there is great variation among children; seems to understand everything that is said to him Vocabulary of some 1000 words; about 80% of utterances are intelligible even to strangers; grammatical complexity of utterances is roughly that of colloquial adult language, although mistakes still occur Language is well established; deviations from the adult norm tend to be more in style than in grammar
29
30
CHAPTER2
about 15 months when the major dendrites are pre- designed to demonstrate lateralization of function has sent. Between 15 and 24 months there is a dramatic emphasized the age at which asymmetry first apincrease in the density of the neuropil. A similar pears. Table 6 gives examples of a number of repreobservation can be made from examination of the sentative functions and the side and earliest age of cortex of the posterior speech zone. Given the cor- demonstrated asymmetry. A central theoretical issue relation between language development and matura- is whether or not functions are disproportionately tion of the language areas, we can infer that language represented in the two hemispheres because they dedevelopment may be constrained, at least in part, by pend on certain anatomical asymmetries that develop the maturation of these areas and that individual dif- independent of environmental stimulation. The fact ferences in language acquisition may be accounted that anatomical asymmetries can be observed in the for by differences in this neural development. Fur- cortex prenatally (Chi, Dooling, & Gilles, 1977; thermore, given the known effect of environmental Wada, Clarke, & Hamm, 1975) and, therefore, exist stimulation on dendritic development, we might also before the expression of the behaviors, implies that predict that those differences in language acquisition asymmetry is relatively innate. Nevertheless, several that have some environmental influence may do so by rna' or problems arise when we try to correlate funcchanging the maturational rate of the dendritic fields tional and anatomical asymmetry. First, the funcwithin these areas. tions that are most lateralized in adults are not easily assessed in children. For example, it is extremely difficult, if not impossible, to determine handedness for writing in infants, unless, of course, one is willing Cerebral Asymmetry to assume that some other indirect measure, such as Just as the asymmetrical function of the adult's hand strength, in this case, will serve as a reliable brain has been a focal point for neurological study, predictor. Second, correlations between function and the development of asymmetry has been a focal point anatomical asymmetry in adults are far from perfect. of developmental studies. As asymmetry is the sub- Although the left planum temporale is thought to be ject of another chapter in this volume (see Kins- the posterior substrate of language functions, it is boume), we shall consider this topic only briefly. larger in only about 70% of right-handed people, Most of the research with children that has been whereas speech is lateralized to the left hemisphere in TABLE 6. Studies Showing Age of Asymmetry for Different Behaviors System Auditory Speech syllables Music Phonemes Words Environmental sounds Visual Rhythmic visual stimuli Face recognition
Somatosensory Dichhaptic recognition Motor Stepping Head turning Grasp duration Finger tapping Strength Gesturing Head orientation
Age
Dominance
Reference
Pre term 22-140 days 22-140 days 4 years 5-8 years
Right ear Left ear Right ear Right ear Left ear
Molfese & Molfese (1980) Entus ( 1977) Entus (1977) Kimura (1963) Knox & Kimura (1970)
Newborn
Right
7-9 years 6-13 years 9-10 years
Left field Left field None
Crowell, Jones, Kapuniai, & Nakagawa (1973) Marcel & Rajan ( 1975) Witelson (1977) Diamond & Carey ( 1977)
All ages
Left
Witelson (1977)
<3 months Neonates 1-4 months 3-5 years 3-5 years 3-5 years Neonates
Right Right Right Right Right Right Right
Peters & Petrie (1979) Turkewitz (1977) Caplan & Kinsbourne (1976) Ingram (1975) Ingram (1975) Ingram ( 1975) Michel (1981)
DEVELOPMENT OF THE CHILD'S BRAIN AND BEHAVIOR
about 99% of right-banders. What then does a similar anatomical asymmetry in the fetal brain imply?
Development of Problem-Solving Ability As each cortical layer within an area develops, it interacts with and modifies the function of the existing structure. Gibson (1977), therefore, suggested that behavior patterns would be expected to emerge exactly in the manner described by Piaget (1952): Behavior patterns characteristic of different stages do not succeed each other in a linear way (those of a given stage disappearing at the time when those of the following one take form) but in the manner of the layers of a pyramid (upright and upside down), the new behavior patterns simply being added to the old ones to complete, correct or combine with them. (p. 329)
Thus, for example, because the deepest layers of the cortex myelinate first, and these are the efferent or output layers, one would expect to observe motor responses preceding the development of perceptual capacity. Indeed, according to Piaget, motor actions must come first, as motor actions provide data from which to build perceptions. The question to consider is just how well the stages of cognitive development coincide with changes in neural maturation. This is a difficult question that has not been extensively studied. Nevertheless, there is at least suggestive evidence that there may be a significant relationship between cortical development and the classical Piagetian stages. (We note that the Piagetian stages of cognitive development are a source of some debate and there are several other conceptual schemes to describe the development of cognition in children (Carey, 1984). We will restrict our discussion to Piaget, however, because we wish merely to demonstrate the type of study that can be done and because we are unaware of any attempt to correlate other schemes of cognitive development to cortical maturation.) Piaget was a biologist by training and considered the acquisition of knowledge and thought to be closely related to brain function. He believed that cognitive development was a continual process and that the child's strategies for exploring the world were constantly changing. These changes were not simply a result of the acquisition of specific pieces of knowledge but rather' at some specifiable points in development, were fundamental changes in the organization of the child's strategies for learning about the world. Piaget identified four major stages of cognitive development: stage I (Sensorimotor; birth to 18 months); stage II (Preoperational or Symbolic; 18
31
months to 7 years); stage III (Concrete Operational; 7 to 11 years); and stage IV (Formal Operational; 11 + years). In stage I the infant learns to differentiate itself from the external world, learns that objects exist when not visible, and gains some appreciation of cause and effect. In stage II, the child begins to represent things with something else, such as a drawing. Stage III is characterized by the child's ability to mentally manipulate concrete ideas such as dimensions of objects and the like. Finally, in stage IV, the child is able to reason in the abstract. Having identified the stages, the challenge for the neuropsychologist is to identify those changes in neural structure that might underlie these apparent qualitative changes in cognitive activity. The first four brain growth stages described earlier coincide with the usual given ages of onset of the four main Piagetian stages (Epstein, 1979). A fifth stage of development, which would correlate with the fifth brain growth stage, was not described by Piaget but has been proposed by Arlin (1975). The concordance of brain growth and Piagetian stage is intriguing but, to date, remains too superficial. We need to know what neural events are contributing to brain growth, and just where they are occurring. Little is known of this in children after 6 years of age but the question remains important to the neuropsychologist seeking to understand the maturation of cortical operations. A detailed hypothetical analysis of stage I has been attempted by Gibson (1977).
Development of Neuropsychological Test Performance Neuropsychologists have developed an amazing array of tests since World War II with which to assess the behavior of patients with cortical injuries (e.g., Lezak, 1983). In principle, it is logical to suppose that if a test is sensitive to restricted cortical lesions in adults, and if a normal child performs poorly on such a test, it could then be inferred that the requisite cortical tissue is not yet functioning normally. This logic is seductive but is not without difficulties. First, the method assumes that tests will be sensitive to focal lesions: Few tests are. Second, a child may perform poorly on a test for many reasons. For example, a child may have difficulty with a verbal test because the speech areas are slow to develop or because he or she has an impoverished environment and has acquired only a limited vocabulary. Furthermore, just because a child does well on a test does not mean that the child's brain is solving the problem in the same manner as the adult brain. Indeed, there are
32
CHAPTER2
examples of tests in which children do well, only to do more poorly the following year, followed later by improvement again. Thus, in their studies of facial recognition in children, Carey, Diamond, and Woods (1980) found that children improved in performance between ages 6 and 10, declined until age 14, and then attained adult levels by age 16. This result can be taken to imply that the younger children were solving the problem in a different manner dian the older children and adults while, presumably, using different cortical tissue. In sum, although there are clear limitations to the inferences that can be made about the development of specific brain regions, we feel that much can be learned using this type of approach. We will illustrate this by focusing on our own studies using tasks that test frontal lobe function and the perception of faces and facial expression.
Frontal Lobe Tests Two tests are especially sensitive to frontal lobe injury, namely the Wisconsin Card Sorting Test and the Chicago Word Fluency Test (Milner, 1964). In the first test the subject is presented with four stimulus cards, bearing designs that differ in color, form, and number of elements. The subject's task is to sort the remaining cards into piles in front of one or another of the stimulus cards. The only help the subject is given is being told whether the choice is correct or incorrect. The test works on this principle: the correct solution is frrst to sort by color; once the subject has figured this out, the correct solution then becomes, without warning, to sort by form. Thus, the subject must now inhibit grouping the cards on the basis of color and shift to form. Once the subject has succeeded at sorting by form, the relevant feature again changes unexpectedly, this time to number of elements. This cycle of color, form, and number is repeated. The subject's score is the number of target categories completed after sorting 128 cards, and the · task is terminated when all the cards have been used or six categories have been completed, whichever comes first. Shifting strategies is particularly difficult for patients with left frontal lobe lesions. In the second test the subject must write as many words as they can beginning with the letter .. S" in 5 minutes. Following this, they must write as many four-letter words beginning with .. C" as possible in 4 minutes and the final score is the total number of words generated. Frontal lobe patients do very poorly on this test. This deficit is not simply a problem of verbal ability, however, as frontal lobe patients perform at normal levels when asked to write the names of as many
objects or animals as they can think of within a fixed time. We note that frontal lobe patients perform normally at many other tests as well. For example, at tests of visual recognition, which are performed poorly by patients with right posterior lesions, frontal lobe patients achieve normal levels of performance. One example is a test of facial closure illustrated in Figure 5. We tested children with these tests and predicted that if the frontal lobes were slow to mature relative to other cortical areas, then children should reach adult levels very late, probably in adolescence on tests of frontal lobe function. In contrast, children should perform at adult levels much sooner on the tests performed normally by patients with frontal lobe lesions. Figure 6 shows that this is indeed the case. Children perform poorly at all tests when very young but improve as they develop. As predicted, performance on tests performed normally by adults with frontal lobe injuries improves more quickly, however, than performance on tests sensitive to frontal lobe injuries.
Face Perception Normal adults have a remarkable capacity to recognize hundreds of faces and to notice and respond to rather subtle changes in facial expression. Patients with right hemisphere lesions, especially posterior lesions, do especially poorly on tasks that require these abilities (Milner, 1980; Kolb & Taylor, 1981). Very young children also have difficulty in recognizing faces (Carey & Diamond, 1977). In view of the apparent right-hemisphere specialization for facial analysis, we studied the ability of children to recognize faces and facial expression. As in the studies using tests of frontal lobe function, we hoped to infer the developmental state of a cortical region, in this case the right hemisphere. Children were given four face-related tests and one verbal test, a test of dichotic listening. The face tests were the Mooney closure test, a split-faces test (Figure 5), a test requiring the matching of similar facial expressions, and a test requiring the identification of the appropriate facial expression for a specific situation. The Mooney closure test simply requires the subjects to identify the gender and approximate age of the hidden face. In the split-faces test (see Kolb, Milner, & Taylor, 1983, for a detailed account), subjects are asked to indicate which of two photographs of faces is most like the original (see Figure 5). The two alternatives are made from combining the left and right halves of the original face with their respective mirror image. Normal adults
DEVELOPMENT OF THE CHILD'S BRAIN AND BEHAVIOR
33
9 \
FIGURE 5. Examples of face perception tests. Left: In this task the subject must choose, for the character whose face is blank, the appropriate expression for the situation depicted by the cartoon, from one of six key photographs. Top righi: Split-faces test. The task is to choose which of the faces in the bottom two composite photographs more closely resembles the face in the photograph on top. The face on the lower right is a composite of the left half of the face above combined with its mirror image. The face on the lower left was constructed in the same manner using the left half of the face depicted above. Bottom right: Mooney closure test. 1be task is to identify the gender and approximate age of the embedded face.
show a bias in their choices, selecting the alternative constructed from two right sides of the original face (i.e., the part falling in the normal left visual field). Patients with right posterior lesions choose at random. In the third test, the subjects are asked to match photographs on the basis of facial expression. They are shown six key photographs, each displaying a distinctive facial expression, and are subsequently asked to match a series of 24 photographs with the
keys on the basis of the emotion that is expressed in each photo (Kolb & Taylor, 1981). In the final test, subjects are shown a cartoon drawing of a situation in which the face of one of the characters is missing (Figure 5). The task is to choose the appropriate expression for the situation from the six key photogmphs used in the previous test. The data of these studies are summarized in Figure 7. Based on the results of the split-faces test, we
34
CHAPTER2
cialized early to analyze faces, or perhaps for complex visual material in general . Children are thus able to identify hidden faces, choose the side of the face in the left visual field, and can match facial expressions quite accurately by about age 5 or 6. In contrast, they have difficulty in identifying what facial expressions go with certain situations. Because it is known that patients with right posterior lesions are especially poor at the first three tests, we might infer that this area of the cortex functions relatively early in development. We have not yet completed our study of adult patients with cortical lesions on the fourth test but we might predict from our developmental study that this test is measuring the function of some other
C/)
w
a:
g, w
'
~ u
,
A
..
10
H
GROUP
60
.. so
C/)
c
~ "'
~:roDI • 6
8
)
e
lit;
~ '9
10
II
12
ll
1411
J:S
ll!o
17
AD
F'L
GROUP
FIGURE 6. Summary of the developmental changes in two tests of frontal lobe function. The open bars differ significantly from the performance of adult control subjects. AD, adult; FL, adults with frontal lobe lesions (frontal lobe lesion data from Milner, 1964). (A) Wisconsin Card Sorting Test. (8) Chicago Word Fluency Test.
c~ " w 10
u: ~ < ::::>
C/)
60
> ..
A
.... u w
suggest that the right hemisphere may be specialized quite early in development for the analysis of faces. Thus, by age 4, children show a significant bias toward the side of the face in the left visual field. This is not simply a matter of the original face being asymmetrical in some way that biases the performance, for even if the negatives are reversed, the result is maintained: the match is made to the side of the face falling in the left visual field. Note, however, that the right hemisphere is not mature for all of the tests at the same time. Performance on the Mooney closure test improves quickly until 5 years of age, and then very slowly improves thereafter. This could reflect a maturation of the basic neural hardware by age 5, with the subsequent improvement being due largely to experience, a conclusion that we favor. The ability to match facial expressions is surprisingly mature by age 6, which is the earliest we have studied, and is consistent with the results of two other tests. Note, however, that 6-year-old children are very bad at selecting the appropriate facial expression for a particular situational context; the major improvement coming at about age 14. The results of the facial recognition study suggest that the right hemisphere of children is spe-
..
~
8
1
l
4
:s
&
1
e
t
10
11
M
.. 10
I~
~ Q.
60
.. ,. 20
B' '
....
..
J¥)
I
..
.z. .. ~
n
GROUP
it
.'
~.
11r ¥·
• • ' 'GROUP
10
II
0>
""
fl
..
0
w
a: a:
10
0 0
.... z
60
0
..
w
a: w
Q.
c
GROUP
" "
,, ...
FIGURE 7. Performance of children on three tests of facial per· ception. For the split-faces test graph the open bars indicate performance that. fails to differ from chance. For the other graphs the open bars indicate performance that differs from adults. AD, adult; FL, adults with frontal lobe lesions; RP, adults with right posterior cortex lesions. Patient data from Kolb, Milner, and Taylor (1983) and Milner (1980). (A) Split-faces test. (B) Mooney closure test. (C) Matching of facial expression with cartoon situations.
DEVELOPMENT OF THE CHILD'S BRAIN AND BEHAVIOR
cortical region, possibly the frontal cortex. If this proves correct, the result would be consistent with the data from other tests of frontal lobe function discussed above.
35
abled children and the presence of left posterior abnormalities.
Early Brain Injury and Behavior
There is little doubt that humans and other animals sustaining brain damage in infancy can show Earlier we described abnormalities in neural mi- more rapid and more complete recovery of a particugration that are probably found throughout the brain lar function than when comparable damage is susbut it is reasonable to predict that there will be condi- tained later in life (Kennard, 1936, 1940). At the tions in which such abnormalities might be restricted same time, there is also little doubt that this apparent to relatively small zones of cortex. In fact, there is sparing of function is not without some cost now reason to suppose that at least some forms of (Fletcher, Levin, & Landry, 1984; Milner, 1974; developmental dyslexia result from abnormal struc- Taylor, 1984; Teuber, 1975; Woods, 1980). Thus, tural development. Drake (1968) examined the brain although recuperation may appear total for many speof a 12-year-old learning-disabled boy who died of cific verbal and academic skills, more global assesscerebral hemorrhage. Autopsy showed that there ments of cognitive status suggest that the recovery is were atypical gyral patterns in the parietal lobes, an often accompanied by new deficits that affect other atrophied corpus callosum, and neurons underlying functions and may be overlooked (St. Jamesthe white matter that should have migrated to the Roberts, 1981; Taylor, 1984). Furthermore, there is cortex. More recently, Galaburda and his colleagues some evidence that brain damage during the first year have reported analogous results from several dyslex- of life, a time during which there are tremendous ic brains (Galaburda & Kemper, 1979; Galaburda & changes in brain morphology, may actually have Eidelberg, 1982; Geschwind & Galaburda, 1985). more severe consequences than similar damage later. Thus, in the brain of a 20-year-old male who pre- Consider the following examples. viously had a reading disability despite average intelligence, they found an abnormal pattern of Effect of Brain Damage on Language cytoarchitecture, especialJy in the posterior speech It is common to find that language deficits reregion of the temporal-parietal cortex. Although other details varied in these cases, the left posterior sulting from cerebral injury in childhood are usually region was always abnormal. These abnormalities short-lived, and that recovery of conversational were believed to be the result of disordered neuronal speech is nearly complete (e.g., Hecaen, 1976). migration and/ or assembly. The right hemisphere Cognitive function is by no means normal, however. was either completely or largely normal in all of these Studies of children with unilateral brain damage in cases. Finally, Geschwind and Galaburda (1985) infancy by Woods and Teuber (Woods, 1980; Woods claimed to have evidence of similar anomalies in & Teuber, 1973) and Rasmussen and Milner (1975, living dyslexic patients, with arteriovenous malfor- 1977) lead to several conclusions in this respect. {1) mations in the left temporal region. Language survives early left-side injury. (2) If leThe finding of left temporal-parietal abnormal- sions are incurred prior to age 5 and include both ity in dyslexics leads to the question of how these language areas in the left hemisphere, language funcpeople, even as children, might perform on tests sen- tions shift to the right hemisphere. (3) If lesions are sitive to focal cortical lesions. Few studies have di- restricted to the anterior or posterior speech zones, rectly compared dyslexic children to adults with left only the affected language area is likely to shift, leadposterior lesions, but studies of dyslexic children ing to bilateral representation of speech. (4) When have found behavioral deficits on tests that are partic- language shifts to the right hemisphere, it is not withularly disrupted by left posterior lesions including out a price for visuospatial functions are impaired. tests of short-term verbal memory, left/right differ- These functions would not be impaired if the damage entiation, and verbal fluency (Sutherland, Kolb, were incurred later in life. (5) Childhood injuries to Schoel, Whishaw, & Davies, 1982; Whishaw & the left hemisphere, occurring after age 5, seldom Kolb, 1984). We must point out that it is likely that change speech representation. The observed recovnot all children with learning disabilities have left ery of language function is assumed to be mediated posterior abnormalities. It would be interesting, by some sort of intrahemispheric reorganization. (6) however, to determine the correlation between neu- Children with lesions incurred before their first birthropsychological test performance in learning-dis- day had verbal and performance IQ scores below the
Abnormal Brain Development and Behavior
36
CHAPTER2
mean of the nonnal population. In contrast, the effects of lesions after the first birthday depend on the side of the lesion. The left-hemisphere lesions lowered both verbal and performance IQ scores whereas later right-hemisphere lesions adversely affected only the performance IQ scores.
Frontal Lobe Damage We have already discussed the functional development of the frontal cortex as inferred from the performance of children on neuropsychological tests sensitive to frontal lobe function. We now ask what might happen to performance on these tests if the frontal lobe was damaged in infancy, before the frontal cortex was functioning as in adulthood. Milner (1974) found that in those patients with childhood damage to the left frontal lobe, there was a shift in language to the right hemisphere, but there was still a marked impairment of the performance of other tests of frontal lobe function such as the Wisconsin Card Sorting Test. We have seen similar results in a small sample of patients with unilateral frontal lobe injuries in childhood (largely head trauma and birth injury) (see Table 7). We have divided these into two groups: those who sustained their injuries before age 2 and those who sustained their injuries between age 2 and 11. The results show that patients with frontal lobe injury at any age were impaired on the Chicago Word Fluency Test and on a test of copying facial movement sequences (see Kolb & Milner, 1981, for details of this test). The results on card sorting were somewhat more variable, the youngest patients doing the best. In addition, like Woods, we found that the patients with early injuries had significantly lower IQs than did those with adult injuries, although our sample was too small to form a statistically testable group with lesions prior to the first birthday. The failure to see dramatic recovery from early frontal lesions finds support from the literature on laboratory
animals: both Kolb and his colleagues (Kolb, 1987; Kolb & Whishaw, 1981 , 1985a,b) and GoldmanRakic and her colleagues (e.g., Goldman, 1974; Goldman-Rakic, Isseroff, Schwartz, & Bugbee, 1983) have found less than complete recovery following early lesions in rodents and primates, respectively. Indeed, Kolb ( 1987) found that frontal lesions acquired just as cortical neuronal proliferation is completed (i.e., on the day of birth for rats) have greater behavioral and morphological effects than similar lesions even a couple of days later. In particular, although the lesion size appears the same at surgery, the early frontal lesion apparently also interrupts cell migration to the remaining sensorimotor cortex, resulting in a larger effective lesion in these rats. It is possible that a similar result might hold for the child's brain as well, particularly in those cases of prenatal injury. To summarize, disruption of the normal growth and differentiation of the human brain is associated with a variety of pathological states, the nature of the behavioral changes depending upon the developmental state of the brain at the time of injury. Although it was once believed that there was nearly complete recovery following cortical injuries during development, the bulk of the evidence does not support this view. Rather, it appears that events that produce abnormal development of the brain will also produce abnormal behavior, although certain behavioral functions such as speech appear to survive differentially relative to other functions, especially visuospatial functions. For the clinical neuropsychologist the assessment of children with brain injuries and the subsequent prediction of outcome is especially difficult for the damage to the frontal and posterior associational cortices because one must wait until at least 10-12 years of age, the time at which these areas assume adultlike function, to determine how well the child will fare. It seems certain, however, that when such children are given a broad assessment battery, there will be significant neuropsychological deficits.
TABLE 7. Neuropsychological Test Performance of Patients with Childhood Injuries to the Frontal Lobe Group
n
IQ
Card sorting
Chicago fluency
Face copying
Control Adult frontal 2- 14 year frontal <2 year frontal
20
114.0 114.5 93.0" 89.811
6.0 1.9" 2.8" 4.5
61.5 27.1" 27.0" 30.6"
80.0 64.5" 58.3" 56.5"
0
Significanlly low score.
8 4 5
DEVELOPMENT OF THE CHILD'S BRAIN AND BEHAVIOR
Conclusion The process of brain maturation is long, lasting at least into early childhood. We have approached the problem of assessing the nature of functional localization in the cortex by examining the way in which structure and behavior emerge in the developing child. Neurons, the elementary components of the brain, are born, migrate, and, as their processes elaborate, establish connectional relationships with other neurons. Behavioral and cognitive capacities follow a similar sequence of development from the rudimentary to the complex. Structure-function relationships can be inferred by matching the developmental timetables of brain anatomy and physiology with that of behavior. In addition, we have demonstrated that neuropsychological tests that are sensitive to focal cortical damage in adults, can be used to assess whether certain brain areas have reached functional maturity in normal, developing children. Further, by studying the abnormal development of the brain and behavior we may make inferences regarding the importance of particular developmental events upon behavior. The study of anatomical and behavioral development of the brain of the child is admittedly far from complete. However, we believe that the data obtained to date are beginning to answer the question about the nature of the brain of the child. The continued study of developmental neuropsychology promises to change our understanding of the biological bases of the development of human behavior.
References Arlin, P. K. ( 1975). Cognitive development in adulthood: A fifth stage? Developmental Psychology, I 1(5), 602-606. Bergstrom, R. M. (1969). Electrical parameters of the brain during ontogeny. In R. J. Robinson (Ed.), Brain and early behavior (pp. 15-37). New York: Academic Press. Berry, M. (1982). Cellular differentiation: Development of dendritic arborizations under normal and experimentally altered conditions. Neurosciences Research Program Bulletin. 20(4), 451-461. Caplan, P. J., & Kinsboume·, M. (1976). Baby drops the ranle: Asymmetry of duration of grasp by infants. Child Development, 47, 532-534. Carey, S. (1984). Cognitive development: The descriptive problem. In M. S. Gazzaniga (Ed.), Handbook of cognitive neuroscience (pp. 37-66). New York: Plenum Press. Carey. S .• & Diamond. R. ( 1977). From piecemeal to configurational representation offaces. Science, /95, 312-314. Carey. S .• Diamond. R .• & Woods. B. (1980). Development of
37
face recognition-A maturdtional component'! Developmental Psychology. (6(6), 257-269. Caveness. W. F. (1969). Ontogeny of focal seizures. In H. H. Jasper. A. A. Ward. Jr .. & A. Pope (Eds.). Basic mechanism.• ofthe epilepsies (pp. 517-534). Boston: Little, Brown. Caviness, V. S., Jr. (1982). Development of neocortical afferent systems: Studies in the reeler mouse. Neumsden_ces Researt·h Program Bulletin, 20(4), 560-569. Caviness. V. S., & Rakic, P. (1978). Mechanisms of co{tical development: A view from mutations in mice. Annual Review of Neuroscience, I, 297-326. Caviness, V. S., Jr., & Sidman, R. L. (1973). Time of origin of corresponding cell classes in the cerebrdl cortex of normal and reeler mutant mice: An autoradiographic analysis. Journal of Comparative Neurology, 148, 141-152. Caviness, V. S., & Williams, R. S. (1979). Cellular pathology of developing human cortex. Research Publications ofthe AsslJdation for Research in Nervous and Mental Diseases, 57. 69-98. Chi, J. G., Dooling, E. C., & Gilles, F. H. (1977). Left-right asymmetries of the temporal speech areas of the human fetus. Archives of Neurology, 34, 346-348. Chugani, H. T., & Phelps, M. E. (1986). Maturational changes in cerebral function in infants determined by ' 8 FOO positron emission tomography. Science. 2}1, 840-843. Conel, J. L. (1939-1967). The postnatal development of the human cerebral cortex (Vols. 1-VIII). Cambridge, MA: Harvard Unjversity Press. Cowan, W. M. (1979). The development of the brain. Scientific American, 241, 112-133. Coyle, J. T. ( 1982). Development of neurotransminers in the neocortex. Neurosciences Research Program Bulletin, 20(4), 479-491. Cragg, B. G. (1975). The density of synapses and neurons in normal, mentally defective and ageing human brains. Brain, 98, 81-90. Crowell, D. H., Jones, R. H., Kapuniai, L. E., & Nakagawa, J. K. (1973). Unilateral cortical activity in newborn humans: An early index of cerebral dominance? Science, 180, 205208. Diamond, R., & Carey, S. (1977). Developmental changes in the representation of faces. Journal of Experimental Child Psychology, 23, 1-22. Drake, W. (1968). ClinicaJ and pathological findings in a child with a developmental learning disability. Journal ofLearning Disabilities, I, 468-415. Entus, A. K. (1977). Hemispheric asymmetry in processing of dichotically presented speech and nonspeech stimuli by infants. InS. J. Segalowitt & F. A. Gruber (Eds.), Language development and neurological theory (pp. 63-73). New York: Academic Press. Epstein, H. T. (1978). Growth spurts during brain development: Implications for educational policy and practice. In J. S. Chall & A. f. Mirsky (Eds.), Education and the brain (pp. 343-370). Chicago: University of Chicago Press. Epstein, H. T. (1979). Correlated brain and intelligence development in humans. In M. E. Hahn, C. Jensen, & B. C. Dudek (Eds.), Development and evolution of brain si:ze: Behavioral implications (pp. 111-131). New York: Academic Press.
38
CHAPTER2
Flechsig, P. (1920). Anatomie des menschlichen Gehirns und Ruckenmarks. Leipzig: Thieme. Fletcher, J. M., Levin, H. S., & Landry, S. H. (1984). Behavioral consequences of cerebral insult in infancy. In C. R. Almli & S. Finger(Eds)., Early brain damage (Vol. I, pp. 189-213). New York: Academic Press. Galaburda, A.M., & Eidelberg, D. (1982). Symmetry and asymmetry in the human posterior thalamus. II. Thalamic lesions in a case of development dyslexia. Archives of Neurology, 39, 333-336. Galaburda, A. M., & Kemper, T. L. (1979). Cytoarchitectonic abnormalities in developmental dyslexia: A case study. Annals of Neurology, 6, 94-100. Geschwind, N., & Galaburda, A.M. (1985). Cerebrallateralization: Biological mechanisms, associations, and pathology. I. A hypothesis and a program for research. Archives of Neurology, 42, 428-459. Gibson, K. R. (1977). Brain structure and intelligence in macaques and human infants from a Piagetian perspective. In S. Chevalier-Skolnikoff & F. E. Poirier (Eds.), Primate biosocial development: Biological, social, and ecological determinants (pp. 113-157). New York: Garland. Goldman, P. S. (1974). An alternative to developmental plasticity: Heterology ofCNS structures in infants and adults. In D. G. Stein, J. J. Rosen, & N. Butters (Eds.), Plasticity and recovery offunction in the central nervous system (pp. 149-174). New York: Academic Press. Goldman, P. S., & Galkin, T. W. (1978). Prenatal removal of frontal association cortex in the fetal rhesus monkey: Anatomical and functional consequences in postnatal life. Brain Research, 152, 451-485. Goldman-Rakic, P. S., &Brown, R. M. (1981). Regional changes of monoamines in cerebral cortex and subcortical structures of aging rhesus monkeys. Neuroscience, 6, 177-187. Goldman-Rakic, P. S., & Brown, R. M. (1982). Postnatal development of monamine content and synthesis in the cerebral cortex of rhesus monkeys. Developmental Brain Research, 4, 339-349. Goldman-Rakic, P. S., Isseroff, A., Schwartz, M. L., & Bugbee, N. M. (1983). The neurobiology of cognitive development. In P. Mussen (Ed.), Handbook of child psychology: Biology and infancy development (pp. 281-344). New York: Wiley. Greenough, W. T. (1976). Enduring brain effects of differential experience and training. In M. R. Rosenzweig & E. L. Bennett (Eds. ), Neural mechanisms of learning and memory (pp. 255-278). Cambridge, MA: MIT Press. Hebb, D. 0. (1949). Organization ofbehavior. New York: Wiley. Hecaen, H. (1976). Acquired aphasia in children and the ontogenesis of hemispheric functional specialization. Brain and Language, 3, 114-134. Hubel, D. H., & Wiesel, T. N. (1970). The period of susceptibility to the physiological effects of unilateral eye closure in kittens. Journal of Physiology (London), 206, 419-436. Huttenlocher, P. R. (1979). Synaptic density in human frontal cortex-Developmental changes and effects of aging. Brain Research, 163, 195-205. Ingram, D. (1975). Motor asymmetries in young children. Neuropsychologia, 13, 95-102. Jacobsen, M. (1978). Developmental neurobiology (2nd ed.). New York: Plenum Press.
Jones, W. H., & Thomas, D. B. (1962). Changes in the dendritic organisation of neurons in the cerebral cortex following deafferentation. Journal of Anatomy, 96, 375-381. Kennard, M.A. (1936). Age and other factors in motor recovery from precentral lesions in monkeys. American Journal of Physiology, 115, 138-146. Kennard, M.A. (1940). Relation of age to motor impairment in man and in subhuman primates. Archives of Neurology and Psychiatry, 44, 377-397. Kimura, D. (l%3). Speech lateralization in young children as determined by an auditory test. Journal of Comparative and Physiological Psychology, 56, 899-902. Knox, C., & Kimura, D. (1970). Cerebral processing of nonverbal sounds in boys and girls. Neuropsychologia, 8, 227-237. Kolb, B. (1987). Factors affecting recovery from early cortical damage in rats. I. Differential behavioral and anatomical effects of frontal lesions at different ages of neural maturation. Behavioral Brain Research, 25, 205-220. Kolb, B., & Milner, B. (1981). Performance of complex arm and facial movements after focal brain lesions. Neuropsychologia, 19(4), 491-503. Kolb, B., Milner, B., & Taylor, L. (1983). Perception offaces by patients with localized cortical excisions. Canadian Journal of Psychology, 37(1), 8-18. Kolb, B., & Nonneman, A. J. (1978). Sparing of function in rats with early prefrontal cortex lesions. Brain Research, 151, 135-148. Kolb, B., & Taylor, L. (1981). Affective behavior in patients with localized cortical excisions: Role of lesion site and side. Science, 214, 89-91. Kolb, B., & van der Kooy, D. (1985). Early cortical lesions alter cerebral morphogenesis and connectivity in the rat. Society for Neuroscience Abstracts, 11, 989. Kolb, B., & Whishaw, I. Q. (1981). Neonatal frontal lesions in the rat: Sparing of learned but not species-typical behavior in the presence of reduced brain weight and cortical thickness. Journal of Comparative and Physiological Psychology, 95(6), 863-879. Kolb, B., & Whishaw, I. Q. (l985a). Earlier is not always better: Behavioral dysfunction and abnormal cerebral morphogenesis following neonatal cortical lesions in the rat. Behavioural Brain Research, 17, 25-43. Kolb, B., & Whishaw, I. Q. (1985b). Fundamentals of human neuropsychology (2nd ed.). San Francisco: Freeman. Lenneberg, E. H. (1%7). Biological foundations of language. New York: Wiley. Lezak, M.D. (1983). Neuropsychological assessment (2nd ed.). New York: Academic Press. Marcel, T., & Rajan, P. (1975). Lateral specialization forrecognition of words and faces in good and poor readers. Neuropsychologia, 13, 489-497. Marin-Padilla, M. (1970). Prenatal and early postnatal ontogenesis of the motor cortex: A Golgi study. I. The sequential developmentofcorticallayers. Brain Research, 23, 167183. Michel, G. F. ( 1981 ). Right handedness: A consequence of infant supine head-orientation preference? Science, 212, 685-687. Milner, B. ( 1964). Some effects of frontal lobectomy in man. In J. M. Warren & K. Akert (Eds.), The frontal granular cortex and behavior (pp. 313-334). New York: McGraw-Hill.
DEVELOPMENT OF THE CHILD'S BRAIN AND BEHAVIOR
39
Milner, B. (1974). Sparing of language function after early uni- Rasmussen, T., & Milner, B. (1977). The role of early left-brain injury in determining lateralization of cerebral speech funclateral brain damage. Neurosciences Research Program Bultions. Annals of the New York Academy of Sciences, 299, letin, 12, 213-217. 355-369. Milner, B. (1980). Complementary functional specializations of the human cerebral hemispheres. In R. Levi-Montalcini Robinson, R. J. (1966). Cerebral function in the newborn child. (Ed.), Nerve cells, transmitters and behavior (pp. 601-625). Developmental Medicine and Child Neurology, 8, 561-567. Vatican City: Pont. Acad. Scientairum. Sidman, R. L., & Rakic, P. (1973). Neuronal migration, with special reference to developing human brain: A review. Brain Molfese, D. L., & Molfese, V. J. (1980). Cortical responses of preterm infants to phonetic and nonphonetic speech stimuli. Research, 62, 1-35. Developmental Psychology, 16(6), 574-581. Spreen, 0., Tupper, D., Risser, A., Tuokko, H., & Edgell, D. Peiper, A. (1963). Cerebral function in infancy and childhood. (1984). Human developmental neuropsychology. New York: New York: Consultants Bureau. Oxford University Press. Peters, M., & Petrie, B. J. F. (1979). Functional asymmetries in St. James-Roberts, I. (1981). A reinterpretation of hemispherecthe stepping reflex of human neonates. Canadian Journal of tomy data without functional plasticity of the brain. 1. IntelPsychology, 33, 198-200. lectual function. Brain and Language, 13, 31-53. Piaget, J. (1952). The origins of intelligence in children. New Sutherland, R. J., Kolb, B., Schoel, M. Whishaw, I. Q., & York: Norton. Davies, D. (1982). Neuropsychological assessment of chilPoliakov, G. I. (1949). Structural organization of the human ceredren and adults with Tourette syndrome: A comparison with bral cortex during ontogenetic development. In S. A. learning disabilities and schizophrenia. In A. J. Freidhoff & Sarkisov, I. N. Filimonov, & N. S. Preobrazenskaya (Eds. ), T. N. Chase (Eds.), Gilles de Ia Tourette syndrome. New Cytoarchitectonics of the cerebral cortex in man (pp. 33-92). York: Raven Press. Medgiz. (In Russian) Taylor, H. G. (1984). Early brain injury and cognitive developPoliakov, G. I. (1961). Some results of research into the development. In C. R. Almli & S. Finger (Eds.), Early brain ment of the neuronal structure of the cortical ends of the dJJmage: Research orientations and clinical observations analyzers in man. Journal of Comparative Neurology, 117, (pp. 325-345). New York: Academic Press. 197-212. Teuber, H.-L. (1975). Recovery of function after brain injury in Poliakov, G. I. (1965). Development of the cerebral neocortex man. Ciba FoundJJrion Symposium, 34, 159-186. during first half of intrauterine life. InS. A. Sarkisov (Ed.), Turkewitz, G. (1977). The development of lateral differentiation Development of the child's brain (pp. 22-52). Leningrad: in the human infant. Annals of the New York Academy of Medicina. (In Russian) Sciences, 299, 213-221. Purpura, D.P. (1976). Structure-dysfunction relations in the visu- Twitchell, T. E. (1965). The automatic grasping responses of al cortex of preterm infants. In M. A. B. Braxier & F. Coinfants. Neuropsychalogw, 3, 247-259. ceani (Eds. ), Brain dysfunction in infantile febrile convul- Wada, 1. A., Clarke, R., & Hamm, A. (1975). Cerebral hemisions (pp. 223-240). New York: Raven Press. spheric asymmetry in humans: Cortical speech zones in 100 Purpura, D. P. (1982). Normal and abnormal development of adult and I 00 infant brains. Archives ofNeurology, 32, 239cerebral cortex in man. Neurosciences Research Program 246. Bulletin, 20(4), 569-577. Whishaw, I. Q., & Kolb, B. (1984). Neuropsychological assessRakic, P. (1972). Mode of cell migration to the superficial layers ment of children and adults with developmental dyslexia. In of fetal monkey neocortex. Journal of Comparative NeuR. N. Malatesha & H. A. Whitaker (Eds.), Dyslexia: A rology, 145, 61-84. global issue (pp. 375-404). The Hague: Nijhoff. Rakic, P. ( 1975). Timing of major ontogenetic events in the visual Williams, R. S., Ferrante, R. 1., & Caviness, V. S., 1r. (1975). cortex of the rhesus monkey. InN. A. Buchwald & M. BraNeoconical organization in human cerebral malformation: A zier (Eds.), Brain mechanisms in mental retardntion. New Golgi study. Neuroscience Abstracts, 1, 776. York: Academic Press. Witelson, S. F. (1977). Early hemisphere specialization and inRakic, P. (1976). Prenatal genesis of connections subserving terhemisphere plasticity: An empirical and theoretical reocular dominance in the rhesus monkey. Nature, 261, 467view. InS. 1. Segalowitz & F. A. Gruber (Eds.), Language 471. development and neurological theory (pp. 213-287). New Rakic, P. (1981). Developmental events leading to laminar and York: Academic Press. areal organization of the neocortex. In F. 0. Schmitt, F. G. Woods, B. T. (1980). The restricted effects of right-hemisphere Worden, G. Adelman, &S. G. Dennis (Eds.), The organizalesions after age one; Wechsler test data. Neuropsychologia, tion of the cerebral cortex (pp. 7-8). Cambridge, MA: MIT 18, 65-70. Press. Woods, B. T., & Teuber, H.-L. (1973). Early onset of compleRakic, P. (1984). Defective cell-to-cell interactions as causes of mentary specialization of cerebral hemispheres in man. brain malformations. In E. S. Gollin (Ed.), Malformations of Transactions of the American Neurological Association, 98, development-Biological and psychological sources and 113-117. consequences (pp. 239-285). New York: Academic Press. Yakovlev, P. E., & Lecours, A.-R. (1967). The myelogenetic Rasmussen, T., & Milner, B. (1975). Clinical and surgical studies cycles of regional maturation of the brain. In A. Minkowski of the cerebral speech areas in man. In K. J. Zulch, 0. (Ed.), Regional development of the brain in early life. OxCreutzfeldt, & G. C. Galbraith (Eds.), Cerebral localization ford: Blackwell. (pp. 238-257). Berlin: Springer-Verlag.
3 Higher Cortical Functions in Children A Developmental Perspective LAWRENCE V. MAJOVSKI
Higher cortical functions in children proceed through different stages of development. Significant limitations in our knowledge exist as to the process of the normal developing human brain with respect to neurophysiological, neurochemical, neuroanatomical, and other related disciplines. Attempts have been made to correlate anatomical and behavioral data in a direct manner, leading to a surfeit of postulations in the literature against a shortage of supporting data for known brain-behavior relationships in children (Taylor, Fletcher, & Satz, 1984). Much emphasis tends to be placed on proposed neural mechanisms accounting for changes regarding development of the human CNS versus description of changes taking place with respect to normal development of the child's brain. Difficulties exist in drawing conclusions because each human brain is unique with respect to its cellular differentiation, acculturation factors, and neural growth patterns (Cooke, 1980). Luria ( 1969a; personal communication, 1977) stressed that what was lacking in the area of clinical child neuropsychology was an integrative scheme outlining a conceptual topography of normal brain development with concomitant motor, sensory, perceptual, and other processes involved in children's higher cortical functions. The main goal of this chapter is to discuss important facts and concepts related to formation of the normal developing brain from conception through childhood. Emphasis wi11 be placed on a normal developmental perspective versus pathological consequences resulting from abnormal influences.
LAWRENCE V. MAJOVSKI • Huntington Medical Re;earch Institutes, Advanced Neurosurgical Laboratories, Pastdena, California 91105
Development of the Human CNS Throughout its development, one of the brain's many functions is to act in generating behavior. The question of how the brain effects control over behavior is central to the study of human neuroscience. The brain can be viewed as a decision-making center for information processing. Understanding the development of the human CNS poses the basic problem of how inhibition brings about the regulation and integration of higher cortical processes involved in the brain's development. Two major themes of importance in this regard are the integrative action of the component parts of the nervous system and that of cellular differentiation, i.e., how the component parts are derived (cellular differentiation) and, through the course of embryological development, the emergence of relationships in the patterns of its organization (Humphrey, 1978: Nilsson, 1978). Some major plaguing problems in understanding the brain's development stem from the lack of a blueprint of nature's original design. Completion of the human brain took place some 50,000 years ago. To date, we still have incomplete knowledge as to the original model. Neuroanatomists, in studying the brain's development, have discovered that the brain develops much more rapidly than other organs. Why this is so is not known at present. The influence of acculturation on the brain is incalculable and is changing the brain's structure more rapidly today than ever before. The human nervous system contains an estimated 1Q9 neurons in the CNS, 30 X 109 in the cerebellum itself, and some 100 X 109 granule cells in the cerebellum (in terms of the macromolecular layers studied). In addition, there are an estimated 10 12 glial cells. It is the association of these cells that makes humans human. 41
42
CHAPTER3
Neuroembryologically, the CNS develops from There is both exteroceptive and proprioceptive senthe medullary plate of the ectoderm, which becomes sibility (Afifi & Bergman, 1980; Carlson, 1977). the neural tube differentiates into the spinal cord and The importance of this phenomenon is seen in brain. the control and regulation of the muscular responses The human nervous system can be divided into achieved via sensory cells in the muscles which prothree major aspects: central, peripheral, and auto- vide a feedback circuit through sensory nerves to the nomic nervous systems. All three act in concert to CNS (Crelin, 1973; Kahle et al., 1978). Luria control behavioral activities (e.g .. motor, sensory, (1973a) and others have pointed out the functional acoustic, optic). It is commonly held that the brain is interrelations between the nervous system, the orgathe organ system that controls human behavior (Got- nism, and the environment from which emerge hightlieb, 1976a,b). Another key in attempting to under- er cortical processes. These three components tostand how the brain works is the problem of inhibi- gether serve conscious perception, voluntary tion and its role in human behavior; exactly how the · movement, and the processing of messages through brain accomplishes control (inhibition) is not fully the process of polysensory integration. To better apunderstood at the present time. preciate the interplay between structure and function, it is important first to understand the development of the brain, starting with the formation of the embryonMorphology ic disk, the nervous system's point of origin (Moore, The CNS is usually defined as the brain (en- 1974; Humphrey, 1978). cephalon) and spinal cord (medulla spinalis). The brain lies in the cranial cavity surrounded by a bony Neuroembryonic Structure Formation capsule. The spinal cord is situated in the vertebral canal surrounded by the vertebrae. Both are covered The nervous system starts developing approxby cranial or spinal meninges that enclose a space imately 18 days after fertilization. The egg is then filled with cerebrospinal fluid (CSF). The peripheral composed of ectoderm and endoderm, with mesonervous system (PNS) is comprised of cranial and derm developing between the two. The nervous sysspinal nerves (31 pairs) with associated ganglia con- tem is derived from the ectodermal layer. During sisting of motor fibers and sensory fibers. There are embryological development, the neural plate, the two types of motor fibers: (1) somatic motor fibers, neural tube, and the neural crest form. The neural which terminate in skeletal muscles; and (2) auto- crest becomes elevated to form the neural folds, nomic fibers, which furnish innervation to cardiac which, in turn, approximate each other in the midline muscle, smooth muscles, and glands. Sensory fibers and then fuse to form the neural tube. Cells at the receive stimuli from receptive organs of various margins of the folds are not included in the wall of the types. Nerves of the PNS supply the head, trunk, and neural tube. The partial fusion of the neural folds limbs. The CNS and PNS together serve conscious occurs approximately 23 to 24 days postfertilization perception, voluntary movement, and the processing (Lowrey, 1978; Moore, 1974, 1975; Hamilton & of sensory-based messages and integration (Kahle, Mossman, 1974; Kahle eta/., 1978; Crelin, 1974; Le Leonhardt, & Platzer, 1978; Arey, 1974; Nilsson, Douarin, 1980). 1978). In its formative stages, the neural tube appears In the autonomic nervous system (ANS), there as a straight structure. However, during its organoare two antagonistic parts, the sympathetic and the genesis, cervical somites deviate. from the shape of parasympathetic systems, which jointly are responsi- the simple neural tube. This portion, destined to beble for preserving a constant internal environment come the brain, forms various bulges and cavities, (homeostasis). All viscera, blood vessels, and glands each of which has significance in the embryological are innervated by the ANS. The human nervous sys- plan of development (Jones & Cowan, 1978). · tem, the organism, and the environment are all funcThree primary bulges appear in the brain region tionally interrelated. of the neural tube: the forebrain (prosencephalon), The human organism not only responds to its midbrain (mesencephalon), and hindbrain (rhombensurroundings, it acts spontaneously on them as well cephalon). When the development of the caudal end throughfunctional circuits. The action that is insti- of the tube is completed, the optic vesicles appear and gated by the CNS (transmitted via efferent nerves) is protrude from each side of the forebrain. Otic invagregistered by the sense organs and information is then ination also occurs at day 28 postfertilization. In returned to the CNS via the efferent nerves. An terms of the ventricles of the future brain, a cephalic itegrative process follows until regulation obtains. flexure and a cervical flexure of the future brain be-
HIGHER CORTICAL FUNCTIONS IN CHILDREN
come visible with cavities at respectively the prosocele of the forebrain, mesocele of the midbrain, and rhombocele of the hindbrain. Each optic vesicle will differentiate to form a characteristic pattern: first, an optic cup, then a stalk, and, later on, an optic nerve, which becomes part of the eyeball. The original connection of each optic vesicle becomes located in the diencephalon, a subdivision of the forebrain. By day 36, the forebrain divides into two parts. The caudal subdivision becomes the diencephalon, and the anterior component further differentiates to form the telencephalic vesicles which eventually become the cerebral hemispheres (Hamilton & Mossman, 1974; R. Y. Moore, 1977; Jones & Cowan, 1978; Humphrey, 1978). Simultaneous with the subdivision of the forebrain, the original cavity (the prosocele) undergoes subdivisions. Two telencephalic vesicles (teloceles) are formed and become the lateral ventricles. The median telocele, which lies between these two teloceles, together with the diocele, becomes the third ventricle. The mesocele develops into the cerebral aqueduct. As the forebrain divides into the telencephalon and diencephalon, the hindbrain is forming into two structures: the anterior metencephalon, which becomes the pons and the cerebellum; and the posterior myelencephalon, which becomes the medulla oblongata. The fourth ventricle forms from the cavity of the metencephalon (metacele), together with the cavity of the myelencephalon (myelocele). By day 34, the cerebellar plate, cervical and mesencephalic flexures, lens invagination, otic vesicles, and olfactory placodes are visible. By day 45, olfactory evagination will have occurred as well as formation of the cerebral hemispheres. Lens fiber migration of retinal cells will begin in earnest at this time. Growth and development by 3 months postfertilization take place in terms of the two smoothwalled telencephalic vesicles. These are easily identified as the cerebral hemispheres. Each emerging hemisphere of the telencephalon will have divided by this time into three parts, each of which has different functions. The first component is the rhinencephalon; the second is the thick basal (striatal) region, which develops into the basal ganglia; and the third is a suprastriatal region, which forms the cerebral cortex and related underlying white matter (Kahle et al., 1978; Crelin, 1974; Arey, 1974). The rhinencephalon starts out as an outgrowth from the telencephalon, e.g., the olfactory lobes. Each olfactory lobe forms part of the wall of the
43
cerebral hemispheres. The second major portion of the rhinencephalon forms part of the wall of the cerebral hemispheres, e.g., the hippocampus. A bulging mass appears on the medial wall of the lateral ventricles, bilaterally. In humans, the rhinencephalon includes, in addition to the bilaterally placed olfactory bulbs and hippocampi, such structures as the bilateral pyriform lobes, midline septeum pellucidum, and midline fornix. The rhinencephalon differentiates into structures referred to as the limbic lobe, which contains interconnections with structures such as the thalamus, epithalamus, and hypothalamus. This constitutes the limbic system. The functional significance of the limbic system is that it is associated with emotional responses and the integration of olfactory information with both visceral and somatic information. The limbic system is involved in "emotional expression," and the hypothalamus is involved in the regulation or control of emotions via hormonal output. The thalamus serves as the portal to the cerebral cortex, but is inextricably bound together in the processing of sensory information, which leads to conscious activity. Other bilaterally placed structures of the rhinencephalon include: the stria terminalis; septum; amygdaloid bodies; medial and lateral olfactory gyria; parahippocampal gyrus; and cingulate gyrus (Carpenter, 1978; Hamilton & Mossman, 1974; R. Y. Moore, 1977; Crelin, 1974; Kahle etal., 1978).
Basal Ganglia The second part of the telencephalon formed is the basal ganglia (or basal nuclei). They are formed in the thickened portion of the striatal region of the telencephalic area, composed of several groups of neuronal cell bodies. One of the major groups of these ganglia (i.e., nerve cell bodies outside the brain and spinal cord) is the corpus striatum, which becomes related to the thalamus of the diencephalon. Up to the third month, the corpus striatum and thalamus are separated by a deep groove. The corpus striatum then bu.lges into the lateral ventricle while the thalamus protrudes into each side of the third ventricle. Beginning about the fourth month, the groove between these two structures will disappear and fuse into a common mass. When fully matured, the basal ganglia comprise the following main structures: caudate nucleus, claustrum, amygdaloid body, and corpus striatum. These structures will be involved with motor control (Nauta, 1986a,b; Carpenter, 1978; Kahle et al., 1978). In the suprastriatal region, the third component of the telencephalon forms all of the externally visi·
44
CHAPTER3
ble cerebral hemispheres. The hemispheres increase in size and completely envelope the mesencephalon and the upper portion of the cerebellum, and the originally smooth surfaces begin to show convolutions at around 7 weeks. The formation of the surface convolutions, known as sulci, and the deeper depressions, tertnedfissures, allows the outei: layer of neurons, e.g., the six cell layers in the cerebral cortex, to increase greatly in depth without a major change in the overall size of the brain in relation to its final volume. By the completion of its development, the cerebral cortex will range in thickness from 1.5 to approximately 4.0 nun with a surface area of 2.3 x 103 to 2.5 x 103 crn2 (Crelin, 1973, 1974). The first major fissure to appear on the lateral aspect of each cerebral hemisphere is the lateral Sylvian sulcus (orfissure) which becomes evident by the third month. The slowly growing floor of the sulcus, which is lateral to the corpus striatum of the basal ganglia, is the insula. It eventually becomes completely covered by adjacent areas of the hemisphere. Beneath the convolutions of the cerebral cortex lies the substrate of the highest centers of cortical integration in the human nervous system. These (six) cortical layers subserve higher cortical functions involving conscious activity, memory, processing of information, decision-making, voluntary action, reflection, and planning (Luria, 1966, 1969b, 1970, 1973a,b).
Ventricle Formation and CSF Important changes are also occurring in various parts of the brain after the third month, particularly ventricle formation. The lateral ventricles each develop three horns that protrude into the various lobes of the cerebral hemispheres: the anterior horn projects into the frontal lobe; the inferior horn into the temporal lobe; and the posterior horn into the occipital lobe. Each lateral ventricle occupies a more lateral position relative to the third ventricle and the formerly broad interventricular foramen (foramen of Monro) becomes a narrow canal. The third ventricle connects with the lateral ventricles of each hemisphere by the foramen of Monro and continues caudally into the cerebral aqueduct of Sylvius, expanding beneath the cerebellum to form the fourth ventricle (Humphrey, 1978). Simultaneously, the egg-shaped thalami bulge into a third ventricle. Eventually the two thalami, with normal development, bridge this ventricle, come in contact with each other, fuse, and produce an interthalamic bridge. The cerebral aqueduct of Syl-
vius becomes a long, slender tube connecting the third and fourth ventricles. Two major foramina become prominent: the left lateral opening (foramen of Luschka) and the median opening (foramen of Magendie), both of the fourth ventricle (Kahle et al. , 1978; Moore, 1975). A region of invagination of the choroid plexus will occur along the choroid fissure of the lateral ventricle. Functionally, the choroid plexus serves as a source of CSF. CSF serves three main functions: it supports the weight of the brain in the skull; it protects the brain from physical trauma during injury to the skull; and it provides a stable chemical environment for the CNS, despite plasma's chemical cornposition changes (Afifi & Bergman, 1980). Ependymal cells lining the brain's ventricles form the medial surfaces of each lateral ventricle, the roofs of the third and fourth ventricles, and portions of the plexus, As these cells grow and invaginate, they are accompanied by blood vessels known as choroidal vesseles. CSF escapes through the median (foramen of Magendie) and lateral (foramen of Luschka) opening of the fourth ventricle into the subarachnoid space surrounding the brain and spinal medulla. The lining of the choroid plexus also forms a physiological barrier known as the brain-barrier system between the CSF and the blood supply to the brain. The brain-barrier system is more permeable in newborn infants than in adults, becoming less permeable as the brain matures. For example, bilirubin in high concentration in infants can cause brain damage because it passes through the brain-barrier system. Similar high levels of bilirubin do not affect the adult brain (R. Y. Moore, 1977b; Langman, 1975; Afifi & Bergman, 1980).
Spinal Cord Formation, Alar and Basal Plates During neuroernbryological development, the neural tube is divided into longitudinal zones: the ventral half of the lateral wall differentiates early into the basal plate. It is considered by some to be the site of origin of the motor nerve cells. The dorsal portion of the lateral wall differentiates later and is termed the . alar plate. It is the site of origin of sensory nerve cells. Between the alar and the basal plates lies an area from which autonomic nerve cells are thought to arise. Viewing the structural plan of the spinal cord and brain stem in this fashion aids our understanding of how various parts of the brain are organized. Because it is held that the basal plate does not participate in the formation of the brain areas beyond the mid-
HIGHER CORTICAL FUNCTIONS IN CHILDREN
brain (metencephalon), the diencephalic and telencephalic vesicles are thought to arise from the alar plate (Kahle et al., 1978). Alar plate cell bodies are comprised of sensory and coordinating (internucial) neurons. These are located in the layer of gray matter (mantle layer). Gray matter is the region of the brain and cord that contains aggregates of nerve cell bodies, as distinct from ganglia, which are nerve cell bodies that lie outside the brain and spinal cord. Differentiation of the diencephalon from the alar plate results in division into dorsal and ventral portions. The dorsal part becomes the thalamus and consists of cell bodies of sensory and coordinating neurons. These nuclei are nerve cell groups within the brain. The ventral portion of the alar plate develops into the hypothalamus, which is comprised of motor control neurons. The dorsal (alar) plate thus becomes the site of sensory and coordinating neuronal cell bodies. The ventral basal plate becomes the site of motor control neuronal cell bodies (R. Y. Moore, 1977b; Crelin, 1973, 1974; K. L. Moore, 1977a). The thalamus exerts control through its projectional (internuncial) neurons, which synapse with parts of the brain other than the cerebral hemispheres, in particular, with the hypothalamus. Thalamocortical interaction exists in what is known as the thalamocortical projection system. Two major structures of importance in terms of cortical-cortical and cortical-subcortical interneuronal connections are the major fiber tracts that arise in the internal capsule and the medial forebrain bundle (MFB). The last projectional pathway from thalamus to cerebral cortex takes place through the nuclealis reticularis thalami. It is thought that through this projection pathway, there is monitoring and modulation of information from lower levels of nervous activity to the upper regions of the CNS. The thalamus also serves a role in pacemaking activities seen in the electroencephalogram (EEG). Another important function is involvement with selective awareness in conscious activity (Bear, 1986; Hamilton & Mossman, 1974; Crelin, 1973; Scheibel & Scheibel, 1961, 1963; Bloom, 1979). A significant portion of the human thalamus is comprised of a group of nuclei that receive proprioceptive and general cutaneous, visceral, visual, and acoustic impulses, which are relayed to the cerebral cortex via other projectional (internuncial) neurons. The structural and functional relationships of the component parts of thalamic nuclei are inextricably comingled (Riss, 1972; Scheibel & Scheibel, 1966, 1972).
45
The hypothalamus, derived trom the alar plate of the diencephalon, is part of the limbic system and is considered to be the headquarters for central motor control of the ANS. It regulates emotional responses and certain visceral functions, such as appetite, thirst, digestion, sleep, sexual drive, heart rate, body temperature, general smooth muscle action of internal organs, and control of the anterior lobe of the hypophysis. The hypothalamus is also involved in the releasing of neuroregulator factors. By the seventh week, the infundibulum (posterior lobe of the hypophysis) appears and develops as an extension of the hypothalamus. The parathyroid structure appears in association with the thyroid gland. During the eighth to tenth weeks, thyroid follicles emerge as weU as production of adrena1ine and noradrenaline (Lemire, Loeser, Alvord, & Leech, 1975; Hamilton & Mossman, 1974; Crelin, 1974). The basal portion of each cerebral hemisphere, situated anterior and lateral to the hypothalamus, is derived from the ventral portion of the alar plate. This basal area is comprised of the basal ganglia. Other important anatomical structures that emerge from the telencephalon include the cerebral cortex and the septal-hippocampal-amygdaloid nuclei complex. Fiber tract systems are also developing with nerve fibers of the brain and spinal cord that have a common origin and a common destination. This anatomical feature should be kept in mind in thinking about the development and differentiation of white matter versus fiber tract systems. It is the cell bodies of neurons-the functional and anatomical unit of the human nervous system-that are involved in human thought, memory, and voluntary and regulatory motor control over the entire nervous system. They are localized in the cerebral cortex, and their functions have been referred to as higher cortical functions in man (Kahle et a/., 1978; Afifi & Bergman, 1980; Crelin, 1974).
Hippocampus The hippocampus plays a significant role in the generation and retrieval of memory. It is also involved in the generation of conscious activity in humans. The brain, as it matures, deals with the analysis of thought, memory functioning, and decision-making with respect to inputs from voluntary and regulatory motor units exercising control over the entire human body. It decides whether or not to engage in or refrain from initiating performance. Speculatively, autism might be linked to the amygdala, hippocam-
46
CHA.PTER3
pus, or septal areas, or faulty information transmission in terms of sensory influx of the structures of the higher cortical areas of the brain during development (Davison & Dobbing, 1968; Altman, Brunner, & Bayer, 1973; R. Y. Moore, 1977b; Cooke, 1980). Major fiber tract systems that become crucial to higher cortical activity in children lie in the area of the corpus callosum (anterior and posterior commissures) and the fimbria. The cerebral hemispheres form the major portion of the brain's volume (R. Y. Moore, 1977b; Hamilton & Mossman, 1974).
Cellular Differentiation of the Nervous System Structural and functional organization of the nervous system are closely related to cellular organization. The neuron is the basic building block of the nervous system. In terms of the nervous system's network, neurons are interconnected in a special way via synapses (Koffler & Nicholls, 1977; Lund, 1978; Cretin, 1973, 1974; R. Y. Moore, 1977b). Inhibitory synapses are as important as excitatory ones. Inhibitory synapses limit and select continual impulse inflow in the sense that chosen signals of importance are transmitted for further information processing. Unimportant signals are suppressed. As the neural tube forms, three cell layers develop and differentiate from its walls: the ependymal layer, the mantle layer, and the marginal layer. These layers form distinct zones: the ependymal zone, the mantle zone (gray matter), and the marginal zone (white matter). The outermost layer becomes the pia mater derived from pial cells. The organizational patterns formed from these zones are best seen from the viewpoints of the spinal medulla (cord), the cerebellar hemispheres, and the cerebral hemispheres (Kahle et al., 1978; Cretin, 1974; Arey, 1974; Moore, 1974).
Spinal Medulla The basic three-zone pattern of this structure will be retained into maturity as follows: the ependymal zone remains as columnar cells lining the lumen of the central canal; the cells of the mantle zone form the gray matter; and the marginal zone becomes the white matter. The gray matter of the spinal medulla assumes the anatomical appearance of an H-shaped mass surrounded by white matter. The association and commissural (internuncial) neurons of the gray matter of the spinal medulla are formed by
other mantle zone neurons. The white matter lacks neurons. Instead, there are bundles of axons arising from those nerve cells located throughout all levels of the spinal medulla and brain to form various fiber tract systems. White matter, in contrast to gray matter's large number of cell bodies (neurons), contains only scattered bodies of cells, which are chiefly supportive in nature, e.g., glial cells, which outnumber neurons in the human nervous system by a ratio of 10: 1. The brainstem's organizational pattern is similar to that of the spinal medulla. With increased specialization of the brain, more complex arrangements of the axonal tracts occur constituting brain white matter.
Cerebellum Research in the past decade indicates that an enormous amount of integration occurs within the cerebellum, specifically with the bursting of Purkinge fibers and the quenching of the Purkinge cell (Ito, 1984). The organizational pattern in the cerebellum shows pronounced deviation, neurohistologically, from the basic structural pattern of the mesencephalon, metencephalon, and medulla oblongata, which develop from neuroblasts of the dorsal portion of the mantle layer. Cells of the cerebellum are involved in motor control, and also have interconnections via the basal ganglia that may serve a role in monitoring and coordinating muscle activity in relation to all forms of sensory input (Kahle et al., 1978; Carpenter, 1978). Two other main deviations from the basic organizational pattern of the neural tube are the development of ventricles of the brain and the tube. The cerebellum serves many functions. It repeats activities in a regulated, precise manner; smooths out all motor activities; is concerned with sensory activity; and may be involved in affectual responses. It is the center for the smooth coordination of muscular responses, especially those involved with subconscious maintenance of normal posture. Due to its primitive three-layer system-molecular, Purkinge, and granular cells-it is also involved in feedback and dampening circuits that serve as a servomechanistic system to control complicated integrative movements such as talking and writing. Many cerebellar cells serve as inhibitory ''tum off' cells such as Mugwump and Golgi II cells (Carpenter, 1978; Ito, 1984). The cerebellum can be viewed as being similar to a computer in that it may even be able to generate programs on its own (Schei· bel, 1978).
HIGHER CORTICAL FUNCTIONS IN CHILDREN
Thalamocortical Fiber System The thalamus receives all types of sensory input relayed by projectional neurons either to various nuclei of the brain stem or to the cerebral cortex. Nearly all of the axons conveying sensory input to each thalamus, cross from the opposite side of the spinal medulla or brain stem. Those fibers that have not crossed include half of optic nerve fibers entering the thalamus from the same side. From each thalamus, projectional neurons relay sensory impulses to the cerebral cortex for which there is a corresponding area of the thalamus. Activation of a minute portion of the thalamus will stimulate the corresponding and much larger portion of the cerebral cortex via the axons of the thalamocortical projectional neurons. The cerebral cortex contains cell bodies of associational neurons, which send their axons through the white matter of the hemisphere and in another part of the cortex of the same side. The cortex also contains cell bodies of commissural neurons, which send their axons via the hemisphere's white matter to end up in the opposite cerebral hemisphere. It is via the commissures that a bridge is formed to allow the functional integration between the two sides of the brain (Carpenter, 1978; Kahle et al., 1978; Crelin, 1973; Arey, 1974). During development, the first commissure to appear is the anterior commissure. It interconnects the olfactory amygdaloid nuclei and cortical portions of the cerebral hemispheres. Second to appear is the hippocampal commissure (region of the fornix), which will unite the two hippocampal, olfactory portions of the hemispheres. Then, posteriorly, in the region of the pineal body, the habenular and posterior commissures interconnect the diencephalon. The last of the commissures and the largest of all is the corpus callosum. It is known that myelination begins in the brain at about embryonic week 16 and typical layers in the cerebral cortex are found at around week 24. At birth, there will be continuing organization of axonal networks, cerebral corticospinal tract development, motor coordination, and myelination patterns (R. Y. Moore, 1977b).
Glial Cells White matter and gray matter are made up almost entirely of cell constituents: white matter consists chiefly of bundles of axons, glial cells, and blood vessels; and gray matter is composed primarily of neuronal cell bodies, dendrites, axons, glial cells, and blood vessels (Afifi & Bergman, 1980).
47
Glial cells are the supporting structures of the CNS. There are three types of glial cells in the human CNS: astrocytes, oligodendrocytes, and microglia cells. Glial cells outnumber neurons 10: 1. Astrocytes and oligodendrocytes derive-chiefly from the spongioblasts in the mantle layer. The latter are ~e rived from neural tube epithelial cells. Micro gilacells primarily arise from mesodermal embryonic connective tissue, from which all layers of blood vessels of the brain and spinal medulla arise. Astrocytes are composed principally of two different functional forms. The first are the fibrous astrocytes, which are abundant in white matter, providing both support and binding for the tracks of nerve fibers. The second,protoplasmicastrocytes, are present in large numbers in the gray matter, and serve many different purposes. They establish close contacts with neuronal cell bodies, blood capillaries, and pia mater. The end-feet of protoplasmic astrocytes, in conjunction with endothelial cells of capillaries, form a highly selective blood-brain barrier (Moore, 1975, 1977a; Hamilton & Mossman, 1974; Crelin, 1973, 1974).
Cerebral Hemispheres During the third month, migrating neuroblasts from the mantle zone pass into the marginal zone, giving rise to the cerebral cortex. Stratification within the cortex proceeds at an ever-increasing rate and, at approximately 6 months, the six layers of cell bodies and their associated interconnections that characterize the cerebral cortex are identifiable. The fmal differentiation of the outer layers continues until the second decade of life (and maybe even longer). The outer three layers become more highly developed in humans than in any other species of animal (K. L. Moore, 1974, 1975, 1977a; R. Y. Moore, 1977b; Arey, 1974; Crelin, 1973, 1974). The pattern of neuronal development in the cerebral cortex during infancy is integral to the complex functions of the cortex, which ultimately consists of a vast information storage and processing ensemble. These will include cognitive reasoning abilities, memory, communication, reflective thinking, individual performance skills. and other functions. From infancy through the first three to five years oflife, the subcortical-cortical interactions serve a prominent role as to the storage of many of the patterns of motor responses that can be elicited at will in order to control motor functions of the developing human body. It is the cortex that gives humans voluntary control over how they will react to sensory-perceptual stim-
48
CHAPTER3
uli, integrate this information, and decide whether or not to act upon it in a deliberate manner (Scheibel & Scheibel, 1973). In humans, voluntary control of muscles is almost exclusively regulated through the descending projectional tract systems arising from neurons in the cerebral motor cortex, e.g., the pyramidal motor system. One is the corticospinal tract that begins to form during embryonic week 9, reaching its outer limits by week 29. Fibers from the projectional neurons located in the cerebral cortex, form fiber tracts and pass from the cerebral cortex to the other parts of the CNS (Majovski & Jacques, 1982; Bear, 1986; Nauta, 1986a,b; Carpenter, 1978; Kahle et al., 1978). Axons of the pyramidal neurons pass from each cerebral hemisphere to form the corticospinal tract portion of each internal capsule situated in the basal ganglia. Farther on down, these two corticospinal tracts pass through the mesencephalon as part of the cerebral peduncles through the lower portion of the medulla oblongata. It is at the level of the medulla oblongata that most of the fibers of the tracts will decussate across the midline to pass down to the opposite side of the spinal medulla and become the lateral corticospinal tract. The uncrossed fibers, which remain ipsilateral, make up the ventral corticospinal tract and eventually cross to the opposite side at the lower levels of the spinal medulla. It is estimated that approximately 30% of the corticospinal tract fibers remain uncrossed, and that 30% are involved in decussation. Axons of the pyramidal motor control neurons of the cerebral cortex synapse with ventral gray column motor neurons (Carpenter, 1978; Arey, 1974). Functionally, the cerebral hemisphere on one side, from the decussation pattern manifested, exerts voluntary motor control over the opposite side of the body and also receives sensory inputs from the opposite side of the body via fibers that are crossed to enter each thalamus (Lund, 1978; K. L. Moore, 1977a; Crelin, 1974). Axons, whether myelinated or unmyelinated, become surrounded by glial cells known as oligodendrocytes. Within the CNs·, commissural, projectional, somatic, associational, and autonomic motor neurons become encapsulated by parts of other cells. The only exceptions are the boutons at synapses and nodes such as those of Ranvier. From the PNS, the neurons become completely encapsulated by parts of other cells, except at the terminal endings and at the nodes of Ranvier (Kahle et al., 1978). Unmyelinated axons are those that are surrounded (sheathed) by parts of either oligodendrocytes (those within the CNS) or neurilemmal cells
(those peripheral to the CNS). In contrast, myelinated axons are those that are sheathed by numerous layers of the cell membranes of either oligodendrocytes or neurilemmal cells. Major differences, neurohistologically, do exist between the myelin sheath formed by each of these. Oligodendrocytes and neurilemmal cells form myelin sheaths by similar processes.
Myelin Sheath Formation. Sheath cells become wrapped around the axon many times, with the sheath cell, the axon, or both, causing the spiraling motion. Penelope's Web, as it is known anatomically, is spun around the myelin sheath. Fiber tracts begin to function maturely at the time they are covered with myelin. The process of myelination in the human brain begins some three months postfertilization. However, at birth, only a few areas of the brain and tract systems are completely myelinated, e.g., brain stem centers serving subcortical functions such as certain primitive reflexes. As the wrapping process occ~rs around the axon, the cytoplasm of the sheath cell retracts such that it is extruded so that the two layers of plasma membrane of the sheath cell fuse together. Myelin is actually formed by numerous fused layers oflipoprotein membrane, and axons covered with fresh myelin assume a white, glistening appearance due to the lipid content. Tracts of the white, myelinated axons make up the majority of white matter of the nervous system. The majority of preganglionic svmpathetic axons are myelinated and are responsible for the appearance of the white ramus communicans. In contrast, the majority of postganglionic sympathetic axons are unmyelinated. Myelination is a process closely associated with the development of the functional capacity of neurons. One of its chief characteristics is the promotion of impulse conduction. Unmyelinated neurons tend to have a low conduction velocity and show fatigue earlier, whereas myelinated neurons fire rapidly and have long periods of activity before fatiguing occurs. Neurons that are capable of rapid transmission of impulses become fully functional at about the time their axons are completely insulated with myelin (Moore, 1975, 1977a; Crelin, 1974; Lemire et al., 1975). Formation of myelin in the spinal medulla begins during the middle of fetal life but is not completed until puberty. The last spinal tracts to be myelinated are the descending motor tracts, such as the corticospinal (pyramidal) and the tectospinal tracts.
HIGHER CORTICAL FUNCTIONS IN CHILDREN
49
These become myelinated during the first two leasing hormones that pass to the endocrine cells of postnatal years. At birth, only a few areas of the 15 the anterior lobe of the hypophysis through the blood dissectable descending tracts are completely mye- vessels and regulate endocrine hormone secretion. linated (Yakovlev & Lecours, 1967; Kahle et al., Thus, the sympathetic and parasympathetic compo1978; Crelin, 1974). In the human brain, myelination nents of the PNS are under the regulation of the neucontinues into the second decade of life and perhaps roendocrine system via the pituitary gland (Kuffler & even beyond that (Yakov1ev, 1962; Yakovlev & Nicholls, 1977; Snyder, 1980; Kaplan, Grumbach, & Aubert, 1976; Daughaday, 1974; Reichlin, 1974). Lecours, 1967). Motor neurons of the cranial nerves show myeIn the region of the substantia nigra, the linization patterns before their sensory counterparts. dopaminergic nigrostriatal system affects motor balOptic nerve fibers begin to show myelinization at ance and affectual response. This site also is implibirth; this will be completed by the end of the second cated in the etiology of schizophrenia and parkinweek. sonism (Majovski, Jacques, Hartz, & Fogwell, Axons of neurons of the cerebral hemispheres 1981). are among the last to become myelinated, beginning Serotonin (5-hydroxytryptarnine) molecules are around birth. At first, only the axons of cortical neu- found in the raphe nuclei of the brain stem. The pathrons of the olfactory, optic, and acoustic areas are ways that originate from them are distributed in a myelinated, followed by those arising from cell manner similar to those for adrenergic neurons. bodies in the somesthetic and motor cortices. Thus, Serotonin produces both inhibition of neuronal acfibers that become myelinated after birth are those of tivity and depression of behavioral activity in the the projectional, commissural, and associational mature brain (Smith & Sweet, 1978b). axons of the cerebral hemispheres. Myelination of -y-Aminobutyric acid (GABA) is a transmitter axons of the associational cortices of the cerebral substance released by inhibitory interneurons, as cortex continues into adulthood (R. Y. Moore, well as by cerebellar Purkinje cells. High concentra1977b; Carpenter, 1978; Crelin, 1973, 1974). tions of GABA are present in the striatonigral pathway within the substantia nigra. GABA is believed to Neurotransmitters and Neurohormones. be exclusively inhibitory (Majovski et al .. 1981). Neurons secrete specific transmitter substances or Acetylcholine (ACh) is released at both excitneurohormones at the axonal terminal endings (bou- atory terminals in the sensorimotor cortex, as well as tons). Boutons are the sites where information is visual cortex, and inhibitory terminals such as the transferred from one neuron to the next and where olivocochlear bundle. In addition, it has been found electrochemical changes in the release properties of in the nucleus basalis of Meynert (Dunn, 1980). the presynaptic terminals take place. Neurohormones Monoamine transmitter substances commonly attach to the membrane of the cell on which the axon produce inhibition of neuronal activity. When this terminates and induce internal changes in that cell. effect is exerted on inhibitory neurons, the net result Neurotransmitter substances serve to either stimulate is often facilitation via noradrenaline. Such a mechaor inhibit the secretory process. If the secretory pro- nism may account for the behavioral arousal process is stimulated, physiochemical changes occur duced by the catecholamines (noradrenaline and within the cell that are intimately related to the dopamine), which are believed to be involved in the changes in the permeability of the cell's membrane facilitation of inhibition (Bloom, 1973, 1979; Smith (Lehninger, 1968; Zucker & Lando, 1986). & Sweet, 1978b; Majovski et al., 1981). Neurohistologically, many neurons that become It has been theorized that dopamine (DA) and highly specialized evolved from glandular cells. Cer- other peptides may play a role in the mechanisms of tain cells of the body are structurally and functionally memory. A large body of evidence implicates pituiintermediate to typical endocrine cells and neurons. tary hormones, particularly adrenocorticotropic horThese cells possess axon terminations in the posterior mone (ACTH), melanocyte-stimulating hormone lobe of the hypophysis (pituitary gland) that are rich (MSH), and vasopressin, in learning. The action of in both cytoplasmic material and hormones. These ACTH, MSH, and vasopressin may improve memohormones are produced by the neurons and pass ry processes by modifying motivational and attenalong the axons to the hypophysis, where they are tional factors. ACTH may act to stimulate the metabstored for release as required. After being released, olism of DA and/or norepinephrine (NE). It is also these posterior lobe hormones pass to the responsive thought by some that vasopressin may act to affect tissues of the body through the vascular system. catecholamine metabolism in a rather complex manOther similar hypothalamic cells secrete so-called re- ner. In this regard, arousal has been shown to be
50
CHAPTER3
associated with the activation of central NE systems and the release of hormones such as ACTH, vasopressin, and glucocorticoids. In the neonate's brain, modulators (neuropeptides) may play a role in terms of the mode of information storage and not a direct effect on information stored at all. Even though not completely understood, it is currently thought that catecholaminergic as well as neurochormonal factors apparently do play some type of role in the storage of memory. Some have suggested that opiate receptors may even play a role in the filtering of sensory stimuli at the cortical level involved with emotion-induced selective attention. This proposition offers the possibility that a neural mechanism exists whereby the limbic-mediated emotional states, essentially for individual and species survival, may influence which sensory stimuli are selected for attention (Bear, 1986; Kety, 1970; Dunn, 1976). By implication, endogenous opiates may exert progressively greater influences at higher levels of sensory information processing in the cortex. Whether this holds true in the neonate's brain over the course of the first several years of development is unknown. However, sensory stimuli at the cortical level may play a role in selective attention, which has a significant bearing on the processes of cognition and learning mechanisms.
Summary Phylogenetically, the brain can be considered as a blueprint of nature's original design, for which we do not have a complete set of plans. To explain the brain's architectural plan today is like looking at a house to which rooms have been added, but without having access to the original plan. The clues that have been left behind concerning the functional aspects of the additions to the phylogenetic blueprint of the brain are as follows. At the midbrain level, anencephalic children will be able to live for only a few days or weeks, but they can exhibit laughing and crying responses. Another clue is the phenomenon of supersegmental control over lower processes of the human nervous system which occurs at the midbrain level. A major consideration as to the human nervous system's functional neuroanatomy is one that has been addressed above in some detailontogenetic development. Another approach to understanding the functional neuroanatomy of the developing brain is in terms of its serving as a communication system where there is biological (neurochemical) information transfer. The principal components involved in this
communication network are axonal conduction, synaptic transmission, and local cell processes. A further approach is to view the spinal cord as a model of neural organization. Chief characteristics are segmental and suprasegmental reflexes. The functional anatomy of sensation is also important for understanding the human CNS and its development with these key features: receptors and transduction processes; pain systems; discriminative sensory systems; and systems for automatic adjustment. Other aspects are the descending motor control systems such as commissural, associational, and projectional fiber systems; basal ganglia: extrapyramidal mechanisms; and brain stem centers of control. The cerebellum is another system crucial in the functional understanding of the developing brain's neuroanatomy. It is involved in the modulation of motor and sensory mechanisms and may even have a role in decision-making responses as well as emotional expression. Another major system is the brain stem system and internal states. These play a prominent role in the regulation of sleep, wakefulness, emotional affect, pain, pleasure, and suppression of pain. Yet another key is the various support systems and the internal milieu, e.g., the autonomic and endocrinological mechanisms that influence neuronal activity and ultimately human behavior. Finally, the developing brain's higher cortical functions are the ultimate expression of human mental information processing consisting of a number of psychological processes (Luria, 1973a,b, 1980).
Factors Affecting Normal Brain Development and Higher Cortical Functions The development of higher cortical pathways used to the best advantage of the neonate rests on improvement of conscious and unconscious programs of behavior. The formation of categories and higher mental processes involved in concept acquisition is limited by unmatured neural organizational patterns that are affected by several variables: timing, nutrition, environment, and genetics, among other factors. Knowledge of these factors with regard to normal infant development of mental abilities and cognitive processes is an essential first step to a clinical exercise in diagnosis and establishment of remediation programs for neuropsychological deficits in infants and young children (Gaddes, 1980;
HIGHER CORTICAL FUNCTIONS IN CHILDREN
Rourke et at., 1983; Spreen, Tupper, Risser, Tuokko, & Edgell, 1984).
General Considerations in Human Brain Growth At the time of adult maturation, the human brain accounts for approximately 2-3% of the body's weight, but utilizes approximately 20% of cardiac output (oxygen consumption), and 70% of the body's glucose, which it is almost exclusively dependent upon in the oxidative phosophorytation process (Davison & Dobbing, 1968; Lemire et at., 1975; Nilsson, 1978; Humphrey, 1978). Even though the body may be starving, the brain receives a disproportionate share of nutrients, thriving almost exclusively on oxygen and glucose. Because of these metabolic requirements and dependence on the above two principal constituents, and the fact that the human neocortex is poorly vascularized, states of anoxia, hypoxia, and hypoglycemia can seriously damage the brain's normal functions, especially in early infancy. A child may suffer from extreme malnutrition and weigh only half of his or her normal weight and yet the brain may only be 15% underweight. Ninety percent of neurons are found in the brain. It is an electrochemical network of some I 0 billion nerve cells, all present at birth, which regulate sensory-perceptual, motor, language, and other functions, as well as the higher psychological processes that we define as human behavior. It works whether we are asleep or not. For example, breathing alone requires the complex coordination of some 90 muscles that must be regulated precisely in order to take just one breath. Processing some 100 bits of information simultaneously, the human brain distinguishes between reality, memory, and fantasy as it matures. As it develops in terms of its inhibitory capacity to effect control over behavior, it regulates the various human drives and emotions that it spawns throughout its developmental course. In some brain regions, 107 cell bodies can fit into a cubic inch; each one of them can be connected to as many as 60 X I 0 3 and none of them exactly alike (Szentagothai & Arbib, 1974; Szentagothai, 1975, 1978; Scheibel & Scheibel, 1973). Some 50 x I0 3 connections are being formed each minute in the visual system at approximately 40 days postconception. Experiments have revealed that at this time, the eyes can respond to a camera's light flash (R. Y. Moore, 1977b; Lund, 1978; Crelin, 1973).
51
The neuroelectrochemical pulses reaching the inner ear pass through at least four increasingly elaborate stages of analysis and refinement before any sound reaches conscious awareness for perceptual discrimination. Insults to the brain early in gestation may arrest development with resulting gross malformations, such as anencephaly (failure to develop a prosencephalic outpouching); holoprosencephaly (failure of the forebrain to separate and develop normal commissures); and lissencephaly (failure of fissures to occur, resulting in a smooth surface of the brain) (Smith, 1976; Langman, 1975; Lemire et al., 1975). Nutritional requirements, as well as effects of malnutrition (especially protein deficiency, which can affect brain weight and result in cognitive deficiencies later in life), are critical from postnatal month 6 to month 18. Throughout childhood and well into adult life, brain function increases tremendously, despite a brain weight gain during this time that remains relatively small (350 to 400 g) (Dodge, Prensky, & Feigin, 1975; Hamilton & Mossman, 1974).
Mass Growth of the Brain Brain weight has been used as a quantitative index of brain growth, as well as a traditional indicator of quantitative aspects of CNS development. The brain of the newborn weighs approximately 300 to 350 g. By 12 months, the weight has more than doubled since birth and is approximately two-thirds that of the adult. The average weight of the adult brain is from 1300 to 1500 g and is related to body size. That is, larger people usually have heavier brains, although there is no connection between brain weight and intelligence. Growth of the CNS during early fetal life reflects an increase in volume in the first trimester from 4% to 16%, compared to only 42% at birth. The brain's weight is 21% that of the body at the sixth fetal month, 15% at birth, and approximately 3% at adulthood (Dodge et at., 1975; Yakovlev, 1962; Yakovlev & Lecours, 1967; Hamilton & Mossman, 1974). The increase in volume of the cerebral hemispheres is slow and steady between fetal months 2 and 6 but rapidly accelerates thereafter. The postnatal growth of the cerebral hemispheres is due mainly to an increase in myelination. The brain stem grows most rapidly between fetal months 2 and 6 and less rapidly thereafter. The cerebellum grows slowly between fetal months 2 and 5, followed by an excep-
52
CHAPTER3
tionally rapid increase in volume commencing at fetal month 6 and continuing until postnatal month 6. The weight of the brain more than doubles during the first nine postnatal months and reaches 90% of its adult weight by the sixth year of life. The hemispheric surface of the brain more than doubles during postnatal growth to reach an adult value of approximately 1600 m2 . This growth is accompanied by an increase in the size and number of gyri, so that the intrasulcal ponion ofthe adult cortex is about the same as that in the newborn. The adult cortical surface area is reached by the second year of life. The entire hemispheric surface is gyrated by approximately fetal week 32, despite the fact that these gyri are less numerous than in the adult brain. Normal thickness of brain tissue between the ventricles and the cortical surface is approximately 4.5 em. There are many reports in the literature that people of normal or above-average intelligence, on cr scans, have only a thin layer of mantle between ventricles and cortical surfaces measuring 1mm versus the normal4.5 em. Several such cases have been documented in the medical literature (Lorber, 1980).
White Matter Development Hemispheric white matter develops slower than cortical gray matter during gestation. Postnatally, white matter will continue to develop long after gray matter has reached a specified volume. The growth of the cortex subsides by the second year of life. That of hemispheric white matter continues even through the second decade due to accumulation of myelinated fibers with their increased diameters. Myelination is closely associated with development of the functional capacity of neurons; they fire more rapidly and have a longer refractory period. Different fiber tracts myelinate at different developmental periods. The component populations of a given tract system may differ as to the timing of myelination (Davison & Peter, 1970; Dobbing, 1975; Dobbing & Sands, 1973). Myelination of tracts typically follows in a caudal-cranial sequence. The schedule for myelination was first elaborated by Fleschig in 1876. Around the fourth fetal month, myelinated fibers appear in the ventral and dorsal spinal roots. Last to receive this investment are the associational fibers of higher cortical centers, e.g., the cerebral cortex's thalami. Some tracts are not fully myelinated until several years after birth (Dobbing & Sands, 1970; Dobbing & Smart, 1974; Crelin, 1973, 1974; Hamilton & Mossman, 1974).
Sensorimotor Functions and the Appearance of Neurological Reflexes Motor control behavior of the newborn is largely under the control of the spinal cord and medulla, whereas motor control in adults occurs at different levels of the nervous system. At birth, several neurological responses are present. The appearance and disappearance of neurological (primitive reflex) signs are essentially transient mechanisms either subserving life-sustaining functions or forming preliminary patterns of future voluntary activity, e.g., the stepping reaction that precedes voluntary step-walking. The most important responses that appear and disappear in early postnatal life are as follows: reflexes of position and movement, e.g., the Moro reflex, asymmetric tonic neck reflex, neck righting reflex, Landau response, palmar grasp reflex, abductor spread of knee jerk, plantar grasp reflex, Babinski response, and parachute reaction. The Landau response and the neck righting reflex are the first to appear around the time of birth and the last to disappear at one to two years (Siqueland & Lipsitt, 1966). Reflexes to sound that appear at the time of birth include the blinking response and the turning response. The reflexes of vision include the following: blinking to threat; horizontal following; vertical following; opticokinetic nystagmus; postrotational nystagmus; lid closure to light; and macular light reflex (Franz, 1963; Barnet, 1966). The feeding reflexes include: rooting response-awake, rooting responseasleep, and sucking response, all of which appear at the time of birth; the last to disappear is the sucking response at approximately 12 months. The level of neural functioning during the neonatal period can be determined only with great approximation. Some of these reflexes will drop out of an infant's behavioral repertoire at around postnatal month 3 or 4. This is presumably due to increasing cortical inhibition of lower centers in the brain. Reflex arcs exist below the cortical level and before integration occurs between subcortical and cortical structures, stimulation of the infant elicits an involuntary, subcortically mediated reflex response. As the maturing cortical centers become integrated with subcortical areas, primitive reflex behavior then becomes inhibited. The fact that most of the activity of a normal newborn can be observed in an anencephalic infant possessing only a brain stem and certain components of the basal ganglia, indicates that the cerebral cortex participates very little, if at all, in the
HIGHER CORTICAL FUNCTIONS IN CHILDREN
function of the CNS during this stage of life. Children with these morphogenetic anomalies do not survive usually more than two months, but while alive, they do show reflex development similar to the patterns of normal infants of the same age (Smith, 1976; K. L. Moore, 1977a; Davison & Dobbing, 1968; Humphrey, 1964; Schulte, 1974; Dodgson, 1962). Concerning development of the cerebral cortex, upper, central, and hindmost regions tend to mature early, such as those concerned with bodily sensations involved in the control of movements, hearing, and vision. Frontal and lower sides (frontal lobe area) in the region of the temporal lobe mature later. In the motor strip area, for example, parts controlling trunk and arm movements appear in advance of parts that will control leg and finger movements.
53
Stratification within the cortex proceeds according to a definite plan in terms of neuronal organization. Neurons with connections in certain parts of the brain differ from others with respect to the cerebral hemispheres. For example, in the posterior portion of the cerebral cortex, the development of the layers proceeds according to a clear-cut, typically six-layer plan with a distinct layer (IV) as the main site of termination of afferent impulses from the subcortical divisions of perceptual analyzer regions. There is, then, a difference between the structural differentiation of the anterior and posterior divisions of the cortex in early ontogenesis. This may account for slowly developing cortical-cortical interconnections essential to the final differentiation of the outer three layers in middle and later childhood (Luria, 1969b; Warren & Akert, 1964; Stuss & Benson, 1986).
Prematurity and Low Birth Weight The fact that babies who are born prematurely but go on to develop without complications suggests that only when the brain is ready, and not before, will it start to develop in earnest. As a rule of thumb, infants who are 4 weeks preterm generally will, after 12 months, "catch up." Judging by the immature state of the cerebral cortex, it can be speculated that the latter plays a minor role in the life of the child at the time of birth. The infant may be completely dependent on the inner regions of the brain and especially its subcortical structures. Not until postnatal month 3 does the roof of the brain begin to intervene in a dominant manner, in the control of arm movements with the first signs of coordination of hand and eye movements (Lemire et al., 1975; Davies & Stewart, 1975; Spreen et al., 1984; Levene & Dubowitz, 1982).
Nutrition and Malnutrition
Human fetal nutrition depends largely on the size and functional capacity of the trophoblasts in the placenta and the villous surface area through which the exchange of nutrients takes place. In the placenta, three phases of growth occur and by about 34 to 36 weeks of gestation, cell division ceases while weight and protein increase nearly until full term. Placentas from infants who had experienced intrauterine growth failure tend to show fewer cells and an increased RNA/DNA ratio compared to control placentas. Study of placentas from malnourished populations in the world has confirmed that fewer cells are present than in normal placentas. Maternal malnutrition, vascular insufficiency, and abnormal influences with regard to intrauterine growth contingent upon the placenta, will curtail cell division in the placenta. During intrauterine life, all fetal organs are in a hyperplastic phase of growth and probably at no other Frontal Lobe Maturation time is the human organism more susceptible to nuThe slowly maturing frontal lobes appear to be tritional stresses. Fetal malnutrition can result from required for a young child to respond correctly to any number of causes, e.g., reduced nutrients within verbal instructions. Luria pointed out that in young the maternal circulation, faulty placental transport of children, the immaturity of brain structures may be specific nutrients, and abnormal maternal circulasuch that it is physically impossible for a child not to tion. Deficiencies in specific nutrients are found to do what he or she is told not to do (Luria, 1960, affect fetal brain development. Reduction in either 1961). This situation bears close resemblance to that total calorie or total protein intake by the fetus can of adults with damage to the frontal lobes. Not until lead to retarded growth (Cravioto & Arrieta, 1979, the front of the brain is well developed can a child 1983). become capable of obeying certain verbal instrucStudies of the effects of malnutrition on the tions. It is usually not until 3 Vz to 4 years of age that growth of human brain cells have been limited. In children are capable of learning to carry out a com- infants who died as a result of failure to thrive (marplex program of actions deliberately, in accordance asmus) during the first year of life, protein, total with unrepeated verbal instructions. RNA, total cholesterol, total phospholipid, and total
54
CHAPTER3
DNA content were proportionally reduced. The rate of DNA synthesis was reduced and cell division showed curtailment in terms of reduction of the number of cells. These factors suggest that if malnutrition persists beyond 8 months postnatally, not only the cell numbers but also size is reduced. Malnutrition in humans tends to reduce the rate of cell division in all brain areas. Early malnutrition affects cell division, myelination, and vulnerability as to the maximum rate of synthesis of DNA. Nutritional deprivation before the age of 18 months causes permanent intellectual impairment.
Cerebral Oxygen Consumption and Blood Flow In the developing brain of the newborn, oxygen consumption is relatively low but gradually increases with maturation. That the neonatal brain is able to tolerate states of anoxia is suggested by the low cerebral oxygen consumption at birth. This ability to tolerate anoxic conditions may also be related to the brain's dependence (prior to birth) on anaerobic glycolysis as an energy source. The level of enzymes needed for aerobic glycolysis just prior to birth shows an increase as the brain's metabolism begins to change from anaerobic to aerobic. Cerebral blood flow (CBF) is relatively low in the newborn but increases with age to a maximum of . . bo 3 3 00 4 105 em /1 g _per ":'mute startmg at a ut or ~ears_of age. J?1srupt1on ~f C~F, oxygen con~umptlon, ~~ tum wdl affect OXIdative phosphorylation ~s the bram dev;lops. The av~rage CBF for an adult IS 45 to 55 cm/100 g ~r mt~ute. In adults, a val~e below2~cm ~~~gwdlbegmaprocesso~meta~hc degradation ~1thm neuronal _mitochondria, !eadmg to cell dea~h 10 a matter of mmute~. Ano~a!tes that pos~ a m~J~r ~hreat to th~ develo~mg bra~n s ?Ietabohc e9uthbnu"?- at. the tt~e ~f btrth are. ~nnatal asphyxta; hypoxtc-tschemtc btrth-~l~ted eptsode_s; and extended ?ecrease of CBF that ts Interrupted~~ terms of cardtac output (Afifi .& Bergman, 1980 • Roberts,. 1986; . Phelps, Mazzlotta, & Schelbert, 1986;_ Fmkelstem, Alpe~, & Ac~erman,_ 1980: Mazz10tta, Phelps, & Miller, 1981, Mazzlotta & Phelps, 1985).
EEG Development Human fetal EEG activity has been recorded as early as day 43 (Bemstine, Borkowski, & Price, 1955). EEG activity in the fetus evolves in a rapid and specific manner. In a 5-month-old fetus, cerebral
activity lacks organization, rhythmicity, and regularity. At 6 months, organization emerges; rhythmic theta activity (4-to-6 Hz) appears in flurries lasting about two seconds. In the seventh month, activity tends to become continuous at about l Hz with voltages ranging from approximately 100 to 200 microvolts. This slower activity is interspersed with faster frequencies around the seventh and eighth months; differences between active and quiet sleeping states become obvious (Ellingson, 1964; Lindsley, 1939). Bursts in the electrical activity pattern at 6 to 7 months are associated with an increase in enzymatic activity in the brain. The major difference between the immature and maturing infant's brain is the definite change in EEG between periods of wakefulness and sleep. As a generalization, the amount of quiet sleep increases with maturation. In the eighth month, during active sleep, fast waves of approximately 2-3 Hz appear with no localization and often imposed on low-voltage faster waves, and become dominant. Frequency measurements of photically evoked responses recorded from the occipital area show a progressive shortening with maturation. Responses to auditory stimuli also show clear differences in wave form, amplitude, and latency of wave components with maturation (Ellingson, 1964; Hagne, 1972). Rate of maturation of brain electrical activity decreases with age. After birth, maturity of changes occurs within the first 3 years. Fewer alterations are · 1 f noted between 3 and 8 years o age. Only mimma differences are apparent from 8 to 21 years. Due to the continually changing aspects of the EEG during . childhood, problems of interpretation are more numerous than for adult recordings. Newborns have only brief periods of wakefulness with eyes open. Most EEG recordings are carried out during sleep. Two well-defined types of sleep are noted in 32- to 39-week premature and in full-term neonates: active sleep and quiet sleep. Early in the newborn's life, 2- to 4-Hz waves are present, which will be replaced by those of 4 to 7 Hz at approximately 5 years of age. Faster activity in the occipital regions of the brain begins to dominate (8 to 12Hz), and alpha rhythm of the mature brain starts to emerge. Occipital alpha rhythm changes rapidly in the first year from a 3- to 4-Hz rhythm to twice that frequency by the end of the first year. Changes in frequency most likely result from brain growth and myelination. Changes in the EEG 's alpha frequency with age have important behavioral consequences; for example, periods of rapid change in EEG activity can help to identify critical periods for behavioral change. One of these periods occurs at the end of the third
HIGHER CORTICAL FUNCTIONS IN CHILDREN
month of life when the alpha rhythm first appears. Another important period extends from the end of the first to the completion of the second year of life when the alpha rhythm attains adult values. Alpha frequency can be construed as an index of brain maturation and thus a reliable marker of reference points for observing critical behavioral changes. Lindsley and colleagues (1939, 1974) proposed that the onset of organized rhythmic occipital activity reflects a significant change in cortical organization and may mark a point at which the infant progresses from a subcortical to a cortical level of functioning. It is suspected that visual behavior in early infancy is processed by subcortical mechanisms with the cortex usually taking over in earnest at about the time rhythmic occipital EEG patterns begin to emerge. There are numerous variations in the EEG pattern that are considered abnormal. Significant alterations can be brought about by changes in level of consciousness. Recordings are usually done in multiple states: wake, drowsy, sleep, and so on. Abnormal EEG findings should be considered only an aid in diagnosis and treatment of CNS disorders. Some abnormalities strongly suggest specific types of CNS pathology. However, there are only a few well-established abnormal patterns that correlate well with the clinical state of an infant: hypsarrhythmic pattern (high-voltage, arrhythmic activity usually associated clinically with massive myoclonic epilepsy); prolonged general bursts of regular 2- to 3-Hz spikeand-wave patterns (associated with petit mal or absence); continuous generalized spiking (sometimes accompanied by a major convulsion, generalized tonic-clonic seizure disorder); and a flat recording containing no distinguishable wave forms {usually consistent with brain death).
Speech and Oral Communication Development Speech (oral communication) is a basic tool in interpersonal relationships and serves as a key indicator of developmental level in the early years of life. Any interference with speech development or subsequent distortion of speech may have a profound effect on social, vocational, and interpersonal development in the child. It is estimated that 5-7% of all children of school age require special help with regard to speech (Neville, 1984; Luria, 1969c). An understanding of the neuropsychological development of the speech process is needed prior to any professional advice or instruction with regard to the causes and treatment of speech-related disorders.
55
Such diverse elements as neurological maturation, acquisition of fine muscle control, and development of symbolic formulation abilities crucial to development of cognitive processing of information are all related to speech. It is these neuroanatomical, neuropsychological, and neurophysiological aspects that provide the content and control from which speech is created. In later stages, especially in acquisition of words and sentences, contributions oflearned behavior emerge (Darley & Fay, 1980; Siegel, 1979: DeVilliers & DeVilliers, 1979; Lenneberg, 1964, 1967; Milner, 1976). In the early prespeech stages, any division of speech development into discrete stages can be misleading. Generally all basic behaviors that appear at different ages function throughout childhood and into adulthood. After the prespeech stage from birth to 3 months, speech starts with the reflex stage. The birth cry is often considered the beginning of speech, but any true expression is doubtful. Shortly after birth, reflex crying appears in response to discomfort or fear. Cries often vary and become differentiated from other noises, such as gurgling, sucking, cooing, and laughing. From 3 to 12 months, the babbling stage occurs. Basic changes in vocal expression are observed in the rapid increase in the number and varieties of sounds. As a child develops early awareness of vocalizations and moves into a period of vocal exploration, practice and repetition are hallmarks. A child at this stage begins to modify imitations and is aware that he/she is "imitating" oneself. In many cases, early imitations of others result from the parents repeating sounds that the child has produced. Later, as the parents initiate imitative responses with familiar new sounds, the basis for learning speech is discernible. In the earlier phase, all kinds of sounds are repeated. Sounds that approximate language are usually selected based on the most intense reinforcement derived by the child. Another step in the development of speech emerges when the child integrates babblings and imitations into sequential patterns that sound more and more like true speech (at about 12 months). Individual sounds tend to be reasonably accurate, vocal quality approaches that of voices heard, and sounds are grouped into nonsense forms and even phrases. Occasionally, what appear to be recognizable words are heard by adults, but it is doubtful that they represent meaningful speech. True speech stages begin to emerge. Characteristic features of the stages include motor control of breathing, phonation, and articulation. Ability to echo, complex mechanical patterning of speech, and
56
CHAPTER3
other skills are practiced to assist in the leap to a distinctly different level of function. Complementary interlanguage is a prerequisite as recognition of familiar objects in the environment requires inner Language. As the child develops into later years, this type of language would be what adults use to ''talk to oneself" (Molfese, Freeman, & Palermo, 1975; Segalowitz & Chapman, 1980). The child later becomes aware that certain sounds spoken by the parents stand for objects (auditory receptive language); inner language and auditory receptive language precede actual production of meaningful words (Luria, 1961, 1982; Luria & Yudovich, 1959). The word stage emerges from approximately 11 to 24 months in which true first words are usually names of concrete objects. Sentences start to appear at approximately 18 to 36 months. With the advent of an increasing vocabulary containing a variety of representations for objects, people and actions in the child's environment provide the opportunity for the child to discover more complex meanings and verbal reinforcement. This is a variable period for the beginning of sentences and there may be periods of little progress after sentences are first used. Complex speech takes place from approximately 24 months up to 7 years. This period in the normal child is characterized by an impressive array of developments. All parts of speech increase at a rapid rate. The word "no" will cover a vast number of situations, behaviorally, and provide a distinct measure of control over individuals in the child's environment. Extrapolations of grammatical structures cause difficulty with some irregular verbs, but the process of such abstracting indicates an increasing complexity of symbolic thinking. Categories are learned, for example, male and female, dog and cat, but also learned is that matching reality to a symbol requires relative concepts as well as absolutes. This quality of assigning symbol to meaning is usually lacking in severely autistic children in their development of speech (DeVilliers & DeVilliers, 1979; Siegel, 1979; Luria, 1960; Wertsch, 1979). As the expansion of potential for expression continues to develop, not only does oral communication serve as an efficient tool for exploring and understanding the people and the world that surround the child, it also becomes a means for controlling and manipulating the environment. It serves to provide an extension in a variety of emotional expressions that goes beyond older but still used methods such as tensing muscles, throwing a tantrum, crying, and engaging in uncontrolled movements (Darling & Fay, 1980). During the child's first 12 to 18 months, there
occurs a transition from visual representation to a verbalization mode in which the child's ability to participate in sequences of interpersonal exchanges via speech is evident. Children use language perhaps as a medium to communicate messages that become more syntactically refined over age. It is the process of assigning a memory code to a symbolic form that raises the question of "how does a child retain what is learned?'' and ''how are coding errors corrected or adaptations to a changing environment effected?'' (Vygotsky, 1980; Grossberg, 1980). It would be a mistake to view speech development as an isolated process. It is integrally linked with physical, psychological, and sociological progress. Disruptions or distortions in any of these areas may have serious repercussions, and it is particularly important that speech be developed during early childhood, since there is compelling evidence suggesting that lack of developmental opportunity or severe inhibitory factors may have serious and permanent effects on Linguistic-symbolic-intellectual development. The child may never become a completely functional human being in the developmental sense if speech is arrested or not developed by 7 years of age (Curtiss, 1979). Professionals and parents involved in a child's life have a significant responsibility to detect aberrant patterns during development. Symptoms of possible difficulty may be noted in several ways during the first year of life. In the prespeech period, the most critical characteristic is the lack of progressive change in the nature of babbling or, conversely, deterioration of vocalization to that representing earlier stages of development. At ages when response to and imitation of sounds of others may be expected, any unusual delay may suggest difficulty in learning to talk. Failure to engage in jargon conversation is likewise an important indicator of problems. It is only when more advanced behaviors have failed to materialize that one becomes concerned in terms of delay. Continuation of earlier forms of vocalizations during later months of the prespeech period are not evidence that progress is delayed. As long as speech is inclusive enough for useful communication and can be reasonably well understood by strangers by 4 years of age, concern over development of a serious problem lessens. In addition to articulatory aspects, there is the possibility, during this period of development, of difficulties in fluency. For example, all children have numerous dysfluencies in speech. At certain times and under certain circumstances, these are more frequent and noticeable. The distinction between normal dysfluencies and stuttering is not always easy to make, partie-
HIGHER CORTICAL FUNCTIONS IN CHILDREN
ularly as there is so much variability from individual to individual. In general, however, major concern is not necessary unless there are signs of struggling behavior, tensions, anxiety, or reactions to specific dysfluencies in the child. Persistent rudimentary sentences may indicate a broader developmental delay but it must be emphasized that the total environmental background must be considered before assuming that a child has basic inadequacies. Recognition of a speech problem versus undue concern over what are essentially individual differences in oral communication patterns can affect a child's cognitive, social, and emotional development. In this sense, appropriate evaluation can be reached by integration of several factors: sequence and nature of speech skills: relationship of these to the child's experience and development; and the acculturation process in which others in the child's environment can influence these developments. One basic principle that seems to override all stages should be the stimulation, encouragement, and provision of opportunities for speech without demand or punishment. In circumstances where the severe form of speech pathology or abnormalities are not present, if speech is rewarding rather than punishing, a child will talk when ready and able to do so. However, if speech sounds have not developed by 4 years of age, a serious disturbance is present. If there is no speech (or poorly established patterns) by 6 years, most likely there will be intellectual and socialization deficiencies of significant proportions (Pinker, 1984; DeVilliers & DeVilliers, 1979).
Acculturation Processes Brain growth and the acculturation process are inextricably bound in human development. Neural networks are being set and affected by specific experiences related to environmental events (Szentagothai, 1978). Acquisition of cognitive operations, which stem from a particular culture, is affecting elaboration, in part, of neural circuits, such as feedback and feedforward loops. These circuits are built during a period of acquisition and development of cognitive neural codes. Brain morphogenesis during a prolonged period of exposure to significant novel experiences can be expected to be modeled in accordance with ongoing experience (Goldman, 1975; Gottlieb, 1976a,b; Taylor, 1969). Data collected from studies of heredity and environment suggest that morphogenetic development of the brain's intellectual nature is attributable to both genetic and social environmental influences, the for-
57
mer having slightly greater effect than the latter (Oates, 1979; Plomin & Rowe, 1979). What this suggests is that several different cortical-cortical and cortical-subcortical systems are operative during the process of learning and information storage. What the infant senses, then, may be in part due to what is neurally "set" to sense or competent to sense via a selective attending process (Pribram, 1976; Siegel, 1979).
Postnatal Perceptual, Cognitive, and Motor Development Perception can be considered a multichannel process of visuomotor, auditory, sensorimotor, and other skills in which motor components of perceptual acts may be seen as a control process over sensory input mechanisms. Sensory input data contribute significantly to the foJlowing: perceptual processing of information; decision-making; readiness-to-act; and motor control commands, all the way from reflexes to the highest neocortical levels involving abstract thought. The significance of these processes as to the phenomenon of controllies in the success (or failure) ofthe infant's behavior. During the first stage (the first 2 years of life), an infant changes from a baby with little awareness of the environment to a child who is aware of the environment due to the developing perceptual systems and neurological processes. If development is normal, the child is capable of discriminating among the various environmental stimuli. During the second stage (2 to 5 years), preconceptual representation as described by developmental theorists such as Piaget and Bruner takes place in which the child develops pictorial (ideograph) images as symbols. The child also begins to advance in language competency. During the third stage (5 to 8 years), symbolic representation occurs in which the child becomes aware that he/she is not alone in the universe and begins to interact with several environmental forces that impinge on the child's development. The fourth stage (7 to 12 years) is characterized by operational thinking in which the child begins to recognize certain relationships between objects and appreciate their relative values, e.g., the concept of mass, size, distance, length, and time (Bower, 1977; Siegel, 1979; Mistretta & Bradley, 1978). Several attempts have been made to match developmental stages in cognition with defined structural changes in the brain (Epstein, 1978; Milner, 1976; Ploog, 1979; Vygotsky, 1974). Thus far, such efforts have had only very limited successes. Major
58
CHAPTER3
obstacles in trying to correlate behavioral, motor, and sensorimotor development in finite stages include nerve processes showing differentiation, synaptic process formation, dendritic arborization, and myelinated pathway development, to name a few. Actual alterations or modifications also can be due to metabolic factors that affect function: decreased CBF; oxygen use by cerebral tissues; glucose use by cerebral tissues; cerebral vascular alterations that can lead to ischemic, hypoxic, or anoxic episodes. As glucose and oxygen are the primary constituents providing the metabolic energy requirements of the brain, rates at which these substrates are used can provide a quantitative assessment of the level of neuronal function in the brain. Maturational changes in cerebral function in infants have been studied by quantitative methods using 2-deoxy-2-fluoro-o-glucose and positron emission tomography (Chugani & Phelps, 1986). Studies on infants at various times during development revealed significant changes in a progressive manner in local cerebral glucose use. Chugani and Phelps (1986) studied 5-week-old, 3-month-old, and older infants. Glucose metabolic activity increased in anatomical regions in agreement with behavioral, anatomical, and neurophysiological alterations that are known to occur in the first 12 months according to established patterns of infant development. Although it is not yet possible to specify which brain 11\echanisms are specifically involved in perceptual processing, it is commonly believed that many of the events are distributed throughout the brain. Perception represents functions drawing from systems at diverse anatomical sites, both in upper and in lower regions of the brain (Livingston, 1978; Majovski & Jacques, 1982; Bower, 1977). Connections conceivably involved with central information processing of context-dependent and context-free events in the environment may involve cortex-to-basal ganglia and frontolimbic pathways. Sensorimotor readiness, in part, is dependent ·on cognitive-spatial mapping properties thought to be carried out in the hippocampal formation and neocortical structures (Izquierdo, 1975; Nauta, 1986a,b; Liben, Patterson, & Newcombe, 1981). Memory underlies the highest functions of the brain, from multiplying two numbers to developing a sense of oneself. All memories come from the world outside of the mind. Whereas visual images leave shadows on the retina for less than a second, sounds taper off into echoes lasting no more than four seconds. A major question arises: how does a coded memory process, specific to certain environmental
events, differ from those that organize and lay down a long-term memory code? Structures implicated as crucial for forming new and enduring memories are the hippocampi. Speculation exists that the hippocampi and amygdala process potential memories and then pass them on to the cortex. It is at this level where thought, planning, decision-making, and other higher cortical functions take place and where sensory impressions leave their "traces" in neurons. How this process works is not understood. Only certain impressions are allowed to flow into the cortex and how others are kept out (inhibited) is a mystery. How neurons are ''tuned in'' to the ongoing events of the environment, such that they can abruptly change in firing patterns upon sensing the smell of food, the perception of fear or threat, the sight of something stimulating, and so on, is unresolved at the present time. A major anatomical site that acts as a "gatekeeper" in this process is thought to be the nucleus basalis of Meynert. One of the implicated chemical messengers in this process is acetylcholine, which has been found to aid neurons in the cortex to retain the imprint of information flowing to the cortex. The question as to how a cognitive code becomes established involves understanding "how" neurons record symbolization of the representational objects from the environment. How representations, symbolically, achieve distinctiveness as to properties of global consistency and stability (once encoded) in the brain remains one of the major challenges for researchers of infant cognitive development (Grossberg, 1980). Representation of environmental events receives continuous updating, in part, due to the thalamocortical processing of spatiotemporal events, such as formation of a spatiotemporal ''envelope'' of reality leading to consciousness. This latter explanation perhaps may account for how the brain, during development, is able to maintain an updated version of reality with global consistency. Lashley, Chow, and Semmes (1958) asserted that the core function of the CNS lies in the spatial.and temporal integration of perception and motor activity in order to provide refined adaptation of behavior. Some important clues related to how brain mechanisms operate in terms of sensory processing and an infant's behavioral output, come from a consideration of noncorrespondence with past experiences. The infant perhaps chiefly discovers through action that stored codes are connected after his or her behavior achieves success. Perhaps it is at that partie-
HIGHER CORTICAL FUNCTIONS IN CHILDREN
ular moment that perception itself is projected in a appropriate symbolic form in a predictive manner for the desired behaviors to follow. Behavior, in this sense, is essential to the shaping process of stored sensory information and not simply its goal. Memory storage might be deposited in the neural substrates of various brain centers that are accessed according to a given "contextlike" paging system such as might be similar to a library cataloging system. Delays in matching stored percepts to sensory input would then be experienced when the context is suddenly altered (Majovski & Jacques, J982). Some crucial steps considered operational in this process include: rapid matching stage; hypothesis formation; internal sorting from possibilities; and testing the selections made via behavioral acts. Siqueland and Lipsitt ( 1966) demonstrated that infants can exhibit learning during the first day of postnatal life. Head turning is a regular response in which hypothesis testing conceivably is occurring. Memory storage undoubtedly is taking place, together perhaps with suppression of interhemispheric transfer of memory codes. The suppression phenomenon introduced here has the potential effect of doubling the capacity of the neocortex for memory storage in its early stages. The anterior commissure of the corpus collosum, which interconnects the two temporal lobes, is believed to participate in memory storage bilaterally in a yet unspecified process. Perhaps the most plausible mechanism for early stimulus identification and refined feature selection involves that of lateral inhibition. In studies that have examined the relationship between abilities of infants and subsequent cognitive functioning, a substantial correspondence has been found between infant behaviors and cognitive and linguistic abilities in early childhood, despite rather low correlations of test scores and measurements from infant intelligence tests (Siegel, 1979). It has been suggested that later cognitive development correlates more highly with early problem-solving skills, whereas language development tends to correlate more highly with a child's understanding of both object-concept and means-end relationships (Siegel, 1979). Very little is actually known about the precise neurophysiological processes that occur in recall and comparison of stored memory data. There is a surfeit of theories and conceptualizations and a shortage of consistent, experimental data that specify the underlying mechanisms. What remains clear, however, is that the processes involve comparisons. They tend to occur at a rather abstract level of cognitive processing
59
that may take place prior to actual conscious registration in the developing infant (Cohen, 1979; Cohen, De Loache, & Strauss, 1979; Mistretta & Bradley, 1978). A consideration of how symbolic information is processed in the cerebral hemispheres raises the issue of cerebral lateralization. Studies have shown that matters are not as localized as was previously thought according to older theories about left and right brain functions. Children as well as adults use both hemispheres. Perhaps in early child development, both hemispheres serve linguistic functions, prior to left hemisphere lateralization in the majority of righthanded individuals for language capacity (Benson & Zaidel, 1985). This would suggest the possibility that greater cortical plasticity is most likely present in the earlier stages of infant language development (Molfese, 1977). Despite the appearance of a high degree of hemispheric specialization as the child matures, the human brain efficiently can be viewed as a ''single-channel system'' in the earlier stages of infant development and later shifts due to changes in its subsequent cognitive and linguistic development (Kinsbourne, 1976; Witelson, 1977; Taylor, 1969). Some of the rather striking consistent correlations between maturation of the brain and development of behavioral processes are as follows: the visual processing area in the brain develops early in the first year; the limbic areas develop later in the first year; the sensorimotor areas develop in earnest in the second year; hearing areas continue to develop into the fourth year (Bronson, 1982; Bushnell, 1982; Levine, 1982; Seines & Whittaker, 1976). The brain's highest areas (those involved in thinking, abstraction, reasoning, problem-solving, and so on) continue to mature well into the teens and perhaps even into the third decade. Development of social competency is really a mixture of maturing of perceptual, sensorimotor, motor, and linguistic mechanisms in the brain, in conjunction with the social conditions of the acculturation process (Oates, 1979; Siegel, 1979).
CNS Maturation in Early Cognitive Development Maturation of cognitive abilities in relation to the growth of the brain, previously discussed, is influential also with regard to emotional and personality development in children (Emde, Gaensbauer, & Harmon, 1976). A sequence of maturation of cognitive abilities has been proposed by Kagan (1985). First, the infant demonstrates the capacity of memory
60
CHAPTER3
for past experiences; second, active memory formation occurs; third, there is a symbolic framework that takes shape; fourth, the infant is able to infer causality; and fifth, the child is able to exhibit selfawareness. Kagan and co-workers assert that these five steps, in the sequence described, occur by the age of 2 in the normal developing brain (Kagan, 1981; Kagan & Moss, 1983; Kagan, Kearsely, & Zelazo, 1978). Kagan asserts that a normal developing child can by 8 months show the ability to retrieve hidden objects, whereas earlier, if "out of sight," it was "out of mind." Beginning approximately at 8-9 months, incoming information is related to knowledge for the first time, giving rise to the emergence of active memory processing. At 8 to 10 months, cardiac acceleration occurs in relation to exposure to the visual cliff experiment. This does not occur prior to 8 months, indicating that the sympathetic nervous system is coming into play more influentially and affects what has been commonly termed the separation anxiety phenomenon. Infants all over the world have been shown to manifest separation anxiety features between 8 and 12 months (Kagan, 1985). As growth of the CNS continues, new capacities emerge. At 17 to 24 months, several important behaviors emerge: appreciation of right versus wrong; appreciation that physical aggression is wrong; appearance of anxiety in relation to failure; the ability to experience empathy; and features of acknowledging anxiety in relation to unsolved problems. At 1-to-3 years, children begin to recognize themselves. It is thought that the maturing brain's anatomical structures permit the concept of selfawareness to emerge, implicating the hippocampus and the thalamocortical projection system. One of the more significant neuroanatomical aspects here is that between 15 and 24 months, all of the cortex's six layers have achieved maturation. Kagan ( 1981) asserted that in a child of approximately 2-3 years, fear is prevalent. It has been shown that the more a child is capable of inhibiting an unfamiliar experience, the better stabilized the child becomes. Separation anxiety (attachment) is cognitively mediated and the environmental context dictates the value placed on the notion of a child's "inhibitedness" (Smith & De Vito, 1984). The latter consideration may be causally related to the formation of aspects of temperament that can be changed with experience. The parasympathetic system has the capacity to quell sympathetic arousal to unfamiliar and nonunderstandable events for the child. The development of temperament heavily in-
fluences a child's behavioral choices in later years in many subtle ways.
Cerebral Asymmetry and Cerebral Lateralization To date, there is no firm conclusion as to the nature and cause of cerebral hemisphere asymmetry. However, the structure and function of each hemisphere are indeed different (Mazziotta & Phelps, 1985). An explanation of the functional differences solely in terms of a dichotomy of verbal or nonverbal nature of information processing also has not been adequately substantiated. Many researchers have proposed theories and models of the development of cerebral asymmetry and its function, including: Lenneberg (1967), Kinsboume (1974), Kimura (1967), Krashen (1973), Witelson (1977), Corballis (1980), Buffery ( 1976), Kinsboume and Hiscock ( 1977), and Moscovitch ( 1977). Studies suggest that within the left and right cerebral hemispheres, at all ages and in both sexes, different functions are served (Benson & Zaidel, 1985; Taylor, 1969; Morgan, 1977; Bradshaw & Nettleson, 1983; Bryden, 1979; Corballis, 1982; Witelson & Pallie, 1973; Kinsboume, 1976). It is believed that bilateral integration of information is mainly subserved by the corpus callosum. Problems can typically arise when growth is delayed or when dysfunctions occur in the between-hemisphere interplay during cognitive information processing. A majority of studies and theories that deal with scientific evidence on cerebral asymmetry have centered on the lateralized aspects of cognitive functioning in children. Studies have shifted in emphasis away from the content-dictated, verbal-spatial dichotomies of the encoding process to a process-determined analytical/ sequential versus gestalt/holistic information processing style (Bogen, 1969; Levy, 1972; LevyAgresti & Sperry, 1968; Luria & Simemitskaya, 1977). Kinsboume and Hiscock (1977, 1978) have presented compelling arguments leading away from the concept of progressive lateralization with age. Kinsboume ( 1982) discussed in some depth the importance of the collaborative efforts of the two hemispheres of the brain. Arguing from the viewpoint of cerebrallateralization theory, Kinsboume stated that mental activities that relate to action in the real world impose demands for integral and coordinated action of both sides of the brain. Luria's (1966, 1973a) position is one of empha-
HIGHER CORTICAL FUNCTIONS IN CHILDREN
sis placed on each hemisphere contributing a different strategy of cognitive information-processing and does not isolate each process within a hemisphere of the brain. He tends to view the human brain as hierarchically organized in order to integrate messages from its lower centers as well as across hemispheres. Luria asserts that dichotomy of functioning does not do justice to the complexity of the human brain's hemispheres. Rather, it is the manner in which the hemispheres organize or represent information versus the type of information organized that is the important distinguishing feature (Luria, 1970, 1973a,b). Witelson's (1977) review of developmental studies of different sensory modalities pertaining to cerebral asymmetries, handedness, sensorimotor, perceptual, and even genetic studies, shows that development of cognitive functions follows a definite order. Several changes occur, some of which can be genetically determined (but influenced by the environment). Hemispheric shifts in which side of the brain handles what type of information, due to alterations in the structural development of the brain, also occur (Bakan, 1971; Annett, 1978). At present, it is reasonable to assume that cerebral dominance is not only related to linguistic processes but also to underlying cognitive ones influenced by the several factors discussed previously. lntermodal hemispheric processing of information can show differential effects when the child reaches 8 years or older, due to the late structural maturation of the corpus callosum. This can lead to various forms of difficulties, for example, dominance problems, dyslexia, and learning difficulties (Satz, Orsini, Saslow, & Henry, 1985; Schonhaut & Satz, 1983). Currently there are conflicting theories and hypotheses related not only to cerebral dominance and hemispheric specialization but also regarding the onset, development, and maturation of the brain's lateralization (Geschwind & Galaburda, 1984).
Conclusion Inadequate encoding of early experience, brain insult, nutritional deficiency, anoxia at birth, perinatal asphyxia, and congenital hereditary defects, etc. each can impose severe restrictions on those capacities essential for processing sensory, motor, visual, acoustic, haptic, and other information. Until normal brain development and the mechanisms for higher cognitive processes are more accurately and adequately described by data from improved experi-
61
mental designs, educational, remedial, diagnostic, and therapeutic regimens wiJI only be partially effective. A necessary first step in making scientific progress is to describe the events and conditions defining a process of psychological function before proceeding to hypotheses and constructs of brain-behavior relationships. Experimental studies of brain mechanisms and developmental issues, coupled with neurophysiological, neuropsychological, and neurobiological data will yield not only broader generalizations, but also specific knowledge that can lead to better understanding of the child's developing brain structures, functions, and cognitive capacities involved in extracting meaning from the external environment (Gottlieb, 1976a,b; Lowrey, 1973).
Research Strategies for Studying Normal Brain Development and Its Functions Research Designs and Methodologies The choice of experimental designs and research methodologies for the investigation of questions in the field of developmental neuropsychology poses significant problems such as replication, proper statistical means of analyzing sample data, group differences, and power as to the conclusions that can be drawn from research studies. Spreen et al. (1984) have recently addressed some of the methodological concerns regarding developmental neuropsychology. The future of clinical research in developmental neuropsychology will radically change in the next few decades, due chiefly to the fact that presently available instruments are being technologically refined for use with different age levels of people studied. The horizon that lies ahead for developmental neuropsychological research and theories that will emerge, will be shaped chiefly through emerging technological innovations. These include imaging of living brain chemistry, metabolism, and neuroreceptors; and the measurement of maturational changes in cerebral functioning by means of mapping brain function onto structure. A radically new data base from which to make generalizations about normal brain-behavior relationships in the developing infant is to be found, chiefly in terms of observations and accurate descrip-
62
CHAPTER3
tions regarding structure, function, and metabolic requirements of tissue competency. A current review of the literature of developmental psychology and developmental neuropsychology reveals a lack of replication in many studies. There exists considerable controversy in the area of human lateralization studies, cerebral asymmetry theories, and implications for brain development. These controversies will abound until research on normal brain-behavior relationships in infants and young children begins to fill in the gaps in our knowledge. Despite the impressive existing array of technological capabilities for measurement, a major aspect is missing. Measurement of the brain's sensory, motor, and cognitive processes provides only indirect assessment of task performance. Even noninvasive techniques, such as the EEG or ERP, are indirect in that they only measure electrophysiological phenomena arising from electrochemical brain activity, not actual cognitive activity (Gevins et al., 1981). Details of cognitive activities must be inferred from knowledge of the underlying physiology and the task being performed by the subject (Fender, 1985). What is needed to overcome these limitations is a systematic, comprehensive investigation correlating patterns ofbrain activity with selected behaviors; and mapping of function onto brain structure. This will allow researchers to acquire information about the functional organization of the working brain. It is possible today to map brain function onto structure by bringing together neuropsychological assessment data about structural changes and correlating structural damage with change in cerebral function and altered cortical processes. Various determinants, which can be quantitatively measured and correlated, will allow more accurate description of the maturational changes in cerebral function in humans with some reasonable behavioral correlates (Fender, 1985; Chugani & Phelps, 1986). Spreen et al. make the following point regarding the importance of replication in competent research: . . . The goals of the replication study are to answer the following questions. Can the original investigator or an independent investigator following the information provided by the original investigator replicate the results of the original study? Have social, cultural, economic, medical, etc., changes in the population made previous findings obsolete or misleading? Are the findings generalizable to a new set of subjects, test items, test settings, etc.? In sum, replication is a powerful tester for determining the relevance of an investigation and for weeding out findings that may show signifi-
cance by pure chance or may have become obsolete. (pp. 88-89)
The key to making a systematic investigation of the mapping of human brain function onto structure during development in order to correlate cortical processes in the functional organization of the brain, lies in the combination of many imaging techniques together with functional source localization and neuropsychological measurement.
Conclusion Relatively little is known about the normal developing brain with respect to higher cortical functions, especially in early infancy. Selected studies and techniques presented here have begun to make an impact on this situation. Significant information on the development of normal brain-behavior functions will come about by joint efforts employing imaging methods (which reveal structure); neuropsychological assessment techniques (Taylor et al., 1984) (which determine consequences of neurological function); neural source localization techniques (which can identify the site of those neural populations that subserve specific brain functions); and metabolic and spectral measurements of human physiology (metabolic tissue competency) (Phelps et al., 1986). It is the possibility of standardized, quantitative correlations of structure, function, metabolism, brain electrical activity, and neuropsychological factors that will ultimately promote a different and novel knowledge base. It will also lead to improvement in our understanding of the development of higher cortical functions from a normal developmental perspective. A major criticism that some have leveled at the usefulness of the correlation between structure, function, and behavior is that such attempts are doomed to fail because it is impossible to localize function in the nervous system. Lesions (or structures) are localized, not functions. However, there is extensive evidence that at least certain stereotypical processes, such as sensory input and execution of motor output, are correlated with highly consistent patterns of electrical cortical activity. This is considered to be hard-wired. As there are characteristic signatures of electrical activity associated with some components of behavior, metabolic (chemical) signatures also have been consistently drawn. It appears that the possibility now exists of linking structure, function, and cerebral physiology with powerful imaging techniques in a
HIGHER CORTICAL FUNCTIONS IN CHILDREN
cohesive, standardized fashion for gathering data so as to produce the type of correlation lacking in the above regard concerning developing brain function and higher cortical processes in children.
References Afifi, A. K., & Bergman, R. A. (1980). Basic neuroscience. Munich: Urban & Schwarzenberg. Altman, J., Brunner, R. L., & Bayer, S. A. (1973). The hippocampus and behavioral maturation. Behavioral Biology, 8, 557. Annett, M. (1978). Genetic and nongenetic influences on handedness. Behavior, 8, 227-249. Arey, L. B. (1974). Developmental anatomy (7th ed.). Philadelphia: Saunders. Bakan, P. (1971). Handedness and birth order. Nature, 229, 195. Barnet, A. B. (1966). Visual responses in infancy and their relation to early visual experience. Clinical Proceedings Children's Hospital National Medical Center, 22, 273. Bear, D. M. (1986). Hemispheric asymmetries in emotional functioning. In B. K. Doane & K. E. Livingston (Eds.), The limbic system: Functional organization and clinical disorders. New York: Raven Press. Benson, D. F., & Zaidel, E. (Eds.). (1985). The dual brain: Hemispheric specialization in humans. New York: Guilford Press. Bernstine, R. L., Borkowski, W. J., & Price, A. H. (1955). Prenatal fetal electroencephalography. American Journal of Obstetrics and Gynecology, 70, 623. Bloom, F. E. (1973). Dynamic synaptic communication: Finding the vocabulary. Brain Research, 62, 299-305. Bloom, F. E. (1979). Neurobiological research and selective attention. H. G. Birch Memorial Lecture, 1979 International Neuropsychological Society, San Francisco. Bogen, J. E. (1969). The other side of the brain: Parts I, II, III. Bulletin of the Los Angeles Neurological Society, 34, 73105; 135-162; 191-203. Bower, T. G. R. (1977). The perceptual world of the child. Cambridge, MA: Harvard University Press. Bradshaw, J. L., & Nettleson, N. C. (1983). Human cerebral asymmetry. Englewood Cliffs, NJ: Prentice-Hall. Bronson, G. W. (1982). The scanning patterns of human infants: Implications for visual/earning. Norwood, NJ: Ablex. Bryden, M. (1979). Evidence for sex differences in cerebral organization. In M. Wittig & A. Peterson (Eds.), Determinants of sex-related differences in cognitive functioning. New York: Academic Press. Buffery, A. W. H. (1976). Sex differences in the neuropsychological development of verbal and spatial skills. In R. M. Knights & D. J. Bakker (Eds.), The neuropsychology of learning disorders. Baltimore: University Park Press. Bushnell, E. W. (1982). Visual-tactual knowledge in 8-,9\12-, and 11-month-old infants. Infant Behavior and Development, 5, 63-75. Carlson, N. R. (1977). Physiology of behavior. Rockleigh, NJ: Allyn & Bacon.
63
Carpenter, M. B. (1978). Core text of neuroanatomy (2nd ed.). Baltimore: William & Wilkins. Chugani, H. T., & Phelps, M. E. (1986). Maturational changes in cerebral function in infants determined by 18FDG positron emission tomography. Science, 231. 840-843. Cohen. L. B. (1979). Our developing knowledge of infant perception and cognition. American Psychologist, 34, 894899. Cohen, L. B., De Loache, J. S., & Strauss, M.S. (1979). Infant perceptual development. In J. D. Osofosky (Ed.), Handbook of infant development. New York: Wiley. Cooke, J. (1980). Early organization of the central nervous system: Form and pattern. In R. K. Hunt (Ed.), Neural development. New York: Academic Press. Corballis, M. C. (1980). Is left-handedness genetically determined? In J. Herron (Ed.), Neuropsychology of left handedness. New York: Academic Press. Corballis, M. C. (1982). Asymmetries in spatial representation: Anatomical or perceptual? In R. N. Malatesha & L. C. Hartlage (Eds.), Neuropsychology and cognition (Vol. 1). The Hague: Nijhoff. Cravioto, J., & Arrieta, R. (1979). Stimulation and mental development of malnourished infants. Lancet, 2, 899. Cravioto, J., & Arrieta, R. (1983). Malnutrition in childhood. In M. D. Rutter (Ed.), Developmental neuropsychiatry. New York: Guilford Press. Crelin, E. S. (1973). Functional anatomy of the newborn. New Haven, CT: Yale University Press. Cretin, E. S. (1974). Development of the nervous system: A logical approach to neuroanatomy. Ciba Clinical Symposia, 26, 1-32. Curtiss, S. (1979). Genie: Language and cognition. UCLA Working Papers in Cognitive Linguistics, 1, 15-62. Darley, F. L., &Fay, W. H. (1980). Speech mechanism. In F. M. Laasman, R. 0. Fisch, D. K. Vetter, & E. S. Benz (Eds.), Early co"elates ofspeech, language, and hearing. Littleton, MA: PSG Publishing. Daughaday, W. H. (1974). The adenohypophysis. In R. H. Williams (Ed.), Textbook of endocrinology. Philadelphia: Saunders. Davies, P., & Stewart, A. L. (1975). Low birthweight infants: Neurological sequelae and later intelligence. British Medical Bulletin, 31, 85. Davison, A. N., & Dobbing, J. (1968). The developing brain. In A. N. Davison & J. Dobbing (Eds.), Applied neurochemistry. Oxford: Blackwell. Davison, A. N., &Peter, A. (1970).Myelination. Springfield, IL: Thomas. DeVilliers, P. A., & DeVilliers, J. G. (1979). Early language. Cambridge, MA: Harvard University Press. Dobbing, J. ( 1975). Prenatal nutritional and neurological development. InN. A. Buchwald & M. A. B. Brazier (Eds.), Brain mechanism in mental retardation. New York: Academic Press. Dobbing, J., & Sands, J. (1970). Timing of neuroblast multiplication in developing human brain. Nature, 226, 639. Dobbing, J., & Sands, J. (1973). Quantitative growth and development of human brain. Archives of Disabled Children, 48, 757.
64
CHAPTER3
Dobbing, J., & Smart, J. L. (1974). Vulnerability of developing brain and behaviour. British Medical Bulletin, 30, 164. Dodge, P., Prensky, A., & Feigin, R. (1975). Nutrition and the developing nervous system. St. Louis: Mosby. Dodgson, M. C. H. (1962). The growing brain: An essay in developmental neurology. Bristol, England: Wright Press. Dunn, A. J. (1976). The chemistry of learning and the formation of memory. In W. H. Gispen (Ed.), Molecular and functional neurobiology. Amsterdam: Elsevier. Dunn, A. J. (1980). Neurochemistry of learning and memory: An evaluation of recent data. Annual Review of Psychology, 31, 343-390. Ellingson, R. J. (1964). Studies of the electrical activity of the developing human brain. Progress in Brain Research, 9, 2653. Emde, R.N., Gaensbauer, T. J., & Harmon, R. J. (1976). Emotional expression in infancy: A behavioral study (Vol. 10). New York: International University Press. Epstein, H. T. (1978). Growth spurts during brain development: Implications for educational policy and practice. In J. S. Chall & A. F. Mirsky (Eds.), Education and the brain: The 77th yearbook of the National Society for the Study ofEducation (Part II). Chicago: University of Chicago Press. Fender, D. H. (1985). Source localization of brain electrical activity. In A. Gevins & A. Remond (Eds.), Handbook of electroencephalography and clinical neurophysiology (Vol. 3A). Amsterdam: Elsevier. Finkelstein, S., Alpert, N. M., Ackerman, R. H., Buonano, F. S., Correia, J. A., Chang, J., Kulas, S., Brownell, G. L., & Taveras, J. M. (1980). Positron imaging of the normal brain-Regional patterns of cerebral blood flow and metabolism. Transactions of the American Neurological Association, 105, 8-10. Franz, R. L. (1963). Pattern vision in newborn infants. Science, 140, 296. Gaddes, W. H. (1980). Learning disabilities and brain function: A neuropsychological approach. Berlin: Springer-Verlag. Geschwind, N., & Galaburda, A. M. (Eds.). (1984). Cerebral dominance. Cambridge, MA: Harvard University Press. Gevins, A. S., Doyle, J. C., Cutillo, B. A., Schaffer, R. E., Tannehill, R. S., Ghannam, J. H., Gilcrease, V. A., & Yeager, C. L. (1981). Electrical potentials in human brain during cognition-New method reveals dynamic patterns of correlation. Science, 213, 918-921. Goldman, P. S. (1975). Age, sex and experience as related to the neural basis of development. InN. A. Buchwald & M.A. B. Brazier (Eds.), Brain mechanisms in mental retardation. New York: Academic Press. Gottlieb, G. (1976a). Conceptions of prenatal development: Behavioral embryology. Psychological Review, 83, 215-234. Gottlieb, G. (1976b). The roles of experience in the development of behavior and the nervous system. In G. Gottlieb (Ed.), Neural and behavioral specificity. New York: Academic Press. Grossberg, S. (1980). How does a brain build a cognitive code? Psychological Review, 87, l-51. Hagne, I. (1972). Development of the EEG in normal infants during the first year of life. Acta Paediatrica Scandinavica, Suppl. l, 232, 5.
Hamilton, W. J., & Mossman, H. W. (1974). Human embryology: Prenatal development of form andfunction·(4th ed.). Cambridge: Heffer. Humphrey, T. (1964). Some correlation between the appearance of human fetal reflexes and the development of the nervous system. In D.P. Purpura & J.P. Schade (Eds.), Amsterdam: Elsevier. Humphrey, T: (1978). Function of the nervous system during prenatal life. In U. Stave & A. A. Weech (Eds.), Perinatal physiology. New York: Plenum Medical. Ito, M. (1984). The cerebellum and neural control. New York: Raven Press. Izquierdo, I. (1975). The hippocampus and learning. Progress in Neurobiology, 5, 37-75. Jones, E. G., & Cowan, W. M. (1978). Nervous tissue.: Development of nervous tissue.In E. Weiss (Ed.), Textbook of histology. New York: McGraw-Hill. Kagan, J. (1981). Th< second year. Cambridge, MA: Harvard University Press. Kagan, J. (1985). The human infant. In A. M. Rogers & C. J. Scheirer (Eds.), The G. Stanley Hal/lecture series (Vol. 5). Washington, DC: American Psychological Association. Kagan, J., Kearsley, R. B., & Zelazo, P. R. (1978). Infancy: Its place in human development. Cambridge, MA: Harvard University Press. Kagan, J., & Moss, H. A. (1983). Birth to maturity. New Haven, CT: Yale University Press. Kahle, W., Leonhardt, H., & Platzer, W. (1978). Color atlas and textbook of human anatomy (Vol. 3). Chicago: Year Book Medical Publishers. Kaplan, S. L., Grumbach, M. M., & Aubert, M. L. (1976). The ontogenesis of pituitary hormone aad hypothalamic factors in the human fetus: Maturation of the central nervous system regulation of anterior pituitary function. Recent Progress in Hormone Research, 32, 161. Kety, S. S. (1970). The biogenic amines in the central nervous system: Their possible roles in arousal, emotion and learning. In F. 0. Schmitt (Ed.), The neurosciences: Second study program. New York: Rockefeller University Press. Kimura, D. (1967). Functional asymmetry of the brain in dichotic listening. Cortex, 3, 163-178. Kinsbourne, M. (1974). Mechanisms of hemisphere interaction in man. In M. Kinsboume & W. L. Smith (Eds.), Hemisphere disconnection and cerebral function. Springfield, IL: Thomas. Kinsbourne, M. (1976). The ontogeny of cerebral dominance. In R. Rieber (Ed.), The neuropsychology of language. New York: Plenum Press. Kinsboume, M. (1982). Hemispheric specialization and the growth of human understanding. American Psychologist, 37, 411-420. Kinsbourne, M., & Hiscock, M. ( 1977). Does cerebral dominance develop? In S. J. Segalowitz & F. A. Gruber (Eds.), Language development and neurological theory. New York: Academic Press. Kinsboume, M., & Hiscock:, M. (1978). Cerebrallateralization and cognitive development. In J. S. Chall & A. F. Mirsky (Eds.), Education and the brain: The 77th vearbook of the
HIGHER CORTICAL FUNCTIONS IN CHILDREN National Society for the Study of Education (Part II). Chicago: University of Chicago Press. Krashen. S. (1973). Lateralization, language, learning, and the critical period: Some new evidence. Language and Learning, 23. 63-74. Kuffler, S. W., &Nicholls, J. G. (1977). From neuron to brain: A cellular approach to the function of the nervous system. Sunderland, MA: Sinauer Associates. Langman, J. (I 975). Medical embryology: Human developmentnormal and abnormal (3rd ed.). Baltimore: William & Wilkins. Lashley, K. S., Chow, K. L., and Semmes, J. (1951). An examination of the electrical field theory of cerebral integration. Psych. Rev., 58, 123-136. Le Douarin, N. (1980). Migration and differentiation of neural crest cells. In R. K. Hunt (Ed.), Neural development. New York: Academic Press. Lehninger, A. L. (1968). The neuronal membrane. Proceedings of the National Academy of Sciences USA, 60, 1069-1080. Lemire, R. J., Loeser, J. D., Alvord, E. C., & Leech. R. W. (1975). Normal and abnormal development of the human nervous system. New York: Harper & Row. Lenneberg, E. H. (1964). The natural history of language. In F. Smith & G. A. Miller (Eds.), The genesis of language: A psycholinguistic approach. Cambridge, MA: MIT Press. Lenneberg, E. H. ( 1967). Biological foundation oflanguage. New York: Wiley. Levene, M. 1., & Dubowitz, L. M. S. (1982). Low-birth weight babies long-term follow-up. British Journal ofHospital Medicine, 24, 487. Levine, S. (1982). Comparative and psychobiological perspectives on development. In A. Collins (Ed.), Minnesota symposium (Vol. 15). Hillsdale, NJ: Erlbaum. Levy, J. (1972). Lateral specialization of the human brain: Behavioral manifestation and possible evolutionary basis. In J. A. Kiger (Ed.), Biology of behavior. Corvallis: Oregon State University Press. Levy-Agresti, J., & Sperry, R. W. (1968). Differential perceptual capacities in major and minor hemispheres. Proceedings of the National Academy of Sciences USA, 61, 1151. Liben, L. S., Patterson, A. H., & Newcombe, N. (1981). Spatial representation and behavior across the life span: Theory and application. New York: Academic Press. Lindsley, D. B. (1939). A longitudinal study of the occipital alpha rhythm in normal children: Frequency and amplitude standards. Journal of Genetic Psychology, 55. 197-213. Lindsley, D. B., & Wicke, J. D. (1974). The electroencephalogram: Autonomous electrical activity in man and animals. In R. F. Thompson & M. M. Patterson (Eds.), Bioelectric recording techniques. Pan B. Electroencephalography and human brain potentials. New York: Academic Press. Livingston, R. B. (1978). Sensory processing, perception, and behavior. New York: Raven Press. Lorber, J. (1980). Is your brain really necessary? Science, 210, 1232-1234. Lowrey, G. H. (1978). Growth and development of children (7th ed.). Chicago: Year Book Medical Publishers. Lund, R. D. (1978). Development and plasticity of the brain: An introduction. New York: Oxford University Press.
65
Luria, A. R. (1960). Verbal regulation of behavior. In M. Brazier (Ed.), The CNS and behavior. New York: Josiah Macy Jr. Foundation. Luria, A. R. (1961). The role ofspeech in the regulation of normal and abnormal behavior. Elmsford, NY: Pergamon Press. Luria, A. R. (l%6). The human brain and psychological processes. New York: Harper & Row. Luria, A. R. (l969a). Origin and brain organization of conscious activity. Evening lecture to the 19th International Congress of Psychology. London: Dorset Press. Luria, A. R. (l969b). Frontal lobe syndromes. In P. J. Vinken & G. W. Bruyn (Eds.), Handbook of clinical neurology (Vol. 2). Amsterdam: North-Holland. Luria, A. R. (l969c). Speech development and the formation of mental processes. In J. Cole & I. Maltzman (Eds.), A handbook of contemporary Soviet psychology. New York: Basic Books. Luria, A. R. (1970). The functional organization of the brain. Scientific American, 222, 66-78. Luria, A. R. (1973a). The working brain. New York: Basic Books. Luria, A. R. (1973b). The frontal lobes and the regulation of behavior. In A. R. Luria & K. H. Pribram (Eds.), The behavioral psychophysiology of the frontal lobes. New York: Academic Press. Luria, A. R. (1980). Higher cortical functions in man (2nd ed.). New York: Basic Books. Luria, A. R. (1982). Language and cognition. New York: WileyInterscience. Luria, A. R., & Simernitskaya, E. G. (1977). Interhemispheric relations and the functioning of the minor hemisphere. Neuropsychologia, 15, 175-178. Luria, A. R., & Yudovich, F.l. (1959). Speech in the development of mental processes in the child. London: Staples. Majovski, L. V., &Jacques, D. B. (1982). Cognitive information processing and learning mechanisms of the brain. Neurosurgery, 10, 663-677. Majovski, L. V., Jacques, D. B., Hartz, G., & Fogwell, L. A. ( 1981). Dopaminergic (DA) systems: Their role in pathological neurobehavioral symptoms. Neurosurgery, 9, 751-757. Mazziotta, J. C., & Phelps, M. E. (1985). Metabolic evidence of lateralized cerebral function demonstrated by positron emission tomography in patients with neuropsychiatric disorders and normal individuals. In D. F. Benson & E. Zaidel (Eds.), The dual brain. New York: Guilford Press. Mazziotta, J. C., Phelps, M. E., & Miller, J. (1981). Tomographic mapping of human cerebral metabolism: Normal unstimulated state. Neurology, 31, 503-516. Metcof, J. (1974). Biochemical markers of intrauterine malnutrition. In M. Winick (Ed.), Current concepts in nutrition (Vol. 2). New 'York: Wiley. Milner, E. (1976). CNS maturation and language acquisition. In H. Whitaker & H. A. Whitaker (Eds.), Studies in neurolinguistics (Vol. I). New York: Academic Press. Mistretta, C. M., &Bradley, R. M. (1978). Effectofearlysensory experience on brain and behavioral development. In G. Gottlieb (Ed.), Studies on the development of behavior and the nervous system (Vol. 4). New York: Academic Press. Molfese, D. L. (1977). Infant cerebral asymmetry. In S. J.
66
CHAPTER3
Segalowitz & F. A. Gruber (Eds.), Language development and neurological theory. New York: Academic Press. Molfese, D., Freeman, R., & Palermo, D. (1975). The ontogeny of brain lateralization for speech and nonspeech stimuli. Brain Language, 2, 356-368. Moore, K. L. (1974). Before we are born: Basic embryology and birth defects. Philadelphia: Saunders. Moore, K. L. (1975). Study guide and review manual of human embryology. Philadelphia: Saunders. Moore, K. L. (1977). The developing human: Clinically oriented embryology (2nd ed.). Philadelphia: Saunders. Moore, R. Y. (1977). The developmental organization of the fetal brain. In L. Gluck (Ed.),lntrauterine asphyxia and the developingfetal brain. Chicago: Year Book Medical Publishers. Morgan, M. (1977). Embryology and inheritance of asymmetry. In S. Hamad, R. Doty, L. Goldstein, J. Jaynes, & G. Lruthamer (Eds.), Lateralization in the nervous system. New York: Academic Press. Moscovitch, M. (1977). The development of Jateralization of language functions and its relation to cognitive and linguistic development: A review and some theoretical speculations. In S. I. Segalowitz & F. A. Gruber (Eds.), Language development and neurological theory. New York: Academic Press. Nauta, W. H. (l986a). Circuitous connections linking cerebral cortex,limbic system and corpus striatum. In B. K. Doane & K. E. Livingston (Eds.), The limbic system: Functional organization and clinical disorders. New York: Raven Press. Nauta, W. H. (l986b). A simplified perspective on the basal ganglia and their relation to the limbic system. In B. K. Doane & K. E. Livingston (Eds.), The limbic system: Functional organization and clinical disorders. New York: Raven Press. Neville, H. (1984). Effects of early sensory and language experience on the development of the human brain. In I. Mehler & R. Fox (Eds.), Neonate cognition: Beyond the blooming buzzing c01!{usion. Hillsdale, NJ: Erlbaum. Nilsson, L. (1978). A child is born. New York: Delacorte Press. Oates, J. (Ed.). (1979). Early cognitive development. New York: Wiley. Phelps, M. E., Mazziotta, J .. &Scheiber!, H. R. (1986). Positron emission tomagraphy and autoradiography: Principles and applicationsfor the brain and heart. New York: Raven Press. Pinker, S. (1984). Language learnability and language development. Cambridge, MA: Harvard University Press. Plomin, R. A., & Rowe, D. C. (1979). Genetic and environmental etiology of social behavior in infancy. Developmental Psychology, 15, 62-72. Ploog, D. (1979). Phonation, emotion, cognition with reference to the brain mechanisms involved. Ciba Foundation Symposium, 69, 79-98. Pribram, K. H. (1976). Modes of central information processing in human learning and remembering. InT. I. Tyler (Ed.), Brain and learning. Baltimore: Graylock Press." Ramon Cajal, S. (1909). Histologie du systeme nerveaux de /' homme et des vertebres. Paris: Maloine. Reichlin, S. (1974). Neuroendocrinology. In R. H. Williams (Ed.). Textbook of endocrinology. Philadelphia: Saunders. Riss, W. (1972). Nonspecific thalamic projection system: Introduction. Brain, Behavior, and Evolution, 6, 329-331. Roberts, E. (1986). Metabolism and nervous system disease: A
challenge for our times (Part I). Metabolic Brain Disease, I, 1-25. Rourke, B. P., Bakker, D. J., Fisk, J. L., & Strang, J.D. (1983). Child neuropsychology: An introduction to theory, research, and clinical practice. New York: Guilford Press. Satz, P., Orsini, D. L., Saslow, E., & Henry, R. (1985). The pathological left-handedness syndrome. Brain and Cognition, 4, 27-46. Scheibel, A. B. (1978). Clinical neuroanatomy lecture. Anatomy and Physiology 103: Winter-Spring Quarter 1978, Basic neurology lecture. UCLA School of Medicine, Los Angeles. Scheibel, M., & Scheibel, A. (1961). On circuit patterns of brain stem reticular core. Annals of the New York Academy of Sciences, 89, 857-865. Scheibel, M., & Scheibel, A. (1963). Some neural substrates of postnatal development. In E. Hoffman (Ed.), First annual review of child development. New York: Russell Sage Foundation. Scheibel, M. E., & Scheibel, A. G. (1966). Patterns of organization in specific and nonspecific thalamic fields. In D. Purpura & M. D. Yahr (Eds.), The thalamus. New York: Columbia University Press. Scheibel, M. E., & Scheibel, A. B. (1972). Input-outputrelations of the thalamic nonspecific system. Brain, Behavior, and Evolution, 6, 332-358. Scheibel, M. E., & Scheibel, A. B. (1973). Dendrite bundles as sites for central program: An hypothesis.lnternational Journal of Neuroscience, 6, 195-202. Schonhaut, S., & Satz, P. (1983). Prognosis for children with learning disabilities: A review of follow-up studies. In M. Rutter (Ed.), Developmental neuropsychiatry. New York: Guilford Press. Schulte, F. J. (1974). The neurological development of the neonate. In J. A. Davis & I. Dobbing (Eds.), Scientific foundations of pediatrics. Philadelphia: Saunders. Segalowitz, S., & Chapman, J. (1980). Cerebral asymmetry for speech in neonates: A behavioral measure. Brain Language, 9, 281-288. Sehies, D. A., & Whittaker, H. A. (1976). Morphological and functional development of the auditory system. In R. W. Rieber(Ed.), The neuropsychology of language. New York: Plenum Press. Siegel, L. S. (1979). Infant perceptual, cognitive, and motor behavior as predictors of subsequent cognitive and language development. Canadian Journal of Psychological Reviews, 33, 382-395. Siqueland, E. R., & Lipsitt, L. P. (1966). Conditioned head turning in newborns. Journal ofExperimental Child Psychology, 3, 356-376. Smith, B. H., & Sweet, W. H. (l978a). Monoaminergic regulation of central nervous system function. I. Noradrenergic systems. Neurosurgery, 3, 109-119. Smith, B. H., & Sweet, W. H. (1978b). Monoaminergic regulation of central nervous system function. II. Serotonergic systems. Neurosurgery, 3, 257-272. Smith, D. W. (1976). Recognizable patterns of human malformation: Genetic, embryologic and clinical aspects (2nd ed.). Philadelphia: Saunders. Smith, 0. A., & De Vito, J. L. (1984). Central neural integration
HIGHER CORTICAL FUNCTIONS IN CHILDREN for the control of autonomic responses associated with emotion. Annual Reveiw of Neuroscience, 7, 43-65. Snyder, S. (1980). Brain peptides as neurotransmitters. Science, 209. 976-983. Spreen, 0., Tupper, D., Risser, A., Tuokko, H., & Edgell, D. ( 1984). Human developmental neuropsychology. New York: Oxford University Press. Stuss, D. T., & Benson, D. F. (1986). The frontal lobes. New York: Raven Press. Szentagothai, J. (1975). The "module concept" in the cerebral cortex architecture. Brain Research, 95, 475-496. Szentagothai, J. ( 1978). The Ferrier Lecture. The neuron network of the cerebral cortex: A functional interpretation. Proceedings of the Royal Society of London, 201, 219-248. Szentagothai, J., & Arbib, M.A. (1974). Conceptual models of neural organization. Neurosciences Research Program, Bulletin, 12, 307-501. Taylor, D. (1969). Differential rates of cerebral maturation between sexes and between hemispheres. Lancet, 2, 140-142. Taylor, H. G., Fletcher, J. M., & Satz, P. (1984). Neuropsychological assessment of children. In L. Halpern & G. Goldstein (Eds.), Handbook of psychological assessment. Elmsford, NY: Pergamon Press. Vygotsky, L. S. (1974). The problem of age-periodization of child development [translated by A. Zender & B. F. Zender]. Human Development, 17, 24-40. Vygotsky, L. S. (1980). Mind in society: The development of
67
higher psychological processes (edited by M. Cole, V. JohnSteiner, S. Scribner, & E. Soubennan). Cambridge, MA: Harvard University Press. Warren, J. M., & Akert, K. (Eds.). (1964). The frontal granular cortex and behavior. New York: McGraw-Hill. Wertsch, J. V. (1979). The regulation of human action and the given-new organization of private speech. In G. Zivin (Ed.), The development of self-regulation through private speech. New York: Wiley. Witelson, S. (1977). Early hemispheric specialization and interhemispheric plasticity: An empirical and theoretical review. InS. J. Segalowitz & F. A. Gruber (Eds.), Language development and neurological theory. New York: Academic Press. Witelson, S., & Pallie, W. (1973). Left hemisphere specialization for language in the newborn. Brain, 96, 641-646. Yakovlev, P. I. (1962). Morphological criteria of growth and maturation of the nervous system in man. Research Publica-
tions, Association for Research in Nervous and Mental Diseases, 39, 3. Yakovlev, P. 1., & Lecours, A. R. (1967). The myelogenetic cycles of regional maturation of the brain. In A. Minkowski (Ed.), Regional development of the brain. Oxford: Blackwell. Zucker, R. S., & Lando, L. (1986). Mechanism of transmitter release: Voltage hypothesis and calcium hypothesis. Science, 231' 574-579.
4 Mechanisms and Development of Hemisphere Specialization in Children MARCEL KINSBOURNE
The Normative Endpoint of Hemisphere Specialization in the Adult The sequences of cognitive development to be considered in this discussion culminate in an endpoint in the mature human nervous system that in its broad definition is no longer in dispute. In the right-handed majority, language-related processes are left latera_lized in almost every case. There does appear to be a nght hemisphere contribution toward certain aspects of verbal behavior, however {Hecaen, 1978; Searleman: 197?, 1983), notably comprehension oflogical relat1onsh1ps, metaphor, and humor (Gardner, Ling, ~amm: & Silverman, 1975), and at the output stage, mtonat1on, particularly when it reflects the emotional tone of the utterance (Ross & Mesulam, 1979). The l~ft hemisphere is also specialized for rapid sequential recognition of familiar input, both verbal and nonverbal, as well as the recall and recognition of order information and formulation of action plans plan~ both ~otor and conceptual. Right hemisphere do?lmance 1s best documented for certain spatial relatiOnal processes particularly in the visual modality, as well as the processing of emotional information (Kinsbourne, 1982). Non-right-banders deviate from the dextral norm in the following manner: in addition to left-sided representation of language, which is as prevalent. in non-right-banders as in right-banders, language IS also represented on the right in some 70% ofthe cases (Gioning, Gloning, Haub, & Quatember, 1_959). I~ non-right-banders as in right-banders, spattal-relattOnal functions are right Iateralized, but inMARCEL KINSBOURNE • Department of Behavioral Neurology, Eunice Kennedy Shriver Center. Waltham, Massachusetts 02254.
volve the left hemisphere also in more than half of all instances (Bryden, Hecaen, & DeAgostini, 1983). Within each hemisphere the territory involved in cognitive function is more extensive in the left- than the right-hander. Gender-related differences in lateralization are more contentious, and the claim that language and visuospatial functions in females are more bilateralized than in males, and even that intrahemispheric organization on the left differs between the sexes, are not as yet well enough substantiated to have made an impact on the study of the development of hemispheric specialization. With respect to devel~ opment, existing knowledge is virtually restricted to language functions on the one hand and spatial~rela tional functions on the other. This discussion will theref~re be confined to considering how peripheral laterahty, notably hand preference, is established in the developing child, and how the differential hemispheric representation of language and spatial-relational functions develops, first in the normal case and then in a number of different types of develo~ mental disability. It is generally accepted that at birth the human infant's cerebrum has not yet assumed the functions that it will ultimately subserve. The functions that will emerge over time can be subdivided into those for which hemispheric specialization has been ascertained, and those for which no lateralization has (as yet) been established. Although a good deal is now known about the changes in number, configuration, and connectivity of neurons in the two developing hemispheres, none of this has been shown to relate to emerg_in_g higher me~tal functions. One can only time the ongm of the vanous lateralized mental skills by observing their appearance at the behavioral level. The developmental sequences involved in the emergence of language and spatial-relational skills have been reasonably well specified. Once they be69
70
CHAPTER4
gin to emerge, and subsequently as they become more refined, it becomes possible to ask, what is the basis of this developing skill? A distinction has to be drawn between hemispheric specialization and lateralization of function to a hemisphere. It cannot be taken for granted that the developing skill in its earliest stages is based on the same cerebral hemisphere as is ultimately the case. The hypothesis that the lateralization to one hemisphere is constant is termed "invariant lateralization" (Kinsbourne, 1975). It has been pitted against an alternative view, namely, that in the earlier stages mental functions in general (and language in particular) are based on the activity of both cerebral hemispheres and that lateralization of verbal functions to the left and spatial relational to the right occurs over time during childhood. This is the "progressive lateralization" hypothesis (Lenneberg, 1967). Virtually all that is known about the evolution of hemispheric specialization in the child can be formulated in terms of these two hypotheses. Insofar as lateralization and its development are subject to biological variation, one can ask whether a particular topography of hemispheric representation of function is more conducive to efficient mental function than Qthers, and whether some of these other "abnormal" topographies characterize certain developmental disabilities and even account for some or all of the behavioral deficits involved. These questions can be addressed most directly in the case of the fully mature nervous system, but lags in achieving that endpoint can also be considered.
The Origin of Bisymmetry
Somatic bisymmetry is an adaptation to the needs of motile organisms. In addition to the obvious advantage of streamlining, the bisymmetric organism is well adapted for the major decision constantly made by organisms as they progress from point to point: to tum right or to tum left (Loeb, 1918). Given that whatever benefits and hazards the environment might present it is likely to do so with equal probability to either side, the organism needs to be able to deploy its sensorimotor capabilities to either side with comparable speed and efficiency (Gardner, 1967). Thus, the receptor equipment and the musculature are both bilaterally arranged, and so are the corresponding control centers in the nervous system (contralaterally in the chordate phylum, ipsilaterally in the other phyla-Hyman, 1940). Sessile organisms are not bisymmetric, and organisms that regress from a motile to a sessile state concurrently lose their bisymmetric organization. Organisms whose life cycle divides into larval and a mature phase and whose mature phase is sessile, exhibit the relationship most strikingly. The motile larva is bisymmetric, the adult form is not. In the case of fishes, which also freely turn up and down, dorsal-ventral symmetry of the body is commonly found (Braitenberg, 1977). If bisymmetry is an adaptation in the service of motility, then two considerations follow: (I) bodily organs not involved in movement or movement control will not necessarily be bisymmetrically structured and (2) bodily parts that control motion may deviate from complete bisymmetry only to an extent short of inducing a maladaptive bias in movement in ecologically valid situations. Examples that docuAsymmetries in the Evolutionary ment both these propositions abound. The viscera of Context vertebrates are by no means bisymmetric, though they are so packed within the bodily cavity as not to Asymmetric cerebral and somatic functioning distort the bisymmetric form and streamlining of the are chiefly of interest in clinical child neuropsychol- organism as a whole. Even parts of the nervous sysogy for any implications they might have for adaptive tem, namely, those that control internal rather than performance and behavior. An appropriate point of externally directed functions (i.e., autonomic), may departure for considering these issues is therefore an depart radically from bisymmetry. Also, as we shall inquiry into the evolutionary origin of these asymme- further consider subsequently, the cerebrum in its tries. Why did they evolve and what might have been most elaborated form in the human is not strictly their relations to the ongoing response of organisms bisymmetric. to environmental pressures? If we could determine what role asymmetry of function plays in adaptive Asymmetry (Somatic) behavior, then we could better predict the ways in which lack or distortion of such asymmetry might Minor asymmetries abound and have been docaffect the functioning of the human infant, child, and umented in all species studied in sufficient detail (Ludwig, 1932). An instructive example relates to adult.
HEMISPHERE SPECIALIZATION IN CHILDREN
the pelvic and pectoral fins of fishes. These, though bilateral, are asymmetric, it being the general rule that the right-sided fins are more bony and muscular than the left (Hubbs & Hubbs, 1944). This is the case even though the fishes' musculature itself is bisymmetric. This appears to be because their asymmetry poses no problem for the function of fins as rudders to direct efficient swimming movements. This is an example of an asymmetry that does not appear to have evolved to meet a specific adaptive need, but appears to exist because the engineering of exact bisymmetry was not needed to meet the particular adaptive pressures in the context of which the species evolved.
Asymmetry (Neural) In behaviorally simple organisms the functions of the nervous system distribute across two domains: the regulation of the internal environment and the control of behavior oriented in space. The former does not call for bisymmetric control, and certain striking brain asymmetries in behaviorally limited species may relate to vegetative function (Braitenberg & Kemali, 1970). In the more behaviorally sophisticated vertebrates, including mammals and birds, the repertoire includes a third domain: higher mental function as involved in communication, memory, and problem solving. Not being targeted toward specific locations in the physical environment, these processes can serve their purpose without being bilaterally represented. Whereas central representation of sensorimotor processes is topographic, representation of higher mental functions is abstract (J. z. Young, 1962). If an abstract representation deviates, even substantially, from bisymmetry, this need not be because the asymmetric topography confers specific adaptive advantage. The relaxation of need for bisymmetry may sufficiently account for deviation from bisymmetry (Kinsbourne, 1974). A striking example of asymmetry of brain representation of function is to be found in certain male songbirds whose song is largely controlled from the left brain (Nottebohm, 1971). The relevant brain area (hyperstriatum ventrale, pars caudale) is also represented on the right side of the brain, though in the intact organism it is not known what its functions are on the right. If the left-sided control of song is destroyed or disconnected from its effector, the right side is capable of taking over control of bird song if the lesion is made in the young organism before singing has fully developed. Here we have a behavior that has a communicative role, and is targeted not toward a particular point in space but simply at ambient
71
space. Bilateral control of such a function would not seem to be necessary and in fact unilateral control prevails. Nevertheless, the mirror image area does develop and is available as a reserve. This is not to suggest that it is there in order to be in reserve in case of left-sided brain damage. Rather, nature is conservative in the manner in which it refines neural control mechanisms, and there was perhaps no provision (or environmental adaptive trigger) to preclude the generally unutilized right-sided area from evolving in parallel with the left. This issue is of interest because there are human phenomena that can be similarly interpreted, notably the left-sided control of speech. If the left speech area is totally destroyed, the right side does seem able to control speech output (Searleman, 1977, 1983). Should we suppose that in the intact state the unilateral facility in actual control of behavior maintains its control by virtue of actively suppressing (inhibiting) its potential rival on the other side (Kinsbourne, 1974)? If this mechanism exists, then its impairment could account for certain forms of behavioral deficit. The analogy with the bird song captures one attribute of human brain organization, the usual virtual restriction of the control of a skill not targeted to a specific external location to one side of the brain. It fails to capture another attribute: the complementarity of human hemispheric specialization (Kinsbourne, 1982). A much simper animal model illustrates the device of complementary specialization. The paired claws of the lobster differentiate into a stout crusher driven by slow muscle fibers, and a slender cutter, largely driven by faster muscle fibers (Govind & Pearce, 1986). The asymmetry develops under central neuron control, and is mediated by lateral differences in the degree of reflex activity. In humans, the cerebral hemispheres supply differentiated but complementary components to skilled behavior, to the point that many real-life activities simply cannot be effectively controlled by one hemisphere alone. Bresson, Maury, Pierant-LeBonniec, and deSchonen ( 1977) found that human infants prefer the right hand for some activities, the left for others. Whether to press the analogy with the lobster one step further is conjectural. Govind and Pearce found that exercising one claw facilitated its development into the crusher. In human, we have no comparable observations. But experimentally controlled studies of this kind are hardly practicable. We have mustered evidence that a child's first words are accompanied by pointing to the named object (Kinsbourne & Lempert, 1979; Lempert & Kinsbourne, 1985). Whether manipulating the side on
72
CHAPTER4
which attention-attracting stimuli appear, or the limb used to point, would qualify the preprogrammed left brain speech laterality is doubtful. But if, as Annett (1973) suggested, sinistrals lack an overriding "right shift factor," it is quite possible that in them lateralization could be influenced in this manner. When one applies these considerations to the human brain, one learns that lateralization of higher mental functions cannot be assumed to be adaptively necessary simply because it happens to be the general rule. Whether the organism that deviates from the norm of lateralization pays a penalty in terms of behavioral control is an empirical issue rather than one that can be taken for granted. It follows that our discussion will have to consider. separately the following two questions: under what circumstances do humans deviate from the usual laterality patterns and when they do, what if any are the consequences for adaptive behavior?
Morphological Asymmetries in the Human The internal organs of the human are subject to well-known major asymmetries. Deviation from the usual pattern has also been well documented, both in the form of a complete lateral reversal (situs inversus), and in terms of such deviations from asymmetry as horseshoe kidney in which the kidney is a single bilaterally symmetric organ connected across the midline. Deviations from the normal position of the internal organs may generate mechanical difficulties, but the functioning of the organs themselves does not seem to depend on their location. In the human brain and musculature, a number of less radical asymmetries exist. No one of them has been convincingly tied to function. The right-hander's body is subtly asymmetric. Most bones and muscles on the right are reported to be somewhat more massive than on the left (Latimer & Lowrance, 1965) and there is evidence that this is not secondary to differential use, as this state of affairs already obtains in the infant (Pande & Singh, 1971). Asymmetries in fingerprints, hair whorls, and other ectodermal structures have been documented. More relevant to brain, the skull is more protuberant anteriorly on the right and posteriorly on the left (LeMay, 1976; LeMay & Culebras, 1977). Correspondingly the right frontal lobe and the left occipital lobe are somewhat bulkier than the corresponding areas on the other side. Function could be inferred from these mor-
phological findings (Galaburda, 1984) if the amount of brain mass in a given area were to correlate with the efficiency with which the individual performs the activities that this area is specialized to control. Also, individuals who lack the asymmetry in question should have a correspondingly different profile of functional capabilities. The evidence is far from conclusive along either of these two lines. In particular, by far the greatest variability with respect to relative size of parts of right and left brain is to be found in the non-right-handed population (McRae, Branch, & Milner, 1968; Hochberg & LeMay, 1975). No one has been able to demonstrate any functional differences between right-banders and non-right-banders that could stem from these morphological variations among the latter. That mere bulk of brain may not be a good index of functional efficiency is not unexpected as the amount of normal variance in intelligence accounted for by overall brain size, though significant, is quite slight. Also, the greater bulk of the male than the female brain is not accompanied by an overall greater intellectual capability. A more refined measure of brain size would perhaps take account of local differences in the amount of infolding of cortex, the gray matter being organized around the folds. There is a dissociation between the size of Broca's area, which is greater on the right, and its infolding, which is greater on the left. However, other areas known to have more bulk on the left side in right-banders, notably the planum temporale (Geschwind & Levitsky, 1968), have not been studied in this way. So we do not know whether, for instance, the right planum is more infolded than the left. In any case, bulk of a brain area by no means necessarily reflects the number of neurons. There is variation in how tightly packed neurons are, let alone in the richness of their connections or the excellence of their organization and normality of their morphology. In the four cases documented by Galaburda, Sherman, Rosen, Aboitiz, and Geschwind ( 1985) in whom dysgenesis of neurons in various left cerebral areas was found at autopsy, brain bulk as observed on CT scan did not deviate from the norm. A further impediment to linking brain asymmetry with differential skill in higher mental functions derives from comparative data. Yeni-Komshian and Benson ( 1976) showed that the planum temporale is larger on the left than on the right in chimpanzees, a species not noted for its verbal ability. In summary, although it is interesting that morphological asymmetries are "invariant" across development, they have not been validated as indices of function, and correspondingly their existence in the newborn (Witelson
HEMISPHERE SPECIALIZATION IN CHILDREN
& Pallie, 1973; Wada, Clark, & Hamm, 1975) cannot be used as evidence that language precursors are lateralized.
Precursors of Lateralization of Function Peripheral Laterality The infant is not capable of the activities that are used to classify more mature individuals into those who are right-handed and those who are not. However, certain biases in infant motor behavior may be precursors of hand preference. The newborn infant is· not capable of behavior so differentiated as to involve the use of one hand and arm only. But within his or her repertoire is a lateral orienting synergism, the asymmetric tonic neck response, which includes turning of head and eyes to one side, extension of the ipsilateral arm and leg and flexion of the contralateral arm and leg. This can be seen as a precursor of locomotion toward one side, though the infant is lying supine. The outstretched arm may be a precursor for reaching and pointing. Be that as it may, Gesell and Ames (1947) frrst observed that spontaneous head turning in infants is more often to the right than to the left, and in a follow-up study with a small sample they found a relationship between the direction of the most frequent head turning in the infant and subsequent hand preference. Notably, all four of the infants who showed predominant leftward head turning subsequently became left-banders. A later study using a larger sample has not found quite so clear an outcome, and in particular, instances of predominantly leftward turning are few. Nevertheless, Liederman and Kinsbourne (l980a) were able to show that head turning asymmetry represents a motor, and not a sensory, bias, and it is indicative that an overall rightward turning bias was found in children of right-handed parents but not in a group of children with one non-right-handed parent (Liederman & Kinsbourne, l980b). It is possible that asymmetric head turning takes place even in utero. The most frequent presentation of the fetal head at birth is left occipita-anterio r (LOA). This indicates that the infant's head is most often turned to the right as it descends (headfirst and backward relative to the mother) through the birth canal. Churchill, Igna, and Senf ( 1962) reported that more LOA than ROA babies turn out to be right-handed at age 2 years. They attributed this to hypothesized hemisphere injury by pressure against the pelvic floor-right hemi-
73
sphere in LOA, left hemisphere in ROA. This does presuppose a staggering amount of birth-related cerebral damage. If the child predisposed to become dextral has a more vigorous rightward turning tendency, even in utero, and vice versa, and this is one determinant of the presentation of the fetal head, the findings can be accounted for without invoking uncorroborated pathology. Subsequent work has shown that 2- and 3month-old infants grasp an object longer with the right than with the left hand (Caplan & Kinsbourne, 1976; Hawn & Harris, 1983) and at 5 months reach more frequently to the right (Cohen, 1966; Seth, 1973; Hawn & Harris, 1983). After pointing has emerged toward the end of the first year, it more frequently is accomplished with the right hand (Bates, O'Connell, Vaid, Sledge, & Oakes, 1986). The situation is complicated by evidence that the hand preferred for activities at the appropriate developmental level fluctuates, perhaps systematically, within a subject during the frrst year of life (Halverson, 1937; Ramsey, 1984; Liederman, 1983). It could be that this reflects epochs in which one or the other hemisphere is in a phase of relatively more active development. A mechanism possibly relating actively developing brain to frequency of corresponding hand use was provided by Kinsbourne ( 1970) who proposed an activational model by which activities in a given hemisphere overflow to hemispheric facilities not primarily involved in the activity in question. Be that as it may, there is now good circumstantial evidence for a developmental sequence of peripheral laterality arising from the asymmetric tonic neck response first evident after the intrauterine age of 32 weeks (Turkewitz, 1977). Contrary to the preexisting concept of handedness emerging from diffuse movement patterns in infancy, its antecedents are already differentiated at birth. In summary, a motor bias that in most individuals is targeted rightward clearly exists as early as at birth or even before, and exerts a major influence on the side of the subsequently preferred hand.
Central Laterality Several infant studies have documented differential response to speech and nonspeech input depending on its side of origin. Entus (1977) used the paradigm of high-amplitude sucking to indicate orienting to a change in stimulus state. Infants 2! months old demonstrated their habituation to a constant sound source by discontinuing the sucking. If that source changed in nature discriminably, the
74
CHAPTER4
sucking was dishabituated. Entus presented tape-recorded speech and music. Given changing speech sounds the interruption of sucking happened earlier if the change occurred in the sound presented to the right ear and given music the same was true for the left. Best, Hoffman, and Glanville (1982) presented similar findings using a heart rate dishabituation paradigm, as did Molfese, Freeman, and Palmero (1975) using amplitude of evoked potential. Amplitudes were higher over the left brain for speech, over the right brain for music, in newborns. MacKain, Studdert-Kennedy, Spieker, and Stem (1983) found infants better able to coordinate visual and auditory aspects of an observed speech act when turning right toward it than when turning left. Segalowitz and Chapman (1980), studying premature infants, found that a verbal input caused a quieting of movements of the right arm and leg and musical input a quieting on the left. These findings are fragmentary and qualitative and not always confirmed (Shucard, Shucard, Cummins, & Campos, 1981). Nevertheless, the cumulative evidence demonstrates that precursors of aspects of verbal behavior are observably present and lateralized as predicted by the invariant lateralization hypothesis as early as at or even before birth. Given the immature and not yet functional state of the cerebrum at that stage in the life span, it becomes evident that the asymmetry must be laid down at a brain stem level. Evidence for an involvement of left thalamic nuclei in verbal behavior and of right thalamus in visual behavior (Ojemann, 1977) fits well with the notion that brain stem mechanisms are involved, if not in the actual mental processing, which is not available to the infant, then in facilitating its occurrence, perhaps by implementing lateralized ascending activation of cortex. The notion that there are lateralized selector mechanisms at a brain stem level, involved in implementing a categorical (hemispheric) mental set (Kinsboume, 1980) could be used to explain difficulties with particular categories of thinking exhibited by children with learning disability, to be discussed subsequently.
Lateralization of Function Emergence of Hand Preference in Children Even when the child is capable of reaching,
grasping, and pointing, movements analogous to the
activities observed in older people when hand preference is determined, the choice of hand used is under
control of factors that will cease to operate as maturation proceeds. Notable is the tendency of infants not to cross the midline when they reach for things (Provine & Westerman, 1979). If the target is slightly to one side of center, the child will reach with the ipsilateral limb regardless of hand preference. It is not that there is any motor constraint on crossing the midline. Ifthe child is already holding a desired object with the ipsilateral arm, he or she does cross the midline in picking up the target with the free hand (Hawn & Harris, 1983). The "prewired" tendency to orient to the side of stimulation (use the hand ipsilateral to the target) in the infant overrides motor preference. This could account for observations such as those of Goodwin (cited in Liederman, 1983) who found that right-hand preference on a reaching task at 19 weeks strongly predicted hand preference at 3 years, but left-hand reaching preference did not. The tendency not to cross the midline seems to be a consequence of brain organization. As such it has to be taken into consideration when failure to establish hand preference occurs in developmentally delayed individuals as the latter may perhaps be subject to developmental lag, the individual remaining under the control of the type of limiting factor that invests the normal infant but not the normal older child. Even when it becomes possible to observe the child's choice of hand in a number of standard unimanual activities, the young child differs from the older individual in sometimes being inconsistent in which hand is used for which activity (a factor separate from the question of which hand is preferred for activities overall). Palmer (1964) observed this so-called ambiguous hand preference (Silva & Satz, 1984) in normal children and we will return to it later in the context of children with mental retardation and autism. Consistent hand preference tends to be established in the preschool years and persist unless the individual is subjected to contrary cultural pressure. Until recently it was customary to encourage if not constrain a child who showed left-handed tendencies to use his right hand instead, generating a miscellaneous series of shifted sinistrals who would use the right hand at least for those activities that are socially conspicuous like writing and holding tableware. Such people might still exhibit more dexterity on the left and their left-hand preference may be revealed by giving them a novel activity to perform. Nowadays this type of pressure has been relaxed in the West, but still persists in the Orient (Teng, Lee, Yang, & Chang, 1976). For this reason, in the West, offspring have a higher probability of being non· right-handed than their parents. Levy (1976) reported that left-banders were 2.2% of the U.S. population in
HEMISPHERE SPECIALIZATION IN CHILDREN
1932, but more than ll% by 1972. This presumably is an effect of relaxation of cultural pressures. Recently documented within the left-handed population is the position of pen in hand, a distinction being made between the inverted position in which the point of the pen is below the tip of thumb and index finger and the noninverted in which the pen is held in the same way as right-banders hold their's. The inverted handwriting posture is considerably more common in males and develops during the grade school period. It is of interest in view of evidence that it reflects certain biological differences in brain organization (Levy & Reid, 1976). With respect to motor behavior the noninverter exhibits a more bilateralized or right hemispheric type of control and the inverter seems to be more ambidextrous (Parlow & Kinsbourne, 1981). The degree of transfer of training between hemispheres also differs between these two sinistral subgroups (Parlow & Kinsbourne, submitted). The prevalence of inverted writing position among left-handed developmentally disabled individuals is not known. It might be of interest, however, because Searleman, Porac, and Coren (1982) found it more common in subjects with a history of birth stress.
Development of Central Laterality Most of the evidence derives from two sources, lateralized brain lesion effects and laterality testing in normal children, and relates to the language function. Laterality paradigms are designed to reveal, in terms of differential performance, effects of functionallateralization on the control of behavior. The fact that a hemisphere is dominant for a particular mental operation is revealed by a bias in how efficiently the relevant task is performed when the pertinent informational signals originate from the right as compared to the left side of the organism. The hemisphere is also in control of the elements of contralateral turning: faster orienting to that side, with consequent faster information pickup; faster response in that direction by limb or gaze. Thus, the task-specific activation of the specialized hemisphere introduces a slight but observable contralateral turning bias (Kinsbourne, 1972). Asymmetry of laterality outcome is the performance consequence of that bias (Kinsbourne, 1970, 1973). In ontogeny, perceptual asymmetries can only begin when children activate one hemisphere more than the other in a particular test situation. It is not necessary that the hemisphere be already specialized
75
along these lines, and actually processing. Laterality effects appear even before the hemispheres are myelinated. As long as the strategy that the child attempts does not change over time, the attendant lateral asymmetry will also not change. For some tasks, however, strategies do seem to change over the childhood years. Particularly in visuospatial tasks, young children may deploy a mixed strategy involving both sides of the brain. Older ones rely more exclusively on the right hemisphere. Thus, a left lateral "advantage" only gradually emerges. That does not imply that the right hemisphere's specialization has only at that point in time progressed to a usable degree of maturation. Indeed, there is evidence from infant studies of precursors to that specialization. Laterality effects are event-related measures of hemispheric usage: hemispheric usage may or may not correspond to hemispheric specialization. It follows that a lateral asymmetry may be present throughout development, indexing the invariant presence of the corresponding specialization. Or it may appear at a certain age. If so, it requires a further research effort to determine whether its appearance is due to a recent lateralization of the mental operations to the indicated hemisphere, or a recently acquired greater reliance on that hemisphere for tasks of the type set. A rather straightforward index of the balance of lateral brain activation is the direction of selective orienting. Kinsbourne (1972) demonstrated that verbal thinking generates righward orienting (gaze and head turning), whereas spatial thought occasioned more left than right orienting behavior. Conversely, rightward head turning was found to facilitate verbal memory relative to left turning (Lempert & Kinsbourne, 1982). This method has as yet been little used in developmental studies. However, Barrera, Dalrymple, and Witelson (1978) did report more left gaze during visual processing offaces by infants, and MacKain et al. (1983) found infants better able to map visual upon auditory components of speech signals when orienting rightward. The most generally used laterality measures for input processing are dichotic listening and visual half-field viewing. In dichotic listening, speech sounds, syllables, or words are simultaneously presented to both ears and the subject is either asked to report as much as he or she can of what is heard ("whole report") or asked to listen selectively under two conditions, to the right ear only and to the left ear only ("selective listening"). In the first case a laterality index is computed to represent the extent to which the subject is able correctly to report input through the right ear as compared to the left. In the
76
CHAPTER4
second case the subject's ability to identify input from the specified ear is compared for the two ears, as well as the incidence of responses that represent interference from the ear not to be attended. Normal right-handed adults as a group exhibit right ear advantage, i.e., they are better able to identify material presented to the right ear than the left under both whole and selective reporting conditions. More intruding stimuli from the right ear are normally reported when selectively listening to the left than vice versa (Treisman & Geffen, 1968). In children as young as age 3, a right ear advantage has been repeatedly demonstrated (Nagafuchi, 1970; Ingram, 1975; Kinsbourne & Hiscock, 1977; Piazza, 1977). Thus, a preponderance of left hemisphere responsibility for verbal auditory processing can be accepted. The question remains: is the degree of this effect as great in children as in adults or is it that although lateralization has already occurred to some extent by age 3, it will subsequently further increase? Whereas the direction of group mean ear advantage is an acceptable index of the side of the cortex dominant for processing the material in question, the degree of asymmetry is a dubious index for "degree of lateralization," which is itself a dubious concept. There are numerous factors that interact with differential hemispheric specialization to generate ear advantages differing in degree, even within the same subject tested under different circumstances or with different dichotic test material. The test-retest reliability for dichotic listening ranges between about 0.5 and 0.8 (e.g., Bakker, Vander Vlugt, & Claushuis, 1978; Hiscock & Kinsbourne, 1980a), hardly commensurate with an index of a fixed structural characteristic. Direction of gaze and direction of movement in the visual environment are both capable of influencing the degree of right ear advantage (Hiscock, Hampson, Wong, & Kinsboume, 1985). Perhaps more important than any of these is task difficulty. The extent to which items from one ear have to be kept in memory while those from another are reported (Inglis & Sykes, 1967) can be a major factor if there is a bias to report a particular ear first (Bryden & Allard, 1981). For all these reasons it would not have been immediately clear how to interpret any interaction between age of child and degree of right ear advantage for verbal material, had such been found. In fact, most competent studies failed to find such an interaction. Instead, the degree of ear advantage is roughly invariant, consistent with the invariant lateralization hypothesis. The proportion of interfering responses from right versus left ear is also invariant across a wide age range in childhood (Hiscock &
Kinsboume, 1977, 1980a; Geffen, 1978; Geffen & Wale, 1979). It follows that the hypothesis of progressive lateralization gains little support from dichotic listening studies both in its original strong form, positing a gradient of lateralization extending until puberty (Lenneberg, 1967), and in its modified form, restricting that gradient to the first 5 years of life (Krashen, 1973). Iflateralization develops at all, its development is completed by age 3 (Porter & Berlin, 1975), the youngest age at which it is feasible to perform dichotic testing in the conventional manner. However, Lokker and Morais ( 1985) tested chil- · dren aged 1f-3 years old dichotically, using selective reaching for an object as the response. They too found a right ear advantage for children of righthanded parents. The visual method of laterality testing is less generally applicable to young children because it relies on the written word and therefore calls for a degree of reading skill that is not developed in most preschool children and only partially developed in the early grades. If grade schoolers are presented with words that they can easily read, then the usual right half-field advantage was found regardless of age by Marcel and Rajan (1975) and Lewandowski (1982). Studies that find progression in the development of the right field advantage (Forgays, 1953; Miller & Turner, 1973; Carmon, Nachson, & Starinsky, 1976; Reynolds & Jeeves, 1978; TomlinsonKeasey, Kelly, & Burton, 1979) are so much in conflict with each other in detail that their diverse outcomes must reflect methodological differences. In any case, no other findings support progressive lateralization into adolescence. The method used for determining lateralization of speech output control in the intact individual is that of verbal-manual interference (Kinsbourne & Cook, 1971; Kinsbourne & Hicks, 1978). Subjects perform a unimanual activity, such as speeded repetitive finger tapping, with one hand or the other, with or without concurrent speaking. If speech control is lateralized, speaking interferes disproportionately with the finger tapping controlled by the same hemisphere (i.e., left lateralized speech will interfer more with right than with left finger tapping). When this paradigm is used, there is already at age 3 differential interference with right-hand performance, indicating that speech is already lateralized to the left at that age. Moreover, for the age range 3 to 12 years, there is no interaction between degree of asymmetry and age, supporting invariance of lateralization for speech production (Hiscock & Kinsboume, 1978, 1980b; White & Kinsbourne, 1980). Less is known about the ontogeny of lateraliza-
HEMISPHERE SPECIALIZATION IN CHILDREN
tion for those nonverbal activities that are regarded as right lateralized. Piazza (1977) found a left ear advantage for the dichotic presentation of environmental sounds in 3-, 4-, and 5-year-olds, and Saxby and Bryden (1984) confirmed this for 5-year-olds (although in an earlier study, Knox and Kimura ( 1970) found somewhat weaker left ear effects in 5- and 6than in 7- and 8-year-olds). For tachistoscopic face recognition, left visual field advantages are found in quite young children, unaffected in degree by age (Marcel, Katz, & Smith, 1974; Young & Ellis, 1976; Young & Bion, 1980; Turkewitz & Ross-Kossak, 1984). With respect to the ability to discriminate shapes by active touching (haptic perceptionWitelson, 1974), the typical left-hand advantage has been found as early as age 2-3 (Rose, 1984). Other studies revealed left-hand advantages for nonsense shapes in preschoolers (Etaugh & Levy, 1981) and grade-schoolers (Witelson, 1974, 1976; Coiffi & Kandel, 1976; Affleck & Joyce, 1979; Flanery & Balling, 1979; Klein & Rosenfield, 1980), only Flanery and Balling finding a developmental trend. Recognition of verbalization shapes yields less consistent data, perhaps because of variability in the degree of left hemisphere participation in the task (e.g., Witelson, 1974; Coiffi & Kandel, 1976). Invariance is generally supported for males, but there is a tendency of younger females to show a right- rather than left-hand advantage. But interpretation is confounded by the tendency of females more than males to use a verbal strategy in coding input (e.g., Caplan & Kinsbourne, 1981). The recognition of briefly exposed faces to either side of the midline is another relevant methodology. This yields no asymmetry until about age 7 or later, subsequent to which a left half-field advantage begins to emerge, earlier in boys than in girls (Carey & Diamond, 1977; Levine, 1985). This finding illustrates a fundamental issue in interpretation. Whereas the gradual emergence of an asymmetry could indeed be interpreted in line with progressive lateralization, it could equally well be interpreted as indicating the emergence of a processor that had not been represented in the less mature brain. If space perception calls for particular processing skills that normally only emerge toward the end of the first decade of life, then presenting this task to younger children will yield a lack of asymmetry by default, rather than indicating that at that age both hemispheres were processing the material in question to a comparable extent. Carey and colleagues have found support for the view that the manner in which children process face information changes qualitatively at the time that the left half-field advantage emerges.
77
If this is so, then what is being indexed is the increasing specialization of the right hemisphere with increasing age, and not a changing neural basis for an already available skill. With further increase in age the degree of left field advantage will not increase. Indeed, it may temporarily diminish at around puberty, in concert with a decrease in performance efficiency (Diamond, Carey, & Back, 1983). This finding, perhaps related to endocrine changes around puberty, is in its details beyond the scope of the present discussion.
Degree of Lateralization The absence of interaction between degree of lateral asymmetry and age in most studies simplifies the task of explanation. In addition to supporting lateralization invariance, it sidesteps the dilemma of interpreting between group differences in the degree of laterality bias in the same direction. The assumption that degree of lateral asymmetry indexes degree of lateralization of the critical task-related mental operation (Shankweiler & Studdert-Kennedy, 1967) has never been substantiated. Indeed, it is unclear what is meant by greater or lesser degree oflateralization. Does the distinction assume that both hemispheres participate in the task, though to a varying extent unequally? If so, are they redundant in their contribution, or complementary? If unilateral brain damage occurs, should the function in question be compromised by damage on both sides, in proportion to the degree of lateralization on each side? If so, no such intimation from lateral brain injury exists. Given the many factors that, for instance, modify asymmetry in a dichotic test-task diffiCulty, task aptitude and motivation, the extent of stimulus dominance, and perhaps whatever else the subject is thinking about and how (happy or sad) he or she is feeling-it is hardly surprising that the literature on degree of lateralization is inconsistent in the extreme. Two related areas to which the concept has been vigorously applied are gender differences and age at puberty difference in degree of lateralization. It is little wonder that the literature in both fields (McGlone, 1980, and peer commentary; Newcombe & Bandura, 1983, respectively) is a morass ofinconsistencies. Operationally, laterality tests (and lateral brain damage effects) can only guide us in a choice between three alternatives: left lateralized, right lateralized, bilateralized. An isolated but intriguing finding relates to lateralization of emotion, which is right-sided in adults
78
CHAPTER4
(Schwartz, Davidson, & Maer, 1975). In infants less than 1 year old, Davidson and Fox ( 1982) monitored power spectrum EEG changes while children discriminated faces. They offered evidence that even at this early age the right hemisphere is more involved than the left in the discrimination of facial affect.
hemisphere territories are more likely to assume a compensatory role the earlier the lesion is, or perhaps the more extensive it is. It may be that right hemisphere territories that normally play some other role are preempted for purposes of the compensatory functioning. Huttenlocher et al. (1986) hemispherectomized infant rats. In this preparation an ipsilateral corticospinal tract develops in addition to the normally present contralateral tract. The investigators Lateralization Probed by Lateral subsequently found that the cortical area that conCerebral Damage tained pyramidal tract cells had greatly expanded, in response to the excision of the contralateral If language lateralization is invariant, then left hemisphere. brain damage should be equally likely, and right Supportive evidence derives from studies of brain damage equally unlikely, to cause aphasia in right hemisphere lesion effects in childhood (Kerright-handed children as in adults (acutely-whether shner & King, 1974; Kohn & Dennis, 1974; Ferro, mechanisms of compensation vary with age is a sepa- Martins & Tavora, 1984; Stiles-Davis, Sugarman, & rate issue). Contrary to earlier impressions (Basser, Nass, 1985). They all found spatial deficits analo1962), this is approximately the case. In Woods and gous to those observed in adults after right hemiTeuber's (1978) series, the incidence of aphasia in sphere damage. children aged 2-14 was about 70 versus 7% percent We conclude that the precursors of language for left- and right-sided damage, respectively, in function rely on activity of the same hemisphere that right-banders. The implication that the left hemi- subserves language in its full maturity. The earliest sphere is specialized early for language is corrobo- manifestation is perhaps a selective activation of that rated by series of children who suffered lateral cere- hemisphere in a verbal context, well before its neural bral damage before the onset of language behavior. substrate has matured to the point that language proThe long-term language outcome is less favorable if cessing is feasible. As language ability differentiates, the early damage was left-sided (Kershner & King, language processing may involve, not a shrinking, 1974; Rankin, Aram, & Horwitz, 1981; Kiessling, but an expanding neural base within that same hemiDenckla, & Carlton, 1983; Varga-Khadem, O'Gor- sphere (Satz & Strauss, 1986). man, & Watters, 1985), at least in terms of syntactic The impressively consistent evidence in favor of proficiency (Dennis & Kohn, 1975; Rankin et al., invariant lateralization for the major functions of 1981; Aram, Ekelman, Rose, & Whitaker, 1985). both hemispheres offers a conveniently simple stanBut the selective syntactic difficulties are somewhat dard of reference against which to evaluate the posovershadowed by the recurrent finding that early lat- sibility that children with a variety of developmental eral damage on either side, even when strictly later- disabilities are characterized by anomalies in lateralized, leads to a moderate impairment across a wide alization, and that these anomalies have a bearing on range of test performances (Aram et al., 1985). This the causation of the behavioral deficit. perhaps indicates that brain organization is not strictly modular, but does draw upon distributed as well as focalized neural processing, at least for purposes of Lateralization in Developmental mental development. Although there is every reason to suppose that Deficit lateralization of language remains invariant throughout its development, the locus of compensation after Introduction damage to the language area of the child's brain does Perhaps because developmental deficits offer so not. Penfield and Roberts ( 1959) reported on the incidence of aphasia after left temporal lobectomy in few clues beyond the surface phenomenology for the epileptic adults with early lesion onset. When lesion reason why they came to be, lateralization has often onset was prior to age 2 years, the probability of been invoked as a possible factor in their pathoaphasia was much less than when the damage had genesis. Almost always it is suggested that the noroccurred subsequently in childhood. It appears that mal lateralization of hemispherically specialized territories in both hemispheres have the potential to cognitive functions failed to occur, the assumption compensate for injury to the language area. Right being that cognitive processing when based on both
HEMISPHERE SPECIALIZATION IN CHILDREN
hemispheres is relatively primitive and necessarily inefficient. The logic of the suggestions depends crucially on the notion that lateralization is normally progressive and that it is this progression, when impaired, that leaves a cognitively deficient end result. As we have seen, the evidence for progressive lateralization is lacking, so that if anomalies of lateralization are to be found in developmental disabilities, some other explanation for such findings has to be sought. In fact, there is plentiful evidence for a relatively greater incidence of unusual forms of lateralization in developmental deficits, but the causes and implications of these differences remain quite obscure. There is a coincidence between absence of the usual left lateralization of language in various disabilities and an increased prevalence of non-righthandedness in the same conditions (though the central and peripheral laterality anomalies by no means correlate perfectly). We first consider the data on hand preference.
Hand Preference in Developmental Disabilities Not only is non-right-handedness more common among the mentally retarded, autistic, and language and learning delayed, but the non-righthanded subgroups of these populations tend to be more severely affected by the deficit in question. For example, Hicks and Barton (1975) found severe and profoundly retarded individuals to be still more often non-right-handed than mild and moderate, who in tum were more non-right-handed than the general population. Bradshaw-McAnulty, Hicks, and Kinsboume ( 1984) confirmed and extended this finding, and further related greater severity of the mental retardation to a greater probability of right-handedness in one or other parent. In infantile autism, nonright-handedness is particularly prevalent (Colby & Parkinson, 1977), and several investigators have found the more lower functioning individuals to be more often non-right-handed (e.g., Fein, Humes, Kaplan, & Lucci-Waterhouse, 1984) (there being general agreement that in autism non-right-handedness is far more common than in the general population (e.g., Tsai, 1982)). Indeed, in mental retardation, and particularly in autism, hand preference even for a single activity is apt to change from trial to trial (ambiguous handedness, according to Silva & Satz, 1984; see also Soper eta/., 1987). Non-right-handedness is relatively common in stuttering and in language delay. In selective reading disability (dyslexia), non-right-handedness is relatively prevalent,
79
again more particularly among the most severely affected children, who are to be found in clinical settings and in special schools for the learning disabled (Satz, 1976). Several sharply contrasting explanations for this conjunction of findings have been offered. (1) Presuming that peripheral non-right-handedness implies a corresponding absence of centrallateralization, the latter itself is incriminated as inducing a processing inefficiency (Orton, 1937). (2) Some left-handedness (Satz, 1972) or left-handedness as such (Bakan, 1971) is as a consequence of early left hemisphere pathology (syndrome of pathological left-handedness of Satz, Orsini, Saslow, & Henry, 1985), and such early pathology is also likely to set up a miscellany of developmental disabilities. (3) An influence active early in development is postulated that tends both to diminish language lateralization and to impair the evolving function of the language hemisphere (Geschwind & Behan, 1982). (4) Susceptibility to becoming non-right-handed and to suffering from a wide range of developmental disorders are consequences of an adverse influence on the fetal brain which are apt to occur when the mother is susceptible to diseases of the immune system (Kinsboume, in preparation). The second and third of these models are restricted in their explanatory value to those developmental deficits that can plausibly be attributed to malfunctioning of the left (language) hemisphere, rather than of the cerebral cortex as a whole. Language and reading disabilities are a case in point, and autism has been similarly regarded (Rutter, Bartak, & Newman, 1971) although the evidence against this is now very strong (Fein et al., 1984). For mental retardation and perhaps autism, in which conditions a more general cerebral deficit seems likely, explanations (2) and (3), targeted on the left hemisphere, lose force. However, there is nothing implausible that more than one of the above postulated mechanisms might come into play. For instance, there is circumstantial evidence that right- as well as left-handed members of relatively sinistral families are more at risk for developmental deficit, or for having more severe deficit should damage occur (Kinsboume, 1986). The damage to which such individuals are vulnerable could in tum, when it implicates the left hemisphere, not only impair cognition, but also cause a shift of hand preference phenotype to sinistral, in a genotypic dextral (pathological left-handedness), and thus increasing the prevalence of non-right-handedness among the affected family members to beyond the level that prevails within their already relatively sinistral family (Bradshaw-McAnulty et al., 1984).
80
CHAPTER4
Central Laterality in Developmental Deficits The measurement of central laterality requires a degree of cooperation from subjects, and perhaps for that reason, has been most often attempted in a relatively mildly disordered population-the learning disabled. These consist of two main subgroups: children with attention deficit and children with control processing difficulties (Kinsbourne & Caplan, 1979). A sample of the former was found to be normally lateralized by a dichotic (Hiscock, Kinsbourne, Caplan, & Swanson, 1979) and a visual (Naylor, 1980) laterality test. Therefore, studies have concentrated on the latter, specifically the reading disabled subgroup (''dyslexics'') under the influence of a persistent theoretical approach that incriminates failure of left language lateralization in its pathogenesis (Orton, 1937; Geschwind & Behan, 1982). According to this view, the visual right side advantage for verbal laterality tests should be lacking in the dyslexic. Although this has sometimes been reported (Olson, 1973; Zurif & Carson, 1970), enough studies have found normal laterality for verbal test in learning-disabled children (e.g., Marcel et al., 1974; Marcel & Rajan, 1975; McKeever & Van Deventer, 1975; Bouma & Legein, 1977; Caplan & Kinsboume, 1982) to indicate that failure of leftsided language lateralization is not a viable explanation for selective reading disability. Several sharply contrasting explanations for this conjunction of findings have been offered. (1) Presuming that peripheral non-right-handedness implies a corresponding absence of centrallateralization, the latter itself is incriminated as inducing a processing inefficiency (Orton, 1937). (2) Some left-handedness (Satz, 1972) or left-handedness as such (Bakan, 1971) is as a consequence of early left hemisphere pathology (syndrome of pathological left-handedness of Satz, Orsini, Saslow, & Henry, 1985), and such early pathology is also likely to set up a miscellany of developmental disabilities. (3) An influence active early in development is postulated that tends both to diminish language lateralization and to impair the evolving function of the language hemisphere (Geschwind & Behan, 1982). (4) Susceptibility to becoming non-right-handed and to suffering from a wide range of developmental disorders are consequences of an adverse influence on the fetal brain which are apt to occur when the mother is susceptible to diseases of the immune system (Kinsboume, in preparation). This does not imply, however, that the leftsided language facility is normal in dyslexic children. It could be structurally compromised, as indicated by
some case reports at autopsy (Galaburda et a/., 1985). Alternatively, the left hemisphere might be undersupplied by ascending activation, rendering it hard for the child to muster verbal skills in full force to solve what is for him or her a difficult verbal problem (Kinsboume, 1980). Such an activational insufficiency might also generate a relatively nonverbal (right hemispheric) cognitive style in such children (Caplan & Kinsbourne, 1982). Obrzut, Hynd, Obrzut, and Pirozzola (1981) found learning-disabled children better able than normally reading controls to listen selectively to left ear input. Obrzut, Hynd, and Zellner (1983) obtained comparable results in visual laterality. The voluntary attentional shift could override rightward attentional bias engendered by the presumably relatively weak left brain activation of the dyslexics. An electrophysiological approach has attempted to distinguish subgroups of dyslexics deficient in right and left hemisphere functioning, respectively (Bakker, Licht, Kok, & Bouma, 1980). Evoked potential studies lend support to the view that dyslexic children may exhibit abnormal responses in one hemisphere when tested, but the stability of those patterns has not been proven (Fried, Tanguay, Boder, Doubleday, & Greensite, 1981; Mecacci, Sechi, & Levi, 1983). If stable, they could reflect lateralized activational deficiencies. Stuttering is another early arising deficit in which maladaptive rivalry between the cerebral hemispheres for control of speech has long been suspected (Travis, 1927). The concept was dramatically supported by case studies of four left-handed stutterers with lateralized cerebrovascular congenital anomalies (Jones, 1966). They were found bilateralized for speech by intracarotid Amytal testing before operation. After operation, repeat Amytal testing showed that speech control had become restricted to the normal (unoperated) hemisphere. Also, after operation, the patients ceased to stutter. However, comparable studies of three more righthanded stutterers without brain damage have not yielded comparable findings (Andrews, Quinn, & Sorby, 1972), and both behavioral and EEG laterality were normal in stutterers (Pinsky & McAdam, 1980). As in dyslexia, it is more likely that in stuttering any cerebral abnormality is of a dynamic rather than static nature, for instance, an abnormality of left hemisphere activation for stuttered speech acts only. We await findings from event-related measures selectively time-locked to stuttered utterances (nonstuttered serving as control). The regional cerebral blood flow study of Wood, Stump, McKeehan, Sheldon, and Proctor (1980), in which stutterers were judged
HEMISPHERE SPECIALIZATION IN CHILDREN
to have inadequate left hemisphere attention (normalized when the stuttering was relieved by haloperidol), is a step in this direction. Lateralization in autism is a contentious issue (Kinsbourne, 1987). Both failure of lateralization and right brain dominance have been proposed, but it seems more likely that heterogeneous patterns of cerebral specialization occur within the autistic population, and that no one pattern of lateralization constitutes a necessary condition for autistic symptomatology to appear.
References Affleck, G., & Joyce, P. (1979). Sex differences in the association of cerebral hemisphere specialization of spatial function with conservation task performance. Jou1711ll of Genetic Psychology, 134, 271-280. Andrews, G., Quinn, P. T., & Sorby, A. (1972). Stuttering: An investigation into cerebral dominance for speech. JoU1711ll of Neurology, Neurosurgery arid Psychiatry, 35, 414418. Annen, M. (1973). Handedness in families. Annals of HU1111l11 Genetics, 37, 93-105. Aram, D. M., Ekelman, B. L., Rose, D. F., & Whitaker, H. A. (1985). Verbal and cognitive sequelae following unilateral lesions acquired in early childhood. Journal of Clinical and Experimei!Ull Neuropsychology, 7, 55-78. Bakan, P. (1971). Handedness and birth order. Nature, 229, 195. Bakker, D. J., Licht, R., Kok, A., & Bouma, A. (1980). Cortical responses to word reading by right- and left-eared normal and reading-disturbed children. Neuropsychologia, 2, l-12. Bakker, D. J., Vander Vlugt, H.• & Claushuis, M. (1978). The reliability of dichotic ear asymmetry in normal children. Neuropsychologia, 16, 753-758. Barrera, M. E., Dalrymple, A., & Witelson, S. F. (1978). Behavioral evidence of right hemisphere asymmetry in early infancy. Paper presented to the Canadian Psychological Association, Ottawa. Basser, S. (1962). Hemiplegia of early onset and the faculty of speech with special reference to the effects of hemispherectomy. Brain, 85, 427-460. Bates, E., O'Connell, B., Vaid, J., Sledge, P., & Oakes, L. ( 1986). Language and hand preference in early development. Developmental Neuropsychology, 2, 1-15. Best, C. T., Hoffman, H., & Glanville, B. B. (1982). Development of infant ear asymmetries for speech and music. Perception and Psychophysics, 35, 15-85. Bouma, H., &Legein, C. P. (1977). Fovealandparafoveal recognition of letters by dyslexics and average readers. Neuropsychologia, 15, 69-80. Bradshaw-McAnulty, G., Hicks, R. E., & Kinsboume, M. ( 1984). Pathological left-handedness and familial sinistrality in relation to degree of mental retardation. Brain and Cognition, 3, 349-356. Braitenberg, V. (1977). The concept of symmetry in neu-
81
roanatomy. Annals of the New York Academy of Sciences, 299, 186-196.
Braitenberg, V., & Kemali, M. (1970). Exceptions to bilateral symmetry in the epithalamus of lower vertebrates. Journal of Comparative Neurology, 138, 137-146. Bresson, F., Maury, L., Pierant-LeBonniec, G., &deSchonen, S. ( 1977). Organization and lateralization of reaching in infants: An instance of asymmetric functions in hand collaboration. Neuropsychologia, 15. 311-320. Bryden, M.P., & Allard, F. A. (1979). Dichotic listening and the development of linguistic processes. In M. Kinsbourne (Ed.), Asymmetricaljunction ofthe brain. London: Cambridge University Press. Bryden, M. P., & Allard, F. A. (1981). Do auditory perceptual asymmetries develop? Cortex, 17, 313-318. Bryden, M.P., Hecaen, H., &DeAgostini, M. (1983). Patterns of cerebral organization. Brain and Language, 20, 249-262. Caplan, P. J., & Kinsboume, M. (1976). Baby drops the rattle: Asymmetry of duration of grasp by infants. Child Development, 47, 532-534. Caplan, P. J., & Kinsboume, M. (1981). Cerebrallateralization, preferred cognitive mode, and reading ability in normal children. Brain and Language, 14, 349-370. Caplan, B., & Kinsboume, M. (1982). Cognitive style and dichotic asymmetries of disabled readers. Cortex, 18, 357-
366.
.
Carey, S., & Diamond, R. (1977). From piecemeal to configurational representation of faces. Science, 195, 312-314. Carmon, A., Nachson, I., & Starinsky, R. (1976). Developmental asj,ects of visual hemifield differences in perception of verbal material. Brain and Language, 3, 463-469. Churchill, J. A., Igna, E., & Senf, R. (1962). The association of position at birth and handedness. Pediatrics, 29, 307-309. Cohen, A. I. ( 1966). Hand preference and developmental status of infants. Journal of Genetic Psychology, 108. 337-345. Coiffi, J., & Kandel, G. (1976). Laterality of stereognostic accuracy of children for words, shapes, and bigrams: Sex differences for bigrams. Science, 204, 1432-1434. Colby, K. M., & Parkinson, C. (1977). Handedness in autistic children. Journal ofAutism and Childhood Schizophrenia, 7, 3-9. Davidson, R. J., & Fox, N. (1982). Asymmetrical brain activity discriminates between positive versus negative affective stimuli in ten month old infants. Science, 218, 1235-1236. Dennis, M., & Kohn, B. (1975). Comprehension of syntax in infantile hemiplegics after cerebral hemidecortification: Left hemisphere superiority. Brain and Language, 2, 475-486. Diamond, R., Carey, S.,&Back,K. J. (1983). Genetic influences on the development of spatial skills during early adolescence. Cognition, 13, 167-185. Entus, A. K. (1977). Hemispheric asymmetry in processing of dichotically presented speech and nonspeech stimuli by infants. InS. I. Segalowitz & F. A. Gruber (Eds.), Language development and ne.urological theory. New York: Academic Press. Etaugh, C., & Levy, R. B. (1981). Hemispheric specialization for tactile-spatial processing in preschool children. Perceptual and Motor Skills, 53, 621-622. Fein, D., Humes, M., Kaplan, E., & Lucci-Waterhouse, L.
82
CHAPTER4
(1984). The question of the left hemisphere dysfunction in infantile autism. Psychological Bulletin, 95, 258-281. FeJTO, J. M., Martins, I. P., & Tavora, L. (1984). Neglect in children. A111Ulls of Neurology, 15, 281-284. Aanery, R. C., & Balling, J.D. (1979). Developmental changes in hemispheric specialization for tactile spatial ability. Developmental Psyclwlogy, 15, 364-372. Forgays, D. G. (1953). The development of differential word recognition. Journal of Experimentol Psyclwlogy, 45, 165168. Fried, 1., Tanguay, P. E., Boder, E.,Doubleday,C., &Greensite, M. (1981). Developmental dyslexia: Electrophysiological evidence of clinical subgroup. Brain and Language, 12, 14-22. Galaburda, A.M. (1984). Anatomical asymmetries. InN. Geschwind & A.M. Galaburda (Eds.), Cerebral dominance: The biologicalfoundations. Cambridge, MA: Harvard University . Press. Galaburda, A.M., Sherman, G. F., Rosen, G. D., Aboitiz, F., & Geschwind, N. (1985). Developmental dyslexia: Four consecutive patients with cortical anomalies. Annals of Neurology, 18(2), 222-233. Gardner, H., Ling, P. K., Flamm, L., & Silverman, J. (1975). Comprehension and appreciation of humor in brain-damaged patients. Brain, 98, 399-412. Gardner, M. (1967). The ambidextrous universe. Baltimore: Penguin. Geffen, G. (1978). The development of the right ear advantage in dichotic listening with focused attention. Cortex, 14, 169177. Geffen, G., & Wale, J. (1979). Developmentofselectivelistening in hemispheric asymmetry. Developmentol Psychology, 15, 138-146. Geschwind, N., & Behan, P. (1982). Left-handedness: Association with immune disease, migraine, and developmental learning disorder. Proceedings of the National Academy of Science, 79, 5097-5100. Geschwind, N., & Levitsky, W. (1968). Human brain: Left-right asymmetries in temporal speech region. Science, 161, 186187. Gesell, A., & Ames, L. (1947). The development of handedness. Journal of General Psychology, 70, 155-157. Gloning, I., Gloning, K., Haub, G., & Quatember, R. (1959). Comparison of verbal behavior in right-handed and nonrighthanded patients with anatomically verified lesion of one hemisphere. Cortex, 5, 43-52. Govind, C. K., & Pearce, J. (1986). Differential reflex activity determines claw and clasper muscle asymmetry in developing lobsters. Science, 33, 354-356. Halverson, H. (1937). Studies of the grasping responses of early infancy. Journal of Genetic Psyclwlogy, 51, 371-392; 393424. Hawn, P.R., & Harris, L. J. (1983). Hand differences in grasp duration and reaching in two- and five-month-old infants. In G. Young, S. Segalowitz, C. M. Cotter, & S. E. Trehub (Eds. ), Manual specialization and the developing brain. New York: Academic Press. H6caen, H. (1978). Right hemisphere contribution to language function. In P. A. Buser & A. Rougeui-Buscr (Eds.), Cere-
bral correlates of conscious experience. Amsterdam: North· Holland. Hicks, R. E., & Barton, A. (1975). A note on left-handedness and severity of mental retardation. Journal of General Psychology, 127, 323-324. Hicsock, M., Hampson, E.• Wong, S.C. P., & Kinsboume, M. (1985). Effects of eye movements on the recognition and localization of dichotic stimuli. Brain and Cognition, 4, I 40155. Hiscock, M., & Kinsboume, M. (1977). Selective listening asymmetry in preschool children. Developmental Psychology, 13, 217-224. Hiscock, M., & Kinsbourne, M. (1978). Ontogeny of cerebral dominance: Evidence from time-sharing asymmetry in children. Developmental Psychology, 16, 70-82. Hiscock, M., & Kinsboume, M. (I980a). Asymmetries of selective listening and attention switching in children. Developmental Psychology, 16, 70-82. Hiscock, M., & Kinsbourne, M. (1980b). Asymmetry of verbalmanual time-sharing in children: A follow-up study. Neuropsychologia, 18, 151-162. Hiscock, M., Kinsboume, M., Caplan, B., & Swanson, J. M. ( 1979). Auditory attention in hyperactive children: Effects of stimulant medication on dichotic listening performance. Journal ofAbnormal Psychology, 88, 27-32. Hochberg, F. H., &LeMay, M. (1975). Arteriographic correlates of handedness. Neurology, 25, 218-222. Hubbs, C. L., & Hubbs, L. C. (1944). Bilateral asymmetry and bilateral variation in fishes. Paper to Michigan Academy of Arts, Sciences and Leners, 30, 229-311. Huttenlocher, P., Raichelson, R. M., Rye, D. B., & Wainer, B. H. (1986). Hemispherectomy in infancy enlarges the distribution of motor neurons in the remaining cerebral hemisphere. Annals of Neurology, 20, 418. Hyman, L. H. (1940). The invertebrates. New York: McGrawHill. Inglis, J., & Sykes, D. H. (1967). Some sources of variation in dichotic listening performance in children. Journal ofExperimental Child Psychology, 5, 480-488. Ingram, D. (1975). Cerebral speech lateralization in young children. Neuropsyclwlogia, 13, 103-105. Jones, R. K. (1966). Observations on stammering after localized cerebral injury. Journal of Neurology, Neurosurgery and Psychiatry, 29, 192-195. Kershner, J. R., & King, A. J. (1974). Laterality of cognitive functions in achieving hemiplegic children. Perceptual and Motor Skills, 39, 1238-1284. Kiessling, L. S.,Denckla, M. B., &Carlton, M. (1983). Evidence for differential hemisphere function in children with hemiplegic cerebral palsy. Developmentol Medicine and Child Neurology, 25, 727-734. Kinsboume, M. (1970). The cerebral basis of lateral asymmetries in attention. Acta Psychologica, 33, 193-201. Kinsbourne, M. (1972). Eye and head turning indicate cerebral lateralization. Science, 176, 539-541. Kinsboume, M. (1973). The control of attention by interaction between the cerebral hemispheres. In S. Kornblum (Ed.), Attention and perjo1'11f1l11Ce W. New York: Academic Press. Kinsboume, M. (1974). Lateral interactions in the brain. In M.
HEMISPHERE SPECIALIZATION IN CHILDREN Kinsboume & W. L. Smith (Eds.), Hemisphere disconnection and cerebra/function. Springfield, IL: Thomas. Kinsboume, M. (1975). The ontogeny of cerebral dominance. Annals of the New York Academy ofSciences, 263, 244-250. Kinsboume, M. (1980). A model for the ontogeny of cerebral organization in non-right-banders. In 1. Herron (Ed.), Neuropsychology of left handedness. New York: Academic Press. Kinsboume, M. (1982). Hemispheric specialization and the growth of human understanding. American Psychologist, 37, 411-420. Kinsboume, M. (1986). Sinistrality and rislc for immune diseases and learning disorders: A pleiotropic gene effect? Annals of Neurology, 20, 416 (abstract). Kinsboume, M. (1987). Cerebral-brainstem interactions in infantile autism. In E. Schopler & G. Mesibov (Eds.), Neurobiological theories ofarousal and autism. New York: Plenum Press. Kinsboume, M., & Caplan, P. J. (1979). Children's learning and attention problems. Boston: Little, Brown. Kinsboume, M., & Cook, J. (1971). Generalized and lateralized effects '-of concurrent verbalization on a unimanual skill. Quarter~y Journal of Experimental Psychology, 23, 341345. Kinsboume, M., & Hicks, R. E. (1978). Functional cerebral space: A model for overflow, transfer and interference effects in human perfOflllance: A tutorial review. In J. Requin (Ed.), Anention and performance VII. Hillsdale, NJ: Erlbaum. Kinsboume, M., &Hiscock, M. (1977). Does cerebral dominance develop? In S. 1. Segalowitz & F. A. Gruber (Eds.), Language development and neurological theory. New York: Academic Press. Kinsboume, M., & Lempert, H. ( 1979). Does left brain lateralization of speech arise from right-biased orienting to salient percepts? Human Development, 22, 270-276. Klein, S. P., & Rosenfield, W. D. (1980). The hemispheric specialization for linguistic and nonlinguistic tactile stimuli in third grade children. Cortex, 16, 205-212. Knox, C., & Kimura, D. (1970). Cerebral proce~ing of nonverbal sounds in boys and girls. Neuropsychologia, 8, 227-237. Kohn, B., & Dennis, M. (1974). Selective impainnents ofvisuospatial abilities in infantile hemiplegics after right hemidecortication. Neuropsychologia, 12, 505-512. Krashen, S. D. (1973). Lateralization, language learning, and the critical period: Some new evidence. Language Learning, 23, 63-74. Latimer, H. B., & Lowrance, E. W. (1965). Bilateral asymmetry in weight and in length ofhuiJian bones. Anotomical Record, 152, 217-224. LeMay, M. (1976). Morphological cerebral asymmetries of modem man, fossil man, and nonhuman primate. Aflllllls of the New York Academy of Sciences, 280, 349-366. LeMay, M., & Culebras, A. (1977). Human brain-pathological differences in the hemispberes demonstrable by cortical arteriograpby. New England Journal of Medicine, 287, 168170. Lempert, H., & Kinsboume, M. (1982). Effect of laterality of orientation on verbal memory. Neuropsychologia, 20, 211214.
83
Lempert, H., & Kinsboume, M. (1985). Possible origin of speech ·in selective orienting. Psychological Bulletin, 97, 62-73. Lenneberg, E. H. (1967). Biological foundations of language. New York: Wiley. Levine, S.C. (1985). Developmental changes in right hemisphere involvement in face-recognition. In C. Best (Ed.), Hemisphere junction and collaboration in the child. New York: Academic Press. Levy, J. (1976). A review of evidence for a genetic component in the determination of handedness. Behavior Genetics, 6, 429453. Levy, J., & Reid, M. (1976). Variations in writing posture and cerebral organization. Science, 194, 337-339. Lewandowski, L. L. (1982). Hemispheric asymmetry in children. Perceptual and Motor Skills, 54, 1011-1019. Liederman, 1. (1983). Mechanisms underlying instability in the development of the hand preference. In G. Young, S. Segalowitz, C. M. Carter, & S. E. Trehub (Eds.), Manual specialization and the developing brain. New York: Academic Press. Liederman, J., & Kinsboume, M. (1980a). The mechanism of neonatal rightward turning bias: A sensory or motor asymmetry? Infant Behavior and Development. 3, 223-238. Liederman, J., & Kinsboume, M. ( 1980b). Rightward motor bias in newborns depends upon parental right-handedness. Neuropsychologia, 18, 519-584. Loeb, J. (1918). Forced movements, tropisms and animal conduct. Philadelphia: Lippincott. Lokker, R., & Morais, 1. (1985). Ear differences in children at two years of age. Neuropsychologia, 23, 127-129. Ludwig, W. (1932). Das Rechts-Links-Problem in Tierreich und beim Menschen. Berlin: Springer. MacKain, K., Studdert-Kennedy, M., Spieker, S., & Stem, D. (1983). Infant intermodal speech perception is a left hemisphere function. Science, 214, 1347-1349. Marcel, T., Katz, L., & Smith, M. (1974). Laterality and reading proficiency. Neuropsychologia, 12, 131-139. Marcel, T., & Rajan, P. (1975). Lateral specialization for recognition of words and faces in good and poor readers. Neuropsychologia, 13, 489-491. McGlone, 1. (1980). Sex differences in human brain asymmetry: A critical survey. Behavioral and Brain Sciences, 3, 215263. McKeever, W. F., & VanDeventer, A. D. (1975). Dyslexic adolescents: Evidence of impaired visual and auditory language processing associated with normal lateralization and visual responsivity. Cortex, 1/, 361-378. McRae, D., Branch, D., & Milner, B. (1968). The occipital horns and cerebral dominance. Neurology, 18, 95-98. Mecacci, L., Secbi, E., & Levi, G. (1983). Abnonnalities of visual evoked potentials by checkerboards in children with specific reading disability. Brain and Cognition, 2. 135-143. Miller, L. K., & Turner, S. (1973). Development of hemifield differences in word recognition. Journal ofEducational Psychology, 65, 172-176. Molfese, D. L., Freeman, R. B., & Palmero, D. S. (1975). The ontogeny of brain lateralization for speech and nonspeecb stimuli. Brain and Language, 2, 356-368. Nagafuchi, M. (1970). Development of dichotic and monaural
84
CHAPTER4
hearing abilities in young children. Acta Otolaryngologica, 69, 409-414. Naylor, H. (1980). Reading disability and lateral asymmetry: An infonnation processing analysis. Psychological Bulletin, 87, 531-545. Newcombe, N., & Bandura, M. M. (1983). Effect of age at puberty on spatial ability in girls: A question of mechanism. Developmental Psychology, 19, 215-224. Nottebohm, F. (1971). Neurallateralization of vocal control in a passerine bird. I. Song. Journal of Experimental Zoology. 177, 229-262. Obrzut, J. E., Hynd, G. W., Obrzut, A., & Pirozzola, F. J. ( 1981). Effect of directed attention on cerebral asymmetries in nonnal and learning-disabled children. Developmental Psychology, 17, 118-125. Obrzut, J. E., Hynd, G. W., &Zellner, R. D. (1983). Attentional deficit vs. learning disabled children: Evidence from visual half-field asymmetries. Brain and Cognition, 2, 89-JCI. Ojemann, G. A. (1977). Asymmetric function of the thalamus. Annals of the New York Academy ofSciences, 299, 380-396. Olson, M. E. (1973). Laterality differences in tachistoscopic word recognition in nonnal and delayed readers in elementary school. Neuropsychologia, 11, 343-350. Orton, S. T. (1937). Reading, writing and speech problems in children. London: Chapman & Hall. Palmer, R. D. (1964). Development of a differentiated handedness. Psychological Bulletin, 62, 257-272. Pande, B. S., & Singh, I. (1971). One-sided dominance in the upper limits of human fetuses as evidenced by asymmetry in muscle and bone weight. Journal ofAnatomy, 109, 457-459. Parlow, S. E., & Kinsboume, M. (1981). Handwriting posture and manual motor asymmetries in sinistrals. Neuropsychologia, 19, 687-696. Parlow, S. E., & Kinsboume, M. Asymmetrical transfer of training between hands: Implications for interhemispheric communication in nonnal brain. (submitted). Penfield, W., & Roberts, L. (1959). Speech and brain mechanisms. Princeton, NJ: Princeton '(Jniversity Press. Piazza, D. M. (1977). Cerebrallateralization in young children as measured by dichotic listening and finger tapping tasks. Neuropsychologia, 15, 417-425. Pinsky, S. D., & McAdam, D. W. (1980). Electroencephalographic and dichotic indices of cerebra11aterality in stutterers. Brain and Language, lJ, 374-397. Provine, R. R., & Westennan, J. A. (1979). Crossing the midline: Limits of early eye-hand behavior. Child Development, 50, 437-441. Ramsey, D. S. (1984). Onsetofunimanual handedness in infants. bifant Behavior and Development, 3, 377-385. Rankin, J. M., Aram, D. M., & Horwitz, S. J. (1981). Language ability in right and left hemiplegic children. Brain and Language, 14, 292-306. Reynolds, D. M. Q., & Jeeves, M.A. (1978). A developmental study of hemisphere specialization for alphabetical stimuli. Cortex, 14, 259-267. Rose, S. (1984). Developmental changes in hemispheric specialization for tactual processing in very young children: Evidence from cross-modal transfer. Developmental Psychology, 20, 568-574.
Ross, E. D., & Mesulam, M. M. (1979). Dominant language functions of the right hemisphere? Prosody and emotional gesturing. Archives of Neurology, 36, 144-148. Rutter, M., Bartak, L., & Newman, S. (1971). Autism: A central disorder of cognition and language? In M. Rutter (Ed.),lnfantile autism: Concepts, characteristics and treatment. London: Churchill. Satz, P. (1972). Pathological lefthandedness: An explanatory model. Cortex, 8, 121-135. Satz, P. (1976). Cerebral dominance and reading disability: An old problem revisited. In R. M. Knights & D. J. Bakker (Eds.), The neuropsychology of learning disorders. Baltimore: University Park Press. Satz, P., Orsini, D. L., Saslow, E., & Henry, R. (1985). The pathological left-handedness syndrome. Brain and Cognition, 4, 27-46. Satz, P., & Strauss, E. (1986, June). The ontogeny ofhemispheric specialization: Some old hypotheses revisited. Paper presented at the Rodin Conference, Cambridge, England. Saxby, L., & Bryden, M.P. (1984). Left-eat superiority in children for processing of auditory emotional material. Developmental Psychology, 20, 72-80. Schwartz, G. E., Davidson, R. J., & Maer, F. (1975). Right hemisphere lateralization for emotion in the human brain: Interactions with cognition. Science, 190, 286-288. Searleman, A. (1977). A review of right hemisphere linguistic capabilities. Psychological Bulletin, 84, 503-508. Searleman, A. (1983). Language capabilities of the right hemisphere. In A. W. Young (Ed.), Functions of the right cerebral hemisphere. New York: Academic Press. Seatleman, A., Porac, C., & Coren, S. (1982). The relationship between birth stress and writing hand posture. Brain and Cognition, 1, 158-164. Segalowitz, S. J., & Chapman, J. S. (1980). Cerebral asymmetry for speech in neonates: A behavioral measure. Brain and Language, 9, 281-288. Seth, G. (1973). Eye-hand coordination and "handedness": A development study of visuomotor behaviour in infants. British Journal of Educational Psychology, 43, 35-49. Shankweiler, D.P., & Studdert-Kennedy, M.A. (1967). A continuum of lateralization for speech perception? Brain and Language, 2, 212-225. Shucard, J. L., Shucard, D. W., Cummins, K. R., & Campos, J. J. (1981). Auditory evoked potentials and sex-related differences in brain development. Brain and Language, 13, 91102. Silva, D., & Satz, P. (1984). Pathological left-handedness and ambiguous handedness: A new explanatory model. Neuropsychologia, 22, 511-515. Soper, H. V.,Satz, P.,Orsini, D. L., VanGorp, W. G., &Gireer, M. F. (1987). Handedness distribution within severe to profound mental retardation. American Journal of Mental Deficiency, 92, 94-102. Stiles-Davis, J., Sugarman, S., & Nass, R. (1985). The development of spatial and class relations in four young children with right hemisphere damage: Evidence for an early spatial constructive deficit. Brain and Cognition, 4, 388-412. Teng, E. L., Lee, P., Yang, K., & Chang, P. C. (1976). Genetic, cultural and neuropathological factors in relation to laterality.
HEMISPHERE SPECIALIZATION IN CHILDREN In D. 0. Walton, L. Rogers, & J. H. Finzi-Fried (Eds.), Conference on human brain function: Brain information service. Los Angeles: BRC Publication, University of California. Tomlinson-Keasey, C., Kelly, R., & Burton, J. (1979). Hemispheric changes in infonnation processing during development. Developmental Psychology, 14, 214-223. Travis, L. E. (1927). Studies in stuttering. Archives ofNeurology and Psychiatry, 18, 67!-690; 998-1014. Treisman, A. M., & Geffen, G. (1968). Selective attention and cerebral dominance in perceiving and responding to speech messages. Quarterly Journal of Experimental Psychology, 20, 139-150. Tsai, L. Y. (1982). Handedness in autistic children and their families. Journal of Autism and Developmental Disorders, 12. 421-423. Turkewitz, G. (1977). The development of lateral differentiation in the human infant. Annals of the New York Academy of Sciences, 299, 309-317. Turkewitz, G., & Ross-Kossak, P. (1984). Multiple modes of right hemisphere infonnation processing: Age and sex differences in facial recognition. Developmental Psycholo&Y· 20, 95=103.
Varga-Khadem. F.• O'Gorman. A.M .. & Watters. G. V. (1985). Aphasia and handedness in relation to hemispheric side, age at injury and severity of cerebral lesion during childhood. Brain, 8(3), 677-696. Wada, J. A., Clark, R., & Hamm, A. (1975). Cerebral hemispheric asymmetry in humans. Archives of Neurology, 32, 239-246. White, N., & Kinsbourne, M. (1980). Does speech output control lateralization over time? Evidence from verbal-manual time sharing tasks. Brain and Language, 10, 215-233.
85
Witelson, S. F. (1974). Hemispheric specialization for linguistic and nonlinguistic tactual perception using a dichotomous stimulation technique. Cortex, 10, 3-17. Witelson, S. F. (1976). Sex and the single hemisphere: Right hemisphere specialization for spatial processing. Science, 193, 425-427. Witelson, S. F., & Pallie, W. (1973). Left hemisphere specialization for language in the newborn. Brain, 96, 641-646. Wood, F., Stump, D., McKeehan, A., Sheldon, S., & Proctor. J. (1980). Patterns of regional cerebral blood flow during attempted reading aloud by stutterers both on and off haloperidol medication: Evidence for inadequate left frontal activation during stuttering. Brain and Language, 9, 141144. Woods, B. T., & Teuber, H. L. (1978). Changing patterns of childhood aphasia. Annals of Neurology, 32, 239-246. Yeni-Komshian, G. H., & Benson, D. A. (1976). Anatomical study of cerebral asymmetry in the temporal lobe of humans, chimpanzees and rbesus monkeys. Science, 192, 387-389. Young, A. W.,&Bion,P. J. (1980). Absence of any developmental trend in right hemisphere superiority for face recognition. Corfu, 16, I 13-221.
Young, A. W.. & Ellis. H. D. (1976). An experimental investi.2&tion of developmental differences in ability to recognize faces presented to the left and right cerebral hemispheres. Neuropsychologia, 14, 495-498. Young, J. Z. (1962). Why do we have two brains? In V. B. Mountcastle (Ed.),lnterhemispheric relations and cerebral dominance. Baltimore: Johns Hopkins Press. ZUrif, E. B., & Carson, G. (1970). Dyslexia in relation to cerebral dominance and temporal analysis. Neuropsychologia, 8, 351-361.
5 Neuropsychology of Child Psychopathology MICHAEL G. TRAMONTANA AND STEPHEN R. HOOPER
Introduction What is the relationship between brain dysfunction and psychopathology in childhood? Notice that the question is not posed as to whether there is a relationship because, as we shall see, it is simple enough to answer that there is. Clearly, there are mental, emotional, and behavioral sequelae for the child who has sustained brain damage or who shows an anomalous course of brain maturation. In some instances, these symptoms may persist and significantly hamper the child's psychosocial adjustment. What is not so clear, though, is how often this is the case and the precise factors that influence it. Does it depend on the extent or type of brain dysfunction? What about the age or other attributes of the child at the time of onset? How might environmental factors potentiate the child's risk? Are certain forms of psychopathology more likely to arise than others? Does the form of psychopathology manifested change over time? Conversely, there is little disagreement that there are multiple etiologies for psychopathology in childhood, and that in some cases, brain dysfunction likely plays an important contributing role. Just how prevalent this is would be impossible to say. Children with psychiatric disorders do show a relatively high rate of brain dysfunction, but estimates of prevalence have varied greatly according to the assessment MICHAELG. TRAMONI'ANA • DepartmcntofPsychology, Bradley Hospital, East Providence, Rhode Island 02915, and Department of Psychiatry and Human Behavior, Brown University, Providence, Rhode Island 02912. STEPHEN R. HOOPER • ClinicalCenterfortheStudyofDevelopmentandLearning, The University of North Carolina, Chapel Hill, North Carolina 27599, and Depanment of Psychiatry, The University ofNorth Carolina School of Medicine, Chapel Hill, North Carolina 27599.
methods used and the subject samples selected for study. Is brain dysfunction more likely to be involved in certain child psychiatric disorders or behavioral syndromes than others? Does its presence help to explain the particular form of psychopathology manifested? How important is it relative to other factors contributing to the child's disturbance? Asking the above questions should not be taken to imply that the field of child neuropsychology and related disciplines are able to respond with definitive answers at this time. Clearly, they are not. Nonetheless, it is important to raise these questions at the outset to underscore the complexity of the topic. This chapter will provide an overview of current knowledge regarding the neuropsychology of child psychopathology. Key conceptual issues will be discussed, including the role and prevalence of brain dysfunction in child psychopathology. Findings pertaining to selected categories of child psychopathology will be summarized and critically evaluated. Although defmitive conclusions are not possible on the basis of existing research, the work to date certainly does serve to set some useful directions for future investigation. These are highlighted, and specific guidelines for research and practice are discussed.
Conceptual Issues Psychiatric Sequelae of Childhood Brain Dysfunction The presence of brain dysfunction in childhood appears to be associated with a greater risk for the development of a psychiatric disorder, far more so than with other physical handicaps (Brown, Chadwick, Shaffer, Rutter, & Traub, 1981; Rutter, Graham, & Yule, 1970; Seidel, Chadwick, & Rutter, 87
88
CHAPTERS
1975; Shaffer, 1978). Moreover, the effects appear to persist and impede the child's long-range adjustment in many important respects (Breslau & Marshall, 1985;Milman, 1979;Shaffereta/., 1985). One of the best investigations on this topic comes from the well-known Isle of Wight epidemiological studies of school-aged children by Rutter and his colleagues (Rutter et al., 1970). Using multiple assessment procedures, and controlling for rater bias, Rutter eta/. found that about 6 to 7% of the general population of children studied had a psychiatric disorder consisting of some persistent emotional, behavioral, or social disability. The rate was nearly twice that (11.5%) for children having chronic handicapping physical conditions not involving the brain. This group consisted of children with disorders such as asthma, diabetes, heart disease, and orthopedic deformities, as well as diseases of the spinal cord or peripheral nervous system. In contrast, the rate of psychiatric disorder was over five times higher (34.3%) in their neuroepileptic group consisting of all children ranging from 5 to 14 years of age with cerebral palsy, epilepsy, or some other frank neurological disorder above the brain stem. Even when eliminating all cases who had an IQ of 85 or less (as low IQ, itself, was found to be associated with an increased risk for psychiatric disorder), the rate of psychiatric disorder was still twice as high in the neuroepileptic group as in the "other physical handicap" group. Among children in the neuroepileptic group, both the severity and the nature of the neurological condition appeared to be related to the risk of psychiatric disorder. Thus, psychiatric disorders were more likely in children with bilateral as opposed to unilateral brain lesions. Among cerebral-palsied children, it was more likely in those with strabismus, impaired language, or specific reading difficulties, and among the children with epilepsy, the risk was greater in those with low IQ or psychomotor seizures. However, the prevalence was actually less among children with the most severely debilitating handicaps, suggesting that these children may be spared from psychiatric difficulties by having conditions that unambiguously free them from competing in otherwise stressful pursuits. In a further study, Seidel et a/. (1975) were better able to control for the possibility that overt stigmata, such as crippling, may have been associated with the higher rate of psychiatric disorder that Rutter eta/. found in their neuroepileptic group (neither obvious crippling nor other overt stigma was common in their ''other physical handicap'' group). Here, they compared two groups of children with
visibly crippling conditions who were alike in all respects except for the presence of brain damage. All children ranged from 5 to 15 years of age and had an IQ of70 or higher. The two groups were matched in terms of age, sex, psychosocial factors as well as the degree of physical disability. Again, based on both teachers' questionnaire responses and psychiatric ratings, the rate of disorder was about twice as high for children with cerebral disorders (mostly cerebral palsy) than for the group with noncerebral or peripheral conditions (including muscular dystrophy, polio, or spina bifida). Clearly, the studies by Rutter et a/. ( 1970) and Seidel et a/. (1975) provided a strong case for an increased risk of psychiatric disorder being associated with the presence of brain damage in childhood. Neither study, however, demonstrated a causal relationship. Although the neurological conditions of the brain-damaged groups typically had an early onset that probably preceded the appearance of any psychiatric disorder, one still could argue that the relationship was merely coincidental. That is, some common vulnerability (whether it be genetic, congenital, or environmental), which may have predisposed a child to cerebral damage, also may have led independently to psychiatric or behavioral disturbance. A more convincing case for the existence of a causal relationship would come from demonstrating that previously normal children with acquired brain injuries are more likely to develop subsequent psychiatric disorders. Children suffering from accidental head injury represent an excellent choice for examining this question, provided that it is recognized that they do not constitute a random sample of the general population. These children, especially those suffering from mild as opposed to severe injuries, often show preexisting problems with impulsivity, aggression, and attention-seeking behavior that make them more susceptible to accidental injury (Klonoff, 1971). The families of these children also differ from the general population in that they show more parental illness and mental disorder, more social disadvantages, and less adequate supervision of the child's play activities. Thus, the absence of adequate controls in many studies reporting intellectual impairment and behavioral disturbance following head injury makes it impossible to determine whether the psychological sequelae stem directly from cerebral damage rather than preexisting difficulties (Rutter, Chadwick, & Shaffer, 1983). Probably the best controlled examination of this topic comes from the prospective studies of headinjured children by Rutter and his colleagues (Brown
NEUROPSYCHOLOGY OF CHILD PSYCHOPATHOLOGY
et al., 1981; Chadwick, Rutter, Brown, Shaffer, & Traub, 1981; Chadwick, Rutter, Shaffer, & Shrout, 1981; Rutter, Chadwick, Shaffer, & Brown, 1980). Children ranging from 5 to 14 years of age who had experienced closed head injuries of sufficient severity to result in a posttraumatic amnesia (PTA) of 7 days or more were compared with a group of children having less severe head injuries (i.e., those with a PTA of less than 7 days, but a duration of at least 1 hour). In addition, these groups were compared with a matched control group of hospital-treated children also suffering severe accidents, but with orthopedic rather than cranial injuries. All children were studied prospectively at 4 months, l year, and 2! years after their injuries. An important feature of this study was the care taken to determine the children's behavior before their accidents. This was done in an unbiased fashion by interviewing parents immediately after their child's injury, but before the child's postinjury psychiatric condition could have been known. The children with severe head injuries did not differ from controls in their preinjury behavior, but they showed more than double the rate of psychiatric disorder at 4 months and at each subsequent followup period. This was true even when children who had psychiatric disorders prior to their accidents were eliminated from the study, thereby focusing specifically on the comparative rate of new psychiatric disorders arising over the course of the follow-up period. There was a rather high threshold for an effect, however, as definite cognitive or psychiatric sequelae were found only in head-injured children having a PTA of at least l week. Whereas persistent psychiatric sequelae were quite common once this range of severity was reached, cognitive impairment lasting for over 2 years generally required aPTA of at least 3 weeks. Head-injured children tended to show greater impairment on timed visual-spatial and visual-motor tests than on verbal tests but, apart from this, no pattern of cognitive deficit specific to head injury was identified. Likewise, the types of psychiatric disorder among the head-injured children were very similar to those found in controls. The only exception to this was in the case of grossly disinhibited social behavior, which was present only in children with very severe head injuries, and may have been linked directly to frontal lobe dysfunction. Children with head injuries showed an increased risk for psychiatric disorder regardless of the age, sex, or social class of the child-factors that ordinarily show a striking mediating effect in the general population. Clearly, the risk was greater among those children with histories of preaccident behavior disorders as well as those experiencing various psy-
89
chosocial adversities within their homes, but the effects were additive rather than interactive. Thus, although psychiatric disorders in childhood have a multifactorial etiology, the evidence from this series of studies indicated that brain injury can play an independent role. Each of the preceding studies dealt with children having known or documented brain damage. One may ask whether a similar relationship exists between psychiatric disorder and so-called ''soft'' neurological signs or minimal brain dysfunction (MBD) in childhood. This has been a subject of much debate, as some investigators have regarded sensory or motor phenomena such as mirror movements, dysdiadochokinesis, dysgraphesthesia, and choreic or athetoid movements to be of little diagnostic value when elicited from patients not having a discrete neurological disorder(e.g., Ingram, 1973). Others (e.g., Rutter eta/., 1970) have argued that it is important to differentiate among different types of signs that are labeled as "soft." Some are considered as soft because: ( 1) they run a developmental course in which the signs may subside as the child grows older; (2) they are rather prevalent among otherwise normal children (with estimates ranging from 8 to 14%); and (3) they also have no clear locus of origin and their neuropathological significance is obscure (Shaffer, 1978). They are not necessarily unreliable, however, and may show consistency over time (Shapiro, Burkes, Petti, & Ranz, 1978). Other signs, such as minor reflex or tone asymmetries, would tend to be less reliable because they are more difficult to detect. Overall, the research on neurological soft signs has shown that: (I) there is a relationship with age, IQ, and sex (with soft signs occurring more frequently among boys), (2) they are more prevalent among children with psychiatric disorders and learning disabilities, (3) they are related to indices of emotional immaturity and dependency in childhood, and (4) the relationship with hyperactivity, aggression, and antisocial conduct is less clear, although soft signs are commonly seen among children who are described as impulsive and distractible (Shaffer, 1978; Shaffer, O'Connor, Shafer, & Prupis, 1983). In a well-controlled prospective study, Shaffer and his colleagues (Shaffer et al., 1985) examined the comparative outcomes in adolescence of children with early soft neurological signs. Children with (n = 83) and without (n = 79) documented soft signs at age 7 received a careful follow-up assessment at age 17. Compared to controls, adolescents with early soft signs had lower IQs and were more likely to have a psychiatric disorder with symptoms of anxiety, withdrawal, and depression. These findings mainly per-
90
CHAPTERS
tained to boys, but all of the girls in the study with an anxiety-withdrawal diagnosis in adolescence showed early soft signs. The relationship was independent of IQ and, when taken together with the presence of anxious-dependent behavior at age 7, the presence of early soft signs was strongly predictive of persistent problems with anxiety and withdrawal. However, no relationship was found with attention deficit disorder or conduct disorder. This was somewhat different than the pattern that Rutter et al. (1970) found in the Isle of Wight study. Children with frank brain damage showed a heterogeneous range of psychiatric disorders without specific features. Hyperactivity and psychosis were more prevalent in their neuroepileptic group, but these appeared to be related more specifically to the presence of mental retardation. However, besides Shaffer et al. (1985), several other studies have found an association between brain dysfunction and the type of behavior problems manifested, although the exact findings have varied according to factors such as age and chronicity. In a 5-year follow-up of children and adolescents with physical disabilities secondary to brain damage, Breslau and Marshall (1985) found that problems with social isolation rather than aggression were more likely to persist. Dorman (1982) found that the relationship between neuropsychological impairment and the type of behavior problems observed varied as a function of age in a group of boys with school problems and no known neurological disorder. Whereas poor neuropsychological performance was associated with externalizing behavior problems in younger (7 to 8) boys, it was associated with internalizing symptoms in the older (9 to 14) subjects. It may be that internalizing rather than externalizing symptoms are more distinctively tied to brain dysfunction as the child grows older and encounters repeated failure and loss in self-esteem. The relationship eventually may become blurred as other factors enter and play a more important determining role in perpetuating the youngster's poor adjustment. Moreover, this process may be accelerated in cases with early histories of more severe disorders. This was suggested in a study by Tramontana, Hooper, and Nardolillo (in press) in which the presence of neuropsychological deficits was found to be associated with more extensive behavior problems among psychiatrically hospitalized boys, regardless of factors such as IQ and SES. However, the relationship mainly applied to younger (8 to 11) as opposed to older (12 to 16) subjects, and specifically involved internalizing rather than externalizing behavior problems.
Taken together, the studies reviewed in this section provide strong evidence that brain dysfunction in childhood is associated with an increased vulnerability for psychiatric disorder. The relationship appears to hold both for children with frank brain damage as well as those with so-called soft neurological signs. The risk is greater for children with more severe neurological disorders (with the possible exception of those with extreme impairment), especially when accompanied by low IQ and other neuropsychological deficits. It also is compounded by factors such as psychosocial adversity and any preexisting tendencies toward behavioral or emotional disturbance. The relationship is not trivial, as the effects appear to persist and influence long-range outcomes (Breslau & Marshall, 1985; Milman, 1979; Shaffer et al. , 1985). This certainly underscores the importance of accurate detection of the functional deficits and behavioral liabilities in the brain-impaired child as a first step in limiting the risk for the later development or progression of a psychiatric disorder (Tramontana, 1983).
Prevalence of Brain Dysfunction among Children with Psychiatric Disorders We tum now to a more complicated issue, namely: How prevalent is brain dysfunction among children with psychiatric disorders? Although brain dysfunction can play a significant contributing role in the development of child psychopathology, it is unclear just how often this occurs. Estimates of prevalence have varied greatly both as a function of the methods and criteria used in identifying brain dysfunction, and in terms of differences in the subject samples selected for study. For example, the prevalence would appear to be rather low if one simply used the presence of positive findings on a routine neurological examination as the basis for establishing neurological involvement. However, such an approach would likely be associated with an underestimation of prevalence because normal neurological examinations are common even among children with documented histories of head injury, encephalitis, or epilepsy (Rutter, 1977). The findings from the few studies that have incorporated noninvasive neurodiagnostic methods such as computed tomography (CT) have been mixed. Much of the research has been with children having autism or other major developmental handicaps for whom enlarged ventricles and other structural deficits have been found in subgroups of the subjects examined (Campbell et al., 1982; Caparulo
NEUROPSYCHOLOGY OF CHILD PSYCHOPATHOLOGY
et al.. 1981; Damasio et al.. 1980; Rosenbloom et al., 1984). Reiss et al. (1983) likewise found ventricular enlargement in a controlled comparison of CT scans for a mixed group of child psychiatric patients. The results were of questionable generalizability, however, because their subjects also tended to be among the more impaired with respect to psychiatric and developmental status, with about half of the group having a confirmed neurological disorder and a third showing mild mental retardation. In another study, CT scans were compared across four subgroups of subjects with childhood disorders (Infantile Autism, Attention Deficit Disorder, Tourette's Syndrome, and Language Disorder) and a control group of medical patients without documented neurological disorder (Harcherik et al.• 1985). No differences were found among the groups with respect to ventricular volume, right-left ventricular ratios, asymmetries, or brain density. The study was very well done from a technical standpoint, but the interpretation of its results is complicated by several factors including: the groups were not matched in age, and the neurological status and level of functioning of subjects (including controls) were poorly specified. In contrast, prevalence rates using neuropsychological criteria tend to be comparatively high. For example, Tramontana, Sherrets, and Golden ( 1980) found a high rate of neuropsychological abnormality in a mixed sample of child and adolescent psychiatric patients without known brain damage. The subjects consisted of 20 hospitalized cases ranging from 9 to 15 years of age who had neither a history of brain damage nor positive findings on a routine neurological examination performed by the admitting psychiatrist. From a neuropsychological standpoint, these were "nonreferred" cases for whom brain dysfunction was not suspected. Nonetheless, about 60% of the subjects showed at least mild impairment (with 25% showing more definite impairment) according to the normative rules established by Selz and Reitan (1979) on the HalsteadReitan Neuropsychological Battery (HRNB). Impaired performance on the HRNB was associated with lower IQ, and was more prevalent among cases whose psychiatric disorders were of at least 2 years' duration and who had a lag of at least 2 years in academic achievement. One may question the meaning of the neuropsychological abnormalities found in· the Tramontana et al. study, especially as to whether they indeed reflected underlying brain anomalies that were missed in a routine neurological examination or review of history. This was explored in a subsequent
91
study with a similar sample of subjects in which neuropsychological results were compared with various quantified indices of brain structure examined through CT (Tramontana & Sherrets, 1985). Psychiatric cases without suspected brain damage again were found to show a high rate of neuropsychological abnormality (at least 50%) when examined on either the HRNB or the Children's Revision of the LunaNebraska Neuropsychological Battery (LNNB-C; Golden, 1981). Impaired performance was more likely among boys, younger subjects, and those with more chronic psychiatric histories. Interestingly, impaired performance was not associated significantly with IQ. The overall results of the two test batteries correlated quite highly, but it was the LNNB-C that corresponded more closely with CT scan results. Specifically, impairment on the LNNB-C was associated with smaller ventricular size and less density variability, suggesting delayed brain maturation. It also was associated with lesser regional densities, especially within the right cerebral hemisphere. The absence of control subjects in the foregoing study does not permit one to conclude that the CT results, although associated with neuropsychological abnormality, were themselves necessarily abnormal according to any established normative standards. Nonetheless, the findings were noteworthy in that the presence of neuropsychological deficits among the psychiatric cases did correspond to variations in brain structure, and were not merely the product of nonneurological factors. This was a remarkable finding, especially in view of the restricted range of the sample, because of the exclusion of cases having documented neurological involvement. Taken together, the studies reviewed in this section indicate that the question of prevalence is inextricably tied to the methods and criteria used in assessing brain dysfunction. Children with cerebral palsy, epilepsy, and other obvious neurological conditions (as evidenced on a routine neurological exam) probably comprise less than 5% of the total population of children with psychiatric disorders (Rutter, 1977). The rate is uncertain, but obviously would be higher if one were to include children with clumsiness, language impairment, mental retardation, and learning disabilities (Gualtieri, Koriath, Van Bourgondieu, & Saleeby, 1983; Rutter et al., 1970) for whom there is at least the suspicion of underlying brain damage. The rate is higher still if one further includes children for whom brain damage is not suspected, but who nonetheless may show various neuropsychological deficits when they are comprehensively assessed. Although these deficits may have a relationship with underlying structural factors (see
92
CHAPTERS
Tramontana & Sherrets, 1985}, they cannot be interpreted as reflecting brain damage per se. Rather, they should be viewed as functional impediments, some of which may be tied to abnormalities in brain structure or maturation, which in any event may play a role in the development of child psychiatric disorders. As was noted before, the presence of neuropsychological deficits has been found to be associated with more extensive behavior problems among younger boys, regardless of factors such as IQ, SES, and whether the deficits can be linked specifically with a history of brain injury (see Tramontana et al., in press). Neuropsychological deficits are important in their own right, as they appear to comprise an important index of increased psychiatric risk.
Relationship to Manifest Psychopathology A number of mechanisms have been suggested whereby brain dysfunction may lead to psychopathology, although evidence as to their relative contributions is uncertain (Rutter, 1977, 1983). These include: (1) behavioral disruption that arises directly from abnormal brain activity; (2) a heightened exposure to failure, frustration, and social stigma due to associated disabilities; (3) the possible effects of brain damage on subsequent temperament and personality development; (4) adverse family reactions ranging from overprotection to scapegoating; (5} the child's own reaction to being handicapped and its effects on his/her actual capacity to cope and compete; and (6) possible adverse effects from treatments themselves (e.g., recurrent hospitalization) that may restrict normal activities and socialization. Thus, the effects may be direct or indirect. They also may be conceptualized as transactional and dynamic. Direct effects, for example, would be seen in the case of frontal lobe damage resulting in pronounced impulsivity and social disinhibition. Other examples include organically induced psychosis or episodic aggressiveness that may arise from temporal lobe epilepsy. In other cases, however, brain dysfunction may play more of an indirect etiological role, one that essentially sets the stage for other factors to come into play that, themselves, act to produce an emotional or behavioral disturbance and perhaps further aggravate existing functional difficulties (Tramontana, 1983). For example, brain dysfunction may give rise to learning disabilities that, in turn, render the child more likely to encounter frustration and failure upon entry into school. This may lead to a conduct disorder
consisting of inattentiveness and defiance as an eventual (albeit indirect) outcome. There also may be a compounded difficulty in those areas of performance that have become anxiety-laden and aversive. Parents and teachers may come to view the child as lazy, apathetic, or otherwise difficult, and thereby generate expectations that would only serve to perpetuate the existing problems. The latter represents a transactional effect, namely, the differential reinforcement elicited from significant others by the brain-impaired child and his/her particular deficits. Lastly, the effects on behavior are dynamic rather than static. Just as the primary symptoms of brain dysfunction may change over time, so too they may vary in terms of their developmental significance and the reactions that they elicit from others, including the child. The pattern of behavioral disturbance itself may vary so that, for example, instead of hypersensitivity, defiance, and misconduct, the child later may show apathy, withdrawal, and resignation. Besides the issue of how brain dysfunction may lead to psychopathology in childhood, there also is the question of what form manifest symptoms may take. Earlier thinking (e.g., Bakwin & Bakwin, 1966; Wender, 1971) suggested that the behavioral manifestations of cerebral dysfunction, whatever the cause, were uniform, and comprised a rather distinctive behavioral syndrome consisting of symptoms such as hyperactivity, inattention, and impulsivity. However, there is little evidence of such a behavioral stereotype for the brain-impaired child. Symptoms such as hyperactivity, inattention, and impulsivity do not distinguish children with either frank brain damage (see Brown et al., 1981; Rutter et al., 1970) or soft neurological signs (see Shaffer et al., 1983, 1985). This is not to say that such symptoms are not common among brain-impaired children-they are, but they also are common features of psychiatric disorder in general, regardless of whether neurological abnormality is present (Rutter, 1977; Werry, 1972). One may argue that the relationship between brain dysfunction and psychopathology in childhood is nonspecific (e.g., Boll & Barth, 1981). That is, the presence of brain dysfunction, regardless of its pattern or cause, may contribute nonspecifically to a lowered adaptive capacity and a greater likelihood of exposure to adverse experiences. In this view, brain dysfunction operates indirectly by creating the functional deficits that make successful adjustment more difficult for the child. Any of a variety of behavioral and emotional problems may result, with the distribution of specific symptoms being similar to what
NEUROPSYCHOLOGY OF CHILD PSYCHOPATHOLOGY
is seen generally among children with psychiatric disorders. There is some support for this position. In the Isle of Wight study (Rutter et al., 1970), children with brain damage showed a heterogeneous range of psychiatric disorders with no specific features. Except for cases falling at the extreme of incapacity, the risk was greater in children with more severe injuries, seizures, and lower IQ. Also, apart from the possible relationship between frontal lobe dysfunction and gross social disinhibition, Brown etal. (1981) found no psychiatric symptoms that were specific to children with closed head injuries. From a different perspective, Tramontana and Hooper (1987) found that groups of adolescents with either conduct disorder or major depression were virtually identical in their pattern of neuropsychological functioning. Thus, although these represented two very different types of psychopathology (each falling at opposite ends of a continuum of externalizing and internalizing symptoms, respectively), there were no distinctive neuropsychological features. On the other hand, it was noted before that a history of soft neurological signs or the presence of neuropsychological impairment tended to be associated specifically with internalizing behavior problems consisting of symptoms such as anxiety, depression, and withdrawal (Shaffer et al., 1985; Tramontana et al., in press). In addition to cognitive deficits, children with physical disabilities secondary to brain damage are likely to show persistent problems with social isolation, but problems with aggression are less likely to persist (see Breslau & Marshall, 1985). Also, results from a study of children with localized (penetrating) head injuries showed a significant association between the presence of depression and lesions ~pecifically involving right frontal and left posterior cerebral regions; this was true regardless of the child's age, sex, and psychosocial factors (Rutter, 1983). No relationship was found, however, between the site of injury and symptoms such as hyperactivity, inattention, aggression, or antisocial conduct. Thus, although we are not suggesting the existence of an alternative behavioral stereotype, it may be that internalizing rather than externalizing symptoms are more distinctively tied to brain dysfunction in childhood, perhaps especially in terms of longer-range outcomes. There also is some indication that specific behavioral features may vary according to the type and localization of injury. The following section provides a selective review of neurodiagnostic findings in major categories of child psychopathology. With this, we then will
93
return to the question of specificity, i.e., the extent to which different types and localization of brain dysfunction are associated with specific patterns of psychopathology, and vice versa.
Findings in Selected Categories of Child Psychopathology This section provides a summary of neurological and neuropsychological findings with respect to psychotic disorders in childhood (including Infantile Autism), Attention Deficit Disorder, conduct disorders, and affective disorders. We also will examine briefly the findings in an assortment of other conditions, including Tourette's Syndrome.
Psychotic Disorders Although the general evidence regarding organic involvement in childhood psychotic disorders is rather compelling (Omitz, 1983), the findings are imprecise and obscured by complex issues surrounding diagnostic definition. This broad category includes conditions such as Pervasive Developmental Disorder, Infantile Autism, and Childhood Schizophrenia. Other suggested subcategories have included developmental psychosis (Noll & Benedict, 1982), disintegrative psychosis (Rutter & Shaffer, 1980), and hypotonic schizophrenia (Cantor, Pearce, Pezzot-Pearce, & Evans, 1981). The relationships among these various subcategories are unclear and, as may be expected, differential diagnosis has tended to be associated with poor reliability and validity (Fein, Waterhouse, Lucci, & Snyder, 1985). By far, the bulk of research investigating neurological and neuropsychological abnormalities in childhood psychosis has focused on Infantile Autism, with a smaller number of studies describing their subjects as childhood schizophrenics.
Childhood Schizophrenia Netley, Lockyer, and Greenbaum ( 1975) postulated that neurological involvement is a necessary antecedent for the development of childhood schizophrenia. Children within this category have been found to show a greater frequency of neurological signs (Hertzig & Walker, 1975; Owens & Johnstone, 1980), abnormal BEGs (Netley et al., 1975), histories of perinatal complications (Torrey, Hersh, & McCabe, 1975), and, compared to other psychiatric
94
CHAPTERS
subjects, tend to show a greater degree of cognitive impairment (Carter, Alpert, & Stewart, 1982). A great deal of attention has been given to identifying the early developmental precursors of schizophrenia in high-risk populations, such as the offspring of schizophrenic parents. These children have been found to show a greater frequency of nonspecific neurological signs (Erlenmeyer-Kimling et al., 1982; Marcus, Hans, Mednick, Schulsinger, & Michelson, 1985), although some investigators have cited mild discoordination and hyperactivity as occurring most frequently (Rieder & Nichols, 1979). Fish (1977), furthermore, reported three specific types of neurological abnormality that distinguished infants at risk for schizophrenia: abnormally quiet state, ~isual-motor problems (particularly on bimanual tasks), and decreased vestibular responsiveness. For the most part, the evidence regarding specific neuropsychological features in childhood schizophrenia has been inconclusive. Whereas one study found that childhood schizophrenics differed from other psychiatric controls specifically in terms of Performance IQ (Carter et al., 1982), another study found that it was Verbal IQ that distinguished children at risk for schizophrenia from normal controls (Gruzelier, Mednick, & Schulsinger, 1979). Likewise, no clear localizing patterns have been found with respect to neurostructural and electrophysiological findings.
Infantile Autism Infantile Autism is a behavioral syndrome typically defined by disturbed social and object relations, poor communication and language abilities, and difficulties in the modulation of sensory input. A variety of etiologies have been proposed, including genetic factors (Levitas et al., 1983), pathological endorphin activity (Gillberg, Terenius, & Lonnerholm, 1985), and autoimmune abnormalities (Warren, Margasetten, Pace, & Foster, 1986). However, regardless of the specific anomaly or etiology that is hypothesized, some form of brain impairment is thought to be involved in nearly all cases (Damasio & Maurer, 1978; Ornitz, 1983). Children with autism tend to have. a significant prenatal or perinatal history and show a high rate of soft neurological signs (Jones & Prior, 1985). Kagan (1981) found two or more signs suggestive of neurological involvement in over 94% of his autistic sample. Garreau et al. (1984) further noted that the presence of neurological impairment was associated with an earlier onset of autistic features.
A number of studies have documented the presence of structural abnormalities in autism, with some of these specifically identifying left hemispheric, and in some cases bilateral, defects particularly involving frontal and temporal regions (Gillberg & Svendsen, 1983; Hauser, DeLong, & Rosman, 1975; Maurer & Damasio, 1982). Others (Hier, LeMay, & Rosenberger, 1979) reported different patterns of morphological asymmetry. In a review of the research literature, Hetzler and Griffin ( 1981) concluded that there was evidence of bilateral temporal lobe involvement in autism. DeLong (1978) instead suggested that there are two subtypes of autism: one with primarily left hemispheric impairment and the other with bilateral impairment. However, there have been disconfirming findings as well (see Harcherik et al., 1985). Although CT scan abnormalities were found by Caparulo et al. (1981), no clear localizable pattern emerged. Other studies found no relationship between morphogotical asymmetries and delayed language development (Tsai, Jacoby, & Stewart, 1983; Tsai, Jacoby, Stewart, & Beisler, 1982), and in some studies no structural abnormality of any kind was found (Harcherik eta/., 1985; Prior, Tress, Hoffman, & Boldt, 1984). Lastly, Coleman, Romano, Lapham, and Simon (1985) found no consistent differences in a postmortem cell count of selected left hemispheric regions between an autistic adult and control subjects. With respect to neurophysiological evidence, it has been estimated that approximately 40 to 50% of child autistics show abnormalities in their EEGs (Tsai, Tsai, & August, 1985). The abnormalities are of a varied nature, but often may include excessive slow-wave activity and less alpha bilaterally (Cantor, Thatcher, Hrybyk, & Kaye, 1986). Among autistic children, the presence of a normal EEG appears to be associated with a more favorable developmental course and higheriQ (Small, 1975). Visual and auditory evoked potentials also have been observed to be impaired in autistics, with visual processing capabilities tending to be more intact than auditory processing (Courchesne, Lincoln, Kilman, & Galambos, 1985; DeMyer, Hingtgen, & Jackson, 1981). Numerous neuropsychological aspects of autism have been reported. Children with autism have been found to have poor motor imitation abilities (Jones & Prior, 1985), disproportionate impairment in sequential processing abilities (Tanguay, 1984), and, as a group, IQs that are significantly lower than normals, but comparatively higher than mentally retarded children (Kagan, 1981). The cognitive profiles of autistic children can be rather varied; indeed, using clustering techniques, Fein, Waterhouse, Luc-
NEUROPSYCHOLOGY OF CHILD PSYCHOPATHOLOGY
ci, & Snyder (1985) differentiated five subgroups of autistic children on the basis of their cognitive profiles. Many of the neuropsychological studies have been directed toward investigating the presence of lateralized deficits in autistic children. The findings have included a reversal of ear advantage for speech sounds (left ear rather than right) on dichotic listening tasks (Blackstock, 1978; James & Barry, 1983; Prior & Bradshaw, 1979); increased prevalence rates of left and mixed handedness of about 20 and 34%, respectively (Soper et al., 1986); and performance profiles on neuropsychological test batteries suggestive of predominantly left hemispheric dysfunction (Applebaum, Egel, Koegel, & Imhoff, 1979; Dawson, 1983; Hoffman & Prior, 1982). In a recent study by Dawson, Finley, Phillips, and Galpert (1986), there was strong evidence of an atypical pattern of hemispheric specialization, with about 70% of autistic children showing right hemisphere dominance for speech. However, the results of some studies have suggested a more complex picture. Fein, Waterhouse, Lucci, Rennington, and Humes (1985) found that, among autistic children, left-banders performed better than right-banders across a variety of tasks. However, those with mixed handedness performed the worst. This suggests that if there is early injury to the left hemisphere, a reversal of handedness ultimately may be associated with a better compensatory development. In another study, Arnold and Schwartz (1983) found that autistic children did not demonstrate the same reversal of ear advantage for speech on dichotic listening demonstrated by children with aphasia and other language disorders. These findings suggest that autism and language disorder are not identical, and that language in autism may not be accommodated by the right hemisphere as is often true in other cases of acquired left hemispheric injury. Kagan's (1981) analysis of speech behavior, memory, and thought processes suggested that the presence of right hemisphere dysfunction in autism should be considered as well. Regardless of the issue of lateralization, the presence of disturbed language functions in autism is critical. Language problems are one of the primary symptoms observed in autistic children (Rutter, 1978), and the degree of language impairment appears to be strongly predictive of the child's prognosis (Rutter, Greenfield, & Lockyer, 1967; Wing, 1971). Bartak, Rutter, and Cox (1975) postulated that a language disability probably constitutes a necessary condition for the development of this behavioral syndrome. They observed that autistic children
95
showed more deviant language development, a severe comprehension deficit, and deficits in the social usage of language. Although related to childhood language disorders, the communication deficits in autistic children are qualitatively different from those seen in developmental dysphasia or acquired aphasia. Specifically, autistic children are deficient in symbolic-representation abilities necessary for language development (Ferrari, 1982), show an impoverishment in verbal mediation and gestural communication (Cohen, Caparulo, & Shaywitz, 1976), and, in echolalic autistics, there is an inappropriate use and modulation of expressive-intonational speech (Simon, 1975). Tager-Flusberg (1981) argued that whereas autistic children show a normal but slower developmental course with respect to phonological and syntactical skills, semantic and pragmatic language development may be particularly deviant. Prizant (1982) further argued that the ritualized patterns and highly rigid behaviors of verbal autistic children are directly related to their limitations in cognitive and linguistic processing. Like Tanguay (1984), Prizant suggested that the peculiarities in autistic speech and language reflect an undue reliance on a simultaneous rather than a sequential mode of cognitive processing. Thus, the research on autism provides a somewhat varied picture with respect to neurological and neuropsychological features. With the possible exception of aberrations in language development and communication skills, there is little agreement as to neurodevelopmental features that are necessarily characteristic of the disorder. Some children with autism show significant anomalies in brain structure, whereas others do not. The same is true with respect to neurophysiological and neuropsychological features. Although some children with autism show evidence of lateralized dysfunction mainly involving the left cerebral hemisphere, this by no means is a defining characteristic of the syndrome (Fein, Humes, Kaplan, Lucci, & Waterhouse, 1984).
Attention Deficit Disorder This syndrome has had a long and controversial history. Nonetheless, it remains one of the most commonly diagnosed child psychiatric disorders (Mattison eta/., 1986). The term Attention Deficit Disorder (ADD) replaced earlier terms such as hyperkinetic reaction and hyperactivity due to the central role that deficits in attention were thought to play in the disorder [Diagnostic and Statistical Man-
96
CHAPTERS
ual-ll/, (DSM-111), American Psychiatric Association, 1980]. DSM-Ill actually describes two global subtypes of the disorder-ADD with and without hyperactivity. However, whereas many studies have documented the presence of deficits in selective and sustained attention in ADD children (Brown & Wynne, 1984; Douglas, 1983), it has been proposed that ADD without hyperactivity be deleted as a diagnostic category in the forthcoming revision of DSMIll because of its ambiguous nature (Barkley, 1986a). ADD has often been linked with minimal brain dysfunction (MBD) because of purported behavioral similarities between children with ADD and those with documented brain damage (Strauss & Lehtinen, 1947). However, problems in documenting the presence of underlying brain dysfunction (e.g., Rutter, 1983; Taylor, 1983) have led to a more descriptive approach in conceptualizing the disorder. Children with ADD have been characterized as showing inattention, impulsivity, and overactivity (Douglas, 1980, 1983); deficits in self-directed instruction (Kendall & Broswell, 1985); poor self-regulation of arousal, particularly in meeting environmental demands (Douglas, 1983); and deficiencies in rule-governed behavior (Barkley, 1981a,b). A number of investigators have questioned whether ADD truly represents a valid syndrome (Rubinstein & Brown, 1984; Rutter & Taylor, 1978), given the heterogeneity of its behavioral manifestations (Halperin, Gittelman, Klein, & Rudel, 1984; Johnson, 1981) and the difficulties in distinguishing it from conduct disorders more generally (Delamater & Lahey, 1983). Given the debates surrounding the validity and diagnostic criteria comprising this disorder, it is not surprising that the search for a neurological basis to ADD has presented confusing (Taylor, 1985), and often undifferentiated findings (Levine, Busch, & Aufseeser, 1982). A number of theories have been proposed, but the exact neuropathology of attentional disturbances in children remains poorly understood (Mesulam, 1981). To date, the evidence seems to be strongest with respect to implicating frontal lobe dysfunction in the increased distractibility and impulsive orienting reactions to irrelevant stimuli seen in ADD children (Passier, Isaac, & Hynd, 1986; Stuss & Benton, 1984; Zambelli, Stamm, Maitinsky, & Loiselle, 1977). Various specific patterns of localization have been proposed, including frontal regions anterior and medial to the precentral motor cortex (Mattes, 1980), as well as frontolimbic pathways (Lou et al., 1984; Newlin & Tramontana, 1980). The evidence with respect to localization has been mixed but promising. Whereas EEG studies
with ADD children have failed to identify focal neurophysiological features (Grunewald, Grunewald, & Rasche, 1975; Mikkelsen, Brown, Minichiello, Millican, & Rapoport, 1982; Montagu, 1975), a recent study of regional cerebral blood flow patterns demonstrated lower perfusion rates in the frontal regions of some ADD children (Lou, Henriksen, & Bruhn, 1984). Also, in a well-designed study, Chelune, Ferguson, Koon, & Dickey (1986) were able to distinguish ADD children from age- and IQ-matched normals with 85% accuracy using tests such as the Wisconsin Card Sorting Test, a measure commonly thought to be sensitive to adult frontal lobe dysfunction. They also demonstrated that, among the ADD children, medication response with psychostimulants was poorer for those showing a greater degree of neuropsychological impairment. This finding is consistent with the observations of Rapoport et al. ( 1980) who found that even normal children show improved performance with psychostimulant medication. It may be that a positive medication response is dependent upon relatively intact frontal lobes, and that the presence of more severe frontal lobe pathology would tend to be associated with a poor medication response. This is a fruitful area for further inquiry.
Conduct Disorders Research in this broad category of child psychopathology has been beset with a number of problems. First, as a diagnosis, it pertains to a very heterogeneous range of disturbances in which the manifestation of socially unacceptable behavior is the primary common feature. Second, the bulk of research has focused on adolescents, particularly the juvenile offender. If one excludes children with ADD, little is known with respect to the neurological and neuropsychological features of conduct disorders manifested at early ages. Third, youngsters with conduct disorders have a higher risk for accidental head injury (Lewis & Shanok, 1977; Lewis, Pincus, & Glaser, 1979; Pincus & Tucker, 1978); thus, although they may show neurological abnormalities on examination, these may bear no direct relationship to the initial conduct disorder. This problem obviously is compounded by the emphasis on studying older as opposed to younger conduct-disordered subjects. With these limitations in mind, the findings for this general category of psychopathology are summarized below. A number of studies have reported abnormal neurological findings in youngsters with conduct dis-
NEUROPSYCHOLOGY OF CHILD PSYCHOPATHOLOGY
orders (Elliott, 1982; Korhonen & Sillanpaa, 1976; Krynicki, 1978; Woods & Eby, 1982). Electrophysiological studies (Coble eta/., 1984; Elliott, 1982; Krynicki, 1978; Luchins, 1983) have found EEG sleep abnormalities, specifically in the expression of slow-wave (delta) activity (Coble et al., 1984); seizure activity that may contribute to recurrent and unprovoked rage attacks (Elliott, 1982); and in some cases, frontal lobe paroxysmal activity, particularly in conduct-disordered adolescents with a significant history of assaultive behavior (Krynicki, 1978). The latter finding bears some relationship to the work of Woods and Eby (1982) and Pontius and Ruttiger (1976) who postulated a delay in the development of normal inhibitory mechanisms (i.e., frontal lobe functions) in repetitively aggressive youngsters. Children with conduct disorders have been reported to show a higher incidence of episodes of disturbed consciousness and, as already noted, to suffer more head injuries than other children (see Lewis & Shanok, 1977; Lewis et al., 1979; Pincus & Tucker, 1978). However, they have not been found to differ from normal controls in terms of perinatal problems, except for more frequently being small for gestational age (McGee, Silva, & Williams, 1984). These findings further serve to suggest that the neurological features in many of these children may postdate the initial onset of their conduct disorders. Conduct-disordered youngsters have been found to have a high rate of learning disabilities (Cannon & Compton, 1980; Robbins, Beck, Pries, Jacobs, & Smith, 1983; Zinkus & Gottlieb, 1978), as well as more generalized problems with language performance (Funk & Ruppert, 1984; Stellern, Marlowe, Jacobs, & Cossairt, 1985; Wardell & Yeudall, 1980). This appears to apply to both nonincarcerated (Robbins et al., 1983) as well as incarcerated (Cannon & Compton, 1980) populations. These findings suggest that the presence of cognitive impairments, perhaps particularly of a verbal nature, places the youngster at risk for acting out impulsively when placed in frustrating or provocative social situations. The degree of impulsivity per se is unrelated to either the type or the number of crimes committed by delinquent youth (Oas, 1985). Rather, it may be that the presence of faulty capacities in verbal reasoning and judgment, along with impulsivity, is a necessary ingredient in the production of chronic antisocial conduct. Thus, although unrelated to the degree of impulsivity, the presence of at least a 15-point inferiority in Verbal IQ versus Performance IQ on the Wechsler Intelligence Scale for Children-Revised (WISC-R) has been found to be predictive of recidivism in adjudicated white delinquent boys (Haynes & Bensch, 1981).
97
Further investigations into the pattern of neuropsychological deficits in conduct disorders have produced mixed results. Berman and Siegal (1976) found that delinquents performed more poorly than normal controls on virtually every task of the HRNB. Whereas prominent deficits were observed in tasks requiring verbal mediation, concept formation, and perceptual organization, only minimal difficulties were found in memory and gross motor coordination. Brickman, McManus, Grapentine, and Alessi (1984) found that more violent youth tended to show more impairment on the LNNB than their nonviolent counterparts, with Expressive Speech and Memory being the distinguishing summary scales. This was true with respect to both male and female offenders. These findings were similar to the results of earlier studies by Lewis, Shanok, and Pincus (1982) and Voorhees (1981). However, in controlling for the presence of psychosis and a history of neurological disorder, Tarter, Hegedus, Alterman, and Katz-Gartis (1983) failed to find differences in neuropsychological, intellectual, and psychoeducational performance across groups of adolescent offenders differing with respect to their type of offense (i.e., violent, nonviolent, sexual). The previously noted problems limit the generalizations that one can make with respect to this category of child psychopathology. It is probably fair to say that, as a group, youngsters with conduct disorders tend to have more limited verbal abilities and a heightened rate of neurological signs (these, however, may arise secondarily as consequences of their behavior disorders). With the possible exception of cases with prominent histories of repetitive, assaultive behavior, the specific role of neurological factors in conduct disorders remains unclear.
Affective Disorders MacAuslan (1975) reported that depressed children have an increased frequency of neurological soft signs when compared to normal controls. Conversely, as was noted earlier, adolescents with early soft signs were more likely to have a psychiatric disorder characterized by symptoms such as anxiety, withdrawal, and depression (Shaffer et al., 1985). Similarly, social isolation rather than aggression is more likely to be a persistent problem in children with physical disabilities secondary to brain damage (see Breslau & Marshall, 1985). Moreover, internalizing rather than externalizing symptoms have been found to be more clearly tied to neuropsychological impairment in psychiatrically hospitalized boys (Tramon-
98
CHAPTERS
tana et al., in press). Also, apart from gross social disinhibition, it will be recalled that depression was the only psychiatric symptom that bore any specific relationship to lesion localization in the series of studies on head injury by Rutter and his colleagues (Rutter, 1983). Thus, depression and other disturbances in affect appear to be important features of the brain-impaired child, perhaps especially in terms of long-range outcomes. Much of the neuropsychological investigation into childhood depression has focused specifically on the question of lateralization of dysfunction. Research demonstrating the specialized role of the right cerebral hemisphere in the processing of human emotion and affective cues, along with reports of right hemispheric dysfunction in adults suffering from depression (Tucker, 1980), prompted inquiries into the existence of such relationships in children. A number of studies have reported impaired nonverbal abilities relative to verbal abilities in children with depression. For example, Kaslow, Rehm, and Siegel (1984) found that higher scores on the Children's Depression Inventory (CDI) were associated with poorer performance on the WISC-R subtests of Block Design, Coding, and Digit Span in a mixed group of childhood depressives. No significant relationships were found for WISC-R Vocabulary or the TrailMaking Test of the HRNB. Blumberg and Izard ( 1985) found a similar pattern of results using the Peabody Picture Vocabulary Test and WISC-R Block Design, with the CDI again serving as the index of depression. In both studies, girls were found to perform more poorly than boys on Block Design. Children of bipolar probands likewise have been found to show a disproportionate inferiority in Performance IQ relative to Verbal IQ when compared to normal controls (Decina et al., 1983). However, the fact that left-handedness was overrepresented in their sample of children at risk for affective disorder would tend to argue against an inference of right hemispheric dysfunction. More generally, the findings in the preceding studies constitute very weak evidence of lateralized right hemispheric dysfunction. The obtained pattern of results simply may reflect the differential sensitivity of performance measures to the effects of depressed concentration and motor speed. Several studies have reported improvements on neuropsychological measures suggestive of both right hemispheric and frontal lobe dysfunction subsequent to treatment with antidepressant medication (Brumback, Staton, & Wilson, 1980; Staton, Wilson, & Brumback, 1981; Wilson & Staton, 1984). Specifically, Staton et al. found that remission of melancholic symptoms was associated with improved performance on WISC-R Similarities,
Comprehension, Block Design, and Coding, as well as on the Matching Familiar Figures Test, the Category Test of the HRNB, and the Visual Reception Subtest of the Illinois Test of Psycholinguistic Abilities. Although the localizing significance of this pattern of results is uncertain, two children in the study were reported to have had a mild left-sided hemiparesis, which also improved subsequent to antidepressant treatment. The latter finding, if replicated, would constitute more convincing evidence of improvement in lateralized dysfunction. A number of electrophysiological studies have been reported as well. Rochford, Weinapple, and Goldstein ( 1981) found greater EEG variance in the right hemisphere than in the left in a heterogeneous group of depressed adolescents. This pattern was distinct from that of normal controls, who demonstrated about equal hemispheric variance, and from adolescents with paranoid symptomatology, who exhibited greater variance in the left hemisphere. However, Knott, Waters, Lapierre, and Gray (1985) found no evidence of specific hemispheric abnormalities in a comparison of EEG patterns and auditory-evoked potentials in matched pairs of siblings discordant for affective disorder. They did find that the bipolar group spent less time in EEG alpha, suggesting a hyperarousal of the nervous system in this form of affective disorder. EEG abnormalities in REM sleep latencies also have been described in depressed adolescents, although this finding has not been documented in prepubescent children (Mendlewicz, Hoffman, Kerkhofs, & Linkowski, 1984). Sackeim, Decina, and Malitz (1982) reviewed much of the earlier literature pertaining to functional brain asymmetry and affective disorders. They concluded that affective disorders, particularly unipolar depression, tend to be associated with right hemispheric cognitive dysfunction and/or electrophysiological overactivation. In contrast, bipolar patients may evidence right- or left-sided hemispheric hyperactivation depending on whether the individual is experiencing a depressive (right hemisphere) or man.ic (left hemisphere) episode. These assertions will require further validation with child and adolescent subjects.
Other Disorders Although not specifically a psychiatric disorder, Tourette's Syndrome is included here because of its bizarre behavioral presentation, particularly in extreme cases, and the heightened psychiatric vulnerability of the afflicted child. It is a rare disorder characterized by facial, body, and vocal tics, with about 50% of the cases demonstrating coprolalia
NEUROPSYCHOLOGY OF CHILD PSYCHOPATHOLOGY
(Woodrow, 1974), and approximately 20-25% exhibiting imitative gestures and copropraxia (Shapiro, Shapiro, & Wayne, 1973). These symptoms vary in intensity, with an exacerbation typically being noted during times of stress (Woodrow, 1974). The specific etiology of this disorder is unknown, although current thinking suggests that it is a neurological disorder (Devinsky, 1983) with a major neurophysiological component (Glaze, Frost, & Jankovic, 1983; Tanner, Goetz, & Klawans, 1982). Some have suggested that it mainly involves subcortical impairment secondary to a neurochemical disturbance, and that the disorder follows a progressive course (Bornstein, King, & Carroll, 1983). However, neither epidemiological work (Lees, Robertson, Trimble, & Murray, 1984) nor serial neuropsychological assessments performed over a 5year period (Newman, Barth, & Zillmer, 1986) have confirmed the presence of a degenerative process. The presence of attentional difficulties often observed in these patients has led some investigators to hypothesize a relationship with ADD, particularly for those ADD children who show motor tics as a side effect of psychostimulant medication (Comings & Comings, 1984). At present, no consistent pattern of neurological or neuropsychological deficits has emerged that characterizes the disorder. Whereas some studies (Ferrari, Matthews, & Barabas, 1984; Incagnoli &Kane, 1983; Sutherland, Kolb, Schoel, Whishow, & Davies, 1982) have identified specific impairments (especially involving visual-spatial and visualmotor abilities), others have suggested a more heterogeneous picture (Joschko & Rourke, 1982). The interested reader is referred to Barkley ( 1986b) for a critical review of the research on Tourette's Syndrome in children. Although inconclusive, the reader also may wish to refer to an assortment of studies examining neurological/ neuropsychological abnormalities in children and adolescents with Conversion Reaction (Regan & LaBarbera, 1984), Borderline Personality (Palombo & Feigon, 1984; Smith, Bemporad, & Hanson, 1982), and Obsessive-Compulsive Disorder(Behar eta/., 1984; Rapoport, Elkins, & Langer, 1981).
Implications for Research and Practice Based on the preceding review, there appears to be little evidence of specificity in the type or pattern of brain dysfunction associated with different catego-
99
ries of child psychopathology. For example, we saw evidence suggestive of left hemispheric dysfunction in disorders as dissimilar as Infantile Autism and Conduct Disorder; frontal lobe dysfunction was reported in one study or another for almost all of the categories of disturbance considered. Clearly, there is wider symptomatic variation across the different categories of child psychopathology than the localization findings would suggest. With the neurodiagnostic findings overlapping to such an extent, these hardly could be used to provide a satisfactory explanation for the different forms of psychopathology manifested. Although this lack of specificity may be an artifact of our limited ''windows'' into the brain, it does support the view of a largely nonspecific, indirect relationship between brain dysfunction and psychopathology in childhood. Undoubtedly, the picture was blurred by differences in subject samples and the methods used in identifying brain dysfunction across studies. Inconsistencies in the use of diagnostic terminology and criteria obviously served to confuse the picture when considering a particular disorder. We should not expect the neurodiagnostic findings to be any more cohesive than the particular behavioral syndrome or category of disturbance to which they refer. Confusion also seems to have resulted from a faulty application of neuropsychological inference in a number of studies. It is one thing to use neuropsychological test data in making inferences regarding lesion localization for cases with documented brain damage; however, even here issues such as cerebral plasticity and individual differences in compensatory development can obscure specific brain-behavior relationships (e.g., Bigler & Naugle, 1985; Rourke et al., 1983). In any event, one is certainly on rather weak ground in making such inferences on cases for whom there is no corroborating evidence as to the presence or localization of injury. A relatively low Verbal IQ is not necessarily associated with left hemispheric impairment, nor is impulsivity necessarily a sign of frontal lobe dysfunction. Although such results may have localizing significance, they easily can be attributed to nonneurological factors as well. Unfortunately, too many investigators have offered interpretations of localized dysfunction solely on the basis of this kind of weak evidence. More rigorous applications of reliable diagnostic criteria, together with a multimethod approach (Tramontana, 1983) that incorporates recent advances in neurodiagnostic technology (including magnetic resonance imaging, positron emission tomography, and topographic EEG mapping), certainly will help to advance our knowledge in this area. This promises to be an exciting line of research
100
CHAPTER 5
offering fresh insights into the neuropathology of child psychiatric disorders. However, more is needed-especially research that focuses on issues of prevention and treatment. The existing research has documented the heightened risk for psychopathology in the child with brain dysfunction. Emphasis should be given to gaining a better understanding of what factors might curtail that risk, and thereby maximize outcomes. At present, the natural history of behavioral disturbances secondary to brain dysfunction is poorly understood. There is some indication that the relationship may weaken or grow more indirect over time, as other factors perhaps come to assume a more important determining role in maintaining problem behaviors (Dorman, 1982; Tramontana et al., in press). The nature of behavioral disturbances also may show some convergence over time, with symptoms such as withdrawal and depression being among the more common outcomes associated with a history of chronic handicap (Breslau & Marshall, 1985; Shaffer et al., 1985). It is important to know more precisely how this process unfolds so that it might be redirected more positively, if not prevented. The whole issue of diff~rential treatment responsiveness as a function of neuropsychological factors certainly warrants further exploration, especially with respect to some of the promising findings that have been reported in the areas of ADD (Chelune et al., 1986) and childhood depression (Staton et al., 1981). Rather than localization of dysfunction per se, greater emphasis should be given to examining the therapeutic and prognostic significance of different profiles of functional deficits, with the emphasis being onfunctional systems. The implications for clinical practice are similar. Accurate detection of the functional deficits and behavioral liabilities in the brain-impaired child is the first step in limiting the risk for the later development or exacerbation of psychopathology. The acronym "ITEM-ize" seems to capture the essence of the strategy, i.e., identify, treat early, and thereby minimize the development of secondary disturbances. The neuropsychologist can play an important role in this regard. Of all the professions that deal with brain-impaired children, the neuropsychologist should be uniquely qualified not only in identifying primary deficits, but in treating secondary emotional and behavioral disturbances that may arise. Issues surrounding valid neuropsychological diagnosis with this population have been discussed elsewhere (Tramontana, 1983), and thus will not be elaborated here. Briefly, the major interpretive problem involves distinguishing the effects of deficit ver-
sus disturbance versus delay in the neuropsychological results of psychiatrically disordered children. The standard application of neuropsychological methods appears to be associated with a greater likelihood of false-positive errors in diagnosis with this population. This is because a psychiatric disturbance in childhood or adolescence, in the absence of brain damage, may itself produce significant impairment on many neuropsychological tests. Impaired performance could result from the disruptive effects of anxiety, depression, or psychogenically based problems with impulse control and attention. Such conditions may not only disrupt present performance, but also could have impeded the past attainment of various skills that are prerequisite to age-appropriate performance in many of the areas that are assessed. It is just as undesirable to overdiagnose brain dysfunction as it is to overlook it when it does exist. This has led some authors (e.g., Tramontana, 1983) to argue for the use of more conservative detection criteria when applying neuropsychological methods in a child psychiatric population. Neuropsychological inferences regarding the presence of brain dysfunction should never be based solely on defective levels of performance, but also should be supported by other features in the neuropsychological results. Following this line of thinking, and operationalizing decision-rules based on Reitan's (1974) four methods of inference (i.e., level of performance, pattern of performance, pathognomonic signs, and right-left differences), Tramontana and Hooper (1987) were able to distinguish psychiatrically hospitalized adolescents with and without documented brain damage on the basis of their neuropsychological results. Using the LNNB, they found that the rate of overall correct classification was improved, with false-positive errors correspondingly reduced, when diagnosing brain impairment was based on the subject meeting criteria on at least two of the detection methods examined. This resulted in a 76% rate of overall correct classification, which was a substantial improvement over the results obtained with the standard cutoff on the battery. There has also been some work examining the use of brief screening procedures in identifying child psychiatric cases who would likely show significant abnormalities on a comprehensive neuropsychological assessment. Capitalizing on the variance shared by measures such as the WISC-R and Aphasia Screening Test with standard batteries such as the HRNB and LNNB (Tramontana, Klee, & Boyd, 1984; Wolf & Tramontana, 1982), Tramontana and Boyd ( 1986) derived a regression formula for predicting neuropsychological abnormality in child psy-
NEUROPSYCHOLOGY OF CHILD PSYCHOPATHOLOGY
chiatric referrals on the basis of such screening procedures. Although not a replacement for a full neuropsychological assessment, this may help aid clinicians in identifying cases for whom a comprehensive neuropsychological assessment is indicated. Overall, the neuropsychology of child psychopathology represents an important and challenging aspect of the broader field of child neuropsychology. It is a complex area of investigation for the researcher and clinician alike, as there are many confounding factors that can obscure the study of brain-behavior relationships in child psychopathology. Our discussion of a number of these hopefully has given the reader an appreciation for the complexity of the topic. However, it is not only a field with problems, but one with prospects as well. The work to date certainly has served to highlight promising areas for inquiry and intervention in which the child neuropsychologist can play an important contributing role.
101
Bigler, E. D., & Naugle, R. I. (1985). Case studies in cerebral plasticity. International Journal of Clinical Neuropsychology, 7, 12-23. Blackstock, E. (1978). Cerebral asymmetry and the development of early infantile autism. Journal of Autism and Childhood Schizophrenia, 8, 339-353. Blumberg, S., & Izard, C. (1985). Affective and cognitive characteristics of depression in 10 and 11 year old children. Journal of Personality and School Psychology, 49, 194-202. (Abstract) Boll, T., & Barth, J. (1981). Neuropsychology of brain damage in children. InS. B. Filskov & T. J. Boll (Eds.), Handbook of clinical neuropsychology (pp. 418-452). New York: Wiley. Bomstein, R. A., King, G., & Carroll, A. (1983). Neuropsychological abnormalities in Gilles de Ia Tourette's syndrome. Journal of Nervous and Mental Disease, 17, 497-502. Breslau, N., & Marshall,l. A. (1985). Psychological disturbance in children with physical disabilities: Continuity and change in a 5-year follow-up. Journal of Abnormal Child Psychology, 13, 199-216. Brickman, A., McManus, M., Grapentine, W., & Alessi, N. (1984). Neuropsychological assessment of seriously delinquent adolescents. Journal ofthe American Academy ofChild Psychiatry, 23, 453-457. References Brown, G., Chadwick, 0., Shaffer, D., Rutter, M., & Traub, M. American Psychiatric Association. ( 1980). Diagnostic and statis(1981). A prospective study of children with head injuries. tical manual of mental disorders. Washington, DC: Author. III. Psychiatric sequelae. Psychological Medicine, I I, Applebaum, E., Egel, A. L., Koegel, R. L., & Imhoff, B. (1979). 63-78. Measuring musical abilities of autistic children. Journal of Brown, R., & Wynne, M. (1984). Attentional characteristics and Autism and Developmental Disorders, 9, 279-285. teachers ratings in hyperactive, reading disabled, and normal Arnold, G., & Schwartz, S. (1983). Hemispheric lateralization of boys. Journal of Clinical Child Psychology, 13, 38-43. language in autistic and aphasic children. Journal of Autism Brumback, R. A., Staton, R. D., & Wilson, H. (1980). Neuropsyand Developmenuzl Disorders, 13, 129-139. chological study of children during and after remission of Bakwin, H., & Bakwin, R. M. M. (1966). Clinical management endogenous depressive episodes. Perceptual and Motor of behavior disorders in children (3rd ed.). Philadelphia: Skills, 50, 1163-1167. Saunders. Campbell, M., Rosenbloom, S., Peery, R., George, A. E., Barkley, R. (198la). Hyperactivity.lnE. G. Mash &L. G. Terdal Kricheff,l.l.,Anderson, L., Small, A.M., &Jennings, S. J. (Eds.), Behavioral assessment of childhood disorders (pp. (1982). Computerized axial tomography in young children. 127-184). New York: Guilford Press. American Journal of Psychiatry, 4, 510-512. Barkley, R. (l98lb). Hyperactive children: A handbook for diagCannon, I., & Compton, C. (1980). School dysfunction in the nosis and treatment. New York: Guilford Press. adolescent. Pediatric Clinics of North America, 27, 79-96. Barkley, R. A. (l986a). Attentiondeficitdisorders.ln E. G. Mash Cantor, D., Thatcher, R., Hrybyk, M., &Kaye, H. (1986). Com& L. G. Terdal (Eds.), Behavioral assessment of childhood puterized EEG analyses of autistic children. Journal of Audisorders (2nd ed.). New York: Guilford Press. tism and Developmental Disorders, 16, 169-187. Barkley, R. A. (1986b). TicdisordersandTourette'ssyndrome. In Cantor, S., Pearce,J., Pezzot-Pearce, F., &Evans, J. (1981). The E. G. Mash & L. G. Terdal (Eds.), Behavioral assessment of group of hypotonic schizophrenics: A pilot study. Schizochildhood disorders (2nd ed.). New York: Guilford Press. phrenia Bulletin, 7, 1-11. Bartak, L., Rutter, M., & Cox, A. (1975). A comparative study of Caparulo, B. K., Cohen, D. J., Rothman, S. L., Young, G., Katz, infantile autism and specific developmental receptive lanJ., Shaywitz, S., & Shaywitz, B. (1981). Computed toguage disorder. I. The children. British Journal of Psychiamographic brain scanning in children with developmental try, 126. 127-145. neuropsychiatric disorders. Journal of the American AcadeBehar, D., Rapoport, J. L .. Berg, C. J., Denckla, M. B., Mann, my of Child Psychiatry, 20, 338-357. L., Cox, C., Fedio, P. Zahn, T., & Wolfman, M.G. (1984). Carter, L., Alpert, M., & Stewart, S. (1982). Schizophrenic chilComputerized tomography and neurological test measures in dren's utilization of images and words in performance of adolescents with obsessive-compulsive disorder. American cognitive tasks. Journal ofAutism and Developmenuzl DisorJournal of Psychiatry, 14, 363-369. ders, 12. 279-293. Berman, A., & Siegal, A. (1976). Adaptive and learning skills in Chadwick, 0., Rutter, M., Brown, G., Shaffer, D., & Traub, M. juvenile delinquents: A neuropsychological analysis. Journal (1981). A prospective study of children with head injuries. II. of Learning Disabilities, 9, 583-590. Cognitive sequelae. Psychological Medicine, JJ, 49-61.
102
CHAPTER 5
Chadwick, 0., Rutter, M., Shaffer, D., & Shrout, P. (1981). A prospective study of children with head injuries. IV. Specific cognitive deficits. Journal of Clinical Neuropsychology, 3, 101-120. Chelune, G., Ferguson, W., Koon, R., & Dickey, T. (1986). Frontal lobe dysinhibition in attention deficit disorder. Child Psychiatry and Human Development, 16, 221-234. Coble, P., Taska, L., Kupfer, D., Kazdin, A., Unis, A., & French, N. (1984). EEG sleep "abnormalities" in preadolescent boys with a diagnosis of conduct disorder. Journal of the American Academy of Child Psychiatry, 23, 438-447. Cohen, D. J., Caparulo, B., & Shaywitz, B. (1976). Primary childhood aphasia and childhood autism: Clinical, biological, and conceptual observations. Journal of the American Academy of Child Psychiatry, 15, 604-645. Coleman, P., Romano, J., Lapham, L., & Simon, W. (1985). Cell counts in cerebral cortex of an autistic patient. Journal of Autism and Developmental Disorders, 15, 245-255. Comings, D., & Comings, D. (1984). Tourette's syndrome and Attention Deficit Disoider with Hyperactivity: Are they genetically related? Journal of the American Academy of Child Psychiatry, 23, 138-146. Courchesne, E., Lincoln, A., Kilman, B., & Galambos, R. (1985). Event-related brain potential correlates of the processing of novel visual and auditory information in autism. Journal ofAutism and Developmental Disorders, 15, 55-76. Damasio, A. R., & Maurer, R. G. (1978). A neurological model for childhood autism. Archives of Neurology, 35, 777-786. Damasio, H., Maurer, R. G., Damasio, A. R., & Chui, H. C. (1980). Computerized tomographic scan findings in patients with autistic behavior. Archives ofNeurology, 37, 504-510. Dawson, G. (1983). Lateralized brain dysfunction in autism: Evidence from the Halstead-Reitan Neuropsychological Battery. Journal of Autism and Developmental Disorders, 13, 269-286. Dawson, G., Finley, C., Phillips, S., &Galpert, L. (1986). Hemispheric specialization and the language abilities of autistic children. Child Development, 57, 1440-1453. Decina, P., Kestenbaum, C., Farber, S., Kron, L., Gargan, M., Sackeim, H., & Fieve, R. (1983). Clinical and psychological assessment of children of bipolar probands. American Journal of Psychiatry, 140, 548-553. Delamater, A., & Lahey, B. (1983). Physiological correlates of conduct problems and anxiety in hyperactive and learning disabled children. Journal of Abnormal Child Psychology, 11, 85-100. DeLong, G. ( 1978). A neuropsychological interpretation of infantile autism. In M. Rutter & E. Schopler (Eds.), Autism: A reappraisal of concepts and treatment. New York: Plenum Press. DeMyer, M., Hingtgen, J., &Jackson, R. (1981). Infantile autism reviewed: A decade of research. Schizophrenia Bulletin, 7, 388-451. Devinsky, 0. (1983). Neuroanatomy of Gilles de Ia Tourette's syndrome. Archives of Neurology, 40, 508-514. Dorman, C. (1982). Personality and psychiatric correlates of the Halstead-Reitan tests in boys with school problems. Clinical Neuropsychology, 4, IIQ-114.
Douglas, V. ( 1980). Higher mental processes in hyperactive children: Implications for training. In R. Knights & D. Bakker (Eds.), Treatment of hyperactive and learning disordered children (pp. 65-92). Baltimore: University Park Press. Douglas, V. (1983). Attention and cognitive problems. In M. Rutter (Ed.), Developmental neuropsychiatry (pp. 280329). New York: Guilford Press. Elliott, F. A. (1982). Neurological findings in adult minimal brain dysfunction and the dyscontrol syndrome. Journal of Nervous and Mental Disease, 170, 680-687. Erlenmeyer-Kimling, L., Comblatt, B., Friedman, D., Marcuse, Y., Rutschmann, J., Simmens, S., & Deni, S. (1982). Neurological, electrophysiological, and attentional deviations in children at risk for schizophrenia. In F. Henn & H. Nasrallah (Eds.), Schizophrenia as a brain disease (pp. 61-98). New York: Oxford University Press. Fein, D., Humes, M., Kaplan, E. Lucci, D., & Waterhouse, L. ( 1984). The question of left hemisphere dysfunction in infantile autism. Psychological Bulletin, 95, 258-281. Fein, D., Waterhouse, L., Lucci, D., Rennington, B., & Humes, M. (1985). Handedness and cognitive functions in POD. Journal of Autism and Developmental Disorders, 15, 323333. Fein, D., Waterhouse, L., Lucci, D., & Snyder, D. (1985). Cognitive subtypes in developmentally disabled children: A pilot study. Journal of Autism and Developmental Disorders, 15, 77-95. Ferrari, M. (1982). Childhood autism: Deficits of communication and symbolic development: I. Distinctions from language disorders. Journal of Communication Disorders, 15, 191208. Ferrari, M., Matthews, W. S., & Barabas, G. (1984). Children with Tourette syndrome: Results of psychological tests given prior to drug treatment. Journal of Developmental and Behavioral Pediatrics, 5, 116-119. Fish, B. (1977). Neurologic antecedents of schizophrenia in children. Archives of General Psychiatry, 34, 592-597. Funk, J. B., & Ruppert, E. (1984). Language disorders and behavioral problems in preschool children. Journal ofDevelopmental and Behavioral Pediatrics, 5, 357-360. Garreau, B., Barthelemy, C., Sauvage, D.,Leddet, I.,&LeLord, G. (1984). A comparison of autistic syndromes with and without associated neurological problems. Journal of Autism and Developmental Disorders, 14, 105-111. Gillberg, C., & Svendsen, P. (1983). Childhood psychosis and computed tomographic brain scan findings. Journal of Autism and Developmental Disorders, 13, 19-32. Gillberg, C., Terenius, L., & Lonnerholm, G. (1985). Endorphin activity in childhood psychosis: Spinal fluid levels in 24 cases. Archives of General Psychiatry, 42, 78Q-783. Glaze, D. G., Frost, J.D., & Jankovic, J. (1983). Sleep in Gilles de Ia Tourette's syndrome: Disorder of arousal. Neurology, 33, 586-592. Golden, C. J. (1981). The Luria-Nebraska Children's Battery: Theory and formulation. In G. W. Hynd & I. E. Obrzut (Eds.), Neuropsychological assessment and the school-age child (pp. 277-302). New York: Grune & Stratton. Grunewald, Z. E., Grunewald, G., & Rasche, A. (1975). Hyper-
NEUROPSYCHOLOGY OF CHILD PSYCHOPATHOLOGY active behavior and EEG arousal reactions in children. Electroencephalography and Clinical Neurophysiology, 38, 149-
159.
103
(Eds.), Gilles de Ia Tourette syndrome (pp. 297-304). New York: Raven Press. Kagan, V. E. (1981). Nonprocess autism in children: A comparative etiopathogenic study. Soviet Neurology and Psychiatry, /4, 25-30. Kaslow, N., Rehm, L., & Siegel, A. (1984). Social-cognitive and cognitive correlates of depression in children. Journal ofAbnormal Child Psychology, 12, 605-620. Kendall, P., & Broswell, L. (1985). Cognitive-behavioral therapy for impulsive children. New York: Guilford Press. Klonoff, H. (1971). Head injuries in children: Predisposing factors, accident conditions, and sequelae. American Journal of Public Health, 61, 2405-2417. Knott, V., Waters, B., Lapierre, Y., & Gray, R. (1985). Neurophysiological correlates of sibling pairs discordant for bipolar affective disorder. American Journal of Psychiatry, 142, 248-250. Korhonen, T., & Sillanpaa, M. (1976). MBD-Iike behavior and neuropsychological performances. Acta Paedopsychiatrica, 42, 75-87. Krynicki, V. E. (1978). Cerebral dysfunction in repetitively assaultive adolescents. Journal of Nervous and Mental Disease, 166, 59-67. Lees, A. J., Robertson, M., Trimble, M. R., & Murray, N. M. (1984). A clinical study of Gilles de Ia Tourette syndrome in the United Kingdom. Journal of Neurology, Neurosurgery and Psychiatry, 47, 1-8. Levine, M.D., Busch, B., & Aufseeser, C. (1982). The dimension of inattention among children with school problems. Pediatrics, 70, 387-395. Levitas, A., Hagerman, R., Braden, M., Rimland, B., McBogg, P., & Matus, I. (1983). Autism and the Fragile X syndrome. Journal of Developmental and Behavioral Pediatrics, 4,
Gruzelier, J., Mednick, S., & Schulsinger, F. ( 1979). Lateralized impairments in the WISC profiles of children at genetic risk for psychopathology. In J. Gruzelier & P. Flor-Henry (Eds.), Hemisphere asymmetries offunction in psychopathology (pp. 105-110). Amsterdam: Elsevier/North-Holland. Gualtieri, C. T., Koriath, U., Van Bourgondieu, M., & Saleeby, N. (1983). Language disorders in children referred for psychiatric services. Journal of the American Academy of Child Psychiatry, 22. 165-171. Halperin, J. M., Gittelman, R., Klein, D. F., & Rudel, R. G. (1984). Reading-disabled hyperactive children: A distinct subgroup of attention deficit disorder with hyperactivity? Journal of Abnormal Child Psychology, 12, 1-14. Harcherik, D. F., Cohen, D. 1., Ort, S., Paul, R., Shaywitz, B. A., Volkmar, F. R., Rothman, S. L. G., & Leckman, J. F. (1985). Computed tomographic brain scanning in four neuropsychiatric disorders of childhood. American Journal of Psychiatry, 142, 731-734. Hauser, S. L., DeLong, G. R., & Rosman, N. P. (1975). Pneumographic findings in the infantile autism syndrome: A correlation with temporal lobe disease. Brain, 98, 667-688. Haynes, J., & Bensch, M. (1981). The P>V sign on the WISC-R and recidivism in delinquents. Journal of Consulting and Clinical Psychology. 49, 480-481. Hertzig, M. E., & Walker, H. A. (1975). Symptom formation as an expression of disordered information processing in schizophrenic children. Journal of Autism and Childhood Schizophrenia, 5, 13-24. Hetzler, B. E., & Griffin, J. L. (1981). Infantile autism and the temporal bilateral lobe of the brain. Journal of Autism and Developmental Disorders, 11, 317-330. 151-158. Hier, D. B., LeMay, M., & Rosenberger, P. B. (1979). Autism Lewis, D., Pincus, J., &Glaser, G. (1979). Violent juvenile delinand unfavorable left-right asymmetries of the brain. Journal quents: Psychiatric, neurological, psychological, and abuse of Autism and Developmental Disorders, 9, 153-159. factors. Journal of the American Academy of Child PsychiaHoffman, W. L., & Prior, M. R. (1982). Neuropsychological try, 18, 307-319. dimensions of autism in children: A test of the hemispheric Lewis, D., & Shanok, S. (1977). Medical histories of delinquent dysfunction hypothesis. Journal of Clinical Neuropsycholand nondelinquent children: An epidemiological study. ogy. 41, 27-41. American Journal of Psychiatry, 134, 1020-1025. lncagnoli, T., & Kane, R. (1983). Developmental perspective of Lewis, D., Shanok, S., & Pincus, J. (1982). A comparison of the the Gilles de Ia Tourette syndrome. Perceptual and Motor neuropsychiatric status of female and male incarcerated deSkills, 57, 1271-1281. linquents: Some evidence of sex and race bias. Journal of the Ingram, T. T. S. (1973). Soft signs. Developmental Medicine and American Academy of Child Psychiatry, 21, 190-196. Child Neurology, 15, 527-530. Lou, H., Henriksen, L., & Bruhn, P. (1984). Focal cerebral hypoJames, A. L., & Barry, R. J. ( 1983). Developmental effects in the perfusion in children with dysphasia and/ or attention deficit cerebral lateralization of autistic, retarded and normal childisorder. Archives of Neurology, 41, 825-829. dren. Journal of Autism and Developmental Disorders, 13, Luchins, D. (1983). Carbamazepine for the violent psychiatric 43-56. patient. Lancet, 1, 766. Johnson, J. A. (1981). The etiology of hyperactivity. Exceptional MacAuslan, A. (1975). Physical signs in association with deChildren, 47, 348-354. pressive illness in childhood. Child-Care-Health-Deve lopJones, V., & Prior, M. (1985). Motor imitation abilities and neument, 1, 225-232. rological signs in autistic children. Journal of Autism and Marcus, J., Hans, S., Mednick, S., Schulsinger, F., & Michelson, Developmental Disorders, 15, 37-46. N. (1985). Neurological dysfunctioning in offspring of Joschko, M., & Rourke, B. (1982). Neuropsychological dimenschizophrenics in Israel and Denmark: A replication analysis. sions ofTourette syndrome: Test-retest stability and implicaArchives of General Psychiatry, 42, 753-761. tions for intervention. In A. J. Friedhoff & T. N. Chase Mattes, J. A. (1980). The role of frontal lobe dysfunction in child-
104
CHAFI'ER S
hood hyperkinesis. Comprehensive Psychiatry, 21, 358369. Mattison, R., Humphrey, F., Kales, S., Handford, H., Finkenbinder, R., & Hemit, R. (1986). Psychiatric background and diagnoses of children evaluated for special class placement. Journal of the American Academy of Child Psychiatry, 25, 514-520. Maurer, R. G., & Damasio, A. R. (1982). Childhood autism from the point of view of behavioral neurology. Journal ofAutism and Developmental Disorders, 12, 195-205. McGee, R., Silva, P. A., & Williams, S. M. (1984). Perinatal, neurological, environmental and developmental characteristics of seven-year-old children with stable behavior problems. Journal ofChild Psychology and Psychiatry and Allied Disciplines, 25, 573-586. Mendlewicz, J., Hoffman, G., Kerkhofs, M., & Linkowski, P. (1984). EEG and neuroendocrine parameters in pubertal and adolescent depressed children: A case report study. Journal of Affective Disorders. 6, 265-272. Mesulam, M. (1981 ). A cortical retrieval for directed attention and unilateral neglect. Annals of Neurology, 10, 309-325. Mikkelsen, E. J., Brown, G., Minichiello, M., Millican, F., & Rapoport, J. (1982). Neurologic status in hyperactive, enuretic, encopretic, and normal boys. Journal of the American Academy of Child Psychiatry, 21, 75-81. Milman, D. H. (1979). Minimal brain dysfunction in childhood: Outcome in late adolescence and early adult years. Journal of Clinical Psychiatry, 40, 371-380. Montagu, J. D. (1975). The hyperkinetic child: A behavioral, electrodermal and EEG investigation. Developmental Medicine and Child Neurology. 17, 299-305. Netley, C., Lockyer, L., & Greenbaum, G. H. (1975). Parental characteristics in relation to diagnosis and neurological status in childhood psychosis. British Journal of Psychiatry, 127, 440-444. Newlin, D. B., & Tramontana, M. G. (1980). Neuropsychological findings in a hyperactive adolescent with subcortical brain pathology. Clinical Neuropsychology, 2, 178-183. Newman, A., Barth, J., & Zillmer, E. (1986). Neuropsychological functioning of an adult with Tourene syndrome: A serial assessment. International Journal of Clinical Neuropsychology, 8, 181-185. Noll, R., & Benedict, H. (1982). Differentiation within the classification of childhood psychoses: A continuing dilemma. Merrill Palmer Quarterly, 27, 176-199. Oas, P. (1985). Impulsivity and delinquent behavior among incarcerated adolescents. Journal of Clinical Psychology, 41, 422-424. Omitz, E. (1983). The functional neuroanatomy of infantile autism. International Journal of Neuroscience, 19, 85-124. Owens, D., & Johnstone, E. (1980). The disabilities of chronic schizophrenia-Their nature and the factors contributing to their development. British Journal of Psychiatry, 136, 384395. Palombo, J., & Feigon, J. ( 1984). Borderline personality development in childhood and its relationship to neurocognitive deficits. Child and Adolescent Social Work Jour~wl. 1, 18-33. Passier, M., Isaac, W., &. Hynd, G. W. (1986). Neuropsychol· ogical development of behavior attributed to frontal lobe
functioning in children. Developmental Neuropsychology, 1, 349-370. Pincus, J., & Tucker, G. (1978). Violence in children and adults. Journal of the American Academy of Child Psychiatry, 17, 277-288. Pontius, A., & Runiger, K. (1976). Frontal1obe system matura· tiona! lag in juvenile delinquents shown in narrative test. Adolescence, 11, 509-518. Prior, M., &. Bradshaw, J. (1979). Hemisphere functioning in autistic children. Cortex, 15, 73-81. Prior, M. R., Tress, B., Hoffman, W. L., &. Boldt, D. (1984). Computed tomographic study of children with classic autism. Archives of Neurology, 41, 482-484. Prizant, B. M. (1982). Gestalt language and gestalt processing in autism. Topics in Language Disorders, 3, 16-23. Rapoport, J., Buchsbaum, M., Weingartner, H., Zahn, T., Ludlow, C., & Mikkelsen, E. (1980). Dextroamphetamine: Its cognitive and behavioral effects in hyperactive boys and normal men. Archives of General Psychiatry, 37, 933-943. Rapoport, J., Elkins, R., & Langer, D. (1981). Childhood obsessive-compulsive disorder. American Journal of Psychiatry, 138, 1545-1554. Regan, J., & LaBarbera, J.D. (1984). Lateralization of conversion symptoms in children and adolescents. American Journal of Psychwtry, 14, 1279-1280. Reiss, D., Feinstein, C., Weinberger, D. R., King, R., Wyatt, R. J., & Brallier, D. (1983). Ventricular enlargement in child psychiatric patients: A controlled study with planimetric measurements. American Journal of Psychiatry, 4, 453-456. Reitan, R. M. (1974). Methodological problems in clinical neuropsychology. In R. M. Reitan & L. A. Davison (Eds.), Clinical neuropsychology: Current status and applications (pp. 19-46). New York: Wiley. Rieder, R. 0., &. Nichols, P. L. (1979). Offspring of schizophrenics. III. Hyperactivity and neurological soft signs. Archives of General Psychiatry. 36, 665-674. Robbins, D. M., Beck, J. C., Pries, R., Jacobs, D.,&. Smith, C. (1983). Learning disability and neuropsychological impairment in adjudicated, unincarcerated male delinquents. Journal of the American Academy of Child Psychiatry, 22. 40-
46.
Rochford, J., Weinapple, M., &. Goldstein, L. (1981). The quantitative hemispheric EEG in adolescent psychiatric patients with depressive or paranoid symptomatology. Biological Psychiatry, 16, 47-54. Rosenbloom, S., Campbell, M., George, A. E., Kricheff, I. 1., Taleporos, E., Anderson, L., Reuben, R. N., & Korein, J. (1984). High resolution CT scanning in infantile autism. Journal of the American Academy of Child Psychiatry. 23. 72-77. Rourke, B. P., Bakker, D. J., Fisk, J. L., & Strang, J.D. (1983). Child neuropsychology: An introduction to theory. research, and clinical practice. New York: Guilford Press. Rubinstein, R., & Brown, R. (1984). An evaluation ofthe validity of the diagnostic category of attention deficit disorder. American Journal ofOrthopsychU!try, 54, 398-414. Rutter, M. (1977). Brain damage syndromes in childhood: Concepts and findings. Journal ofChild Ps.vchology and Psychiatry, 18. 1-21.
NEUROPSYCHOLOGY OF CHILD PSYCHOPATHOLOGY
105
Rutter, M. (1978). Diagnosis and definition of childhood autism. the treatment of borderline children. American Journal of Journal of Autism and Childhood Schizophrenia, 8, 139Psychotherapy, 36, 181-197. Soper, H., Satz, P., Orsini, D., Henry, R., Mvi, J., & Schulman, 161. Rutter, M. (1983). Issues and prospects in developmental neuroM. (1986). Handedness patterns in autism suggest subtypes. psychiatry. In M. Rutter (Ed.), Developmental neuropJournal of Autism and Developmental Disorders, 16, 155sychiatry (pp. 577-598). New York: Guilford Press. 167. Rutter, M., Chadwick, 0., & Shaffer, D. (1983). Head injury. In Staton, R. D., Wilson, H., & Brumback, R. A. (1981). Cognitive M. Rutter (Ed.), Developmental neuropsychiatry (pp. 83improvement associated with tricyclic antidepressant treatment of childhood major depressive illness. Perceptual and 111). New York: Guilford Press. Rutter, M., Chadwick, 0., Shaffer, D., & Brown, G. (1980). A Motor Skills, 53, 219-234. prospective study of children with head injuries. I. Design Stellern, J., Marlowe, M., Jacobs, J., & Cossairt, A. (1985). and methods. Psychological Medicine, 10, 633-645. Neuropsychological significance of right hemisphere cogRutter, M., Graham, P., & Yule, W. (1970). A neuropsychiatric nitive mode in behavior disorders. Behavioral Disorders, 10, study in childhood (Clinics in Developmental Medicine Nos. 113-124. 35-36). London: Spastics International Medical Publica- Strauss, A., & Lehtinen, L. (1947). Psychopathology and educations/Heinemann Medical Books. tion of the brain-injured child. New York: Grune & Stratton. Rutter, M., Greenfield, D., & Lockyer, L. (1967). A five to Stuss, D., & Benson, D. (1984). Neuropsychological studies of fifteen-year follow-up study of infantile psychosis. II. Social the frontal lobes. Psychological Bulletin, 95, 3-28. and behavioral outcome. British Journal of Psychiatry, JJ3, Sutherland, R., Kolb, B., Schoel, W., Whishow, I., &Davies, D. 1183-1199. (1982). Neuropsychological assessment of children and Rutter, M., & Shaffer, D. (1980). DSM-111, astepforwardorback adults with Tourette syndrome: A comparison with learning in terms of the classification of child psychiatric disorders? disabilities and schizophrenia. In A. J. Friedhoff & T. N. Journal of the American Academy of Child Psychiatry, /9, Chase (Eds.), Gilles de Ia Tourette syndrome (pp. 311-322). 371-394. New York: Raven Press. Rutter, M., & Taylor, E. (1978). Hyperkinetic disorder in psychi- Tager-Flusberg, H. (1981). On the nature oflinguistic functioning atric clinic attenders. Developmental Medicine and Child in early infantile autism. Journal ofAutism and DevelopmenNeurology, 20, 279-299. tal Disorders, 11, 45-56. Sackeim, H. A., Decina, P., &Malitz, S. (1982). Functional brain Tanguay, P. (1984). Toward a new classification of serious psyasymmetry and affective disorders. Adolescent Psychiatry, chopathology in childreil. Journal of the American Academy 10, 320-335. of Child Psychiatry, 23, 373-384. Seidel, U. P., Chadwick, 0., & Rutter, M. (1975). Psychological Tanner, C. M., Goetz, C. G., & Klawans, H. L. (1982). Chodisorders in crippled children: A comparative study of chillinergic mechanisms in Tourette syndrome. Neurology, 32, dren with and without brain damage. Developmental Medi1315-1317. cine and Child Neurology, 17, 563-573. Tarter, R. E., Hegedus, A.M., Alterman, A. 1., & Katz-Garris, Selz, M., & Reitan, R. M. (1979). Rules for neuropsychological L. (1983). Cognitive capacities of juvenile violent, nondiagnosis. Journal of Consulting and Clinical Psychology. violent, and sexual offenders. Journal ofNervous and Mental 47, 258-264. Disease, 171, 564-567. Shaffer, D. (1978). "Soft" neurological signs and later psychiTaylor, E. (1985). Syndromes of overactivity and attention deficit. atric disorder-A review. Journal of Child Psychology and In M. Rutter & L. Herson (Eds.), Child and adolescent psyPsychiatry, 19, 63-65. chiatry: Modern approaches (2nd ed.) (pp. 424-443). OxShaffer, D., O'Connor, P. A., Shafer, S. Q., & Prupis, S. (1983). ford: Blackwell. Neurological ''soft signs'': Their origins and significance for Taylor, H. (1983). MBD: Meanings and misconceptions. Journal behavior. In M. Rutter (Ed.), Developmental neuropsychiaof Clinical Neuropsychology, 5, 271-287. try (pp. 144-163). New York: Guilford Press. Torrey, E. F., Hersh, S. P., & McCabe, K. D. (1975). Early Shaffer, D., Schonfeld, I., O'Connor, P. A., Stokman, C., Trautchildhood psychosis and bleeding during pregnancy: A proman, P., Shafer, S., & Ng, S. (1985). Neurological soft spective study of gravid women and their offspring. Journal signs. Archives of General Psychiatry, 42, 342-351. of Autism and Childhood Schizophrenia, 5, 287-297. Shapiro, A., Shapiro, E., & Wayne, H. (1973). Treatment of Tramontana, M. G. (1983). Neuropsychological evaluation of Tourette's syndrome with haloperidol: Review of thirty-four children and adolescents with psychopathological disorders. cases. Archives of General Psychiatry, 28, 92-97. In C. J. Golden & P. J. Vicente (Eds.), Foundations of Shapiro, T., Burkes, L., Petti, T. A., & Ranz, J. (1978). Conclinical neuropsychology (pp. 309-340). New York: Plenum sistency of "nonfocal" neurological signs. Journal of the Press. American Academy of Child Psychiatry, 17, 70-78. Tramontana, M.G., & Boyd, T. A. (1986). Psychometric screenSimon, N. (1975). Echolalic speech in childhood autism: Considing of neuropsychological abnormality in older children. Ineration of possible underlying loci of brain damage. Archives ternational Journal of Clinical Neuropsychology, 8, 53-59. of General Psychiatry, 32, 1439-1446. Tramontana, M.G., & Hooper, S. R. (1987). Discriminating the Small, J. (1975). EEG and neurophysiological studies of early presence and pattern of neuropsychological impairment in infantile autism. Biological Psychiatry, /0, 385-397. child psychiatric disorders. International Journal of Clinical Smith, H. F., Bemporad, J. R., & Hanson, G. (1982). Aspects of Neuropsychology, 9, 111-119.
106
CHAPTERS
Tramontana, M. G., Hooper, S. R., & Nardolillo, E. M. (in to criminal disorders: The assessment of impulsivity and its press). Behavioral manifestations of neuropsychological imrelation to crime. Advances in Behavioral Research and Therapy. 2, 159-177. pairment in children with psychiatric disorders. Archives of Clinical Neuropsychology. Warren, R., Margasetten, N., Pace, N., & foster, A. (1986). Tramontana, M.G., Klee, S. H., & Boyd, T. A. (1984). WISC-R Immune abnormalities in patients with autism. Journal of interrelationships with the Halstead-Reitan and Children's Autism and Developmental Disorders, 16, 189-197. Luria Neuropsychological Batteries. International Journal of Wender, P. (1971). Minimal brain dysfunction in children. New Clinical Neuropsychology, 6, 1-8. York: Wiley. Tramontana, M.G., & Sherrets, S.D. (1985). Brain impairment Werry, J. S. (1972). Organic factors in childhood psychoin child psychiatric disorders: Correspondences between neupathology. In H. C. Quay & J. S. Werry (Eds.), Psychoropsychological and CT scan results. Journal ofthe American pathological disorders of childhood. New York: Wiley. Academy of Child Psychiatry, 24. 590-596. Wilson, H., & Staton, R. D. (1984). Neuropsychological changes Tramontana, M. G., Sherrets, S. D., & Golden, C. J. (1980). in children associated with tricyclic antidepressant therapy. Brain dysfunction in youngsters with psychiatric disorders: International Journal of Neuroscience, 24, 307-312. Application of Selz-Reitan rules for neuropsychological diWing, L. (1971). Perceptual and language development in autistic agnosis. Clinical Neuropsychology, 2, 118-123. children: A comparative study. In M. Rutter (Ed.), Infantile Tsai, L. Y., Jacoby, C. G., & Stewart, M. A. (1983). Morautism: Concepts, characteristics, and treatment. London: phological cerebral asymmetries in autistic children. BiologChurchill Livingstone. ical Psychiatry, 18, 317-327. Wolf, B. A., & Tramontana, M.G. (1982). Aphasia Screening Tsai, L. Y., Jacoby, C. G., Stewart, M. A., & Beisler, J. M. Test interrelationships with complete Halstead-Reitan test (1982). Unfavourable left-right asymmetries of the brain and results for older children. Clinical Neuropsychology, 4, 179autism: A question of methodology. British Journal of Psy186. chiatry, I40, 312-319. Woodrow, K. (1974). Gilles de Ia Tourette's disease-A review. Tsai, L., Tsai, M., & August, G. (1985). Brief Report: ImplicaAmerican Journal of Psychiatry, 131, I 000-1003. tions of EEG diagnosis in the subclassification of infantile Woods, B. T., & Eby, M. D. (1982). Excessive mirror moveautism. Journal ofAutism and Developmental Disorders, 15, ments and aggression. Biological Psychiatry, I7, 23-32. 339-344. Zambelli, A. J., Stamm, J. S., Maitinsky, S., & Loiselle, D. L. Tucker, D. M. (1980). Lateral brain function, emotion, and con(1977). Auditory evoked potentials and selective attention in ceptualization. Psychological Bulletin, 89, 19-46. formerly hyperactive adolescent boys. American Journal of Voorhees, J. (1981). Neuropsychological differences between juPsychiatry, 134, 742-747. venile delinquents and functional adolescents. Adolescence, Zinkus, P. W., &Gottlieb, M.l. (1978). Learning disabilities and I6, 57-66. juvenile delinquency. Clinical Pediatrics, 17. 775-780. Wardell, D., & Yeudall, L. (1980). A multidimensional approach
6 Neuropsychological Sequelae of Chronic Medical Disorders RICHARD A. BERG AND JOHN C. LINTON
Introduction In the past, neuropsychologists were largely concerned with the evaluation and study of neurological conditions that result in impairment in intellectual functioning. The vast majority of both clinical work and published research was done with adults. By comparison, child clinical neuropsychology is still in its infancy. Both clinicians and researchers who have worked with children have tended to focus their energies largely on the brain itself and have viewed most problems of concern to them as occurring within the brain or some other portion of the central nervous system (CNS). However, child neuropsychologists are becoming increasingly involved in the evaluation of individuals who suffer from diseases that may affect any part of the body. Although the brain and other parts of the body are separate in terms of anatomy, they function as an integrated whole. Thus, when other organ systems are affected by a disease process, the brain in general, and cognitive functioning in particular, also may become impaired. This impairment may result from damage to brain tissue from the disease itself, or alternatively, brain dysfunction may occur as a secondary effect of a disease process elsewhere in the body. For instance, the failure of other organ systems to provide nutrients to the brain may result in diminished cognitive functioning. The notion of multiply interactive systems is primary to the discussion of diseases and conditions presented in this chapter. As
RICHARD A. BERG AND JOHN C. LINTON • Department of Behavioral Medicine and Psychiatry, West Virginia University Medical Center-Charleston Division, Charleston, West Virginia 25326.
nothing in the human body functions in total independence, there can be no single causal mechanism. Easily acknowledged on one level, this concept is both pervasive and essential to the understanding of brain-body relationships. In the assessment of a child with medical problems, however, it is important that clinicians consider multiple causes for any noted neuropsychological dysfunction. Additionally, psychiatric and social problems may impact on a child's behavior and overall functioning. The determination of the presence and severity of any brain effects thus requires knowledge of the possible contribution of a variety of factors including the disease itself, those organ systems directly and indirectly affected, the specific phase of the illness, any current medical treatment, premorbid personality, the coping capacity of the child, and the child's estimated functional level prior to the illness. In many disease conditions, the cognitive sequelae only have been assumed due to clinically reported mental or behavioral changes of some children with the disease. There has been comparatively little research on the neuropsychological effects of individuals suffering from a great many nonneurological diseases. Even when CNS effects are reported as possible or frequent, there is little understanding of the specific types of cognitive deficits likely to occur with differing disease processes, and even less is known about recovery patterns or residual effects. . In this chapter we will discuss the functioning of each major organ system in the body and the ways in which its malfunction may potentially impact on brain functioning. Additionally, we will attempt to pull together the comparatively little research that has been done on disease processes specific to an organ system and the neuropsychological effects that have been reported in the literature. As the reader will
107
108
CHAPTER 6
note, in a number of cases specific neuropsychological data is not available. In those instances, clinical symptomatology that implicates possible neuropsychological dysfunction will be presented.
The Brain and CNS Basic brain structure and function are reviewed in other sections of this book. We are concerned here with the function of the brain in relation to other body processes. The brain has unusual energy requirements. Although it comprises only about 2% of total body weight, it receives roughly 15% of cardiac output, and accounts for 20% of the body's oxygen consumption (Freedman, Kaplan, & Sadock, 1976). As a consequence of this high energy demand, brain cells are extremely sensitive to alterations in their energy supply, mainly oxygen and glucose. Even mild energy deficits can impair the function and integrity of brain cells (Ariel & Strider, 1983). In the normally functioning brain, energy is obtained through a process of oxidation of glucose to carbon dioxide and water. The energy is then expended in the transportation of various compounds across cell membranes and for the synthesis of other cell constituents. Because oxygen and glucose are transported to brain cells by blood, an adequate cerebral blood flow is essential for brain metabolism. Additionally, adequate availability of nutrients is dependent upon proper functioning of the digestive system. When parts of the brain are damaged, other organ systems may or may not be disrupted depending upon which portion(s) of the brain is involved. If, for example, damage occurs to cortical areas, generally only an individual's cognitive and sensorimotor skills are affected. Subcortical damage may disrupt the automatic functioning of other systems, however, such as heart rate, blood pressure regulation, breathing, hormonal balance, water regulation, or immune response. Such disruption can lead to further disability and potentially to more damage to the brain. If peripheral nerves are damaged, only the area served by those nerves typically demonstrates impairment. However, if a major organ system is involved, it may begin to function improperly, creating imbalances in other systems that may, in tum, impact on brain functioning. Thus, it appears that all parts of the body in some way contribute to maintaining brain functioning and vice versa, as a disturbance in one system is highly likely to lead to a disruption elsewhere.
Infections of the CNS The effects of encephalitis on the developing brain have been of interest since the outbreak of epidemic encephalitis following World War I, which resulted in high mortality among children and a high frequency of subsequent psychiatric morbidity (Graham, 1983). Ebaugh (1923), Kennedy (1924), and Strecker (1929) all reported studies of children who had been followed for a number of years after the initial acute illness. The acute phase of encephalitis was characterized by sleepiness, fever, and other signs of localized CNS involvement, which was followed by a gradual onset of a number of significant personality changes. Ebaugh (1923) reported a wide range of behavioral and emotional sequelae that involved insomnia with nocturnal agitation, affective disorders of the depressive type, hysterical reaction, and unwarranted fearfulness as well as mental retardation. Since the 1920s, reports of epidemic encephalitis and subsequent behavioral sequelae have been sporadic and Graham (1983) noted that encephalitis is now generally considered a rare cause of childhood cognitive disturbance. Levy (1959) described 100 children with hyperkinetic and antisocial behavior disorders to whom he ascribed the cause as encephalitis. However, since that study, doubt has been raised as to the actual etiology of the disorders manifested by these children. Sabatino and Cramblett ( 1968) reported that 14 children who had contracted documented cases of California encephalitis virus between the ages of 5 and 14 years demonstrated auditory perceptual deficits as well as unspecified deficits in visual perception. A variety of emotional disorders were also reported including nervousness, hyperactivity, restlessness, and disruptive behavior together with learning problems. Meningitic infections also must be considered when discussing potential adverse effects of CNS infections. One of the most comprehensive studies to date, however, has not revealed any significant cognitive deficits in children (Lawson, Metcalfe, & Pampiglione, 1965). Ninety-nine children whose infections had been contracted between the ages of 2 months and 15 years were followed for I to 8 years after recovery from the illness. No significant declines in intellectual functioning were noted nor were any specific cognitive deficits identified. There was, however, some suggestion that those children who had nonbacterial infections were more likely to show learning difficulties than those who had contracted bacterial illnesses. About 21% of the children dem-
NEUROPSYCHOLOGICAL SEQUELAE OF CHRONIC DISORDERS
onstrated EEG abnormalities; however, there was no correlation between EEG findings and intellectual, behavioral, or learning deficits. A smaller-scale study on 18 children with meningitis found that the only adverse sequelae occurred in the few children who had had seizures in the acute phase of the disease (Fee, Mariss, Kardash, Reite, & Seitz, 1970).
Brain Tumors After malignancies of the blood-forming tissues such as leukemia, brain tumors are the most common type of malignancy in children (Graham, 1983). It has been calculated that there are approximately 600 new cases in the United States annually (Till, 1975). Approximately 60% of childhood brain tumors occur in the subtemporal part of the brain, and of these most are either medulloblastomas or cerebellar astrocytomas. The remainder consist of subtentorial tumors or tumors of the brain stem and adjacent structures. About 3% of brain tumors are metastatic, in marked contrast to that found in adults. Surprisingly, it is unusual for children with brain tumors to present with symptoms of intellectual decline or behavioral change (Graham, 1983). Headache and vomiting are the most common present symptoms, and although there may be some accompanying irritability, this does not typically lead to any diagnostic confusion (Till, 1975). There is one distinct exception to this, however-gliomas in the pontine area. Malignancies in this brain region generally present with what are described as striking personality changes (Arseni & Goldenberg, 1959; Cairns, 1950; Lassman & Arjona, 1967). Characteristically, the symptom pattern seen involves a period of withdrawal, apathy, and lethargy followed by aggression, hyperactivity, temper tantrums, and physical violence. These tumors usually occur between the ages of3 and 13 years. The course of this tumor may last several years and the outcome is generally fatal despite treatment with radiation therapy. Other types of brain stem tumor are likely to present with gait disturbance and symptoms such as squint indicating cranial nerve involvement, but behavioral changes involving lethargy, · irritability, inability to concentrate, enuresis, and sleep disturbance also have been known to occur (Panitch & Berg, 1970). The prognosis for brain malignancies in children is poor at best, despite the use of the best available treatments such as surgery and irradiation. The clearest information is available concerning the most common brain tumor in children-medullobla s-
109
tomas. Studies suggest that the 5-year survival rate is between 40 and 75%, and 30 and 50% after 10 years (Bloom, Wallace, & Henk, 1969; Hope-Stone, 1970). In one study, 18 of 22 survivors were reported to be without serious deficits (Bloom et al., 1969). Two children of those followed were found to have partial disability, and two others demonstrated significant intellectual deterioration. It is noteworthy that the two children who demonstrated intellectual decline were those who were diagnosed at the youngest ages (11 and 15 months). These findings with respect to age are similar to those reported in some studies of children with leukemia comparing early and later diagnosed children: children diagnosed at an early age appeared to be more at risk for the development of cognitive dysfunction (Eiser, 1979; Meadows et at., 1981). (Further discussion about the effects of leukemia and its treatment can be found later in this chapter.) Matson and Crigler (1969) studied children treated for craniopharyngioma and found no particular psychological or behavioral problems even though the survivors were frequently partially sighted and required hormone replacement therapy. Systematic studies of a child's neuropsychological status following treatment for a brain malignancy are quite rare and clearly are needed. Much of the current literature also notes the need for further research on the effects of brain tumors, particularly to determine whether those survivors who do not manifest directly observable deficits show deficits of a far less obvious nature.
Neuromuscular Diseases There are a variety of neuromuscular diseases that afflict children and it is beyond the scope of this chapter to detail the effects of each on the cognitive functioning of children. One disease entity, Duchenne muscular dystrophy (DMD), will be offered as a possible model for the effects such diseases can have on the developing brain. It is important to note that the sequelae of such diseases tend to have variable identifiable effects depending on a wide variety of factors including age at diagnosis, age at testing, and so on. DMD is a hereditary disease causing progressive muscular weakness and degeneration of skeletal muscle tissue. Its course generally includes confinement to a wheelchair by age 11 and death in the late teens. It affects males almost exclusively. Many studies of intellectual functioning in DMD patients have reported diminished or retarded intellec-
110
CHAPTER 6
tual development, supporting the position of Duch- rebral hemorrhages, and blindness have been reenne in 1872 (Dubowitz, 1979). The group IQ score ported in severe cases (Aita, 1964). Preliminary regenerally averages about 85, one standard deviation search conducted with children diagnosed as having below the mean of the general population (Dubowitz, sickle cell disease indicated that the overall intellec1977; Karagan, 1979). tual capabilities of these children were lower than Although there is support for the notion of intel- those of an age- and sex-matched group of black lectual impairment in DMD patients, there is no com- children (Berg & Wilimas, 1983). In this pilot promon consensus of this in the research literature. ject, the Wechsler Intelligence Scale for ChildrenSome have reported a decline in intellectual perfor- Revised (WISC-R) and the children's version of the mance (e.g., Black, 1973; Florek & Karolak, 1977) Luria-Nebraska Neuropsychological Battery were whereas others have found no significant differences administered both to a group of 30 black children in longitudinal studies (e.g., Cohen, Molnar, & Taft, with the disease who had not undergone hyper1968; Worden & Vignos, 1%2). transfusion of packed red cells and to a similar group A recent study attempted to clarify the picture of black children without the disease who had been somewhat by studying 14 younger and 11 older chil- selected based on age and sex. In all cases, children dren with DMD. It was found that younger children with the disease performed significantly more poorly with DMD performed more poorly on tasks requiring on all IQ measures. No significant differences on the some language and attentional-organizational skills, Luria-Nebraska tests were found although the results but not on visual-motor tasks. The older group had were reported as following the same direction as that generally higher IQ levels in the average range and found on the WISC-R. the younger group had low-average IQ scores (Sollee, Latham, Kindlon, & Bresnan, 1985). These authors noted that individuals with DMD do not appear Polycythemia Vera to suffer from fixed, global cognitive deficits. Just as an abnormal decrease in erythrocytes can Rather, deficits appear to vary at different ages with lead to abnormal cognitive status, the converse is also no specific patterning evident to date. true. An abnormal increase in red blood cell production in the bone marrow (polycythemia vera) can cause erythrocytes to clump together, creating a sitBlood and Circulatory System uation that slows blood flow and impedes circulation. Despite the fact that in this condition there is an adeThe primary function of the blood system is that quate oxygen supply, there is difficulty in breathing of a carrier and delivery service of transporting oxy- and the individual may become cyanotic. Brain funcgen from the lungs to tissues and returning carbon tioning may become lowered due to insufficient cirdioxide, conveying metabolites to tissues, and re- culation or blockage of a cerebral blood vessel. Aita turning waste products for disposal. It has other ( 1964) discussed a number of commonly reported important functions such as maintaining water con- neurological symptoms in diseases that result in extent of the tissues, harboring the body's defense cells, cess erythrocytes. These include headaches, dizzicarrying hormones that regulate a variety of body ness, visual and hearing difficulties, and parasfunctions, and helping to maintain and regulate body thesias. Children who tend to hemorrhage easily may temperature. A disruption either in the blood or with- show more focal deficits such as aphasia and hemiin the circulatory system can directly impact on brain paresis, or they may exhibit a progressing dementialike condition as more and more cerebral tissue is functioning. destroyed by repeated hemorrhaging (Aita, 1964; Ariel & Strider, 1983).
Anemia
Anemia, or an abnormal decrease in red blood cells, can produce variable CNS effects. Erythrocytes (red blood cells) contain hemoglobin, which carries oxygen, and any unusual decrease in the number of these cells can result in an overall lowering of brain functioning due to cerebral hypoxia (deficiency in oxygen supply). Convulsions, diffuse organic brain syndromes, focal vascular lesions, ce-
Excessive Increases or Decreases in Platelets A severe reduction in the number of platelets or defects in coagulation factors in the blood may result in spontaneous bleeding. This can be a primary disease (e.g., thrombocytopenia purpura) or it can develop secondarily to another disease process such as leukemia (discussed below), toxic chemical ex-
NEUROPSYCHOLOGICAL SEQUELAE OF CHRONIC DISORDERS
posure, irradiation, infection, or massive blood transfusion. If such bleeding occurs within the brain, it may be single or multiple, small or large, and may resemble a focal stroke (Aita, 1964). The cognitive sequelae of such an incident can be tremendously varied depending on the location and size of the hemorrhage and can range from comparatively minor to pervasive with the patient existing in a vegetative state (Walton, 1977). In the more minor occurrences of such bleeding, temporary general confusion, paresis, and convulsions can be seen (Heron, Hutchinson, Boyd, & Aber, 1974). Matoth, Zaizov, and Frankel (1971) reported that 20 children with chronic thrombocytopenia had been found to have learning and behaviorial problems when compared with patients with other medical disorders. Their study, which used the WISC, Bender Visual-Motor Gestalt, and Human Figure Drawing tests, revealed no statistical differences between the two groups on any of the tests, however. It was noted that over two-thirds of the group of children with thrombocytopenia exhibited ''soft'' neurological signs of minimal brain dysfunction. Over half of this group also demonstrated mild, diffuse EEG abnormalities. To date, there has been no longterm follow-up of such groups to determine if noted behavioral or cognitive abnormalities persist into adulthood. An excessive increase in the number of platelets can result in the formation of a thrombus and subsequent blockage of a blood vessel. Tissue supplied by the blocked vessel will then receive an insufficient supply of blood, and an ischemic condition wherein the tissue starves may result. If the tissue dies or is damaged, an infarction is the result. Thrombus formation anywhere in the body can be serious because of the high tendency for the thrombus to pass through the heart and be carried to the lungs or brain. When cerebral blood vessels become blocked and an infarction occurs, deficits in cognitive functioning can occur (Walton, 1977). If the blocked vessel supplied a small area of the brain, the cognitive sequelae will generally resemble those seen with a fairly discrete cerebral lesion. Where the blocked cerebral vessel supplied a larger region of the brain, pervasive deficits can result.
Leukemia Leukemia is a disease in which there is an uncontrolled multiplication of certain white blood cells resulting in their accumulation in large numbers (LODAT, 1981). An abnormal growth and division
111
of lymphoblasts (one type of white blood cell) in the bone marrow results in acute lymphocytic leukemia (ALL). ALL is perhaps the most prevalent form of malignancy in children and the one that has been the most heavily researched with respect to the effects of the disease and its treatment. This is because the therapy for the disease has become so effective that long-term disease-free survival can be expected in at least 50% of patients (Bowman, 1981; Mauer, 1980). Survival of these children has permitted an increasing emphasis on the late sequelae of ALL and its treatment. Treatments for other childhood malignancies have not been quite so effective. Thus, research into the cognitive sequelae has been highly restricted. The research investigating the long-term effects of ALL and its treatment may, however, be used as an at least temporary model for the effects of other malignant disease processes. A complicating factor in the study of the neuropsychological functioning of children with ALL lies in the standard treatment regimen for the disease. Common treatment involves the intrathecal and intravenous administration of a neurotoxic medication, methotrexate. Coupled with this is the administration of at least 1800-2400 rads of cranial irradiation, which is done prophylactically in an attempt to destroy those leukemic cells that may have migrated to the brain. Some recent reports have suggested that cranial irradiation may not be necessary in the treatment of what is referred to as "standard-risk" leukemia and that such prophylactic measures need only be employed with ''high-risk'' patients whose disease is diagnosed at a more advanced stage and is more likely to have invaded the CNS (Copeland, Pfefferbaum, Aetcher, Jaffee, & Culbert, 1982). In any event, it becomes clear that it is very difficult, if not impossible, to assess the effects of the disease alone in such instances. Despite the fact that ·a good deal of research has been conducted on the effects of ALL, the actual effects of the disease still tend to be inconsistent. Eiser and Lansdown (1977) and Goff, Anderson, and Cooper (1980) found that when leukemic children who had received CNS irradiation were evaluated, significant deficits emerged. This was particularly true if the disease was diagnosed and treated prior to the age of 5 years. Deficits included declines in overall intellectual abilities as well as a pattern of distractibility and memory deficits. Other investigators have noted deficits involving IQ declines (Meadows et al., 1981) and motor speed deficits (Eiser, 1978). In contrast, several investigators have found that the disease and its treatment lead to no documented dysfunction (Obetz et al .. 1979; Ivnik, Col-
112
CHAPTER 6
ligan, Obetz, & Smithson, 1981). A longitudinal study in which a group of leukemic children were followed from original diagnosis and treatJ;IIent for a period of at least 3 years found no specific patterns of deficits or any IQ declines (Berget a!., 1983). However, almost half of the children had performance patterns suggestive of the presence of mild, specific learning dysfunction when performance patterns were analyzed individually rather than as a group. Copeland eta!. ( 1982) suggested that the cranial irradiation is the principal factor leading to cognitive dysfunction in children with ALL. In a carefully controlled investigation, two groups of children treated for leukemia were evaluated. One group received the standard treatment including prophylactic cranial irradiation and the other group received no irradiation. The former group were found to perform significantly worse on a number of neuropsychological measures including generally lower IQ scores and poorer nonverbal memory. Significant differences were also found on the Wechsler Information, Similarities, Arithmetic, Block Design, and Object Assembly subtests. In general, the research conducted to date indicates that there is likely a higher than normal chance that any type of treatment for leukemia and/ or the disease itself is potentially detrimental to a child's neuropsychological development. Research into these effects is currently ongoing in a large number of facilities and is attempting to delineate more specifically the areas of functioning that may be affected as well as the influences of various types of treatment for childhood malignancies.
Endocrine System The endocrine or ductless gland system is primarily involved in the production of hormones for correlating and regulating bodily processes. Such glands include: the pituitary, which lies in a depression of the sphenoid bone between the roof of the mouth and the hypothalamus; the pineal, which is just posterior to the pituitary; the adrenals, which are attached to the top of each kidney; the thyroid, located in front of the trachea just below the voice box; the parathyroids, which are embedded in the thyroid; the thymus, found near the lower part of the trachea; the pancreas, found in the curvature between the stomach and small intestine; the ovaries, located near the uterus; and the testes, suspended in the scrotum. Study of this system reveals complex interre-
lationships among the various endocrine glands. Hormones are exceedingly powerful agents, and in some instances their activities cover practically the entire body. In most cases, they interact normally and well. Production of hormones is usually regulated by the bodily requirements for each, and when this need is met, production is decreased or antihormones are released. Uncontrolled excesses or insufficiencies of glandular secretion are responsible for a variety of disorders of development and metabolism, most of which have implications for the integrity of the brain. One such disorder-diabetes mellitus-has received a great deal of attention by neuropsychologists of late.
Diabetes Mellitus Diabetes mellitus is a disease complex resulting from abnormalities in carbohydrate metabolism, due to insufficient production of insulin in the pancreas. Because of this lack of insulin, diabetics have chronically high blood glucose levels (hyperglycemia), and excrete a great deal of unmetabolized sugar as well as many salts and minerals essential to health. Diabetes mellitus is a heterogeneous group of disorders rather than a single disease, and its exact cause is unknown (Miller & Sperling, 1986). However, two general classifications of diabetes are common. Both forms have the potential to injure large and small blood vessels, leading to deterioration of peripheral and autonomic nerves, the cardiovascular system, the eyes, and the kidneys (Cirillo et al., 1984; Pfeifer et al., 1984). As such, diabetics are at increased risk for heart disease, stroke, kidney dysfunction, blindness, and peripheral neuropathy. Adult onset, also known as type II or non-insulin-dependent, diabetes mellitus (NIDDM) is the most common, accounting for over 90% of all diabetics, and affecting about 5 million adults in the United States. Occurring typically in overweight individuals past the age of 40, the onset is subtle, and diagnosis is often made secondary to problems with the vascular system. NIDDM is characterized by diminished but not absent secretion of insulin by the pancreas. Treatment is usually by diet change and the use of medication for the stimulation of insulin production; exogenous insulin is not necessary. Juvenile onset, also known as type I or insulindependent, diabetes mellitis (IDDM) is a common chronic disease estimated to affect 150,000 children and adolescents (Cerreto & Travis, 1984) and
NEUROPSYCHOLOGICAL SEQUELAE OF CHRONIC DISORDERS
400,000 adults (Carter Center, 1985) in the United States. Males and females are equally affected, and peak presentation is seen at the time of puberty, although IDDM is diagnosed from early childhood through early adulthood (Miller & Sperling, 1986). In IDDM the pancreas stops producing insulin entirely. Presentation of symptoms is usually clear and dramatic, with polyuria, polydipsia, polyphagia, and rapid weight loss over a period of about 1 month. If untreated, severe hyperglycemia can lead to ketoacidosis, diabetic coma, and death. This previously fatal disorder was converted to a manageable chronic disease after 1922 with the availability of exogenous insulin (Johnson & Rosenbloom, 1982). For the youngster with lOOM, management of near-normal metabolism involves constant monitoring of bodily systems and daily insulin injections. Insulin needs vary with nutrition, exercise, physical health, and emotional state. As mentioned, insufficient insulin can lead to dangerous hyperglycemia. Conversely, too much insulin or too little food, or an imbalance of food, exercise, and insulin can result in a marked decrease in blood sugar (hypoglycemia). Hypoglycemia can progress from an insulin reaction, with mental confusion and anxiety, to hypoglycemic seizures, to insulin coma (Miller & Sperling, 1986). This metabolic seesaw may have important implications not only for the psychological adjustment, but also for the brain of the diabetic youngster. Extensive investigation of neuropsychological functioning in lOOM is currently being carried out by at least three research teams, namely those of Christopher Ryan at the University of Pittsburgh; Joanne Rovet at the Hospital for Sick Children in Toronto; and Clarissa Holmes at the University of Iowa. An excellent review of this area is that by Ryan and Morrow (1987a). Discussing the history of research in this area, they point out that early on, diabetes per se was thought to be benign with respect to impact on brain functioning, the only connection thought to be secondary to involvement of renal or cardiovascular disease in older patients who had the disease for many years. A series of studies from the 1930s through the 1960s challenged this notion by comparing the intelligence of diabetic children with general norms, in sum yielding results that were equivocal. Although the findings were inconclusive, two important methodological innovations were introduced in these series. These were the use of nondiabetic siblings controls to control for effect of family influences and
113
socioeconomic status, and the attempt to experimentally relate specific diabetic characteristics such as age of onset and duration of illness to outcome (Ack, Miller, & Weil, 1961). ·One important global finding from such work was that age of onset seemed to be an important variable, with those diagnosed as diabetic before the age of 5 having average IQs l 0 points lower than their siblings. A trend was also seen suggesting an inverse relationship between number of hypoglycemic seizures experienced and measured intelligence. Thus, although no clear evidence of specific neurobehavioral dysfunction in diabetic children emerged, researchers were made aware that there may be some neurobehavioral differences between diabetic and normal children, and that the age of onset and the number of hypoglycemic seizures noted in the history may relate to the extent of this difference (Ryan & Morrow, 1987a). Because global measures of intelligence were used in that series of studies, differences between diabetics and controls could not be assigned to specific structural or operational changes in the brain. However, a series of EEG studies did fmd a significantly higher number of clinically abnormal EEGs in a group of diabetic children as compared to agematched normal controls, further finding that the variable most related to this EEG difference was the number of severe hypoglycemic episodes (EegOlofsson & Peterson, 1966), orthenumberofsevere episodes of both hypoglycemia and hyperglycemia (Haumont, Dorchy, & Pelc, 1979). Ryan and Morrow (1987b) posited that this IQ and EEG evidence seemed to demonstrate that diabetic children and adolescents showed a greater tendency than their nondiabetic age mates to have mild, diffuse brain dysfunction, and that multiple episodes of severe hypoglycemia were in some fashion responsible for the development of this ''diabetic encephalopathy.'' They noted with some surprise that although these findings fmnly established a basis for further investigation of this diabetes-related organic syndrome, such research essentially dried up for no apparent reason during the 1970s, in favor of studies of the psychosocial aspects of diabetes (see Cerreto & Travis, 1984). Given these early findings, and the fact that a medical colleague found a high incidence of school difficulty in his diabetic patients, Ryan and his associates at the Children's Hospital in Pittsburgh began a series of neurobehavioral studies to reassess the degree to which diabetic youngsters are at risk to develop cognitive deficits secondary to CNS defects.
114
CHAPTER 6
The goal of their series of studies was to describe particular neuropsychological difficulties found in this population, and to relate these problems if possible to specific variables associated with each case. Based upon previous findings and some preliminary work, they focused on the examination of age at onset, duration of disease, and degree of metabolic control as they related to cognitive functioning, which was measured by neuropsychological testing. To test the notion that both age at which diabetes mellitus is diagnosed and duration of the disease are potent variables, Ryan, Vega, and Drash (1985) administered a comprehensive neuropsychological battery to 125 randomly selected diabetic adolescents, all of whom had been diabetic for at least 3 years, but ranged in their age of onset from 2 months to 14 years. They divided their subjects into an "early onset" group (diagnosed before age 5, n = 46), a "later onset" group (diagnosed after age 5, n = 79), and a sibling control group (n = 83). A factor analysis of the cognitive measures used in their testing battery generated five clusters of tests, namely general intelligence, visuospatial processes, learning and memory, attention and school achievement, and mental and motor speed. Statistical analyses found significant differences between early and late onset subjects on all five clusters. Further, cutting scores of two standard deviations below the control mean were assigned to each of the 20 tests that discriminated early from late onset subjects, with at least three such low scores necessary to be seen as "impaired." On the basis of these rules, 24% of the early onset were seen as impaired, whereas only 6% of both the late onset and controls met impairment criteria. Further analysis of these data suggested that age at onset and disease duration differentially affected the testing results. Age at onset appeared to predict results of tests measuring .. fluid intelligence," described by the authors as adaptive abilities used to process relatively unfamiliar information in novel ways, such as scanning and identification of visual stimuli. Duration of illness seemed more able to predict performance on tests tapping ''crystallized intelligence," defined as the use of well-practiced skills depending largely on stored knowledge, such as reading and spelling skills, and sequencing ability. Regarding the differential effects of duration and age at onset, there is some evidence to indicate that the relationship between duration and crystallized intelligence can be accounted for by the fact that school attendance is a factor in both. Ryan, Longstreet, and Morrow (1985) found that diabetics missed significantly more school than
matched controls over time, and further that cognitive and achievement test findings in this group were best predicted by measures of school attendance. Therefore, perhaps like most chronically ill youngsters, diabetics miss a significant amount of school (Gortmaker & Sappenfield, 1984), and this attendance problem may reduce their ability to master classroom-related learning, or crystallized intelligence. Thus, longer duration of diabetes would lead to greater attendance problems, and more difficulty in school. Ryan, Vega, and Drash (1985) and Ryan and Morrow (1987), on the other hand, suggested that the findings regarding age of onset and performance on tasks assessing fluid intelligence may be due to structural or functional disturbances in the brain. This mild brain damage may develop from multiple episodes of severe hypoglycemia and resultant hypoglycemic seizures early in life. There is some evidence (Temand, Go, Gerich, & Haymond, 1982) that younger diabetics are more sensitive to the effects of insulin, and therefore have more reactive hypoglycemic seizures. This is consistent with the finding of Ryan, Vega, and Drash ( 1985) that early onset diabetics had more of a history of hypoglycemic seizures. In general, the work of Ryan and colleagues suggests that cognitive deficits in diabetics can be seen as early as age 10. Rovet and her colleagues in Toronto have undertaken the study of even younger diabetic patients, in an effort to further examine neurobehavioral findings in this group. Rovet, Ehrlich, and Hoppe ( 1988) administered an extensive series of neuropsychological tests to a diabetic sample including children as young as 6. They divided the sample into 27 early onset (pre-age 4), 24 later onset (postage 4), and 30 sibling controls. In contrast to Ryan's studies, they found no differences among the three groups on intelligence or achievement, actually finding that diabetics outperformed controls on tasks measuring verbal ability. However, they found some interesting results related to gender. Early onset girls performed less well on spatial tasks and had a lower performance IQ than later onset or control girls, but this finding did not hold for the boys. These early onset girls also had more academic problems, including failed grades and special education placement, than the other groups. A multiple regression analysis for each sex separately found that for both genders taken together, the best predictor of verbal performance was socioeconomic status; for girls only, the best predictor
NEUROPSYCHOLOGICAL SEQUELAE OF CHRONIC DISORDERS
of spatial performance was age at onset; and again for both genders, the best predictor of spatial performance was seizures before the age of 5. Rovet (personal communication, February 9, 1988) is currently conducting a prospective study, following newly diagnosed diabetic children to better examine at what point neurobehavioral deficits are evidenced, rather than depending on retrospective study. Other interesting results regarding gender and age at onset include those of Ryan and Morrow (1987b), who found that early onset diabetic adolescent girls had significantly poorer self-esteem, as measured by the Piers-Harris subscales of Physical Appearance and Anxiety, than did early onset boys, later onset boys and girls, and controls. However, the extent to which this represents a result of greater cognitive deficit, bodily changes differentially experienced by girls over time, or a unique coping reaction in the face of chronic illness is undetermined. Ryan and Morrow (1987b) summarized this literature by stating that both their and Rovet's teams have found age of onset of diabetes to be an important risk factor for the development of significant neurological deficits in both children and adolescents. They speculated that this strong association between diabetes early in life and brain dysfunction may be accounted for by two different phenomena, namely that the brain of a young child is very sensitive to the deleterious effects of any sort of metabolic insult, and that this sensitivity may be greater in females; or that the young child has a heightened responsivity to insulin, and thus has more hypoglycemic seizures, with resultant increase in damage to the brain. These findings seem to be consistent with increasing evidence that the time from birth to 5 years may constitute a "critical period" for the development of serious brain dysfunction from a variety of causes with a number of outcomes. It is worthy of note that even though early onset is viewed as an important variable, diabetes appears to affect performance even in those who are diagnosed later in life (Franceschi et al., 1984). Ryan, Vega, Longstreet, and Drash ( 1984) tested 40 diabetic adolescents (aged 12-19) who were classified as late onset (diagnosed at 5-12), and a group of 40 matched nondiabetic controls using a neuropsychological battery and critical flicker fusion. They found that all subjects performed within normal limits, but that the diabetic sample had verbal IQs lower than controls (which may be related to the duration effects and school attendance noted above), but also did less well on psychomotor tasks, performing more slowly than controls. It has been hypothesized that diabetics may develop a charac-
115
teristic personality style that may account for some of this psychomotor task performance difference, which will be addressed briefly below. In general, it appears that early onset diabetes produces mild brain dysfunction as measured on cognitive tests, but no such findings were evident in this later onset group. Another variable to receive research attention is that of degree of metabolic control. As mentioned earlier, poor metabolic control implies a tendency to hyperglycemia, which is implicated in the risk for disorders of the vasculature, both large and small vessels, and hypoglycemia, which has been shown to have clearly deleterious effects on the brain. The work by the teams of Ryan and Rovetdoes not show a relationship between degree of metabolic control and cognition in children and adolescents. However, work by Holmes and her team at Iowa suggests that a recent history of poor metabolic control may increase the risk of mild neuropsychological disturbances in young adults. Holmes (1986) compared two matched groups of diabetic men in their early 20s, one classified as in "good" and the other in ''poor'' control as measured by tested hemoglobin A1C levels. This test measures the relative degree of metabolic control, or avoidance of blood glucose extremes, over the preceding 3 months. In her study, Holmes found that the poor control group had lower scores on the Information and Vocabulary subtests of the WAIS, and also did worse on reaction time tasks. Ryan and Morrow (1987) suggested caution in interpreting this finding, as metabolic control was only measured for the 3 months before testing. They speculated that perhaps if these subjects were out of control as children, they may have attended poorly, and now retrieve poorly as adults. In a typical day the blood glucose level of a diabetic child may vary widely, being dependent upon food, insulin dose, and exercise (Miller & Sperling, 1986). Cerebral metabolism depends upon the ability of serum glucose to circulate freely in the brain. Because little glucose is stored in the brain, when its supply has been compromised there is a lag time before fatty acids are utilized as a backup source (Holmes, Koepke, Thompson, Gyves, & Weydert, 1984). Temporary change in cerebral functioning might therefore be possible at the time of testing, perhaps affecting psychomotor tasks if not global intelligence. In a series of studies (Holmes, Hayford, Gonzalez, & Weydert, 1983; Holmes et al., 1984; Holmes, Koepke, & Thompson, in press) Holmes and her colleagues inspected the acute neuropsychological consequences of deviant blood glucose levels.
116
CHAPTER6
By use of an automatic insulin/ glucose infusion system, they were able to stabilize young diabetics for extended periods at one of three blood glucose levels: hypoglycemia (55-60 mg/dl); euglycemia (110 mg/dl); or hyperglycemia (300 mg/dl). All subjects were tested in all three conditions, using a balanced design in which neither experimenter nor subject knew in which level the subject was. They found that relative to euglycemia, subjects tended to perform more slowly in the hypoglycemic state on a variety of tasks involving simple mental calculation and word production, as well as responding in choice reaction time situations. A tendency was noted for subjects to perform in a similar fashion in the hyperglycemic state. Ryan and Morrow (l987a) summarized such research by stating that the hypoglycemic state reduces efficiency and increases response time on brief tasks. They further speculated about the possibly more dramatic effects that this state might produce on more lengthy tasks under conditions of greater fatigue, a situation that may have occurred during previously described measurements of cognitive functioning in diabetics by use of neuropsychological testing. Finally, there has been some question as to the relative contribution of nonorganic variables to performance on measures of neurobehavioral functioning. A number of studies have commented upon behavior and personality styles among diabetics. Some have stressed problems in the family (Lancet Editorial, 1980; Winter, 1982) and school (Weitzman, 1984), whereas others have focused upon IDDM as a risk factor in certain clinical syndromes such as eating disorders appearing in adolescent females (Daneman, Johnson, & Garfinkel, 1985). Still other authors have tried to conceptualize the effects of diabetes as the child and adolescent attempts to cope with normal developmental tasks at different cognitive stages (Johnson & Rosenbloom, 1982; Cerreto & Travis, 1984). Some of the selfesteem problems found in early onset diabetic females may be of interest here. Research in this area of diabetic personality functioning has been fraught with methodological problems, and at least one study (Skenazy & Bigler, 1985) found that diabetics are no more poorly adjusted than other chronic disease groups, and further that degree of psychological adjustment was not predictive of performance on a battery of neuropsychological tests. Ryan and Morrow (l987b) commented that they have observed a "cautiousness" in their young diabetic patients, perhaps reflecting the youngsters' daily need for constant attention to de-
tail. However, they admit that this is more a matter of clinical lore than evidence. In summary, it is clear that over the past 7 years a good deal of progress has occurred in understanding the neuropsychological correlates of insulin-dependent juvenile onset diabetes in children and adults. Specific neurobehavioral impairments have been identified, as have several diabetes-related variables that appear to be important risk factors for the development of such impairments. Ryan and Morrow (l987a) reiterated that age at onset of IDDM is a most potent variable, with those diagnosed before the age of 5 to be much more likely to show evidence of cognitive impairment than those with onset later than age 5. They suggested that this diabetic encephalopathy yields deficits in a wide selection of cognitive domains, with performance disrupted on measures of attention, learning, memory, problem-solving, visuospatial, and visuomotor efficiency. Thus, some early onset diabetics have lower IQs than their siblings or nondiabetic peers. But not all early onset diabetics have diminished intellectual functioning, and in fact most do not. However, evidence from both electrophysiological and neuropsychological measures suggests that those who have had multiple episodes of serious hypoglycemia early in life are likely to be impaired on a wide range of tasks. These findings imply clearly that because young children are quite insulin sensitive, keeping an excessively tight metabolic control (rigidly preventing hyperglycemia) may increase the risk of starting serious and perhaps debilitating hypoglycemic episodes in these patients. Most diabetic children and adults are late onset, and show relatively subtle impairments. When detected, they tend to appear on difficult information-processing tasks requiring the subject to complete novel assignments as rapidly as possible. The slowness noted may be involuntary (reflective of a transient hypoglycemic state) or voluntary (due to learned caution in the face of decision-making situations). It is also possible that information-processing mechanisms in this population may be disrupted by a complex biochemical disturbance, resulting from a long history of poor metabolic control. And finally, performance may be impaired due to increased absences from school, with related academic problems. All of these hypotheses are grist for the research mill, for at this time there is no strong evidence that extensive structural damage to the brain is directly causative of the subtle deficits sometimes found in patients with late onset IDDM.
NEUROPSYCHOLOGICAL SEQUELAE OF CHRONIC DISORDERS
Respiratory System The respiratory system is comprised of the upper airway, including the nose, pharynx, larynx, and epiglottis; the lower airway, including the trachea, the primary or main bronchi, the segmental bronchi, and bronchioles; and the lungs, located within the thoracic cavity on either side of the heart. The exchange of gases provided by this system is vital to the brain. The cardiovascular system's function is to supply oxygen to body tissues via circulating blood. The blood also removes carbon dioxide produced by metabolism, and transports this waste product to the lungs. Here the carbon dioxide is replaced by oxygen, and the newly oxygenated blood is recirculated to the tissues, including the brain. A network of airways provides the pathway for the transport and exchange of oxygen and carbon dioxide. Under normal circumstances, by inhalation the upper airway provides air for the lower airway, where it is conducted through smaller and smaller pathway branches in each lung field. The final branches of bronchioles (terminal respiratory bronchioles) end in clusters of alveoli, or air-filled sacs. Thus, the working area of the lung is a network of air tubes and blood vessels, through which blood ultimately reaches the alveoli, which are the primary structures for the exchange of carbon dioxide and oxygen (Luckman & Sorenson, 1980). Given that the need for oxygen by the brain is so great, a significant disruption in the functioning of the respiratory system may have negative cerebral consequences.
Bronchial Asthma Asthma is the most common chronic disease of childhood, estimated to occur in 5% of adults and children in the United States, over 10 million people, of whom over 2 million are under the age of 16. The onset of asthma is usually within the first 5 years, although it can occur at any age. A more favorable prognosis appears to be related to an early onset, unless significant asthma attacks begin before the age of 2 (King, 1980). Asthma accounts for nearly one-fourth of all days absent from school by children, and it ranks ~ird ~~ong all chronic illnesses as a cause of physictan vtstts (Sadler, 1982). It also contributes greatly to acute visits to emergency rooms, days in the hospital, and problems related to psychosocial adjustment (Rubin et al., 1986). Although the symptoms of asthma have been
117
recognized for centuries, there is currently no commonly agreed upon definition of the syndrome, nor any. consensus with regard to whether asthma is primarily a medical or a psychological disorder (Creer, 1982). A great deal has been written about the onset of asthma from several psychological perspectives, notably psychoanalytic and behavioral (Sadler, 1982), and a variety of psychological interventions have been presented to deal with this condition (King, 1980). Basically, asthma is a bronchial disorder characterized by airway obstruction that is intermittent, variable, and reversible. The lung pathology may occur in the central or larger airways, and the peripheral or smaller airways. Obstruction of these bronchial airways may be due to smooth muscle constriction, swelling of tissue, swelling of the mucosa, and dried mucus plugs (Chai, 1975; King, 1980). The one common denominator is a peculiar hyperreactivity of the airways, whether to physical, chemical, pharmacologic, or immunologic stimuli (Sadler, 1982). The clinical symptoms present as spasms of difficult breathing, coughing, and wheezing, with the attacks lasting from several minutes to several hours, although in a condition known as "status asthmaticus" obstruction may last for days or weeks. Attacks can vary along a continuum of severity from very mild to very severe, the latter increasing the risk of brain damage or even death (Bierman, Pierson, Shapiro, & Simons, 1975). Although the notion that cerebral anoxia secondary to bronchial asthma attacks may lead to neurobehavioral deficits seems to be defensible and stimulating, very little solid research has been conducted in this area. Dunleavy and Baade (1980) stated that there have been a number of studies reporting on the adaptive behaviors of asthmatic children, particularly as applied to the classroom situation. Most are speculative, whereby observed maladaptive behaviors and learning problems are assumed to be related to organic damage, which is further assumed to be a result of transient hypoxia accompanying their severe asthmatic attacks. In an effort to make use of assessment instruments better designed to detect brain-behavior relationships, Dunleavy & Baade (1980) evaluated a sample of asthmatic children using the Halstead Neuropsychological Test Battery for Children. Their goal was to identify patterns of neurobehavioral deficit characteristic of severely asthmatic children 9 to 14 years of age. Nineteen severely asthmatic subjects and 19
118
CHAPTER 6
matched nonasthmatic controls who had no history of organic damage were administered the Halstead Battery. Significantly poorer test performance was noted for the asthmatic group, with eight Halstead Battery tests showing most difference between the groups. Three of these tests, Trail Making, Tactual Performance, and WISC Mazes, were more sensitive than the others in discriminating asthmatic from nonasthmatic subjects. The authors concluded that the primary deficits of impaired asthmatic children are in visualizing and remembering spatial configurations, in incidental memory, and in planning and executing visual and tactile motor tasks. Using both a classification battery, developed from four of the eight Halstead Battery tests that showed greatest discrimination between groups, and blind clinical analysis, Dunleavy and Baade (1980) identified 7 of the 19 asthmatics (35%) as impaired, whereas only one of the controls was so labeled. They further compared the test score means of the seven neuropsychologically impaired asthmatic children with the Halstead Battery test score means of 9- to 14-year-old brain-damaged (cerebral tumor, traumatic injury, inflammatory disease) group studied by Boll (1974). Their asthmatic sample performed better than did the Boll sample, in line with the clinical assessment of very mild to mild brain damage for the asthmatic children in their.study. The authors also mentioned that five of their asthmatic subjects reported that they had experienced periods of unconsciousness and had "turned blue" during their attacks. Of these five, four were classified by their Halstead scores as impaired. This finding was thought to add credence to the notion that loss of consciousness and cyanosis, which occurs during some severe asthmatic attacks, can contribute to later occurrence of organic dysfunction. Suess and Chai (1981) suggested the conclusions of Dunleavy and Baade were premature, because the possibility of similar performance deficits as a function of antiasthma medications was not taken into account. In essence, the treatment and not the disease, may account for the obtained neurobehavioral deficits. Dunleavy (1981) responded that in their sample they found no relationship between drug use, as obtained from detailed medical history, and neuropsychological test performance. Further, he reported that of the seven children classified as impaired, only three were receiving antiasthma medication, and of the 12 asthmatic children who showed no evidence of performance deficit, seven were receiving such medication. However, Chai (personal communication,
March 21, 1986) and his research team at the National Jewish Hospital in Denver have recently completed a 3-year study of the effects of antiasthma drugs such as steroids on information retention in asthmatic children. Their preliminary analyses suggested that the use of such medications has no noticeable effect on retention in reading and writing, but quite significant effects on the automatic memory required for retention of math skills. In summary, limited research has demonstrated that some severely asthmatic children exhibit very mild to mild brain-damage-like behaviors, that certain such behaviors are more likely to be seen than others, and that these deficits can be predicted to a degree by previous episodes of loss of consciousness and cyanosis. Other research has suggested that such findings are consistent not only with the assumption of underlying change in cerebral structure or function, but also as an iatrogenic effect of antiasthma drugs over time.
Cystic Fibrosis Cystic fibrosis (CF) is the most commonly seen lethal genetic syndrome of infants, children, and young adults. It is most prevalent in Caucasian youngsters, with one case of CF in every 1500-2000 live births; it is much less common in black and Oriental populations. Inheritance appears to be by an autosomal recessive gene, suggesting a specific biochemical defect, but no single, unifying hypothesis exists at this time to account for the pathogenesis of CF (Matthews & Drotar, 1984). CF is a very complex condition affecting a wide variety of bodily functions. It causes abnormalities in the exocrine gland network, pancreas, liver, gastrointestinal tract, reproductive system, and especially in the respiratory system. Chronic obstructive pulmonary problems seem to account for the majority of morbidity and mortality in the CF patient. In the past, children with CF simply died young. But the introduction of the sweat test in 1954 permitted early diagnosis, and coupled with newer treatment regimens, children with this disease are now surviving fairly well into late adolescence, young adulthood, and in some cases into early mid-life (Taussig & Landau, 1986). A tremendous adjustment to the disease is required because it requires lifetime care, a great deal of medicine, a strict diet, a mechanical apparatus to assist breathing, daily postural drainage and breathing exercises, and living in constant danger of respiratory infections. Individuals with CF are prone to such infections, and their breathing is often altered as a result of
NEUROPSYCHOLOGICAL SEQUELAE OF CHRONIC DISORDERS
increased airway resistance. Some of the issues discussed above with asthmatics regarding decreased cerebral oxygenation and its effect on cortical integrity may be germane here, although symptoms seem to be less severe. CF is not seen as a disease that directly affects the brain. In fact, Breslau (1985) and Breslau and Marshall ( 1985), in studies of psychiatric disorders in children with physical disabilities, used CF subjects as non-brain-involved chronic disease controls. They found that healthy and CF subjects, previously diagnosed as troubled, improved in their mentation and psychological adjustment over a 5-year period, whereas brain-damaged subjects showed no such improvement. However, Matthews and Drotar (1984) suggested that CF children express some of their difficulty in adjusting to this multisystem disease by the development of learning problems in school. As is seen with other chronic diseases, these learning problems may in fact be symptomatic of psychological difficulties, absenteeism, and decreased sensory stimulation; however, they may also be due to mild neurobehavioral deficits. Currently, no research specifically addresses this issue in children with CF.
Cardiovascular System The main functions of the cardiovascular system are to pump blood through the body, to pick up and deliver fluids, gases, chemicals, and nutritive substances, and to increase or decrease blood flow in response to activity levels of the body. The cardiovascular system is composed of the heart, large arteries and veins, smaller arterioles and venules, and the capillaries. The manner of blood flow and its regulation are crucial factors to be considered in discussing cardiovascular functioning and the effects of dysfunction. · Blood traveling through vessels exerts different pressures and moves at different speeds according to the size of the vessel. The cardiovascular system acts to maintain a relatively constant and limited range of pressures and blood flow velocities within the vessels. Any increase in friction, such as occurs with blockages, narrowing, or roughness along the vessel walls, increases the workload of the system and can lead to failure.
CNS Effects Disease or malfunction anywhere in the cardiovascular system tends to initiate a vicious cycle of
119
adjustments that cause the heart to work harder to compensate for these changes, resulting in further damage. A compromised heart eventually leads to a compromised brain. Although the brain will still receive a greater share of materials needed by the body, prolonged cardiac dysfunction will ultimately lower the amount available to brain cells. Insufficient oxygen and nutrients are likely to produce results of diffuse neuropsychological dysfunction in children with cardiovascular disease or irregularities (Cravioto & Arrieta, 1983). The child is likely to have cognitive deficits in a wide variety of functions, although these may be mild unless damage to the heart (or reduction of blood flow) has been severe. Many of the cognitive deficits in individuals with cardiac problems may not even be noticed because of the concern over other more attention-demanding physical symptoms. Mild deficits that are noticeable are often temporary and tend to be viewed with less concern (Ariel & Strider, 1983). When circulation to the heart itself is blocked and tissue is damaged (''heart attack''), there is often an extreme drop in blood pressure. This may produce symptoms of dizziness or massive changes in mental functioning such as delirium or dementia. The lack of oxygen to brain tissue may produce focal deficits such as aphasia, sensorimotor disturbances, or visual difficulties (Rowland, 1984). Such effects can be either temporary or permanent. A cerebral hemorrhage may occur from the increased pressure and destruction of cerebral blood vessels, producing either diffuse or focal effects that tend to be more permanent (Rowland, 1984). Because these deficits are generally more disrupting to the child's ability to function, they are more likely to lead to a concern to the child, parent(s), and/or physician, and are often the symptoms that lead to requests for evaluation. H such diseases progress slowly, then compensation usually occurs and the child may appear to have normal cognitive functioning. This is particularly true for children as the developing brain tends to be somewhat more amenable to recovery of function than the more mature adult brain. It must be cautioned, however, that there is a growing body of literature to suggest that the developing brain may not be as "plastic" as once was thought (Golden, 1981). Another outcome of cardiac disease may be the development of bacterial endocarditis, an infection of cardiac tissue wherein bacteria collect in damaged valves or in the pericardia} sacs. In addition to creating inflammation and edema, bacteria may spread to other parts of the body by means of blood circulation. If the brain is entered, the result is usually a septic embolism (a blockage creating infection in that area),
120
CHAPTER 6
widespread meningitis, or development of a focal brain abscess (Rowland, 1984). Neuropsychological effects can be quite variable, ranging from focal deficits that resemble an adult stroke to a diffuse encephalopathy (widespread inflammation). The variations possible make it a quite difficult condition to accurately diagnose. In such instances, it has been suggested that evaluation be conducted on a follow-up basis to determine if additional damage has occurred or to assess the extent and severity of residual impairment (Ariel & Strider, 1983; Golden, 1981). The development of hemorrhages in the brain from hypertensive destruction of vessels also produces variable effects. Although hypertension in children is comparatively rare, it occurs with enough frequency to merit some discussion. Hemorrhages typically result in focal deficits, but these can be singular or multiple. Mild or severe disruption of cognitive functional systems can result depending on where bleeding occurs (Walton, 1977). Acute hypertensive encephalopathy may produce massive edema and pressure effects leading to severe diffuse deficits, convulsions, decerebrate rigidity, coma, or death from cerebral hemorrhage. Hypotension, or low blood pressure, generally has only mild or unnoticeable effects on brain functioning (Ariel & Strider, 1983); however, it can produce diffuse impairment of moderate to severe degrees as well. Children may complain of amnesia, excessive fatigue, fainting, convulsions, or loss of specific cognitive abilities, all of which indicate that ischemia to brain tissue has likely occurred (Gold, 1984). As such sequelae are variable and often fluctuating, the diagnosis of brain dysfunction or permanent impairment is a difficult one to make. Cardiovascular difficulties also can modify blood constituents, producing brain ischemia because of the alteration in blood flow or inability of erythrocytes to carry oxygen. Many such problems may first be labeled as "psychiatric" or emotional disorders, because the child exhibits depressive symptoms or confusion as the first symptoms (Taylor, 1979). Surgery for cardiovascular problems also carries a certain risk. Circulation of blood to the brain may become impaired while the patient is connected to a heart-lung machine. Thrombosis, embolism, anoxia, or toxic/allergic reactions to anesthesia or injected medications may occur. Infections can develop that spread to the brain, or the heart may simply fail to regain its normal rhythm after surgery. All of these may lead to cognitive deficits of varying degree and location (Ariel & Strider, 1983). Brobeck (1979) reported that EEG tracings re-
vealed more abnormalities after cardiac surgery than prior to the surgical procedure. He further noted that if the patient's EEG does not return to normal within 3 to 4 weeks after surgery, the likelihood of cerebral damage having occurred is quite high. Studies using neuropsychological instruments have found that signs of cerebral dysfunction prior to surgery place the child at even higher risk for the development of later cerebrovascular problems as well as at a higher risk for death during the surgical procedure (Kilpatrick, Miller, Allan, & Lee, 1975). In the adult literature, there have been a number of investigations conducted on patients who have undergone surgery for occlusions or narrowing of the internal carotid arteries. Such patients are often so diagnosed because they experience transient ischemic attacks with such symptoms as dizziness, memory loss or disorientation, mild speech problems, visual changes, or mild sensorimotor deficits. If untreated, a cerebral stroke is likely (Thompson, Patman, & Talkington, 1978). To date, little such work has been conducted with children, and therefore, inferences must be drawn from the adult literature. Those studies that have been done with adults to determine if surgical intervention improves or changes cognitive status have reported mixed findings. Several studies have reported significant improvement (e.g. , Bornstein, Benoit, & Trites, 1981; Goldstein, Kleinknecht, & Gallo, 1970; Owens et al., 1975) whereas a similar number of investigations have found either no improvement or deterioration in function occurs (e.g., Drake, Baker, Blumenkrantz, & Dahlgren. 1968; Murphy & Maccubbin, 1966). One relatively recent study concluded that any reported performance increase was likely a function of test-retest practice effects and not true improvement in functioning (Matarazzo, Matarazzo, Gallo, & Weins, 1979). Despite the large number of studies on adults, the issue still appears to be highly controversial. Children with similar cardiovascular dysfunction, therefore, may suffer from much the same forms of dysfunction; however, there remains the caveat that one is dealing with a developing brain. As the brain appears to develop functional capacity during development, early damage may impede this development. Damage incurred at later ages may result in a loss of established functional capabilities. However, it is generally felt that the prognosis for recovery of cognitive function or reacquisition oflost functions depends primarily on an interactive effect of a number of variables including type, location, and extent of the damage, to name a few (Rourke, Bakker, Fisk, & Strang, 1983).
NEUROPSYCHOLOGICAL SEQUELAE OF CHRONIC DISORDERS
Finally, in cases of cardiovascular system abnormality or ·malfunction that has led to cognitive impairment, personality changes such as depression, irritability, anxiety, and so on may occur with some frequency (Lishman, 1978). These changes often are noticed before actual intellectual deficits appear and may be attributed to the child's inability to adjust to the illness. In some cases, the changes may be the sequelae of damage to brain tissue and this possibility needs to be fully explored.
Lymphatic and Connective Tissue Systems The lymphatic system, often referred to as the immune system, is similar to the blood and cardiovascular systems in its structure, but differs greatly in function. Its primary purposes are to defend the body against invasion by injurious agents, to gather and destroy worn-out cells, and to produce antibodies. It also stores extra red blood cells and produces hormones that help to regulate the development of new red blood cells. The lymph system is composed of the spleen, lymph vessels and nodes, and defensive cells (Rowland, 1984).
Spleen The spleen is the main storage center for new red cells and the destruction center for old ones. It also makes some types of white cells (the lymphocytes). In emergency situations, large numbers of red cells are dumped into the bloodstream to ensure adequate oxygen supply.
Lymph Vessels The lymph system has its own vessel system that drains fluid from the tissue spaces. These vessels form into larger ducts that eventually merge into the blood. Along the vessels are lymph nodes, which act to help prevent large particles or foreign bodies from entering the bloodstream. Lymph cells in the nodes are generally effective in eliminating most foreign bodies with the exception of viruses. Lymph ducts and nodes are found almost everywhere in the body except in the CNS. The lymph capillary system is placed in such a manner so that virtually all materials that enter the body through the skin or mucosa must first pass through the lymphatic system. As the lymph system is in essence a dumping/disposal sys-
121
tern, any disruption of it can disrupt water and electrolyte balance within the body, which in tum can increase pressure on the capillary system, shut down blood flow, and eventually result in death.
Disorders of the Connective Tissue System Connective tissue refers to fibrous tissue that provides support for holding cells together and forms a protective covering around the body and internal organs. Connective tissue cells are found everywhere in the body, but large amounts of them are found in bones and joint tissues. The connective tissue system is composed of ligaments, tendons, cartilage, skin, blood vessels, internal membrane linings, and sheath coverings of organs and muscles. They also constitute a large portion of organs such as the eyes, lungs, heart, kidneys, and liver (American Rheumatism Association Committee, 1973). Disorders of the connective tissue can be inherited or acquired. The genetic maladies are comparatively rare and are beyond the scope of this chapter. Acquired conditions generally include rheumatoid arthritis, systemic lupus erythematosus, progressive systemic sclerosis, polymyositis, dermatomyositis, Sjogren's syndrome, amyloidosis, various form of vasculitis, and rheumatic fever. Although different in terms of severity and the age groups that can be affected, these diseases all display features associated with inflammation and destruction or alteration of connective tissue. Common symptoms include fatigue, fever, muscular weakness, joint swelling and pain, skin lesions, gastrointestinal erosions with hemorrhages, peripheral vascular dysfunctions, neuropathies, and blood cell disorders such as anemia and thrombocytopenia. The course of the illness may vary greatly from individual to individual, with periods of both remission and exacerbation, a chronic mild illness, severe and rapidly progressing deterioration, or fluctuations between mild and severe episodes. Some children with these diseases may become severely disabled due to crippling joint deformities or loss of function in a major organ system, such as the kidneys. Initial symptoms can mimic many other diseases because they are so variable, and thus are difficult to diagnose in many instances (Gilroy & Meyer, 1975).
CNS Effects Comprehensive research on the neuropsychological effects of diseases of lymph and connective tissue systems in children is minimal, at best. The
122
CHAPTER 6
medical literature indicates that the effects on the brain and CNS tend to be variable and generally unpredictable (Gilroy & Meyer, 1975; Rutter, 1983; Walton, 1977). The small vessel inflammation and destruction that can occur in many of these diseases can produce focal ischemic lesions in many organs, causing them to malfunction and reduce their support to the brain. Vessels in the brain may also be affected, although pathological studies have been inconsistent in confirming this with most disease types with the exception of giant cell arteritis, a condition rarely found in children (American Rheumatism Association Committee, 1973). Hypertension is a frequent outcome of these diseases the effects of which were described above. Compression or ischemia may result in peripheral neuropathy, with sensory or motor losses in digits or limbs (Graham, 1983). Diffuse or focal cerebral infections may occur as a result of the suppression of immune responses from drugs taken during treatment. Although evidence of the disease process in the brain itself has not yet been confirmed, studies have indicated the presence of immune complexes associated with connective tissue and lymphatic system disease processes in the choroid plexus of the brain (Atkins, Kondon, Quismorio, & Friou, 1982; Bresnihan et al., 1979; Winfield, Lobo, & Singer, 1978). Psychoses, depression, and mental confusion have frequently been reported as sequelae of these conditions. Reactions to corticosteroids, antihypertensives, antidiuretics, anti-inflammatory agents, and other medication used for treatment have been found to produce changes in emotional or mental state, although it is difficult to separate this from the effects of the disease itself. CNS effects have been most often reported with systemic lupus erythematosus (Ariel & Strider, 1983). The reported sequelae include emotional disorders, convulsions, choreiform movements, and cerebrovascular accidents with focal neurological deficits. These usually occur in children with highly active and severe disease. In the early stages oflupus, the CNS effects may be mild and transient, or may so resemble a psychiatric disorder that the patient is treated as such (Bennett, Bong, & Spargo, 1978; Hughes, 1979).
Gastrointestinal System The gastrointestinal or digestive system extends from the mouth to the anus, and has as its principal
function the provision of the body with fluids, nutrients, and electrolytes. It is lined with secreting cells and glands, and has accessory organs, all of which contribute to this function of providing the body with that which it needs to function. A second function is to dispose of the waste residues from the digestive process (Luckman & Sorenson, 1980). Some of the important parts of this system are the stomach, the small bowel, the colon, the rectum, and the anus. A variety of chronic diseases interfere with the function of the GI tract, yielding psychological as well as physical problems (Whitehead & Bosmajian, 1982; Reinhart, 1982; Raymer, Weininger, & Hamilton, 1984).
Inflammatory Bowel Disease Ulcerative colitis and Crohn's disease are usually considered together as inflammatory bowel disease. The cause of both is unknown, and the clinical symptoms presented are similar, including severe cramps and discharge of mucus and blood. Pediatric presentations often mimic a number of other conditions, including idiopathic growth failure, before a diagnosis is made (Kelts & Grand, 1980). Both diseases are serious and potentially lifethreatening, and often necessitate surgery as well as a variety of other medical and psychological treatments. Further, the risk of cancer of the bowel is greatly increased in such patients. Although inflammatory bowel disease is usually thought of as a syndrome of young adults, cases are found in all age groups. The greatest incidence is in adolescence, although 15% of all cases occur in those below the age of 15, and this proportion seems to be on the rise (Hagenah, Harrigan, & Campbell, 1984). A number of extraintestinal manifestations of irritable bowel disease exist, including disorders of the liver and biliary tract, skin lesions, joint swelling, and later arthritis. But perhaps most important from a neuropsychological point of view is the retardation in growth seen in 15 to 30% of these children. This manifestation often precedes the diagnosis of the disease, and may be independent of severity of symptoms or use of steroids in treatment. Some commonly cited reasons for this growth failure are malabsorption, intestinal protein loss, abnormal hormonal response, anorexia, and undernutrition (Kelts & Grand, 1980). Although the implications of this retarded growth for the brain have not been addressed, it would appear that loss of needed nutrition, particu-
NEUROPSYCHOLOGICAL SEQUELAE OF CHRONIC DISORDERS
123
larly in early childhood, may have a deleterious ef- alysis. It is important to note that there is typically a fect on cerebral development. However, no studies to time lag between dialysis treatment and subsequent date have found specific neurobehavioral problems improvement or deterioration of the child's behavior in children with inflammatory bowel disease. or mental status changes. Performance on tests of visual-motor speed and accuracy was found to be best 24 hours after treatment (Lewis, O'Neil, Dustman, & Beck, 1980). Renal System Despite the clear necessity of the dialysis procedure, it too has its complications. It is now well The kidneys are bean-shaped organs located recognized that an encephalopathy referred to as diagainst the posterior wall of the abdomen. Their prin- alysis disequilibrium syndrome may be a consecipal function is to assist in the maintenance of water quence of the procedure (Marshall, 1979). This synbalance, blood pH level, and electrolyte balance. The drome has a characteristic course involving functional unit of the kidney is the nephron, which is intermittent slowing of speech with stuttering and a mass of filter tubes. Green (1978) estimated there to word-finding difficulties, which tends to develop be approximately I million nephron units in each over a period of 2 to 3 months. Symptoms appear to kidney. be more prominent either during or immediately Blood goes to the kidneys directly from the aorta after dialysis (Ariel & Strider, 1983). These speech by way of the renal arteries. The latter divide into a difficulties progress to problems in the production of net of capillaries shaped as a tiny ball. This ball of sentences, and myoclonus and dyspraxic movecapillaries is called the glomerulus. The blood re- ments occur. Additionally, severe memory loss, turns from the kidneys via the renal vein. Blood flow concentration problems, and, at times, psychoticlike is slowed as it goes through the glomerulus, which behavior can develop in children. Other sympallows water ions and smaller molecules to diffuse tomatology appears toward the end of a dialysis through a "filtration system" of sorts in which un- treatment and may subside over several hours, but wanted material is passed on to the bladder. Other the confusion, when it appears, may persist for sevsubstances are returned to the blood for further use. eral days (Ariel & Strider, 1983). Evidence suggests Kidney damage is referred to as nephrosis. In- that electrolyte imbalance and edema of the brain fections of the kidney system (nephritis) may impair may cause the disequilibrium syndrome (Raskjn & the functioning of the kidneys. When kidney function Fishman, 1976). is impaired and urine production is reduced, the There is also a dialysis dementia that may be waste products typically excreted are retained in the seen in those undergoing long-term dialysis (Alfrey body and can lead to potentially fatal disorders. Gen- et al., 1972; Mahurkar, Dhar, & Salta, 1973). Alerally, when kidney malfunction occurs, dialysis is though noted primarily in adults, dialysis dementia used to either augment or replace the filtration aspect has been found to occur in children. It is characof kidney function. terized first by a disturbance in speech, with facial Severe psychological as well as neuropsycho- grimacing, convulsions, and eventually dementia. logical complications are known to arise from di- The symptoms occur initially during the dialysis proalysis. Depression is common, both as a function of cess and tend to clear after a period of time, but the kidney dysfunction as well as the stress of the eventually the remission no longer occurs and the procedure on the patient. The patient may then be syndrome continues on its course. To date, no sucunable or unwilling to follow the necessary medical cessful treatment has been found (Ariel & Strider, regimen, which can further exacerbate the disease. 1983). Other findings, however, show that symptoms and Subdural hematomas have been reported in dibehavior can be related to interference with CNS alysis patients (Talalla, Halbrook, Barbour, & functioning (Marshall, 1979). Kurze, 1970). Two explanations have been offered It has been well documented that uremia results for this. Many patients undergoing dialysis are given in severely impaired cognitive functioning (Ariel & anticoagulant drugs so that the shunts implanted in Strider, 1983). Characteristic symptoms such as the arm can remain free from thrombosis. Addisluggish mentation, lethargy, anorexia, nausea and tionally, patients with kidney failure are subject to vomiting, tremor, sleepiness, or convulsions have abnormal bleeding. The symptoms may look similar been widely reported (Ginn, 1975; Raskin & Fish- to dialysis equilibrium and are often difficult to difman, 1"976). These symptoms are relieved by di- ferentiate clinically. However, worsening or per-
124
CHAPTER 6
sistence of symptoms between dialysis treatments may signal the presence of a hematoma as opposed to the disequilibrium syndrome. Evaluation of neuropsychological functioning in such children can aid in documenting functional level. Assessment of mental status and neuropsychological functioning in children with uremia can provide useful diagnostic treatment recommendations. Obtaining good baseline measures early in the diagnosis of renal disease will allow for more valid assessment of any later occurring deterioration as well as response to dialysis treatment when initiated (Ariel & Strider, 1983). The timing of the assessment of dialysis patients must be considered as the cognitive functioning of individuals tends to fluctuate greatly at periods in the dialysis regimen, yielding different performance patterns. Ryan, Souheaver, and DeWolf (1981) reported comparative neuropsychological assessments on chronic hemodialysis patients, undialyzed uremic patients, and medically-psychiatrically ill patients. Significant differences between the medical-psychiatric and the other groups were noted. Dialysis patients performed better than undialyzed patients on some tasks but were impaired relative to the medicalpsychiatric group. These findings indicate that chronic dialysis patients cannot be considered to have normal neuropsychological functioning.
References Ack, M., Miller, 1., & Weil, W. B. (1%1). Intelligence of children with diabetes mellitus. Pediatrics, 28, 764-770. Aita, J. A. (1964). Neurologic manifestations ofgeneral diseases. Springfield, IL: Thomas. Alfrey, A. C., Mishell, J. M., Burks, J., Contigaglia, R. H., Lewin, E., & Holmes, J. H. (1972). Syndrome of dyspraxia and multifocal seizures associated with chronic hemodialysis. Transactions, American Society for Artificial Internal Organs, /8, 257-261. American Rheumatism Association Committee. (1973). Primer on the rheumatic diseases (7th ed.). Atlanta: Arthritis Foundation. Ariel, R., & Strider, M.A. (1983). Neuropsychological effects of general medical disorders. In C. J. Golden & P. J. Vincente (Eds.), Foundations of clinical neuropsychology (pp. 273308). New York: Plenum Press. Arseni, C., & Goldenberg, M. (1959). Psychic disturbances in infiltration of the brain stem. ActaNeurochirurgica, 7. 292300. Atkins, C. J., Kondon, J. J., Quismorio, F. P., & Friou, G. J. ( 1972). The choroid plexus in systemic lupus erythematosus. Annals of Internal Medicine, 76, 65-72. Bennett, R. M., Bong, D. M., & Spargo. B. H. (1978). Neuropsy-
chiatric problems in mixed connective tissue disease. American Journal of Medicine, 65, 955-962. Berg, R. A., Ch'ien, L. T., Bowman, W. P., Ochs,J., Lancaster, W., Goff, J. R., & Anderson, H. R., Jr. (1983). The neuropsychological effects of acute lymphocytic leukemia and its treatment-A three year report: Intellectual functioning and academic achievement. International Journal of Clinical Neuropsychology, 5, 9-13. Berg, R. A., & Wilimas, J. J. (1983. May). Sickle cell disease and neuropsychological function. Paper presented at the MidSouth Conference on Human Neuropsychology, Memphis, Tennessee. Bierman, C., Pierson, W., Shapiro, G., & Simons, E. (1975). Brain damage from asthma in children. Journal of Allergy and Clinical Immunology, 55, 126. Black, F~ W. ( 1973). Intellectual ability as related to age and stage of disease in muscular dystrophy: A brief note. Journal of Psychology, 84, 333-334. Bloom, H. J. G., Wallace, E. N. K., & Henk, J. M. (1969). The treatment and prognosis of medulloblastoma in childhood. American Journal of Roentgenology, /05, 43-62. Boll, T. J. (1974). Behavioral correlates of cerebral damage in children aged 9 through 14. In R. M. Reitan & L.A. Davidson (Eds.), Clinical neuropsychology: Current status and applications. Washington, DC: Winston. Bomstein, R. A., Benoit, B. G., & Trites, R. L. (1981). Neuropsychological changes following carotid endarterectomy. Canadian Journal of Neurological Science, 8, 127-132. Bowman, W. P. (1981). Childhood acute lymphocytic leukemia: Progress and problems in treatment. CMA Journal, 124, 129-142. Breslau, N. (1985). Psychiatric disorder in children with physical disabilities. Journal of the American Academy of Child Psychiatry, 24, 87-94. Breslau, N., & Marshall, I. A. (1985). Psychological disturbance in children with physical disabilities: Continuity and change in a 5-year follow-up. Journal of Abnormal Child Psychology, 13, 199-216. Bresnihan, B., Hohmeister, R., Cutting, J., Travers, R. L., Waldburger, M., Blacj, C., Jones, T., & Hughes, G. R. (1979). The neuropsychiatric disorder in systemic lupus erythematosus: Evidence for both vascular and immune mechanisms. Annals of Rheumatic Diseases. 38. 301-306. Brobeck, J. R. (Ed.). (1979). Best and Taylor's physiological basis of medical practice (lOth ed.). Baltimore: Williams & Wilkins. Cairns, H. (1950). Mental disorders with tumours of the pons. Folia Psychiatrica. Neuro/ogica. Neurochirurgica Neer/andica (English abstract), 53. 193-203. Carter Center. ( 1985). Closing the gap: The problem of diabetes mellitus in the United States. Diabetes Care. 8, 391-406. Cerreto, M. C., & Travis, L. B. (1984). Implications of psychological and family factors in the treatment of diabetes. Pediatric Clinics of North America. 31. 689-710. Chai, H. ( 1975). Management of severe chronic perennial asthma in children. Advances in Asthma and Allergy. 2. 1-12. Cirillo, D., Gonfiantini, E., DeGrandis, D., Bongiovanni, L., Robert, J., & Pinelli, L. (1984). Visual evoked potentials in diabetic children and adolescent~. Diabetes Care, 7. 273-275.
NEUROPSYCHOLOGICAL SEQUELAE OF CHRONIC DISORDERS Cohen, H. J., Molnar, G. E., & Taft, L. T. (1968). The genetic relationship of progressive muscular dystrophy (Duchenne type) and mental retardation. Developmental Medicine and Child Neurology, 10, 754-765. Copeland, D. R., Pfefferbaum, B., Fletcher, J., Jaffee, N., & Culbert, S. ( 1982, April). Neuropsychological assessment of long-term survivors of leukemia. P:·per presented at the American Society of Clinical Oncoloe '', St. Louis. Cravioto, J., & Arrieta, R. (1983). Malnutrition in childhood. In M. Rutter (Ed.), Developmental neuropsychiatry, New York: Guilford Press. Creer, T. L. (I 982). Asthma. Journal of Consulting and Clinical Psychology, 50, 912-921. Daneman, G. M., Johnson, L. E., & Garfinkel, P. E. (1985, May). Is diabetes a precipitant of eating pathology? Paper presented at the meeting of the American Pediatric Society, Washington. DC. Drake, W., Baker, M., Blumenkrantz, J., & Dahlgren, H. (1968). The quality and duration of survival in bilateral carotid occlusive disease: A preliminary survey of the effects of thromboendarterectomy. In J. Toole, R. Siekert, & J. Whisnant (Eds.), Cerebral vascular disease. New York: Grune & Stratton. Dubowitz, V. (1977). Mental retardation in Duchenne muscular dystrophy. In L. P. Rowland (Ed.), Pathogenesis of human muscular dystrophies. Amsterdam: Excerpta Medica. Dubowitz, V. ( 1979). Involvement of the nervous system in muscular dystrophies in man. New York: New York Academy of Sciences. Dunleavy, R. A. (1981). Neuropsychological correlates of asthma: Effect of hypoxia or drugs? Journal of Consulting and Clinical Psychology. 49, 137. Dunleavy, R. A., & Baade, L. E. (1980). Neuropsychological correlates of severe asthma in children 9-14 years old. Journal of Consulting and Clinical Psychology, 48, 214-219. Ebaugh, F. (1923). Neuropsychiatric sequelae of acute epidemic encephalitis in children. American Journal of Diseases of Children. 25, 89-97. Eeg-Oiofsson, 0., & Peterson, I. (1966). Childhood diabetic neuropathy: A clinical and neuropsychological study. Acta Paediatrica Scandinavica, 55, 163-176. Eiser, C. (1978). Intellectual abilities among survivors of childhood leukemia as a function of CNS irradiation. Archives of Disease in Childhood. 53, 391-395. Eiser, C. (1979). Intellectual development following treatment for childhood leukaemia. In J. M.A. Whitehouse & H. E. M. Kay (Eds.), CNS complications of malignant disease. New York: Macmillan Co. Eiser, C., & Lansdown, R. (1977). Retrospective study otintellectual development in children with acute lymphoblastic leukemia. Archives of Disease in Childhood, 52, 525-529. Fee, W. E., Mariss, M. 1., Kardash, S., Reite, M., & Seitz, C. (1970). The long-term prognosis of aseptic meningitis in childhood. Developmental Medicine and Child Neurology, 12. 321-329. Florek, M., & Karolak, S. (1977). Intelligence level of patients with Duchenne type of progressive muscular dystrophy (PMD-D). European Journal of Pediatrics, 126. 275-282. Franceschi, M., Cecchetto, R., Minicucci, F., Smizne, S., Baio,
125
G., & Canal, N. (1984). Cognitive processes in insulin-dependent diabetes. Diabetes Care, 7, 226-231. Freedman, A.M., Kaplan, H. I., & Sadock, B. J. (1976). Modern synopsis of comprehensive textbook of psychiatry (Vol. 2). Baltimore: Williams & Wilkins. Gilroy, J., & Meyer, J. S. (1975). Medical neurology (2nd ed.). New York: Macmillan Co. Ginn, H. E. (1975). Neurobehavioral dysfunction in uremia. Kidney International, 7, 217-221. Goff, J. R., Anderson, H. R., Jr., & Cooper, P. F. (1980). Distractibility and memory deficits in long-term survivors of acute lymphoblastic leukemia. Developmental and Behavioral Pediatrics, 1. 158-163. Gold, A. M. (1984). Stroke in children. In L. P. Rowland (Ed.), Merritt's textbook of neurology (7th ed.). Philadelphia: Lea & Febiger. Golden, C. J. (1981). Diagnasis and rehabilitation in clinical neuropsychology (2nd ed.). Springfield, IL: Thomas. Goldstein, S. G., Kleinknecht, R. A., & Gallo, A. E., Jr. (1970). Neuropsychological changes associated with carotid endarterectomy. Cortex, 6, 308-322. Gortmaker, S. L., & Sappenfield, W. (1984). Chronic childhood disorders: Prevalence and impact. Pediatric Clinics ofNorth America, 31, 3-16. Graham, P. J. (1983). Specific medical syndromes. In M. Rutter (Ed.), Developmental neuropsychiatry, New York: Guilford Press. Green, J. H. (1978). Basic clinical physiology. New York: Oxford University Press. Hagenah, G. C., Harrigan, J. F., & Campbell, M. (1984). Inflammatory bowel disease in children. Nursing Clinics of North America, 19, 27-39. Haumont, D., Dorchy, H., & Pelc, S. (1979). EEG abnormalities in diabetic children: Influence of hypoglycemia and vascular complications. Clinical Pediatrics, 18, 750-753. Heron, J. R., Hutchinson, E. C., Boyd, W. N., & Aber, G. M. (1974). Pregnancy, subarachnoid hemorrhage, and the intravascular coagulation syndrome. Journal ofNeurology, Neurosurgery, and Psychiatry, 37, 521-527. Holmes, C. S. (1986). Neuropsychological profiles in men with insulin-dependent diabetes. Journal of Consulting and Clinical Psychology, 54, 386-389. Holmes, C. S., Hayford, J. T., Gonzalez, J. L., & Weydert,J. A. ( 1983). A survey of cognitive functioning at different glucose levels in diabetic persons. Diabetes Care, 6, 180-185. Holmes, C. S., Koepke, K., & Thompson, R. G. (1986). Simple versus complex performance impairments at three blood glucose levels. Psychoneuroendocrinology, 11, 343357. Holmes, C. S., Koepke,K-M., Thompson,R.G.,Gyves,P . W., & Weydert, J. A. (1984). Verbal fluency and naming performance in type I diabetics at different blood glucose concentrations. Diabetes Care, 7. 454-459. Hope-Stone, H. F. (1970). Results of treatment of medulloblastomas. Journal of Neurosurgery, 32, 83-88. Hughes, G. V. R. (1979). Connective tissue diseases (2nd ed.). Oxford: Blackwell. Ivnik, R. H., Colligan, R. C., Obetz, S. W., & Smithson, W. A. (1981). Neuropsychologic performance among children in
126
CHAPTER 6
remission from acute lymphocytic leukemia. Developmental and Behavioral Pediatrics, 2, 29-34. Johnson, S. B., & Rosenbloom, A. L. (1982). Behavioral aspects of diabetes mellitus in childhood and adolescence. Psychiatric Clinics of Nonh America, 5, 357-369. Karagan, N.J. (1979). Intellectual functioning in Duchenne muscular dystrophy: A review. Psychological Bulletin, 86, 250259. Kelts, D. G., & Grand, R. J. (1980). Inflammatory bowel disease in children and adolescents. Current Problems in Pediatrics, 10, 1-40. Kennedy, R. (1924). Prognosis of sequelae of epidemic encephalitis in children. American Journal of Diseases of Children, 29, 158-172. Kilpatrick, D. G., Miller, W. C., Allan, A. N., & Lee, W. H. (1975). The use of psychological test data to predict openheart surgery outcome: A prospective study. Psychosomatic Medicine, 37, 62-73. King, N. J. (1980). The behavioral management of asthma and asthma-related problems in children: A critical review of the literature. Journal of Behavioral Medicine, 3, 169-189. Lancet Editorial. (1980). Behavioral disorders in diabetic children. Lancet, 2, 188-189. Lassman, L., & Arjona, V. E. (1%7). Pontine gliomas of childhood. Lancet, 1, 913-915. Lawson, D., Metcalfe, M., &Pampiglione, G. (1965). Meningitis in childhood. British Medical Journal, I, 557-562. Levy, S. (1959). Post-encephalitic behavior disorder-a forgotten entity: A report of 100 cases. American Journal of Psychiatry, 115, 1062-1067. Lewis, E. G., O'Neil, W. M., Dustman, R. E., & Beck. E. C. (1980). Temporal effects of hemodialysis on measures of neural efficiency. Kidney International, 17, 357-363. Lishman, W. A. (1978). Organic psychiatry: The psychological consequences of cerebral disorder. Oxford: Blackwell. LODAT, Handbook and Information Committee. (1981). Living one day at a time. Milwaukee: American Cancer Society. Luckman, J., & Sorenwn, K. C. (1980). Medical-Surgical Nursing. Philadelphia: Saunders. Mahurkar, S. D., Dhar, S. K., & Salta, R. (1973). Dialysis dementia. Lancet, 1, 1412-1415. Marshall, J. (1979). Neuropsychiatric aspects of renal failure. The Journal of Clinical Psychiatry, 40, 81-85. Matarazzo, R. G., Matarazzo, J.D., Gallo, A. E., Jr., & Weins, A. N. (1979). IQ and neuropsychological changes following carotid endarterectomy. Journal of Clinical Neuropsychology, 1, 97-116. Matoth, Y., Zaizov, R., & Frankel, J. J. (1971). Minimal cerebral dysfunction in children with chronic thrombocytopenia. Pediatrics, 47, 698-706. Matson, D. D., & Crigler, J. F. (l%9). Management of craniopharyngioma in childhood. Journal of Neurosurgery, 30, 377-390. Matthews, L. W., & Drotar, D. (1984). Cystic fibrosis-A challenging long-term chronic disease. Pediatric Clinics ofNorth America, 31, 133-152. Mauer, A. M. ( 1980). Therapy ofacute lymphoblastic leukemia in childhood. Blood, 56, l-10. Meadows, A. T., Gordon, J., Massari, D. J., Littman, P., Fer-
guson, J., & Moss, K. (1981). Declines in IQ scores and cognitive dysfunctions in children with acute lymphocytic leukemia treated with cranial irradiation. Lancet, 2, 10151018. Miller, J. D., & Sperling, M. A. (1986). Diabetes mellitus in children. In V. C. Kelley (Ed.), Practice ofPediatrics. 6. 119. New York: Harper and Row. Murphy, F., & Maccubbin, D. A. (1966). Carotid endarterectomy: A long-term follow-up study: In J. Shillito (Ed.), Clinical neurosurgery (Vol. 13). Baltimore: Williams & Wilkins. Obetz, S. W., lvnik, R. J., Smithson, W. A., Colligan, R. C., Groover, R. V., Gilchrist, G. J., Houser, D. W., Burgert, F. 0., & Klass, D. V. (1979). Neuropsychologic follow-up study of children with acute lymphocytic leukemia: A preliminary report. The American Journal of Pediatric Hematology/Oncology, 1, 207-213. Owens, M., Pressman, M., Edwards, A. E., Tourtellotte, W., Rose. J. G., Stern, D., Peters, G., Stabile, B. E., & Wilson, S. E. (1975). The effect of small infarcts and carotid endarterectomy on post-operative psychological test performance. Journal of Surgical Research, 28, 209-216. Panitch, H. S., & Berg, B. 0. (1970). Brain stem tumors of childhood and adolescence. American Journal ofDiseases of Children, 119, 465-472. Pfeifer, M. A., Weinberg, C. R., Cook, D. L., Reenan, A., Halter,J. B., Ensinck,J. W., &Porte, D. (1984). Autonomic neural dysfunction in recently diagnosed diabetic subjects. Diabetes Care, 7, 447-453. Raskin, N.H., & Fishman, R. A. (1976). Neurologic disorders in renal failure. The New England Journal of Medicine, 294, 294-210. Raymer, D., Weininger, 0., & Hamilton, J. R. (1984). Psychological problems in children with abdominal pain. Lancet, /, 439-440. Reinhart, J. B. (1982). Disorders of the gastrointestinal tract in children: Consultation-liaison experience. Psychiatric Clinics of North America, 5, 387-397. Rourke, B. P., Bakker, D. J., Fisk, J. L., & Strang, J.D. (1983). Child neuropsychology: An introduction to theory, research, and clinical practice. New York: Guilford Press. Rovet, J. F., Ehrlich, R. M., & Hoppe, M.G. (1988). Specific intellectual deficits in children associated with early onset insulin-dependent diabetes, Child Development, 59, 226234. Rowland, L. P. (1984). Signs and symptoms in neurological diagnosis. In L. P. Rowland (Ed.), Merritt's textbook of neurology (7th ed.). Philadelphia: Lea & Febiger. Rubin, D. H., Leventhal, J. M., Sadock, R. T., Letovsky, E., Schottland, P., Clementine, I., & McCarthy, P. ( 1986). Educational intervention by computer on childhood asthma. Pediatrics, 77, 1-10. Rutter, M. (1983). Issues and prospects in developmental neuropsychiatry. In M. Rutter (Ed.), Developmental neuropsychiatry. New York: Guilford Press. Ryan, C., Longstreet, C., & Morrow, L. (1985). The effects of diabetes mellitus on the school attendance and school achievement of adolescents. Child: Care, Health and Development, 11, 229-240.
NEUROPSYCHOLOGICAL SEQUELAE OF CHRONIC DISORDERS Ryan, C., & Morrow, L. (1987a). Neuropsychological characteristics of children with diabetes. In M. L. Wolraich & D. K. Routh (Eds.), Advances in Developmental and Behavioral Pediatrics (Vol. VIII). Greenwich, cr: JAI Press. Ryan, C., & Morrow, L. A. (1987b). Self-esteem in diabetic adolescents: Relationship between age at onset and gender. Journal of Consulting and Clinical Psychology, 54, 730731. Ryan, C., Vega, A., & Drash, A. (1985). Cognitive deficits in adolescents who developed diabetes early in life. Pediatrics, 75, 921-927. Ryan, C., Vega, A., Longstreet, C., & Drash, A. (1984). Neuropsychological changes in adolescents with insulin dependent diabetes. Journal ofConsulting and Clinical Psychology, 52, 335-342. Ryan, J. J., Souheaver, G. T., & DeWolf, A. S. (1981). Halstead-Reitan test results in chronic hemodialysis. Journal of Nervous and Mental Disease, 169, 311-314. Sabatino, D., & Cramblett, H. (1968). Behavioural sequelae of Californian encephalitis virus infection in children. Developmental Medicine and Child Neurology, 10, 331-337. Sadler, J. E. (1982). Childhood asthma from the point of view of the liaison child psychiatrist. Psychiatric Clinics of North America, 5, 333-343. Skenazy, J. A., & Bigler, E. D. (1985). Psychological adjustment and neuropsychological performance in diabetic patients. Journal of Clinical Psychology, 41, 391-396. Sollee, N.D., Latham, E. E., Kindlon, D. J., & Bresnan, M. J. (1985). Neuropsychological impairment in Duchenne muscular dystrophy. Journal of Clinical and Experimental Neuropsychology, 7, 486-496. Strecker, E. (1929). Behaviour problems in encephalitis. Archives of Neurology and Psychiatry, 21, 137-144. Suess, W. M., & Chai, H. (1981). Neuropsychological correlates of asthma: Brain damage or drug effects? Journal ofConsulting and Clinical Psychology, 49, 135-136.
127
Talalla, A., Halbrook, H., Barbour, B. H., & Kurze, T. (1970). Subdural hematoma associated with long-term hemodialysis for chronic renal disease. Journal of the American Medical Association, 212, 1847-1849. Taussig, L. M., & Landau, L.l. (1986). Cystic fibrosis. In V. C. Kelley (Ed.), Practice of Pediatrics, 2, 1-33, New York: Harper and Row. Taylor,- I. W. (1979). Mental symptoms and electrolyte imbalance. Australian and New Zealand Journal ofPsychiatry, 13, 159-160. Temand, C., Go, V. L. W., Gerich, J. E., & Haymond, M. W. (1982). Endocrine pancreatic response of children with onset of insulin-requiring diabetes before age 3 and after age 5. Journal of Pediatrics, 101, 36-39. Tew, B. (1979). The "cocktail party syndrome" in children with hydrocephalus and spina bifida. British Journal of Disorders in Communication, 14, 89-101. Thompson, J. E., Patman, R. D., & Talkington, C. M. (1978). Carotid surgery for cerebrovascular insufficiency. Current
Problems in Surgery, 15, 1-68.
Till, K. (1975). Pediatric neurosurgery. Oxford: Blackwell. Walton, J. N. (1977). Brain's diseases of the nervous system (8th ed.). New York: Oxford University Press. Weitzman, M. (1984). School and peer relations. Pediatric Clinics of North America, 31, 59-69. Whitehead, W. E., & Bosmajian, L. S. (1982). Behavioral medicine approaches to gastrointestinal disorders. Journal ofConsulting and Clinical Psychology, 50, 972-983. Winfield, J. B., Lobo, P.l., & Singer, A. (1978). Significance of anti-lymphocyte antibodies in systemic lupus erythematosus. Anhritis and Rheumatism, 21, 215-216. Winter, R. J. (1982). Special problems of the child with diabetes. Comprehensive Therapy, 8, 7-13. Worden, D. K., & Vignos, P. J. (1962). Intellectual function in childhood muscular dystrophy. Pediatrics, 29. 968-9n.
7 Neuropsychological Bases of Common Learning and Behavior Problems in Children MARION J. SELZ AND SHERYL L. WILSON
One of the few consistencies found in the literature on learning disabilities is that there is little agreement about the meaning, behavioral correlates, or etiology of this term. There is even dispute over whether it is an homogeneous or a heterogeneous diagnostic entity (e.g., Fisk & Rourke, 1983; Goldman, Thibert, & Rourke, 1979; Pirozzola & Campanella, 1981; Satterfield, Cantwell, & Satterfield, 1974). Within the literature, "learning disability" has often been used synonymously with hyperactivity, attention deficit disorder, minimal brain dysfunction, and minimal brain damage, as well as specific learning disorders. This has further supported the idea that there is a single phenomenon that manifests itself in these various forms. The tendency toward a unitary view of learning disabilities is most likely the result of almost a century of research that has been primarily confined to developmental reading disorders (Pirozzola & Campanella, 1981). More recently, increasing attention has been given to the possibility that these separate terms may be describing subgroups subsumed under this general clinical category (Pirozzola, 1979; Rourke, 1975; Satz & Morris, 1981). An early attempt to classify the diverse symptoms found in the general category "learning disability" was proposed by Peters, Romine, and Dyckman (1975). They simply divided the entire range of childhood disturbances encompassed by the
MARION J. SELZ • Rehabilitation Psychology, St. Mary's Hospital, Tucson, Arizona 85703. SHERYL L. WILSON • Department of Psychology, University of Arizona, Tucson, Arizona 85721.
general term into learning disabilities and hyperkinesis (or hyperactivity). Research and clinical observation have found that some children are affected by the so-called pure form of learning disability, whereas other children are affected by the so-called pure form of hyperactivity. But there are some children who are affected by both disorders and still others who have specific deficits that do not appropriately fit within a general label of "learning disability.'' In an attempt to account for the diversity within both hyperactivity and learning disability, Denckla ( 1979) devised a clinical inferential system that identifies ten common syndromes. Perhaps the most current and widely used classification system is the American Psychiatric Association's third edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-III) (1980). Within this system, learning disabilities are presented as specific, but not exclusive, developmental disorders, whereas hyperactivity is subsumed under a separate category, Attention Deficit Disorder (ADD). The relationship between neuropsychological deficits and learning disabilities will be reviewed in this chapter. First, the early models and research that initially demonstrated the neuropsychological basis of learning disabilities will be briefly discussed. Second, the view that the concept of "learning disability" is comprised of various subgroups that can be differentiated on the basis of specific neuropsychological deficits will be presented. Third, behavioral disorders, emphasizing ADD, will be discussed, within the framework of Luria's functional systems of the brain model.
129
130
CHAPTER 7
Reitan's Theoretical Model and Original Work Interest in brain-behavior relationships, from both experimental and clinical points of view, is not a recent development in psychology. However, formal assessment techniques concerning evaluation of the status of brain functions in individual persons have a rather recent history, dating back only approximately 50 years. Perhaps the best known American pioneer in the study of human brain-behavior relationships is Ward Halstead (1908-1969). Based on naturalistic observation, Halstead developed a battery of behavioral indicators designed to differentiate individuals with cerebral damage from those without such damage (Halstead, 1947). He recognized that brain damage in its various forms can affect sensory, central processing, or motor functions, and so included measures of these three stages of the behavioral cycle in his original battery. Ralph Reitan, who began his neuropsychological research in Halstead's laboratory, went on to develop a research approach oriented essentially toward systematic subdivision of the term brain damage (Reitan, 1975). He understood the potential contribution of a standardized battery approach, which served as the basis of his research. Reitan's first study utilizing this approach compared the results obtained with Halstead's tests by a group of 50 subjects with documented cerebral damage or dysfunction with a group of 50 subjects who showed no past or present signs or symptoms of cerebral damage or dysfunction. The results showed "striking" differences between the two groups (Reitan, 1955). Reitan (1955) concluded that "Halstead's battery was sufficiently sensitive to the effects of organic brain damage to provide an objective and quantitative basis for detailed study of relationships between brain function and behavior'' (p. 8). Reitan's work in this area resulted in a modification and expansion of Halstead's original battery, which is now known as the Halstead-Reitan Neuropsychological Test Battery. This is appropriate for adults and adolescents 15 years and older. During the 1960s, Reitan and his colleagues at the Indiana University Medical Center expanded their work to include children (Reed, Reitan, & Kl~ve, 1965; Reitan, 1964; Reitan & KI0ve, 1968). This resulted in the development of two batteries for children: the Halstead Neuropsychological Test Battery for Children (ages 9-14) and the Reitan-Indiana Neuropsychological Test Battery for Children (ages 5-8). The adult battery as well as the battery for older children
(9-14) was based on seven of Halstead's original ten measures (three were not retained because Reitan's 1955 validational research found that these tests did not reliably discriminate between brain-damaged and control subjects). Other tests were added or new ones developed (particularly for the younger children's battery), in accordance with Reitan's conceptual framework (Reitan, 1980). According to this framework, a comprehensive neuropsychological test battery must meet three criteria: first, the battery must be comprehensive in that the full range of abilities subserved by the brain must be measured; second, the battery must include tests that are sensitive to general or overall brain functions as well as tests capable of detecting deficits in particular parts of the brain; and third, the tests must be technically designed to be amenable to valid neuropsychological inference and interpretation based on the four methods of inference, as described by Reitan (1967), Reitan and Davison (1974), and Reitan and Heineman (1968). In regard to meeting the first two criteria, the batteries developed by Reitan currently contain measures of abstraction and concept formation, visualspatial and spatial-kinesthetic problem solving, attention and discrimination, memory, language skills, and sensory-perceptual and motor skills. By these measures, the broad range of abilities as well as specific areas responsible for particular functions are thought to be tapped. The third criterion, application of the four methods of inference, is perhaps one of Reitan's most important contributions to the field of neuropsychology because it provides a means for valid interpretation of the test batteries. (For a detailed description of the four methods of inference, as well as their application to the neuropsychological study of learning disabilities, see Selz, 1981.)
Early Developmental Neuropsychological Studies In the refinement and validational research, as mentioned previously, Reitan initially demonstrated that the neuropsychological test battery was able to differentiate between adult subjects with and without brain lesions (Reitan, 1955). Subsequent studies related test findings using the adult battery to such conditions as site of lesion, causal effects, age of onset, premorbid condition, and severity and extent of lesion (Doehring & Reitan, 1962; Reitan, 1958; Reitan & Davison, 1974; Reitan & Fitzhugh, 1971). Some of the issues addressed by these early adult
BASES OF COMMON LEARNING AND BEHAVIOR PROBLEMS
studies, for example, whether the neuropsychological approach could differentiate between normal subjects and brain-damaged subjects, could also be applied to children, but questions unique to children began to emerge. The first such issues were: Does cerebral damage to developing abilities in children have effects similar to damage to acquired abilities in adults? And, are there neuropsychological deficits in children who have negative neurological findings, but whose behavior is in some ways similar to that of children with structural brain damage? The research pertinent to these issues will be briefly reviewed, as the answers to these initial questions were necessary before the more sophisticated questions pertaining to learning disabilities could be formulated. In 1965, Reed et al. matched 50 brain-damaged children between the ages of I 0 and 14 years with 50 normally functioning controls and compared the performance of each pair on 27 measures from the Wechsler-Bellevue Scale and the neuropsychological battery. This analysis found that the braindamaged subjects were most frequently impaired in language functions. Because brain damage in adults tends to markedly impair adaptive and problem-solving abilities rather than language skills, these initial results suggested that brain damage sustained in childhood may have a different effect than brain damage sustained in adulthood. In a replication and extension of this study, Boll (1974) found comparable results, supporting the validity of the Wechsler scales and neuropsychological measures as tests for brain damage in older children. Boll and Reitan ( 1970) found significant differences on the performance of motor and sensoryperceptual tasks when they compared a sample of brain-damaged children with a control sample. Specifically, one measure of sensory-perceptual function (Tactile Finger Localization) and three measures of motor functions (Finger Tapping Test, Tactual Performance Test, and Grip Strength Test) were able to differentiate between the two groups. In regard to the younger children's battery, Reitan ( 1964) compared 29 brain-damaged children between 5 and 8 years of age with 29 controls on the Wechsler Intelligence Scale for Children (WISC) and neuropsychological measures. He found significant differences between the groups on all but one of 41 measures (the insignificant comparison involved grip strength using the dominant hand). Consistent with the results found by Reed eta/. ( 1965), the measures of verbal functions frequently showed the largest group differences. In an earlier study utilizing the same groups of subjects, Reitan (1971) found that motor and sensory-perceptual measures alone were
131
able to classify subjects according to group membership with 70-80% accuracy. These studies demonstrate that groups of braindamaged children consistently achieve lower scores than age-matched controls on the neuropsychological and Wechsler measures. Significant group differences, however, do not automatically translate into clinical applications. For example, the fact that three motor tasks differentiated normal from braindamaged children (Boll & Reitan, 1970) does not suggest that these measures would comprise an adequate brain-damage test. For an accurate and comprehensive diagnosis of a child's problems, the full battery and utilization of the four methods of inference are necessary. The question concerning the possible difference between the effects of brain damage on developing abilities in children and those to acquire abilities in adults was initially raised by the results of the Reed et al. (1965) study. Several subsequent studies attempted to further explore this issue. In 1965, Fitzhugh and Fitzhugh compared 30 pairs of subjects between 15 and 29 years of age on 22 measures from the Wechsler scales and from the neuropsychological battery for adults. In one group, all subjects had sustained brain damage before age 10; in the other group, all subjects had sustained brain damage after age 12. The latter group achieved higher mean scores on all 22 measures. The differences between the groups were significant on several Wechsler performance scales, the Traii Making Test, Part A, and TPT-Memory. These results suggest that greater impairment of abilities occurs with early onset than with late onset of brain damage. This indicates that the longer a person has a normal brain, the more time there is to acquire information and skills. Further support for these conclusions was provided in 1966 by Reed and Fitzhugh in a study comparing four groups of brain-damaged patients with age-matched controls. The brain-damaged groups consisted of older children with mild brain damage; older children with moderate brain damage; adults with recent, mild brain damage; and adults with longstanding moderate brain damage. Results indicated that the controls performed at a higher level than the brain-damaged groups, and that the mildly impaired groups were superior to the moderately impaired groups. The most important aspect of the findings, however, related to the patterns of deficits within the groups. The two groups of brain-damaged children differed from the controls most strikingly on measures of language and symbolic skills. The differences between the normal and brain-damaged
132
CHAPTER 7
children were smaller on the Tactual Performance Test, Finger Tapping, and Category Test. Similarly the moderately impaired adults showed their greatest impairment on language-related tests. But the mildly impaired adults had a different pattern. They differed from their control groups primarily on measures of immediate problem-solving ability such as the Category Test, Tactual Performance Test-Time, and Wechsler Block Design. The results of this study are somewhat difficult to interpret, as severity of cerebral impairment was confounded with age of onset in the adult samples. However, the consistency of findings across several studies suggests that brain damage in young children impairs their ability to acquire knowledge in the manner that knowledge is presented in the schools, and thus the skills emphasized in the school years-language and symbolic skill-show some of the greatest deficits. In contrast, if brain damage occurs in adulthood after a normal childhood and adolescence, language and acquired knowledge are usually spared (with the exception of cases of damage to the language areas in the brain) and deficits appear primarily in adaptive and problem-solving ability. Once the validation studies began providing evidence that the neuropsychological analysis could successfully differentiate between brain-damaged children and normal children, researchers began applying this method to other childhood disorders. One area that seemed particularly appropriate was that of learning disabilities. One of the first studies designed to test this method of analysis was conducted by Doehring in 1968. He compared 39 boys between the ages of 10 and 14 who were retarded in reading but otherwise normal, with a control group of 39 boys with normal reading skills. Doehring found that the normal readers were significantly superior to the retarded readers on 62 out of 103 measures, and that the pattern of deficit included visual and verbal impairment. The retarded readers were superior to the normal readers on several measures using somesthetic input. Two experienced neuropsychologists gave subjects blind ratings of ''no cerebral dysfunction'' or ''definite cerebral dysfunction. '' The trend for the judges to rate the normal subjects as having no cerebral impairment and the retarded readers as having definite cerebral impairment was significant. Reitan and Boll (1973) evaluated the neuropsychological correlates of minimal brain dysfunction (MBD) in children 5 to 8 years of age. Their study included four groups of approximately 25 children each. The categories were normal; brain-damaged; MBD children whose primary referral was due to academic problems; and MBD children who were
described as having mainly behavioral problems in the classroom. The test results were subjected to two types of analysis: an analysis of variance supplemented by t tests, and a blind overall judgment of whether each child appeared to have normal, mildly impaired, or abnormal brain functions. Using statistical methods to analyze the level of performance, the authors found that the normal subjects generally achieved the best scores and the brain-damaged subjects the lowest scores, with the two MBD groups usually scoring in between. The brain-damaged groups consistently performed significantly more poorly than the other three groups, but the differences between the two MBD groups and the control subjects generally failed to reach significance. The blind judgment analysis achieved finer distinctions between the groups. Using only the test protocols, Reitan identified 64% of the normal children as having normal brain functions, approximately 85% of the MBD subjects as having mildly impaired brain functions (averaging the two groups), and 96% of the brain-damaged children as having abnormal brain functions. Finally, in research designed to specifically address the neuropsychological difference between normal, learning-disabled, and brain-damaged children in the 9- to 14-year age groups, Selz (1977) found significant differences between these three groups. Using analysis of variance on three summary IQ scores and 10 neuropsychological measures, there were significant differences between the groups beyond the .01 level on II of the 13 measures, and a significant difference beyond the .015 level on one measure.. The t-test comparisons indicated that the learning-disabled subjects performed in a relatively normal manner on measures with a strong motor component. The performance of the disabled learners resembled that of the brain-damaged subjects on tests with strong cognitive or attentional demands. A discriminant analysis that utilized the 13 measures classified subjects according to group membership with 80% accuracy. The 20% classification errors tended to be in the direction of classifying subjects as less impaired than their group membership implied, for example, assigning a learning-disabled subject to the control category (Selz & Reitan, 1979).
Refinements in the Study of Learning Disability Once the evidence supporting the ability to differentiate learning-disabled children from normal children using neuropsychological analysis was
BASES OF COMMON LEARNING AND BEHAVIOR PROBLEMS
found, researchers began developing new applications for this method of assessment. Just as Reitan had begun his career by developing an approach to aid in systematic subdivision of the term brain damage, the most recent application of neuropsychological methods centers upon attempting to subdivide learning disabilities into discrete categories. The next section of this chapter will focus on the research from the Neuropsychology Laboratory of Windsor Westem Hospital Centre, one of the major laboratories involved in this effort. This work was chosen for its comprehensive treatment of learning disabilities, and is an example of the work being done in several different settings. One of the Neuropsychology Laboratory's major contributors to research utilizing neuropsychological assessment of children with learning disabilities is Byron Rourke, who has provided evidence supporting the division of learning disabled into distinct subgroups. He has also provided the rationale for these distinctions, in that the classifications provide information pertinent to both prognosis and rehabilitation. Rourke's work will be presented in three parts: reading disorders, spelling disorders, and arithmetic disorders.
Reading Disability: Analysis and Remediation Initially, Rourke began his research program based on the theoretical question of whether it was the case that some or all of the deficits exhibited by children who are classified as "learning disabled" are the result of cerebral impairment (Rourke, 1975). The investigation of this question led to the utilization of neuropsychological assessment as well as the implementation of Reitan's four methods of inference. The first important application of this method of investigation was a 4-year longitudinal study of reading disability. Due to its importance as an early demonstration of the use of neuropsychological methodology in the study of learning-disabled children, this work will be presented in some detail. In the initial phase of the investigation, there were 30 subjects in the normal-reading (NR) group and 29 subjects in the retarded-reading (RR) group. Subjects were assigned to their respective groups based on their performance on two subtests of the Metropolitan Achievement Test (MAT). Subjects were matched for age; their ages were between 7 years, 2 months and 8 years, 4 months. The study included testing at intervals of2 years, 3 years, and 4 years after the initial assessment. By the final testing
133
period, 24 NR and 20 RR subjects from the original groups participated in the retesting. One of the important deviations of this study from previous work in the area was Rourke's decision to include only subjects with Full Scale WISC scores between 91 and 117. Rourke defended this decision on the basis that it would be reasonable to assume that "dull" and "bright" children would differ markedly on a large number of variables, rather than just on reading ability, and that the results of this investigation were intended to apply to groups of "average" children. The overall results of this study showed that in general, the children in the RR group were further behind the children in the NR group at the time of the final assessment than they had been initially. This was attributed to the fact that most of the NRs made more than 4 years of progress in reading over the 4year period, whereas the PRs made less than 2 years' progress. Several subjects originally classified as RR made significant advances in reading performance during this 4-year period, but it was found that there were variables that reliably differentiated between these RRs and those who made little if any progress. This investigation provided important methodological support for the utilization of Reitan's four methods of inference (level of performance, difference scores, comparisons of performance on the two sides of the body, and pathognomic signs). These plus developmental analyses provided important information that may not have been available were such rigorous methods not practiced. Another issue addressed in this investigation by Rourke was the relative merits of the positions in the controversy between the "developmental lag" model and the "deficit" model. Satz and Van Nostrand (1972) had postulated an underlying "lag mechanism'' in brain maturation that they thought forecast the later onset of dyslexia. Based on this, they conceptualized dyslexia within the framework of a developmental model rather than that of 3 ''deficit'' model. They suggested that the underlying lag in brain maturation causes a delay in the rate of acquisition of developmental skilJs rather than a loss or impairment in these skills. The "deficit" model, supported by Doehring (1968), Reed (1968), and Reitan (1964), suggests that there is some sort of cerebral dysfunction underlying the acquisition of age-appropriate reading skills. The "deficit" view differs from the "developmental lag" position in that there is no necessary expectation that the children who suffer from the deficit(s) will ever catch up with their normal age mates in those skills that are required for age-appro-
134
CHAPTER 7
priate reading, but is similar to the "developmental lag" view in that it also predicts a less than ageappropriate level of reading performance throughout the elementary school years. The "deficit" position supports the possibility that children will exhibit a deficit in the cerebral structures or systems that subserve some of the abilities necessary for reading, but allows for adaptation and/ or compensation for the deficits in question. Drawing from the work of Luria ( 1966) and Wepman ( 1964), the possibility of enlisting other neuronal structures or systems that in the ordinary course of events would not subserve these abilities is entertained. In order for Rourke to relate this investigation to the assessment of these two positions, seven developmental lag-deficit paradigms were constructed. Then the results of Satz, Friel, and Rudegair' s ( 1974) study as well as Rourke's (1975) longitudinal study were compared in the light of these paradigms (Rourke, 1976a). In general, it was found that there was some support for the "developmental lag" position in the case of fairly simple, early-emerging abilities. Specifically, the position presented by Satz and Van Nostrand (1972) postulates a particular lag in the development of the left cerebral hemisphere. Both Rourke's ( 1975) study as well as Staz and colleagues' (1974) study lent support to the notion that dysfunction of the left cerebral hemisphere is particularly involved in the genesis of reading retardation. However, Rourke did not find that RRs, as a group, eventually "caught up" in those abilities thought to subserve the reading function, nor did they "catch up" in the ability to read. Thus, the weight of the evidence of Rourke's ( 1975) study favors a "deficit" rather than a "developmental lag" position. Another study that grew out of this longitudinal investigation used measures of reading, spelling, and psychometric intelligence as well as the Underlining Test (Doehring, 1968), in an attempt to determine their relative predictive accuracy for both the NR group and the RR group (Rourke & Orr, 1977). The results indicated that performance on the Underlining Test was the best predictor of eventual achievement levels in reading and spelling when compared to the other tests used (i.e., reading, spelling. and psychometric intelligence). It was also found that the Underlining Test was particularly sensitive with the RRs, providing accurate predictions about the substantial gains in reading by the one subgroup of RRs. Although the results of this study were exciting, Rourke and Orr ( 1977) advised caution in the drawing of clinical inferences (including prognosis) for individual cases. This caution was necessary because of the restricted range of IQs and the truncated and
restricted distributions of initial MAT subtest scores. Also to be considered are the restricted age span and fairly small number of measures chosen for comparison (Rourke, 1976b). The final study to be discussed in this section constitutes Rourke's first attempt to design a study specifically to identify subgroups of reading-disabled children (Petrauskas & Rourke, 1979). The study used a multivariate classification procedure (Q-type factor analysis) similar to that employed by Doehring and Hoschko (1977). The sample was comprised of 150 children between 7 and 8 years of age. There were 133 RRs chosen from a clinic population and 27 NRs who were tested in the original longitudinal investigation. The results indicated three types of RRs that could be differentiated from NRs and one another. Following is a brief description of each subtype (including the subtype containing the NRs).
Subtype 1 Subtype I could be characterized as having relatively well-developed visual-spatial and eye-hand coordination skills; average or near-average tactilekinesthetic abilities, abstract reasoning, and nonverbal concept foirnation; near-average word definition ability; mildly impaired word-blending ability, immediate memory for digits, and store of general information; and moderate to severe impairment in verbal fluency and sentence memory.
Subtype 2 Subtype 2 could be characterized as having average or near-average kinesthetic, psychomotor, visual-spatial constructional and word-defining abilities, and nonverbal problem-solving and abstract reasoning skills within a context that provides immediate positive and negative feedback; borderline to mildly impaired immediate memory for digits and other "sequencing" skills, store of general information, sound-blending verbal fluency, and concept formation when substantial verbal coding is required and/or when no positive and negative feedback is provided; and moderate to severe impairment in finger recognition, immediate visual-spatial memory, and memory for sentences.
Subtype 3 Subtype 3 could by characterized as having average or near-average finger recognition (left
BASES OF COMMON LEARNING AND BEHAVIOR PROBLEMS
hand), kinesthetic, visual-spatial constructional, vocabulary and sound-blending abilities, and nonverbal concrete formation within a context of immediate positive and negative feedback; borderline to mildly impaired finger recognition (right hand), immediate memory for digits, eye-hand coordination under speeded conditions, store of general information, and nonverbal abstraction and the shifting of set without the benefit of positive and negative feedback; mild to moderate impairment in verbal fluency, sentence memory, and immediate visual-spatial memory; and moderate to severe impairment in concept formation that involved substantial verbal coding. Subtype 4 Subtype 4 could be characterized as having average or above-average finger recognition, kinesthetic, sequencing, eye-hand coordination, visualspatial constructional, visual memory, auditory-verbal receptive, and concept-formation abilities. The results of the Petrauskas and Rourke ( 1979) study illustrated the complex nature of the neuropsychological dimensions of the groups of clinical disorders known primarily as "reading disability." At the very least, this analysis provided evidence indicating that reading-disabled children do not constitute a homogeneous group. It is also clear that deficiencies in psycholinguistic skills play a major role in reading difficulties at this age level. Further work (Fisk & Rourke, 1978) found that psycholinguistic disabilities are also the source of major problems for different subtypes of learning-disabled children ranging from 9 to 14 years of age. These results provide important clinical applications. If the various subtypes of reading disability have various etiologies, then perhaps reading-disabled children will respond differentially to various forms of teaching and intervention. Investigations designed to address this issue are needed and are part of Rourke's ongoing research strategy (Rourke & Strang, 1983). Also, Rourke and his colleagues have been able to provide speculations regarding the likelihood of dysfunction maximally involving different systems, suggesting a relationship between the left temporal lobe and reading disability associated with subtype I; a relationship between the left temporoparieto-occipital area and reading disability associated with subtype 2; and a relationship between the left frontal lobe and reading disability associated with subtype 3 (Rourke & Strang, 1983). Of course, these attempts at localization are only speculative at this time, but they provide a basis for further validation and research.
135
In regard to specific remediation recommendations, at this time there seem to be very little available. It is stressed that the nature of the deficit determines the remedial teaching methods (Aaron, 1981 ), which has led to a lack of any systematic approach. This is in part due to the lack of information pertaining to the different types of deficits that are manifested as reading disability. The work of Rourke is an attempt to understand these diverse deficits. However, as he has stated, the research focused on development of remedial programs has just begun (Rourke & Strang, 1983). At this time the basic approach is to differentiate between children who respond to phonetic approaches and those who respond to whole word approaches (Aaron, Grantham, & Campbell, 1978). This is very similar to the remedial approach to spelling disability, which will be detailed in the next section.
Spelling Disability: Analysis and Remediation Several studies conducted by Rourke and his colleagues have attempted to analyze spelling errors from the point of view of their degree of phonetic accuracy. The impetus for this line of research emerged from clinical practice with learning-disabled children and research evidence derived from studies of adults with well-documented lesions (e.g., Kinsbourne & Warrington, 1963; Rourke, 1976c). Both sources have found consistent evidence indicating the relative "intactness" of children (and adults) whose misspelling was essentially phonetically accurate (e.g., gelusy for the word jealousy). This is contrasted to misspellings that are phonetically inaccurate (e.g., htvowe for jealousy). In particular, research has documented that adults with lesions of the so-called "language areas" of the left cerebral hemisphere exhibit a tendency to spell in a phonetically inaccurate way (e.g., Kinsboume & Warrington, 1963). Rourke began a line of research based on these findings that was designed to investigate the existence of two types of spelling disability. In the first study (Burgher & Rourke, 1976), the spelling production of the subjects from the longitudinal investigation of reading retardation (Rourke, 1975) was examined. It was found that RRs had significantly fewer phonetically accurate spelling errors than NRs. Also, there was a "leveling off' in the degree of phonetic accuracy of the misspellings after the second year of testing (i.e., ages 9-10). The next study (Sweeney & Rourke, 1978) was cross-sectional in nature and was based on two
136
CHAPTER 7
groups of studies. The first group of studies was conducted by Newcombe (1969) and Nelson and Warrington (1974) and found a positive relationship between spelling retardation characterized by phonetically inaccurate errors and a general impairment in language functioning. The second group was conducted by Boder (1973) and demonstrated a positive relationship between spelling performances characterized by an excess of phonetic accuracy and poor memory for how words looked. The subjects for the study conducted by Sweeney and Rourke (1978) were children at age levels corresponding to grades four and eight. There were three groups, one a control group consisting of normal spellers (NS) and two experimental groups of disabled spellers. One of the experimental groups consisted of phonetically inaccurate (PI) spellers; the other was composed of phonetically accurate (PA) spellers. It was found that the PI spellers were inferior to the PA spellers and the NS on most measures of psycholinguistic ability at the older (grade eight) age level. Upon closer examination, it was discovered that the PA spellers performed at levels similar to those of NS on tests in which the correct responses could be brief, single-word answers, whereas the PI spellers were inferior to both the PA spellers and the NS. In contrast, the levels of performance of PA spellers could not be distinguished from those of PI spellers and were inferior to NS on those measures that required a fairly complex formulation of an answer to a question, and when the correct answer could not be determined exclusively by utilizing the information contained in the question itself. Sweeney and Rourke ( 1978) concluded that the older PA spellers experienced significant difficulty in associating spoken language with the analysis of visual-spatial information; in relating the verbal information provided with other information; and encoding relatively complex word strings in response to verbal information provided. Returning to the younger group, it was found that the arithmetic subtest of the Wide Range Achievement Test (WRAT) differentiated the three groups, with the PI spellers exhibiting the poorest performance. This suggested that one important deficiency in the PI spellers at this age level might be their ability to benefit from formal instruction in the use of rules and, consequently, in logical-grammatical reasoning. This prompted further analysis of these subjects at the fourth grade level (Sweeney, McCabe, & Rourke, 1978) in which the subjects' performance on the Logico-Grarnmatical Sentence Comprehension Test (Wiig & Semel, 1974) was compared. There-
suits of this study indicated that whereas PI spellers were significantly inferior to NS in their logicalgrammatical reasoning, PA spellers were not inferior. This led Sweeney et at. (1978) to pose two non-mutually exclusive possibilities: (I) that younger PI spellers are deficient in the ability to benefit from formal instruction in rules by which different formsof symbolic information are processed; and (2) that PI spellers have difficulty in processing language that can be viewed as somewhat more complex than everyday conversation. In general, these studies indicate the heuristic value of a qualitative analysis of the spelling errors of children if the nature of the underlying deficiencies of spelling retardation is to be explicated. Also, it is clear that the age at which children were tested was a very important consideration with respect to the relationship between degree of phonetic accuracy of misspellings and performance on the dependent variables (Rourke, 1978). Sweeney and Rourke (1985) formulated an extensive remedial program that will be presented here in brief (for further detail, refer to the original source). The limited ability of PI spellers to carry out specific operations on linguistic information (e.g. , phonemic segmentation) suggests that this subtype of disabled spellers would encounter significant difficulty in learning the rudiments of applying such operations to specific verbal information, including single words, through conventional classroom instruction. Also, these children would have difficulties in generalizing from conventional instruction to new verbal information, including unfamiliar words. Therefore, it may be appropriate to provide PI spellers with the opportunity to "operate" excessively on each word to be learned (e.g., by synthesizing repeatedly the individual speech sounds of the words presented in sequence, and by segmenting repeatedly the phonemic composition of the word presented orally). Once these operations can be carried out efficiently on a particular word, it would probably also be beneficial to develop the ability to read the word "by sight" in order to promote fluency in reading. If concentrated training in the application of these basic operations on unfamiliar words does not improve performance, then it would be reasonable to restrict instruction to intensive development of a sight word reading strategy. Sweeney and Rourke ( 1985) proposed the utilization of remedial methods involving intense visual analysis to aid the PA spellers. They refered to the "right hemisphere strategies" proposed by Bakker, Moerland, and Goekoop-Hoefkens (1981), which
BASES OF COMMON LEARNING AND BEHAVIOR PROBLEMS
consist of (l) development of a sight word strategy; (2) an attempt to increase the salience of the visualspatial features of word configurations, particularly for unfamiliar words and words with low phonemegrapheme correspondence; and (3) encouragement of visualization of the graphic features of words to be spelled and to attend to disregarding how the word looks. Other strategies that may benefit PA spellers include flashcard exercises, activities that require recognition of words in tachistoscopic presentations that gradually decrease in exposure time, and exercises requiring simultaneous reading aloud with an instructor who gradually decreases his/her input to promote independent oral reading. It may also be beneficial to have the child, while blindfolded, simultaneously trace the spatial features of relatively unfamiliar words made from sandpaper, while repeating the words, and then to write these words out correctly, again while repeating the words orally. This may serve to increase the salience of the physical forms of word configurations (Sweeney & Rourke, 1985). Both PI and PA spellers demonstrated difficulty in assimilating verbal information and in making generalizations from this information. This may be addressed by multisensory teaching in conjunction with dramatization. Although this method would provide some overlap in the teaching aids for both PI and PA spellers, it should be stressed that the emphasis for each group is quite different. Specifically, although both groups have demonstrated receptive language difficulties, the deficiencies of the PI spellers focus on their inability to carry out very basic, receptive linguistic operations. In contrast, the PA spellers have difficulty in associating the spoken word with an analysis of visual-spatial information, relating verbal information provided with existing information, and encoding relatively complex word strings in response to verbal information provided (Sweeney & Rourke, 1985). Investigating the development of linguistic and cognitive abilities in children who exhibit qualitatively distinct spelling disorders would seem to be a natural course to follow in delineating underlying deficiencies, with a view to the formulation of a specific plan of remediation. It should be emphasized, however, that the patterns of abilities and deficits emerging from the studies reviewed and the associated remedial strategies presented, are not considered exhaustive. Replications of these studies and additional investigations geared to specifying further the nature of the deficits related to different types of spelling disorders would certainly be warranted.
137
Arithmetic Disability: Analysis and Remediation There is a notable lack of research pertaining specifically to arithmetic disability as compared to studies of retarded reading and spelling (Rourke, 1978). One reason for this may be the lack of educational concern due to the emphasis in Western culture on reading and spelling (Rourke & Strang, I983). The fact that arithmetic is often viewed as a type of language activity (e.g., the Arithmetic subtest of the WISC-R is considered part of the verbal scale) may also contribute to the neglect of this area of learning. Rourke and his colleagues at the Neuropsychology Laboratory have conducted important work in this area. The first of these important studies was that of Rourke and Finlayson ( I978) who attempted to determine if children who exhibited arithmetic retardation within the context of differing patterns of reading and spelling performances would also exhibit differing patterns of brain-related abilities. For this study, children between the ages of 9 and I4 who had been classified as learning disabled were divided into three groups on the basis of their patterns of performance in reading and spelling tasks relative to their performance in arithmetic. The children in Group 1 were uniformly deficient in reading, spelling, and arithmetic. The children in Group 2 exhibited significantly better performance in arithmetic (which was still below age expectation) relative to reading and spelling. The children in Group 3 exhibited normal reading and spelling but marked impairment in arithmetic. Also, Groups 2 and 3 did not differ in their level of arithmetic performance, although they were superior to the Group I in arithmetic performance. The three groups were equated for age and Full Scale IQ on the WISC. There were 16 dependent measures that provided the basis of comparison. Analysis of the three groups' performance on the dependent measures found: (I) the performance of Groups 1 and 2 were superior to Group 3 on measures of visual-perceptual and visual-spatial abilities, and (2) Group 3 performed at a superior level to Groups 1 and 2 on measures of verbal and auditoryperceptual abilities. These results were interpreted as being consistent with the view that the children in Group 3 may have a relatively dysfunctional right cerebral hemisphere, and that children in Groups 1 and 2 were suffering from the adverse effects of a relatively dysfunctional left cerebral hemisphere (Rourke & Finlayson, I978). Of particular interest within this context was the fact that the two groups who had been equated for deficient arithmetic perfor-
138
CHAPTER 7
mance (i.e., Groups 2 and 3) exhibited vastly different performances on verbal and visual-spatial tasks. These differences were clearly related to their patterns of reading, spelling, and arithmetic, rather than being limited to only their arithmetic performance. A second study pertaining to this question was conducted shortly after the first by Rourke and Strang (1978), utilizing level of performance comparisons as well as comparison of performance on the two sides of the body to determine differences between the three groups of children from the first study (Rourke & Finlayson, 1978). Rourke and Strang (1978) were specifically interested in the ability of motor, psychomotor, and tactile-perceptual tests to differentiate between the two groups, and they found that there were no significant differences evident on the simple motor tasks. However, on more complex psychomotor measures as well as on a composite tactile-perceptual measure, the children in Group 3 exhibited marked impairment. Finally, there was evidence consistent with the view that Group 3 children were suffering from the adverse effects of a relatively dysfunctional right cerebral hemisphere within the context of satisfactory left hemispheric functioning. Opposed to this was the evidence that Group 2 children exhibited indications that would be consistent with the opposite pattern of hemispheric integrity. A third study (Strang & Rourke, 1983) using the same children from Groups 2 and 3 compared their performance on the Halstead Category Test (Reitan & Davison, 1974). This investigation found that children in Group 3 made significantly more errors on the test. Also, whereas the level of performance of the Group 2 children was age appropriate on this nonverbal problem-solving task, Group 3 children's performance was approximately one standard deviation below the mean. Specifically, these children had particular difficulty on the subtests requiring "high-order" visual-spatial analysis. The overall consistency of these three studies (Rourke & Finlayson, 1978; Rourke & Strang, 1978; Strang & Rourke, 1983) is striking. All three studies indicated that learning-disabled children who are better on the Arithmetic subtest of the WRAT compared to the Reading and Spelling subtests (i.e., Group 2) performed poorly on tests thought to tap abilities primarily subserved by the left cerebral hemisphere. Children whose only deficit appeared on the Arithmetic subtest of the WRAT, with Reading and Spelling within normal limits (i.e., Group 3), did poorly on measures of abilities thought to be subserved primarily by the right hemisphere. Rourke and his colleagues have conducted qualitative analyses on the performance of Group 3 children on the WRAT
Arithmetic subtest, in an attempt to gain further insight into this neglected disability. The results and implications of this approach will not be discussed here; the reader is referred to Strang and Rourke (1985) for a detailed description. Of particular importance from a psychoeducational standpoint is the pattern of results that emerged in these studies in the case of the children with outstanding or "specific" deficiencies in arithmetic (Group 3). It is clear that these children exhibited a number of adaptive deficiencies that should render them the focus of more serious concern than had been the case. It is quite probable that they are not seen as in need of early help with their scholastic career because they read and spell at normal levels. Also, it may be the case that specific deficiencies in arithmetic are thought by many to result from socioemotional disturbances, genetic predispositions, or motivational shortcomings, which are not within the normal province of educational intervention and are often thought to "pass with time" (Rourke, 1978). And this may very well be the case in some instances. However, it is also clear that at least a subset of such children would appear to have a number of significant brain-related deficiencies that can and should be recognized and dealt with early in their academic career. Strang and Rourke (1985) presented two principles that underlie the arithmetic difficulties of children who exhibit neuropsychological arithmetic difficulties similar to Group 3 children. First, it must be considered that mechanical arithmetic should be made as much a verbal task as possible. Second, the teaching of such children must be highly systematic and rather concrete, sometimes requiring physical aids to illustrate mathematical concepts (for a more detailed treatment of this issue, refer to Strang & Rourke, 1985). The striking differences between the two groups that were equated for level of performance in arithmetic calculations (i.e., Groups 2 and 3) should also serve as a note of caution regarding the composition of learning-disabled groups of children for study. Had subjects from Groups 2 and 3 been combined to form a group retarded in arithmetic for comparison with a normal control group or other learning-disabled children, it is clear that the differences evident between these two groups would simply have been cancelled. The important conclusion that emerges from this type of analysis is that it is no longer acceptable within the field of development or learning disabilities to constitute groups of children for study solely on the basis of their levels of performance on academic tasks. One must at least be prepared to
BASES OF COMMON LEARNING AND BEHAVIOR PROBLEMS
139
designate qualitatively distinct types of reading or ated with speech and reading in normal individuals as arithmetic disabilities, if such exist, before launching measured by more invasive techniques such as sointo a measurement of the supposed correlates of dium Amytal testing (Wada & Rasmussen, 1960) and these problems. Failure to do so may not only pose regional cerebral blood flow (rCBF) (Lassen, lngvar, severe and unnecessary limitations on the conclu- & Skinhoj, 1978). sions that can be drawn from such studies, it may also Another electrophysiological procedure often increase the probability that Qlatantly false "find- used to investigate the basis of learning disability has ings" will be generated and propagated (Strang & been evoked potential (EP). This procedure has enRourke, 1985). countered the same type of difficulties encountered by the EEG studies, but the difficulties have similarly 6een somewhat ameliorated by the BEAM technique (Duffy et al., 1979). Further refinement of this proElectrophysiological Correlates of cedure has resulted in the finding that brain stem Learning Disability evoked responses (BSER) appear to be a viable electrophysiological procedure to assess learning disA relatively new approach to the study of learn- ability (Obrzut, Wilson, Lord, & Caraveo, in press). ing disabilities is from a neurobiological perspective, Until recently, auditory brain stem responses have which supports the neuropsychological etiology of been evoked by the binaural presentation of a brief, some childhood learning disorders (Gaddes, 1980, discrete sound stimulus and recording of the poly1981; Hynd & Cohen, 1983; Obrzut & Hynd, 1983; phasic change in the electrical activity of the brain Geschwind & Galaburda, 1985). Electrophysiolog- stem during the first 10 msec following the stimulus. ical and neuropathological evidence is accumulating These BSERs are thought to be the far-field reflection that supports the existence of the clinical syndrome, of sequential electrical events at successively higher "learning disability." Although this line of investi- levels of the brain stem auditory pathway. gation is beginning to provide some gross differentiaRecent refinement in this approach involved tion within this syndrome (i.e., language versus spa- presenting the auditory stimulus monaurally to the tial difficulties), the research is unable at this time to left or right ear, and determining any asymmetry in provide the fine distinctions found in the neuropsy- the BSER that occurs in response to the monaural chological research. stimulus. To date only a few attempts to assess learnOne of the current methods in this line of re- ing-disabled populations via this procedure have search has focused on developing and refining elec- been made and these reveal a marked symmetry in trophysiological procedures to investigate the basis neurologically normal persons and characteristic of learning disabilities. Initial work in this area used asymmetry in a special population (stutterers) EEG data as indices of cortical dysfunction. Major (Decker & Howe, 1981). The limited research in difficulties encountered have been concerned with evoked responses in learning-disabled populations the interpretation of the temporal and spatial rela- has focused upon abnormalities in laterality or asymtionship of vast arrays of electrode patterns in on- metry in the cerebral cortex. This rather gross apgoing brain activity. To deal with this problem, proach has not yet provided any useful diagnostic Duffy, Burchfield, and Lombroso (1979) developed taxonomy for the various subtypes of learning disthe brain electrical activity mapping (BEAM) tech- ability, including disabilities involving confusion in nique, which can summarize, reduce, and visually the processing of auditory information. display spectral, spatial, and temporal information of There are a number of other electrophysiologbrain activity from 20 different scalp locations. This ical techniques available for the study of the brain, information is then statistically compared to a control e.g., computerized axial tomography (CAT), maggroup (Torello & Duffy, 1985). netic resonance (MR) imaging, and positron emisThe BEAM recording technique has demon- sion tomography (PET) scans. These tools may be strated diagnostic utility within a well-defined popu- considered invasive because of the potential for exlation of dyslexics, finding that dyslexia is a neu- posure to radiation. In addition, these techniques rerophysiological problem with topographically quire that a state-dependent condition (e.g., same specific brain areas of dysfunction (Duffy, Denckla, state cognitive task) exist for approximately 30 min Bartels, & Sandini, 1980). Discrete brain areas were to provide time for the radioactively labeled glucose identified in the "pure" dyslexic subjects as being compound to enter the metabolically activated cells significantly different from the age-matched control for detection by the sensors (Torello & Duffy, 1985). subjects. These areas have been classically associ- For these reasons, the research focusing on learning
140
CHAPTER 7
language and right dominance for spatial processing is somehow arrested in the dyslexic, resulting in increased symmetry. Further. there has been a connection between learning disabilities and non-righthandedness (Geshwind & Behan, 1982). But as the Although the emphasis of the electrophysiolog- authors noted, the findings in these studies do not ical research has been on differentiating normal imply that non-right-handedness causes learning dislearners from disabled learners, other research has abilities. Rather, non-right-handedness is simply a focused upon studying the developmental process of marker of an alteration in dominance. Geschwind the brain. Specifically, it has been found that investi- and Galaburda (1985) further stated that there is a gation into the asymmetrical development of the strong genetic component associated with learning brain may lead to new hypotheses concerning the disabilities, which is sex-limited (a gene is expressed etiology of learning disabilities (Geschwind & Ga- less often in one sex). They proposed that learning disorders are more common in males because of laburda, 1985). In their recent review, Geschwind and Galabur- some male-related factor in development, possibly da ( 1985) presented evidence for the following: ( 1) testosterone (e.g., Dorner, 1980; Gorski, Harlan, & anatomic asymmetry of the brain; (2) asymmetry of Jacobson, 1980; Pfaff, 1966; MacLusky & Naftolin, the fetal brain; (3) male-female differences in the 1981; McEwen, 1981; Raisman & Field, 1973). brain; (4) laterality in developmental disorders; (5) When one considers the consequences of these specabnormal cytoarchitecture in childhood dyslexia; (6) ulations, it would appear that learning disorders may patterns of maturation of the brain; (7) cell death in not be preventable, making the need for greater unthe developing brain; (8) reorganization of the brain derstanding of the etiology as it pertains to remediaafter intrauterine lesions; (9) hormonal influences on tion even more important. brain structures; (10) genetic studies of handedness; (11) chemical asymmetry of the brain; (12) evolution of asymmetry; (13) associations of left-handedness; and (14) relationships of laterality to special talents. Attention Deficit Disorder (ADD) As can be seen from this list, the review by GeWithin the neuropsychological literature, ADD schwind and Galaburda (1985) was quite comprehensive. It is not our intention to attempt a summary of is considered the major behavioral disorder in childthe fmdings; our discussion will be limited to those hood (Goldman, Thibert, & Rourke, 1979). The findings dealing directly with developmental disor- DSM-lll includes under ADD the childhood disorders and childhood dyslexia. The interested reader is ders that have previously been labeled hyperactivity, hyperkinesis, minimal brain dysfunction, minimal referred to the original article for greater detail. Geschwind and Galaburda ( 1985) postulated brain damage, minimal cerebral dysfunction, and that the brain abnormalities in dyslexia are develop- minor cerebral dysfunction. The clinical signs of mental in nature and can be attributed to alterations in ADD are developmentally inappropriate inattention, the cortex and connectionally related subcortical excessive overactivity, and impulsivity. The DSMstructures resulting from disturbances in neuronal 111 identifies two subtypes of ADD, one with and one migration and assembly. Further, they stated it is without hyperactivity. The DSM-lll also provides a likely that in some individuals the postulated slowing third category that reflects the findings that in most of the rate of migration ofneurons to the left cortex is cases, hyperactivity decreases markedly or disapeven more marked than in the normal state; this ex- pears after pubescence without a comparable reduccessive delay can lead to lateralized developmental tion in either inattention or impulsivity (ADD, Rearrest and malformation, thus leading to childhood sidual Type). The cause of ADD may be brain damage, psydyslexia and possibly to other developmental disorders. This hypothesis is based on the case studies of chosocial frustrations, or a mixture of the two (Pirfive patients (Drake, 1968; Galaburda, 1983; Ga- ozzola & Campanella, 1981). Although the scope of laburda & Eidelberg, 1982; Galaburda & Kemper, neuropsychology includes differential diagnoses be~ tween the different causes, the major contribution of 1979). As can be seen, the emphasis of this type of neuropsychology probably lies in its ability to exresearch is finding the neurological basis for learning plain the disorder in terms of brain-behavior reladisabilities. At this time, it appears that the normal tionships. In order to do this, both the developmental asymmetrical development with left dominance for processes of the brain and the differences between the disabilities has not used these methods. As can be imagined, it would be difficult to justify the use of these invasive methods with a normal population, as well as children with difficulties discernible by less dangerous methods.
BASES OF COMMON LEARNING AND BEHAVIOR PROBLEMS
fully mature brain (adult) and the maturing brain (child) must be considered.
Developmental Processes Gontrary to a notion once widely held, the brain is not fully developed at birth (Boll & Barth, 1981). In fact, research now supports the model of a developmental process that begins at conception and may end as early as 12-15 years of age or as late as the mid-20s. Therefore, neuropsychological evaluation needs to identify not only what skills should be present at what age, but also at what ages what skills are appropriately or pathologically absent. Further, the brain does not mature in a vacuum, but is the result of its interaction with the child's environment. At present, no one-to-one relationship has been established between periods of physical growth in the brain and psychological maturation (Golden, 1981). Although this may appear to make generalizations about the relationship between children's behavior and their brain integrity extremely difficult, by combining a developmental model with a theory of how the systems within the brain function, the neuropsychologist can successfully diagnose behavior disorders such as ADD and hypothesize the neurological causes. One well-formulated theory of how the developing brain forms higher mental processes has been proposed by Luria (1964). He conceptualized the brain as composed of three complex functional systems, and that the brain's functions "may suffer as a result of the destruction of any link which is a part of the structure of a complex functional system" (p. 11). A brief description of Luria's theory will provide an understanding of how injury to the brain, and in particular to the reticular activating system (RAS), may produce the diverse syndrome now classified as ADD. Luria (1964) conceptualized the functional systems of the brain as consisting of three units. The first unit is designated the arousal unit, which consists of those parts of the brain identified as the RAS. This system is a collection of diffuse intertwined structures that act to raise or lower cortical arousal as well as filtering sensory input, especially from those senses that are always "on" (tactile/kinesthetic and auditory). This prevents the cortex from being flooded with constant, irrelevant stimuli, which can interfere with cognitive processing. It is now considered that the arousal unit is the most -important element in the development of normal attention. Injury to the RAS during the prenatal or perinatal period may result in ADD. One symptom of
141
this disorder, hyperactivity, has received a great deal of scientific attention. It has been hypothesized that hyperactivity may result from two separate mechanisms (Satterfield, Cantwell, & Satterfield, 1974). In the frrst, the child is "underaroused," which implies that the reticular system is not producing sufficient cortical arousal or is not allowing enough stimuli through to maintain cortical arousal. This induces a state of "sensory deprivation," a situation that is highly aversive if prolonged. In response, the child will attempt to generate additional input and arousal in the only manner available, generally motor movements, including vocalizations. The opposite of the underaroused child is the overaroused child. In this case, the difficulty appears to be in the ftltering system, which fails to screen stimuli, thereby flooding the cortex with an excess of information. The child becomes stimulus bound, unable to focus on any one thing. However, the behavior is very similar to the underaroused child in that this child is unable to sit quietly for any length of time, which also increases the activity level of the child. Drug studies have provided evidence supporting this hypothesis of two separate forms of hyperactivity. It has been found that approximately 70% of children displaying hyperactive behavior respond positively to stimulant medication, and 30% either show little improvement or actually become worse (Knights& Viets, 1975; Satterfield eta/., 1974). The explanation offered for these results is that hyperactive children who respond best to stimulant medication are those who have low CNS arousal levels before treatment. Satterfield et al. (1974) hypothesized that associated with low CNS arousal levels there is insufficient CNS inhibitory control over motor function and that CNS arousal and inhibition vary together. Consistent with this, it was found that the children who responded well to stimulant medication had the greatest amount of restlessness as reported by school teachers and had excessive movement-generated EEG artifacts in the laboratory. Satterfield et al. (1974) cited animal neurophysiological studies in which stimulation effects at the reticular formation both increased cortical arousal and enhanced the inhibition of sensory signals at synapses in the sensory pathways. All of this appears to indicate that a majority of children manifest hyperactive behavior due to low CNS arousal and insufficient inhibitory control over motor outflow and sensory input due to RAS dysfunction. The remaining children may already be overaroused, which explains the ineffectiveness of stimulant drug treatment, and is also related to a problem in the RAS. The second unit discussed by Luria is the most
142
CHAPTER 7
widely and frequently researched area of the brain (Reynolds, 1981). This unit is responsible for most early learning skills as well as higher-order cognitive information processing. According to Luria's conceptualization, the second unit is subdivided into three types of areas: primary, secondary, and tertiary. The primary areas act as sensory reception areas-input is received on a general "point-topoint" basis from the appropriate sensory area. There are three primary areas, each devoted to a specific sense. The primary auditory area is in the temporal lobe; the primary visual area is in the occipital lobe; and the primary tactile/kinesthetic area is in the parietal lobe. The secondary areas analyze and integrate the information from the primary areas. It is thought that the secondary areas of the sensory input unit process information sequentially, which allows the brain to be aware of stimulus change and to link events temporally. Also, there is more specialization in the role of the hemispheres at the secondary level. For most individuals, the left hemisphere predominates in the analysis of verbal material, and the right hemisphere predominates in the analysis of nonverbal material. However, this is also a developmental process and careful consideration of the skills required of the functional system must be recognized. For example, the earliest stages of reading require the recognition of unfamiliar shapes, a task that is primarily done in the right hemisphere. However, once the letter is learned, it becomes a verbal symbol primarily detected by the left hemisphere (Golden, 1981). The tertiary level of the sensory input unit, located primarily in and around the parietal lobe, is responsible for cross-modality integration. This area plays a primary role in many of the tasks commonly subsumed under "intelligence." For example, auditory-visual integration is necessary for reading, whereas auditory-tactile integration is necessary for writing. Arithmetic, as well as body location in space and visual-spatial skills, depends upon visual-tactile integration. Further hemispheric differentiation of tasks also occurs at this level. The left hemisphere is responsible for grammar, syntax, and other language-related skills, as well as the understanding of arithmetic symbols and processes. The right hemisphere is responsible for the visual-spatial relationship of parts, the spatial nature of arithmetic (e.g., "borrowing" or "carrying over"), verbalspatial skills, facial recognition, recognition of emotional (nonverbal) facial and postural reactions, and analysis of unusual or unknown pictures. Injuries to the tertiary area, depending on location and severity, can lead to loss or impairment of any of the above
skills, usually through the loss of the ability to effectively integrate across two or three sensory modalities. Luria postulated that "the third block of the brain, comprising the frontal lobes is involved in the formation of intentions and programs for behavior'' (1970, p. 68). This unit, which is labeled the output/planning unit, is also conceptualized as consisting of primary, secondary, and tertiary areas. The primary area of this unit is the motor output area of the brain. Signals are sent from this area, through the motor tracts of the brain, to the specific muscles needed to perform any given behavior. The secondary area is responsible for organizing the sequence of motor acts. The tertiary area of the output/planning unit represents the highest level of development and is, in many ways, dramatically different from the primary and secondary areas in terms of functions. The tasks of the tertiary area are planning, decisionmaking, evaluation, temporal continuity, impulse and emotional control, focusing of attention, and cognitive flexibility. As the frontal lobes develop, they assume dominance over the arousal unit of the brain. Thus, at about puberty, many symptoms of reticular system dysfunction may disappear as the tertiary areas achieve behavioral dominance. This provides an explanation for the clinical observation that hyperactivity associated with ADD often disappears at or around puberty. Injuries to the frontal lobes prior to adolescence are virtually impossible to detect behaviorally. Children may show "symptoms" associated with frontal lobe dysfunction (e.g., irritability, hyperemotionality, impulsiveness, rage), but these are usually the result of injury to the arousal unit during the neonatal period or the limbic system thereafter. Such symptoms may or may not improve when frontal lobe development occurs depending upon the integrity of the frontal lobes themselves. Therefore, prediction of adolescent hyperactivity based on early childhood behavior is very difficult without definite neuropathological information. For that matter, even the absence of ADD in childhood, implying an intact RAS, does not necessarily preclude the possibility of behavioral difficulties in adolescence if some form of insult has occurred. In general, although the concept that attentional difficulties may interfere with cognitive processes has been recognized for several decades, it is only within the last 10-15 years that ADD has been differentiated from learning disabilities. This new perspective seems to be a direct result of the advances within neuropsychological research. Not only has neuropsychology provided the tools for differential diagnosis
BASES OF COMMON LEARNING AND BEHAVIOR PROBLEMS
of specific learning disabilities and attentional deficits, but it has also provided the means to differentiate the causes of these disabilities (environmental versus physiological). This then allows for both valid prediction of behavior based upon brain integrity and appropriate intervention procedures.
143
Burgher, P., & Rourke, B. P. (1976). A comparison of the phonetic accuracy of spelling errors of normal and retarded readers; A four year follow-up. Unpublished study, University of Windsor. Decker, T. N., & Howe, S. W. (1981). Auditory tract asymmetry in brain stem electrical responses during binaural stimulation. Journal of the Acoustical Society of America, 69, 1084-
1090.
Conclusions At this time it seems that there is enough mounting evidence to support the neuropsychological basis of learning disability. As the most recent research has found, this can no longer be considered a homogeneous entity, but must be regarded as a category comprised of various deficits and etiologies. Therefore, the area of remediation needs to recognize these differences and provide appropriate approaches. At this time, research has just begun to address this need. Of particular concern should be the question of how the field of neuropsychology can make this emerging information available to educators and clinicians so that appropriate interventions are provided for the learning-disabled child.
References Aaron, P. G. (1981). A neuropsychological key approach. In G. W. Hynd & I. E. Obrzut (Eds.), Neuropsychological assessment and the school-age child: Issues and procedures. New York: Grune & Stratton. Aaron, P. G., Grantham, S. L., &Campbell, N. (1978). Differential treatment of dyslexia of diverse etiologies. Unpublished manuscript. American Psychiatric Association, Committee on Nomenclature and Statistics. (1980). Diagnostic and statistical numual of mental disorders (3rd ed.). Washington, DC: Author. Bakker, D. I., Moerland, R., & Goekoop-Hoefkens, N. (1981). Effects of hemisphere-specific stimulation on the reading performance of dyslexic boys: A pilot study. Journal of Clinical Neuropsychology, 3, 155-159. Boder, E. (1973). Developmental dyslexia: A diagnostic approach based on three atypical reading-spelling patterns. Developmental Medicine and Child Neurology, 15, 663-687. Boll, T. I. (1974). Behavioral correlates of cerebral damage in children aged 9 through 14. In R. M. Reitan & L. A. Davison (Eds.), Clinical neuropsychology: Current status and applications. Washington, DC: Winston. Boll, T. I., & Barth, I. T. (1981). Neuropsychology of brain damage in children. InS. B. Filskov & T. I. Boll (Eds.), Handbook of clinical neuropsychology. New York: Wiley. Boll, T. J., & Reitan, R. M. (1970). Motor and sensory-perceptual deficits in brain-damaged children. Paper presented at the meeting of the Midwestern Psychological Association.
Denckla, M. B. (1979). Childhood learning disabilities. In K. Heilman & E. Valenstein (Eds.), Clinical neuropsychology. New York: Oxford University Press. Doehring, D. G. (1968). Patterns of impairment in specific reading disability. Bloomington: Indiana University Press. Doehring, D. G., & Hoschko, I. (1977). Classification of reading problems by the Q-technique offactor analysis. Cortex, 13, 281-294. Doehring, D. G., & Reitan, R. M. (1962). Concept attainment of human adults with lateralized cerebral lesions. Perceptual Motor Skills, 14, 27-33. Dorner, G. (1980). Sexual differentiation of the brain. Vitamins and Hormones, 381, 325-385. Drake, W. E. (1968). Clinical and pathological findings in a child with a developmental learning disability. Journal ofLearning Disability, I, 9-25. Duffy, F. H., Burchfield, J. L., & Lombroso, C. T. (1979). Brain electrical activity mapping (BEAM): A method for extending the clinical utility of EEG and evoked potential data. Annals of Neurology, 5, 309-321. Duffy, F. H., Denckla, M. B., Bartels, P. H., & Sandini, G. (1980). Dyslexia: Regional differences in brain electrical activity by topographic mapping. Annals ofNeurology, 7. 412430. Fisk, J. L., & Rourke, B. P. (1978, February). Mental retardation and learning disabilities: A neurodevelopmental perspective. Paper presented at the meeting of the International Neurospychology Society, Minneapolis. Fisk, I. L., & Rourke, B. P. (1983). Neuropsychological subtyping of learning-disabled children: History, methods, implications. Journal of Learning Disabilities, 16, 529-531. Fitzhugh, K. B., & Fitzhugh, L. (1965). Effects of early and later onset of cerebral dysfunction upon psychological test performance. Perceptual and Motor Skills, 20, 1099-1100. Gaddes, W. H. (1980). Learning disabilities and brain function: A neuropsychological approach. Berlin: Springer-Verlag. Gaddes, W. H. (1981). An examination of the validity of neuropsychological knowledge in educational diagnosis and remediation. In G. W. Hynd & I. E. Obrzut (Eds.), Neuropsychological assessment of the school-age child: Issues and procedures. New York: Grone & Stratton. Galaburda, A. M. (1983). Developmental dyslexia: Current anatomical research. Annals of Dyslexia, 33, 41-53. Galaburda, A.M., & Eidelberg, D. (1982). Symmetry and asymmetry in the human posterior thalamus: II. Thalamic lesions in a case of developmental dyslexia. Archives ofNeurology. 39. 333-336. Galaburda, A. M., & Kemper, T. L. (1979). Cytoarchitectonic abnormalities in developmental dyslexia: A case study. Annals of Neurology, 6, 94-100. Geschwind, N., & Behan, P. 0. (1982). Lefthandedness: Associa-
CHAPTER 7
144
tion with immune disease, migraine, and developmental learning disorder. Proceedings of the National Academy of Sciences, 79, 5097-5100. Geschwind, N., & Galaburda, A.M. (1985). Cerebrallateralization: Biological mechanisms, associations, and pathology: A hypothesis and a program for research. Archives of Neurology, 42, 428-459. Golden, C. J. (1981). The Luria-Nebraska.Children's Battery: Theory and formulation. In G. W. Hynd & J. E. Obrzut (Eds.), Neuropsychological assessment and the school-age child: Issues and procedures. New York: Grune & Stratton. Goldman, J. 0., Thibert, R. J., & Rourke, B. P. (1979). Platelet serotonin levels in hyperactive children. Journal ofPediatric Psychiatry, 4, 285-296. Gorski, R. A., Harlan, R. E., & Jacobson. C. D. (1980). Evidence for the existence of a sexually dimorphic nucleus in the preoptic area of the rat. Journal ofComprehensive Neurology, 193,
529-539. Halstead, W. C. (1947). Brain and intelligence. Chicago: University of Chicago Press. Hynd, G. W., & Cohen, M. (1983). Dyslexia: Neuropsychological theory, research and clinical differentiation. New York: Grune & Stratton. Kinsbourne, M., & Warrington, E. K. (1963). The developmental Gerstmann syndrome. Archives of Neurology, 8, 490-501. Knights, R. M., & Viets, C. A. (1975). Effects of pemoline on hyperactive boys. Pharmacology, Biochemistry and Behavior, 3, 1107-1114. Lassen, N. A., Ingvar, D. H., & Skinhoj, E. (1978). Brain function and blood flow. Scientific American, 239, 62-71. Luria, A. R. (1964). The working brain. New York: Basic Books. Luria, A. R. (1966). Higher cortical functions in man. New York: Basic Books. Luria, A. R. (1970). The functional organization of the brain. Scientific American, 222, 66-78. MacLusky, N. J., & Naftolin, F. ( 1981). Sexual differentiation of the central nervous system. Science, 211, 1294-1303. McEwen, B. S. (1981). Neural gonadal steroid actions. Science, 2ll, 1303-1311. Nelson, H., & Warrington, E. K. (1974). Developmental spelling retardation and its relation to other cognitive abilities. British Journal of Psychology, 65, 265-274. Newcombe, F. (1969). Missile wounds to the brain. New York: Oxford University Press. Obrzut, J. E., & Hynd, G. W. (1983). The neurobiological and neuropsychological foundation of learning disabilities. Journal of Learning Disabilities, 16, 515-520. Obrzut, J. E., Morris, G. L., Wilson, S. L., Lord, J. M., & Caraveo, L. E. (1987). Brainstem-evoked response in the assessment of learning disabilities. International Journal of Neuroscience, 32, 11-823. Peters, J. E., Romine, J. F., & Dyckman, R. A. (1975). A special neurological examination of children with learning disabilities. Developmental Medicine and Child Neurology. 17,
63-78. Petrauskas, R. J., & Rourke, B. P. (1979). Identification of subgroups of retarded reading: A neuropsychological multivariate approach. Journal of Clinical Neuropsychology. I.
17-37.
Pfaff, E. W. (1966). Morphological changes in the brains of adult male rats after neonatal castration. Journal ofEndocrinology. 36, 415-416. Pirozzola, F. J. (1979). The neuropsychology of developmental reading disorders. New York: Praeger. Pirozzola, F. J., & Campanella, D. J. (1981). The neuropsychology of developmental speech disorders, language disorders, and learning disabilities. In G. W. Hynd & J. E. Obrzut (Eds.), Neuropsychological assessment and the school-age child: Issues and procedures. New York: Grune & Stratton. Raisman, G., & Field, P. M. (1973). Sexual dimorphism in the neuropil of the preoptic area of the rat and its dependence on neonatal androgen. Brain Research, 54, 1-29. Reed, H. B. C., Jr., & Fitzhugh, K. B. (1966). Patterns of deficits in relation to severity of cerebral disfunction in children and adults. Journal of Consulting Psychology. 30, 98-102. Reed, H. B. C., Reitan, R. M., & Klove, H. (1965). Influence of cerebral lesions on psychological test performance of older children. Journal of Consulting Psychology, 29, 247-
251. Reed, J. I. (1968). The ability deficits of good and poor readers. Journal of Learning Disabilities, 2, 134-139. Reitan, R. M. (1955). Investigation of the validity of Halstead's measures of biological intelligence. AMA Archives of Neurology and Psychiatry, 73, 28-35. Reitan, R. M. (1958). Qualitative versus quantitative mental changes following brain damage. Journal of Psychology. 45.
339-346. Reitan, R. M. (1964). Manual for administering and scoring the Reitan-lndiana Neuropsychological Battery for Children (Aged 5 through 8). Indianapolis: University oflndiana Medical Center. Reitan, R. M. (1967). Psychological assessment of deficits associated with brain lesions in subjects with normal and subnormal intelligence. In]. L. Khanna(Ed.),Braindamageandmental retardation: A psychological evaluation. Springfield, IL: Thomas. Reitan, R. M. (1971). Sensorimotor functions in brain-damaged and normal children of early school age. Perceptual and Motor Skills, 33, 655-664. Reitan, R. M. (1975). Assessment of brain-behavior relationships. In P. McReynolds (Ed.), Advances in psychological assessment (Vol. III). San Francisco: Jossey-Bass. Inc. Reitan, R. M. (1980). Applications of neuropsychological testing in psychiatric diagnosis. Unpublished manuscript. Reitan, R. M., & Boll, T. J. (1973). Neuropsychological correlates of minimal brain dysfunction. Annals of the New York Academy of Sciences, 205. 65-88. Reitan, R. M., & Davison, L. A. (Eds.) (1974). Clinical neuropsychology: Current status and application. Washington. DC: Winston. Reitan, R. M., & Fitzhugh, K. B. (1971). Behavioral deficits in groups with cerebral vascular lesions. Journal of Consulting and Clinical Psychology. 37, 215-223. Reitan, R. M., & Klove, H. (1968). Identifying the brain-injured child. In H. M. Robinson & H. K. Smith (Eds.), Clinical studies in reading l/1. Chicago: University of Chicago Press. Reitan, R. M., & Heineman, C. (1968). Interactions of neurological deficits and emotional disturbances in children with
BASES OF COMMON LEARNING AND BEHAVIOR PROBLEMS learning disorders: Methods for their differential assessment. In J. Helmuth (Ed.), Learning disorders (Vol. III). Seattle: Special Child Publication. Reynolds, C. R. (1981). The neuropsychological basis of intelligence. In G. W. Hynd&J. E. Obrzut(Eds.),Neuropsychological assessment and the school-age child: Issues and procedures. New York: Grone & Stratton. Rourke, B. P. (1975). Brain-behavior relationships in children with learning disabilities: A research program. American Psychologist, 30, 911-920. Rourke, B. P. ( 1976a). Reading retardation in children: Developmental lag or deficit? In R. M. Knights & D. 1. Bakker (Eds.), Neuropsychology of learning disorders: Theoretical approaches. Baltimore: University Park Press. Rourke, B. P. (1976b). Interactions between research and assessment. Journal of Pediatric Psychology, I, 7-11. Rourke, B. P. (1976c). Issues in neuropsychological assessment of children with learning disabilities. Canadian Psychological Review, 17, 89-102. Rourke, B. P. (1978). Neuropsychological assessment of children with learning disabilities. In S. B. Filskov & T. J. Boll (Eds.), Dyslexia: An appraisal of current knowledge. New York: Oxford Univeristy Press. Rourke, B. P., & Finlayson, M.A. T. (1978). Neuropsychological significance of variation in patterns of academic performance: Verbal and visuo-spatial abilities. Journal of Abnormal Child Psychology, 6, 121-133. Rourke, B. P., & Orr, R. R. (1977). Prediction of the reading and spelling performances of normal and retarded readers: A four year follow-up. Journal of Abnormal Child Psychology, 5, 9-20. Rourke, B. P., & Strang, J.D. (1978). Neuropsychological significance of variations in patterns of academic performance: Motor, psychomotor, and tactile-perceptual abilities. Journal of Pediatric Psychology, 3, 62-66. Rourke, B. P., & Strang, J. D. (1983). Subtypes of reading and arithmetic disabilities: A neuropsychological analysis. In M. Rutter (Ed.), Developmental neuropsychiatry. New York: Guilford Press. Satterfield, J. H., Cantwell, D. P., & Satterfield, B. T. (1974). Pathophysiology of the hyperactive child syndrome. Archives of General Psychiatrv, 3/, 839-844. Satz, P., Friel, J., & Rudegair, F. (1974). Differential changes in the acquisition of developmental skills in children who later became dyslexic: A three year follow-up. In D. Stem, J. Rosen, & N. Butters (Eds. }, Plasticity and recovery offunction in the central nervous system. New York: Academic Press. Satz, P., & Morris, R. (1981). Learning disability subtypes: A review. In P. J. Pirozzola & M. C. Wittrock (Eds.), Neuro-
145
psychological and cognitive processes in reading. New York: Academic Press. Satz, P., & Van Nostrand, G. (1972). Developmental dyslexia: An evaluation of a theory. In P. Satz & J. Ross (Eds.}, The disabled learner: Early detection and intervention. Rotterdam, The Netherlands: Rotterdam University Press. Selz, M. (1977). A neuropsychological model of learning disability: Classification ofbrain function in 9-14 year old children. Unpublished doctoral dissertation, University of Washington. Selz, M. (1981). Halstead-Reitan neuropsychological test batteries for children. In G. W. Hynd & 1. E. Obrzut (Eds.), Neuropsychological assessment and the school-age child: Issues and procedures. New York: Grune & Stratton. Selz, M., & Reitan, R. M. (1979). Neuropsychological test performance of normal, learning-disabled, and brain-damaged older children. Journal ofNervous and Mental Disease, I 67, 298-302. Strang, J.D., & Rourke, B. P. (1983). Concept-formation/ nonverbal reasoning abilities of children who exhibit specific academic problems with arithmetic. Journal ofClinical Child Psychology, I2, 33-39. Strang, J.D., & Rourke, B. P. (1985). Arithmetic disability subtypes: The neuropsychological significance of specific arithmetical impairment in childhood. In B. P. Rourke (Ed.), Neuropsychology of learning disabilities. New York: Guilford Press. Sweeney, J. E., McCabe, A. E., & Rourke, B. P. (1978). Logical-grammatical abilities of retarded spellers. Unpublished study, Windsor Western Hospital Centre and University of Windsor. Sweeney, J. E., & Rourke, B. P. (1978). Neuropsychological significance of phonetically accurate and phonetically inaccurate spelling errors in younger and older retarded spellers. Brain and Language, 6, 212-225. Sweeney, J. E., & Rourke, B. P. (1985). Spelling disability subtypes. In B. P. Rourke (Ed.), Neuropsychology of learning disabilities. New York: Guilford Press. Torello, M. W., & Duffy, F. H. (1985). Using brain electrical activity mapping to diagnose learning disabilities. Theory into Practice, 24, 95-99. Wada, J., & Rasmussen, T. (1960). lntracarotid injection of sodium Amytal for the lateralization of cerebral speech dominance. Journal of Neurosurgery, 17, 266-282. Wepman, J. M. (1964). Discussion of "the modality concept." In H. A. Robinson (Ed.), Meeting individual differences in reading. Chicago: University of Chicago Press. Wiig, E. H., & Semel, E. M. (1974). Development of comprehension of logico-grammatical sentences by grade school children. Perceptual and Motor Skills. /8, 171-176.
8 Measurement and Statistical Problems in Neuropsychological Assessment of Children CECIL R. REYNOLDS
The field of neuropsychology as practiced clinically has been driven in large part by the development and application of standardized diagnostic procedures that are more sensitive than medical examinations to changes in behavior, in particular higher cognitive processes, as related to brain function. The techniques and methods so derived have led to major conceptual and theoretical advances in the understanding of normal and abnormal patterns of brainbehavior relationships. Despite the apparent utility of many of the neuropsychological tests discussed in this volume, their psychometric properties leave much to be desired. Much of their utility comes from the clinical acumen and experience of their users and developers, a situation that has, historically, made clinical neuropsychology more difficult to teach than should be the case. In fact, much oftoday's practice and yesterday's theoretical advances in clinical neuropsychology stem from intense and insightful observation of brain-damaged individuals by such astute observers as Ward Halstead, A. R. Luria, Hans Teuber, Karl Pribram, Roger Sperry, and others. These superstars of clinical neuropsychology were state-ofthe-art researchers (though the state-of-the-art was often crude) to be sure, but their greatest inspirations came from their constant monitoring and informal interactions with the behavior of persons suffering from a variety of neurological trauma and disease. Halstead roamed the halls of Otho S. S. Sprague making notes as he observed behavior among braindamaged individuals; Luria gained great insights into brain function with his rather informal, sometimes CECIL R. REYNOLDS • Department of Educational Psy-
chology, Texas A&M University, College Station, Texas 77843.
impromptu, bedside examination and discussions with soldiers with head injury; Sperry and his students followed and observed a series of ''split-brain'' patients going about their daily activities, even to the point of observing some as they dressed themselves and while engaged in leisure activities with others. Major advances have occurred because of the sheer clinical acumen of these individuals. Many clinical neuropsychologists continue to evaluate patients quite profitably on the basis of observation and informal assessment. Others have devoted themselves to more purely actuarial approaches to clinical work and research (e.g., Reitan, Rourke, and Satz). Most clinicians engage the complementarity of the two approaches, a modus operandi that has made neuropsychologists more and more accepted and contributing members to medical staff in teaching hospitals and clinics. Clearly, clinical neuropsychology has been successful in earning its riches in medicine and in psychology largely due to the combination of empirical research and clinical acumen in the field. Neuropsychological techniques have infiltrated assessment in special and remedial education as well (e.g., Hynd, 1981). At the same time, many of the practices in neuropsychology, clinical and research, have been criticized extensively from inside (e.g., Parsons & Prigatano, 1978; Reynolds, 1982; Willson & Reynolds, 1982) and outside the discipl!ne (e.g., Coles, 1978; Sandoval, 1981) for a lack of attention to certain principles of research design in the field and the failure to incorporate the many advances in psychometric methods of the past 25 years. To be sure, our research methods and statistical tools have improved greatly since Halstead's early work; yet our ability (or our inclination) to apply them uniformly or to our 147
148
CHAPTER 8
best advantage certainly has not kept pace with the growth in our clinical acumen and with theoretical advances in the field. Neuropsychologists have shown an increasing interest in the educational problems of children categorized as learning disabled as well, bringing neuropsychological methods to bear on the recurring questions of neuropsychological dysfunction within this population. Clinical neuropsychological assessment of educational disorders such as learning disability offers a prime opportunity to meld theory and clinical acumen with good psychometric practice, but has not, apparently, come about. The failure to reach this coalescence in clinical neuropsychology has serious implications for the credibility and, ultimately perhaps, the survival of the clinical application of neuropsychological principles in medical and educational settings. Perhaps it is because of the youth of the field or its placement primarily in the medical setting, where good research design and statistical methods have only recently been discovered, that has retarded coalescence. Problems of statistical methods and design in test development in clinical neuropsychology have been noted with increasing frequency (e.g., Reynolds, 1982, 1986). In reviewing the Halstead-Reitan Neuropsychological Test Battery (HRNB), Dean (1985) remarked that the "manual for the HRNB lacks the basic psychometric documentation needed in interpretation. Moreover, interpretations are more dependent on the psychologist's knowledge and clinical acumen than reported psychometric properties for the battery'' (p. 645). The other major battery in the discipline, the Luria-Nebraska Neuropsychological Battery (LNNB), fares no better; as Adams (1985) remarked, "the methodological errors committed in the construction of the test [the LNNB] are both numerous and substantive" (p. 879). Other scales in common use by clinicians are equally guilty. The normative data for the Benton Test of Facial Recognition, Mirsky's Continuous Performance Test, Purdue Pegboard Test, the Wide Range Achievement Test, and numerous other measures used in neuropsychological testing are far below contemporary standards. It is a monument to the clinical acumen and tenacity of clinical neuropsychologists, and perhaps also the insensitivity of many medical practitioners to behavioral changes, that the field has survived and in fact prospered over the last 50 years. The issues to be delineated in the following pages deal primarily with pragmatic concerns that affect the clinical practice of neuropsychology in patient care certainly, but also the study of brain-behavior relationships. These issues principally revolve
around measurement problems evident in the neuropsychological literature, the resolution of which could enhance progress in research and practice in the field. The solutions are neither novel nor unknown nor are the problems restricted to neuropsychology. A number of difficulties in present practice are apparently the result of either a lack of psychometric sophistication among those in the field, an ignoring of certain well-known measurement principles, or some combination of these two reasons. The following discussion will present several examples of what may be seen as a lack of sophistication in or attention to measurement issues in neuropsychology and propose alternative procedures. Rather than employ a single battery or procedure as an example throughout, a variety were chosen to illustrate the widespread nature of the problem and not to appear to be ''picking on" any specific application. A series of statistical issues related to diagnostic research problems is next presented along with recommendations for improving this line of research as well. Lest this work appear too negative, it is worth noting that neuropsychology has emerged as a major field within psychology and that the procedures critiqued herein have been and remain useful in clinical and research domains. The clinical acumen, insight, and dedication of the practitioners who use these scales are considerable and are not being questioned. Indeed, they have moved the field substantially in many ways. Nevertheless, the fact remains that our methods and techniques could be better-by following some well-known, widely accepted methods.
Normative Data and Standardization Samples The systematic development and presentation of
normative data has received far too little attention in
neuropsychology. Perhaps this is due to the rather tedious, mundane nature of such tasks, but, nevertheless, the lack of good normative data in neuropsychology is a distinct handicap to the field. Certainly one encounters reports of "normative data" in the professional literature. However, these reports either are typically based on very small samples (some even as small asN = 10 at yearly age intervals) or do not employ normal individuals. Too much of our neuropsychological data are based on impaired individuals; we do not know enough about how normal individuals respond to most neuropsychological tests. The latter issue is more serious clinically than most clinicians realize because in addition to the
NEUROPSYCHOLOGICAL ASSESSMENT OF CHILDREN
other problems it creates, it results in a lack of items with sufficient difficulty for assessment of premorbidly high functioning individuals with less than massive neurological trauma. Most of the children with premorbid IQs or I 30 or more who suffer general cerebral trauma but lose only 20 to 25 IQ points or less can easily go undetected in neuropsychological testing, i.e., they can appear normal and go untreated or even lose existing services once these levels have been reached in the case of initial massive trauma. Without adequate normative data drawn from large-scale samplings of the population, the clinician and the researcher are also unable to assess the effects of demographic variables such as race or ethnicity, gender, and socioeconomic status on neuropsychological test performance. Demographic variables do have a significant influence on test performance on any number of tasks. Often, neuropsychologists ignore such factors during test interpretation or believe that because brain function is being evaluated, demographic variables may be irrelevant. Systematic effects of many demographic variables have been noted on numerous tasks as illustrated by even simple tasks like Coding and Digit Symbol (some of the most sensitive of all the Wechsler tasks to neurological trauma) from the Wechsler Scales where females (both black and white) consistently outscore males. Whether using a level of performance or an ipsative profile analysis, ignorance of such robust findings could mislead the clinician. Very little research has considered the influence of demographic variables on more strictly neuropsychological test results and some of the primary books in the field do not discuss the issue or its relationship to diagnosis (e.g., Golden, 1981). For such tests as the Wechsler scales, the major studies of demographic influence on scores have occurred as a function of research involving the standardization samples of these instruments (see Reynolds & K~tufman, 1986, for a review). The failure to provide good, stratified samples in the development and standardization of neuropsychological tests has been a major inhibiting factor in efforts to understand demographic influences. Other writers have reached similar conclusions. The manual for the HRNB contains no standardization or normative data, yet age and other demographic variables are correlated with the test results. This greatly complicates test interpretation for individuals (Dean, 1985). By good normative data, reference is made to the application of stratified, random sampling techniques now common, and applied to such tests as the WISC-R, the McCarthy Scales, and the K-ABC. Indeed, the recent standardization of the K-ABC
149
{Kaufman & Kaufman, 1983) is an excellent model of the development of normative data. Good standardization samples provide a reliable standard against which to judge the performance of others and have additional benefits including at least the following: {1) communications between researchers, {2) training of clinical neuropsychologists, and (3) the deflation and exposure of a variety of clinical myths. After a brief discussion of the first two of these benefits, I will tum to a more extensive presentation of the "myth deflation advantage." Communication among researchers is a difficult and expensive task but a necessary one; indeed, it is in communication among us that the foundation of the "community of scholars" must lie. It would certainly enhance the clarity of research communications in the field if a good normative reference sample were available against which research samples could be contrasted {provided other appropriate demographic variables were adequately controlled). The development of scaled scores for neuropsychological tests based on such a sample would simplify matters as well. The issue of training also is ultimately one of communication. The presence of normative data would make learning easier. Although the accuracy of the statement is not known, it has often been said that when asked about how to become a good clinical neuropsychologist, Ralph Reitan replied, "Work in the field for 30 years.'' Although this would probably work, much of this time would be spent in developing a set of "clinical norms" in one's own mind about how normal and various groups of impaired individuals perform on such tests as the HRNB. This is necessary because of the lack of a standardization sample for such scales as the HRNB and the Bader Test of Reading-Spelling Patterns (a scale clearly intended to assess developmental phenomena), and the less than adequate sample for such popular scales as the LNNB. The transmission of the knowledge and the clinical skills of neuropsychology could be greatly enhanced by the presence of accurate, high-quality normative data. Norms also have the advantage of allowing us to evaluate certain aspects of the "clinical mythology" of assessment. For some time, statistically significant Verbal-Performance IQ differences on the Wechsler Scales were believed {and, unfortunately, still are by many) to be indicative of brain damage, neurological dysfunction, or almost certainly a learning disability if a child were involved. From the standardization sample of the WISC-R, Kaufman ( 1976a) developed normative data for the frequency of occurrence of these differences. Prior to reporting these data, he took an infonnal poll of clinicians
150
CHAPTER 8
asking what they believed, on the basis of their clinical experience, the typical Verbal-Performance IQ difference would be for normal children. The response indicated a belief that other than small differences were considered unusual. A 3- or 4-point difference was the typical response. Differences of 15 points have long been thought to be clinically significant and have been used (e.g., Dean, 1978) to document the presence of a learning disability. Actual analyses of the frequency of occurrence of Verbal-Performance differences by Kaufman (l976a) with the 2200 normally functioning children in the WlSC-R standardization sample revealed a very different picture. The average difference was more than 9 points, with 12-point differences (the difference required for significance at p = 0.05) occurring with one of three children, and 15-point differences (p = .01) occurring with one of four children. This should not have been surprising as this is essentially the same distribution of difference scores that was reported 20 years earlier for the WISC, but that had gone largely ignored until Kaufman published his analyses of the WISC-R standardization sample. Note that the availability of a proper standardization sample made the investigation possible at all. This again points to the need to develop good normative data from which to evaluate one's clinical insights. Below is given an example of another recently developed index of neurological dysfunction that was not normed until very recently, as an example of how one goes about developing and reporting such data.
Gutkin and Reynolds' (1980) Norming of the Selz and Reitan Index of Neurological Dysfunction During the last several decades, hundreds of attempts have been made to develop diagnostically useful patterns based upon Wechsler subtest scores (Matarazzo, 1972). In general, these attempts have not been successful (Kaufman, 1979; Sattler, 1974). A variety of scatter indexes have been developed and investigated as potentially useful diagnostic indicators for exceptionality. The Wechsler scales, andespecially the Wechsler Intelligence Scale for Children-Revised (WISC-R) (Wechsler, 1974), have been extensively investigated with regard to utility of scatter indexes in diagnosis. Scatter indexes from the WISC-R that have been investigated include VerbalPerformance IQ discrepancies (Kaufman, 1976a;
Piotrowski, 1978; Reynolds, 1979a; Reynolds, Hartlage, & Haak, 1980), the range of subtest scores, i.e., highest minus lowest subtest score (Kaufman, 1976b; Tabachnick, 1979), and the "number of deviant signs" or number of subtests deviating significantly from the mean of all subtests (Kaufman, 1976b). Range of subtest scores in particular has attracted substantial attention as a potential technique with which normals and different pathological groups could be distinguished. Although some prior research has found statistically significant differences between diagnostic groups on this scatter index, other studies, such as Thompson (1980), have not. Even in those studies where significant differences in Wechsler subtest scatter have been found across groups of normal, brain-damaged, emotionally disturbed, and other categories of child psychopathology, the small actual differences and resulting substantial overlap of distributions have made scatter indexes such as the range of little diagnostic utility. As such results become more widely known, the search for more sensitive, sophisticated indexes of scatter has broadened and statistics such as the profile variance technique (Plake, Reynolds, & Gutkin, 1981) have been developed. Selz and Reitan ( 1979) presented another Wechsler scatter index that seemed to facilitate the accurate diagnosis of neurological dysfunction when combined with other perceptual and neurological tests. Specifically, scatter was calculated by subtracting the lowest subtest score from the highest subtest score and dividing the result by the mean of all subtests (i.e., it is calculated as the range/mean). Selz and Reitan reported three levels of diagnostic criteria in their study. A scatter index calculated with this technique that equaled or exceeded 1.0 was taken as a mild indication of neurological dysfunction. A scatter index that equaled 1.4 was interpreted as being consistent with the existence of a ''probable'' neurological problem. A scatter index equaling or exceeding 1.76 (rounded to 1.8 for the Gutkin & Reynolds, 1980, study) was viewed as part of a symptom complex indicating definite neurological impairment. One of the most common shortcomings in the Wechsler scatter pattern body of research has been the failure of investigators to validate the abnormality of various diagnostic indicants with a normal population. Often, seemingly abnormal levels of subtest scatter have been found to be quite common among normal individuals (Field, 1959; Kaufman, 1976a,b; Reynolds, 1979a) prompting the Gutkin and Reynolds (1980) study.
NEUROPSYCHOLOGICAL ASSESSMENT OF CHILDREN
The subjects for their investigation were the white (N = 1868) and black (N = 305) children from the WISC-R standardization sample of 2200 children. The characteristics of these children are described in great detail elsewhere (Wechsler, 1974). It is noteworthy, however, that these groups accurately reflect the 1980 United States census and are thus excellent, nationally representative samples of normal white and black children. The sample of 2200 was chosen to be a stratified random sample of children of the United States with sample stratification occurring by age (20 at each year between 6! and 16}), race, sex, geographic region of residence in the United States, urban versus rural residence, and socioeconomic status (as determined by occupation of the head of household). As per the procedure separately described by Selz and Reitan (1979), a scatter index was calculated for each subject by subtracting his/her lowest subtest score from his/her highest subtest score and dividing the result by his/her mean subtest score. This calculation was performed for both the 10 regularly administered subtests and the 12 total subtests comprising the WISC-R. A series of one-way ANOV As was calculated to determine if subtest scatter varied as a function of the demographic and intellectual characteristics of subjects, for the stratification variables are known to be differentially related to overall performance on the various IQ scales (Reynolds & Gutkin, 1979). Socioeconomic status is typically related to level of performance on cognitive tests, whereas race has its greatest impact on pattern of performance (Reynolds, 1981a). Specifically, subjects were grouped according to age (less than 10, 10-12, greater than 12), sex (male, female), race (white, black), place of residence (urban or rural), and Full-Scale IQ (FSIQ) (less than 85, 85-115, and greater than ll5). Because Kaufman (1976b) found significant differences in Verbal-Performance IQ scatter as a function of FSIQ, one-way ANOVAs yielding significant results were further examined with a covariance analysis with FSIQ serving as the covariate. Demographic variables that yielded significant results with covariance analysis were used to segregate the study's data. Analysis of variance on the dimensions of place of residence, sex, FSIQ, occupation of the head of household, age, and race revealed significant differences for the latter four variables. Using the FSIQ as a covariate resulted in nonsignificant differences for occupation of the head of household, but statistically significant differences remained for the dimensions of age and race. Because further examination revealed that the means at the different age levels
151
TABLE 1. Means and Standard Deviations for Scatter Index Using 10 Subtests FSIO
<8.5 Mean S.D. 85-115 Mean S.D. >115 Mean S.D. Totals Mean S.D.
Whites
Blacks
Totals
0.98 0.36
1.01 0.38
0.99 0.37
0.64 0.23
0.70 0.23
0.65 0.23
0.51 0.16
0.62 0.20
0.51 0.16
0.65 0.27
0.84 0.35
0.68 0.29
never differed from each other by more than 0.04, subsequent data analyses were broken out only according to subject race and FSIQ. Means, standard deviations, and the percentage of subjects equaling or exceeding each Selz and Reitan (1979) diagnostic criterion as found by Gutkin and Reynolds ( 1980) are represented according to subject race and FSIQ category in Tables 1-5. As indicated by the data analysis, the utility of the Selz and Reitan (1979) scatter index varies with the level of criteria employed and the characteristics of the subjects examined. The most stringent criterion (i.e., scatter index equal to or greater than 1.8) appears to set a standard that is almost ~ever reached in the normal population except for 2-5% of the subjects in the lowest IQ group.
TABLE 2. Means and Standard Deviations for Scatter Index Using 12 Subtests FSIO
<85 Mean S.D. 85-115 Mean S.D. >115 Mean S.D. Totals Mean S.D.
Whites
Blacks
Totals
1.07 0.35
1.10 0.37
1.09 0.35
0.71 0.23
0.78 0.22
0.72 0.23
0.58 0.17
0.70 0.14
0.58 0.17
0.72 0.27
0.93 0.34
0.76 0.29
152
CHAPTER 8
TABLE 3. Percentage of Subjects Equaling or Exceeding Scatter Index of 1.0 10 subtests FSIO <85 85-115 >115
12 subtests
Whites
Blacks
Whites
Blacks
49 9 0
55 15
62 14 2
64
0
21 0
Using the second most stringent criterion (i.e., scatter index equals or exceeds 1.4) also yields satisfactory results with all but the lowest IQ group. As with the highest criterion, normal subjects with IQs of 85 and above virtually never equal or exceed the Selz and Reitan index of 1.4. It is noteworthy, however, that both black and white subjects in the lowest IQ group do exceed the 1.4 criterion at a rate that calls into question the validity of this index for this particular group, unless one assumes a rather high incidence of neurological impairment in children with IQs below 85, even though most of these children were functioning normally. The most lenient of the Selz and Reitan criteria (i.e., scatter index equals or exceeds 1.0) appears to completely lack validity with the lowest IQ group. Depending upon whether one uses 10 or 12 subtests and whether the subjects were white or black, between 49 and 64% of the sample exceeded the 1.0 cutoff. Clearly, the use of this standard with this group of normal children would lead to an unacceptable number of false-positives. The middle IQ group also meets or exceed the 1.0 criteria in numbers that call the validity of the index into serious question. Only with the highest IQ group was the 1.0 standard sufficiently infrequent. Although the differences are not highly pronounced, statistically significant differences were
TABLE 4. Percentage of Subjects Equaling or Exceeding Scatter Index of 1.4 10 subtests FSIO <85 85-115 >115
TABLE 5. Percentage of Subjects Equaling or Exceeding Scatter Index of 1.8
12 subtests
Whites
Blacks
Whites
Blacks
12 I 0
18 I 0
21 I 0
23 0 0
10 subtests FSIO <85 85-115 >115
12 subtests
Whites
Blacks
Whites
Blacks
4 0 0
3 0 0
2 0 0
5 0 0
evidenced on the Selz and Reitan index between blacks and whites, with the former group consistently showing higher index scores (see Tables 1 and 2) across the entire IQ range. Even smaller, but statistically significant, differences were found as a function of the children's age. No consistent pattern emerged in regard to this variable, although the youngest group most often evidenced the highest Selz and Reitan index scores. However, the overall data from the Gutkin and Reynolds ( 1980) study indicate that the utility of the Selz and Reitan index varies substantially according to subject characteristics, especially FSIQ. As pointed out by Selz and Reitan, their scatter index is in need of cross-validation with other samples before the clinician can use it with confidence. The presentation of normative data in the detail presented here will be necessary for such batteries as the HRNB and LNNB if we are to advance not only the study of brain-behavior relationships in the normally functioning human brain but the clinical acumen of the neuropsychologist as well. Designing and conducting a normative study on such a large scale as to be useful is time-consuming and quite expensive. Few and far between are the times when money is available for the wholesale assessment of normally functioning individuals. Neuropsychologists must move actively to seek federal funding for normative studies, and test publishing houses must become convinced of the viability of neuropsychological test construction projects. Major publishing houses have been responsive to the needs of psychology in some instances resulting in large investments in test construction projects such as the Wechsler Scales. Clinical neuropsychologists must demand that neuropsychological tests meet the same psychometric standards as many other scales and move toward the development and norming of such scales. Nowhere is this more needed generally than in neuropsychology and child neuropsychology in particular.
NEUROPSYCHOLOGICAL ASSESSMENT OF CHILDREN
Reliability of Neuropsychological Measures Reliability may be the single most influential of psychometric concepts because of its relationship to all other psychometric characteristics. It is the foundation of validity, and classical psychometric theory is known as reliability theory. The problem of reliability, particularly of internal consistency reliability such as represented in Cronbach 's alpha and the various Kuder-Richardson formulas, has not been well attended in clinical neuropsychology. Observed test score variance can be divided into two components, true score variance and error variance. Only true score variance is "real," systematic, and related to true differences among individuals. Only true score variance can be shared or related between two variables; thus, we see that the criterion-related validity of any test is restricted as a function of the square root of the product of the reliabilities of the two measures (i.e., (rxxryy)l). Reliability of neuropsychological measures is an area that has received attention in the literature as an area of special need (e.g., Parsons & Prigatano, 1978; Reynolds, 1982). Reliability of neuropsychological measures is equally important to individual diagnosis and to research, as reliability will influence the likelihood that any experimental or treatment effects will be detected. In short, reliability is the foundation on which validity, the most important of measurement concepts, is built. Nonetheless, even in the research literature, reliability data are seldom presented and the most frequently used of the various batteries, the HRNB, does not even have a discussion of reliability in its manual. Reliability of the LNNB is reported but is based on highly heterogeneous groups, across a too wide age span, and is likely spuriously inflated. The BoderTest of Reading Spelling Patterns, developed by a pediatric neurologist as a neuropsychological measure of educational deficits, provides a good example of how not to assess reliability of neuropsychological measures, although the authors must be acknowledged for at least attending to the issue. In studying the reliability of the Boder Test, Boder and Jarrico ( 1982) reported on several aspects of reliability. Test-retest (represented as r 12) or stability of scores is reported for 2-month and 1-year intervals. The sample size for the 2-month study was 50 and for the long-term study, N was 14. Three aspects of the test (the Boder not actually being divided into subtests) were evaluated, those yielding scores for Reading Level, Correctly Spelled Known Words, and the number of Good Phonetic Equiv-
153
alents produced by the child. Both test-retest reports of reliability are based on wide age spans; for Reading Level an r 12 of 0.98 is reported for ages 6-9, and 0.96 for ages 10-15. These rs are inflated to an unknown degree by the correlation of Reading Level with age. The use of a wide age range with rs based on raw scores or other age-related scores such as grade equivalents and age equivalents is a common method of exaggerating the observed reliability of test scores. When corrected for the confounding with age, which is certainly correlated to some substantial degree with Reading Level, these reliabilities are far less impressive. The original r 12 for correctly spelled known words is but 0.64 for ages 6-9 and 0.84 for ages 10-15. For Good Phonetic Equivalents, the values are 0.89 and 0.85. After correction for spurious correlations with age, it would be surprising to find these reliabilities within any commonly accepted range for diagnosing individual cases. Also, the Ns are extremely small (ages 6-9, N = 27; ages 10-15, N = 23) for making any decision about the stability of scores on the Boder. The long-term rs were 0.81 for Reading Level, 0.62 for spelling, and 0. 79 for Good Phonetic Equivalents, values that were cleary unacceptable for individual cases, but based only on N = 14, an unacceptable sample size in any case. Internal consistency (corrected split-half) reliability estimates are reported for 46 cases randomly selected (by Boder & Jarrico's description) from Roder's private patient files. If truly random, the age range of these 46 children is 6 to 18 years. This tremendous age range greatly inflates internal consistency estimates and is unacceptable for assessing reliability (Willson & Reynolds, in press). Though considerably inflated, the values were 0.97 for Reading Level, 0.82 for spelling (known words), and 0. 92 for Good Phonetic Equivalents. The degree of spurious inflation is indeterminate. The errors made in the estimation of the reliabil- · ity of the Boder Test are serious but are all too common, not only in neuropsychology but in several areas of testing (e.g., see Reynolds, 1983). Lack of attention to standard psychometric methods seems all too rampant in clinical neuropsychology and is retarding developments in the field. One other piece of relevant data is reported in the Boder Test manual's reliability section-the agreement on diagnosis of one of three reading disorders for the two testings with the test-retest reliability samples. Chi-square tests were used, appropriately, to evaluate changes in classification across testings, yet even these results are interpreted improperly. According to Boder and Jarrico (1982, p. 95), when the chi-square is evaluated, "a significant result shows high agreement be-
154
CHAPTER 8
tween the classifications at the two testings. ' ' Rather, the significant chi-squares only show a statistically significant relationship between the two classifications; that is, the two sets of classifications were not independent. The actual number of children given the same classification on each testing was not reported and is necessary for a more proper interpretation of these results. The reliability of the Boder Test remains an open question specifically as does the reliability of the most popular of neuropsychological tests generally. At first glance, this seems attributable to the lack of presence of the major test publishers in neuropsychology. The HRNB is essentially an inhouse production of Reitan; the Boder Test is published by Grune & Stratton and is their only test product. The LNNB, on the other hand, is produced by Western Psychological Services, one of the 10 largest test publishers in the country. Tests such as the revision of the Wechsler Memory Scale by the Psychological Corporation and the Kaufman Assessment Battery for Children (American Guidance Service), put out by the major houses involved in the development and publication of individually administered tests, fare much better in the appraisal of their reliability and their corresponding methods of estimation. Authors are ultimately responsible for the psychometric care used in devising their tests, and closer attention to the details of reliability seems clearly necessary. Because the validity of a test is restricted as a function of the square root of the product of the reliability coefficients of the test and the criterion [i.e., the theoretical limit placed on a test's validity coefficient is equal to (rxxryy)i]' one method of improving the validity of existing neuropsychological tests then obviously is to work toward enhancing the reliability of these scales. Too frequently, neuropsychologists rely on the "clinical" nature of certain tests and procedures to the extent that such important concepts as reliability are overlooked. Reliability of our testing and assessment procedures is equally important to research in neuropsychology. The most direct implication of reliability for research is in the detection of experimental effects. As reliability decreases, so does the likelihood that a significant effect will be found in any experimental or clinical treatment. Marcel Kinsbourne has made reference to just this problem in consistently detecting hemispheric differences on certain tasks under a specified set of conditions ( 1981 , personal communication). As anyone acquainted with the neuropsychology literature will be quick to recognize, theresults of research employing dichotic listening procedures and tachistoscopic split-visual field presenta-
tion methods are not in great agreement. Dichotic listening and split-visual field methods are both very unreliable from a purely psychometric perspective. The reliabilities reported (albeit infrequently) in the literature are seldom better than 0. 5 to 0. 65. The reliability problems here possibly lie with the techniques themselves but a more likely problem seems to be the stimulus materials, i.e., the test that is presented through these methods. The proper application of traditional psychometric methods in the construction of tests to be presented through these methods would undoubtedly enhance the reliability of these techniques. Increases in the reliability of neuropsychological measures could increase the discriminability of the tests in studies of differential diagnosis as well. Error variance cannot contribute to the general problem of distinguishing among clinical groups, although, as reviewed later in this chapter, error variance can contribute spuriously in single studies without internal replication. Although increasing reliability will most certainly not alleviate all of the interpretive problems existing in this literature, it is better not to base arguments over the interpretation of data on what is essentially error variance and little else.
Scaling Problems in Neuropsychological Testing Children are in a constant state of development and change in many ways but perhaps most dramatically in their neurological and higher cortical development. Children are acquiring knowledge at the most rapid pace of their lifetime and their reasoning processes and insights into their learning grow in a dramatic manner. All are nevertheless moving at an uneven pace. Consequently, the scaling of any tests or measurement devices designed to aid the assessment of brain-behavior relations is crucial. This is true regardless of whether one takes a "key approach," looks simply at level of function, or assesses profiles of performance. The scaling of neuropsychological tests has been sporadic, with some scaled well, some poorly, and a significant number not at all. Even when scaling is handled well from a technical perspective, the quality of the standardization sample providing the estimates of population parameters from which standard scores are subsequently determined will influence the utility of the derived scores. Raw scores, e.g., the number correct, a time to completion, or a number of errors, are problematic
NEUROPSYCHOLOGICAL ASSESSMENT OF CHILDREN
but common. The HRNB does not provide any transformed scores. Without standard score transformation, it is difficult to make any meaningful comparisons of scores. The dominant approaches to interpretation of the HRNB include assessing levels of performance and contrasting performance among the various tasks. Raw scores cannot be compared or assessed directly for a variety of reasons, the most potent being the lack of comparability of the raw score distributions among the tasks of the battery and for any one task across age. As Dean (1985) noted, in reviewing the HRNB, "without standard score transformation data, it is difficult to make any meaningful comparison between scores on individual tests" (p. 645). The use of raw scores is not necessarily wrong and is typically superior to the use of inaccurate score transformations. Appropriate use of raw scores does require extensive work and numerous calculations on the part of the person interpreting the test scores. The LNNB provides standardized or scaled scores in the form of the familiar T-score (mean = 50, standard deviation = 10) although the scaling is questionable due to the shape of the distributions obtained and the sample used in their derivation. Other attempts to scale neuropsychological measures have been made but use inappropriate scales, particularly age or grade equivalents (GEs). Neuropsychological reports, even textbook examples, often contain performance reported in GEs. The BoderTest can again provide an illustration of some of the problems with scaling as practiced in clinical neuropsychology. The Boder Test does not actually provide any type of standard score although the test's authors treat the various scores as though they are standardized or scaled scores. The Boder Test provides a Reading Level (RL) (analogous to aGE), a Reading Quotient (RQ), and a Reading Age (RA). Without adequate normative data, which the Boder does not possess, these scores are not very meaningful. Even if carefully normed, using state-of-the-art methods, these scores have serious limitations and should be used with extreme caution if at all and never as the featured scores for any contemporary scale, neuropsychological or otherwise.
RLandRA The RL and RA of the BTRSP have similar problems. RL has the greatest difficulties even if calculated on the basis of good normative data, so the RL (the Boder Test's analogue of aGE) will be featured here. Given the interdependence of the RL and the RA as calculated on the Boder Test, their problems are almost identical.
155
The GE, i.e., RL, of the BTRSP is based on the grade level at which the words from the reading lists of the Boder are estimated to be introduced into the curriculum, and assumes that half of the children master these words for reading and spelling. True GEs are based on actual student knowledge of the curriculum content as reflected in mean scores on achievement tests. When content is introduced and when it is actually mastered by 50% of the pupils may not be closely related. Actual performance must be assessed. GEs as a score on which to base decisions about individual pupils have serious deficiencies that have been presented in detail in a variety of sources (e.g., Angoff, 1971; Reynolds, 1981a, 1982; Thorndike & Hagen, 1977). Though frequently treated as a standard score, GEs are not standard scores, and attempts to standardize them (Burns, 1982) have been largely unsuccessful (Reynolds & Willson, 1983); and indeed the true meaning of the GE is often distorted if understood at all. Most of the problems with GEs can be traced to one of two factors, or both: (1) GEs are calculated independent of the dispersion or distribution of scores about the mean, and (2) the regression of the age, grade, and raw score is nonlinear, and varies across subject matter within grade as well as across grade within subject matter or content areas (AEs have analogous problems). Essentially, this tells us the GEs are on an ordinal scale of measurement and not an interval scale as so frequently interpreted. This makes many common uses of GEs entirely inappropriate. Boder and Jarrico (1982, p. 5) defined significant reading retardation as performance two years below grade level for age according to the Boder Test RL. Other diagnoses are dependent upon the RL and its discrepancy with expected level of performance as well. The "two years" criterion for a reading disorder has been a common ground for diagnosis for some time and only recently abandoned (see also Reynolds, 1984). The use of GE scores at a constant discrepancy level irrespective of actual grade placement produces considerable irregularity for diagnosis, however. The distortions in interpreting discrepancies between GE scores and grade placement are readily apparent in Table 6, which was developed from data available in the norms or technical manuals of the Wide Range Achievement Test (WRAT), Pea-· body Individual Achievement Test (PlAT), Woodcock Reading Mastery Test (WRMT), and the Stanford Diagnostic Reading Test (SORT). As is typical ofGE scores, some occasional interpolation was necessary to derive the exact values in Table 6. It is apparent, however, that a third grader who reads
156
CHAPTER 8
TABLE 6. Standard Scores Corresponding to Performance "Two Years below Grade Level for Age" on Four Major Reading Testsa Standard scoresb Grade placement
1.5 2.5 3.5 4.5
Two years below placement
Wide Range Achievement Test
Pk.5 K.5
65
1.5
69 73 84 88 86 87 90 85
5.5 6.5
2.5 3.5 4.5
7.5 8.5
5.5 6.5
9.5 10.5 11.5 12.5
Peabody Individual Achievement Test"
Stanford Diagnostic Reading Teste
72 64
9.5
85
75 85 88 89 91 93 93 93
10.5
85
95
7.5 8.5
Woodcock Reading Mastery Testd
77 85 91 94 94 96
95 95 95
64 64 77 91 92 93
95 95 92 92
a Adapted from Reynolds ( 1981 a). bAll standard scores have been converted for ease of comparison to a common scale having a mean of I00 and a standard deviation of 15.
d"fotal test.
"two years below grade level for age" has a much more severe problem than say a seventh or eighth grader reading two years below grade level. In fact, a twelfth grader with an IQ of 90 on a Wechsler scale and reading two years below grade level for age has no reading problem at all, but rather reads at a level slightly higher than what might be expected. Standard scores are by far the more accurate representation of an individual's achievement level than GEs because they are based not only on the mean at a given level but also on the distribution of scores about the mean. Thus, in the case of deviation standard scores, such as the Wechsler IQs, the relationship between standard scores is constant across age, and there are no excuses for the failure to provide such scores. Certainly the Boder Test provides no rationale for the lack of standard scores or even the preference forGEs. Neither do most clinical neuropsychologists have an adequate rationale for continued use of AEs and GEs in reports or in their application to profile analysis. GEs are also inappropriate for use in any other sort of discrepancy analysis of an individual's test performance or key or profile analyses for the following reasons: 1. The growth curve between age and achievement in basic academic subjects flattens out at upper grade levels. This can also be observed in Table 6 where it is seen that there is very little change in
standard score values corresponding to two years below grade level for age after about grade 7 or 8. In fact, GEs have almost no meaning at this level, for reading instruction typically stops by high school and GEs are really only representing extrapolations from earlier grades. An excellent example of the difficulty in interpreting GEs beyond about grade 10 has been provided by Thorndike and Hagen ( 1977) using an analogy with AEs. Height can be expressed in AEs just as reading can be expressed in GEs. However, although it might be helpful to describe a tall firstgrader as having the height of an 8}-year-old, how does one then characterize the 5' 10" 14-year-old female, for at no age does the mean height of females equal 5' 10"? Because the average reading level in the population changes very little after junior high school, GEs at these ages become virtually nonsen· sical, with large fluctuations in GEs sometimes resulting from a raw score difference of 2 or 3 points on a 100-item test. 2. GEs assume that the rate of learning is constant throughout the school year and that there is no gain or loss during summer vacation. 3. As partially noted above, GEs involve an excess of extrapolation, especially at the upper and lower ends of the scale. However, because tests aie not administered during each month of the school year, scores between the testing intervals (often a full year) must be interpolated on the assumption of con-
NEUROPSYCHOLOGICAL ASSESSMENT OF CHILDREN
stant growth rates. Interpolation between extrapolated values on an assumption of constant growth rates is at best a highly perilous activity. The assumption underlying score derivations on the Boder Test that each word read correctly represents 2 months of academic achievement is even more perilous, and most likely cannot be substantiated. For the Boder Test, this adds to the error and the unfounded assumptions already present in properly derived GEs, which the RL of the Boder Test is not. Popular achievement tests in neuropsychology have similar problems. The front of the protocol for the WRAT, for example, notes that standard scores only and not the so-called grade rating (the WRAT's GE) should be used for interpretive purposes. 4. Different academic subjects are acquired at different rates and the variations in performance differ from one content area to another. As a consequence, "two years below grade level for age" may be a much more serious deficiency in mathematics than, say, in reading comprehension. The degree of academic deficit in Reading versus Spelling, as on the BoderTest, denoted by the "two years" marker, will differ as well. 5. GEs exaggerate small differences in performance between individuals and for a single individual across tests. Some test authors (e.g., Jastak & Jastak, 1978) even provide a caution on test record . forms that standard scores only, and not GEs, should be used for comparison purposes.
RQ The Boder Test also provides an RQ calculated in accordance with one of three formulas offered in the Boder Manual, with the choice of derivation given to the examiner. Giving a choice of three formulas to the examiners is problematic in itself. Though some general guidelines are provided concerning when to use each formula, the choice is left to the examiner, and it is entirely probable that, faced with a similar or even identical set of scores, different examiners will arrive at different RQs; the same examiner may well fall to the same plight over time, being inconsistent in the choice of formulas given a common set of circumstances. However, this is one of the more minor problems with the Boder Test RQ. The RQ is derived from a faulty score to begin with, the RA, as noted in the previous section. The RQ, 1 as calculated to be (RA/CA) x 100, will not have a constant standard deviation across age and is not a 1CA,
Chronological age.
157
standard score as thought by many. It is conceptually the antiquated notion of a ratio IQ that was abandoned many years ago in favor of more refined standard score systems. The standard deviation of the first version of the RQ will almost certainly range from at least 10 to 29, leading to the confusing (and unaccounted for in the Boder Test's diagnostic system) state wherein, depending upon age, RQs of 80 and 90 represent the same overall level of performance at different ages. The actual range of standard deviations could be larger or smaller; whichever it turns out to be is less important than the fact that the standard deviation will not be constant across age and that the standard deviation at any age is unknown. Use of either of the two alternative formulas given below for calculating the RQ is even more problematic. These formulas 2 2RA RQ = MA + CA X 100 RQ =
MA
3RA
+ CA + Grade
(l)
Age
X
100
(2)
are quite similar to expectancy formulas proposed by the U.S. Office of Education (1976) in their attempt to define severe discrepancies between aptitude and achievement. Commentary on such formulas has shown them to be grossly inadequate for use in any kind of normative reporting or discrepancy analysis (e.g., Algozzine, Forgnone, Mercer, & Trifiletti, 1979; Cone & Wilson, 1981; Danielson & Bauer, 1978; Hanna, Dyck, & Holen, 1979), and far more sophisticated approaches are needed (Reynolds, 1984). The standard deviation of the scores derived from these formulas will also vary and is unknown. The same number of children will not be identified at each IQ level or each age level using the BTRSP classification rules. There is no established validity for either formula; they are only intuitive in their appeal. Given the problems of age- and grade-based equivalency scores and the amount of severe criticism they have received in the literature, it is difficult to imagine a justification for their use in place of standard scores. Certainly standard scores should be provided at a minimum with AEs and GEs and related derived scores (e.g., RQ) provided as supplementary if at all. Because AEs and GEs are representing only ordinal scale data (and thus cannot be averaged or otherwise manipulated with any confidence except 2
MA, Mental age.
158
CHAPTER 8
under special conditions), it is particularly important that these scores not be used for comparative purposes. Standard scores could not be reported for the Boder Test because there are no normative data on which to base these calculations.
Ratios and Quotients Probably the best known of all scores to the layperson is the ratio IQ, originated for use with the Binet scales early in this century. As every introductory psychology course student knows, IQ = (Mental Age/Chronological Age) x 100. This forms the ratio or quotient from which the designation IQ was derived. Such ratios or quotients have numerous psychometric problems and are no longer used by the major test publishers but do persist in certain areas of neuropsychology, even to the point of developing ratios of "hold" to "don't hold" subtests for estimating premorbid functioning. Such ratios are nonsensical for most interpretive purposes, however, and lead primarily to confusion. Although they may be used to rank individuals who take a common test no comparisons beyond rank on the common mea~ sure ar~ possible-including profile analysis or any compansons across tests. The so-called ratio IQ is a l"l!tio. of ?umbers with radically different underlying dtstnbuttons and mathematical properties. Chronological age is a ratio scale of measurement. Mental ag~ is on an ordinal s~ale of measurement. Creating a l"l!tto of two such dtsparate scales is a conceptual mghtmare of some proportion. The standard deviation of the distribution of such ratios will also vary across age. The 1937 Stanford-Binet Intelligence Scale, which yielded such a ratio IQ, showed a stan?ard deviation that ranged from about 9 to 32 dependmg upon the age of the individual assessed. The familiar standard deviation of 16 used then, and now, by the Binet Scales, was the average standard deviation across ages 2 years to about 16 years. Gross inaccuracies of interpretation are facilitated by such scales and they are not standard or scaled scores in any sense of contemporary uses of these terms. Various ratio scores and quotients such as the early IQ remain in use in neuropsychological assessment but do not possess the properties of standard scores that make the latter so useful in all areas of testing and assessment.
Standard or Scaled Scores The primary advantage of standardized or scaled scores lies in the comparability of score interpreta-
tion across age. By standard scores is meant scores of the Wechsler Deviation IQ genre, referred to more properly as age-corrected deviation scaled scores. This designation is used because the mean and the standard deviation of the scaled score distribution are reset or rescaled periodically, typically every 2 to 4 months at preschool ages and every 4 to 6 months therea_fter until the adult years when much larger age groupmgs may be used. Standard scores of the deviation IQ type have the same percentile rank across age, for they are based not only on the mean but the variability in scores about the mean at each age level. For example, a score that falls two-thirds of a grade level below the average grade level has a different percentile rank at every age. Standard s~ores are more accurate and precise. When constructmg tables for the conversion of raw sc~res into standard scores, interpolation of scores to amve at an exact score point is typically not necess_ary '.whereas t?e opposite is true of GEs. ExtrapolatiOn ts also typtcally not necessary for scores within three standard deviations of the mean, which accounts for more than 99% of all scores encountered. Standard scores are on an equal interval scale in many cases (see Gordon, 1984), making profile analysis possible across subtests of a common scale. Ipsative analysis of performance only makes mathematical sense with an interval or higher scale of meas~rement. Score comparisons among different battenes or subparts of different measuring devices are also possible provided the reliability of each measure is known and, for some purposes, the correlation between the various pairs of scores must also be known. If ~e ~li~bility coefficients are comparable, the score dtstnbuttons are normal, and the percentile rank of scores is known, Table 7 can be used to place scores on a commonly expressed metric, i.e., a scale having the same mean and standard deviation. The choic~ of s~dard score scales is often arbitrary but is sometimes dtctated by the standard deviation of the raw score distribution, such that the score points should not be artificially spread over too many stand~d score points nor should too many raw score pomts be collapsed into a single scaled score point. Such problems are usually avoided by the choice of an appropriate scale and finding a suitable scale is easy_en~ugh that it is seldom a serious problem in the apphcatton of scaled scores to most practical problems of assessment. There are few instances when other score systems are superior to scaled scores, and methods are now available for the use of scaled scores even at the most extreme points in the distributions of intelligence, achievement, and other special abilities (e.g., see Reynolds & Clark, 1985, 1986).
NEUROPSYCHOLOGICAL ASSESSMENT OF CHILDREN
159
TABLE 7. Conversion of Standard Scores Based on Several Scales to a Commonly Expressed Metric X=O S.D.= I
X=-100 S.D.= 15
X= 100 S.D.= 16
X= 100 S.D.= 20
X=500 S.D.= 100
X= 50 S.D.= 10
X= 50 S.D.= IS
X=36 S.D.=6
X= 10 S.D.=3
Percentile
2.6 2.4 2.2 2.0 1.8 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0 -0.2 -0.4 -0.6 -0.8 -1.0 -1.2 -1.4 -1.6 -1.8 -2.0 -2.2 -2.4 -2.6
139 136 133 130 127 124 121 118 115 112 109 106 103 100 97
142 138 135 132 129 126 122 119 116 113 110 106 103 100 97
152 148 144 140 136 132 128 124 120 116 112 108 104 100
760 740 720 700 680
76 74 72 70
89 86 83 80 77 74 71
52 51 49 48 47
>99 99 99 98 96 95 92 88 84 79 73 66 58
96
580 560 540 520 500 480
18 17 17 16 15 15 14 14 13 12 12
94
94 90
92 88
460 440
84
80 76 72 68
91 88 85 82 79 76 73 70 67 64
61
87 81 78 74 71 68
65 62 58
84
660 640
620 600
420 400
380 360 340
64
320
60 56 52
300
48
280 260 240
68 66 64
62 60 58 56 54 52 50 48 46
44 42 40 38 36 34
32 30 28 26 24
68
65 62 59 56 53
so
47 44 41 38 35 32 29 26 23 20 17 14 II
46
44 43 42 41 40 38 37 36 35 34 33 31 30 29 28 26 25
24 23 21 20
II II
10
9 9 8 8 7 6 6 5 5 4 3 3 2
rank
so
42
34
27 21 16 12 8 5 4 2
methodological and statistical problems in clinical neuropsychology-one that, oddly enough, is nearly the opposite of the preceding discussions. In differential diagnosis, rather than being naive about the One of the major problems of psychologists in underlying psychometrics or scaling, the tendency professional practice has historically been that of dif- has been to perform analyses that are too sophistiferential diagnosis of mental disorders. In the area of cated for the data. clinical neuropsychology, differential diagnosis has To their credit, researchers have, over the last been of particular importance. As Golden (1981) decade, brought to bear the most sophisticated statispointed out, one of the major areas of research in tical methodologies directly on the problem of diagclinical neuropsychology has had as its purpose the nosis and classification of mental disorders in the development of clinical tests and procedures to dif- form of various multivariate analytical techniques. In ferentiate reliably between brain-injured and neu- the quest to provide accurate diagnosis of neurologically intact individuals and to separate brain- rological disturbances, a large set of behaviors is injured groups into subsamples according to loca- typically assessed. Rourke (1975), in discussing tion, cause, time of onset, and, in some cases, prog- more than two decades of research in differential nosis. Many neuropsychologists still earn a consider- diagnosis, indicated that children referred to his laboable portion of their "keep" differentiating organic ratory are typically administered "the WISC, the from nonorganic psychiatric referrals and evaluating Peabody Picture Vocabulary Test, the Halstead Neuthe nature and extent of lesions for the neurology ropsychological Test Battery for Children, the Reitan service. This does bring out another major area of Indiana Neuropsychological Test Battery, the Wide
Differential Diagnosis: Determining Membership in Clinical Populations
160
CHAPTER 8
Range Achievement Test, an examination for sensory-perceptual disturbances, the Klove-Mathews Motor Steadiness Battery, and a number of other tests for receptive and expressive language abilities'' (p. 912). Multivariate classification techniques are very powerful in the determination of group membership. Unfortunately, with such a large set of variables, small numbers of subjects can all be grouped and classified purely on the basis of random or chance variation that takes maximum advantage of correlated error variances. Thus, the need for large numbers of subjects in such research is a crucial one. In the study of clinical disorders, however, one is frequently limited to relatively small samples of design. Although most researchers acknowledge this difficulty, few realize the devastating effects of subject/variable ratios approaching I on the generalizability of studies of differential diagnosis. This is not to say that excellent studies have not been done. Studies of discriminability by Satz and his colleagues (e.g., Satz & Friel, 1974; Satz, Taylor, Friel, & Fletcher, 1978) use large numbers of variables but have considerable subject populations. LargeN studies of clinical populations are the exception rather than the rule. Willson and Reynolds ( 1982) evaluated the effects of small samples on the validity of research attempting to discriminate among clinical disorders on the basis of neuropsychological test performance and have reported on many of the statistical problems that seem to plague the area.
Some Statistical Considerations In predicting group membership from a set of variables (e.g., neuropsychological test scores) there are several considerations. First, procedures that use samples of the target populations involve sampling error in the estimation of the relationships being examined. This means that results are expected to fluctuate from sample to sample due to the random differences inherent in the samples. The usual measure of prediction is the squared multiple correlation (R 2). In applying results of a particular sample to a second sample, R2 is expected to decrease because correlation is a maximizing operation-R2 was made as big as possible for the first sample capitalizing on correlated error variances whenever possible. It is unlikely the same fit of the data will occur in a second sample. Thus, rules for classification derived from a particular sample cannot necessarily be expected to generalize to any other sample without a cross-validation effort to demonstrate such an effect.
A second consideration in prediction occurs when the prediction used a strategy for selecting a small number of variables from a much larger initial set of variables (e.g., Purisch, Golden, & Hammeke, 1979). In the same, some correlations underestimate the population value and others overestimate it. In stepwise regression or discriminant procedures and related multivariate methods, the overestimates are always chosen. When a large number of predictors is available, stepwise procedures maximize the chance of selecting random or near-random predictors. These are variables that do not predict well in the population but by chance correlate highly with the outcome in the particular sample being used. The degree of decrease in R 2 from sample to population can be estimated. The most commonly used estimate is from Wherry ( 1932; see also Lord & Novick, 1968, p. 286) and is R~
= 1 - (1 - R2)(N- 1)/(N- K- 1)
(3)
N is the number of observations, K the number of predictors, R 2 the observed squared multiple correlation between outcome and predictors, and R~ the population squared multiple correlation. This formula holds for either multiple regression or discriminant analysis. Formula (3) has been widely cited. A review by Cattin ( 1980) suggested that for small Nand large K, another approximation should be used:
R2
= (N-K-3) p4 + p2 (N-2K-2) p 2 + K
(4)
where N-3
p2 =I- N-K-1 (1-R2) [
20 - R 2 )
8(1 - R2 ) 2
1+ N-K-1 + (N-K+1)(N-K+3)
]
Although R2 is biased, the amount of bias is on the order of .01-0.02 for N = 60 and K =50. Of special interest is the case where there are more predictors than people. In equation (4), the shrunken R2 may become negative or greater than 1.0. What this really means is that mathematically with more predictors than observations of the outcome, there is no unique solution to a best prediction. In discriminant analysis· this may result in perfect classification entirely at random by the predictors. Mathematically, this results from having more parameters to estimate than data points. Either one is
NEUROPSYCHOLOGICAL ASSESSMENT OF CHILDREN
forced to make enough side conditions to constrain the solution or one accepts a solution that results from a particular order of entering predictors. As it is mathematically impossible to estimate all regression coefficients, there will be
K_ K! N- N!(N-K!) different solutions that would provide perfect classification but would not generalize to any other samples. In particular, it may be quite likely to find a solution that maximizes R2 based entirely on chance correlations if there are enough correlations from which to choose. Even when there are fewer predictors than subjects, the shrunken R2 estimate will rapidly approach zero as the number of predictors becomes a significant proportion of the number of subjects. When small samples of subjects are involved, as with many neuropsychological studies, the use of a large number of tests as predictors frequently fulftlls this condition. Multiple regression and discriminant analysis have been discussed interchangeably to this point, but some distinctions need to be made about them. Formally, they are identical in two-group prediction, e.g., brain-damaged versus non-brain-damaged. For more than two groups, discriminant analysis must be used. There have been a number of different classification rules proposed using discriminant analysis. These pertain to assumptions about prior probabilities for population composition and about homogeneity of within-group covariances. In any case the R of relationship between predictor and between group distance is computed. It is a canonical correlation (see Cooley & Lohnes, 1971, p. 249). Because there may be more than one discriminant function, there will be a canonical R for each. Their squares do not necessarily add together to get a total R2, as there may be redundancy between functions (Cooley & Lohnes, 1971, p. 170), but their squared sum is the maximum possible R 2 •This may be useful as a liberal estimate because if it can be shown that fl.2 is near zero, there is no need to estimate the stu~y·s R2, because it will produce an even smaller estimate A2 of RP. For two groups, multiple regression and discriminant analysis yield the same results. For more than two groups, the canonical R2 is still useful as an approximation to a multiple regression R2. This interpretation can be a useful one but is rare in clinical neuropsychology. as multiple group discriminant
161
analysis is so seldom applied by clinical researchers. The omission is significant in that. although researchers in neuropsychology may not be familiar with or do not apply this technique for other reasons, it may be quite useful in discriminating among several populations.
The Willson and Reynolds Examples of Classification Problems To illustrate the problems !hat can be created by these statistical considerations, Willson and Reynolds (1982) examined all articles in three journals (Psychology in the Schools, Journal of Consulting and Clinical Psychology, and Clinical Neuropsychology) for the years 1979-1981. They selected studies that used test batteries, socioeconomic or demographic variables, or a combination of these variables to predict clinical group membership and that could help illustrate the difficulties of such work. Nine studies were found. The studies are listed in Table 8 (from Willson & Reynolds, 1982). Also listed are sample sizes (N), total number of predictors used in stepwise procedures (KT), and number of predictors used in the final or discriminant equation (KF). In one case, RF was determined indirectly through a 2 x 2 classification table that was reported in the studies. The tetrachoric correlation was computed and squared (see Glass, 1978, for a discussion of estimating correlation effects). Table 8 lists several other statistics that represent estimated values of R~ and their significance via their associated F-statistic. The statistic At represents the estimated R shrunken by equation (3) to account for all the predictors originally considered. Because in no study was the overall reported for all predictors, it was necessary to use the R~ based on the final regression. Thus, Rt underestimated the shrunken R~ to some degree. Its upper bound is given by R~ the shrunken estimate based on the number of predictors actually used. This is an overestimate of the actual shrunken R2. A second set of ~tatistics was calculated from the nine studies to estimate loss of classification power due to shrinkage and is presented in Table 9. For the reported R~ and the Rt the t-statistic equivalent was computed according to
Rt
t
=
[
R2JK
(l-R2 )/(N-K-1)
] 112
(5)
Although most studies had only two groups, in Selz and Reitan (1979), the t-statistic was based on a
162
CHAPTERS
TABLE 8. Summaries of Prediction Studies from Three Special Population Journals Study
Dean (1978) Selz & Reitan (1979)" Wallbrown, Vance, & Pritchard (1979) Purisch, Golden, & Hammeke (1979) Taylor & Imivey (1980)
Dunleavy, Hansen, & Baade (1981) Fuller & Goh (1981) Golden, Moses, Graber, & Berg (1981) Malloy & Webster (1981)'"
(a) (b) (a) (b) (c) (d) (e) (f)
(a) (b) (a) (b)
Sample size
Total number of predictors
Number of predictors used
120 75 200 100 100 30 30 30 30 30 30
14 37 8 282 14 16 3 16 3 16 2 37 22 11 11 14 14
4 37 3 40 14
24 80 60 120 36
36
5 I
5 1 2 1 3 12 2 2 14 14
R~
•2
•2
(reported)
RT
RF
0.25* 0.57 0.19* 1.00* 0.88* 0.44* 0.08 0.30 0.14* 0.25* 0.11 0.82* 0.38* 0.55* 0.68* 0.57
0.09 0.57 0.13
0.21* 0.57 0.17*
0.94*d
()b
0.84* 0.00 0.00 0.00 0.02 0.00 0.03 ()b
0.05 0.37* 0.62* 0.57 0.94*
()b
0.84* 0.26 0.05 0.10 0.12 0.22* 0.09 0.79* 0.19 0.54* 0.68* 0.57 0.94*
"Trinomial classification table was reported; it was converted to a binomial (normal versus brain-damaged or ID) and the tetrachoric ~lation computed, wbich was squared to obtain R~. Because it was based on a praliction equation from another study, no shrinkage was expected. bAn R2 of 0.00 is expected in an ovenletermined system, in wbich there are more predictors than subjects. Perfect classification is always possible. •Binomial classification was reported. The tetrachoric com:lation was computed as in footnote a. "The R2 values were estimated from a misclassification rate of 20% with 36 subjects. Although actual study was trinomial, the R2 represents the equivalent for binomial classification for ease of computing. *p < 0.05.
reduction of three groups to two (nonnal versus abnormal). Then, an effect size was computed, (6)
as defined by Glass (1978). This statistic is the number of standard deviations separating the two groups. Finally, the percentile point under the normal curve for halfthe effect is presented. This is the point that minimizes misclassification assuming equal cost for either false-positive or false-negative errors, and the equal population base rates. Of the 17 R~ obtainable from the studies, 12 were initially significant. After correcting the shriilkage, only 4 were significant as Ri-, and 8 as R~. Thus, half the results reported in these studies are attributable to chance large correlations. Under the most optimistic of circumstances, the upper limit of shrinkage R2 estimate shows a mean R2 of 0. 37 versus a mean obtained R 2 value of 0.48 for all studies considered. The lower bound estimate of the shrunken R 2 yields even more pessimistic results, demonstrating a mean value of but 0.25. The chance variation that can appear on the surface to be reliable discrimination with powerful multivariate techniques is thus rather considerable. The importance of large
subject/variable ratios and proper cross-validation becomes immediately obvious in considering the results summarized in Tables 8 and 9. Interested professionals must consider with special care the prediction rules generated from those studies when R dropped to nonsignificance. In examining the misclassification rates (see Table 9), there is a change from about one-third expected in the original studies (35%) to almost half (44%) using the corrected values, the chance rates under no knowledge, and very unimpressive when contrasted against base rates in referral populations. This is not surprising given the considerable decline in effect sizes shown in Table 9. It must be reiterated that the shrinkage occurs in research in which correlation maximizing procedures have been used: stepwise multiple regression, stepwise discriminant analysis, and canonical correlation. The R2 does not shrink in a fixed variable study in which all variables are included and in which order is unimportant (balanced ANOVA design) or in which order is predetermined (path analysis design, causal model design). Diagnosis seeks to find the best empirical discriminators, but it is most subject to chance. The shrunken estimate of R2 and the expected
NEUROPSYCHOLOGICAL ASSESSMENT OF CHILDREN
163
TABLE 9. Expected Misclassifications from Nine Studies
t-equivalent for
·z RT
·z Rr
·z RT
RF
·z
·z RT
(a) (b)
2.76 0 6.27 1.15 3.12 1.69<" 1.21c
(c)
0.13c
(d) (e) (f)
2.84C
0.86 0 6.27 1.15 1.89 0.00 0.00 0.00 0.42 0.00 0.65 0 0.37 1.61 4.00 1.19 5.04 1.38
0.50 0 1.25 0.27 0.44 0.62 0.44 0.27 0.71 0.73 0.64 2.05 0.25 1.49 2.04 0.40 1.68 0.81
0.16 0 1.25 0.27 0.27 0.00 0.00 0.00 0.15 0.00 0.24 0 0.06 0.42 0.73 0.40 1.68 0.33
39% 50% 27% 45% 41% 38% 41% 45% 36% 36% 37% 16% 45% 23% 15% 42% 20% 35%
47% 50% 27% 45% 45% 50% 50% 50% 47% 50% 45% 50% 49% 42% 36% 42% 20% 44%
(a) (b)
Selz & Reitan ( 1979)h Wallbrown, Vance, & Pritchard (1979) Taylor & Imivey (1980)
Dunleavy, Hansen, & Baade (1981) Fuller & Goh ( 1981) Golden, Moses, Graber, & Berg (1981) Malloy & Webster (1981)b Mean
Two-population % misclassification for
'2
RF Dean (1978) Purisch, Golden, & Hammeke (1979)
Estimated effect" size for two populations for
(a) (b) (a) (b)
1.95c
1.16C 5.01 1.14 5.78 11.15 1.19 5.04 3.11
aEffect = t(lln, + l/n2 ) 112 (Glass, 1978). hNo shrinkage occurred. ··single group statistics; effects were computed as if for two groups. Single group results are smaller than reported here.
misclassification rate are part of the technique of Should prediction studies be cross-validated pricross-validation. Cross-validation requires two inde- or to publication? Is it the responsibility of the rependent samples. Ideally both are drawn indepen- searcher to provide this evidence? If actuarial rules dently from the same population. Often a single sam- for diagnosis are used, the obvious answer is yes. ple is split into two halves. In either case the Even in more purely clinical decision-making, the regression is computed on one sample and the repeatability of one's results from a referral pool canweights applied to the scores of the second popula- not be ignored. The Selz and Reitan ( 1979) research tion to predict the outcome or group membership as is an example where this procedure was followed, appropriate. The R 2 is a one-sample estimate of the with quite credible results. The application of such a R 2 in the nonvalidation sample, but it is not nearly so prediction rule to new populations requires new convincing. First, it is a statistic itself that may vary; cross-validation, however. To those who argue it is second, it uses information from one sample, not difficult to obtain subjects in rare disorder categonearly as good as that available from two samples. In ries-hold the results until a second population is clinical samples theN is typically so small that split- sampled. There will be no real loss to the discipline. ting it is not a good idea. The regression weights and On the contrary, there will be a net gain, for only the R 2 will become even more changeable as the sub- cross-validated results will achieve publication. ject/variable ratio is halved. This leaves two-sample Given the sometimes harsh attacks on the application cross-validation. Samples should be drawn from the of clinical neuropsychological techniques (e.g., same population initially. There may also be consid- Coles, 1978; Sandoval, 1981) to the rehabilitation of erable value in determining the generalizability of the learning and related problems, it certainly behooves results to other populations in an effort to improve the researchers to be careful in deciding when research is clinical utility of the classification rules. These are ready to report. The external validity of one's "findseparate problems; the sampling procedure in the first ings" must be clear. case is obvious and has been discussed. Sampling in Cross-validation, when it is presented as evithe second case will be dictated by the specific design dence for the consistency of results in one populaof the study. tion, does not provide evidence for generalizability to
164
CHAPTER 8
other populations. This can be done by additional approaches. Roffe and Bryant (1979), in a study of studies. There is no reason to believe the regression profile reliability for the McCarthy Scales, used the weights that discriminate two groups will discrimi- Pearson correlation. This approach is problematic nate either from a third. This point is not restrictive to from several perspectives. If the Pearson correlation prediction studies but occurs in all behavioral is employed, profiles must first be "standardized" as research. described by Nunnally (1978). However, even under A related issue should be obvious and of great these circumstances, the Pearson correlation is likely significance to the practicing clinician in neuropsy- to be inaccurate, for it does not consider level, disperchology and other subdisciplines involved in differ- sion, shape, and accentuation of the various profiles. ential diagnosis. When actuarial rules for the diag- Several very powerful techniques have been develnosis of psychological and neuropsychological oped that accurately determine the similarity of prodisorders appear in refereed professional journals, files, including D 2 (Cronbach & Gieser, 1953; Osthose in applied settings, especially those keeping good & Suci, 1952) and the coefficient of pattern closest to current developments in the field, may feel similarity, rP (Cattell, 1966). These two approaches very confident in applying such rules in diagnosis and in particular are accurate, sophisticated measures of in the development of treatment plans. Clearly, the profile similarity and should be employed whenever Willson and Reynolds ( 1982) results show that in the possible. absence of proper cross-validation, many diagnoses or classifications may be made on the basis of random relationships. This constitutes an unacceptable situation for all involved, but especially for the patient and Summary his/her physician. It has become customary over the years to end reports of research with cautionary statements and call for further research. If diagnostic or classificaProfile Reliability tion studies with small samples and large numbers of A related problem when multiple scores are variables employing powerful, sophisticated multibeing used in classification or individual diagnoses variate classification techniques are to continue to and decision-making is the reliability of the set of appear without concurrent cross-validation, and mulscores for each individual considered. Perhaps sta- tivariate profiles are to be considered, much stronger bility is a better conceptualization, as the question is cautions are needed to avoid the inadvertent leading whether (or how much) the profile of scores, taken as of the diagnostician into potential malpractice. The a whole, would change and whether this change most obvious, and the most sound solution is not to would affect clinical decision-making. Stability of publish such small-N studies without concurrent repprofiles over at least very short periods of time lication, and not to rely on profiles with unknown (largely depending upon the clinical disorder under stability in differential diagnosis research or clinical investigation) needs investigation and yet has gone practice. Many problems related to measurement and stalargely ignored. The problems of such research, which may at first seem simple, are difficult ones, but tistics in clinical neuropsychological research and in clinical diagnosis have been reviewed here. Many can be solved. The most difficult problem is that of differential other problems exist but a substantial portion of these practice effects among the various scales that go into difficulties can be resolved by avoiding the problems making up the score profile. This introduces meth- noted in this chapter. By so doing, other, now fuzzy, odological artifacts that require statistical control issues, methodological, statistical, and clinical, through estimation of regression effects for each part should be brought into a sharper focus and new probof the battery prior to comparing the two profiles lems can be identified. The failure to resolve basic obtained-in such a case, only the second profile measurement issues in clinical neuropsychological should be corrected. Following this set of correc- research can do nothing except restrain progress in tions, profile stability or reliability can then be as- the field at a time when sophisticated technology is sessed. Profile reliability is essentially a multivariate experiencing explosive growth all around us. Major problem and thus requires a multivariate solution. A publishers must be persuaded to enter the field in a variety of statistical approaches may be used and the big way and users of neuropsychological tests must specific purpose of the study may dictate different demand quality instrumentation.
NEUROPSYCHOLOGICAL ASSESSMENT OF CHILDREN
165
research. In L. Shulman (Ed.), Review ofResearch in Education, 5, 351-379. Golden, C. 1. (1981). Diagnosis and rehabilitation in clinical Adams, R. L. (1985). Review of the Luna-Nebraska Neuropsyneuropsychology (2nd ed.). Springfield, IL: Thomas. chological Battery. In J. V. Mitchell (Ed.), The ninth mental Golden, C. J., Moses, J. A., Jr., Graber, B., & Berg, T. (1981). measurements yearbook. Lincoln: University of Nebraska Objective clinical rules for interpreting the Luria-Nebraska Press. Neuropsychological Battery: Derivation, effectiveness, and Algozzine, B., Forgnone, C., Mercer, C., & Trifiletti, J. (1979). validation. Journal of Consulting and Clinical Psychology, Toward defining discrepancies for specific learning dis49, 616-68. abilities: An analysis and alternative. Learning Disability Gordon, R. A. (1984). Digits backward and the Mercer-Kamin Quarterly, 2, 25-31. law: An empirical response to Mercer's treatment of internal Angoff, W. H. (1971). Scales, norms, andequivalentscores.lnR. validity ofiQ tests. In C. R. Reynolds & R. T. Brown (Eds.), L. Thorndike (Ed.), Educational measurement (2nd ed.). Perspectives on Bias in Mental Testing. New York: Plenum. Washington, DC: American Council on Education. Gutkin, T. B., & Reynolds, C. R. (September, 1980). Normative Bersoff, D. N. ( 1982). The legal regulation of school psychology. Data for Interpreting Reitan's Index of Wechsler Subtest In C. R. Reynolds & T. B. Gutkin (Eds.), The handbook of Scatter. Paper presented to the annual meeting of the Amerischool psychology. New York: Wiley. can Psychological Association, Montreal. Boder, E., & Jarrico, S. (1982). Boder test of reading-spelling Hanna, G. S., Dyck, N. J., & Holen, M. C. (1979). Objective patterns. New York: Grone & Stratton. analysis of achievement-aptitude discrepancies in LD classiBums, E. (1982). The use and interpretation of standard grade fication. Learning Disability Quanerly, 2, 32-38. equivalents. Journal of Learning Disabilities, 15, 17-18. Hynd, G. (Ed.). (1981). Neuropsychology in the schools. School Cattell, R. B. (1966). Handbook ofmultivariate experimental psyPsychology Review, 10(3). chology. Chicago: Rand McNally. Jastak, J. F., & Jastak, S. (1978). Wide Range Achievement Test. Cattin, P. (1980). Note on the estimation of the squared crossWilmington, DE: Jastak. validated multiple correlation of a regression model. PsychoKaufman, A. S. (1976a). A new approach to the interpretation of logical Bulletin, 87, 63-65. test scatter on the WISC-R. Journal ofLearning Disabilities, Coles, G. S. (1978). The learning disabilities test battery: Em9, 160-167. pirical and social issues. Howard Educational Review, 4, Kaufman, A. S. (1976b). Verbal-performance IQ discrepancies on 313-340. the WISC-R. Journal of Learning Disabilities, 9, 739-744. Cone, T. E., & Wilson, L. R. (1981). Quantifying a severe disKaufman, A. S. (1979). Intelligent testing with the WISC-R. New crepancy: A critical analysis. Learning Disability Quarterly, York: Wiley-Interscience. 4, 359-371. Kaufman, A. S., &Kaufman, N. L. (1983). Kaufman assessment Cooley, W. W., & Lohnes, P. R. (1971 ). Multivariate data analybattery for children: Interpretive manual. Circle Pines, MN: sis. New York: Wiley. American Guidance Service. Cronbach, L. J., & Gieser, G. C. (1953). Assessing similarity Lord, F. M., & Novick, M. R. (1968). Statistical theories of between profiles. Psychological Bulletin, 50, 456-473. mental tests. Reading, MA: Addison-Wesley. Danielson, L. C., & Bauer, J. N. (1978). A formula-based classiMalloy, P. F., & Webster, J. S. (1981). Detecting mild brain fication of learning disabled children: An examination of the impairment using the Luria-Nebraska Neuropsychological issues. Journal of Learning Disabilities, 11, 163-176. Battery. Journal of Consulting and Clinical Psychology, Davis, F. B. (1959). Interpretation of differences among average 49,768-770. and individual test scores. Journal of Educational PsycholMatarazzo, J. D. (1972). Wechsler's measurement and appraisal ogy. 50, 162-170. of odult intelligence. Baltimore: Williams & Wilkins. Dean, R. S. (1978). Distinguishing learning-disabled and emoNunnally, J. C. (1978). Psychometric theory (2nd ed.). New tionally disturbed children on the WISC-R. Journal of ConYork: McGraw-Hill. sulting and Clinical Psychology, 46, 381-382. Osgood, C. E., & Suci, G. J. (1952). A measurement of relation Dean, R. S. (1985). Review of the Halstead-Reitan Neuropsychodetermined by both mean differences and profile interpretalogical Test Battery. In J. V. Mitchell (Ed.), The ninth mental tion. Psychological Bulletin, 49, 251-262. measurements yearbook. Lincoln: University of Nebraska Parsons, 0. A., & Prigatano, G. P. (1978). Methodological conPress. siderations in clinical neuropsychological research. Journal Dunleavy, R. A., Hansen,J. L., & Baade, L. E. (1981). Discrimiof Consulting and Clinical Psychology, 46, 608-619. nating powers of Halstead Battery tests in assessment of 9 to Piotrowski, R. J. (1978). Abnormality of subtest score differences 14 year old severely asthmatic children. Clinical Neuropsyon the WISC-R. Journal ofConsulting and Clinical Psycholchology, 3, 9-12. ogy, 46, 569-570. Fuller, G. B., & Gob, D. S. (1981). Intelligence, achievement, Plake, B. S., Reynolds, C. R., & Gutkin, T. B. (1981). A techand visual-motor performance among learning disabled and nique for the comparison of profile variability between indeemotionally impaired children. Psychology in the Schools, pendent groups. Journal of Clinical Psychology, 37, 14218, 262-268. 146. Glass, G. V. (1978). Integrating findings: The meta-analysis of Purisch, A. D., Golden, C. J., & Hammeke. T. A. (1979). Dis-
References
166
CHAPTER 8
crimination of schizophrenic and brain injured patients by a standardized version of Luria's neuropsychological tests. Clinical Neuropsychology, 1, 53-59. Reynolds, C. R. (1979a). Interpreting the index of abnormality when the distribution of score differences is known: Comment on Piotrowski. Journal ofConsulting and Clinical Psychology, 47, 401-402. Reynolds, C. R. (1979b). Objectivity of scoring for the McCarthy Drawing Tests. Psychology in the Schools, 16, 367-368. Reynolds, C. R. (198la). The problem of bias in psychological assessment. In C. R. Reynolds & T. B. Gutkin (Eds.), The handbook of school psychology. New York: Wiley. Reynolds, C. R. ( 1981 b). Screening tests: Problems and promises. In N. Lamberts (Ed.), Special Education Assessment Matrix. Monterey, CA: CTB McGraw Hill. Reynolds, C. R. (1982). The importance of norms and other traditional psychometric concepts to assessment in clinical neuropsychology. In R. N. Malatesha & L. C. Hartlage (Eds.), Neuropsychology and cognition (Vol. II). The Hague: Nijhoff. Reynolds; C. R. (1983). Some new and some unusual educational and psychological tests. School Psychology Review, 12, 481488. Reynolds, C. R. (1984). Critical measurement issues in learning disabilities. Journal of Special Education, 18, 451-476. Reynolds, C. R. (1986). Clinical acumen but psychometric naivete in neuropsychological assessment of educational disorders. Archives of Clinical Neuropsychology, 1, 121-138. Reynolds, C. R., & Clark, J. H. (1985). Profile analysis of standardized intelligence test performance of very low functioning individuals. Journal of School Psychology, 23, 277283. Reynolds, C. R., & Clark, J. H. (1986). Profile analysis of standardized intelligence test performance of very high functioning individuals. Psychology in the Schools, 23, 5-12. Reynolds, C. R., & Gutkin, T. B. (1979). Predicting the premorbid intellectual status of children using demographic data. Clinical Neuropsychology, 1, 36-38. Reynolds, C. R., & Gutkin, T. B. (1980). Statistics related to profile interpretation of the Peabody Individual Achievement Test. Psychology in the Schools, 17, 316-319. Reynolds, C. R., Hartlage, L. C., & Haak, R. (1980, September). Lateral preference as determined by neuropsychological performance and aptitude/achievement discrepancies. Paper presented to the annual meeting of the American Psychological Association, Montreal. Reynolds, C. R., & Kaufman, A. S. (1986). Clinical assessment of children's intelligence with the Wechsler Scales. In B. Wolman (Ed.), Handbook of intelligence. New York: Wiley. Reynolds, C. R., & Willson, V. L. (1983, January). Standardized grade equivalents: Really! No. Well, sort of, but they lead to the valley of the shadow of misinterpretation and confusion. Paper presented to the annual meeting of the Southwestern Educational Research Association, Houston. Roffe, M. W., & Bryant, C. K. (1979). How reliable are MSCA
profile interpretations? Psychology in the Schools, /6, 14-18. Rourke, B. P. (1975). Brain-behavior relationships in children with learning disabilities: A research program. American Psychologist, 30, 911-920. Sandoval, J. (1981, August). Can neuropsychology contribute to rehabilitation in educational settings? No. Paper presented to the annual meeting of the American Psychological Association, Los Angeles. Sattler, J. M. (I914).Assessmentofchildren's intelligence. Philadelphia: Saunders. Satz, P., Taylor, H. G., Friel, J., & Fletcher, J. (1978). Some developmental and predictive precursors of reading disabilities: A six year follow-up. In A. L. Benton & D. Pearl (Eds.), Dyslexia: An appraisal of current knowledge. New York: Oxford University Press. Selz, M., & Reitan, R. M. (1979). Rules for neuropsychological diagnosis: Classification of brain functions in older children. Journal of Consulting and Clinical Psychology, 47, 258-
264. Tabachnick, B. G. ( 1979). Test scatter on the WISC-R. Journal of Learning Disabilities, 12, 60-62. Taylor, R. L., & lmivey, J. K. (1980). Diagnostic use of the WISC-R and McCarthy Scales: A regression analysis approach to learning disabilities. Psychology in the Schools, 17, 327-330. Thompson, R. J. (1980). The diagnostic utility ofWISC-R measures with children referred to a developmental evaluation center. Journal of Consulting and Clinical Psychology, 48,
440-447. Thorndike, R. L., & Hagen, E. P. (1977). Measurement and evaluation in psychology and education (4th ed.). New York: Wiley. USOE. ( 1976). Education of handicapped children: Assistance to states: Proposed rulemaking. Federal Register, 41, 5240452407. Wallbrown, F. H., Vance, H., & Pritchard, K. K. (1979). Discriminating between attitudes expressed by normal and disabled readers. Psychology in the Schools, 4, 472-477. Wechsler, D. (1974). Wechsler Intelligence Scale for ChildrenRevised. New York: Psychological Corporation. Wherry, R. J., Sr. (1932). A new formula for predicting the shrinkage of the coefficient for multiple correlation. Annals of Mathematical Statistics, 2, 404-457. Williams, N. H. (1976). Arousal and information processing in learning disabled children. Unpublished doctoral dissertation, University of Alberta, Edmonton, Canada. Willson, V. L., & Reynolds, C. R. (1982). Methodological and statistical problems in determining membership in clinical populations. Clinical Neuropsychology, 4, 134-138. Willson, V. L., & Reynolds, C. R. (in press). Classroom applications of educational measurement. New York: MacMillan. Wright, L., Schaefer, A. B., & Solomons, G. (1979). Encyclopedia of pediatric psychology. Baltimore: University Park Press.
9 Models of Inference in Evaluating Brain-Behavior Relationships in Children EILEEN B. FENNELL AND RUSSELL M. BAUER
Introduction Neuropsychologists generally measure behavior as a means for making inferences about brain function. Regardless of whether such measurement takes place in the clinic or the laboratory, the basic process is the same: behavioral and cognitive performances that are readily observable "stand in," as it were, for the less observable ''brain states'' they are thought to reflect. Once measurement is completed, the quantitative and qualitative relationships among such performances are assembled according to certain rules in order to make probabilistic statements about brain function. The rules that are applied in the given case depend on the inferential model one uses in relating behavioral performance to brain function. This basic process characterizes all of neuropsychology, transcends theoretical persuasion, and, in fact, is a fundamental aspect of the clinical-inferential method in general. It may seem trivial to point this out because making inferences about brain functioning from behavioral data is so much a part of the neuropsychological approach. However, because such inferences are so routinely made, it is important not only to articulate the various levels at which they play a part in our thinking, but also to understand the specific theoretical assumptions on which they are based. The goal of this chapter is to describe major inferential models that relate behavior to· brain function in the child neuropsychology area. In working toward this EILEEN B. FENNELL AND RUSSELL M. BAUER • Department of Clinical and Health Psychology, University of Florida, Gainesville, Florida 32610
goal, we will first describe basic issues in clinical inference as they relate to child neuropsychology. In doing so, we will outline a ''hypothesis-testing'' approach to neuropsychological assessment that borrows from classical methods of inductive inference. Finally, we will articulate some of the basic models of inference of particular relevance to child neuropsychological assessment.
Basic Issues in Clinical Inference Clinical-Inferential Methods ''Inference'' refers to the process of arriving at a conclusion by reasoning from evidence. Inferences generally take place according to organized systems of rules that stipulate (a) the kind of evidence on which conclusions can be drawn, (b) the kinds of conclusions that are possible given certain evidence, and (c) a set of logical connections between evidence and conclusions. In child neuropsychology, the performance of the child on cognitive or neuropsychological tests is the "evidence" on which inferences about brain functioning are based. It is assumed that the ''conclusions'' of interest are couched in terms of some aspect of brain function that is not directly observable by the neuropsychologist. In general, we do not observe or measure ''functions''; we see only the behavioral indicators of spared and impaired brain functions. Thus, for example, we make inferences about language function on the basis of performance on tests of language ability; we may infer that there is some disturbance in brain-based attentional mechanisms when the child cannot stay on task, repeat 167
168
CHAPTER 9
digits, or shows inconsistent performance across a set of homogeneous test items. In what follows, we presume that there are complex differences between children and adults in terms ofhow brain pathology leads to neuropsychological and behavioral deficits. Thus, the meaning or utility of a behavioral "sign" that has been well-validated with adults may be different when that sign is applied to child neuropsychology cases (Rourke, 1983; Aetcher & Taylor, 1984). Similarly, the predictive value of knowing that a patient has had a specific brain insult may critically depend on the age at which the damage was incurred (Woods & Teuber, 1973; Rasmussen & Milner, 1977). Despite the fact that the behavioral effects of the brain disease are complexly dependent on such factors as age and stage of development, we believe that one fruitful approach to understanding childhood neuropsychological data is the classical inferential method commonly used in adult neurology (Adams & Victor, 1977). Briefly, this method involves the collection of clinical data in terms of signs (e.g., neuropsychological impairments) and symptoms (clinical complaints). These findings are then correlated with similar signs and symptoms occurring in neurological disorders in which the underlying anatomy is known. By making analogy with these better known disorders, and by reasoning from anatomic data, the findings in an individual case can be interpreted in terms of some pathophysiological mechanism. This particular approach can be used with the individual case or can form the framework of a general conceptualization of a disorder that is as yet poorly understood. An excellent example is seen in a neurobehavioral model of autism (Damasio & Maurer, 1978; Maurer & Damasio, 1982). These authors observed various abnormalities in a group of autistic children, including disturbances of motility (stereotyped movements, abnormal posturing and gait), attention (unpredictable response to sensory stimuli, gaze aversion), communication (mutism, use and comprehension of nonverbal signs), and social behavior (poor cooperative play, failure to initiate social interaction). They related each of these signs and symptoms to findings in specific acquired neurological diseases (e.g., basal ganglia disease, acquired mutism from mesial frontal lobe lesions) in which the pathogenesis and localization were more firmly established. On the basis of this analysis, they proposed a specific neuroanatomy for autism, which included the mesolimbic cortex (mesial temporal and frontal lobes), the neostriatum, and the anterior and medial thalamic nuclei. For our purposes, the specif-
ic merits of this hypothesis, and its ability to explain core features of autism, are not at issue. What is important, however, is that the hypothesis was derived by starting with observable signs and symptoms and by inferring from them a possible functional anatomy.
Levels of Inference Up to this point, we have discussed clinical inference as if it were a simple process of reasoning from neuropsychological test performance to brain function. In fact, there are several types of inference involved here, each of which exists at a different level of analysis. For purposes of discussion, consider a 5-year-old male child who has received a closed head injury in a vehicular accident. He is given ·a battery of neuropsychological tests including assessment of intellectual ability, memory, language, visual and auditory perception, attentional ability, and sensorimotor skill. Results indicate low average intellectual ability, attentional and recent memory problems, and poor beginning reading skills. We are asked to relate the child's current status to the recent head injury and to assist in educational planning. There are three basic levels of analysis involved in utilizing our test data to answer such questions. At the first level, we are concerned with the degree to which the behavior elicited by the test battery is representative of the domain of behavior that would have been elicited given unlimited testing time (Cronbach, Rajaratnam, & Gieser, 1963). The basic issue here is whether our test findings are generalizable to other settings or conditions. At the second level of analysis, we are interested in the specific meaning of each of the test findings. Each test finding might have a formal statistical (actuarial; see Wiggins, 1973) relationship with a specific form of brain impairment or may suggest a qualitative feature seen in other known brain disease (Adams & Victor, 1977). In either event, we are inferring what the outcome of each test means in terms of some nontest behavior or variable. This is a process that Holt ( 1968) termed primary• iriference. Thus, we make inferences regarding the status of verbal memory ability (and its constituent variables) by individually noting performance on tests to which this ability contributes. At the third level, we are concerned with integrating the diverse test findings to arrive at a general interpretation or conceptualization of their meaning. We are concerned here with the degree to which the pattern of spared and impaired abilities suggests a specific neuropsychological mechanism that can best account for
EVALUATING BRAIN-BEHAVIOR RELATIONSHIPS
the test findings and the clinical complaints. Holt ( 1968) indicated that this level demands a knowledge of the range of expectable syndromes, what their constituent variables are, and some means of measuring the strength of each variable. By drawing on this knowledge and on his [sic] knowledge of theory, the diagnostician puts together his primary inferences and in an act of secondary inference locates the subject with reference to diagnostic syndromes. (p. 15)
169
ment approach. Regardless of whether one adopts a "quantitative" or a "qualitative" approach, or some mixture of the two, an important concern is the manner in which clinical data contribute to hypotheses about the nature of the child's neuropsychological status. In this section, we outline a general approach to forming and testing hypotheses derived from neuropsychological test data that seems equally well suited to both quantitative and qualitative models. This approach is based on our belief that there is no fundamental distinction between scientific and clinical hypotheses. That is, we believe that the same inferential processes one uses in the research laboratory for distinguishing between viable and invalid scientific hypotheses can be used in deriving and testing clinical hypotheses (see also Landy, 1986, who takes a hypothesis-testing approach to the process of test validation). Our approach is based on a model of inductive inference outlined by Platt (1964), a model that abounds in the physical sciences, particularly molecular biology and high-energy physics. Platt called his approach strong inference because its systematic application seems related to rapid scientific advance in the fields that utilize its strengths. It is based primarily on disconjirmatory logic (Popper, 1959) and consists mainly of the sequential evaluation of hypotheses that survive disconfirmation in experimental (clinical) test. Strong inference consists of the systematic application of the following steps:
It is important to note that the specific inferences the clinician makes at each of these three levels will differ depending on the purpose and targets of assessment and will at least in part be related to the assumptions the clinician makes about the relationship between test behavior and brain function. Such differences are most apparent at the second and third levels. For example, clinicians who favor an actuarial or statistical approach to test interpretation will be most concerned with the formal quantitative relationships between test signs and specific forms of neuropsychological impairment. Those who favor qualitative approaches might be more interested in demonstrating the presence of one or more "pathognomic signs'' that are considered crucial indicators of functional impairment. In either case, an attempt is made to relate the pattern of test findings to previously available data on children with head injuries. The quantitative clinician may attempt to determine the degree to which other children with documented head injuries obtained the same or similar clinical a. Devising alternative hypotheses profile. The qualitative clinician might focus efforts b. Devising "crucial experiments" with alteron determining the degree to which this child is simnative possible outcomes, each of which will ilar to other head-injured children in terms of the exclude one of the alternative hypotheses specific neurobehavioral mechanisms that can best c. Carrying out "clean" experiments account for their neuropsychological deficit pattern. d. Recycling the procedure with surviving In actual practice, the responsible clinician frehypotheses quently uses a mixture of actuarial and clinical methods in making inferences from test data (Meehl, These four steps are recognizable to all neuropsycho1957; Lezak, 1983). logists as the basic elements of inductive inference. The difference, however, is in the systematic, formal Fundamentals of Hypothesis Formation: The application of all of these steps to every clinical problem. Logic of Strong Inference We will illustrate the utility of this approach by In the previous section we briefly considered describing a series of simple "experiments" dethree levels at which inferences about brain function signed to determine more precisely the nature of a may be made from test performances. The specific specific neuropsychological deficit. As an example, nature and content of such inferences will depend in assume we have a child who performs poorly on large part on the assumptions the clinician makes WISC-R Block Design and assume that we have reaabout the relationship between test behavior and son to believe from the medical history that a signifibrain function. That is, inferential processes in neu- cant neuropsychological factor is involved. Because ropsychological assessment are inextricably related Block Design measures more than one ability, a to the theoretical or conceptual basis of one's assess- question of some relevance might be to determine
170
CHAPTER 9
more specifically the nature of the child's failure. On an a priori basis, the Block Design test can be thought of as tapping motor, visuoperceptual, visuomotor, constructional, and problem-solving abilities. A strong inference approach to isolating the reason(s) for a deficit in Block Design would proceed first by devising a series of tests that systematically eliminate one or more of these constituent abilities and then observing the resulting effects on performance. Assuming relatively good control over difficulty level, performance on Block Design could be contrasted with performance on visuoperceptual tasks without motor demands (e.g., form discrimination, visual synthesis). The relative role of visuomotor versus motor abilities could be assessed by contrasting performance on motor tests not requiring visual tracking (e.g., fmger tapping, fine finger movements) with tasks with high visuomotor demands (e.g., drawing, visual reaching, grooved pegboard), and so on. By a systematic approach that involves the ruling out of alternative hypotheses for the Block Design performance, attention is gradually directed toward surviving hypotheses. For example, a pattern of good performance on visual, problemsolving, and motor tasks would tend to rule these abilities out as explanations, leaving the hypothesis that it was the visuomotor or constructional aspect of Block Design that specifically gave rise to the poor performance. It is our view that almost any complex behavior can be broken down in this way so that more precise hypotheses about performance deficits can be achieved. The most important point is that a strong inference approach, with its emphasis on disconfirmation, is quite different from the logic typically used in traditional approaches to neuropsychological testing. Traditional approaches typically utilize a pattern of test performances (e.g., lateralized motor or sensory findings, poor nonverbal memory performance) as confirmatory evidence for a particular hypothesis. This is the fundamental assumption underlying the so-called "sign" approach. Our view is that although this approach may result in correct inferences, it does so inefficiently and at the risk that alternative explanations for a test sign or performance have not been entirely ruled out.
Qinical Judgment in Neuropsychology We have implied in the previous section that the theoretical model the clinician espouses will be an important factor governing the kinds of inferences made about neuropsychological test data. Although
this makes the clinician an additional source of variance in test interpretation, which some (e.g., Rourke, Bakker, Fisk, & Strang, 1983) find somewhat undesirable, we believe that the "cognitive activity of the clinician" is an integral and inevitable aspect of neuropsychological test interpretation. This issue has received little systematic attention within neuropsychology, though some guidance is available from the contemporary application of the Brunswick Lens Model (Brunswick, 1956) to the practice of psychodiagnosis (Hammond, Hursch, & Todd, 1964; Hursch, Hammond, & Hursch, 1964; Meehl, 1960; Wiggins, 1973). The basic idea is that inference in neuropsychology is not only dependent on the specific relationship among test signs and brain function (so-called criterion-oriented validity), but also on the manner in which the neuropsychologist uses such test signs in arriving at interpretive statements. There are two basic issues involved here. First is the degree to which the clinician accurately uses the test results to arrive at a clinical diagnosis. This issue can be understood if one assumes that the separate test performances function as variables that, separately and in combination, predict some criterion (e.g., brain function). The intercorrelations among test performances, and their individual relationships to the criterion, determine the relative importance (weighting) each performance has in predicting the criterion. In an ideal setting, the neuropsychologist utilizes the various test performances in a manner that accurately reflects the separate and combined relationship such performances actually have with the criterion. In this ideal world, the clinician's inferences directly reflect the empirical validities of the various test performances vis-a-vis the criterion; the test performances that bear stronger relationships with the criterion are given more weight than are those that correlate less highly. However, in actual clinical practice, the clinician may not have precise knowledge of the predictive relationship between test performance and criterion. What might result from this situation is a method of combining test performances that does not accurately reflect their predictive validity with respect to the criterion. In this case, it becomes important to distinguish empirical validity (the statistical relationship that exists between predictor variables and criterion) and cue utilization (the relationship between test performance and the inferences made by the clinician) (see Wiggins, 1973, p. 157). A second issue is the manner in which test results are combined to arrive at a clinical conclusion. Do neuropsychologist& combine test findings in the linear fashion implied by multiple regression ac-
EVALUATING BRAIN-BEHAVIOR RELATIONSHIPS
counts of empirical validity, or do they adopt a more complex, nonlinear method of combining data in which certain test scores are viewed as a priori more important than others? An example of a nonlinear approach to test score interpretation is the use of the ''pathognomic sign'' approach in neuropsychology. Pathognomic signs are performances that are seen rarely, if ever, in persons with normal brain function. When they do appear, therefore, they more than likely suggest some brain impairment. For example, the appearance of aphasia is regarded as a pathognomic sign of left hemisphere impairment. In terms of this discussion, pathognomic signs could be weighted very heavily and could form the basis of the clinical judgment even in the absence of other supportive clinical evidence. In this situation, the clinician could elect, perhaps unwisely, to ignore the relationships among other test performances and the criteria if pathognomic signs are present.
Summary In discussing these basic issues, our purpose has not been to argue for one or another approach to interpretation, but rather to articulate the processes entering into the inferential process in child neuropsychology. Our view is that it is important to be explicitly aware of the crucial role such processes play in making sense out of neuropsychological test data. In large part, the concepts that are invoked to explain neuropsychological test performances (e.g., attention, memory capacity) are unobservable and must be inferred from overt behavior. Whether one elects to deal at the individual test level or at the level of pattern analysis, the same basic inferential processes are involved. In this section, we have purposely emphasized the context, rather than the content, of inductive inference in the clinical setting. As we have stated, this is important because the inferential method is such a fundamental aspect of the neuropsychological approach. We have outlined a strong inference model that, for us, is a useful way of conceptualizing the process of hypothesis formation and hypothesis testing in child neuropsychology. Finally, we have attempted to point out some of the general issues involved when a clinician attempts to derive meaningful interpretations of multiple test performances. With these broad issues as a background, the specific nature and content of the inferences made by the neuropsychologist will in large part depend on the level of data analysis and on the conceptual model that governs the assessment approach. We now tum to a discussion of the major inferential models of relevance to the child neuropsychologist.
171
Models of Inference in Child Neuropsychology As Tarter and Edwards (1986) noted, clinical neuropsychological assessment is descriptive, correlational, and inferential rather than explanatory, causal, and direct. Neuropsychological procedures provide descriptive information regarding behavior in both normal children and neurologically impaired children. This description of behavior on standardized and formal observations (tests) is then correlated with suspected or known pathological lesions derived from other tests (concurrent validity) in order to enable the neuropsychologist to: ( 1) infer the presence or absence of brain pathology from test signs (primary inference) or (2) classify the individual child according to test and historical variables into some classification group (e.g., brain injured, learning disabled, attention deficit disordered) by a process of secondary inference. Although the inferential process may be similar in adult and child neuropsychology, major differences exist in the context of that process.
The Inferential Context of Child Neuropsychology Several recent texts devoted to child or developmental neuropsychology emphasize the critical differences in the neuropsychological organization and functioning of children as compared to adults (Rourke et al., 1983; Spreen, Tupper, Risser, Tuokko, & Edgell, 1984; Hartlage & Telzrow, 1986). Furthermore, as the analytic focus for the child neuropsychologist is on a developing brain, inferential processes regarding brain function must rest on models that account for differences in development at different ages (Segalowitz & Gruber, 1977; Vander Vlugt, 1979; Dean, 1986) rather than rely on the fixed models of adult brain functions and pathologies (Lezak, 1983; Heilman & Valenstein, 1985). In addition, unlike adult neuropsychology, child neuropsychology is still in need of basic descriptions of the neuropsychological effects of developmental and acquired pathologies of childhood (Boll & Barth, 1981; Boll, 1983; Menkes, 1985; Netley & Rovet, 1983; Pirozzolo, Campanella, Christensen, & LawsonKerr, 1981; Rutter, 1983; Spreen et al., 1984) as well as with regard to understanding the role of individual differences in affecting the descriptive and inferential processes in child neuropsychology (Bakker, 1984; Bolter & Long, 1985; Clark, 1984; Rourke & Adams, 1984; Dean, 1986). Finally, only recently
172
CHAPTER 9
has an emphasis been placed on the discriminative validity of various test signs and historical indicators of certain subgroups of childhood disorders such as learning disabilities (Rourke, 1981; Rourke, Fisk, & Strang, 1986; Morris, Blashfield, & Satz, 1986) or minimal brain dysfunctions (Denck1a, 1979; Satz & Fletcher, 1980; Ross & Ross, 1982; Chadwick & Rutter, 1983). To the extent that the content of neuropsychological knowledge of the developing brain is still emerging, primary and secondary inference in child neuropsychology is affected by maturational and experiential variables to a more significant degree than is typically assumed in adult neuropsychology. At the same time, methods of assessing neuropsychological functioning in children were historically rooted in the work on adult assessment (lncagnoli, Goldstein, & Golden, 1986). It is not surprising, therefore, to find that there are many clinicians who have approached child neuropsychology with techniques of assessment and models of inference that were borrowed from adult batteries or with "scaled-down" versions of adult tests, theresults of which form the data for their clinical inferences (Boll & Barth, 1981; Boll, 1983; Rourke et al., 1983).
terpretation of large numbers of clinical groups can be gathered (Tarter & Edwards, 1986; Hartlage & Telzrow, 1986). However, a fixed battery approach, particularly one that is empirically rather than theoretically based, may not be designed to assess agerelated differences sensitively in problem-solving behaviors because the emphasis is often on quantitative discrimination. Frequently educational and other experiential variables are not treated in the content of the battery itself and often the battery may not be capable of addressing specific referral questions such as prescription of remediation programs for a developmental impairment. Finally, many fixed batteries suffer from dependence on the match between the validation and cross-validation samples and the base rate of clinical problems in the sample in which it is applied. Thus, for example, batteries developed at a hospital-based referral clinic may or may not be as capable of detecting cognitive disorders in a schoolbased or psychiatry inpatient setting. In the flexible battery approach, a core set of standardized tests are administered to which are added a selected set of additional tests designed to enhance examination for specific referral questions (Rourke et a/., 1986) or as problem behaviors are elucidated from the core battery (Hartlage & Telzrow, 1986). Like the fixed battery, such an approach may be empirically derived or theoretically Assessment Methods in Child based or some mixture of the two such as when a Neuropsychology screening battery, empirically derived, is followed Broadly speaking, assessment approaches in by a theoretically based complement of tests dechild neuropsychology can be classified into three signed to test the best fit to a syndrome type. The major types: a fixed battery approach, a flexible bat- advantage of such a flexible battery is the ability of tery approach, and an individualized or patient-cen- the neuropsychologist to employ both a nomothetic tered approach. In the fixed battery approach, the as well as an ideographic approach to child assesssame set of tests, designed to tap a very broad spec- ment in order to allow for evaluation of broad dimentrum of functions and abilities, is administered to sions of behaviors as well as specific subcomponents each child regardless of the referral question. This of an ability or symptom related to a specific referral approach (reviewed in more detail in· a subsequent question. chapter by Golden) may be either empirically or theIn the individualized or patient-centered aporetically based. In the former instance, the test bat- proach to assessment (Goodglass, 1986), the set of tery is selected according to its ability to separate test procedures employed is driven by two interacting groups and is typified by the work of Reitan and his factors: the referral question that includes the child's associates. Theoretically based batteries, in contrast, history and presenting symptoms and the test perforare founded upon a theory of development as it re- mance (successes and failures) itself (Luria, 1973; lates to rather broad or narrow dimensions of behav- Christensen, 1975). More than any other approach, ior and is typified by the Florida Longitudinal Project the patient-centered assessment requires that the Battery (Satz & Morris, 1981). Fixed batteries may clinician have substantial clinical knowledge of the also include formal decision rules for clinical in- specific as well as nonspecific effects of brain lesions terpretation (Rourke, 1981; Rourke et al.. 1986). on brain development and neuropsychological funcThe major advantages of a fixed battery approach are tioning. The patient-centered approach has, as a prithe breadth and depth of functions assessed, the nor- mary goal, the isolation of a specific neurobehavioral mative data bases frequently provided, and the ease mechanism to account for the pattern of test findings. with which a systematic data base for clinical in- In this sense, the logic underlying this approach is
EVALUATING BRAIN-BEHAVIOR RELATIONSHIPS
most easily adapted for use in the strong inference model outlined earlier. A number of lesion-related factors have been shown to have a differential impact on the developing brain including: age at time of insult (e.g., prenatal versus adolescence), type of lesion (e.g., vascular versus infectious), site oflesion (e.g., primary versus association cortex), and etiology (e.g., anoxia versus trauma). As yet, however, detailed descriptions of the neuropsychological functions and organization of behaviors in many children with development or acquired neuropathologies are not yet available. As a result, the highly clinical-inferential approach of the patient-centered battery is infrequently seen in practice.
Quantitative Inferences in Child ~europsychology
There are four major inferential approaches that focus on quantitative aspects of performance in child neuropsychological assessment. These interpretive approaches derive primarily from the fixed battery methods in clinical assessment outlined by Reitan and others and focus on: (I) level of performance, (2) differential score patterns, (3) comparisons between sides of the body, and (4) comparisons of performanceacrosstime(Boll, 1983;Rourkeetal., 1986). In the level ofperformance approach, the child's performance on a variety of measures is individually compared to normative data available for agematched normal subjects or for selected clinical comparison subgroups (Satz & Morris, 1981; Spreen et al., 1984). Oneriskofthis approach is a high number of false-positive errors due to the large number of factors (e.g., psychiatric disturbance) besides brain dysfunction that can lead to poor performance. The differential score or panern approach evaluates an individual's set of performances on a battery of tests and may compare relative strengths or weaknesses in the total performance or may attempt to match a pattern or profile of scores to a clinical subtype. This is frequently done in the learning disabilities literature. This latter inferential approach is often difficult due to the fact that all measures do not equally allow for a comparison between the idealized referrant group and the individual case (Morris, Blashfield, & Satz, I 986) or due to the absence of a well-defined ideal subtype to which the individual case is compared (Boll, 1983). Comparisons between sides of the body may be made on the basis of speed of performance (e.g., finger tapping rate), skill of performance (e.g., man-
173
ual dexterity), preferred performance (e.g., handedness), or accuracy of performance on sensory or motor tasks (e.g., tactile form discrimination). Transfer of learning from one side to the other may also be compared (e.g., dominant versus nondominant hand time to complete Tactual Performance Test, Reitan & Davison, 1974). In this analysis, differential performance between the sides of the body or in transfer of learning is the basis for inferring differential functional integrity at the level of the brain (Reitan & Davison, 1974). The fourth approach, the longitudinal approach, compares performance on a battery of tests over time. In this approach, pretest versus posttest comparisons can be made of the effects of a known acquired lesion (e.g., surgical excision of an epileptic lesion), of the effects of recovery or restoration of function following an acquired lesion (e.g., recovery of memory functions subsequent to head trauma), or of the effects of a treatment intervention (e.g., pharmacological therapy for attention deficit disorder with hyperactivity). In all these quantitative approaches, the reliability and validity of the interpretations made rest on the data base available. Although normative data continue to be developed, there is still a paucity of data about the neuropsychological performance of various childhood neurologic disorders as well as a paucity of longitudinal studies of these clinical groups against which normal versus abnormal inferences can be derived.
Qualitative Inferences in Child ~europsychology
Reflecting the emphasis of American psychology on a psychometric approach, it is an unfortunate truth that only recently have developmental neurologists begun to collect data on the qualitative aspects of performance in both normal and abnormal development and attempted to relate these findings to neurologic models of brain development and to cognitive development (Fletcher & Taylor, 1984; Waber & Holmes, 1985). The focus of this approach is the emphasis on the qualitative features of performance (''how" a test is performed) rather than solely on the quantitative features of performance ("what" is achieved). By examining these qualitative features of performance, inferences are derived about the processes involved in executing a given behavioral task. These processes are then related to the differential functions of the right or left cerebral hemispheres or to other subcortical functional processes (Van der
174
CHAPTER 9
Vlugt, 1979). To some degree, this approach has also been applied to an analysis of errors in performance, for example, among different subgroups of learningdisabled children (Hynd, Obrzut, Hayes, & Becker, 1986) and to understanding the neurobehavioral deficits of autistic children (Damasio & Maurer, 1978). To some extent, this qualitative approach also underlies the pathognomic sign approach in models of developmental delay versus deficit (Hartlage & Telzrow, 1986), in the analysis of "soft" versus "hard" neurologic signs among learning-disabled children (Denckla, 1979; Shafer, Shaffer, O'Connor, & Stokman, 1983; Shaffer, O'Connor, Shafer, & Prupis, 1983), and in the search for behavioral signs of "organicity" such as rotations in drawings (Boll, 1983). In this latter context, the neuropsychologist infers that the presence of the ''sign'' indicates the presence of brain damage or brain dysfunction. As Rourke et al. (1983) pointed out, however, the risk of relying upon this approach is the likelihood of increasing false-negative errors, for the absence of sign is interpreted to reflect the absence of pathology. In fact, it is now recognized that behavioral signs of brain pathology are age-related and may appear, disappear, and reappear at different stages of development due to the late versus early effects of a lesion as well as the capacity of the developing brain to adapt and to compensate for brain pathology (Stein, Rosen, & Butters, 1974; Boll, 1983). Finally, as noted with regard to quantitative inference, qualitative inference also depends on carefully documented empirical data relating qualitative aspects of performance to normal and abnormal development, which are still not widely available to clinical practitioners (Spreen et al. , 1984).
Inferential Fallacies in Child Neuropsychology Fletcher and Taylor (1984) pointed out a number of inferential fallacies about the relationship between a child's test performance and the integrity of that child's brain. These fallacies include: (1) the differential-sensitivity fallacy, (2) the similar-skills fallacy, (3) the special-sign fallacy, and (4) the brain-behavior isomorphism fallacy. In the differential-sensitivity fallacy, it is assumed that neuropsychological test findings associated with brain lesions in adults will be useful signs of brain disease in children. Instead, Fletcher and Taylor ( 1984) argued that one must document that a particular measure is a sensitive neurobehavioral measure in children whether or not it is helpful in the
description of brain pathology in adults. Although it has been argued that the type and locus of childhood brain pathologies often do not lead to a picture of focal deficits (Boll & Barth, 1981; Boll, 1983), wherever possible, appropriate neurological and neurodiagnostic criteria must be utilized to assess the sensitivity of behavioral tests to brain pathology. The similar-skills fallacy focuses on the belief that tests developed and normed on adult subjects measure the same abilities in children. Thus, for example, age norms that step down from adult age groups to younger children have been published for such widely used adult measures as the Wechsler Memory Scale (Ivinskis, Allen, & Shaw, 1971). In this example, it is assumed that children process these task demands in a fashion similar to adult subjects, despite clear evidence of age-based differences in the capacity and processes of verbal memory in children (Kail, 1984). Other major differences between children and adults are widely recognized in such important behavioral domains as language (Segalowitz, 1983), right hemisphere functions (Witelson, 1977) and early versus later acquired reading skills and the role of the right versus the left hemisphere (Bakker, 1984). The special-sign fallacy occurs when neuropsychologists utilize specific test behaviors (e.g., rotations in drawings) as signs of brain pathology or infer from the presence of minor, accompanying or correlated signs that the major pathology from which these signs derive is present. A variant of this same fallacy occurs throug)J the overreliance on analogies between signs of CNS pathology in adults and the assumption that such signs must also mean CNS pathology in children. As Fletcher and Taylor (1984) and Boll (1983) noted, there is very little evidence that similar behavioral pathologies seen in adults and children reflect similar etiologies. We do not want to imply that lessons learned from adult neurology should not be applied to children-only that they must be applied with caution. The major advantage of such application would be to generate new hypotheses capable of being put to subsequent disconfirmatory trials. Finally, the brain-behllvior isomorphism fallacy consists of mistaking dysfunction on behavioral tests as prima facie evidence of brain dysfunction. Instead, there is a need to document that behavioral dysfunction observed on a. test is related to brain pathologies and not to other sources of variability such as experiential, socioeducational, or emotional factors. As Fletcher and Taylor ( 1984) noted, there is no simple relationship between the degree or extent of brain involvement and the degree of behavioral
EVALUATING BRAIN-BEHAVIOR RELATIONSHIPS
disorder among children with brain pathology. This fallacy is often embedded in the language employed by the child neuropsychologist such that descriptions of behavioral dysfunctions are equated inferentially to etiology (developmental delay) or to diagnosis (brain damage) (Hartlage & Telzrow, 1986). Boll ( 1983) cautioned against several other persistent myths in the field of child neuropsychology. One myth suggests that there are certain predictable characteristics of brain-damaged children rather than recognizing that the overriding effect of brain damage upon psychological functioning is to increase the variability of behavior. A second myth asserts that brain damage in children causes characteristic hyperactive motor behavior rather than recognizing that changes in simple, complex, and integrated motor skills may occur in many but not all neurologic disorders, including psychiatric/behavioral disorders without evidence of neurologic abnormality. Still another myth asserts that perceptual dysfunctions are the major difficulties produced by brain damage in children rather than recognizing that no single pattern of neuropsychological deficits is characteristic of brain damage in children. One final myth is that brain damage causes serious emotional disturbance, which ignores the complex interplay between intrapersonal factors (age, type of lesion), interpersonal factors (family, school environment), and adaptational abilities that can lead to the same spectrum of emotional and conduct disorders seen in non-brain damaged children. Although many of these myths were founded in the use of single-test behavioral indices of brain pathology, they continue to persist in part due to the need for more comprehensive data on the developental effects of brain lesions in children.
Summary In its present state, the typical approach to assessment of children for the presence of neuropsychological dysfunction involves the collation of data from developmental/clinical history with data derived from a battery of neuropsychological tests. On the basis of a quantitative analysis of the test data and to a lesser extent an analysis of the fit between qualitative aspects of test performance with known clinical subtypes as well as with historical data and presenting symptom, the child neuropsychologist then can proceed by a process of primary inference to relate test data to brain function and further, by secondary inference, assign the individual child to a classification group.
175
The context of the primary or secondary inferential process in child neuropsychology is influenced by several critical factors that include: the lack of a consistently employed neurologic model of brain development to relate to behavioral data; the differential effects of lesions according to age at insult; type and etiology of lesion on a developing brain; the need for better descriptive data on a wide variety of clinical populations in developmental neuropathologies; and the need for better specification of quantitative as well as qualitative aspects of normal and abnormal behavior. In the absence of substantive knowledge of the influence of such critical factors, inferential models in child neuropsychology are still primarily descriptive in nature. The beginning emergence of an integration of description and adequate neuropsychologic tests (Freides, 1985) with classification schemes should facilitate prescriptions for remediation and better understanding of the effects of intervention on the remediation of neurobehavioral deficits resulting from neurologic pathologies in infancy, childhood, and adolescence.
References Adams, R. D., & Victor, M. (1977). Principles of neurology. New York: McGraw-Hill. Bakker, D. 1. (1984). The brain as a dependent variable. Journal of Clinical Neuropsychology, 6, 1-16. Boll, T. 1. (1983). Neuropsychological assessment of the child: Myths, current status, and future prospects. In C. E. Walker & M. C. Roberts (Eds. ), Hamlbook of clinical child psychology (pp. 186-208). New York: Wiley-Interscience. Boll, T. J., & Barth, 1. (1981). Neuropsychology of brain damage in children. InS. B. Filskov &T. 1. Boll (Eds.),Handbookof clinical neuropsychology (Vol. I, pp. 418-452). New York: Wiley-Interscience. Bolter, 1. F., & Long, C. 1. (1985). Methodological issues in research in developmental neuropsychology. In L. C. Hartlage & C. F. Telzrow (Eds.), The neuropsychology of individual differences (pp. 42-59). New York: Plenum Press. Brunswick, E. (1956). Perception and the representative design of psychological experiments. Berkeley: University of California Press. Chadwick, 0., & Rutter, M. (1983). Syndromes of brain dysfunctions in children. In M. Rutter (Ed.), Developmental neuropsychiatry. New York: Guilford Press. Christensen, A. L. (1975). Luria's neuropsychological investigation. New York: Spectrum. Clark, C. M. (1984). Statistical models and their application in clinical neuropsychological research and practice. In S. B. Filskov & T. 1. Boll (Eds.), Hamlbook of clinical neuropsychology (Vol. II, pp. 577-605). New York; Wileylnterscience.
176
CHAPTER 9
Cronbach, L. J., Rajaratnam, N., &Gieser, G. C. (1963). Theory Meehl, P. E. (1957). When shall we use our heads instead of the of generalizability: A liberalization of reliability theory. Britformula? Journal of Counseling Psychology, 4, 268-273. ish Journal of Statistical Psychology, 16, 137-163. Meehl, P. E. (1960). The cognitive activity of the clinician. AmerDamasio, A. R., & Maurer, R. G. (1978). A neurological model ican Psychologist, 15, 19-27. for childhood autism. Archives of Neurology, 35, 777Menkes, J. H. (1985). Textbook of child neurology. Philadelphia: 786. Lea & Febiger. Dean, R. S. (1986). Foundation and rationale for neuropsychologMorris, R., Blashfield, R., & Satz, P. (1986). Developmental ical bases of individual differences. In L. C. Hartlage & C. F. classification of reading-disabled children. Journal of ExTelzrow (Eds.), The neuropsychology of individual difperimental and Clinical Neuropsychology, 8, 371-392. ferences (pp. 8-39). New York: Plenum Press. Netley, C., & Rovet, J. (1983). Relationships among brain organiDenckla, M. B. (1979). Childhood learning disabilities. In K. zation, maturation rate, and the development of verbal and Heilman & E. Valenstein (Eds.), Clinical neuropsychology nonverbal ability. In S. Segalowitz (Ed.), Language functions and brain organization (pp. 245-266). New York: Aca(pp. 535-573). New York: Oxford University Press. Fletcher, J. M., & Taylor, H. G. (1984). Neuropsychological demic Press. Pirozzolo, F. J., Campanella, D. J., Christensen, K., & Lawsonapproaches to children: Towards a developmental neuropsyKerr, K. (1981). Effects of cerebral dysfunction on neuchology. Journal of Clinical Neuropsychology, 6, 39-56. rolinguistic performance in children. Journal of Consulting Freides, D. (1985). Desirable features in neuropsychological tests. Journal of Psychopathology and Behavioral Assessment, and Clinical Psychology, 49(6), 791-806. 7(4), 351-364. Platt, J. R. (1964). Strong inference. Science, 146, 347-353. Goodglass, H. (1986). The flexible battery in neuropsychological Popper, K. R. (1959). The logic ofscientific discovery. New York: assessment. InT. lncagnoli, G. Goldstein, & C. J. Golden Harper. (Eds.), Clinical application of neuropsychological test batRasmussen, T., & Milner, B. (1977). The role of early left-brain teries (pp. 121-134). New York: Plenum Press. injury in determining lateralization of cerebral speed funcHammond, K. R., Horsch, C. J., & Todd, F. J. (1964). Analyzing tions. Annals of the New York Academy of Sciences, 299, the components of clinical inference. Psychological Review, 355-369. 71, 438-456. Reitan, R. M., & Davison, L.A. (1974). Clinical neuropsycholHartlage, L. C., & Telzrow, C. F. (1986). Neuropsychological ogy: Cu"ent status and applications. Washington, DC: assessment and intervention with children and adolescents. Winston. Sarasota, FL: Professional Resource Exchange. Ross, D. M., & Ross, S. A. (1982). Hyperactivity: Current issues. Heilman, K. & Valenstein, E. (1985). Clinical neuropsychology research and theory (2nd ed.). New York: Wiley(2nd ed.). New York: Oxford University Press. Interscience. Holt, R. R. (1968). Editor's foreward. In D. Rapaport, M. M. Rourke, B. P. (1981). Neuropsychological assessment of children Gill, and R. Schafer, Diagnostic psychological testing. New with learning disabilities. In S. B. Filskov & T. J. Boll York: International Universities Press. (Eds.), Handbook of clinical neuropsychology (Vol. I, pp. Horsch, C. J., Hammond, K. R., & Horsch, J. L. (1964). Some 453-478). New York: Wiley-lnterscience. methodological considerations in multiple-cue probability Rourke, B. P. (1983). Reading and spelling disabilities: A develstudies. Psychological Review, 71, 42-60. opmental neuropsychological perspective. In U. Kirk (Ed.). Hynd, G. W., Obrzut, J. E., Hayes, F., & Becker, M.G. (1986). Neuropsychology of language, reading, and spelling. New Neuropsychology of childhood learning disabilities. In D. York: Academic Press. Wedding, A.M. Horton, &J. Webster(Eds.), The neuropsyRourke, B. P., & Adams, K. M. (1984). Quantitative approaches to the neuropsychological assessment of children. In R. M. chology handbook: Behavioral and clinical perspectives. Berlin: Springer. Tarter & G. Goldstein (Eds. ), Advances in clinical neuropsychology (Vol. 2, pp. 79-108) ..New York: Plenum Press. Incagnoli, T., Goldstein, G., & Golden. C. J. (1986). Clinical Rourke, B. P., Bakker, D. J., Fisk, J. L., & Strang, J.D. (1983). application of neuropsychological test batteries. New York: Child neuropsychology: An introduction to theory. research Plenum Press. Ivinskis, A., Allen, S., & Shaw, E. (1971). An extension of and clinical practice. New York: Guilford Press. Rourke, B. P., Fisk, J. L., & Strang, J.D. (1986). NeuropsychoWechsler Memory Scale norms to lower age groups. Journal logical assessment of children: A treatment oriented apof Clinical Psychology, 26, 354-357. proach. New York: Guilford Press. Kail, R. (1984). The development of memory in children. San Rutter, M. (Ed.). (1983). Developmental neuropsychiatry. New Francisco: Freeman. Landy, F. J. (1986). Stamp collecting versus science: Validation York: Guilford Press. as hypothesis testing. American Psychologist, 41, ll83Satz, P., & Fletcher, J. M. (1980). Minimal brain dysfunction: An appraisal of research concepts and methods. In H. E. Rie & E. ll92. D. Rie (Eds.), Handbook of minimal brain dysfunction: A Lezak, M.D. (1983). Neuropsychological assessment (2nd ed.). New York: Oxford University Press. critical review (pp. 669-714). New York: Wiley. Satz, P., & Morris, R. (1981). Learning disability subtypes: A Luria, A. R. (1973). The working brain. New York: Basic Books. Maurer, R. G., & Damasio. A. R. (1982). Childhood autism from review. In F. J. Pirozzolo & M. Wittrock (Eds.), Neuropsythe point of view of behavioral neurology. Journal of Autism chological and cognitive processes in reading (pp. 109and Developmental Disorders, 12, 2ll-221. 141). New York: Academic Press.
EVALUATING BRAIN-BEHAVIOR RELATIONSHIPS Segalowitz, S. (Ed.). (I 983). Language functions and brain organization. New York: Academic Press. Segalowitz, S. J., & Gruber, F. A. (Eds.). (1977). Language development and neurological theory. New York: Academic Press. Shafer, S. Q., Shaffer, D., O'Connor, P. A., & Stokman, C. J. (1983). Hard thoughts on neurological soft signs. In M. Rutter (Ed.), Developmental neuropsychiatry (pp. 133-143). New York: Guilford Press. Shaffer, D., O'Connor, P. A., Shafer, S. Q., &Prupis, S. (1983). Neurologic "soft signs": Their origins and significance for behavior. In M. Rutter (Ed.), Developmental neuropsychiatry (pp. 144-163). New York: Guilford Press. Spreen, 0., Tupper, D., Risser, A., Tuokko, H .• & Edgell, D. (1984). Human developmental neuropsychology. New York: Oxford University Press. Stein, D. G., Rosen, J. F., & Butters, N. (Eds.). (1974). Plasticity and recovery offunction in the central nervous system. New York: Academic Press. Tarter, R. E., & Edwards, K. L. (1986). Neuropsychological
177
batteries. In T. Incagnoli, G. Goldstein, & C. J. Golden (Eds.), Clinical application of neuropsychological test batteries (pp. 135-153). New York: Plenum Press. Vander Vlugt, H. (1979). Aspects of normal and abnormal neuropsychological development. In M. S. Gazzaniga (Ed.), Handbook of behavioral neurobiology, 2, 99-117. Waber, D.P., & Holmes, J. M. (1985). Assessing children's copy of productions of the Rey-Osterrieth complex figure. Journal of Clinical and Experimental Psychology, 7, 264-280. Wiggins, J. S. (1973). Personality and prediction: Principles of personality assessment. Reading, MA: Addison-Wesley. Witelson, S. F. (1977). Early hemisphere specialization and interhernisphere plasticity: An empirical and theoretical review. In S. Segalowitz & F. A. Gruber (Eds.), Language development and neurological theory (pp. 213-287). New York: Academic Press. Woods, B. T., & Teuber, H. L. (1973). Early onset of complementary specialization of cerebral hemispheres in man. Transactions of the American Neurological Association, 98, 113-117.
II Neuropsychological Diagnosis
10 Halstead-Reitan Neuropsychological Test Batteries for Children NANCY L. NUSSBAUM AND ERIN D. BIGLER
The purpose of this chapter is to review the children's dardize a battery of measures to assess various asversion of the Halstead-Reitan neuropsychological pects of brain functioning in children. The major test batteries for children. The Halstead-Reitan Neu- theoretical basis of the Halstead-Reitan and the ropsychological Test Battery for Children 9 to 14 Reitan-lndiana is the proposition that behavior has years of age (Reitan & Davison, 1974) and the an organic basis. Thus, performance on behavioral Reitan-Indiana Test Battery for Children ages 5 measures can be used to assess brain functioning. through 8 (Reitan, 1969) are two children's batteries Obviously, in order to infer brain functioning based based on the adult version of the Halstead-Reitan on behavioral measures, it was necessary to validate (Reitan & Wolfsun, 1985). These two batteries will . these measures on children with known brain be discussed in terms of their development and valid- damage. ity. This discussion will be followed by a description of the measures, their administration, scoring, and the functional domains they are purported to mea- Validation Studies sure. Finally, the interpretation of test results obtained from the Halstead-Reitan and the Reitan-InThe validation of the Halstead-Reitan for childiana will be discussed. dren was first reported by Reed, Reitan, and Klove The Halstead-Reitan Neuropsychological Test (1965) for 9- to 15-year-old children with known Battery for Older Children (9 to 14) and the Reitan- brain damage, and by Klonoff, Robinson, and Indiana Neuropsychological Test Battery for Young- Thompson (1969) for 5- to 9-year-old children. er Children (5 to 8) are two of the most commonly These studies demonstrated the validity of using neuused neuropsychological test batteries for children. ropsychological variables from the Halstead-Reitan These batteries were developed by Ralph Reitan to differentiate brain-damaged from non-brainbased on the adult version of the Halstead-Reitan damaged children. Subsequently, numerous studies Neuropsychological Test Battery (Halstead, 1947; have shown the discriminant validity of the Reitan & Davison, 1974; Reitan & Wolfsun, 1985). Halstead-Reitan and the Reitan-Indiana in separatThe children's batteries include modifications and a ing children with known brain damage from nondownward extension of the adult Halstead-Reitan, brain-damaged children (Boll, 1974; Reitan, 1974; as well as the addition of some supplementary mea- Selz, 1981; Selz & Reitan, 1979a,b). For example, sures not included in the adult version (see Table 1). Boll (1974) matched 27 brain-damaged children with The reason for the development of the 27 normal children on the basis of age, sex, race, and Halstead-Reitan and the Reitan-Indiana was to stan- handedness. Significantly poorer performance by the brain-damaged group was reported for finger oscillation (dominant and nondominant), the tactual perforNANCYL. NUSSBAUM • Learning Diagnostic Center/ Aus- mance test (dominant, nondominant, and both ERIN 0. hands), finger recognition (dominant), fingertip tin Neurological Clinic, Austin, Texas 78705. BIGLER • Department of Psychology, University ofTexas at number writing (dominant and nondominant), SeaAustin, Austin, Texas 78712; and Austin Neurological Clinic, shore Rhythm Test, and Speech Sounds Perception Austin, Texas 78705. Test. Similar results were found by Reitan ( 1974) in a 181
182
CHAPTER 10
TABLE 1. Subtests of the Halstead-Reitan Neuropsychological Test Batteries for Children Halstead battery<' (9-14 years) Category test Tactual performance test Fingertapping test Speech sounds perception test Seashore rhythm test Trail-making test, A & B Sttength of grip test Sensory perceptual exam Tactile fmger localization test Fingertip number writing test Tactile fonn recognition test Aphasia screening test
Reitan-Indiana battery<' (5-8 years) Category test Tactual perfonnance test Fingertapping test
Marching test Sttength of grip test Sensory perceptual exam Tactile finger localization test Fingertip symbol writing test Tactile form recognition test Aphasia screening test Color form test Progressive figures test Matching pictures test Target test Matching figures and matching V's test Drawing of star and concentric squares
"The Wechsler Intelligence Scale for Olildren-Rcvised (Wechsler, 1974), the Wide Rallge Ac:bievemellt Test-Revised (Jastak & Wilkinson, 1984), and the Lateral Dominance Test (Harris, 1947) are often included in the comprehensive neuropsychological evaluation of children.
study with children aged 5 to 8, matched for age and sex. Furthermore, in a study of children with questionable rather than defmite neurological impairment, it was found that a neuropsychological test battery correctly identified the presence or absence of ~pairment, even when the initial subjective clinical impressions did not suggest deficits (Tsushima & Towne, 1977). Also, in a recent study (Nici & Reitan, 1986) using intellectual, achievement, and neuropsychological measures, it was found that measures of motor functioning and general neuropsychological abilities were the best discriminators of 9- to 14-year-old brain-damaged children versus nonbrain-damaged children. It should be added that although the Reitanlndiana and the Halstead-Reitan were originally developed to assess brain damage in children, they also have been used extensively to evaluate various aspects of purely behavioral functioning. As such, in general clinical practice, the assessment of functional aspects of behavior may be the most widely used application of these measures. Through the use of the Reitan-Indiana and the Halstead-Reitan, a great deal of information can be obtained concerning certain aspects of sensory functioning, motor abilities, auditory processing, attention, spatial abilities,
memory, visuospatial abilities, visuomotor abilities, conceptual processing, sequential processing, and language functioning. Therefore, although the Reitan-lndiana and the Halstead-Reitan are often used in the evaluation of organic dysfunction in children, they also have a great deal of clinical utility as measures of behavioral competencies in children. In regard to this use of the Reitan-Indiana and the Halstead-Reitan to assess behavioral functioning, Fletcher and Taylor (1984) proposed that the greatest clinical utility of these instruments is in their usefulness in defining the ability structure of the child. They argued that the widest clinical application of the children's neuropsychological batteries is related to the clinical sensitivity of these measures to the child's behavioral strengths and weaknesses. In summary, as can be seen from the previous discussion, the Reitan-Indiana and the HalsteadReitan have a wide variety of clinical applications. These test batteries can be useful in the behavioral assessment of children with known brain damage, as well as in the evaluation of the ability structure of the child without known brain damage. In the next section, types of behaviors measured, the administration and scoring of the Halstead-Reitan and the ReitanIndiana will be discussed in greater detail.
HALSTEAD-REITAN TEST BATTERIES
183
Subtests from the Halstead-Reitan Neuropsychological Test Battery for Children Ages 9 through 14
Domain Measured. The tactual performance test is a measure of tactile, motor, spatial, and memory functioning.
Nonnative data for the measures included in the Halstead-Reitan have been developed by Spreen and Gaddes (1969) and Knights and Norwood (1980).* See Tables 2 and 3 for an example of a raw score conversion table, in which raw scores are converted to standard scores (X= 100, S.D. = 15) using normative data.
Fingertapping Test
Category Test
Description. This test includes 168 items pre-
sented visually to the child. The child must respond by selecting a number ( 1, 2, 3, or 4) that corresponds with the visual stimulus. Feedback is given on each item regarding the correctness or incorrectness of the response.
Scoring. The child's raw score is the total number of errors made. Normative data are also available for each of the six subtests of the category test (Knights & Norwood, 1980). Domain Measured. The category test is a measure of concept formation. The child must abstract principles related to number concepts, spatial position, and unusualness of the stimuli. There is also a memory component involved in the last subtest of this measure. Tactual Performance Test
Description. On this test, the child is required to complete a six-figure formboard while blindfolded. The test is carried out first with the dominant hand, then the nondominant hand, and finally with both hands. Next, the board is removed and the child is asked to draw from memory the shapes and their correct locations. Scoring. The child's raw score is the amount of time taken with the dominant, nondominant, and both hands. The total time is the sum of the trials with the dominant, nondominant, and both hands. Also, the memory raw score is the total number of blocks recalled, and the location raw score is the number of blocks reproduced in their correct locations. *Address for nonnative data: Dr. Robert M. Knights Psychological Consultants, Inc., 52 Hopewell Ave., Ottawa, Ontario K15 2Y8, Canada ($6.50).
Description. This test requires a child to tap a mounted key as quickly as possible with the index finger of the dominant and nondominant hand. Five trials are given for each hand. Scoring. The child's raw score is the average of the five scores from the dominant and nondominant hand. Domain Measured. This task is a measure of fine motor speed and coordination. Speech Sounds Perception Test
Description. On this test, the child must discriminate nonsense words presented on a tape recorder. The child is given four choices to select from and must underline the correct stimulus. Scoring. The child's raw score is the total number of items correct out of 30. Domain Measured. Auditory discrimination, sound-symbol matching, and attentional abilities are assessed on this task. Seashore Rhythm Test
Description. The child is presented pairs of rhythms on a tape recorder and must discriminate whether they are the same or different. Scoring. The child's raw score is the total number of items correct out of 30. Domain Measured. This test is a measure of nonverbal auditory perception, attention, and concentration. Trails A
Description. On this test, the child must connect circles containing the numbers 1 through 15 as quickly as possible. Scoring. The child's raw score is the number of seconds taken to complete the task, and the number of errors made. Domain Measured. This task includes components measuring visual perception, motor speed, sequential skills, and symbol recognition.
184
CHAPTER tO
TABLE 2. Finger Oscillation-Dominant Handa Electric tapper Age'>
6
S.D.
30.9 3.27
x
Raw score
20 21 22 23 24
25 26 27 28 29
30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48
49
so
51 52 53 S4
55 56 57 58 59 60
•
55 59
64 68
73 78 82 87 91 96
too
lOS
110 114 119 123 128 133 137 142 146 151
**
••
•• •• •• ••
••
••
•• •• •• •• ••
**
••
••
•• **
7 33.6 3.93
•
• 56 60
63 67 71 15 79 82 86 90 94
Manual tapper
10 37.7 4.98
38.9 5.51
59 62
• 58
61 65
• •
• •
64
68 72
59 62 65
63 65
68
68
8
9
37.9 5.45
34.3 4.37
•
• 56
67 70 73 76 78 81 84
98 102
87 89
lOS
92
109 113 117 121 124 128 132 136 140 144
•• •• •• •• •• •• ••
•• ••
•• **
•• •• •• **
95 98 100 103 106 109 111 114 117 120 122 125 128 131 133 136 139 141 144
••
•• •• •• ** ••
•
75 78 82 85 89 92 96 99
102 106 109 113 116 120 123 126 130 133 137 140 144
••
•• •• •• •• •• •• ••
•
56
71 74 71 80
83 86 89 92
95 98 101 104 107 110 ll3 116 119 122 125 128 131 134 137 140 143
•• •• ••
••
•• •• ••
••
••
**
••
**
**
II
•
57
60
12 41.3 5.47
•
•
•
•
*
55 58 61
71 73 16 79 81
74
84
71
87 89 92
95 98
100 103 106 108 lll
ll4 116 119 122 125 127 130 132 135 137 141 143
**
••
••
**
••
64 66
69 72
80 83 85 88 91 94 96 99
102 lOS
107 110 113 116 118 121 124 126 129 132 135 137 140 143
•• **
••
•Standard sans (X = 100, S.D. • IS) were c:alculated based on the normative data developed by S}IRell and Gaddes (1969). •· standard score greater than three standard deviations below lbe mean; ••, standard scote greater than three standard deviations above lbe mean. •six- to eight-year-old norms are for lbe electric fin&ertappet; 9- to 12-year-old norms are for lbe manual fingertapper.
HALSTEAD-REITAN TEST BATTERIES
185
TABLE 3. Right Hand0 Fingertip symbol writing Age
X errors S.D.
5 6.00 3.00
6 3.50 2.50
7 2.70 2.00
8 2.05 1.50
130 125 120 115 110 105 100 95
121 115 109 103 97 91 85 79 73 67 61 55
120
121
113
lll
Raw score
0 I
2 3 4 5 6 7 8 9 10 11 12 13 14
90
85 80
75 70 65 60
55
IS •Standard sco~
Fingertip number writing
••
•• •• ••
105 98
90
83 75 68 60
•• **
••
•• ••
••
••
101 91 81 71 61
•• •• •• •• •• •• •• •• ••
9 4.36
3.73
117 113 109 105 101 97
93
90 86
82 78 74 70. 66
62 58
10 3.25 3.01 116 111 106 101 96
91 86 81 76 71 66
61 56
••
•• ••
11
2.75 2.00 121 113 106 98 91 83 76 68
61
••
•• •• •• •• ••
••
12 2.00 1.00
13 0.50
14 0.50 0.40
113 86 59
119 81
o.ss
130 115 100 85 70 55
••
•• •• •• ••
•• •• ••
•• ••
•• •• •• ••
••
•• •• •• ** •• •• ••
••
••
•• •• •• •• ••
•• ••
•• •• *"' ••
•• ••
100, S.D. = 1') were calculated based on tbe normative data deY!'loped by Knights and **• standanl ~greater diaD duee standanl deviations below tbe mean.
(X =
Norwood (1980).
Trails B
Sensory Perceptual Exam
Description. This test requires the child to connect alternating letters (A to G) and numbers (l to 8).
Tactile, auditory, and visual perception are measured by this task.
Scoring. As in Trails A, the child's raw score is the number of seconds taken to complete the task, and the number of errors made.
Tactile Perception. The child is asked to close the eyes and to report whether the right hand, left hand, right face, or left face is being lightly touched. Following unilateral trials to determine whether the Domain Measured. Trails B also measures vi- child can perceive unilateral stimulation, bilateral trisual perception, motor speed, sequencing skills, and als are randomly interspersed with unilateral trials. symbol recognition. It also is a measure of simulBilateral trials include the stimulation of both hands taneous processing and cognitive flexibility. and contralateral stimulation of the hand and face, the so-called double simultaneous stimulation (DSS) procedure. (In addition, we have found it useful to Strength of Grip Test include ipsilateral hand/face trials. This appears to Description. The child's grip strength is mea- require more sensitive tactile discrimination.) sured using a dynamometer adjusted for hand size. Scoring. A raw score for each body side is calThree alternating trials are allowed for the dominant culated by summing the total number of errors made and nondominant hand. on unilateral and bilateral trials. (Ipsilateral errors are Scoring. A mean raw score is obtained for each not included in this raw score when converting raw hand. scores to standard scores.) Domain Measured. Differential hand strength is assessed by this measure.
Domain Measured. Differential tactile perception is measured by this task.
186
CHAPTER 10
Auditory Perception. The examiner lightly rubs his/her fingers together by the child's right ear, left ear, or both ears. The child is asked to close his eyes and to report where the sound is coming from. Following unilateral trials to determine whether the child can perceive unilateral stimuli, bilateral trials are randomly interspersed with unilateral trials. Bilateral stimulation constitutes the DSS procedure for auditory stimulation. Scoring. A raw score for each ear is calculated by summing the total number of errors made on unilateral and bilateral trials. Visual Perception. The child's visual fields are tested by quadrant. Then the child is asked to report peripheral, unilateral, and bilateral single movements by the examiner at eye level, above eye level, and below eye level. Bilateral stimulation constitutes the DSS procedure for visual stimulation. Scoring. A raw score for the right and left visual fields are calculated by summing the total number of errors made on unilateral and bilateral trials. Domain Measured. Differential visual perception in the right and left visual fields is measured by this task. Tactile Finger Localization Test
Description. With the child observing, the examiner numbers the child's fingers. The child must then close the eyes and report the number of the finger being stimulated. Scoring. The raw score is the sum of the errors made on each hand. Domain Measured. This test is a measure of tactile perception, tactile localization, and attention for each body side. Fingertip Number Writing Test
Description. The child watches while the ex-
aminer traces the numbers 3, 4, 5, and 6 on the palm of the child's hand. The child is then asked to close the eyes and report the number written in a set order on the fingertips of the right and left hands.
Scoring. The raw score is the sum of the errors made on each hand. Domain Measured. Aspects of complex tactile perception and concentration are assessed by this task for each body side.
Tactile Form Recognition Test
Description. The child places a hand through an opening in a board, and the examiner places either a small cross, triangle, square, or circle in the child's hand without the child seeing what object has been placed there. The child must then point to the correct object on the board with the other hand. The task is carried out two times with each hand. Scoring. The raw score is the sum of errors made with each hand, and the number of seconds taken to identify the object with the right and the left hand. Domain Measured. This test is a measure of attention, tactile perception, and reaction time for each body side. Aphasia Screening Test
Description. This test includes 32 items requiring naming, copying, spelling, reading, writing, repeating, verbal comprehension, and right/left discrimination. Scoring. Selz and Reitan (l979b) developed a scoring system for the items from the Aphasia Screening Test. Domain Measured. The Aphasia Screening Test is a useful screening measure for dyspraxia (both spelling and constructional), dysnomia, dysgraphia, dyslexia, dyscalculia, ideational dyspraxia, expressive aphasia, receptive aphasia, dysarticulation, visual dysgnosia, auditory dysgnosia, and right/left disorientation.
Subtests from the Reitan-Indiana Neuropsychological Test Battery for Children 5 through 8 The Reitan-lndiana (Reitan, 1969) is a downward extension of the Halstead-Reitan Neuropsychological Test Battery (see Reitan & Wolfsun, 1985) developed for children 5 to 8 years of age. The modifications discussed below were necessary because of the developmental differences found between younger and older children (Reitan & Davison, 1974; Boll, 1981). In addition, as in the older children's battery, normative data for measures included in the Reitan-Indiana have been developed by Spreen and Gaddes ( 1969) and Knights and Norwood (1980).
HALSTEAD-REITAN TEST BATTERIES
Category Tests
Description. The number of test items was reduced to 80 items in five categories. On the first subtest the child is required to identify colors by selecting a corresponding lever. The following four subtests involve principles of size, shape, color, or memory. Scoring. (See the previous description of scoring for the Category Test.) Domain
Measured. Nonverbal
reasoning, learning, memory, and concept formation are assessed by this task.
Tactual Performance Test
Description. The same six-figure formboard employed in the older children's battery is used on this test, but the board is presented horizontally. (See the previous discussion of the tactual performance test for a more detailed description of the task, the scoring, and the abilities measured.)
187
Sensory Perceptual Exam (SPE). The tactile and auditory tasks of the SPE remain unchanged. Only a minor modification of the visual task has been made requiring that visual stimulation be presented at eye level only. In addition, if the young child has difficulty verbally reporting the body part touched or the side of stimulation (i.e., right, left), they are allowed to point to or raise their hand on the side of stimulation. Tactile Finger Localization. The procedure for this task remains unchanged from the older children's battery. Fingertip Symbol Writing. This task is very similar to the Fingertip Number Writing Test from the older children's battery. The child is required to close the eyes and report which symbols (X's and O's) are written on the child's fingertips. (See the older children's battery for a description of the scoring procedure and the domain measured.) Tactile Form Recognition. The procedure for this task remains unchanged from the older children's battery.
Fingertapping Test
Description. The same procedure is used with
Aphasia Screening Test
younger children, as was described for older children, except an electric fingertapping device is used to compensate for the younger child's poorer fine motor coordination. (The scoring and domain measured are the same as described previously,)
Description. Some items were deleted or simplified for the younger children. Items included in this test require the child to write, copy simple geometric figures, identify pictures, read letters and simple words, carry out simple mathematical functions, identify body parts, and identify right/left body side.
Speedt Sounds Perception Test, Seashore Rhythm Test, Trails A and B
Scoring. Knights and Norwood ( 1979) developed a scoring system for the younger children's version of the Aphasia Screening Test.
These tests were omitted in the younger children's battery.
Marching Test
Description. On the Marching Test the child is required to follow a sequence of circles connected by lines up a page, by touching each circle as quickly as possible. Scoring. The raw score is the number of errors
and time taken to complete the task.
Domain Measured. This test is a measure of upper extremity gross motor functioning and coordination. Strength of Grip. This test remains unchanged
from the older children's battery.
Domain Measured. This is a screening instrument for constructional dyspraxia, dysgraphia, dyslexia, dysnomia, dyscalculia, right/left disorientation, receptive aphasia, expressive aphasia, visual dysgnosia, and body dysgnosia. Color Form Test
Description. A board is presented with different colored geometric shapes. The child must alternate between touching shapes and colors selectively attending to one aspect of the stimulus (e.g., color), while ignoring the other (e.g., shape). Scoring. The child's raw score is the total number of errors made and the amount of time taken to complete the task.
CHAPTER 10
188
Domain Measured. This test is a measure of attention, ability to inhibit, visual sequencing, cognitive flexibility, and upper extremity gross motor coordination. Progressive Figures Test
Description. The child is presented a sheet of
paper on which are printed eight large geometric shapes (e.g., circle), with a smaller shape (e.g., square) inside. The child must use the small inside figure as a cue for moving to the outside shape of the next figure.
Scoring. The child's raw score is the total
number of errors made and the amount of time taken to complete the task.
Domain Measured. This test measures visual perception, motor speed, cognitive flexibility, attention, and concentration. Matching Pictures Test
Scoring. The raw score is the number of errors and time taken to complete the task. Domain Measured. This test is a measure of visual perception and reaction time. Drawing of Star and Concentric Squares
Description. The child must copy figures of varying complexity. Scoring. The raw score is the number of errors and time taken to complete the task. Domain Measured. This test is a measure of visual perception, fine motor coordination, and constructional praxis.
Interpretation of Children's Performance on Neuropsychological Batteries In the past, a number of approaches have been
Description. The child must match pictures be- applied to the interpretation of children's perfor-
ginning with identical pictures and progressing to matching pictures from more general categories.
Scoring. The raw score is the total correct out of a possible 19.
mance on neuropsychological tasks (Reitan & Davison, 1974; Selz, 1981; Rourke, Bakker, Fisk, & Strang, 1983; Fletcher & Taylor, 1984; Teeter, 1986). These various approaches and their limitations will be discussed below.
Domain Measured. This test is a measure of visual discrimination, reasoning, and categorizing Level of Performance skills. Target Test
Description. The child is shown an 18-inch card with nine dots printed on it, and is given a sheet with the same dot configuration. The examiner taps out a design on the stimulus card and the child must draw the design on the response sheet. Scoring. The raw score is the number of items correct.
Domain Measured. Visual/spatial memory is
measured by this task.
Matching Figures and Matching V's Test
Description. On the Matching Figures Test the child must match figures printed on blocks with figures printed on a card. The figures become progressively more complex. The matching V's task requires the child to match V's that vary in the width of the angle.
In this approach, the child's level of performance on neuropsychological measures is compared to normative data, such as the norms developed by Spreen and Gaddes (1969), and Knights and Norwood (1980). If the child's performance falls significantly below (e.g., two standard deviations below the mean) what would be expected for their age, then a deficit is diagnosed in the particular area measured by the neuropsychological task. For example, on the fmgertapping test, the mean performance for a 9year-old is 34.3 (S.D. = 4.37) (Spreen & Gaddes, 1969) for the dominant hand. Using the level of performance approach, if a 9-year-old obtained a score of25, their fine motor speed with the dominant hand would be interpreted as impaired. There are a number of disadvantages to relying solely on this method for the interpretation of children's neuropsychological test performance. First, because of the large variability in normal children's performance on neuropsychological measures, it is often difficult to interpret the individual child's
HALSTEAD-REITAN TEST BATTERIES
scores. For example, the mean performance for an 8year-old on the Tactual Performance Test (dominant hand) is 5. 71 minutes but the standard deviation of 6 minutes is greater than the mean (Knights & Norwood, 1980)! If a two-standard deviation cutoff is adhered to for interpreting a deficient level of performance on this task, the child would be required to continue for over 17 minutes in order to diagnose impaired performance. Another problem that is encountered when using only the level of performance to interpret children's neuropsychological performance is the tendency to yield a large number of false-positives. This is largely because so many factors other than impairment can negatively influence the child's performance (e.g., motivation, attention, frustration tolerance) (Rourke et at., 1983).
Pathognomonic Signs The pathognomonic signs (e.g., hemiplegia, aphasia, hemisensory deficit) approach refers to the identification of specific deficits that are not commonly seen in normal individuals. For example, articulation errors on the Aphasia Screening Test could be interpreted as an aphasic "sign." Again, one of the limitations of this approach would be the large variability seen in the normal population. For example, because of the wide variation in the development oflanguage abilities in children, it is often quite difficult to interpret specific errors made by the individual child. There may be a tendency to interpret as abnormal an isolated error made by the child. Conversely, it has also been found that the sole use of the pathognomonic signs approach has a tendency to yield a large number of false-negatives. For example, Boll (1974) found that 26 of 35 children with known brain damage were misclassified as normal when the sign approach was used to classify subjects. One reason for this is that such conditions as infantile hemiplegia may show considerable plasticity and recovery of function (see Bigler & Naugle, 1985), and their neuropsychological test performance may not conform with what would be expected with a more recently acquired brain lesion.
Patterns of Performance In this approach, the relationship among performance on neuropsychological measures is examined. If large discrepancies are noted, then strengths and weaknesses are interpreted. For example, a child who does very well on the Speech Sounds Perception Test and the verbal items from the Aphasia Screening
189
Test, but who performs poorly on the constructional tasks of the Aphasia Screening Test and poorly on the Tactual Performance Test Memory and Locations subtests, may be interpreted as having adequate auditory processing, but deficits in the area of nonverbal, visual/spatial functioning. This approach is of limited use in the interpretation of neuropsychological performance of very young children and children with severe disabilities (Rourke et al., 1983). However, it has been used quite extensively in the subgrouping of children with learning disabilities (Rourke, 1984; Nussbaum & Bigler, 1986).
Comparison of Right and Left Body Sides This approach compares the relative performance of one body side to the other on motor and sensory-perceptual tasks. For example, a large discrepancy between the right (dominant) and the left (nondominant) hand on the fingertapping test, with the right-hand score less than the left-hand score, may indicate impairment in left-hemisphere functioning. However, once again, because of the wide variability in the performance of normal children, discrepancies between performance on the right and left body sides may often be difficult to interpret for the individual child. This is especially true of the younger child.
Multiple Inferential Approach Boll (1974, 1981) proposed a multiple inferential approach to the interpretation of children's neuropsychological performance because of the limitations of the isolated use of the approaches discussed previously. In the multiple inferential method, the complimentary use of level of performance, pathognomonic signs, pattern analysis, and right/left comparisons is employed in the interpretation of neuropsychological performance. This approach minimizes the limitations that are encountered when these methods are used in isolation.
Rules Approach The rules approach developed by Selz and Reitan ( 1979b) also combines a number of inferential methods in the interpretation of children's neuropsychological performance. The rules are based ori the four methods of inference discussed earlier: level of performance, pathognomonic signs, pattern analysis, and right/left differences.
190
CHAPTER 10
The rules were initially developed and validated as an objective system for classifying children as normal, learning disabled, or brain damaged. They consist of a four-point scoring system in which 37 aspects of neuropsychological performance are rated on a scale from 0 (normal to excellent performance) to 3 (very abnormal performance). For example, a child who made nine or more errors on the Tactile Finger Recognition task would receive a score of 3 (impaired) using the rules system. In summary, the rules approach is an attempt at providing an objective system to measure the degree of impairment in the neuropsychological performance of children.
Functional Organizational Approach The functional organizational approach is another method that has been proposed for the interpretation of children's neuropsychological performance (Fletcher & Taylor, 1984). In this model, three variables are taken into consideration: (1) the manifest form of the disability [e.g., the impulsive behavior of the child with attention deficit disorder (ADD)]; (2) the behavioral and cognitive correlates of the manifest disability (e.g., performance on neuropsychological measures with an attentional component); (3) the biological substrate underlying the covariance of the first two variables (e.g., neurochemical imbalance, structural abnormality, lag in cortical maturation). This model avoids the tenuous inference involved in the interpretation of direct brain-behavior relationships in children using behavioral data. Rather, the biological or neurological substrate is seen as influencing the manifest disability by imposing limits on the basic behavioral competences of the child. Additional moderator variables, such as the family system and the educational setting, are also taken into consideration. Furthermore, the relationship between performance on neuropsychological measures and the manifest disability is not seen as causal but instead as correlational. Thus, performance on neuropsychological measures is used to clarify various functional aspects of the child's manifest disability. For example, the ADD child's performance on neuropsychological measures can help to determine the degree of cognitive impulsivity present in the disorder. In this example the functional aspects of the child's performance would be stressed in the interpretation of the neuropsychological data. Also the importance of interpreting neuropsychological data in a developmental context is emphasized in this model.
Summary In summary, the Halstead-Reitan and the Reitan-Indiana were developed by Reitan (Reitan & Davison, 1974) based on the adult version of the Halstead-Reitan Neuropsychological Battery (Reitan & Wolfsun, 1985). Since the initial modifications of the Halstead-Reitan Neuropsychological Battery, a great deal of work has been done in the developing normative data (Spreen & Gaddes, 1969; Knights & Norwood, 1980) and interpretive models (Fletcher & Taylor, 1984; Obrzut & Hynd, 1986) for the children's neuropsychological batteries. In this chapter the various measures included in the Halstead-Reitan and the Reitan-Indiana were reviewed. A description of each measure was given, as well as a brief discussion of scoring and domains measured by the various tests. Finally, a number of approaches to the interpretation of children's neuropsychological performance were reviewed.
References Bigler, E. D., & Naugle, R. I. (1985). Case studies in cerebral plasticity. Clinical Neuropsychology, 7, 12-23. Boll, T. J. (1974). Behavioral correlates of cerebral damage in children age 9-14. In R. M. Reitan & L.A. Davison (Eds.), Clinical neuropsychology: Current suztus and applications. Washington, DC: Winston. Boll, T. J. (1981). The Halstead-Reitan neuropsychological battery. InS. Filskov & T. J. Boll (Eds.), Handbook of clinical neuropsychology. New York: Wiley-Interscience. Fletcher, J. M., & Taylor, H. G. (1984). Neuropsychological approaches to children: Towards a developmental neuropsychology. Journal of Clinical Neuropsychology, 6(1), 39-56. Halstead, W. C. (1947). Brain and intelligence: A quantitative study of the frontal lobes. Chicago: University of Chicago Press. Harris, A. J. (1947). Ha"is test of lateral dominance, manual of directions for administration and interpretation. New York: Psychological Corporation. Jastak, S., & Wilkinson, G. J. (1984). The Wide Range Achievement Test-Revised. Wilmington, DE: Jastak. Klonoff, H., Robinson, J. C., & Thompson, G. (1969). Acute and chronic brain syndromes in children. Developmental Medicine and Child Neurology, 11. 198-213. Knights, R. M., & Norwood, J. A. (1979). Neuropsychological test battery for children: Exmniner' s manual. Ottawa: Knights Psychological Consultants. Knights, R. M., & Norwood, J. A. (1980). Revised smooth normotive dauz on the neuropsychological test battery for children. Ottawa: Knights Psychological Consultants. Nici, J., & Reitan, R. M. (1986). Patterns of neuropsychological abilities in brain-disordered versus normal children. Journal of Consulting and Clinical Psychology, 54, 542-545.
HALSTEAD-REITAN TEST BATTERIES
191
Nussbaum, N. L., & Bigler, E. D. (1986). Neuropsychological Child neuropsychology: An introduction to theory, research and behavioral profiles of empirically derived subgroups of and clinical practice. New York: Guilford Press. learning disabled children. International Journal of Clinical Selz, M. (1981). Halstead-Reitan neuropsychological test battery Neuropsychology, 8, 82-89. for children. In G. W. Hynd&J. E. Obrzut(Eds.),NeuropsyObrzut,J. E., &.Hynd, G. W. (Eds.). (1986). Child neuropsycholchological assessment and the school-age child: Issues and ogy (Vol. II). New York: Academic Press. procedures. New York: Grone & Stratton. Reed, H. B. C., Reitan, R. M., & Klove, H. (1965). Influence of Selz, M., & Reitan, R. M. (1979a). Neuropsychological test percerebral lesions on psychological test performances of older formance of normal, learning-disabled and brain-damaged children. Journal of Consulting Psychology, 29, 247older children. Journal ofNervous and Mental Disease, 167, 251. 298-302. Reitan, R. M. ( 1969). Manual for administration of neuropsycho- Selz, M., & Reitan, R. M. (1979b). Rules for neuropsychological logical test batteries for adults and children. Indianapolis, diagnosis: Classification of brain function in older children. IN: Author. Journal of Consulting and Clinical Psychowgy, 47, 258Reitan, R. M. (1974). Psychological effects of cerebral lesions in 264. children of early school age. In R. M. Reitan & L. A. Davison Spreen, 0., & Gaddes, W. (1969). Developmental norms for fif. (Eds.), Clinical neuropsychology: Current status and apteen neuropsychological tests for ages 6 to 15. Conex, 5, plications. Washington, DC: Winston. 171-191. Reitan, R. M., & Davison, L. A. (Eds.). (1974). Clinical neuro- Teeter, P. A. (1986). Standard neuropsychological batteries for psychology: Current status and applications. Washington, children. InJ. E. Obrzut&G. W. Hynd (Eds.), Child neuroDC: Winston. psychology (Vol. II). New York: Academic Press. Reitan, R., & Wolfsun, D. (1985). The Halstead-Reitan neuro- Tsushima, W. T., & Towne, W. S. (1977). Neuropsychological psychological test battery: Theory and clinical interpretllabilities of young children with questionable brain disorders.. tion. Tuscon, AZ: Neuropsychology Press. Journal of Consulting and Clinical Psychology, 45, 757Rourke, B. P. (Ed.). (1984). Subtype analysis of learning dis762. abilities. New York: Guilford Press. Wechsler, D. (1974). Wechsler Intelligence Scale for ChildrenRourke, B. P., Bakker, D. J., Fisk, J. L., ~Strang, J.D. (1983). Revised. New York: Psychological Corporation.
11 The Nebraska Neuropsychological Children's Battery CHARLES]. GOLDEN
The Nebraska Neuropsychological Children's Batteries: Current Status and Applications For the past 7 years, researchers at the University of Nebraska and elsewhere have been attempting to develop objective neuropsychological test batteries for children that draw on some of the ideas and techniques employed by A. R. Luria. Development of the children's battery was heavily influenced by the earlier development of the adult Luria-Nebraska Neuropsychological Battery (LNNB) as well as the work done by Lawrence Majovski. Research on the battery has been discussed in the literature (e.g., Plaisted, Gustavson, Wilkening, & Golden, 1983) as well as the test manual (Golden, 1986) in great detail, and therefore the present chapter will focus primarily on a more important issue: the interpretive strategies used with the battery. Specifically, the present chapter will look at several issues: (I) the process of development of the battery; (2) a description of the battery with a brief review of current research; (3) methods of interpreting the test battery; and (4) current research studies aimed at improving the test battery and extending its usefulness to lower ages.
Development of the Battery The original development of the battery was begun by administering the adult LNNB to children from ages 5 to 12. It was discovered that children 8 and up could do a majority of the procedures used in CHARLES J. GOLDEN • Departments of Psychology. Sociology, and Anthropology, Drexel University, Philadelphia, Pennsylvania 19104.
the adult battery. It was also found that those items in general corresponded to those that one would expect from Luria's theories on brain development. It was also found that below age 8, drastic changes were needed in the battery content to have a useful test. Thus, it was initially decided to develop a test down to age 8. Similarly, it was found that 13- and 14-yearolds could perform perfectly normally on the adult battery (which was originally intended to extend down to age 15). At the 12-year-old level, children began to show difficulties with the adult battery (although above-average 12-year-olds can also perform perfectly normally). Thus, it was decided that the adult battery could be used down to age 13 and that the new children's battery should aim at ages 8 to 12. A battery for younger children was postponed until completion of this work and is described in the last section of this chapter. Items were deleted from the adult battery that appeared to be too difficult for initial normative youngsters in this age range. When possible, similar but easier items were substituted. We were also privileged to consult with Dr. Lawrence Majovski who was working on developing a qualitative approach to the assessment of children based on his studies with Luria. We were able to adapt and add several additional items and areas of examination to the test from his suggestions. This initial work consisted of three successive versions of the test, which were evaluated on groups of normal and impaired children until the fourth and published version was completed.
Description of the Battery The final version of the children's battery consisted of 11 basic scales, just as the adult battery does, and 149 procedures. However, most of these 193
194
CHAPTER 11
procedures consist of numerous items so that the actual number of items exceeds 500. Administration time takes about 1! to 3 hours depending on levels of cooperation and levels of impairment. This version was given to 125 children. The group consisted of 25 normal children at each of five age levels: 8, 9, 10, 11, and 12. Performance norms were developed on each procedure and scale based on the performance of this group. The first task was the development of scale scores for each item. It was decided to have different scale scores for each age group on each item so that scores for a given individual could be directly compared. For each procedure, a score of 0 was set to mean a performance within one standard deviation of the mean score for the age group. A score of 1 represented a performance between one and two standard deviations below the mean, and a score of 2 represented scores more than two standard deviations below the mean. Each item of the test was assigned to 1 of the 11 basic scales. Originally, this was done on the basis of our experience with the adult battery and on our theoretical belief of where items should load. From this assignment, scale raw scores were calculated by adding up the scaled score on each procedure to yield a total raw score. Procedures were then correlated with each of the raw scores to ensure that procedures correlated highest with the scale they were assigned to, so that items could be reorganized when necessary. After final scale assignments were determined, scale T scores were generated by first calculating the means and standard deviations of each of the original 125 normal subjects. An ANOV A for each scale score by age indicated no significant differences between the scale mean scores for each age group, and F tests indicated no significant differences among group variances. As a result, the conversion of scale raw scores to T scores was done on the basis of all125 subjects rather than for each age group alone. Each of the 11 scales is multifactorial in structure. This was done for several reasons. First, each scale was conceived not as covering a specific skill but rather as a domain of skills in a given area (such as motor function). Second, this allowed the test to yield stable test scores (which is related to the number of items on the scale as well as the individual stabilities of the items) with fewer items in each skill area. This has the positive effect of allowing for a broader coverage of skills in a reasonable period of time. This has the drawback, however, of not covering any one area in as much detail as possible. This is remedied by simply following the LNNB with specif-
ically selected testing in areas in which more information is needed after examining the LNNB performance, and by using qualitative observations to enrich the data generated by quantitative analysis alone. Although we will describe the test scales in somewhat greater detail later, we present a synopsis of the scales here for those readers not familiar with the child or adult battery. Scale Cl (Motor): This scale is designed to measure basic fine motor speed, unilateral and bilateral coordination, imitation skills, verbal control of motor movements, and construction skills. Scale C2 (Rhythm): Items on this scale evaluate the children's ability to make simple tonal discriminations, maintain a melodic pattern in singing, counting tones, and reproduction of rhythmic patterns. Scale C3 (Tactile): This scale assesses tactile sensitivity through localization of stimuli, two-point discrimination, pinprick and pressure sensation, movement detection, and stereognostic skills. Scale C4 (Visual): Items evaluate visual recognition skills, identification of altered pictures, closure, and the use of spatial relationships. Scale C5 (Receptive): This examines the ability to discriminate phonemes, analyze simple sounds and words, and follow increasingly complex instructions. Scale C6 (Expressive): This scale examines the ability to pronounce words and sounds properly, repetition of sentences, naming, automatic speech, and the use of complex expressive structures. Scale C7 (Writing): These items measure basic copying skills, the ability to analyze letter sequence, the ability to write from dictation, and the ability to spell. Scale C8 (Reading): This scale evaluates basic reading recognition skills as well as sound synthesis. Scale C9 (Arithmetic): The items assess basic arithmetic skills from number recognition through multiplication and simple algebra. Scale CIO (Memory): This scale measures short-term verbal and nonverbal memory with and without interference. Scale Cll (Intelligence): This scale is comprised of items similar to the Wechsler Intelligence Scale for Children (WISC) Picture Arrangement, Picture Completion, Vocabulary, Comprehension, Arithmetic, and Similarities subtest items. Other items measure basic ability to generalize and make deductions from data. In addition to the basic scales, the 149 items were factor analyzed in a population of 719 brain-
THE NEBRASKA NEUROPSYCHOLOGICAL CHILDREN'S BATTERY
damaged and notmal children. The resulting factors were impressive in that few of the factor scales used items from multiple scales, suggesting that item placement was essentially correct. Some factor scales simply repeated what the regular scales already yielded, whereas some failed to achieve reasonable stability. Those scales that were both stable and yielded new information were kept for further study. A second analysis involved the factor analysis of each scale alone. Many of the resultant factors duplicated factors found in the first analysis and were discarded, as were factors that were insufficiently stable. At the end of this process, II additional scales were derived, 2 of which were cross scale factors and 9 of which were intrascale. For each of these 11 scales, T scores were derived on the basis of the performance of 240 normal children in the overall sample. The 11 factors were briefly described as: (1) academic achievement, the largest cross scale factor including reading, writing, arithmetic, and expressive items; (2) spatial organization, the second major cross scale factor; (3) purposeful unspeeded movement; (4) motor speed; (5) drawing quality; (6) drawing time; (7) rhythm perception and reproduction; (8) basic tactile function; (9) basic receptive language skills; (10) repetition; and (11) abstract verbal skills. At present, other scales are also in the process of development, including scales to measure left and right hemisphere sensorimotor function and scales for the analysis of chronicity of disorders. However, there are no localization scales planned for the children's battery as with the adult battery because localized lesions do not generally cause consistent deficits in children due to the interaction of developmental processes with brain development and the time and location of the injury. For those readers interested in the details of general research on the battery, the test manual (Golden, 1986) offers the most complete and detailed account of this work. Other reviews may be found in Plaisted eta/. (1983). In general, research has examined the ability of the test to discriminate between braindamaged and normal subjects (with hit rates of about 86%), and other studies have examined correlations between the LNNB and such tests as the PIAT and the WISC-R. As reported in the test manual, this work has generally confirmed the validity of the LNNB scales. Other work has evaluated the effectiveness of the test with such groups as children who are learning disabled or who have epilepsy.
195
Interpretation Of prime importance with any test battery are the methods of interpreting the battery. This is especially important with the LNNB-CR for many because the process is often different from the procedures used with other tests. In interpreting the LNNB-CR, little confidence is placed in formal interpretations of elevations on individual scales. A major reason for this is that these are clearly not homogeneous scales with many items intended to measure just one ability but rather a domain of abilities. As a result, a single interpretation of an elevation on any particular scale would be ludicrous and possibly lead to obscure diagnostic errors. Thus, pattern analysis of the scales and items, combined with a qualitative analysis of the test performance, is the major approach to interpreting scale profiles.
Levels of Interpretation When interpreting the LNNB-CR, or other similar batteries, it is important to be aware that the many levels on which the battery can be interpreted depend on the needs, as well as the skill and knowledge, of the user. The first level is primarily concerned with ascertaining whether significant brain injury exists in a given child as a screening procedure to differentiate neuropsychological from other possible disorders. This level is not appropriate with children for whom there has been known significant brain injury, as the question is clearly unnecessary at that point. Rather, this level of interpretation is used most often when there may have been a brain injury, in screening children for a possible cause of unusual behavior, or when there has been a brain injury to which there are attributed no "significant" effects on neurological examination. This level is used most often by individuals who are not neuropsychologists and is the level most frequently discussed and taught in programs and internships in clinical, school, and counseling psychology. When there is a high likelihood of brain damage in cases screened in this manner, it is imperative that the child be referred for a more sophisticated evaluation of the data. On the other hand, because most children are seen initially by nonneuropsychologists, this level of practice is quite important in the screening to determine which children should be referred for further evaluation. The second level of interpretation involves sim-
196
CHAPTER 11
pie description of what the child can and cannot do, without drawing any conclusions or reaching any integrative statements. The third level of interpretation takes the second level to the next logical step: identification of the probable underlying causes of the child's overall behavior. This step requires extensive understanding of the various brain-behavior relationships. The interpretation process generally evolves from a theoretical orientation of how the brain functions and the ways in which information is processed through the central nervous system. Finally, the fourth level of interpretation involves the integration of all findings and conclusions into a description of how the brain of the individual is functioning. This is a difficult task in most cases, because the result of brain damage is affected by a variety of factors including: (1) the combination of areas injured; (2) when the injury occurred; (3) the degree to which alternate functional systems are available or are later formed (spontaneously or through training); (4) the kind of injury; (5) the emotional reaction of the child; (6) the adequacy of the examination situation; (7) the stimuli presented; (8) the time of day the testing was done; (9) the organization of the brain before brain injury; ( 10) the nature of brain dominance in the individual; and (11) the presence of subcortical or peripheral impairment. In addition, any conclusions can be complicated by a myriad of factors that can radically alter one's decisions. In many cases, the last level of interpretation concerns the understanding of basic underlying deficits rather than simply determining location. In general, localization is useful not so much because of a need to know exactly where the lesion is but because it may allow the generation of hypotheses about the child's problems and future that can be subsequently evaluated and related to known patterns of recovery in specific disorders. These hypotheses can be tested by observation of the child's behavior.lt is this level that is ultimately important; to know where a lesion is but to be wrong about its effects is useless. It is much more useful to understand the effects and implications of the injury and to be concerned less about their anatomical localization. In most cases, lesion localization is only an intermediate step useful in hypothesis generation. It should be noted, however, that lesion localization and description is useful in itself in some cases. Primary among these are legal cases in which it is essential to know whether a given deficit or deficit pattern can reasonably be related to a given injury or injuries. These conclusions should be left to expert neuropsychologists. This is especially true
when working with children where localization is problematical.
Identifying Brain Damage
Use of the Critical Level Adjusting for Age. The first step in identifying
when a profile is statistically abnormal and likely to be indicative of brain damage is based on establishing a valid critical level for the child. The critical level represents the highest LNNB-CR score that can be considered normal for the battery. In contrast to some other tests, this cutoff level is variable with the LNNB-CR, and is adjusted for age. The computerized scoring program for the LNNB-CR calculates the critical level. If the handscoring version of Form I is used, the critical level may be calculated using the following formula: Critical level= 82.02- (0.14 x age in months)
Identifying Deviant Scores. Once the critical level has been established, determining the probability of brain damage is relatively simple. The number of scales on the battery that exceed the criticallevel is counted, yielding the number of abnormal scores. The scores that are considered at this point are the basic clinical scales (Cl through Cll).
In general, three or more scores above the critical level are thought to be indicative of brain damage, whereas zero or only one elevated scale suggests the probable absence of brain damage. If the critical
level has been chosen correctly, the accuracy of this decision is about 75 to 85% of all cases. In addition to the problems in setting the educational level in some cases, there are also important exceptions to the rules for determining the presence or absence of brain damage. There are cases that have two scale elevations. In such cases, the profile is tentatively considered borderline. Before labeling anyone as brain damaged, even in a relatively straightforward case, one should carefully consider the effects of such a diagnosis on the individual. Such persons or their protocols should generally be referred to an experienced neuropsychologist for diagnosis. In any case, any diagnosis of brain damage in a profile should be qualified to reflect the neuropsychological sophistication of the person using the test; as noted earlier, use of neuropsychological tests does not make one a neuropsychologist. It is much safer to avoid diagnoses in reports as much as possible and to use the finding of
THE NEBRASKA NEUROPSYCHOLOGICAL CHILDREN'S BATTERY
possible brain dysfunction as a reason for further referral of the protocol to an interpreter with more expertise.
Interpreting Scale Patterns
Factors Affecting Scale Interpretation It is important to recognize that injuries in any part of the brain can potentially affect the scores on any of the scales of the battery. This is due to the relatively homogeneous content of the scales with respect to secondary skills that are measured in conjunction with the primary skill as denoted by the scale label. Despite this caution, hypotheses may be generated from individual scales and overall patterns of scores if proper caution is used in recognizing the wide ranges of factors that affect the neuropsychological performance on a given scale. For example, a child may have such severe expressive language problems that any item that requires any verbal response, no matter how simple, may be missed. Similarly, severe receptive language problems may make it impossible to adequately communicate instructions to the child. Although the administrative procedures attempt to minimize the effects of these disorders, in some cases it is not. possible to eliminate these factors. Similarly, children with severe peripheral deficits or brain stem injuries may appear to have more severe cognitive injuries than are actually present. These problems are unfortunately common to all standardized tests. To an extent, the identification of these factors through the qualitative scoring is possible, although interpretation of these indices is not as well established as for the quantitative scores. ' Other factors that cause changes in scale elevations as well as overall patterns include a wide variety of neurological factors. One of the most important factors is the duration of the brain disorder, and whether or not it is still present. In general, the acute disorders, i.e., those that are continuing, will affect LNNB-CR scores much more significantly than will disorders from which the child has had 3 to 6 months or more to recover. The size of the lesion must be considered along with the question of chronicity. In individuals with disorders that resolve themselves without structural damage to the brain, LNNB-CR scores will return essentially to normal, reflecting the child's recovery of all major skills. The examiner should nevertheless be alert to specific items, even on normal scales, that the child may fail to adequately perform and that may
197
reflect residual damage. It is important to notice these factors, as they are often useful in explaining specific problems the child may be having that were not noticed before the damage occurred, as well as in designing specific rehabilitation training. Another important aspect is the location of the damage. Brain damage in each hemisphere is expressed very differently on the pattern of scores, as is brain damage in different locations within a single hemisphere. In general, the LNNB-CR is more sensitive to the disruptive effects of left hemisphere lesions. The cause of the brain dysfunction also represents a major problem with respect to the battery's results. Brain dysfunction may be caused by a wide range of problems: from overt structural damage to metabolic disorders and idiopathic problems that have no clear genesis. In general, those disorders that destroy brain tissue cause much more damage to the brain and therefore cause more highly significant deviations on the LNNB-CR battery. Disorders such as idiopathic epilepsy, which may not have a clear structural focus or clear cause, may cause relatively little damage. A final consideration is the premorbid level of the individual. Specifically, an individual with higher skills prior to a given brain injury will have higher skills afterward than would a person with overall lower initial cognitive skills; the person with higher cognitive skills can more easily reorganize brain function to adapt to the loss in specific areas. An individual who is extremely intelligent prior to an injury may show only motor and sensory signs with relatively few cognitive deficits in the abnormal range even after a significant injury, because the person's skills have simply been reduced from aboveaverage to average. In these cases, one must be sensitive again to the pattern of items missed by the child that may indicate brain dysfunction. One obvious problem with children is that we may not have an idea of what the "premorbid" level was. For the reasons discussed above, interpretation of the LNNB-CR typically focuses more on scale patterns and on intrascale variability than on interpretation of scale elevations per se. Scale patterns have the advantage of allowing the user to make deductions about the reason a given scale was impaired. For example, in profiles in which CS (Receptive Speech) is the highest score, one can hypothesize that deficits on other scales may be attributed to the loss in receptive skilJs. As with any other procedure, this leads only to hypotheses, but can offer valuable insights in attempting to understand the child's basic underlying deficits. The major patterns on the
198
CHAPTER 11
LNNB-CR are discussed below in the context of the highest scale among the basic clinical scales.
Developmental
Issues
in
Interpretation.
Another substantial problem in interpretation is the role of developmental issues. There are several major problems that must be recognized when the test is interpreted. All of these problems stem from the fact that children's brains are not fully developed until midway in their teenage years. Thus, there is a difference in what brain skills can be affected at given ages and in the long-term impact of such injuries when the injury occurs at different developmental stages. It is not the purpose of this chapter to review theories of brain development, but such issues must be considered in the interpretation of any child neuropsychology battery. In the context of these theoretical concerns, one can approach the interpretation of the LNNB-CR (or any other neuropsychological test for children). The focus of such interpretation should not be localization, but rather a detailed analysis of the major deficits that underlie the child's function. Such interpretations must take into account the factors discussed in this chapter and not solely rely on face valid interpretations of item content and one's supposition of what an item measures. One must be alert to how the child performs the items (whether the answer is right or wrong) and integrate this information with knowledge about developmental stages. Even more so than in adults, children's behavior after brain injury can be unpredictable and must be considered as its own case study. Even with these limitations, one can examine scale patterns to generate hypotheses. Each hypothesis that is generated must be tested against the actual patterns of items in the scales and the qualitative data on how the items were performed. The scale pattern serves to only generate hypotheses.
injuries to the right hemisphere or to the left parietal area. Drawings that are accurate but done slowly may simply reflect motor dysfunction of the dominant hand and the opposite cerebral hemisphere (or, sometimes, compulsiveness). Because of the nature of the items on the C I scale, it tends to be sensitive to many different types of brain dysfunction. Primary sensitivity is to sections of the posterior frontal lobe, but lesions of the temporal and parietal lobes, as well as the anterior frontal lobe, will also cause significant elevations in the score. However, extreme elevations (scores exceeding SOT) will usually only be caused by lesions in the motor system. Elevations on the C 1 scale are best interpreted in comparison to elevations on C3 (Tactile Functions). When Cl is elevated but C3 is not, this is suggestive of difficulties with motor tasks. This comparison can be very useful in initially localizing a deficit in the anterior-posterior dimension. Clients displaying pure parietal lobe dysfunction will rarely achieve a Cl score above 60T, although specific items involving kinesthetic feedback will be most frequently missed. On examination, the items on the battery will usually show a clear pattern in these posterior injuries that is highly effective in localizing a given disorder. When both of these scales (C1 and C3) are highly elevated, generalized impairment of motor and sensory areas is suggested, but this is often in the context of diffuse deficits. If only these four scales are affected, then peripheral disorders affecting motor and sensory skills need to be considered, as well as the possibility of subcortical diseases.
Cl (Rhythm). The C2 scale is much more simply organized than the C1 scale. Item 35 involves the analyses of groups of tones. The child must compare two groups of tones, saying whether one is higher or lower. Items 36 through 38 require the child toreproduce tones. Whereas the initial items involve the perception of tonal qualities, these latter items inClinical Scales volve the expression of tonal relationships. Items 39 through 40 involve the evaluation of Cl (Motor Functions). The Cl scale is one of the most complex scales on the LNNB-CR. A wide acoustic signals. The child must identify the number variety of motor skills reflect both right and left hemi- of beeps in groups of sounds. The last two items in sphere performance. The first three items involve the C2 scale deal with the perception and reproducsimple movements of the hands. These items are es- tion of rhythm. Item 41 measures the ability of the pecially sensitive to disorders in or near the posterior child to reproduce rhythmic patterns. This item refrontal lobe. In many cases, evidence of lateralized quires both the perception of rhythmic patterns and motor disorders may be detected by examining the the reproduction of sounds, usually using the dominant hand. The item can be missed by individuals raw scores on these items. Items 21 through 32 assess construction dys- with deficits in either hemisphere. Item 42 asks the praxia. Items that are performed very poorly often child to make a series of rhythms from verbal comreflect severe spatial disorganization characteristic of mands. The combination of verbal and rhythmic con-
THE NEBRASKA NEUROPSYCHOLOGICAL CHILDREN'S BATTERY
tent on this item also makes it sensitive to injuries in either hemisphere. Of all the basic clinical scales, C2 is the most sensitive to disorders of attention and concentration. When giving these items to such individuals, it is often useful for the examiner to stop the administration between the stimuli in each item and not go on to the next item until the individual's attention has been secured. Because there are usually only two choices in each item, items cannot be repeated. Consequently, it is important to ensure that the first administration is carried out as accurately as possible. When elevations of the C2 scale are the highest in the proftle, they are most often associated with right hemisphere injuries that are usually more anterior than post~rior. This is especially true when the highest scales are some combination of C2, C9 (Arithmetic), and CIO (Memory). However, this same pattern may be seen in left anterior lesions as well, although in those cases it is accompanied by at least subtle, if not gross, deficits in some form of verbal skills. When the C2 deficit is combined with C4 (Visual Functions) scale elevation, then the lesion may be either anterior or posterior, with a more posterior lesion being more likely with higher elevations onC4. C3 (Tactile Functions). Items 43 through 56 involve different levels of cutaneous sensation. Individuals must identify where they are touched, how hard they are touched, and so forth. Injuries to the anterior parietal area will cause significant elevations on this scale as will injuries to the middle parietal areas that Luria (1973) designated as the "secondary area'' of the parietal lobe. Individuals with damage in and around the angular gyrus may have particular problems with verbal/tactile items. The last two items on the C3 scale involve the stereognostic perception. Individuals with old injuries to the parietal lobe on either side may have difficulty meeting the time requirements. Proftles with highest scores on the C3 scale are interpreted in conjunction with the relative standings on the Cl scale. If C3 is greatly elevated over Cl, then this points to a posterior lesion. This generally remains true even when the C I scale equals the C3 scale, especially if the Cl deficits arise from construction difficulty and sequencing rather than motor paralysis. Deficits may be due to an inability to concentrate, which should also result in inconsistent behavior, or to an inability to integrate and identify all stimuli. In the latter case, the deficit will have a similar effect but will also show up as a rule in other
199
naming and identification tasks while causing less difficulty in purely spatial tasks. When the deficit is purely spatial, such as in proftles with C3 and C4 (Visual Functions) elevations, the injury is likely to be right parietal-occipital, although this pattern may also reflect subcortical involvement of one or both hemispheres. When naming is strongly involved, left parietal deficits should be considered. All such hypothesized localizations assume a normally dominant left hemisphere. Such patterns can be changed significantly by mixed or right hemisphere dominance for speech. C4 (Visual Functions). The C4 scale evaluates a range of visual functions. Items 59 and 60 ask the child to identify objects by viewing either an object itself or a picture of an object. The person need not identify the object by name but rather can describe function or indicate recognition in other ways. Despite this, naming must be considered a component of these items. If the child is not able to do these items, later items on the battery that are more sensitive to right hemisphere function may be missed simply because of left hemisphere involvement. Thus, interpretation of the scale must depend on the child's performance on the simple, initial items. Later items require a great deal more visualspatial perception than do these first two items, although naming is still required. Item 61 presents pictures that are difficult to perceive. Item 62 presents objects that overlap one another and that the child with poor visual-spatial skills has difficulty identifying. Item 63 examines the ability to determine that two figures are mirror-image versions of one another. Item 64 is a modification of items from the Raven Progressive Matrices (Raven, 1960). It is also a strong measure of visual-spatial organization and right hemisphere function. Item 65 involves spatial rotation without any speech components. Individuals may point to the correct answer or circle it as necessary (or say it if this is not possible). Poor performance on this task is suggestive of impairment of visual-spatial skills. Proftles in which the C4 scale is .highest, in combination with any secondary scale, generally reflect impairment in the right hemisphere or the occipital areas of the left hemisphere. The C4 scale can be elevated in other left hemisphere injuries, but rarely will it be the highest scale overall. In right hemisphere injuries, deficits on only more complex visual tasks suggest either a mild parietal involvement or injury to anterior areas. These lesions are usually accompanied by elevation on the Cl scale suggestive of right hemisphere lesions. Subcortical
200
CHAPTER 11
lesions that interfere with visual processing can also cause patterns suggestive of right hemisphere injury, as can severe peripheral visual problems.
CS (Receptive Speech). C5 items evaluate the ability of the child to understand receptive speech, from simple phonemic analysis to the understanding of complex sentences with inverted English grammar. Items 66 through 71 concern the understanding of simple phonemes. For items 66 through 70, the individual hears simple phonemes and must then repeat or write them. It is important to note if individuals are able to either say or write phonemes but not to do both. Item 71 tests the ability to understand phonemes spoken at different levels of pitch. It is not unusual for individuals with significant damage in the right temporal area to miss this item. Items 72 through 77 involve the understanding of simple words and sentences. The child must do relatively simple tasks of naming, pointing, and identifying, and must define simple words. The intent of these items is simply to ensure that the child is hearing correctly and interpreting correctly what is said to him or her. Beginning with item 78 and continuing through the end of this scale to item 83, the individual is given increasingly more difficult instructions. These items assess the child's ability to understand and to perform or answer as requested. All these items can be affected by damage to the left hemisphere, but several items can also be affected by right hemisphere dysfunction. For example, item 79 requires some spatial orientation on the part of the child. If the child appears to understand the sentence but disrupts the spatial requests made, the possibility of right hemisphere dysfunction must be suspected. When this scale is highest, as well as significantly elevated above the critical level by at least 15 points, deficits are usually associated with left hemisphere injury. Lesser elevations, caused by difficulty with the more complex items, can appear as the highest scale in right anterior injuries. This can be especially true in mild elevation combinations of C5 and C 10 (Memory), C5 and C2 (Rhythm), C5 and C4 (Visual Functions), or C5, C11 (Intellectual Processes), and C9 (Arithmetic). In the most significant elevations, however, left hemisphere involvement is generally indicated. An important caveat in evaluating speech in children without a history of normal language achievement is differentiating between problems due to environment and nonneurological physical factors and those due to brain-based deficits. One common problem is the child with hearing difficulties due to
multiple infections requiring tubes or infections that caused partial or complete deafness. In these children, language abnormalities may simply be due to an inadequate chance to develop the relevant skills. Similar problems arise from backgrounds with inadequate verbal stimulation. It is very difficult to differentiate between deficits due to a poor premorbid history and those due solely to brain damage. C6 (Expressive Speech). The C6 scale evaluates the individual's ability to repeat simple phonemes and words and to generate automatic as well as more complex speech forms. Initial items simply require the repetition of sounds or words spoken by the examiner. Beginning with item 89, the child must repeat the same list of words and sounds by reading them rather than hearing them. If an individual is able to pass either one of these sections, significant expressive speech deficits are not present. For example, the individual who is able to read but not to receive auditory information will be able to do the second section. Inversely, the child who is unable to read will be able to do only the first section. Therefore, one must carefully examine the pattern of answers to see if the errors are confined to one modality or the other. Beginning with item 93, the child must repeat increasingly more difficult sentences. Item 94 examines the ability to name from a description rather than from a visual presentation of the object. Items 95 through 98 ask individuals to count and say the days of the week, first forward then backward, all a form of automatic speech. Items 99 through 104 evaluate the ability to produce speech spontaneously under three conditions: after looking at a picture, after hearing a story, and after being given a topic to discuss. If other items on this scale are performed without difficulty, and yet the child experiences problems with these items, there is the possibility of low intelligence. The final section involves complex systems of grammatical expression; the child must fill in words that are missing in a sentence or make up a sentence from words that are given to the child. In general, C6 scores are sensitive to injuries in the left hemisphere only. It is rare to see a high C6 score in individuals with unilateral right hemisphere injuries. Exceptions to this are individuals who had difficulty reading prior to their injury, or whose disorders have somehow interfered with auditory perception or have had generalized effects (e.g., pressure effects from a tumor). However, examination of the patterns of the items on the battery can easily eliminate these possibilities. In the absence of these types of conditions, elevation on the C6 scale, es-
THE NEBRASKA NEUROPSYCHOLOGICAL CHILDREN'S BATTERY
pecially above a score of70T, is almost always indicative of a left hemisphere injury. Very mild C6 elevations may be associated with right hemisphere lesions, with the major errors occurring in the last items of the scale (spontaneous speech, sequencing, and fill-in items). One assumption underlying interpretation of all the language scales is that the child was originally fluent in English, as all of the current research is based on nativeborn speakers. C7 (Writing). The C7 scale evaluates the ability to analyze words phonetically in English and then to do copying of increasing difficulty. Initially, children are asked to copy simple letters, then combinations of letters and words, and then write their first and last names. They are then asked to write sounds, words, and phrases from dictation. In general, disorders of writing localize to the temporal-parietal-occ ipital area, especially in and around the angular gyrus of the left hemisphere. However, specific disorders may indicate problems in other areas. For example, the ability to write from written material but not from auditory material suggests a specific lesion in the temporal lobe. Conversely, the ability to write from dictation but not from written material suggests a lesion in the occipital or occipital-parietal areas of the cerebral cortex. If the child is, in general, able to write but has difficulty forming letters and changing from one letter to another, there could be a problem in kinesthetic feedback in which the child mixes up letters that are formed by similar motor movements. If the child is simply unable to write at all due to paralysis, this, of course, is suggestive of a lesion in the motor strip area of the posterior frontal lobe. Finally, if the child writes at an angle to the page, suggesting some spatial problems, and has no other writing disorders, this can be related to right hemisphere dysfunction. Lack of the abiJity to read or write one's name is often indicative of a general dementia or, in some cases, a disorder of automatic writing that may occur with injuries in both hemispheres. Motor writing errors are generally associated with the hemisphere opposite the child's normal writing hand, although care must be taken in injuries that cause the child to change writing hands. In these cases, writing may remain poor but reflects an injury in the ipsilateral hemisphere. Motor writing problems may arise simply as a result of motor problems reflecting the functions of the motor areas of the brain, but may also arise in injuries involving kinesthetic and tactile feedback. Motor writing deficits in
201
which the writing itself is motorically intact but spatially disrupted (at large angles to the horizontal, or where words are written over one another) may reflect injuries to the right (or spatially dominant) hemisphere.
CB (Reading). The C8 scale closely parallels the C7 scale. The child is asked to generate sounds from letters that the examiner reads aloud. This. generally measures the ability of the child to show integration of letters and auditory analysis functions of the temporal and parietal areas of the left hemisphere. The child is then asked to name simple letters, read simple sounds, and read simple words and letter combinations that have meaning. Finally, the child must read entire sentences as well as paragraphs. If the child is able to read simple words but not entire sentences or paragraphs, possible injuries include disorders of visual scanning that make it impossible for the child to grasp more than one word at a time. Generally, deficits on the C8 scale, in a child who could read prior to an injury, are almost always associated with a left hemisphere injury, usually posterior. The exceptions to this are deficits that occur because of spatial disruption (inability to follow a line, which shows most clearly in the paragraph reading) or neglect of the left side (which should be corrected by the examiner if the test is administered correctly). Both suggest right hemisphere dysfunction. However, we are not justified in making such conclusions in an individual who never was able to read unless there is other evidence to confirm a given injury. C9 (Arithmetic). The C9 scale is the most sensitive of all the LNNB-CR scales to educational deficits. Even in normally educated individuals, this is the scale most likely to appear in a severely pathological range when there is, in fact, no damage. This scale starts with the child simply writing down numbers from dictation in both Arabic and Roman numerals. Several items have been employed to identify the spatial dysfunctions that are possible. The child is asked to write 17 and 71 , 69 and 96. Thereafter, the scale requires the person write down numbers of increasing complexity. As numbers become more complex, it is possible to see if the child places the numbers in the correct sequence, again looking for possible spatial deficits that can be caused by right hemisphere or left occipital-parietal dysfunction. In the next section, the child is asked to compare numbers, an operation that is basic to the left occipital-parietal area. In items 124 and 125, the child is asked to do simple arithmetic problems. These are problems that most individuals can proba-
202
CHAPTER 11
bly do from memory. The last item is the presentation of serial 3's. The C9 scale appears to be potentially sensitive to lesions in all parts of the brain, as well as to preexisting deficits common to about 20% or more of the normal population whose performance is well below grade level expectations.
subtest of the Wechsler Adult Intelligence Scale (WAIS). Performance in this area is further evaluated by items 144 through 146, in which the child must find the logical relationships between specific objects and the groups to which they belong. The last items on the scale, items 147 through 149, involve simple arithmetic problems very similar to those seen on the C10 (Memory). The CIO scale is basically in- WISC-R Arithmetic subtest. Overall, the C11 scale is highly sensitive to disvolved with short-term memory functions. The first orders in both hemispheres but is most sensitive to items on C I 0 look at the ability of the child to memodisorders in the left hemisphere. Injuries in the parrize a list of seven simple words and to predict his or ietal lobes will cause maximum dysfunction. The her performance. Items 129 through 131 involve imdetermination of laterality, however, must be made mediate sensory trace recall. The items test word by investigating specific items to judge whether those memory and visual memory. initial items that are right hemisphere oriented sugItems 132 and 133 involve simple verbal memogest adequate visual interpretation skills. If these ry under two conditions of interference. Several diffiskills appear to be intact, then the scale is likely to be culties in short-term memory are seen in these items. reflecting a left hemisphere dysfunction alone. If Finally, item 135 is a measure of the individual's these are the only items missed on the scale, the ability to associate the verbal stimulus with a picture. possibility of isolated right frontal dysfunction must This item can be interfered with by either left or right be seriously considered. hemisphere dysfunction and is sensitive to high-level The Cll scale score correlates about 0. 7 with disturbances in memory skills. WISC-R Full Scale IQ. However, this scale, because Overall, the C 10 scale is most sensitive to verbal of its greater flexibility in administration, may yield dysfunction because of its importance in a majority of higher IQ estimates in individuals with impairment in the items. However, nonverbal dysfunction caused expressive or receptive speech skills. Although the by right hemisphere lesions will show up in a moderscale can reflect impairment in either hemisphere, a ately elevated CIO score of about 60T, with a pattern high elevation combined with C2 (Rhythm), C4 (Viof missing the nonverbal items. It is always important sual Functions), CIO (Memory), and C9 (Arithmetic) to look at the pattern of the items missed before vengenerally points to right hemisphere dysfunction, turing the hypothesis of a possible etiology. whereas elevations combined with C6 (Expressive C11 (Intellectual Processes). The items in Cll Speech), C8 (Reading), and C7 (Writing) indicate should be differentiated from items in a standard in- left hemisphere damage. telligence test. All the items on this scale have been selected because they are able to discriminate between brain-damaged and normal subjects. Thus, Qualitative Analysis rather than giving a level of intelligence that can be associated with a child's learning history, the items The LNNB-CR lends itself to a qualitative analtend to give a functional intellectual level. ysis along with the quantitative analysis discussed in Initial items in the scale involve the understand- this chapter. The consideration of the qualitative facing of thematic pictures. The first item asks the child tors becomes the next step in the diagnostic process. to interpret a picture in his or her own words; items Here, the interest is not in whether a child got a 137 and 138 ask the individual to put pictures into a certain score on a certain item but rather how that series that makes sense, similar to the items in Picture score was achieved. One of the great advantages of Arrangement. Item 139 asks the child to tell what is this battery (inherent in its design as well as in the comical or absurd about certain pictures. Deficits of administration and testing-the-limits techniques sugvisual scanning can also be seen in individuals who gested) is that the same test procedures lend themare not able to appreciate the complexity of a picture selves to both quantitative and qualitative analyses. and who, thus, tend to focus on one area. Item 140 Although it is possible to interpret the battery from requires the child interpret a story. Items 141 and 142 only one method or the other, the use of only one ask for similarities. Item 141 involves simple concept technique does not take full advantage of the posformations and definitions; items 142 and 143 call for sibilities within the battery nor does it yield the maxcomparisons and differences between objects in imum amount of useful information in any given much the same way as do items on the Similarities case. The two approaches complement one another
THE NEBRASKA NEUROPSYCHOLOGICAL CHILDREN'S BATTERY
and allow the user of the LNNB-CR to enjoy the best of both methods, avoiding that continuing, yet ultimately futile, argument over which approach is better or which approach should be employed. In scoring qualitative errors, there is a wide range of possibilities aimed at gaining a better understanding of the .. why" behind a child's error. Qualitative analysis can also aid in the evaluations of responses that are correct in terms of the quantitative scoring but still unusual, such as the child who reads a word on the C8 scale but stutters in pronouncing it, or the child who can describe an object and its uses on the C4 scale but is unable to give its name. The disadvantage of qualitative inferences is the lack of formal scoring criteria and reliability across examiners. At present, there is no way in which such problems can be completely eliminated, but there are ways in which such problems can be minimized. The Qualitative Scoring Summary allows the user to score over 60 categories of qualitative observations that can be made during administration of the battery. Although this could be cumbersome in a child who is severely impaired, it is quite useful in the child who shows mild to moderate damage, or who is apparently showing no damage at all, in identifying a specific deficit that may have neuropsychological implications. These categories are not exhaustive by any means, but do represent the areas we have found to be the most useful and reliable. The presence of such a scoring system does not, however, eliminate the need for intelligent observation on the part of the user, nor the need for neuropsychological knowledge to interpret this information. It should be clear from the previous description that there are a number of major complexities with a qualitative system and these have generally inhibited the development of such systems in the past or caused them to be quite unreliable across examiners. It is expected that the details for each qualitative item scoring will allow this system to be somewhat more rigid but also more reliable. Interpretation of qualitative information on more than a behavioral level is quite difficult. At present, all information on such interpretations is largely unsystematized in the sense that different clinicians or researchers use different definitions and different theoretical frameworks. At present, learning to properly observe and interpret the qualitative aspects of behavior is done through clinical experience with children and the reading of such experiences reported in specific cases by clinicians such as Luria. An academic understanding of the qualitative aspects of behavior through reading and classes yields a basic framework from
203
which to approach later supervised clinical experience. A standard examination such as the LNNBCR, which allows the observation of the same basic behaviors across many diagnostic groups, is also an aid in making these clinical observations. It is very important that the clinician learn to observe and record the child's approach to the items, especially if that approach differs from those seen in the normal child. (The examiner must have adequate experience with normals to make this comparison.) This should be done even if the examiner does not understand the meaning of the behavior or its significance. Sometimes the significance becomes clear after the quantitative analysis is completed, or it may become clear upon consultation with one's supervisor or a consultant. By doing this on a systematic basis, the user will begin to appreciate the meaning of the child's behavior and to develop the ability to perform such analyses independently. After a qualitative analysis has been made, it should be integrated with the quantitative analysis. It is our strong belief that neither form of data is inherently .. superior" in any given case. In some cases, the qualitative data help to explain inconsistencies that cannot be resolved in the quantitative results. In other cases, the quantitative data suggest an alternate approach to an observation that clears up the interpretation of a qualitative aspect of behavior. Only when the two sets of data have been integrated has a fully effective initial evaluation been completed.
Prior History Even at this point, there still remains another step in the diagnostic process-the reconciliation of the conclusions of the above techniques with the history. This can be done in two ways: (I) knowing the history when the case begins and considering it throughout the diagnostic process, or (2) analyzing the case with a minimum of information and checking the detailed history afterward. (Doing any case completely blindly is not recommended.) Both techniques have their drawbacks. If too much history is known, one may be so biased that the inconsistencies between the history and data are overlooked or deemed unimportant. Historical information and the conclusions made available by others prior to the neuropsychological assessment may be right or wrong. A lesion may exist as reported, or may not. The child's developmental history may be accurate or may contain serious errors. The relative accuracy of information does depend on the source of that information as well
204
CHAFfER 11
as its nature. In all cases, however, it is important to pencil version and a version primarily given by comdouble-check all such information. puter under the supervision of a technician or psyConsequently, our bias suggests working from a chologist. The computer version has the advantage of basic history for important aspects that concern the ensuring that the test materials are given properly and validity of testing. In the evaluation process, we treat timed properly, freeing the examiner to do more obconclusions from the data as simply hypotheses to be servation and analysis of the patient's performance. confirmed or discarded. This leaves the clinician It is expected that the new version will be released for more flexibility to take his or her data seriously and to general use in about 3 years. learn from those data all that is possible. If, at the end, discrepancies are found between history and neuropsychological findings, the clinician should investigate the history and the findings for errors that Conclusions may cause this discrepancy, and look for conditions outside neuropsychology that may have affected one The Luria-Nebraska Children's Battery is a test or the other source of data. that evaluates a wide range of skills aimed at assessing the neuropsychological competence of children between the ages of 8 and 12. The battery offers a New Research Approaches variety of quantitative and qualitative scores by At present, we are in the development of a new which to detail the performance of children and to battery that represents a substantial expansion of the integrate that performance with historical data. The old one. The new battery has been expanded to more battery has been shown to be highly successful in heavily cover areas not well involved in Luria's origi- diagnosis, but does need to be supplemented by other nal examination: more complex and detailed analysis tests when detailed analysis of single areas is necesof visual-spatial functions, more detailed analysis of sary or preferable. The battery lends itself to invarious aspects of aphasia, reading comprehension, terpretation on a variety of levels. Thus, it can be motor writing, complex memory functions, and useful to people with varying backgrounds providing higher intellectual skills. The new test combines caveats on use are followed carefully. In the long run, the battery represents an initial items from both the children's and adult Luna-Netest that will be expanded and improved with each braska battery along with new items. A factor analyversion. This is based on the philosophy of the author sis of the items as a whole yielded 37 scales, which form the basis of the new test. Internally, these scales that there is never a perfect test battery, and that are more like the factor scales, representing purer and change of current tests on the basis of research is both necessary and desirable. more specific abilities. The test differs in that instead of giving every item, items are given on the basis of the individual's abilities (as with the Stanford-Binet, for example). Within each area patients may get only very difficult, References very easy, or moderately hard items depending on their performance. This allows the test to contain Golden, C. J. (1986). Manual for the Luria-Nebraslw Neuropsychological Battery: Childrens Revision. Los Angeles: WPS. items suitable from ages 3 through adulthood, and Plaisted, J. R., Gustavson, J. L., Wilkening, G. N., &Golden, C. applicable to wide ranges of performance in brainJ. (1983). The Luria-Nebraska Neuropsychological Batinjured individuals. The giving of only a part of the tery-Childrens Revision: Theory and current research finditem pool allows for a reasonable testing time (2 to 3 ings. Journal of Clinical Child Psychology, 12, 13-21. hours). Raven, J. C. (1960). Guide to the Standard Progressive Matrices. London: Lewis. The test will come in two versions, a paper and
12 Applications of the Kaufman Assessment Battery for Children (K-ABC) in Neuropsychological Assessment CECIL R. REYNOLDS, RANDY W. KAMPHAUS, AND
BECKY L. ROSENTHAL
The Kaufman Assessment Battery for Children (KABC) (Kaufman & Kaufman, 1983a) is a recently published, individually administered clinical test of intelligence and achievement designed specifically for use with children from 2i to 121 years of age. The K-ABC was developed from a theoretical framework that to a large degree reflects a coalescence of the work of Luria and Vygotsky and American researchers with interests in cerebral specialization as interpreted and integrated by Alan and Nadeen Kaufman. The K-ABC is thus of interest to clinical neuropsychologists. This chapter will review the theoretical framework of the K-ABC, provide an overview of the test, its methods of development and standardization, its technical properties, and describe its conceptual and empirical relationship to neuropsychology. Additionally, techniques for rehabilitation of academic disturbances will be introduced as viewed from the K-ABC model. The previous work of the K-ABC authors, Alan S. and Nadeen L. Kaufman, is well known in the area of intellectual assessment. Alan Kaufman had a major impact on the assessment of children's intelligence long before the K-ABC. As a Research Associate at The Psychological Corporation, Kaufman
CECIL R. REYNOLDS • Department of Educational Psychology, Texas A&M University, College Station, Texas 77843. RANDY W. KAMPHAUS AND BECKY L. ROSENTHAL • Department of Educational Psychology, University of Georgia, Athens, Georgia 30602.
was project director for development of the WISC-R and the McCarthy Scales, tasks to which he was well suited because of his studies at Columbia under the tutelage of the psychometrician Robert L. Thorndike. Subsequently, Kaufman's (1975) article on the factor analysis of the WISC-R standardization sample (still one of the most cited articles in WISC-R research) provided an important reminder to clinicians that, as was the case with the old WISC, the updated WISC-R produced three factors. Hence, it is not always possible to interpret the verbal and performance IQs of the WISC-R as unitary dimensions. A possibly more legendary article by Kaufman (1976) cautioned psychologists against overinterpreting verbal and performance IQ discrepancies that were quite common in the general population. Alan and Nadeen Kaufman's first book, Clinical Evaluation of Young Children with the McCarthy Scales ( 1977), provided a logical framework and summary of major research findings for what was at the time a new, and somewhat unique, measure of intelligence. PriortotheK-ABC, however, Kaufman was probably best known for his text Intelligent Testing with the WISC-R (l979a). Anastasi (1982) says of Kaufman's book that the most important feature of [Kaufman's] approach is that it calls for individualized interpretations of test performance, in contrast to the uniform application of any one type of pattern analysis . . . . The basic approach described by Kaufman undoubtedly represents a major contribution to the clinical use of intelligence tests. (p. 466)
205
206
CHAPTER 12
Throughout his writings, however, one detects a sense of dissatisfaction with many aspects of existing intelligence tests. In an article published in the Journal of Research and Development in Education (1979b), Kaufman set the stage for the development of the K-ABC. In that article, he maintained that individual intelligence testing had been remarkably resistant to change, despite advances related to the understanding of intelligence in fields such as psychology and neurology. Kaufman argued that substantive theoretical advances in intelligence research had gone unheeded in the conservative test-publishing industry and that the field lacked any true innovations since the work of Binet around the turn of the century. In developing his intelligent testing philosophy (an outgrowth of Wesman's (1968) intelligent testing approach), Kaufman gave great emphasis to the need for theory-driven assessment and interpretation of children's intelligence and adherence topsychometric theory (Reynolds, 1987). Kaufman ( 1979b) then set the stage for the emphasis of theory in the development of the K-ABC. The Kaufmans' emphasis on assessing intelligence from a strong theoretical base is one characteristic of the K-ABC that, perhaps more than anything else, distinguishes it from its predecessors; it is the derivation of that theo-. ry that creates much of interest to the neuropsychologist.
tual schemes of the functional organization of the brain are probably the most comprehensive currently available" (Adams, 1985, p. 878). Much of Luria's work grew from earlier writings of Sechenov (translation, 1965) and Vygotsky (translation, 1978). Majovski (1984), who studied with Luria in Russia for several years near the end of Luria's life, evaluated the K-ABC regarding its relevance to Luria's approach with children and to child neuropsychology in general, and gave the scale high marks. Luria defmed mental processes in term of two sharply delineated groups, following Sechenov's suggestions. The first is the integration of elements into simultaneous groups. He further qualified Sechenov's original meaning, indicating that simultaneous processing meant the synthesis of successive elements (arriving one after the other) into simultaneous spatial schemes whereas successive process- . ing meant the synthesis of separate elements into successive series. The latter qualifications are crucial in seeing the match of the various subtests of the KABC mental processing scales to this processing dichotomy. In the K-ABC the Kaufmans define these processes in a manner similar to Luria and provide a standardized assessment of these functions. Simultaneous processing here refers to the mental ability of the child to integrate input simultaneously to solve a problem correctly. Simultaneous processing frequently involves spatial, analogic, or organizational abilities (Kaufman & Kaufman, 1983b) as well as Theoretical Framework problems solved through the application of visual imagery. The Triangles subtest on the K-ABC (an The K-ABC intelligence scales are based on a analogue of Wechsler's Block Design task) is a promodel of sequential and simultaneous information totypical measure of simultaneous processing. To processing. The theoretical underpinnings of the prosolve these items correctly, one must mentally intecessing model were gleaned from a convergence of grate the components of the design to "see" the research and theory in a variety of areas including whole. Such a task seems to match up nicely with both neuropsychology and cognitive psychology. Luria's qualifying statement of synthesis of separate The neuropsychological theory employed by the elements (each triangle) into spatial schemes (the Kaufmans was distilled from two lines: the informalarger pattern of triangles, which may form squares, tion processing approach of Luria (e.g., Luria, 1966) rectangles, or larger triangles). Whether the tasks are and the cerebral specialization work done by Sperry spatial or analogic in nature, the unifying charac(1968, 1974), Bogen (1969), Kinsboume (1975), teristic of simultaneous processing is the mental synand Wada, Clarke, and Hamm (1975). thesis of the stimuli to solve the problem, independent of the sensory modality of the input or the output. Lurian Theories Sequential processing, on the other hand, emLuria's theory was of paramount importance in phasizes the arrangement of stimuli in sequential or the formulation of the Kaufmans' approach to assess- serial order for successful problem solving. In every ing intelligence with the K-ABC. This seems only instance, each stimulus is linearly or temporally reappropriate given the stature of Luria's theory. "Al- lated to the previous one (Kaufman & Kaufman, exander R. Luria's theory of higher cortical function- 1983b) creating a form of serial interdependence ing has received international acclaim. His concep- within the stimulus. The K-ABC includes sequential
K-ABC
processing subtests that tap a variety of modalities. The Hand Movements subtest involves visual input and a motor response; the Number Recall subtest involves auditory input with a response involving the auditory output channel only; Word Order involves the visual channel for input and an auditory response. Therefore, the mode of presentation or mode of response is not what determines the scale placement of a task, but rather it is the mental processing demands of the task that are important (Kaufman & Kaufman, 1983b). By providing systematic variation of modality of input and modality of response, the KABC provides a clinical vehicle for locating intact complex functional systems as well as specifying where any potential breakdown may have occurred in a faulty functional system. Qualitative evaluation of a child's performance on the K-ABC can be most useful in such instances and can lead to beneficial rehabilitation plans. No one with an intact brain uses only a single type of information processing to solve problems. These two methods of information processing are constantly interacting (even in the so-called splitbrain following commissurotomy, the two hemispheres of the brain often "whisper" to each other even if they cannot talk), though one approach will usually take a lead role in processing. Which method of processing takes the lead role can change according to the demands of the problem or, as is the case with some individuals, persist across problem type (i.e., forming what Das, Kirby, and Jarman (1979) refer to as habitual modes of processing). In fact, any problem can be solved through either method of processing. In most cases, one method is clearly superior to another. It is the latter case that makes the K-ABC a valuable tool-the two mental processing scales are primarily, not exclusively, measures of sequential or simultaneous processing. Pure scales, i.e., scales measuring only one process, do not exist. Careful observation of a child's performance, which should be the order of the day during any evaluation, will be particularly important to any neuropsychological assessment or neuropsychological interpretation of K-ABC test results; observation in many cases will be a primary source of information regarding which mental processes a child invoked on any given task, regardless of its scale. An equally important component of the K-ABC is the Achievement Scale. This scale measures abilities that serve to complement the intelligence scales. Performance on the achievement scales is viewed as an estimate of children's success in the application of their mental processing skills to the acquisition of knowledge from the environment (Kaufman, Kauf-
207
man, & Kamphaus, 1985). This scale contains measures of what have been identified traditionally as verbal intelligence, general information, and acquired school skills. Keeping in mind that it is not possible to separate entirely what you know (achievement) from how well you think (intelligence), the Kaufmans attempted to differentiate the two variables better than traditional measures of intelligence had. From a clinical neuropsychological standpoint, the K-ABC allows one to assess information processing skills without as much contamination from prior learning. Measurement of children's academic skills, however, is a traditional component of comprehensive neuropsychological assessment. The inclusion of the Achievement Scale in the K-ABC affords the opportunity to observe the application of processing skills to complex learning tasks, assess functional academic levels, and estimate long-term memory ability. For a thorough review of interpretation of the K-ABC Achievement Scale, see Kamphaus and Reynolds (1987). Majovski (1984) noted the high degree of "fit" between Luria's theory and the K-ABC and recommended that the test be used as an integral part of the neuropsychological battery for children. The K-ABC is not, under any circumstance, a complete or substitute battery for a comprehensive neuropsychological assessment; it is a good complement to nearly any choice of neuropsychological instruments. Majovski found the K-ABC particularly useful in contrasting problem-solving skills with acquisition of facts and in evaluating how a child solves a particular problem.
Cerebral Specialization Theories Support for use of the K-ABC in the context of neuropsychological assessment also comes from a variety of sources in cerebral specialization research. In a comprehensive review of research concerning the lateralization of human brain functions, Dean (1984) concluded that the K-ABC was well suited to clinical use and in research with children. Research evidence to be reviewed in a subsequent section provides additional but still tentative support for Dean's conclusion. It has been proposed that sequential and simultaneous processing are lateralized to the left and right hemispheres, respectively (e.g., Reynolds, ·1981b; Sperry, 1974; Sperry, Gazzaniga, & Bogen, 1969). Many other dichotomies have been suggested. The most prominent of these are displayed in Table I. Some find research on cerebral specialization difficult to coalesce. Indeed, the many seeming contra-
208
CHAPTER12
TABLE 1. Definitions of the Two Types of Mental Processing That Underlie the K-ABC Intelligence Scales from the K-ABC Manual and from Several Theoretical Perspectives" Source
K-ABC: Kaufman & Kaufman
Labels for process Sequential
(1983a)
Simultaneous
Celebral Specialization: Nebes (1974) (summarizing model of Bogen, Levy-Agresti, and Speny)
Analytic/propositional/ left hemisphere Synthetic/appositional/ right hemisphere
Luria/Das: Das, Kirby, & Jarman (1979)
Successive
Simultaneous
Cognitive Psychology: Neisser
Sequential/serial
(1967)
Parallel/multiple
Defmitions Places a premium on the serial or temporal order of stimuli when solving problems. Demands a gestalt-like, frequently spatial, integration of stimuli to solve problems with maximum efficiency. Sequentially analyzes input, abstracting out the relevant details to which it associates verbal symbols in order to more efficiently manipulate and store the data.
Organize and treat data in terms of complex wholes, being in effect a synthesizer with a predisposition for viewing the total rather than the parts. Processing of information in a serial order. The important distinction between this type of information processing and simultaneous processing is that in successive processing the system is not totally surveyable at any point in time. Rather, a system of cues consecutively activates the components. The synthesis of separate elements into groups, these groups often taking on spatial overtones. The essential nature of this sort of processing is that any portion of the result is .at once surveyable without dependence upon its position in the whole. Viewed as a constructive process, it constructs only one thing at a time. The very definition of 'rational' and 'logical' also suggest that each idea, image, or action is sensibly related to the preceding one, making an appearance only as it becomes necessary for the aim in view. A spatially serial activity is one which analyzes only a part of the input field at any given moment ... On the other band, sequential refers to the manner in which a process is organized; it is appropriate when the analysis consists of successive, intem:lated steps. Carries out many activities simultaneously, or at least independently.
•From Kaufman (1984). Reprinted with permission.
dictions in the results of cerebral specialization stud- Goldstein, Jaynes, & Krauthamer, 1977; Kinaies have prompted at least one leading researcher to bourne, 1978; Schwartz, Davidson, & Maer, 1975; remark thus: .. (To] say that the field of hemispheric Segalowitz & Gruber, 1977). One will fmd in the specialization is in a state of disarray and that the literature a large number of studies of hemispheric results are difficult to interpret is an understatement. specialization attempting to provide anatomical loThe field can best be characterized as chaotic'' calization of performance on specific, yet higher(Tomlinson-Keasey & Clarkson-Smith, 1980, p. 1). order, complex tasks. Much of the confusion in the On the other hand, reviews by Dean (1984) and Rey- literature stems from the apparently conflicting data nolds (1981b) have noted some consistencies, as dis- of many of these studies. However, the dynamic played in. Table 1. functional localization principle of Luria, and knowlFor the vast majority of individuals, the left ce- edge that any specific task potentially can be perrebral hemisphere appears to be specialized for lin- formed through any of the brain's processing modes, guistic, propositional, serial, and analytic tasks and should give some insight into the conflicting results the right hemisphere for more nonverbal, apposi- that appear in the literature. In this regard, it is most tional, synthetic, and holistic tasks (Bever, 1975; important to remember that cerebral hemispheric Bogen, 1969; Gazzaniga, 1970; Hamad, Doty, asymmetries of function are process-specific and not
K-ABC
stimulus-specific. Shure and Halstead (1959) noted early in this line of research that manipulation of stimuli was at the root of hemispheric differences, a notion that is well supported by current empirical research (e.g., Ornstein, Johnstone, Herron, & Swencionis, 1980) and thought (e.g., Reynolds, 198Ib). The confusion of the content and sensory modality through which stimuli are presented with the process by which they are manipulated, particularly in the secondary and tertiary regions of each lobe of the neocortex, seems to be at the root of the chaos. How information is manipulated while in the brain is not dependent on its modality of presentation and not necessarily on its content, though the latter may certainly be influential. The variation in content and method of presentation of the tasks that make up the three scales of the K-ABC allow one to tease out any modality or content effects that might nevertheless occur for a specific child, though clearly the emphasis of the K-ABC is on process, not content. A review of Table 1 reveals that a process-oriented explanation provides an organizing principle superior to a focus on content. The "content-driven" attempts at explaining hemispheric differences fail to recognize the possibilities for processing any given set of stimuli or particular content in a variety of processing modes. Bever (1975) emphasized this point and elaborated on two modes of information processing that are of interest here due to their similarity to simultaneous and successive cognitive processes. According to Bever (1975), cerebral asymmetries of function result from two fundamental lateralized processes: holistic and analytic processing. Lateralization of these two methods of information processing is incompatible and cannot coexist in the same physical space. Analytic processing appears analogous to successive processing and is lateralized, in most individuals, to the left hemisphere. Holistic processing is analogous to simultaneous processing and is typically lateralized to the right hemisphere. The K-ABC also taps most of the functions identified by Dean ( 1984) in his review of the cerebral specialization literature, with the exceptions of depth, haptic, and melodic perceptions. These skills are assessed by other traditional neuropsychological batteries, although such tasks are virtually nonexistent for the very young child. Careful observation may still provide insight into neuropsychological processing deficits especially if one pays particular attention to the manner in which errors are made. Qualitative and quantitative data are complementary, not interchangeable; the intelligent testing philosophy of Kaufman (1979b) is just as crucial to neuro-
209
psychological assessment as to any other area of clinical evaluation. Das et al. ( 1979) do not agree that simultaneous and successive processing are represented in the right and left hemispheres, respectively, but rather believe that each mode of processing is prominently represented in both hemispheres. According to Bever's (1975) line of reasoning, this is an impossible state of affairs in the normally functioning human brain. Additionally, the hemispheric-lateralization literature is consistent with the notion of a successive-processing left hemisphere and simultaneous-processing right hemisphere relationship. Das et al. (1979) have developed their theory exclusively on the basis of group data, yet they attempt to discredit hemispheric lateralization of cognitive processing by calling upon anecdotal individual case data. Hardly anyone would contend that hemispheric specialization for cognitive processing is the same in every individual. However, this seems to be the requisite state of affairs for Das et al. to accept the hypothesis of lateralization of simultaneous and successive processing. This hardly seems necessary. The sheer weight of evidence at present (cited later in this chapter) indicates that, for the vast majority of individuals, the lateralization of simultaneous and successive processes is to the right and left hemispheres, respectively. Children may also form habitual modes of information processing that detract from their efficiency in learning new material or in solving novel problems (Reynolds, 1981b). This would be one explanation of large sequential-simultaneo us score differences that should also be explored, perhaps through protocol analysis or some related technique during a testing of the limits type of procedure. Children who attempt to solve sequential tasks using simultaneous processing approaches or vice versa are likely to have academic as well as behavioral problems. n any event, a problem is evident when this occurs, one that needs more detailed study for the individual child in order to explain the reason for this persistent approach. Normally functioning individuals appear to be able to use the two modes of information processing separately or in conjunction with one another or possibly shift at will depending on the type of information to be processed (Gazzaniga, 1974, 1975), though such decisions are more likely to be made at an unconscious level in interaction between the stimuli to be processed and the child's preference for a processing approach. At.the highest level of function, the two modes of processing operate in a complementary manner, achieving maximal interhemispheric integration of processing or, in Bogen, Dezure, Tenouten, and Marsh's (1972) terminology, "cerebral complementarity.'' For example, right hemisphere
210
CHAPTER 12
function (simultaneous processing) is important in contributing to letter and word recognition during the formative stages of learning to read. It is a more complex function handled primarily through successive processing in the intermediate learning stage, due to its linguistic nature. Highly skilled readers who have mastered the component skills of reading, making it an automatic function, demonstrate extensive use of both processes in reading (Cummins & Das, I977). When first learning to read, successive processing is most important and many children with difficulties in learning to read have problems with successive processing (Cummins & Das, I977). This is also consistent with the findings of higher Wechsler performance than verbal IQ in most groups of reading-disabled children. Performance IQ is almost certainly more closely related to the simultaneous processing of information than to successive processing whereas the converse relationship holds for the verbal IQ, the latter being a language task as is reading. Most (though certainly not all) young children with reading and related language problems have greater difficulty with sequential processing (Kamphaus & Reynolds, 1987). The traditional verbal-nonverbal distinction between the two hemispheres, although in part accurate, is likely an overcharacterization and simplification of hemispheric differences (Dean, I984). Hemispheric differences seem more related to the method by which information is processed than to the specific modality of presentation (Brown & Hecaen, 1976; Dean, 1984; Reynolds, 1981b). The K-ABC seems to be the best available measure of intelligence to quantify this bimodal functioning. Although specific neural substrates related to K-ABC performance remain to be detailed, the scale appears useful as a tool to gain insight into the relative efficacy of the two hemispheres.
pantomime and that are responded to motorically), to assess the intelligence of children with speech or language handicaps, of hearing-impaired children, and of those who do not speak English. It is particularly useful as part of the assessment of children suspected of aphasia and certain of the apraxias. However, the Nonverbal scale is useful as an estimate of general intellectual level only and cannot be subdivided into Sequential or Simultaneous Processing scales. All of the K-ABC global scales (Sequential Processing, Simultaneous Processing, Mental Processing Composite, Achievement, and Nonverbal) yield standard scores, with a mean of IOO and standard deviation of I5, to provide a commonly understood metric and to permit comparisons of mental processing with achievement for children suspected of learning disabilities. Furthermore, use of this metric allows for easy comparison of the K-ABC global scales to other major tests of intelligence and to popular individually administered tests of academic achievement, provided that the standardization samples are comparable (see Reynolds, I984). The availability of age-corrected deviation scaled scores for all intraindividual score comparisons offers another advantage to the neuropsychologist. The use of other equivalent-type scores, such as age or grade equivalents, has long been a problem for interpretation of tests in child clinical neuropsychology (see Reynolds, I98Ia, 1984). Standard scores are clearly the most appropriate mechanism for score comparison and the analysis of fluctuations in individual performance, one of the hallmarks of current (and past) neuropsychological approaches to test interpretation. . The K-ABC is comprised of 16 subtests, not all of which are administered to any age group. Children aged 2! are given 7 subtests, age 3 receives 9 subtests, ages 4 and 5 receive II subtests (but not precisely the same set of tasks, due to developmental and neuropsychological processing changes), age 6 receives I2 subtests, and the peak of 13 subtests is given to those aged 7 through 121. Also, attempting to be sensitive to children's development, testing An Overview of the Scales time ranges from about 30 minutes for 2!-year-olds to 1 hour 20 minutes (including the Achievement scale) The intelligence scales of the K-ABC consist of for 7- to 12!-year-olds. The Mental Processing subsubtests that are combined to form scales of Sequen- tests yield standard scores with a mean of 10 and tial Processing, Simultaneous Processing, and the standard deviation of 3, modeled after the familiar Mental Processing Composite, a summary score re- Wechsler scaled score. Achievement subtests, on the flective of the sequential and simultaneous scales. On other hand, yield standard scores with a mean of tOO the separate Achievement scale, subtests are com- and a standard deviation of 15, which permits direct bined to form a global Achievement score. The K- comparisons of the mental processing global scales ABC also includes a special short form of the Mental with individual achievement areas. Kamphaus and Processing Composite, known as the Nonverbal Reynolds ( I987) have developed a new organization scale (comprised of tasks that can be administered in of the K-ABC subtests into a more traditional system
K-ABC
that allows derivation of a verbal IQ and a Reading Composite in addition to a Global Composite IQ and the typical K-ABC scales. Those who prefer a Wechsler-type approach to test data may find this reorganization of interest. The Kaufmans have organized the K-ABC as shown in Figure l.
Administration and Scoring
211
Mental Processing Scale Sequential Processing Scale Hand Movements (ages 2Yz-12Vz years)* Imitating a series of hand movements in the same sequence as the examiner performed them Number Recall (ages 2Vz-12Vz) Repeating a series of digits in the same sequence as the examiner said them Word Order (ages 4-12Y2) Touching a series of pictures in the same sequence as they were named by the examiner, with more difficult items employing a color-interference task Simultaneous Processing Scale Magic Window (ages 2Vz-4) Identifying a picture that the examiner exposes by moving it past a narrow slit or "window," making the picture only partially visible at any one time Face Recognition (ages 2Vz-4)* Selecting from a group photograph the one or two faces that were exposed briefly in the preceding photograph Gestalt Closure (ages 2Vz-12Vz) Naming the object or scene pictured in a partially completed "inblot" drawing Triangles (ages 4-12Vz)* Assembling several identical triangles into an abstract pattern that matches a model Matrix Analogies (ages 5-12Vz)* Selecting the picture or abstract design that best completes a visual analogy Spatial Memory (ages 5-12Vz)* Recalling the placement of pictures on a page that was exposed briefly Photo Series (ages 6-12Vz)* Placing photographs of an event in chronological order. Achievement Subtests Expressive Vocabulary (ages 2Vz-4) Naming the object pictured in a photograph Faces & Places (ages 2Vz-12Vz) Naming the well-known person, fictional character, or place pictured in a photograph or illustration Arithmetic (ages 3-12Vz) Answering a question that requires knowledge of math concepts or the manipulation of numbers Riddles (ages 3-12Vzl Naming the object or concept described by a list of three characteristics Reading/Decoding (ages 5-12Vz) Naming letters and reading words Reading/Understanding (ages 7-12Vz) Acting out commands given in written sentences
Administration and scoring procedures for the K-ABC are available in the K-ABC Administration and Scoring Manual (Kaufman & Kaufman, 1983a). One important aspect of K-ABC administration that deserves special mention, however, is the notion of teaching items. The first three items of each mental processing subtest (the sample and the first two items appropriate for a child's age group) are designated as teaching items. On these items the examiner is required to teach the task if the child fails on the first attempt at solving the item. By "teaching the tas'k" it is meant that the examiner is allowed the flexibility to use alternate wording, gestures, physical guidance, or even a language other than English to communicate the task demands to the child. The examiner is not allowed to teach the child a specific strategy for solving the problem, however. This built-in flexibility is particularly helpful to preschoolers, minoritygroup children, and exceptional children, who sometimes perform poorly on a task in a traditional IQ test, not because of a lack of ability, but because of an inability to understand the instructions given. Kaufman and Kaufman ( l983b) discuss the concept of teaching items in greater detail and note, as is evident from Table 3(later in this chapter), that this built-in flexibility has not adversely affected the reliability of the K-ABC. The extensive use of sample practice and teaching items on the K-ABC helps to ensure that the various subtests actually measure what they were intended to measure. Many intelligence tests contain basic language concepts, such as "next," "same," "alike," "opposite," "backwards," and "after," that less than half of children in kindergarten and a significant number of primary-grade children do not understand (Kaufman, 1978). Thus, a child may perform poorly on a test because of a very specific language deficit, despite the fact that the test was intended to measure psychomotor speed, memory, spatial ability, or some other intellectual ability. Violations of standardized procedure to explain the directions to children make the obtained scores essentially unusable, as the amount and direction of error FIGURE 1. K-ABC subtests. Asterisks denote subtests that also introduced through such procedures are unknown make up the nonverbal scale.
212
CHAPTER 12
and are not constant across children (or across examin~rs). Because the K-ABC was standardized by usmg the sample and teaching items to ensure the child's understanding of the task, influences on performance are built into the normative data and the error introduced is included in the standard errors of measurement reported in the K-ABC Interpretive Manual (Kaufman & Kaufman, 1983b). The K-ABC basal and ceiling rules, referred to as starting and stopping points in the K-ABC Administration and Scoring Manual (Kaufman & Kaufman, 1983a), are also somewhat different from those of many existing intelligence tests. The first rule_ for administering the K-ABC subtests is very straightforward: examiners are instructed to start and stop testing at the items designated as starting and ~topping points for the child's age group. The set of Items between the starting and stopping points are, therefore, designed based on standardization data, to represent a full range of difficulty for the child's age group. The first basal and ceiling rule is very straightforward, but it is also rigid. Hence, several supplemental rules are given to allow examiners to find items of appropriate difficulty for children at the ends of the distribution of ability. The K-ABC also incorporates a very simple discontinue rule, one that is the same for all K-ABC subtests. As noted earlier, the Nonverbal scale is intended for use with children for whom administration of the regular K-ABC (and virtually all other well-normed, standardized measures of intelligence) would be inappropriate: those who are hearing impaired, have speech or language disorders, other communication handicaps, or are limited English proficient. The Nonverbal scale yields a global estimate of intelligence; however, a method for profile interpretation of subtest scaled scores is offered in the K-ABC Interpretive Manual (Kaufman & Kaufman, 1983b). Most well-normed intelligence tests that are applicable to communications-handicapped children are very narrow and give a quite limited view of these children's intelligence (e.g., the Columbia Mental Maturity Scale). Although the K-ABC Nonverbal scale has limitations in this regard, of those tests of mental ability with adequate technical/psychometric characteristics the K-ABC Nonverbal scale provides the broadest sampling of abilities. This breadth of assessment should enhance studies of these children and their development. The lack of adequately normed scales with any breadth of assessment has been a hindrance not only to clinical assessment of children with communication disorders but also to research in the area (Reynolds & Clark, 1983). The Nonverbal scale of the K-ABC is the best-normed, psycho-
metrically most sophisticated nonverbal scale presently available. Several features of the K-ABC also make it a very attractive scale for use in evaluating very low-functioning adolescents. Reynolds and Clark ( 1985) described_ a useful approach to applying the K-ABC (and certam other tests) to this difficult population; a special record form is available from the publisher of the K-ABC for this procedure and several other special, newly derived scales. Standardization Th~ K-ABC was standardized on a sample of 2000 children, using primarily 1980 U.S. Census figure_s. Th~ sample was s~atified by age, sex, geographic regiOn, race/ethmc group, parental educational attainment (used as a measure of socioeconomic status (SES)), community size and educational placement (regular class placeme~ vers_us plac~ment in a variety of programs for exceptional children). Educational placement is an infrequently used stratification variable. Typically, exceptional children are excluded from the standardization samples for individually administered tests. An attempt was made to include representative proP?rtions of learning-disabled, mentally retarded, gifted and talented, and other special populations in ~e standardiza~on sample according to data provided by the National Center for Education Statistics and the U.S. Office of Civil Rights. An overview of the K-ABC standardization sample, indicating its match to the U.S. Census data for the variables of gCC?graphic region, race/ethnic group, parental education, and community size, is presented in Table 2. O~e~l: the mat~h is quite good, although high-SES mmonties (specifically blacks and Hispanics) were statistically significantly oversampled. The real effect was small, however, resulting in an overestimation of black and Hispanic populations' total scores by around two points on the Mental Processing Composite (overestimation here referring to the mean scores of these groups had their representation in the standardization sample been a perfect match to the 1980 U.S. Census Bureau statistics).
Reliability Split-half reliability coefficients for the K-ABC global scales ranged from 0.86 to 0.93 (mean = 0.90) for preschool children, and from 0.89 to 0.97 (mean = 0.93) for children aged 5 to 12!. Mean internal consistency reliability coefficients for the global scales and the subtests are shown in Table
213
K-ABC
TABLE 2. Representation of the K-ABC Standardization Sample (Ages 2% through 12%) by Geographic Region, Race/Ethnic Group, Parental Education, and Community Size
Stratification variable
K-ABC sample
U.S. population
N
%
(%)
Region East North Central South West
401 565 628 406
20.0 28.2 31.4 20.3
20.3 26.5 34.0 19.2
Parental education Less than high school High school education Some college College degree
384 813 413 390
19.2 40.6 20.6 19.5
2l.l 4l.l 19.8 18.0
3. A test-retest reliability study was conducted with 246 children retested after a 2- to 4-week interval (mean interval 17 days). The results of this study showed good estimates of stability that improved with increasing age. For the Mental Processing Composite, coefficients of .83, .88, and .93 were obtained for the Achievement scale at each age group. Further details of the test-retest study can be found on pp. 81-84 of the K-ABC Interpretive Manual (Kaufman & Kaufman, 1983b). The test-retest reliabilitY coefficients for the global scales, and to a lesser extent the internal consistency (split-half) coefficients, show a clear developmental trend, with coefficients for the preschool ages being smaller than those for the school-age range. This trend is consistent with the known variability over time that characterizes preschool children's standardized test performance in general. As shown in Table 3, the reliability coefficients of the K-ABC subtests typically meet or exceed those for comparable intelligence tests with (Kaufman & Kaufman, l983b) subtest reliabilities ranging from 0.72 to 0.89 for preschool children and from 0. 71 to 0. 92 for school-age children. Test-retest coefficients for the subtests given in the K-ABC Interpretive Manual (Kaufman & Kaufman, l983b) show the same predictable developmental trend identified for the global scales, and are consistent with the values for such traditional intelligence scales as the various Wechsler Scales and the McCarthy Scales.
Stratification variable Race or ethnic group White Total minorities Black Hispanic Native American, Asian, or Pacific Islander Community size Central city Suburb or small town Rural area
K-ABC sample
U.S. population
N
%
(%)
1,450 550 3ll 157 82
72.5 27.5 15.6 7.8 4.1
73.1 26.8 14.5 9.1 3.2
579 876
28.9 43.8
27.9 43.8
545
27.2
28.3
Validity The K-ABC Interpretive Manual (Kaufman & Kaufman, 1983b) includes the results of 43 validity studies, an impressive amount of prepublication research that is all too uncommon in test manuals. Studies were conducted on aspects of construct, concurrent, and predictive validity. In addition, several of the studies were conducted with samples of exceptional children, including samples classified as hearing impaired, physically impaired, gifted, mentally retarded, and learning disabled. Topics considered under construct validity include developmental changes, internal consistency, factor analysis, and convergent and divergent relationships with other measures. Of particular interest are the data on factor analysis and correlations given in detail in the K-ABC manuals and elsewhere (e.g., Kamphaus & Reynolds, 1987; Reynolds, 1984); only a synopsis of the findings in the K-ABC Interpretive Manual is given here. A recent book by Kamphaus and Reynolds ( 1987) details a wealth of K-ABC validity data. Serious K-ABC users are referred to that volume for a book-length treatment of K-ABC research data and clinical findings. Several prepublication factor analytic studies were conducted with early research editions of the KABC (Kaufman, Kaufman, Kamphaus, & Naglieri, 1982; Naglieri, Kaufman, Kaufman, & Kamphaus, 1981). In addition, Kamphaus, Kaufman, and Kauf-
214
CHAPTER 12
TABLE 3. Average Reliability Coefficients for the K-ABC Scales and Subtestsa
Scale or subtest
Preschool children (ages 2!1 through 4) N = 500 mean r across age
School-age children (ages 5 through 12!1) N = 1500 mean r across age
Global scalesb Sequential processing Simultaneous processing Mental processing composite Achievement Nonverbal
0.90 0.86 0.91 0.93 0.87
0.89 0.93 0.94 0.97 0.93
Mental processing subtestsc Magic window Face recognition Hand movements Gestalt closure Number recall Triangles Word order Matrix analogies Spatial memory Photo series Achievement subtestsc Expressive vocabulary Faces & places Arithmetic Riddles Reading/decoding Reading/understanding
0.72 0.77 0.78 0.72 0.88 0.89 0.84
0.85 0.77 0.87 0.83
0.76 0.71 0.81 0.84 0.82 0.85 0.80 0.82
0.84 0.87 0.86 0.92 0.91
•The values shown for preschool children are the mean coefficients for three age groups (2~. 3, and 4), and the values shown for school-age children are the mean coefficients for eight age
groups.
bComposite score reliability coefficients were computed based on Guilford's (1954) formula. cAll coefficients for the subtests were derived using the split-half method and corrected by the Spearman-Brown formula.
man (1982) factor analyzed the published edition of the K-ABC using the 2000 children from the standardization sample. All of these exploratory studies support the division of the intelligence scales into the Sequential and Simultaneous Processing scales. An overview of the findings from the Kamphaus et al. (1982) study is presented in Table 4. Subsequent confirmatory factor analytic research provides strong support for the two-factor sequential and simultaneous processing model (Willson, Reynolds, Chatman, & Kaufman, 1985) but is less enthusiastic in support of a distinct Achievement scale. Hence, it is possible to find a better mathematical fit to the subtest's structure when the Achievement scale subtests are included; however, the psychological meaning of such structures is high-
ly suspect, as they are being developed after the fact in a manner reminiscent of Monday morning quarterbacking and not from a prior theoretical base, as did the Kaufmans. The subtests of the Achievement scale do show their largest loadings on a separate, clearly identifiable factor, as Kaufman and Kaufman (1983a) proposed, yet each shows large secondary and tertiary loadings on the two mental processing factors. Though some would interpret the Achievement scale as a good measure of verbal intelligence or perhaps even "g" (e.g., Keith, 1985), this seems ill advised as it involves so many assumptions regarding equal opportunity to acquire certain knowledge and is, in addition, extremely inferential, relying primarily upon acquired factual knowledge rather than the manipulation of information to solve a problem.
K-ABC
215
TABLE 4. Mean Sequential/Simultaneous Factor Loadings for Preschool and School-Age Childrena School-age (ages 5 through Preschool (ages Scale Sequential processing Hand movements Number recall Word order Simultaneous processing Magic window Face recognition Gestalt closure Triangles Matrix analogies Spatial memory Photo series a Factor loadings
italicized.
2~
through 4)
12~)
Sequential
Simultaneous
Sequential
Simultaneous
0.60 0.64 0.69
0.19 0.28 0.32
0.37 0.77 0.75
0.43 0.15 0.26
0.21 0.28 0.23 0.36
0.63 0.40 0.59 0.47
0.08 0.20 0.30 0.24 0.26
0.53 0.72 0.57 0.60 0.69
were obtained by principal factor analysis with varimax rotation. Factor loadings of 0.35 and above are
Of particular interest to various clinicians is the relationship of the K-ABC to the WISC-R. Numerous studies involving the K-ABC and WISC-R are reported in the K-ABC Interpretive Manual (Kaufman & Kaufman, 1983b). In a study of 182 children enrolled in regular classrooms, the Mental Processing Composite (MPC) correlated 0. 70 with WISC-R Full-Scale IQ (FSIQ). Hence, the K-ABC Mental Processing scales and the WISC-R share a 49% overlap in variance. These findings indicate that the KABC does bear a substantial relationship to the widely used WISC-R; yet these data also indicate that the K-ABC is hardly a duplicate of the WISC-R, but rather possesses its own unique contribution to the field of intelligence measurement. Also of interest in this sample is the standard score difference between the MPC and FSIQ. The K-ABC, based on 1980 U.S. Census data, was shown to be about 3 points tougher (mean MPC = 113.6) than the WISC-R (mean FSIQ = 116.7), based on a sample of 182 children from regular classes (Kaufman & Kaufman, 1983b). Kamphaus and Reynolds (1987) found that the K-ABC norms are a couple of points tougher than older tests such as the 1972 Stanford-Binet. The one exception to this rule is the McCarthy Scales where no consistent pattern has emerged. The K-ABC norms are also quite similar to those for tests normed in the 1980s such as the Stanford-Binet Fourth Edition.
Relationship of Individual Subtests to Neuropsychology Although the simultaneous/ sequential processing dichotomy was the most important factor in subtest selection, many of the separate subtests of the K-ABC resemble traditional tasks in neuropsychological assessment. Kaufman, O'Neal, Avant, and Long (1987) summarized these similarities. Luria used tasks similar to Hand Movements and Word Order to assess motor conditions and higher cortical functions of the left temporal lobe. Number Recall and Matrix Analogies are generally considered marker tests for sequential and simultaneous processing, respectively. These tasks are also included in a Luria-based successive-simultaneous test battery. Tasks like Gestalt Closure have long been accepted as measures of simultaneous processing and right hemisphere processing. Research with tests similar to Face Recognition and Face & Places is continuing in an effort to establish the localization and lateralization for the recognition of familiar and unfamiliar faces. Finally, Kaufman et al. (1987) noted that Triangles is an adaptation of Kohs' s Block Design test (Kohs, 1927). A version of the block design test is included in all of the Wechsler Scales and in Goldstein's tests for brain damage (Goldstein, 1948). Knowledge of the background of these tasks adds to the clinician's interpretation of the K-ABC at the subtest level. Further investigation
216
CHAPTER 12
of the localization value and developmental progression of these tasks as presented in the K-ABC is certainly warranted.
Specific Neuropsychological Research on the K-ABC The amount of neuropsychological research devoted to the K-ABC is limited at present, but the few studies available are generally supportive of its role in adding useful information to general neuropsychological batteries. Studies relating the K-ABC to the child forms of the Halstead-Reitan Neuropsychological Battery are noticeably absent from the literature and are certainly needed. Such studies are a necessary component of any comprehensive assessment of utility of the K-ABC to determine how it precisely fits into the neuropsychological assessment process. Several studies are available relating the K-ABC to the Lucia-Nebraska Neuropsychological BatteryChildren's Revision (LNNB-CR) and several have compared the K-ABC and WISC-R with relatively well-defined samples of neurologically impaired children. Other studies have looked at the K-ABC alone with neurologically impaired children.
Correlation with the Luna-Nebraska Scales Of particular importance to understanding how the K-ABC might contribute to neuropsychology, clinically as well as in the research setting, will be understanding how the K-ABC is related to existing neuropsychological scales. To be useful in neuropsychological assessment, the K-ABC should be related to existing measures, but not so closely that its use is merely redundant with preexisting scales. Although no data are available relating the K-ABC to the Halstead-Reitan techniques, several studies have been reported comparing the K-ABC with the LNNB-CR. The first such report was by Snyder, Leark, Golden, Grove, and Allison (1983). Synder et al. ( 1983) evaluated 46 elementary school children (ages 8 to 12!) who had been referred for a variety of learning difficulties. All children were administered the K-ABC, WISC-R, and LNNB-CR as described in Golden (1978). Correlations between the K-ABC and LNNB-CR ranged from -0.01 (LNNB-CR Writing scale with K-ABC Simultaneous) to -0.64 (LNNB-CR Intelligence with K-ABC Sequential and MPC). The LNNB-CR Intelligence scale correlated highest of all LNNB-CR scales with the K-ABC global scales (SIM -0.54,
SEQ -0.64, MPC -0.64, Nonverbal -0.51, ACH -0.26). Correlations with the LNNB-CR are negative in direction because high scores on the LNNBCR are indicative of increasingly high levels of pathology, i.e., the LNNB-CR is scored negatively, not positively as are most mental tests such as the KABC. In a stepwise multiple regression, from three to five LNNB-CR scales were required to maximize the prediction of each of the K-ABC scales. The pattern of relationships is much as one would anticipate. Results of these analyses are shown in Table 5. After the intelligence scale, the visual and the motor scales contributed the greatest to the prediction of the Simultaneous scale; for the Sequential scale, following intelligence, the Rhythm and the Receptive Speech scales were the best predictors of performance. The MPC and K-ABC Nonverbal scales showed the same pattern as the Simultaneous scale, and not surprisingly because these scales overlap so much with the Simultaneous scale. The K-ABC ACH scales correlated from -.50 to -.58 with the school-related scales of the LNNB-CR (e.g., Expressive Language, Reading). In the multiple regression, academic skiJJs
TABLE 5. Multiple Correlations between Subtest Standard Scores on the Luria-Nebraska Neuropsychological BatteryChildren's Revision (Predictors) and K-ABC Global Scale Standard Scores (Criteria) (N = 46) 8 K-ABC scale Sequential processing
Simulataneous processing
Mental processing composite Achievement
Nonverbal
Luria-Nebraska Predictor
R
Intelligence Rhythm Receptive speech Motor Tactile Intelligence Visual Motor Intelligence Motor Visual Arithmetic Receptive speech Writing Memory Intelligence Visual Motor
0.582 0.656 0.677 0.694 0.714 0.521 0.641 0.694 0.638 0.697 0.733 0.579 0.620 0.655 0.679 0.499 0.608 0.669
•Adapled from Snyder et al. (1983).
K-ABC
again dominated prediction of the ACH scales. The K-ABC mental processing scales also were related significantly to each WISC-R IQ in this study; correlations ranged from 0.35 between SEQ and PIQ to 0. 72 between the MPC and FSIQ and the Nonverbal and FSIQ. The K-ABC ACH scale correlated 0.66 with FSIQ, 0.77 with VIQ, but only 0.28 with PIQ. After examining the overall pattern of correlations in the study, Snyder et al. concluded that the relationships revealed were "basically consistent" with the model of intelligence on which the K-ABC was based and the theoretical perspective of Luria in particular. Snyder et al. also concluded that the KABC provides additional information, beyond the WISC-R and LNNB-CR, that should be useful to the clinical neuropsychologist. We agree. The pattern of correlations as well as the magnitude of correlations is encouraging. The K-ABC is clearly related to children's neuropsychological function as determined by the LNNB-CR, but not so much so that K-ABC scores are simply redundant with other neuropsychological test results; the K-ABC apparently has something to add. Another recent study supports this conclusion. In a similar study with a larger sample (65 children), Leark, Snyder, Grove, and Golden (1983) provide more detailed information. Table 6 displays the correlation matrix between the K-ABC global scales and the subscales of the LNNB-CR. Several
interesting patterns emerge here. The LNNB-CR subscales that are known to be the most sensitive to brain impairment (Pathognomonic and Intellectual) are clearly the most closely related to performance on all of the K-ABC global scales in the Leark et al. study. However, there is very little overlap in item content from these scales to the K-ABC and yet the K-ABC seems sensitive to deficits in cortical functioning, at least at the level of the higher information processing functions of the brain. The SEQ-SIM distinction and the separate Achievement scale of the K-ABC receive support from the pattern of correlations in Table 6 as well. The school-related subscales of the LNNB-CR correlate considerably higher with the K-ABC ACH scale than with the mental processing scales. Clear differentiations occur elsewhere as well. When evaluating correlations with the LNNBCR Rhythm scale, one sees that the SEQ scale is significantly related to the Rhythm scale (r = -0.40) whereas the SIM scale is not (r = -0.13). A similar pattern is observed for the LNNB-CR Receptive Speech scale. Additionally, the K-ABC Nonverbal scale is more highly correlated with the LNNB-CR Motor scale than is any other K-ABC scale. The Nonverbal and the SIM scales are more closely related to the LNNB-CR Visual scale than is the SEQ scale. Most of these relationships are intuitively obvious but their actual occurrence, especially given the moderate magnitude of the relationships, is
TABLE 6. Correlations between K-ABC Global Scales and LuriaNebraska Neuropsychological Battery-Children's Revision Summary Scales (N = 65)0 Correlation with K-ABC global scale Luria-Nebraska scale Motor Rhythm Tactile Visual Receptive Expressive Writing Reading Arithmetic Memory Intelligence Pathognomic Left Right "Adapted from
Sequential processing
Simultaneous processing
MPC
Achievement
Nonverbal
-0.382 -0.405 -O.ll5 -0.252 -0.515 -0.323 -0.324 -0.210 -0.307 -0.471 -0.567 -0.598 -0.171 -0.075
-0.424 -0.132 -0.320 -0.498 -0.355 -0.154 -0.144 -0.060 -0.152 -0.300 -0.570 -0.469 -0.379 -0.319
-0.456 -0.282 -0.270 -0.461 -0.482 -0.260 -0.248 -0.066 -0.258 -0.427 -0.645 -0.606 -0.335 -0.244
-0.242 -0.370 -0.221 -0.192 -0.600 -0.614 -0.539 -0.618 -0.607 -0.629 -0.439 -0.649 -0.222 -0.091
-0.481 -0.199 -0.321 -0.489 -0.427 -0.259 -0.246 -0.012 -0.202 -0.356 -0.599 -0.656 -0.352 -0.301
Leark et at. (1983).
217
218
CHAPTER 12
certainly encouraging with respect to potential contributions of the K-ABC to neuropsychological assessment and research.
Relationships with Other Neuropsychological and Neurological Test Results Several early studies have related the K-ABC to other neuropsychologically based scales and to hard evidence of neurological impairment. Much more is needed but progress is evident in these few studies. Telzrow, Redmond, and Zimmerman (1984) examined test score patterns of children classified into Boder's three subtypes of dyslexia-dysphonetics, dyseidetics, and mixed. Telzrow et al. attempted to determine whether the WISC-R or the K-ABC would be more closely aligned with Boder's neuropsychological classification of reading disorders. WlSC-R scores were grouped into Bannatyne's four categories of neuropsychological functions, as these groupings (Verbal Conceptualization, Sequencing, Spatial, and Acquired Knowledge) seem most clearly related to various schemes for regrouping the Wechsler subtests to Boder's diagnostic categories. These children could not be differentiated on the basis of their Bannatyne patterns on the WISC-R. Boder' s subtypes were randomly distributed across Bannatyne's suggested patterns. On the K-ABC a significant relationship (p < 0. 01) occurred between Boder classification and the pattern of SequentialSimultaneous score differences on the K-ABC. In particular, Boder' s dysphonetic dyslexics were far more likely to display a SIM greater than SEQ pattern than were the other diagnostic groups. Though Telzrow et al. used a small sample (N = 23) the results are impressive with large effect sizes. The pattern of K-ABC results was precisely in keeping with predictions of underlying neuropsychological deficits in Boder's classification scheme. A followup study with a second sample showed similar results with one major difference (Telzrow, Century, Harris, & Redmond, 1985). In the latter study, 27 children with reading disorders were compared according to their performance on the WISC-R and the K-ABC,just as in the previous study. In the Telzrow et al. ( 1985) work, considerable consistency was found among the K-ABC, WISC-R, and the Boder results. Threequarters of the dyslexic children who displayed a Bannatyne WISC-R pattern of Spatial greater than Sequential also showed a K-ABC pattern of Simultaneous greater than Sequential scores. In each case, it was the dysphonetic reading-disabled children who
showed the higher Spatial and Simultaneous scores. Results with dyseidetic children are less clear on this point, most likely due to the small number of readingdisabled children showing this particular problem. The results of these two studies, considered· in concert, though ambiguous with regard to the WISC-R, provide support for the use of the K-ABC in evaluating children with neuropsychologically related reading problems. Hooper and Hynd ( 1985) also examined the utility of using the K-ABC in differential diagnosis of developmental dyslexia according to Boder's subtypes. Their sample consisted of 30 normal readers and 87 reading-disabled students (two or more grade levels below their grade placement). On the basis of the Boder Reading-Spelling Pattern test, 32 of the reading-disabled students were classified as "nonspecific'' (significantly low achievement in reading, but reading and spelling pattern typical of normal readers), and 55 were classified as dyslexics. Of the dyslexics, 30 were classified as dysphonetics, 5 as dyseidetics, and 20 as alexics. Evaluation of the performance of all of these children on the K-ABC indicated the discriminatory value of the Sequential Processing scale but not the Simultaneous Processing scale in terms of Boder's subtypes. Normal readers scored significantly higher than all of the readingdisabled subtypes on the Sequential Processing scale. Sequential/simultaneous discrepancies, however, did not discriminate between the subtypes of dyslexia. The small number of dyseidetics in the study may have contributed to this outcome. When combined with the results ofTelzrow et al. (1985), Hooper and Hynd's results provide support for the simultaneous > sequential pattern in dyslexia (particularly dysphonetic subtype) but the importance of the simultaneous < sequential pattern in dyslexia (as it may apply to dyseidetics) is much less clear. Hooper and Hynd note, "It would be of interest to examine the discriminant validity of the K-ABC with empirically, as opposed to clinically, derived subgroups of developmental dyslexia." In a conceptually related work, Dietzen ( 1986), using dichotic listening tasks to assess hemispheric specialization, found a positive correlation between the K-ABC simultaneous processing scores and hemispheric specialization for nonverbal processing for 75 children of low SES. Dietzen reported a significant positive relationship between sequential processing on the K-ABC and degree of hemispheric specialization for verbal information in this same sample. Additional studies have added to our knowledge through the use of physical evidence of neurological damage and relating it to K-ABC results. Morris and
K-ABC
Bigler (1985) investigated whether the K-ABC SEQ and SIM scales can be related to left and right hemisphere functioning and whether the K-ABC is better able to indicate neuropsychological deficits than the WISC-R. In this study, 79 children ages 6 to 12 years were administered the WISC-R, the K-ABC, and several neuropsychological measures of left and right hemisphere functioning. Neurological data, including EEGs and CAT scans, were also available. Neuropsychological test scores were collapsed into two composite scores for each subject, right hemisphere (RH) and left hemisphere (LF). Twenty-five children who were right-handed and neurologically impaired were divided into three groups according to their KABC scores: SIM >SEQ, SEQ> SIM, and SIM = SEQ. A one-way MANOVA revealed a significant difference among these groups on the RH and LH scores (p < 0.05) but not for WISC-R groups using VIQ-PIQ differences for classification (p = 0.41). Further analyses revealed that the key to undertanding these differences was the inability of the WISC-R to detect RH dysfunction. Whereas both scales seemed to pick up LH dysfunction, only the K-ABC could diagnose RH problems at a statistically significant level. These results also are consistent with lateralization of sequential and simultaneous processing of the left and right hemispheres, respectively, giving evidence also consistent with the findings of Leark et al. (1983). Shapiro and Dotan ( 1985) provided a replication of sorts of the Morris and Bigler ( 1985) study, though their sample size (N = 27) makes statistical comparisons less relevant. The pattern of Shapiro and Dotan's results is most interesting, however. These researchers compared two groups of children with neurological impairment defined as focal versus nonfocal on the K-ABC and on the Wechsler Scales. All had neurological exams and most had EEGs orCAT scans, or both. Two groups were formed. The nonfocal group was determined to be children with normal exams with the exception of soft neurological signs that were not unilateral in nature. Most of these 13 children were diagnosed as ADD, LD, BD, or developmental delay. The 14 children in the focal group had lateralized deficits on the neurological exam and a focal abnormality on the EEG or structural damage on the CAT scan. The nonfocal group contained 10 males and 3 females (a typical occurrence) and the focal group, 7 males and 7 females. Of particular interest here, Verbal-Performance IQ differences on the Wechsler Scales were not related to the presence of focal neurological findings. "However, ... , one finds that of the eight children with significantly lower sequential than simultaneous scores, six had predominantly left hemisphere find-
219
ings. The two children with right-hemisphere findings were non-right-banders. Of the six children with significantly lower simultaneous scores, four showed right hemisphere findings. Of the two with left hemisphere findings, one was left-handed and the other had primarily left occipital findings with some visual impairment" (Shapiro & Dotan, 1985, p. 6). The KABC Simultaneous scale should always show considerable impairment for children with visual problems. Even with the small sample size available, highly significant (p < 0.005) relationships occurred between left-brain focal findings and lower sequential than simultaneous scores and focal right-brain problems and lower simultaneous than sequential scores. Shapiro and Dotan noted that the relationship was stronger for males than for females. They concluded in part that ''the lack of relationship of verbal/performance discrepancies on the Wechsler tests to neurological findings may reflect the lack of homogeneity of function in those scales as compared to the K-ABC" (p. 7). These results and the subsequent conclusions are clearly in accordance with the results of Morris and Bigler and the two studies by Telzrow et al. Taken as a whole, empirical results available thus far are impressive in their support of the potential of the K-ABC in contributing to the tasks of the neuropsychologist. Not only does the K-ABC model appear useful, but the psychometric integrity of the K-ABC lends the scale to other theoretical approaches, thus avoiding the fate of the inadequately developed ITPA. The K-ABC seems to be related, even at this early stage in its career, to neuropsychological functioning both theoretically and empirically. The moderate but consistent relationships with other neuropsychological batteries bode well for its use, indicating that it does provide additional or at least distinctive information. It also seems more closely related to recent neuropsychological models of higher cognitive processes than the Wechsler series. The Wechsler series is much more researched at this point, however, and we should proceed cautiously with the K-ABC; existing data are very promising and dictate that research should continue full speed ahead.
Clinical Neuropsychological Applications of the K-ABC The general field of clinical neuropsychology is seen by many as a set of tests and related techniques for relating observed, quantifiable behavior to the
220
CHAPTER 12
integrity of an individual's functional neurological organization and structure. The efficacy of these techniques in the diagnosis and remediation of learning disturbances has been well documented over the last decade (e.g., Bradley, Battin, & Sutter, 1979; Gaddes, 1981; Golden, 1978; Hartlage, 1982; Hartlage & Reynolds, 1981; Knight & Bakker, 1980; Reitan & Davison, 1974; Rourke & Orr, 1977). It is not surprising, then, that many individuals believe that to become a successfully functioning clinical neuropsychologist, it is only necessary to master the technical skills involved in the administration and scoring of tests like the Halstead-Reitan Neuropsychological Test Battery. Such thinking is a gross oversimplification of the profession and function of clinical neuropsychology. Much more than being a set of techniques, the principal tool of neuropsychology is the paradigm it offers for viewing and interpreting individual test data. Without the provision of a strong paradigm, clinical neuropsychology could not have progressed to the point of applicability and generalizability that has emerged today. As with other areas and subspecialties of psychology, several competing paradigms and theories exist in clinical neuropsychology, and any one of these may be the most appropriate for interpreting data for any single child (e.g., Ayers, 1974; Das et al., 1979; Luria, 1966; Pribram, 1971; Reynolds, 1981b; Tarnopol & Tarnopol, 1977; Wittrock, 1980). It is thus crucial to achieve an understanding of the various neuropsychological models of higher-order human information processing in order to be effective in translating test results into meaningful educational programs, i.e., teasing out the aptitude x treatment interaction for a single individual. The relationship of the KABC to a particular neuropsychological model, Vygotsky's Zones of Proximal Development, will be explored in some detail later in this chapter. Contemporary neuropsychological theories of the intellect and its function are presented in a number of sources (e.g., Das et al., 1979; Kolb & Whishaw, 1985; Luria, 1966; Pribram, 1971; Reynolds, 1981b; Wittrock, 1980) and will not be reviewed further here. Well-grounded, empirically evaluated theoretical models of neurological function enable one to make specific predictions about how children will perform under a given set of learning circumstances. Of course, one will not always be correct; but a good theory from which to work allows the psychologist to narrow down the number of alternative hypotheses (methods of remediation) considerably. Without a viable set of theories, one would be a completely stimulus-bound technician relying en-
tirely on trial-and-error experience and more or Jess shooting in the dark whenever encountering a child with a set of behavioral nuances and test scores not previously seen. The neuropsychological approach proffered here and by the K-ABC generally is one that matches up cognitive neuropsychological strengths with methods of presenting and acquiring information that rely most heavily on these strengths. Neuropsychological strengths in a child's ability spectrum may be in traditional areas of ability like linguistic processing or in a particular cognitive/learning style. Cognitive and/ or learning styles are now being more fully explored but seem almost certainly to be tied strongly to the underlying neuropsychological integrity, development, and preferences of the individual (Guyer & Friedman, 1975; Reynolds, 1980, 198lb). Of course, merely detecting cerebral dysfunction (or minimal brain dysfunction or minimal brain damage) is not a very useful exercise from an educational or rehabilitative perspective. It is the accurate description of the dysfunction (an integral part of diagnosis) leading to an educational program to enhance the child's acquisition of skills in a subject matter area that is of significance. The K-ABC and the WISC-R provide excellent instruments for determining a baseline standard of general mental ability against which to compare other, more specific scales, during the process of ipsative test score interpretation. The multidimensional scaling of these tests also makes them amenable to a variety of neuropsychological interpretive strategies (Hartlage, 1982; Kaufman, 1979a) despite the K-ABC's reliance on Luria's approach. As Hartlage (1982) has discussed, one can compare the functional integrity of the left and right temporal lobes through the Similarities versus Picture Arrangement contrast, or left and right parietal lobe function by contrasting the child's performance on Arithmetic and Block Design. Knowing that Arithmetic significantly exceeds Block Design or even that the child's left parietal lobe function presents as superior to right parietal lobe function is of little import in and of itself. Rather, the inferences that can be drawn from such a finding are the important focal point of the evaluation, and can be turned toward the design of educational programs. Certain cognitive skills tend to cluster together within an individual's overall functional level. Neuropsychological and certain other approaches to assessment allow one to make inferences about skills that have not actually been evaluated by knowing the correlates of these skills. Look-say, whole word, configurational approaches to reading seem to be moderated much more by the posterior right parietal
K·ABC
lobe in coordination with sections of the right occipital and temporal lobes than in the left hemisphere counterparts of these structures. Knowing that these structures in the right hemisphere function more effectively (efficiently or at a higher level) thus gives rise to the designation of a method for teaching reading to the child in question that can capitalize on the identified neuropsychological strengths. Additionally, there is no reason why behavioral methods cannot be employed as motivational strategies in such programs. The neuropsychological interpretation of many common psychological tests can generate good hypotheses for choosing a particular method of instruction, and operant psychology or one of its variants can assist in promoting the student's interest in and learning through the specified method. The K-ABC, designed with such approaches in mind, a priori, should help in such models. It can also offer information on the intact nature of a variety of processes for young children that are poorly tapped by most existing scales. Extensions of tasks such as Hand Movements, Word Order, and Spatial Memory can be useful tools rather than methods of testing. Attention seems to be to the area of cognitive rehabilitation of deficit skills most influenced by cognitive retraining and such an addition of teaching tasks can prove useful. The K-ABC is a nice fit to Reynolds' ( 1981 b) approach, an approach recently elaborated and integrated with cognitive and behavioral models (Reynolds, 1986).
Vygotsky's Zones of Proximal Development Vygotsky (translation, 1978) has offered another neuropsychologically related concept that may be amenable to assessment with the K-ABC, with some modification in the standardized administration procedures. Vygotsky espoused what continues to be a widely held ''truth'' even today as regards children's learning. "A well known and empirically established fact is that learning should be matched in some manner with the child's developmental level'' (Vygotsky, 1978, p. 85). Hunt (1961) in his nowclassic volume concurred and conceptualized this concept in part as the "problem of the match." Although true for normally developing children, such a inatch may be absolutely crucial for promoting the development of children with an immature or traumatized central nervous system. In formulating the concept of zones of proximal development (ZPD), Vygotsky rejected what he considered to be the three primary alternative developmental theories of learning (learning broadly de-
221
fined, as it should be, and not restricted to school learning). The first rejected theory centers on the assumption that processes of child development are independent of learning, which is considered an external process, not an activity involved in development. Accordingly, learning uses the achievements of development in acquisition of new knowledge rather than providing impetus for modifying the course of development. Vygotsky classified Piagetian theory under this theoretical rubric. The second rejected approach are theories assuming that learning is development, approaches that reduce development to a simple accumulation of all possible responses by the child. The third rejected theoretical proposition is based on a combination of the two prior positions, in which development is based on two different but related processes, maturation, which depends upon the development of the central nervous system, and learning, also considered here as a developmental process. In this apparently reciprocal relationship, maturation makes new learning possible, which then stimulates and pushes forward the maturation process. The latter is the approach that led educators in the past to conclude that the study of certain subjects (e.g., Latin) was of great value for mental development. Vygotsky framed the concepts of ZPD to provide a more adequate view of the relationship between learning and development. This concept can lead us to better cognitive retraining and rehabilitation programs for children who have resisted more traditional approaches. In determining and using ZPDs, two developmental levels must first be established. The first is the child's current, actual level of development. Vygotsky viewed this level as that already completed as a result of previous developmental cycles. This is a level determined by rigidly administered, standardized measures of intelligence and achievement. It is the level determined by what children can do on their own. The second level of development to determine is the level the child can achieve if an adult or more advanced, accomplished peer provides help through demonstration, asking leading questions, or actual collaboration leading the child to discover the answer. It is the difference between these two levels that is important to use here, and that difference is the ZPD. ZPD is ''the distance between actual developmental level as determined by independent problem solving and the level of potential development as determined through problem solving under adult guidance or in collaboration with more capable peers" (Vygotsky, 1978, p. 86). The upper limit of the ZPD today becomes the actual developmental level of tomorrow. The fundamental principles of
222
CHAPTER 12
ZPD are the basis for the approaches taken by Feuerstein, Haywood, Campione, and others to cognitive assessment and cognitive training though some key differences exist (see Reynolds, 1986, for a brief review). In neuropsychological work in particular, it can be of help to apply this principle in rehabilitation efforts and the K-ABC is structured in such a way as to give the best available assessment of the ZPD. Typically, the ZPD is determined by contrasting scores on the same test under standardized versus nonstandardized limit-testing procedures. The requirement of initial testing under standardized procedures confounds the latter. The use of two different tests is confounded by differences in normative samples and other technical differences between the scales (e.g., see Reynolds, 1984, 1985, for discussions of these problems). The structure of the K-ABC offers a good solution to these problems. The first developmental level, independent problem solving, can be determined from administration of the K-ABC Achievement scale or preferably from the more restricted Verbal Intelligence scale described in the recent text, Clinical and Research Applications of the K-ABC (Kamphaus & Reynolds, 1987), the latter being more appropriate in this instance because it does invoke considerable reasoning skills and is so much akin to Cattell's concept of crystallized intelligence. The second level of development can be ascertained perhaps more accurately with the K-ABC Mental Processing scales than with any other currently available technique, but only through the use of a nonstandard administration. To assess the second developmental level, administer the K-ABC mental processing subtests using the following procedures: I. Begin all children with the sample item just as directed for the standardized administration. 2. Use the teaching item of the K-ABC as instructed with the following exception. If the child determines the solution to the problem with help or simple repetition, then give credit for successful completion of the item. If you must give the child the correct answer rather than lead him or her to develop a correct response, do not assign passing credit. Under these conditions it is permissible to give strategies and practice trials on subtests that are heavily memory dependent such as Hand Movements or Number Recall. 3. After the sample and teaching items, continue to use the teaching procedures when-
ever a child misses an item when administered under standardized conditions. Give credit for the item as above, i.e., whenever you can lead the child to the correct response, but do not assign credit when you must demonstrate or recite the correct response. 4. Continue administration of each subtest until the child fails to obtain credit on two consecutive items (a ceiling we have estimated to be appropriate based on the growth curves of the various mental processing subtests and the length of time available for testing) when assistance has been provided by the examiner. 5. Compute the raw and scaled scores for the mental processing subtests and the MPC just as though a standardized administration had been conducted. This procedure obviously does not allow one to report standard scores for the MPC or other K-ABC mental processing scales. The Verbal Intelligence scale (Kamphaus & Reynolds, 1987) or Achievement scale can be reported as needed, however. The difference between the MPC, as derived above, and the Verbal or Achievement scale approximates the child's ZPD, and is thus easily quantified for research purposes. On a more practical level, it provides relatively clear guidance in developing short-term and intermediate goals for cognitive rehabilitation programs. The next step in the developmental process is revealed through assessment of the ZPD, the near point in development. In cognitive retraining it is certainly important to know which skills are most likely to be responsive to rehabilitation efforts at specific points in the retraining process. Again the ZPD provides clues. Individual subtests as well as subscales of the MPC can be contrasted with the regularly obtained Verbal Intelligence Composite or Achievement scale score to ascertain areas needing greater attention or even in the evaluative process to determine what progress has been made and in what areas. Given the uneven nature of progress in development and cognitive rehabilitation, the latter can be useful in determining where next to focus rehabilitative efforts. The assessment can be repeated periodically as well. By this conceptualization, we are constantly seeking to move children forward to a level that is within their reach by constantly reassessing the child's reach so that we continue the "match" (a Ia Hunt, 1961) between development and learning. This can be a useful guide for learning-disabled chil-
K-ABC
dren and mentally retarded children as well and is certainly not restricted to children with neuropsychological problems, and is an approach that seems, at least at this stage, similar to approaches proposed by Haywood and Switzky (1986).
Implications for Educational Rehabilitation One of the goals for the K -ABC was to develop a children's intelligence test that yields scores that can provide guidance to educational interventions (Kaufman & Kaufman, 1983b). Chapter 7 of the K-ABC Interpretive Manual provides a framework for educational intervention. The Kaufmans support a strength model of remediation as opposed to deficit-centered ability training models (approaches focusing on the remediation of underlying cognitive processing deficiencies rather than the specific behavioral deficit), which have permeated much of past and present special education practice. Findings in neurology, genetics, and related areas have repeatedly suggested major limitations of the deficit model (e.g., Adams & Victor, 1977; Hartlage, 1975; Hartlage & Givens, 1981; Hartlage & Hartlage, 1973a,b, 1978). Viewed from contemporary neuropsychological models, the deficit approach to remediation is doomed to failure, as it takes damaged or dysfunctional areas of the brain and focuses training specifically on these areas. Not only does knowledge of neurology predict failure for such efforts (Kolb & Whishaw, 1985), evaluations of tht;!ltl approaches have found them to be quite ineffectual in the remediation or learning protJiems (e.g., Glass & Robbins, 1967; Levine, Brooks, & Shonkoff, 1980). One need not embrace localizationist approaches to the diagnosis and descriptive etiology of learning problems, however, in order to employ the neuropsychological model. Most current neuropsychological models do not subscribe to strict localizationist approaches, but rather employ dynamic localization concepts similar to that of Luria ( 1966, 1970). The paramount discovery thus far, however, is that deficit approaches to remediation do not seem to be very effective. Such approaches have been criticized as being potentially harmful to the child as well (Hartlage & Reynolds, 1981). A more meaningful, efficacious approach to a child's learning problems is provided by adopting a strength model of remediation. The strength model is based on abilities that are sufficiently intact so as to subserve the successful accomplishment of the steps
223
in the educational program, so that the interface between cognitive strengths (rather than weaknesses) determined from the assessment and the intervention strategy is the cornerstone of meaningfulness for the entire diagnostic-interventi on process. Placed in the language of Luria's neuropsychological model of intelligence, it is necessary to locate an intact complex functional system capable of taking over and moderating the learning process needed to acquire the academic skills in question. The K-ABC philosophy is reflected in the attitude that the best remedial program for a child who cannot read is to teach the child to read, but to do so using methods and materials optimally related to the child's best information processing skills. The focus is clearly on direct instruction in the child's area of academic deficit allowing children to exploit their perference for processing in a particular way. The structure of the K-ABC provides theoretical guidance to this admittedly muddy area of educational research and practice, guidance that is sorely needed (see Reynolds, 1981b), and guidance that is focused on instruction, not peripheral activities. The K-ABC provides a clear model for using neuropsychological data and theories to make inferences regarding important aptitudes for the individual Ieamer. A model for matching neuropsychological aptitudes to treatment approaches that is nicely complemented by the K-ABC has been presented by Hartlage and Telzrow (1983). Their model teaches ''circumvention'' of dysfunctional areas of the brain to develop compensatory (not remedial) skills and then to capitalize on the child's strengths. Hartlage and Telzrow's approach is very much in line with thE K-ABC phila~mphy and may be useful pllfticularly to the neuropsychologist in designing rehabilitation approaches from K-ABC results, thus expanding the potential utility of the K-ABC in clinical application in neuropsychology.
Summary We have much to learn about the neurological substrates of the K-ABC. Yet for a scale so young, there is a great deal of support for the continued use and exploration of the K-ABC as a tool in the psychological and educational assessment of children. Very much remains to be done. As the years progress, we are confident that the K-ABC will continue to show evidence of utility in helping neuropsychologists gain a better picture of children's brain-behavior relationships, leading to improved rehabilitation pro-
224
CHAPTER 12
grams. We must, of course, keep in mind the tenuous
Gazzaniga, M. S. (1975). Recent research on hemispheric lateralization of the human brain: Review of the split brain. UCLA Educator, 17, 9-12. Wilkening, this volume), the age group addressed by the K-ABC, but also one of the most perplexing and Glass, G. F., & Robbins, M.P. (1967). A critique of experiments on the role of neurological organization in reading perfordemanding age groups of all those encountered by the mance. Reading Research Quarterly, 3, 5-52. clinical neuropsychologist. Golden, C. J. (1978). Diagnosis and rehabilitation in clinical neuropsychology. Springfield, IL: Thomas. Goldstein, K. (1948). After effects of bnun injuries in wars: Their evaluation and treatment. New York: Grune & Stratton. Guyer, B. L., &Friedman, M.P. (1975). Hemispheric processing Adams, K. (1985). Review of the Lucia-Nebraska Neuroand cognitive styles in learning disabled and normal children. psychological Battery. In J. V. Mitchell (Ed.), Ninth mental Child Development, 46, 658-668. measurements yearbook. Lincoln, NE: Buros Institute of Hamad, S., Doty, R. W., Goldstein, L., Jaynes, J., & Mental Measurements. Krauthamer, G. (Eds.). (1977). Loteralization in the nervous Adams, R. D., & Victor, M. (1977). Principles of neurology. system. New York: Academic Press. New York: McGraw-Hill. Hartlage, L. C. (1975). Neuropsychological approaches to preAnastasi, A. (1982). Psychological testing (5th ed.). New York: dicting outcome of remedial educational strategies for learnMacmillan. ing disabled children. Pediatric Psychology, 3, 23-28. Ayers, A. J. (1974). Sensory integration and learning disorders. Hartlage, L. C. (1982). Neuropsychological assessment techLos Angeles: Western Psychological Services. niques. In C. R. Reynolds & T. B. Gutkin (Eds.), The handBever, T. G. (1975). Cerebral asymmetries in humans are due to book of school psychology. New York: Wiley. the differentiation of two incompatible processes: Holistic Hartlage, L. C., & Reynolds, C. R. (1981). Neuropsychological and analytic. In D. Aaronson & R. Rieber (Eds.), Developassessment and the individualization of instruction. In G. W. mental psycholinguistics and communication disorders. New Hynd & J. E. Obrzut (Eds.), Neuropsychological assessment York: New York Academy of Sciences. of the school-aged child: 1ssues and procedures. New York: Bogen, I. E. (1969). Theothersideofthe brain: Parts I, II, and III. Grune & Stratton. Bulletin of the Los Angeles Neurological Society, 34, 73- Hartlage, L. C., & Telzrow, C. F. (1983). The neuropsychologi105, 135-162, 191-203. cal basis of educational intervention. Journal of Learning Bogen, J. E., Dezure, R., Tenouten, W., &Marsh, I. (1972). The Disabilities, 16, 521-528. other side of the brain. IV. Bulletin of the Los Angeles NeuHartlage, P. L., & Givens, T. S. (1981). Common neurological rological Society, 37, 49-61. problems of school age children. In C. R. Reynolds & T. B. Bradley, P. E., Battin, R. R., & Sutter, E. G. (1979). Effects of Gutkin (Eds.), The handbook of school psychology. New individual diagnosis and remediation for treatment of learnYork: Wiley. ing disabilities. Clinical Neuropsychology, 1, 25-31. Hartlage, P. L., & Hartlage, L. C. (1973a). ComparisonofhyperBrown, J. W., & Hecaen, H. (1976). Lateralization and language lexic and dyslexic children. Neurology, 23, 436-437. presentation. Neurology, 26, 183-189. Hartlage, P. L., & Hartlage, L. C. (1973b). Dermatog/yphic Cummins, J., & Das, J.P. (1977). Cognitive processing and readmarkers in dyslexia. Paper presented at the meeting of the ing difficulties: A framework for research. Alberta Journal of Child Neurology Society, Nashville, Tennessee. Educational Research, 23, 245-256. Hartlage, P. L., & Hartlage, L. C. (1978). Clinical consultation to Das, J. P., Kirby, J. R., & Jarman, R. F. (1979). Simultaneous pediatric neurology and developmental pediatrics. Journal if and successive cognitive processes. New York: Academic Clinical Child Psychology, 12, 52-53. Haywood, H. C., & Switzky, H. N. (1986). The malleability of Press. Dean, R. S. (1984). Functionallateralization of the brain. Journal intelligence: Cognitive process as a function of polygenicof Special Education, 8, 239-256. experiential interaction. School Psychology Review, 15, Dietzen, S. R. (1986). Hemispheric specialization/or verbal se245-255. quential and non-verbal simultaneous information process- Hooper, S. R., & Hynd, G. W. (1985). Differential diagnosis of ing styles of low income 3 to 5 year olds. Doctoral dissertasubtypes of developmental dyslexia with the Kaufman Astion, Washington State University. sessment Battery for Children (K-ABC). Journal of Clinical Gaddes, W. H. (1981). An examination of the validity of neuroChild Psychology, 14, 145-152. psychological knowledge in educational diagnosis and re- Hunt, J. M. (1961). Intelligence and experience. New York: mediation. In G. W. Hynd & J. E. Obrzut (Eds.), NeuropsyRonald Press. chological assessment of the school-aged child: Issues and Kamphaus, R. W., Kaufman, A. S., & Kaufman, N. L. (1982, procedures. New York: Grune & Stratton. August). A cross-validation study of sequential-simulGazzaniga, M.S. (1970). The bisected brain. Englewood Cliffs, taneous processing at ages 2l-12l using the Kaufman Assessment Banery for Children. Paper presented at the meetNJ: Prentice-Hall. Gazzaniga, M. S. (1974). Cerebral dominance viewed as a deciing of the American Psychological Association, sion system. In S. Dimond & I. Beaumont (Eds.), HemiWashington, DC. sphere functions in the human brain. London: Halstead Press. Kamphaus, R. W., & Reynolds, C. R. (1987). Clinical andre-
nature of localization in the 2! to 12! group (see
References
K·ABC
search applications of the K-ABC. Circle Pines, MN: American Guidance Service. Kaufman, A. S. (1975). Factor analysis of the WISC-R at eleven age levels between 6l and 16! years. Journal of Consulting and Clinical Psychology. 43. 145-147. Kaufman, A. S. ( 1976). Verbal-performance IQ discrepancies on the WISC-R. Journal ofConsulcing and Clinical Psychology, 44. 739-744. Kaufman, A. S. (1978). The importance of basic concepts in the individual assessment of preschool children. Journal of School Psychology, 16. 207-211. Kaufman, A. S. (l979a).1ntel/igent testing with the WISC-R. New York: Wiley-Interscience. Kaufman, A. S. (1979b). Cerebral specialization and intelligence testing. Journal ofResearch and Development in Education, 12, %-197. Kaufman, A. S. (1983). Some questions and answers about the Kaufman Assessment Battery for Children (K-ABC). Journal of Psychoeducational Assessment, 1, 205-218. Kaufman, A. S., & Kaufman, N. L. (1977). Clinical evaluation of young children with the McCarthy scales. New York: Grune & Stratton. Kaufman, A. S., & Kaufman, N. L. (1983a). Kaufman Assessment Baltery for Childrn~ (K-ABC) administration and scoring manual. Circle Pines, MN: American Guidance Service. Kaufman, A. S., & Kaufman, N. L. (1983b). K-ABC interpretive manual. Circle Pines, MN: American Guidance Service. Kaufman, A. S., Kamphaus, R. W., & Kaufman, N. L. (1985). The Kaufman Assessment Battery for Children (K-ABC). In C. S. Newmark (Ed.), Major Psychological Assessment Instruments. Newton, MA: Allyn & Bacon. Kaufman, A. S., Kaufman, N. L., Kamphaus, R. W., &Naglieri, J. A. (1982). Sequential and simultaneous factors at ages 312!: Developmental changes in neuropsychological dimensions. Clinical Neuropsychology, 4, 74-81. Kaufman, A. S., O'Neal, M. R., Avant, A. H., & Long, S. W. ( 1987). Introduction to the Kaufman Assessment Battery for Children (K-ABC) for pediatric neuroclinicians. Journal of Child Neurology, 2, 3-16. Keith, T. z. (1985). Questioning of the K-ABC: What does it measure? School Psychology Review, 14, 9-20. Kinsbourne, M. (1975). Cerebral dominance,leaming, and cognition. In H. R. Myklebust (Ed.), Progress in learning disabilities. New York: Grune & Stratton.. Kinsboume, M. (1978). Biological determinants of functional bisymmetry and asymmetry. In M. Kinsbourne (Ed.), Asymmetrical function of the brain. London: Cambridge University Press. Knight, R. M., & Bakker, D. J. (1980). Treatment of hyperactive and learning disordered children: Cu"ent research. Baltimore: University Park Press. Kohs, S. C. (1927). Intelligence measurement. New York: Macmillan. Kolb, B., & Whishaw, I. Q. (1985). Fundamentals of human neuropsychology (2nd ed.). San Francisco: Freeman. Leark, R. A., Snyder, T., Grove, T., & Golden, C. I. (1983, August). Comparison of the K-ABC and standardized neuropsychological batteries: Preliminary results. Paper pre-
225
sented at the annual meeting of the American Psychological Association, Anaheim, CA. Levine, M.D., Brooks, R., & Shonkoff, I. P. (1980). A pediatric approach to learning disorders. New York: Wiley. Luria. A. R. (1966). Human brain and psychological processes. New York: Harper & Row. Luria, A. R. (1970). The functionill organization of the bmin. Scientific American. 18, 266-268. Majovski. L. V. (1984). The K-ABC: Theory and applications for child neuropsychologicill assessment and research. The Journal of Special Education, 18, 266-268. Morris, I. M., & Bigler, E. (1985, January). An investigation of
the Kaufman Assessment Battery for Children (K-ABC) with neurologically impaired children. Paper presented at the annual meeting of the International Neuropsychological Society, San Diego. Naglieri, J. A., Kaufman, A. S., Kaufman, N. L., & Kamphaus, R. W. (1981). Cross-validation of Das' simultaneous and successive processes with novel tasks. Alberta Journal of Educational Research, 27, 264-271. Ornstein, R., Johnstone, I., Herron, I., & Swencionis, C. (1980). Differential right hemisphere engagement in visuospatial tasks. Neuropsychologia, 18, 49-64. Pribram, K. (1971). Language ofthe brain. Englewood Cliffs, NJ: Prentice-Hall. Reitan, R., & Davison, L.A. (1974). Clinical neuropsychology: Current status and applications. Washington, DC: Winston. Reynolds, C. R. (1980, July). The neuropsychological basis of intelligence and a reconceptualization of dominance. Invited address to the Utah State University Conference on Brain Research and Teaching, Logan. Reynolds, C. R. (1981a). The fallacy of "two years below grade level for age'' as a diagnostic criterion for reading disorders. Journal of School Psychology, 19, 350-358. Reynolds, C. R. ( 1981 b). Neuropsychological assessment and the habilitation of learning: Considerations in the search for the aptitude x treatment interaction. School Psychology Review, 343-349. Reynolds, C. R. (1982), The importance of norms and other psychometric concepts to assessment in clinical neuropsychology. In R.N. Malatesha & L. C. Hartlage (Eds.), Neuropsychology and cognition (Vol. II). The Hague: Nijhoff. Reynolds, C. R. (1984). Critical measurement issues in learning disabilities. Journal of Special Education, 18, 451-476. Reynolds, C. R. (1986). Transactional models of intellectual development, yes. Deficit models of process remediation, no.
School Psychology Review, 15, 256-260. Reynolds, C. R. (1987). Intelligent testing. In C. R. Reynolds & L. Mann (Eds.), Encyclopedia of special education. New York: Wiley-Interscience. Reynolds, C. R., & Clark, J. H. (Eds.). (1983). Assessment and
programming for children with low incidence handicaps. New York: Plenum Press. Reynolds, C. R., & Clark, J. H. (1985). Profile analysis of standardized intelligence test performance of very low functioning individuals. Journal of School Psychology, 23, 277283. Rourke, B. P., & Orr, R. R. ( 1977). Prediction of the reading and spelling performances of normal and retarded readers: A four-
226
CHAPTER 12
age follow-up. JounuJI of Abnormal Child Psychology, 5, 9-20. Schwartz, G. E., Davidson, R. J., & Mear, F. (1975). Right hemisphere lateralization for emotion in the human brain: Interactions with cognition. Science, 190, 286-288. Sechenov, I. (1965). Reflexes of the brain. Cambridge, MA: MIT Press. (Original work published in 1863.) Segalowitz, S. J., & Gruber, F. A. (Eds.). (1977). Language development and neurological theory. New York: Academic Press. Shapiro, E. G., & Dolan, N. (1985, October). Neurologicalfindings and the KmiftrumAssessment BaJteryfor Children. Paper presented at the National Association ofNeuropsychologists, Philadelphia. Shure, G. H., &Halstead, W. C. (1959). Cerebrallateralizationof individual processes. Psychological Monographs: General andApplied, 72, (12). Snyder, T. J., Leark, R. A., Golden, C. J., Grove, T., & Allison, R. (1983, March). CorrelationsoftheK-ABC, WJSC-R. and LuriD-Nebraska Children's Batteryfor exceptioMI children. Paper presented at the meeting of the National Association of School Psychologists, Detroit. Sperry, R. W. (1968). Hemisphere decoMection and unity in conscious awareness. American Psychologist, 23, 723-733. Sperry, R. W. (1974). Lateral specialization in the surgically separated hemispheres. In F. 0. Schmitt & F. G. Worden (Eds.), The neurosciences: Third study program. Cambridge, MA: MIT Press. Sperry, R. W., Gazzaniga, M. S., & Bogen, J. E. (1969). Interhemispheric relationships: The neocortical commissures: Syndromes of hemispheric discoMection. In P. Vinken & G.
W. Beuyn (Eds.), Handbook of clinical neurology. New York: Wiley-Interscience. Tarnopol, L., & Tarnopol, M. (1977). Introduction to neuropsychology. In L. Tarnopol & M. Tarnopol (Eds.), Brain/unction and reading disabilities. Baltimore: University Park Press. Telzrow, C. F., Century, E., Harris, B., & Redmond, C. (1985, April). Relationship between neuropsychological processing models and dyslexia subtypes. Paper presented at the National Association of School Psychologists, Las Vegas. Telzrow, C., Redmond, C., & Zimmerman, B. (1984, October). Dyslexic subtypes: A comparison of the Bannatyne, Bader, and Kaufman models. Paper presented at the aMual meeting of the National Academy of Neuropsychologists, San Diego. Tomlinson-Keasey, C., & Clarkson-Smith, L. (1980, February). What develops in hemispheric specialization? Paper presented at the meeting of the International Neuropsychological Society, San Francisco. Vygotsky, L. S. (1978). Mind in society (M. Cole, V. JohnSteiner, S. Scribner, & E. Souberman, Eds.). Cambridge, MA: Harvard University Press. Wada, J., Clarke, R., & Hamm, A. (1975). Cerebral hemisphere asymmetry in humans. Archives ofNeurology. 37, 234-246. Wesman, J. (1968). Intelligent testing. American Psychologist, 23. 267-274. Willson, V. L.,Reynolds,C. R.,Chatman,S., &Kaufman, A. S. (1985). Conftrmatory analysis of sequential and simultaneous processing factors for children ages 2! to 12! years. Journal of School Psychology, 23, 216-269. Wittrock, M. C. (1980). The brain and psychology. New York: Academic Press.
13 Neuropsychological Applications of Common Educational and Psychological Tests CATHY F. TELZROW
There are two major orientations to neuropsychological assessment. The first, best exemplified by Reitan (1979), Rourke (1981; Rourke, Bakker, Fisk, & Strang, 1983), and the Luria-Nebraska school (Golden, 1981; Gustavson et al., 1984), employs a standardized battery of tasks designed to identify brain impairment. Typically, an invariant battery of identical tasks is utilized for all subjects regardless of the presenting problem or the history of the child. The second approach to neuropsychological assessment of children favors the use of a flexible combination of traditional psychological and educational tests. The composition of this battery varies depending upon a number of child variables, including the age, history, functioning level, and presenting problem of the particular child. Although the neuropsychological battery approach to assessment is important historically because of its major contributions to the understanding of brain-behavior relationships, its application with children may not be appropriate in all instances. Criticisms of the standardized battery approaches have focused on the time and cost requirements (Goldstein, 1984; Hartlage & Telzrow, 1986; Slomka & Tarter, 1984; Sutter & Battin, 1984), the redundancy of procedures in some cases, and the cursory appraisal of functions in others (Goldstein, 1984; Hartlage & Telzrow, 1986; Slomka & Tarter, 1984; Taylor, Fletcher, & Satz, 1984). Perhaps of greater import, however, is the theoretical orientation such an approach presumes and its sometimes questionable validity with pediatric populations. Fletcher CATHY F. TELZROW • Cuyahoga Special Education Ser-
vice Center, Maple Heights, Ohio 44137.
(I 985) made a distinction between neuropsychologists who hypothesize from a central nervous system (CNS) impairment orientation to children's behavior and those who reason the other way, developing hypotheses about children's neuropsychological functioning from careful, systematic behavioral observations. A flexible neuropsychological assessment approach that employs traditional educational and psychological tests appears more consistent with the latter orientation, which Fletcher and his colleagues (1985; Taylor et al., 1984) argue is better suited to the unique developmental characteristics of children. To illustrate the differences between these two orientations, consider that many of the tasks on standardized batteries are not appropriate for young children with significant impairment or those with specific sensory deficits. Administration of a Halstead-Reitan or a Luria-Nebraska Children's Battery thus results in much missing data, and it may not be possible to derive meaningful information about the neuropsychological abilities of such children from these techniques. In contrast, use of a flexible battery of developmental and psychological tests permits the clinician to select tasks appropriate to the functioning levels and response limitations of the child, resulting in a more complete assessment of neuropsychological strengths and weaknesses. In addition, this procedure provides a description of the neuropsychological assets of children, rather than emphasizing the diagnosis and localization of brain impairment, for which standardized batteries have been noted. As a result of this delineation of residual neuropsychological strengths, the flexible battery approach translates more directly into educational and 227
228
CHAPTER 13
vocational interventions, a major goal of neuropsychological assessment (Dean, 1986; Klesges, Fisher, Pheley, Boschee, & Vasey, 1984; Telzrow, 1985a). Finally, because this orientation employs traditional educational and psychological tests that are more familiar to school personnel, the flexible battery approach to neuropsychological assessment is more directly applicable to school settings. Because schools provide valuable rehabilitation opportunities for children with neuropsychological impairment, the ease with which assessment data can be transferred into these settings is critical. This chapter describes the standardized test components of a flexible neuropsychological assessment approach, including major variables to consider in the selection and interpretation of traditional measures. The scope of this chapter necessitates limiting discussion to psychological and educational tests. However, in clinical practice, the author typically utilizes other sources of data, including a detailed history, observation, and selected neuropsychological tasks, such as finger tapping and measures of sensory-perceptual functioning. For a more complete treatment of the integration of these multiple data sources, the reader is referred to Hartlage and Telzrow (1986).
Major Areas of Function and Associated Tests The major neuropsychologically relevant behaviors in children that can be tapped by traditional educational and psychological tests include cognitive ability (with additional analysis of information processing patterns), academic achievement, language and motor skills, and social-emotional behavior. A description of each of these functions, together with its relevance for neuropsychological assessment of children, follows in this section. Included within the discussion of each area are specific measures that have demonstrated utility for assessing such functions. Attempts were made to include only instruments with sound psychometric properties that are in fairly general use in educational and clinical practice.
Cognitive Functioning Cognitive functioning encompasses intelligence and more specific components of information processing, such as memory, visuospatial processing, and linguistic ability. Assessment of cognitive functioning includes determination of overall level of
functioning relative to the normative population, as well as analysis of intraindividual strengths and weaknesses in cognitive processing. A general level of performance score is important as a means of making predictions about expected achievement levels for children with developmental problems, and in determining loss of function for children with acquired impairment. It also serves as an important baseline for the interpretation of other data. However, this type of assessment alone is not sufficient, for a general cognitive ability score may be misleading, in that unique neuropsychological deficits may contribute to a decrement in overall performance (refer to the Stanford-Binet scores of R.G. in Figure 3 to illustrate this point). For this reason, analysis of individual cognitive processing strengths and weaknesses is critical. In performing such analyses, Kaufman (1979), Reynolds (1981), and others have stressed the importance of ipsative or intraindividual comparisons, for instance, comparing an individual's pattern of subtest or scaled scores with his or her own mean performance. Conducting profile analyses with this degree of precision minimizes such interpretive errors as assuming chance variability reflects significant processing differences. Various ability scales conceptualize cognitive processes differently. Some of these scales were developed from theoretical models of intelligence (e.g., Stanford-Binet Intelligence Scale, 4th Ed.; Thorndike, Hagen, & Sattler, 1986). Others were formulated from theoretical constructs, with empirical (i.e., factor analytic) rationale for the assignment of tasks to specific scales (e.g., Kaufman Assessment Battery for Children; Kaufman & Kaufman, 1983). The following cognitive ability scales have been found to be of value in conducting neuropsychological evaluations of children.
Bayley Scales of Infant Development (Bayley, 1969)
Standardized scores for mental and motor ability can be derived from the Bayley Scales for infants and toddlers up to 30 months old. Age equivalent scores can be computed for older children functioning at these developmental levels. The Bayley is appropriate for children with a wide variety of developmental problems. However, because of its heavy visual demands, it is not considered an appropriate instrument for children with visual impairments. As is true with most infant scales, there is a heavy reliance on motor skills.
APPLICATIONS OF EDUCATIONAL AND PSYCHOLOGICAL TESTS
Use of the Bayley in its traditional format provides a comparison of children's mental and motor performance. Although this is of some value for young children with significant motor impairment (e.g., hypotonic children or those with cerebral palsy), several of the mental scale items require motor responses; hence, the separation is not pure. Reuter, Stancin, and Craig (1981) developed an extremely useful scoring adaptation for the Bayley that permits the derivation of developmental age scores for five domains: cognitive, language, social, fine motor, and gross motor. Use of the Kent Scoring Adaptation of the Bayley Scales of Infant Development provides a systematic analysis of individual strengths and weaknesses. It has been found particularly useful by the author in differentiating among developmental language disorders, autistic syndromes with disordered language, and mental retardation in young children. Children with severe expressive language syndrome, as described by Rapin and Allen (1983), tend to have scores on the language domain depressed relative to the other domains. Those with childhood autism tend to have unique depressions in the language and social domains, as well as deficits on several of the cognitive items, although they may do quite well on the form boards and other tasks with heavy visuospatial loading. Mentally retarded children, though deficient in all areas, are not uniquely low on language and social domains; in fact, these may represent relative strengths for this population.
Kaufman Assessment Battery for Children (K-ABC; Kaufman & Kaufman, 1983) The K-ABC is a measure of intellectual ability and achievement. The cognitive processing portion of the K-ABC will be discussed in this section, with the achievement scale described subsequently. The K-ABC was designed from a theoretical model integrating data and hypotheses from cognitive psychology, cerebral specialization research, and clinical neuropsychology (Kamphaus & Reynolds, 1984). Drawing heavily from the Luria school and the work of followers of this model, Kaufman and Kaufman attempted to develop a measure of intellectual ability that divides information processing into sequential and simultaneous types. The well-standardized scale, released with an impressive number of validity studies, nevertheless encountered sharp criticism from those who found fault with the authors' theoretical model of intelligence (Goetz & Hall, 1984; Sternberg, 1984) and with the empirical support for the Sequential and Simultaneous processing scales
229
(Bracken, 1985; Keith, 1985). Other critiques, particularly those conducted by persons with clinical orientations (Kaufman, 1984), have been somewhat more optimistic about the scale's contribution to neuropsychological assessment of children (e.g., German, 1983; Majovski, 1984; Telzrow, 1984). After nearly three years of use with children with a wide variety of developmental and acquired problems, it is the author's opinion that the K-ABC is of mixed value in the neuropsychological assessment of children. For selected children, the hypothetical profile (e.g., Sequential > Simultaneous, or vice versa) predicted from history or other test data proves true. For other children, however, the anticipated pattern is not demonstrated (e.g., see case of J.G. in Figure 2), and thus far common elements or predictor variables in populations for whom the model works and those for whom it does not have not been identified fully. Investigations of the K-ABC and other psychometric tests for reading-disabled children have revealed predicted patterns, suggesting some construct validity for the K-ABC Mental Processing scale. For example, dyslexic children who demonstrate a V < P profile on the Wechsler Intelligence Scale for Children-Revised (WISC-R; Wechsler, 1974) have been reported to have a Sequential < Simultaneous split on the K-ABC (Stoiber, Bracken, & Gissal, 1983; Telzrow, Century, Harris, & Redmond, 1985). Comparison of reading-disabled children's KABC scores with their cognitive factor scores on the WISC-R 1 derived from recategorization of subtests using the procedures outlined by Bannatyne (1974), also demonstrated expected similarities (e.g., WISCR sequential factor based on Arithmetic, Digit Span, and Coding subtests was similar to the K-ABC Sequential score) (Telzrow et al., 1985). Furthermore, relationships between the K-ABC and subtypes of reading disorders have been demonstrated, again in a direction predicted by the K-ABC's theoretical model (i.e., dysphonetic dyslexic has Simultaneous >Sequential profile) (Telzrow et al., 1985). There are a number of cases, however, in which the predicted profile on the K-ABC is not demonstrated. In a study of left- and right-hemiplegic children, Lewandowski and DeRienzo (1985) reported the presence of the anticipated Sequential > Simultaneous pattern for left-hemiplegic children, but not the corollary for the right-hemiplegic group. Hooper and Hynd (1985) analyzed the K-ABC's relationship to the Boder Test of Reading-Spelling Patterns (Boder & Janico, 1982), and found the K-ABC differentiated dyslexic from normal readers but did not distinguish among dyslexic subtypes. In the author's
230
CHAPTER 13
experience, children who show significant V > P patterns on the WISC-R frequently do not reflect comparable Sequential > Spatial profiles on the KABC. In addition, youngsters for whom there is other evidence of substantial language impairment often may not have a Mental Processing Composite (MPC) that is significantly lower than the Nonverbal Score on the K-ABC. There appear to be a couple of possible explanations for these inconsistencies. First, decisions about the assignment of the KABC subtests to the Sequential and Simultaneous scales, respectively, were made largely on the basis of factor analysis. These analyses sometimes revealed that specific subtests (e.g., Hand Movements, Matrix Analogies, Photo Series) were not factorially pure throughout the age span of the K-ABC. In fact, the authors pointed out (Kaufman & Kaufman, 1983) that Hand Movements, assigned to the Sequential scale, actually has substantial loading on the Simultaneous factor at certain ages as well, a finding that may describe developmental variability in cognitive processing (Kaufman, Kaufman, Kamphaus, & Naglieri, 1982). As a result of this lack of factor discreteness, derived scores sometimes are a little difficult to interpret. The overlap between the K-ABC Sequential and Simultaneous scales can be demonstrated by examining the learning-disabled (LD) profile reported for this instrument. The K-ABC manual describes an LD profile as one where the Simultaneous scale is slightly higher than the Sequential scale, where Gestalt Closure represents the highest subtest score, where Sequential scores, in general, are lower relative to the other scores, and where Riddles, a subtest on the Achievement scale that has the highest simultaneous loading, is the highest achievement subtest. It is the author's experience that this profile often does prove to be true for LD children. However, one observation that has been made concerns the distribution of subtest scores throughout the Simultaneous scale. There often seems to be a dichotomy between subtests with high perceptual content (Gestalt Closure, Triangles, and Spatial Memory), and those with greater sequential processing potential (Matrix Analogies, Photo Series). For many LD children, the latter tend to be lower than the items with more visuospatial demands. As a result, the total Simultaneous score is depressed, resulting in only a slight difference, or perhaps none at all, between the two mental processing scales. This unusual distribution of subtest scores also can be demonstrated by examining profiles of children with spina bifida who have had ventriculo-peritoneal shunts inserted. As a rule, children with by-
drocephalus demonstrate a visuospatial weakness (Spiegler,-Baron, Hammock, & McCullough, 1985), presumably associated with right lateralized shunt insertion (Batshaw & Perret, 1981; Hartlage & Telzrow, 1986). The case ofD.S., shown in Figure I, illustrates the dichotomy of scores between two factors, one apparently a linguistic, sequential factor, and a second visuospatial factor. D. S. , a young boy with a shunt, demonstrates perceptual deficits on the Gestalt Closure, Triangles, and Spatial Memory subtests. However, his superior verbal reasoning skills produced a score well above average on the Matrix Analogies subtest. Similarly, his mastery of oral language skills resulted in superior scores on the Expressive Vocabulary and Riddles tasks, although his visuospatial deficits have made early learning of letter and number symbols more difficult. These examples were used to illustrate the fact that the Sequential and Simultaneous scales do not appear to be pure factors. The Simultaneous scale, in particular, seems to be an amalgam of perceptual processing, more equivalent to the WISC-R Performance scale, or, particularly, Bannatyne' s spatial factor from this scale (Picture Completion, Block Design, and Object Assembly) and subtests with more sequential loading (e.g., Matrix Analogies, Photo Series). Other hypotheses for the sometimes inconsistent pattern of scores on the K-ABC relative to other common cognitive scales have been proposed. Keith (1985) questioned the K-ABC's sequential versus simultaneous constructs, and suggested the two identiKaufman Assessment Battery for Children Sequential
ss
Simultaneous
Hand Movements Number Recall Word Order
11 15 9
Gestalt Closure Triangles Matrix Analogies Spatial Memory
Mental Processing Composite
Faces & Places Arithmetic Riddles Reading/Decoding
6 6 12
4
Simultaneous Processing = 80 ± 7
Sequential Processing = 110 ± 7
Subtest
ss
= 92 ± 6
ss 109 ± 11
87 ± 8 124 ± 10 93 ± 6
FIGURE I. D.S.: Spina bifida with ventriculo-peritoneal shunt. CA. 5-3.
APPLICATIONS OF EDUCATIONAL AND PSYCHOLOGICAL TESTS
fied factors (with slightly different subtest loadings) perhaps more accurately are identified as verbal memory and nonverbal reasoning scales. Bracken (1985) suggested the Kaufmans' "sequential" tasks lack cognitive complexity and instead are tasks of short-term memory. Obrzut, Obrzut, and Shaw's (1984) finding that the K-ABC Sequential score did not correlate at significant levels with the WISC-R Verbal IQ suggests these two scales are measuring somewhat discrete skills. In the author's experience, the subtests on the K-ABC represent extremely novel tasks for children. As a result, some youngsters, particularly those who are very young or who come from disadvantaged backgrounds, may not respond well to such unfamiliar tasks (e.g., Hand Movements). Also, there appears to be a heavy short-term memory demand on the K-ABC. Because many children with neuropsychological impairment have memory dysfunctions, this variable may impact children's performance on the scale. . In summary, the K-ABC is a relatively recent scale of cognitive ability that purports to tap sequential and simultaneous processing abilities. The degree to which the scale succeeds has been debated, although the recency of the battery suggests considerably more research needs to be conducted before any definitive conclusions about its applicability with neuropsychologically impaired children can be drawn (see Reynolds, Kamphaus, & Rosenthal, this volume). From clinical experience, the author has found the K-ABC Mental Processing scale to be of mixed value in the assessment of children with a variety of developmental and acquired problems.
Leiter International Performance Scale (Leiter, 1969) The inclusion of the Leiter in this discussion of cognitive measures represents a compromise. Although this instrument does not possess the same degree of psychometric sophistication demonstrated by the other measures described in the chapter, it nevertheless is included because of its unique properties and the author's experience that it can be of important clinical value in the assessment of children with neuropsychological impairment. The Leiter is a nonverbal scale of cognitive ability that requires no language in either the administration or the subject's response. It relies on the use of perceptual tasks, graduating from rather pure perceptual matching tasks at the early ages to items of greater abstract reasoning and concept formation at the upper levels. Throughout the test, extensive demands
231
on visual perceptual skills are made; hence, the test is not considered appropriate for children who may have visual impairments, or those who may have perceptual difficulties as a result of neurological trauma. In the author's experience, the Leiter is particularly useful in the assessment of children who have significant language impairments. This may be true of deaf or hard of hearing children, those with developmentallanguage disorders, or those with atypical developmental syndromes such as autism (Shah & Holmes, 1985). Use of the Leiter in conjunction with other carefully selected psychological tests can assist the clinician in making a differential diagnosis among these types of disorders. In young deaf or hearing-impaired children, the author frequently uses the Leiter together with another age-appropriate verbal measure of cognitive ability (e.g., the Bayley in children below age 2i, the Binet or the Wechsler in older children). By comparing performance across the two measures, it is possible to determine whether the hearing loss represents a discrete impairment, or is one component of more pervasive problems. Children with hearing impairment alone, especially congenitally deaf children, tend to score within normal limits or often above on the Leiter (as the Leiter tends to inflate cognitive scores at the early ages), while scoring much lower on tasks with greater language demands. Developmental performance on other tasks (e.g., motor tasks or items sensitive to nonverbal mental milestones) tends to be age-appropriate, although for preschoolers some self-care skills (e.g. , toileting, dressing) may be slightly delayed because of the communication requirements essential for such skill acquisition. When used in conjunction with verbal scales of cognitive ability, the Leiter is helpful in identifying children with specific communication disorders who may have adequate nonverbal cognitive skills (Elbert & Willis, 1984). As with the hearing-handicapped children described above, this population tends to score within normal limits on the Leiter, while performing quite poorly on any language tasks, such as picture naming. The case of J.G. (Figure 2) illustrates the Leiter and K-ABC scores of a hyperlexic child with a pervasive oral language deficit. Separating children such as J. G. who have specific language disorders from those with more pervasive disorders such as autism may be possible through use of the Leiter, together with such instruments as the K-ABC Achievement scale, and measures of communicative intent and social responsiveness (Telzrow, 1985b). In the author's experience, children whose failure to
232
CHAPTER 13
Kaufman Assessment Battery for Children
ss
Sequential Hand Movements Number Recall Word Order
Sequential Processing = 83 ± 7
7 8 7
Simultaneous
ss
Gestalt Closure Triangles Matrix Analogies Spatial Memory
7 8 9
5
Simultaneous Processing = 81 ± 7
Mental Processing Composite = 80 ± 6 Nonverbal Score = 81 ± 7 Subtest Expressive Vocabulary Faces & Places Arithmetic Riddles Reading/Decoding Reading/Understanding
ss 68 ± 9 74 ± 11 107 ± 8 76 ± 10 144 ± 6 Age equivalent = 7-6
Leiter International Performance Scale CA =5-2 MA =5-3 10 = 102 Soder Test of Reading-Spelling Patterns Reading Quotient= 135 Known Words = 80% Unknown Words = 20% FIGURE 2. J.G.: Severe expressive communication disorder with hyperlexia. CA, 5-2.
develop speech is associated with an expressive syndrome or even auditory-verbal agnosia (Rapin & Allen, 1983) tend to perform consistently on all Leiter items up to their general level of nonverbal cognitive ability. In contrast, autistic children demonstrate a much more variable performance pattern, and tend to fail even early Leiter items requiring social experience (e.g., Picture Completion at Year 3, Genus and Clothing at Year 5). Pure perceptual tasks, however, may be passed by autistic children beyond their chronological age levels. In summary, the Leiter is less than adequate psychometrically. However, its unique task demands make it a useful clinical tool for evaluating the nonverbal cognitive ability of children. When used in conjunction with other measures, it can provide useful information in performing differential diagnoses among mental retardation, specific language disorders, and atypical developmental syndromes such as childhood autism.
McCarthy Scales of Children's Abilities (MSCA; McCarthy, 1972) The MSCA is a measure of cognitive ability designed for use with children aged 2! to 8!. This scale conceptualizes cognitive processing into five scales on the basis of a theoretical model derived from the author's clinical experiences. The Verbal, Perceptual-Performance, and Quantitative scales are discrete in content, and combine to derive a total score or General Cognitive Index (GCI). Two additional scales, Memory and Motor, are obtained from subtests that either are totally separate from or overlap with the other scales. Factor analysis of a portion of the standardization sample supports the existence of a general cognitive factor, Memory, and Motor components across all ages (McCarthy, 1972). Cross-validation studies support the existence of independent Verbal, Perceptual-Performance, and Motor factors as well (Naglieri, Kaufman, & Harrison, 1981). Another study of 300 6- to 8~-year-old children demonstrated support for the general (GCI), Verbal, and Motor factors, but questioned the interpretation of the Memory, Perceptual-Performance, and Quantitative scales for this population (Keith & Bolen, 1980). Other evidence suggests that interpretation of a discrete Quantitative scale may be problematic for preschool children (Kaufman & Kaufman, 1977) and for school children with GCis in the lowest 16th percentile (Naglieri et al., 1981). In the author's experience, the MSCA is most useful in the assessment of mildly impaired children ages 4 and older. Younger children and those with more pervasive handicaps, such as those with significant communication disorders or attention deficits, have not been found to perform optimally on this long, demanding scale. Furthermore, the MSCA may not have sufficient floor to derive valid scores for moderately retarded children (Beckett, Reuter, & Stancin, 1984). For the assessment of suspected learning disabilities, however, or other kinds of mild handicaps, the McCarthy may provide a relevant analysis of strengths and weaknesses, particularly in light of the separate Memory and Motor scales. Early investigations of the MSCA reported significantly lower GCis for LD than for non-LD populations (Gob & Simons, 1980). Furthermore, Gob and Simons (1980) reported an absence of distinctive pattern or scatter characteristics on the MSCA for LD groups, and clinicians should be mindful of the significant scatter demonstrated by populations of normal children (Kaufman, 1976). No significant scatter differences were found between "at-risk" preschoolers and the MSCA standardization sample
APPLICATIONS OF EDUCATIONAL AND PSYCHOLOGICAL TESTS
(Prasse, Siewert, & Ellison, 1983), although a greater proportion of at-risk children showed scatter greater than or equal to 21 points. Prasse et al. (1983) suggested that the MSCA, particularly the Motor Scale, may contribute to the identification of neurologically compromised 4-year-old children.
Stanford-Binet Intelligence Scale-Fourth Edition (Binet; Thorndike et al., /986) The recent revision of the Stanford-Binet was viewed by most as long overdue, given there had been few changes in individual items since Terman and Merrill's 1937 edition (Sattler, 1982). The fourth edition represents a significant departure from previous versions, and, although validity data are just beginning to be accumulated, early impressions are that the latest Binet has the potential for significant contributions to the neuropsychological assessment of children. The fourth edition was developed from a theoretical model relying heavily on g or general cognitive ability. The work of Cattell and others is used to conceptualize intelligence as comprising fluid and crystallized abilities, as well as a memory component. The four major scales on the fourth edition are Verbal Reasoning and Quantitative Reasoning (presumably indices of crystallized ability), Abstract/Visual Reasoning (theoretically a measure of fluid ability), and Short-Term Memory. A composite or partial composite score can be derived from any combination of area scores. Results of preliminary factor analyses conducted during the preparation of the technical manual for the fourth edition indicated the presence of a strong g factor throughout, as well as a verbal reasoning component across all ages. A short-term memory factor was identified at all but preschool ages, although items sensitive to spatial memory (e.g., the Bead Memory subtest) were not correlated with the other subtests on this scale (Memory for Digits, Memory for Sentences, Memory for Objects), which appear to be more sensitive to temporal memory. Evidence for the Quantitative and Abstract/Visual Reasoning scales were inconsistent across the ages analyzed. Additional validity data on the revised Binet are not available at this writing, hence the following impressions should be considered tentative. In limited clinical use the author has found the fourth edition to reveal patterns of neuropsychological strengths and weaknesses that are consistent with data from other, proven measures. Children who demonstrate a V < P
233
split on the Wechsler, for example, tend to have a significant Verbal Reasoning < Abstract/Visual Reasoning difference on the Binet. The Quantitative scale, early experience suggests, is less precise, probably because of the extensive language demands on the Quantitative subtest of this scale. Similarly, children with a Verbal Reasoning< Abstract/Visual Reasoning profile (e.g., see scores ofR.G. in Figure 3) may have a significant difference between ShortTerm Memory subtests that are sensitive to spatial as opposed to temporal memory. The author's early impressions are that the fourth edition may not be as useful as its predecessor for evaluating the general cognitive ability of young delayed children, a group for whom the 1960/1972 version was used frequently. Although it is at this writing an impression based on a limited sample, the fourth edition seems not to have adequate floor for this population.
Wechsler Inteiiigence Scale for Children-Revised (WISC-R; Wechsler, 1974) The WISC-R has been perhaps the most widely used measure of intellectual ability, and has been incorporated into the Reitan neuropsychological battery for older children (Selz, 1981; Selz & Reitan, 1979). Because of its excellent psychometric properties and the plethora of validity data on this instrument, the WISC-R probably remains the measure of choice in the assessment of children's cognitive strengths and weaknesses. The Verbal-Performance dichotomy employed in the WISC-R often is useful in demonstrating unique deficits in either linguistic or perceptual skills. Verbal-Performance splits have been identified in a number of populations of children with neuropsychological impairment, including those with learning disabilities (V < P or V > P, depending on subtype; Rourke, 1983), Turner's syndrome (V > P; Lewandowski, Costenbader, & Richman, 1985; McGlone, 1985), hydrocephalus (V > P; Spiegler et al., 1985), and children with Duchenne's dystrophy (V < P; Leibowitz & Dubowitz, 1981). Children with diffuse neuropsychological impairment such as attention deficit disorder (ADD) have been reported to exhibit unique deficits on the Arithmetic, Digit Span, Coding triad on the WISC-R (Lufi & Cohen, 1985). Ownby and Matthews (1985) suggested that this so-called "third factor" on the WISC-R is a complex cognitive index that may be associated with the ability to change mental set and to sustain attention during higher cognitive tasks. Such abilities ap-
234
CHAPTER 13 Stanford-Binet Intelligence Scale: Fourth Edition Verbal Reasoning
ss
Vocabulary Comprehension Absurdities
26 29 34
Verbal Reasoning SAS
. Abstract/Visual Reasoning Pattern Analysis Copying Matrices
= 54
52 (est., no ceiling) 48 (est., no ceiling) 48
Abstract/Visual Reasoning SAS = 98 est.
Quantitative Reasoning
ss
Quantitative
36
Quantitative Reasoning SAS
ss
= 72
Short-Term Memory Bead Memory Memory for Sentences Short-Term Memory SAS
Test Composite
ss 47 24
= 66
= 69
FIGURE 3. R.G.: Severe expressive communication disorder with autism. CA, 17-0.
pear to share common characteristics with the plan-
ning and integrative functions attributed to the frontal cortex (Goodglass & Kaplan, 1979). In general, WISC-R profiles such as those just described are most tenable for populations of children with developmental problems. For children with acquired impairments, particularly those dating from an early age, even lateralized insults may not translate into the hypothesized V-P split. Although such children appear to have relative sparing of presumably lateralized functions, a generalized decrement in intellectual ability may be evidenced (Aram, Ekelman, Rose, & Whitaker, 1985; Lewandowski & DeRienzo, 1985; Satz, 1985; Wilkening & Berg, 1985). Hence, specific impairment (e.g., depressed Verbal IQ subsequent to an early left lateralized trauma) appears to give way to more generalized losses. One exception to this rule may be children with acquired severe head injuries, who have been reported to show a significant deficit on Performance but not Verbal IQ relative to those with mild to moderate injuries (Bawden, Knights, & Winogron, 1985). Other studies also suggest that Performance IQ is especially vulnerable to the effects of head injuries (Berger-Gross & Shackelford, 1984; Chadwick, Rutter, Brown, Shaffer, & Traub, 1981). This V > P pattern appears to be associated with greater sensitivity of the Performance Scale's tasks to brain impairment rather than to locus of brain injury. In addition to interpretation of the separate Verbal-Performance scales on the WISC-R, many clinicians find that analysis of four cognitive factors proposed by Bannatyne (1974) contributes to understanding the neuropsychological strengths and
weaknesses of individual children. Of particular significance for describing subtypes of LD children are the relative scores on Bannatyne's sequential (comprising WISC-R Arithmetic, Coding, and Digit Span subtests) and spatial (formed by Picture Completion, Block Design, and Object Assembly subtests) factors. Populations of LD children have been reported to demonstrate a spatial > sequential pattern on the WISC-R (Bannatyne, 1978; Zingale & Smith, 1978). More recent investigations suggest this pattern may be revealed most often in reading-disabled subtypes (Fischer, Wenck, Schurr, & Ellen, 1985; Rugel, 1974; Stoiber et al., 1983; Telzrow, Century, Redmond, Whitaker, & Zimmerman, 1983). Individual subject characteristics, particularly IQ and gender, may relate to the cognitive patterns demonstrated on the Bannatyne recategorization of the WISC-R (Fischer et al., 1985).
Woodcock Johnson Psychoeducational Battery: Tests of Cognitive Ability (W]PB-TCA; Woodcock & Johnson, 1977) The WJPB-TCA, although not a measure of intellectual ability comparable to the WISC-R or the Binet (Hessler, 1982), nevertheless may prove to be of value in the appraisal of children's neuropsychological strengths and weaknesses. The WJPB-TCA is divided into four major cluster scores: Verbal Ability, Reasoning, Memory, and Visual-Perceptual Speed. Two of the clusters-Verbal Ability and Reasoning-use suppressor variables in the derivation of the cluster scores. Suppressor variables are nega-
APPLICATIONS OF EDUCATIONAL AND PSYCHOLOGICAL TESTS
tively weighted subtests designed to homogenize the two clusters so they are "purer" measures of verbal and reasoning ability. Individuals who score high on the main subtests within a cluster are presumed to have lower scores on the suppressor variable, and vice versa. To the degree this theoretical premise does not hold true for a given child, the cluster score may under- or overestimate that individual's performance on specific subtests (Breen, 1985). Score artifacts also may occur for children whose scores fall at either end of the distribution (Kampwirth, 1983). In any of these cases, computation of individual subtest scores is recommended (Hessler, 1982; Marston & Ysseldyke, 1980), although clinicians should be mindful of the fact that standard errors of measure.ment are higher for subtest than for cluster scores. Hessler ( 1982) identified three major profiles on the WJPB-TCA. The first, which has been associated with learning disabilities involving reading and spelling, is characterized by a low Verbal Ability cluster score relative to the Reasoning cluster score. A second profile is the reverse, with a low Reasoning cluster score compared to Verbal Ability. Such children may have adequate skills in the language areas such as reading and spelling, although they may exhibit arithmetic problems. A third type of pattern, which has been associated with diffuse neuropsychological impairment such as may be observed in children with attention deficits, is characterized by low scores on the Memory and Visual-Perceptual Speed areas relative to the other two cluster scores. Critics of the WJPB-TCA suggest the entire battery is verbally loaded, and thus tends to be biased against youngsters with weaker language abilities (Phelps, Rosso, & Falasco, 1985). Although the Reasoning cluster score of the scale appears upon initial examination to be sensitive to nonverbal cognitive ability, comparison of this cluster score and its two main subtests (Analysis-Synthesis and Concept Formation) with the WISC-R and the Halstead Category Test did not support this hypothesis (Harr & Telzrow, 1986). The WJPB-TCA Reasoning cluster score was shown to be correlated at significant levels with both the WISC-R Verbal and Performance IQ scores, and in fact showed higher correlations with the WISC-R sequential than spatial subtests. The Analysis-Synthesis subtest, in particular, shared substantial variance with the WISC-R Verbal IQ (r = 0. 72). This analysis revealed virtually no relationship between the WJPB-TCA Reasoning cluster or its subtests and scores on the Category Test. Harr and Telzrow concluded that the Reasoning cluster of the WJPB-TCA does not appear to be sensitive to nonverbal cognitive ability.
235
The Visual-Perceptual Speed cluster of the WJPB-TCA is somewhat unique among cognitive measures, and may offer potential for identifying neuropsychological deficits in this area, particularly those associated with conditions such as Tourette's syndrome or performance decrements attributed to specific medications. Such a measure of performance may have implications for vocational planning as well. The Memory cluster on the WJPB-TCA comprises a limited sample of short-term auditory memory, and hence clinicians should be cautious about overinterpreting these results. However, weaknesses in this area may be demonstrated by ADD or LD populations.
Academic Achievement Establishing levels of academic achievement is important for determining the presence of specific learning disabilities and loss of function following a specific trauma, as well as for helping to plan and monitor specific interventions. This section describes psychometrically sound measures of academic achievement, including both comprehensive achievement batteries and individual measures. K-ABC (Kaufman & Kaufman, 1983)
As noted above, in addition to its mental processing scale, the K-ABC also includes a comprehensive achievement battery. As such, it is the only published instrument that provides both intellectual ability and achievement data on the same standardization population. (Although this claim is made for the WJPB, many experts dispute the fact that the WJPB-TCA is a measure of intellectual ability in the traditional sense.) The K-ABC provides standard scores for these subtests: Expressive Vocabulary (ages 2-6 through 4-11 only), Faces & Places (a measure of general information), Arithmetic, Riddles (an index oflistening integration ability), Reading/Decoding (ages 5 and up), and Reading/ Understanding (ages 7 and up). In the author's experience, the K-ABC achievement scale is useful in generating and supporting hypotheses about a variety of neuropsychological problems in children. The following patterns on this scale have been identified. Youngsters with learning disabilities, as noted above, tend to have highest scores on the Riddles subtest. Because this is a listening task, and LD children often have problems in this area, such a pattern may seem incongruous. However, items on the Riddles subtest may be repeated, and
236
CHAPTER 13
the heavy simultaneous (i.e., integrative) emphasis tive to Mathematics and Knowledge. Such a profile, on this task seems to favor the strengths of these typically observed in youngsters with languageyoungsters. In contrast, Arithmetic and, particularly, based learning disorders affecting reading and writthe Reading subtests may be depressed for LD popu- ten language, is consistent with the low Verbal Abillations. Young children with severe communication ity-high Reasoning pattern on the cognitive portion disorders tend to score poorly on the Expressive Vo- of the WJPB. Mathematics learning disabilities are cabulary, Faces & Places, and Riddles subtests, al- reflected in a significantly weak performance on the thpugh their scores on Arithmetic and the Read- Mathematics cluster. Such students are reported to ing/Decoding subtests may be more within normal have a cognitive profile characterized by a high Verlimits at early ages, due to the high visuospatial con- bal Ability cluster score relative to the Reasoning tent in early number and letter recognition. The K- cluster score. Two other achievement profiles on the ABC is particularly useful in identifying preschool WJPB-TA are described by Hessler (1982). The first, hyperlexic children, whose scores may be at ceiling observed in children whose incidental learning abililevels on the Reading/Decoding subtest and above ties are adequate despite unique deficits in formal average on the Reading/Understanding subtest learning, is reflected in low scores in Reading, Math(given the gestural response mode and early single- ematics, and Written Language clusters relative to word items), despite significant deficiencies on Ex- the Knowledge cluster. Finally, students may obtain pressive Vocabulary, Faces & Places, and Riddles low scores on all four of the WJPB-TA clusters. Pre(Telzrow, 1985b) (seecaseofJ.G. inFigure2). Chil- sumably, such students have significant neuropsydren with unique visuospatial deficits tend to have an chological impairment, such as may be observed in opposite pattern on the K-ABC Achievement scale,. mental retardation or pervasive learning disabilities. and may exhibit deficits on Arithmetic and Reading tasks that rely on visuospatial skills, although tasks sensitive to oral language (i.e., Expressive Vocabu- Boder Test of Reading-Spelling Patterns (Boder & lary, Faces & Places, and Riddles) are preserved (see ]arrico, 1982) case of D.S. in Figure 1). Inclusion of the Boder, as with the Leiter, described earlier, represents a compromise of psychometric imprecision and clinical utility. Although W]PB: Tests of Achievement (W]PB-TA; criticized because of its psychometric weaknesses Woodcock & Johnson, 1977) (Reynolds, 1984), the Boder has nevertheless been The WJPB-T A derives cluster scores in four found to be of value in clinical practice (Hiltebeitel, achievement areas: reading, mathematics, written 1985; Nockleby & Galbraith, 1984; Telzrow et al., language, and knowledge (comprising social studies, 1983). This instrument, developed by pediatric neuscience, and humanities subtests). Cluster scores are rologist Elena Boder after extensive observation of not differentially weighted for variable intracluster the reading and spelling habits of dyslexic children, subtest performance; hence, a significant difference is designed to identify the presence and type of neubetween individual subtests comprising a cluster rologically based reading disorder. The scale uses score may result in the cluster score's misrepresent- graded word lists, balanced for phonetically regular ing those separate abilities. In such cases, separate and irregular words, and individual spelling lists that scores should be derived for the individual subtests are prepared for each subject based on reading perfor(Hessler, 1982; Marston &Ysseldyke, 1980). This mance. Three criteria (reading quotient, percent may occur, for example, for children with certain known words spelled correctly, percent unknown types of learning disabilities who perform adequately words spelled as good phonic equivalents) are used to on rote arithmetic tasks despite limited ability to ap- categorize children into these reading types: normal ply arithmetic concepts. In this instance the total reader, dysphonetic dyslexic, dyseidetic dyslexic, mathematics cluster score would underestimate com- mixed dyslexic, dyslexic reader with an unspecified putation skills and overestimate the score on the Ap- pattern, and a nonspecific reading disorder with a normal reading/spelling pattern. plied Problems subtest. In clinical use, the dyslexic reading types identiHessler's ( 1982) description of the cognitive profiles on the WJPB-TCA has corollaries on the fied by the Boder have been demonstrated to exhibit achievement portion of the battery. One pattern re- predicted profiles on the WISC-R (Telzrow et al., ported for LD children is characterized by low scores 1983), finger tapping (Telzrow et al., 1983), and the o_n the Reading and Written Language clusters rela- K-ABC (Telzrow et at., 1985). For example, chil-
APPLICATIONS OF EDUCATIONAL AND PSYCHOLOGICAL TESTS
dren whose reading is characterized by the poor sound-symbol associations and sound blending weaknesses typical of Boder' s auditory phonetic dyslexia tend to show a lower Verbal than Performance WISC-R IQ and demonstrate a spatial strength on Bannatyne's factors. Auditory phonetic dyslexics have been hypothesized to have a left hemisphere dysgenesis (e.g., Rourke, 1983), and a finger tapping profile suggestive of such a pattern frequently is observed. Such children also have been reported to exhibit a Sequential < Simultaneous profile on the K~ABC (Telzrow et al., 1985). This profile on the WISC-R, K-ABC, and Boder is depicted in the case study of S. W. at the conclusion of this chapter. In the author's experience, use of the Boder with beginning readers requires a special caution. Youngsters who are just beginning to read and spell often find sound-symbol associations difficult. Hence, their reading-spelling patterns, if scored strictly on the bases of the rules prescribed by Boder and Jarrico (1982), may be consistent with the authors' description of a dyslexic reader. However, frequently such children may be exhibiting a normal stage of reading development rather than a specific deficit (e.g., Bradley, 1983). Other data should be considered to help differentiate between neuropsychological dysfunction and a failure of acquisition.
Oral and Written Language Skills The broad area of oral and written language skills is diverse and complex, and cannot be treated comprehensively in this chapter. However, psychologists working with preschool and school-aged children should be able to identify signs of specific communication deficits of a neuropsychological origin as distinct from impoverished language as a result of poor language models or generally depressed cognitive ability. This section describes traditional tests of oral and written language that might be used for this purpose.
Test of Language Development (TOLD; Hammill
& Newcomer, 1982, 1984)
The TOLD, which is published in primary (TOLD-P) and intermediate (TOLD-I) versions, is designed as a screening measure of children's receptive and expressive language skilJs. ProfiJe analysis using ipsative comparison techniques might be used to determine whether a given youngster exhibits unique weaknesses on individual subtests. Reynolds ( 1983) developed a helpful table that enables clini-
237
cians to determine whether a given subtest score on the TOLD-Pis significantly higher or lower than the individual's own mean performance. Analysis of unique strengths and weaknesses in this fashion might lead to hypotheses about language deficits affecting phonological, semantic, or syntactic language systems (Elbert & Willis, 1984). Significantly poor performance on the Word Discrimination or Word Articulation subtest may suggest a phonological disturbance, whereas low scores on Grammatic Understanding and Grammatic Completion tasks may be associated with disruption of the syntactic language systems. Semantic skills presumably are reflected by scores on the Oral and Picture Vocabulary subtests.
Peabody Picture Vocabulary Test-Revised (PPVT-R; Dunn & Dunn, 1981) The PPVT-R is a test of receptive vocabulary appropriate for individuals aged 2! to 50. The multiple-choice pictoral format may utilize a pointing, oral (i.e., indication by letter designation), or yes/no response, so it is possible to adapt the test for braininjured persons who have significant upper extremity involvement. The PPVT-R can contribute to neuropsychological evaluation of children in a number of ways. In normal individuals, PPVT-R scores are moderately correlated with intelligence (median correlation with major tests in the mid 0.60s; Dunn & Dunn, 1981). Hence, major discrepancies between measured IQ and PPVT-R scores might suggest neuropsychological weaknesses in general language capacity or inadequate incidental learning opportunities. Such comparisons should be made in light of the average mean score differences reported between the PPVTR and cognitive scales (Bracken & Prasse, 1984). Comparison of PPVT-R scores with performance on nonlanguage tests such as the Beery has been suggested as a means of developing hypotheses about neuropsychological strengths and weaknesses (Hartlage, 1981; Satz & Fletcher, 1982).
Test of Written Language (TOWL; Hammill & Larsen, 1978) The TOWL is one of the few standardized, normative measures available to assess the written language skills of younger children. The instrument was standardized using 1602 (for scaled score norms) or 1712 (for grade equivalent score norms) children aged 8! to approximately 14!. The TOWL provides a
238
CHAPTER 13
balanced assessment of written language mechanics (spelling, grammar, and punctuation) and creative components (thematic maturity, vocabulary, and thought units). A written language quotient, or standardized age score, is derived from the five principal subtests; scores for Thought Units and Handwriting areas are reported separately. In the author's experience, use of the TOWL may be particularly helpful in identifying dysgraphia, especially of the type occurring without dyslexia. For such children, unique weaknesses in spelling and handwriting may be identified, although their performance on other TOWL subtests may be ageappropriate or above.
The Token Test for Children (DiSimoni, 1978)
The test authors (Hammill et al., 1980) advocate the use of ipsative interpretation, particularly using composite scores, and suggest as a guideline that differences of more than one standard deviation from the individual's own mean scores may reflect significant intraindividual variability. Hence, analysis of the relationship among composite scores may reveal generalized language deficits or specific impairment in written language areas such as reading and writing.
Motor/Visual Motor Skills Assessment of motor skills is important as an indication of general neurodevelopmental delay in young children, or as a sign of unique motor deficiency exclusive of cognitive functioning (e.g., in some cases of cerebral palsy or musculardystrophy). Visual motor performance can be used to establish a baseline for nonverbal cognitive skills, as well as an index of discrete abilities in this area. Specific measures sensitive to motor and visual motor performance are described in this section.
As DiSimoni indicates in the manual accompanying this instrument, the title "Token Test" has become a generic, in that there are numerous versions and adaptations from the original instrument described by DeRenzi and Vignolo in 1962. The DiSimoni form of the Token Test is essentially Noll's adaptation of the measure, standardized on 1304 children aged 3 to 12! without known language or learn- Physical Dexterity Tasks-System of ing problems (DiSimoni, 1978). When used appro- Multicultural, Pluralistic Assessment (SOMPA; priately, the Token Test can provide helpful data Mercer & Lewis, 1978) regarding the ability of children to follow oral directions, to understand basic language concepts, and to These tasks, designed for children aged 5-12, integrate verbal information. It has been reported su- are purported to identify neuropsychologically releperior to teacher judgment in identifying preschool vant motor behaviors. The tasks are divided into six children with receptive language difficulties (DiSim- areas: ambulation, equilibrium, placement, fine oni & Mucha, 1982). When scores on this task are motor sequencing, finger-tongue dexterity, and incompared with performance on nonverbal measures voluntary movement. Risk status is based on the perof mental development for young children, some hy- formance of the normative sample and age, as most potheses can be generated about the child's relative of the tasks are significantly age-related. strengths and weaknesses in neuropsychological The SOMPA Physical Dexterity tasks are of the processing. type frequently used to identify "soft neurological signs" in children (Gaddes, 1985; Spreen, Tupper, Risser, Tuokko, & Edgell, 1984). Such signs have Test of Adolescent Language (TOAL; Hammill, been associated with a variety of neuropsychological Brown, Larsen, & Wiederholt, 1980) deficits, including hyperactivity, subnormal IQ, and The TOAL is one of the few empirical, na- behavior problems (Shaffer, O'Connor, Shafer, & tionally standardized scales of adolescents' oral and Prupis, 1983). However, false-positives are comwritten language abilities. Although it is a long, diffi- monly identified, so cautious interpretation of chilcult test, and many clinicians have reported anec- dren's performance on such tasks is warranted. dotally that it identifies far too many false-positives, it probably remains the language measure of choice for adolescents. The test comprises eight subtests designed to survey in comprehensive fashion semantic and syntactic aspects of written and oral language via both receptive and expressive modes. Ten different composite scores can be derived from various combinations of subtests.
Developmental Test of Visual Motor Integration (VMI or Beery; Beery, 1982) The Beery is a form-copying task appropriate for children aged 2 and older; a developmental ceiling occurs at approximately age 13, and scores are not provided above this level. Performance on this
APPLICATIONS OF EDUCATIONAL AND PSYCHOLOGICAL TESTS ·
test provides a good estimate of general developmental level in younger children (Routh, 1984), as drawings of geometric designs and human figures tend to be highly correlated with age. Children with unique deficits in motor or visual perceptual skills (e.g., children with cerebral palsy or a late acquired left hemiparesis) would represent an exception to this rule. The VMI has been reported to be a valid predictor of future academic performance in kindergartenage children, and there is evidence that it is particularly related to subsequent mathematics achievement (Klein, 1978). Such a pattern between visuospatial skills, as reflected in VMI scores, and arithmetic achievement is consistent with the observations of Rourke and his colleagues (Rourke & Finlayson, 1978; Rourke & Strang, 1983). Use of the Beery in conjunction with nonmotor tasks of visual perception can help delineate specific neuropsychological skills. Youngsters who perform poorly on the Beery and the WISC-R Coding subtest, and yet score within normal limits on the Picture Completion, Object Assembly, and Block Design subtests on the WISC-R may exhibit discrete motor impairment. Conversely, children who show depressed scores on all these tasks may have a generalized visuospatial deficit regardless of response format.
Motor Scales of Selected Batteries In addition to the two specific measures mentioned above, a number of developmental instruments, particularly those designed for young children, include motor scales. Examples include the Bayley and McCarthy scales, described above, and others such as the Vineland Adaptive Behavior Scale (Sparrow, Balla, & Cicchetti, 1984) and the Battelle Developmental Inventory (Newborg, Stock, Wnek, Guidubaldi, & Svinicki, 1984).
239
been significantly correlated with achievement. Traits such as persistence and adaptability have been reported to be the best temperament predictors of achievement (Martin & Holbrook, 1985). In a metaanalysis of a wide variety of kindergarten and first grade predictors, including IQ, language skills, soft neurological signs, and behavioral characteristics, Hom and Packard (1985) found that social behavioral traits such as attention/distractibility were among the best predictors of subsequent school achievement. Furthermore, findings suggest that a behavioral checklist is superior to children's cognitive or neuropsychological performance in identifying brain damage confirmed by CAT scan (Klesges & Fisher, 1981). Other studies have shown a relationship between neuropsychological functioning and social behavior. For example, Gill and Lueger (1985) reported that adolescents with conduct disorders perform poorly on neuropsychological tasks sensitive to frontal lobe functions. Other support for the role of the frontal lobes in mediating behavior comes from Slomka, Tarter, and Hegedus (1984), who described attention variables such as distractibility and hyperactivity as dramatically apparent in a case of an adolescent with agenesis of the frontal lobes. Similar behavior problems were reported for an 11-year-old girl with frontal lobe dysfunction subsequent to surgery (McKay et al., 1985). Because all human behavior-be it cognitive, perceptual, or social-arises from the central nervous system (Gaddes, 1985), consideration of only some of these elements (e.g., only cognitive or achievement variables) may result in an incomplete description of neuropsychological abilities. Children's social-emotional behavior constitutes an important variable on which they are judged; thus, comprehensive assessment of such behavior in children is especially critical. This section describes psychometrically sound empirical measures of socialemotional behavior.
Social-Emotional Behavior Clinicians are becoming increasingly aware of the behavioral sequelae of neuropsychological dysfunction on social and interpersonal skills. One indication of the focus on neuropsychological aspects of behavior is evidenced by studies of children's temperament (Martin, 1983; Rothbart & Posner, 1985). A variety of temperament characteristics, including activity level, degree of responsiveness, and general mood, have been associated with individual differences in neuropsychological functioning. Teacher ratings of children's temperament, presumably genetically influenced behavioral characteristics, have
Child Behavior Checklist (CBCL) (Achenbach & Edelbrock, 1983) The CBCL is available in parent-report (ages 416), teacher-report (ages 6-16), direct observation (ages 4-16), and self-report (ages 11-18) formats. Factor analysis of the parent-report scales of 2300 children was used to cluster items into several scales sensitive to both internalizing and externalizing dimensions. Similar analyses were conducted for the teacher-report form for 1800 children (McConaughy, 1985). Profiles of specific populations have been
240
CHAPTER 13
significant developmental delays requiring selection of an instrument designed for chronologically younger populations. The author has utilized the Bayley, for example, in assessing autistic children as old as 18 years. This procedure is preferable to selecting an age-appropriate instrument with insufficient floor to describe the neuropsychological functioning of a given individual. A second variable to consider when selecting instruments within each of the areas of function concerns the response format of the test and specific handicaps of the child that may be inconsistent with these. To illustrate, tests of receptive language that require a pointing response (e.g., the Token Test for Children) may be inappropriate for youngsters with impaired use of upper extremities. A distinction should be made between nonessential task demands Revised Behavior Problem Checklist (RBPC; Quay and tests of specific neuropsychological abilities. & Peterson, 1979) The example of the Token Test illustrates non-essential task demands; that is, a pointing response is not The RBPC is a recent revision and expansion of essential to the evaluation of language understandits predecessor, the Behavior Problem Checklist. ing. Consideration of the child's specific skills in Factor-analytic studies of the behavior problems of relation to non-essential task demands is necessary in special populations (including psychiatric inpatients order to obtain an accurate description of the child's and outpatients, LD students, and developmentally neuropsychological abilities. It would be inaccurate, disabled children) produced six scales: Conduct Disfor example, to conclude that a child's receptive lanorder, Socialized Aggression, Attention Problemsguage skills were deficient because of his or her inImmaturity, Anxiety-Withdrawal, Psychotic Beability to point and manipulate tokens. However, it havior, and Motor Excess. The scales have been would not be inappropriate to administer a test of reported to differentiate between normal and clinic motor ability to the child, for such a measure samples of children (Quay, 1983). In addition, conwould-quite appropriately-describe the signifistruct validity for several of the RBPC's scales is cant motor limitations of the individual (Willis, reported in a comparison with DSM-ill categories Culbertson, & Mertens, 1984). (Quay, 1983). Concurrent validity for the RBPC has In addition to the child's age and non-essential been demonstrated by high correlations with behav- task demands in light of specific impairments, a third ioral observation data in various contexts (Lahey & variable to consider when making test selections conPiacentini, 1985). The absence of representative na- cerns the validity of specific instruments for populational norms for this scale has been described as a tions of children with various neuropsychological potential limitation to its clinical utility (Lahey & disorders. In the case of some of the measures dePiacentini, 1985). scribed, such validity h,as been demonstrated through extensive investigation with children with a wide variety of developmental and acquired disorders. For Selecting Combinations of Tests others, the recent release of the instrument has not made the accumulation of validity data possible, alWhen assessing the neuropsychological abili- though clinical experience with these instruments ties of children, a comprehensive survey of behavior suggests that they may contribute to the description is desirable. Hence, collecting data about all the of neuropsychological abilities in children. A fourth variable to consider when selecting functioning areas just described is recommended. The choice of specific tests within each area requires specific test instruments concerns the purpose of the consideration of a number of child variables and test evaluation and the specific question(s) it is intended to answer. Clinicians in hospitals or rehabilitation characteristics. The first factor to consider is the age or develop- settings may be called upon to establish levels of loss mental1evel of the child. Although the two are com- of function and to monitor recovery or, for degenerparable in most children, some youngsters exhibit ative conditions, progressive deterioration. Psycholidentified on the CBCL, including delinquent, ADD with hyperactivity, and a subtype characterized by significant aggressiveness that is elevated on the Depressed, Social Withdrawal, and the Aggressive scales (Achenbach & Edelbrock, 1983). Further analysis has suggested the teacher form of the CBCL may help distinguish between subtypes of ADD (with and without hyperactivity) (Edelbrock, Costello, & Kessler, 1984). Others have found that LD children may be elevated on various of the CBCL scales, although especially on the hyperactivity scale (McConaughy & Ritter, 1986). Many of the items on this scale are sensitive to attention deficits, poor impulse control, and deficient interpersonal skills associated with neuropsychological deficits in children.
APPLICATIONS OF EDUCATIONAL AND PSYCHOLOGICAL TESTS
ogists working in educational settings must respond to questions regarding the need for special programming and adaptive equipment and the viability of certain interventions. Knowledge of the degree to which specific instruments can help the psychologist be responsive to such referral questions is a critical ingredient in conducting neuropsychological evaluations. Finally, choosing the standardized test components of a flexible battery of traditional educational and psychological tests should incorporate knowledge of the test interactions and the ways in which scores from various instruments complement one another and contribute to a description of the child's functioning. Such awareness is gleaned not only from research data, but also from clinical practice and experience in using diverse instruments with various populations of children.
Developing Hypotheses about Neuropsychological Functioning from Combinations of Traditional Tests Several premises about neuropsychological organization (e.g., lateralized abilities, a comparison of developmental milestones and current performance) can guide the clinician in developing hypotheses relative to the nature, etiology, and chronicity of observed neuropsychological deficits (e.g., Hartlage & Telzrow, 1986). One example of the application of these guidelines is demonstrated by a neuropsychological profile of specific dyslexia, characterized by a significantly lower WISC-R Verbal than Performance IQ; a Bannatyne factor pattern favoring spatial skills; generally more efficient visuospatial processing (e.g., on the Beery VMI) relative to language ability (e.g., TOLD-I); and a dysphonetic reading-spelling pattern on the Boder Test (Telzrow et al., 1983). The test data shown for S. W. in Figure 4 illustrates such a profile. Characteristic patterns of this sort have been identified for various groups of children with neuropsychological impairment, and as further study is conducted on subtyping conditions such as learning disabilities (Rourke, 1985), head injuries (Rourke et al., 1983), and hyperlexia (McClure & Hynd, 1983), greater specificity of such profiles is anticipated. However, sole reliance on known neuropsychological profiles is limiting, as such patterns have not been identified for all variations of impairment, and those that have been described have been reported to be unreliable when applied to groups of children,
241
Wechsler Intelligence Scale for Children-Revised
ss
Verbal
6
Information Similarities Arithmetic Vocabulary Comprehension Digit Span
ss
Performance Picture Completion Picture Arrangement Block Design Object Assembly Coding
9
6 9 7
12 14 14 19
6
5 Performance 10 = 121
Verbal 10 = 84
Full Scale 10 = 99
Kaufman Assessment Battery for Children Sequential Subtests
ss
Hand Movements Number Recall Word Order
9 7 6
Sequential Score= 83 ± 8
Simultaneous Subtests
7 7
Simultaneous Score= 91 ± 6
Nonverbal Score
Faces & Places Arithmetic Riddles Reading/Decoding Reading/Understanding
13 11 6
Gestalt Closure Triangles Matrix Analogies Spatial Memory Photo Series
Mental Processing Composite
Achievement Subtests
ss
= 86
= 86
± 6
± 6
ss 96 ± 8 79 ± 8
89 ± 9 76 ± 8 77 ± 8
Soder Test of Reading-Spelling Patterns Reading Quotient = 68 Known Words = 40% Unknown Words = 0% FIGURE 4. S.W.: Developmental dyslexia, auditory phonetic
type. CA, 10-9.
possibly because of the influence of such moderating variables as age, gender, and rehabilitation experiences. Individual differences are just that-individual-and although it sometimes is possible to generalize across groups of children with known neurological conditions, a more viable and useful approach employs intraindividual profile analysis to identify neuropsychological strengths and weaknesses (Reynolds, 1985). When integrating data from traditional educational and psychological tests, the clinician utilizes
242
CHAPTER 13
identified intraindividual variability to provide a description of impaired neuropsychological functions as well as residual assets. As noted early in this chapter, the flexible battery approach is less identified with determining the presence and loci of brain impainnent as with a comprehensive behavioral description of how impaired abilities and residual strengths affect routine functioning. For children, the immediate outcomes of such a description should include a prediction about the child's acquisition of new skills in specific areas (e.g., language, motor development), a discussion of implications for school learning and activities of daily living (e.g., self-care skills, interpersonal behavior), and suggestions about optimal intervention strategies (Telzrow, 1985a). Such assessment data also can be used to help adolescents select vocational options that are not contraindicated by their individual neuropsychological profiles (Hartlage & Telzrow, 1984, 1986). Finally, because enjoyable leisure activities are critical for optimal personal adjustment, clinicians may wish to counsel parents, children, and adolescents about the neuropsychological relevance of data from traditional psychological and educational tests for making avocational choices (Hartlage & Telzrow, 1986).
References Achenbach, T. M., &Edelbrock, C. (1983).MfllllllllfortheChild Behavior Checklist. Burlington, VT: University Associates in Psychiatry. Aram, D. M., Ekelman, B. L., Rose, D. F., & Whitaker, H. A. (1985). Verbal and cognitive sequelae following unilateral lesions acquired in early childhood. Journal of Clinical and Experimental Neuropsychology, 7, 55-78. Bannatyne, A. (1974). A note on recategorization of the WISC scaled scores. Journal ofLearning Disabilities, 7, 272-274. Bannatyne, A. (1978). The spatially competent LD child. Academic Therapy, 14, 133-155. Batshaw, M. L., & Perret, Y. M. (1981). Children with handicaps: A medical primer. Baltimore: Brooks. Bawden, H. N., Knights, R. M., & Winogron, H. W. (1985). Speeded perfonnance following head injury in children. Journal of Clinical and Experimental Neuropsychology, 7, 39-54. Bayley, N. (1969). Manual for the Bayley Scales of Infant Development. New York: Psychological Corporation. Beckett, L., Reuter, J., & Stancin, T. (1984). The use of the McCarthy Scales of Children's Abilities to assess moderately retarded children. Psychology in the Schools, 21, 305-310. Beery, K. E. (1982). Revised administration, scoring, and teaching mammlfor the Developmental Test ofVisuai-Motor Integration. Chicago: Follett. Berger-Gross, P., & Shackelford, M. (1984, February). Closed
head injury in children: Intelligence and academic achievement. Paper presented at the International Neuropsychological Society, Houston. Boder, E., & Jarrico, S. (1982). The Boder Test of ReadingSpelling Panerns, Manual. New York: Grone & Stratton. Bracken, B. A. (1985). A critical review of the Kaufman Assessment Battery for Children (K-ABC). School Psychology Review, 14, 21-36. Bracken, B. A., & Prasse, D.P. (1984). Peabody Picture Vocabulary Test-Revised: An appraisal and review. School Psychology Review, 13, 49-60. Bradley, L. (1983). The organization of visual, phonological, and motor strategies and learning to read and to spell. In U. Kirk (Ed.), Neuropsychology of language, reading, and spelling (pp. 235-254). New York: Academic Press. Breen, M. J. (1985). The Woodcock-Johnson Tests of Cognitive Ability: A comparison of two methods of cluster scale analysis for three learning disability subtypes. Journal of Psychoeducational Assessment, 3, 167-174. Chadwick, 0., Rutter, M., Brown, E., Shaffer, D., & Traub, M. (1981 ). A prospective study of children with head injuries. II. Cognitive sequelae. Psychological Medicine, 11, 49-61. Dean, R. S. (1986). Perspectives on the future of neuropsychological assessment. In B. S. Plake & J. C. Witt (Eds. ), Buros-Nebraska series on measurement and testing: Future of testing and measurement (pp. 203-244). Hillsdale, NJ: Erlbaum. DeRenzi, E., & Vignolo, L. (1962). The Token Test: A sensitive test to detect receptive disturbances in aphasics. Brain, 85, 665-678. DiSimoni, F. G. (1978). The Token Test for Children. Allen, TX: DLM Teaching Resources. DiSimoni, F. G., & Mucha, R. (1982). Use of the Token Test for Children to identify language deficits in preschool age children. The Journal of Auditory Research, 22, 265-270. Dunn, L., & Dunn, L. (1981). Manual/or the Peabody Picture Vocabulary Test-Revised. Circle Pines, MN: American Guidance Service. Edelbrock, D., Costello, A. J., & Kessler, M. D. (1984). Empirical conoboration of attention deficit disorder. Journal of the American Academy of Child Psychitltry, 23, 285-290. Elbert, J. C., & Willis, D. J. (1984). Psychological evaluation of children with speech and language disorders. InS. J. Weaver (Ed.), Testing children (pp. 104-120). Kansas City, MO: Test Corporation of America. Fischer, W. E., Wenck, L. S., Schurr, K. T., & Ellen, A. S. (1985). The moderating influence of gender, intelligence, and specific achievement deficiencies on the Bannatyne WISC-R recategorization. Journal ofPsychoeducational Assessment, 3, 245-255. Fletcher, J. M. (1985). (ReviewofChildneuropsychology.)Journal of Clinical and Experimental Neuropsychology, 7, 577584. Gaddes, W. H. ( 1985). Learning disabilities and brain function: A neuropsychological approach (2nd ed.). Berlin: SpringerVerlag. German, D. J. (1983). Analyses of word-finding disorders on the Kaufman Assessment Battery for Children (K-ABC). Journal of Psychoeducational Assessment, 1, 121-133.
APPLICATIONS OF EDUCATIONAL AND PSYCHOLOGICAL TESTS Gill, K. J., & Lueger, R. J. (1985, October). Frontal lobe cog· nitive dysfunction in conduct disorder adolescents. Paper presented at the meeting of the National Academy of Neuropsychologists, Philadelphia. Goetz, E. T., & Hall, R. J. (1984). Evaluation of the Kaufman Assessment Battery for Children from an information-processing perspective. The Journal of Special Education, 18, 281-296. Gob, D. S., & Simons, M. R. (1980). Comparison of learning disabled and general education children on the McCarthy Scales of Children's Abilities. Psychology in the Schools, 17, 429-436. Golden, C. J. (1981). The Luria-Nebraska Children's Battery: Theory and formulation. In G. W. Hynd & J. E. Obrzut (Eds.), Neuropsychological assessment and the school-age child (pp. 277-302). New York: Grone & Stratton. Goldstein, G. (1984). Comprehensive neuropsychological assessment batteries. In G. Goldstein & M. Hersen (Eds.), Hand· book ofpsychological assessment (pp. 181-210). Elmsford, NY: Pergamon Press. Goodglass, H., & Kaplan, E. (1979). Assessment of cognitive deficit in the brain-injured patient. In M.S. Gazzaniga (Ed.), Handbook of behavioral neurobiology (Vol. 2, pp. 3-22). New York: Plenum Press. Gustavson, J. L., Golden, C. J., Wilkening, G. N., Hermann, B. P., Plaisted, J. R., Maclooes, W. D., &Leark, R. A. (1984). The Luna-Nebraska Neuropsychological Battery-Children's Revision: Validation with brain-damaged and normal children. Journal ofPsychoeducational Assessment, 2, 199208. Hammill, D. D., Brown, V. L., Larsen, S.C., & Wiederholt, J. L. (1980). Test of Adolescent lAnguage. Austin, TX: PROED. Hammill, D. D., & Larsen, S. C. (1978). Manual for the Test of Written Language. Austin, TX: PRO-ED. Hammill, D. D., & Newcomer, P. L. (1982). Test of lAnguage Development-Primary. Austin, TX: PRO-ED. Hammill, D. D., & Newcomer, P. L. (1984). Test of lAnguage Development-Intermediat e. Austin, TX: PRO-ED. Harr. G., & Telzrow, C. (1986, April). Common variance among WJPB cognitive scale, WISC-R, and Category Test. Paper presented at the meeting of the National Association of School Psychologists, Hollywood, FL. Hartlage, L. C. (1981). Neuropsychological assessment techniques. In C. R. Reynolds & T. B. Gutkin (Eds.), Handbook of school psychology (pp. 296-313). New York: Wiley. Hartlage, L. C., & Telzrow, C. F. (1984). Rehabilitation of per· sons with learning disabilities. Journal ofRehabilitation, 50, 31-34. Hartlage, L. C., & Telzrow, C. F. (1986). Neuropsychological assessment and intervention with children and adolescents. Sarasota, FL: Professional Resource Exchange. Hessler, G. L. (1982). Use and interpretation of the WoodcockJohnson Psycho-educational Battery. Hingham, MA: Teaching Resources Corporation. Hiltebeitel, C. L. (1985, October). Characteristicsojdysphonetic learning disabled children. Paper presented at the meeting of the National Academy of Neuropsychologists, Philadelphia. Hooper, S. R., & Hynd, G. W. (1985). Differential diagnosis of
243
subtypes of developmental dyslexia with the Kaufman Assessment Battery for Children (K-ABC). Journal of Clinical Child Psychology, 14, 145-152. Hom, W. R., & Packard, T. (1985). Early identification ofleaming problems: A meta-analysis. Journal of Educational Psychology, 77, 597-607. Kamphaus, R. W., & Reynolds, C. R. (1984). Development and structure of the Kaufman Assessment Battery for Children. The Journal of Special Education, 18, 213-228. Kampwirth, T. J. (1983). Problems in the use of the WoodcockJohnson suppressors. Journal of Psychoeducational Assess· ment, 1, 337-339. Kaufman, A. S. (1976). Do normal children have "flat" ability profiles? Psychowgy in the Schools, 13, 284-285. Kaufman, A. S. (1979).1ntelligent testing with the WISC-R. New York: Wiley. Kaufman, A. S. (1984). K-ABC and controversy. The Journal of Special Education, 18, 409-444. Kaufman, A. S., &Kaufman, N. L. (1977). Clinical evaluation of young children with the McCarthy Scales. New York: Grune & Stratton. Kaufman, A. S., & Kaufman, N. L. (1983). Kaufman Assessment Batteryfor Children: lnterpretive manual. Circle Pines, MN: American Guidance Service. Kaufman, A. S., Kaufman, N. L., Kamphaus, R. W., & Naglieri, J. A. (1982). Sequential and simultaneous factors at ages 312!: Developmental changes in neuropsychological dimensions. Clinical Neuropsychology. 4, 74-81. Keith, T. Z. (1985). Questioning the K-ABC: What does it measure? School Psychology Review, 14, 9-20. Keith, T. Z., & Bolen, L. M. (1980). Factor structure of the McCarthy Scales for children experiencing problems in school. Psychology in the Schools, 17, 320-326. Klein, A. E. (1978). The validity of the Beery Test of VisualMotor Integration in predicting achievement in kindergarten, ftrst, and second grades. Educational and Psychological Measurement, 38, 457-461. Klesges, R. C., & Fisher, L. P. (1981). A multiple criterion approach to the assessment of brain damage in children. Clinical Neuropsychology, 3, 6-l I. Klesges, R. C., Fisher, L., Pheley, A., Boschee, P., & Vasey, M. ( 1984). A major validational study of the Halstead-Reitan in the prediction of CAT-scan assessed brain damage in adults. The International Journal of Clinical Neuropsychology, 6, 29-34. Lahey, B. B., & Piacentini, J. C. (1985). An evaluation of the Quay-Peterson Revised Behavior Problems Checklist. Journal of School Psychology, 23, 285-289. Leibowitz, D., & Dubowitz, V. (1981). Intellect and behavior in Duchenne muscular dystrophy. Developmental Medicine and Child Neurology, 23, 577-590. Leiter, R. G. (1969). Examiner's manual for the Leiter International Performance Scale. Chicago: Stoelting. Lewandowski, L., Costenbader, V., & Richman, R. (1985). Neuropsychological aspects of Turner syndrome. The International Journal of Clinical Neuropsychology, 7, 144-147. Lewandowski, L. J., & DeRienzo, P. J. (1985). WISC-R and KABC performances of hemiplegic children. Journal of Psychoeducational Assessment, 3, 215-221.
244
CHAPTER 13
Lufi, D., & Cohen, A. (1985). Using the WISC-R to identify attentional deficit disorder. Psychology in the Schools, 22, 40-42. Majovski, L. V. (1984). The K-ABC: Theory and applications for child neuropsychological assessment and research. The Journal of Special Education, 18, 257-268. Marston, D., & Ysseldyke, J. (1980). Derived subtest scores for the Woodcock-Johnson Psycho-educational Battery. Hingham, MA: Teaching Resources Corporation. Martin, R. P. (1983). Temperament: A review of research with implications for the school psychologist. School Psychology Review, 12, 266-273. Martin, R. P., & Holbrook, J. (1985). Relationship of temperament characteristics to the academic achievement of firstgrade children. Journal ofPsychoeducational Assessment, 3, 131-140. McCarthy, D. (1972). McCarthy Scales of Children's Abilities. New York: Psychological Corporation. McClure, P. H., & Hynd, G. W. (1983). Is hyperlexia a severe reading disorder or a symptom of psychiatric disturbance? Nosological considerations. Clinical Neuropsychology, 5, 145-149. McConaughy, S. H. (1985). Using the Child Behavior Checklist and related instruments in school-based assessment of children. School Psychology Review, 14, 479-494. McConaughy, S. H., & Ritter, D. R. (1986). Social competence and behavioral problems of learning disabled boys aged 611. Journal of Learning Disabilities, 19, 39-45. McGlone, J. (1985). Can spatial deficits in Turner's syndrome be explained by focal CNS dysfunction or atypical speech lateralization? Journal of Clinical and Experimental Neuropsychology, 7, 375-394. McKay, S., Stelling, M. W., Bauman, R. J., Carr, W. A., Walsh, J. W., & Gilmore, R. L. (1985). Assessment of frontal lobe dysfunction using the Luria-Nebraska Neuropsychological Battery-Children's Revision: A case study. International Journal of Clinical Neuropsychology, 7, 107-111. Mercer, J. R., & Lewis, J. F. (1978). System of Multicultural Pluralistic Assessment. New York: Psychological Corporation. Naglieri, J. A., Kaufman, A. S., &Harrison, P. L. (1981). Factor structure of the McCarthy Scales for school-age children with low GCis. The Journal of School Psychology, 19, 226-232. Newborg, J., Stock, J. R., Wnek, L., Guidubaldi, J., & Svinicki, J. (1984). Battelle Developmental Inventory. Allen, TX: DLM Teaching Resources. Nockleby, D. M., & Galbraith, G. G. (1984). Developmental dyslexia subtypes and the Boder Test of Reading-Spelling Patterns. Journal of Psychoeducational Assessment, 2, 91-
100.
Obrzut, A., Obrzut, J. E., & Shaw, D. (1984). Construct validity of the Kaufman Assessment Battery for Children with learning disabled and mentally retarded. Psychology in the Schools, 21, 417-424. Ownby, R. L., & Matthews, C. G. (1985). On the meaning of the WISC-R third factor: Relations to selected neuropsychological measures. Journal of Consulting and Clinical .Psychology, 53, 531-534. Phelps, L., Rosso, M., & Falasco, S. L. (1985). Multiple regres-
sion data using the WISC-R and the Woodcock-Johnson tests of cognitive ability. Psychology in the Schools. 22. 46-49. Prasse, D.P., Siewert, J. C., & Ellison, P. H. (1983). McCarthy performance and neurological functioning in children born "at risk." Journal of Psychoeducational Assessment, 1. 273-283. Quay, H. C. (1983). A dimensional approach to behavior disorder: The Revised Behavior Problem Checklist. School Psychology Review, 12, 244-249. Quay, H. C., & Peterson, D. R. (1979). Revised Behavior Problem Checklist. Author. Rapin, I., & Allen, D. A. (1983). Developmental language disorders: Nosologic considerations. In U. Kirk (Ed.), Neuropsychology of language, reading, and spelling (pp. 155-184). New York: Academic Press. Reitan, R. M. (1979). Manual for administration of neuropsychological test baneries for adults and children. Tucson, AZ: Reitan Neuropsychology Laboratories, Inc. Reuter, J., Stancin, T., & Craig, P. (1981). Kent scoring adaptation of the Bayley Scales of Infant Development. Kent, OH: Kent Developmental Metrics. Reynolds, C. R. (1981). A note on determining significant discrepancies among category scores on Bannatyne's regrouping of WISC-R subtests. Journal ofLearning Disabilities, 14, 468469. Reynolds, C. R. (1983). Statistics for the enhancement of profile analyses on the Test of Language Development-Primary. Psychology in the Schools, 20, 5-9. Reynolds, C. R. (1984). Psychometric characteristics of the Boder Test of Reading-Spelling Patterns. School Psychology Review, 13, 526-529. Reynolds, C. F. (1985, August). Toleration in technology of instruction: Implications of the NAS report for special education. Paper presented at the meeting of the American Psychological Association, Los Angeles. Rothbart, M. K., & Posner, M. I. (1985). Temperament and the development of self-regulation. In L. C. Hartlage & C. F. Telzrow (Eds.), The neuropsychology of individual differences: A developmental perspective (pp. 93-123). New York: Plenum Press. Rourke, B. P. (1981 ). Neuropsychological assessment of children with learning disabilities. In S. B. Filskov & T. J. Boll (Eds.), Handbook of clinical neuropsychology (pp. 453478). New York: Wiley. Rourke, B. P. (1983). Reading and spelling disabilities: A developmental neuropsychological perspective. In U. Kirk (Ed.), Neuropsychology of language, reading, and spelling (pp. 209-234). New York: Academic Press. Rourke, B. P. (Ed.). (1985). Neuropsychology of learning disabilities: Essentials of subtype analysis. New York: Guilford Press. Rourke, B. P., Bakker, D. J., Fisk, J. L., & Strang, J.D. (1983). Child neuropsychology. New York: Guilford Press. Rourke, B. P., &Finlayson, M.A. I. (1978). Neuropsychological significance of variations in patterns of academic performance: Verbal and visual-spatial abilities. Journal of Abnormal Child Psychology, 6, 121-133 . Rourke, B. P., & Strang, J. D. (1983). Subtypes of reading and
APPLICATIONS OF EDUCATIONAL AND PSYCHOLOGICAL TESTS
arithmetical disabilities: A neuropsychological analysis. In M. Rutter (Ed.), Developmental neuropsychiatry (pp. 473488). New York: Guilford Press. Routh, D. K. (1984). Appraisal of children with learning and behavior problems. InS. J. Weaver (Ed.), Testing children (pp. 15-23). Kansas City, MO: Test Corporation of America. Rugel, R. (1974). WISC subtest scores of disabled readers: A review with respect to Bannatyne's recategorization. Journal of Learning Disabilities, 7, 48-55. Sattler, J. M. (1982). Assessment of children's intelligence and special abilities. Boston: Allyn & Bacon. Satz, P. ( 1985, August). Recovery from early brain damage: Some facts and reflections. Invited address at the meeting of tbe American Psychological Association, Los Angeles. Satz, P., &Fletcher, J. M. (1982). Manua/fortheFloridaKindergarten Screening Battery. Odessa, FL: Psychological Assessment Resources. Selz, M. (1981). Halstead-Reitan Neuropsychological Test Batteries for children. In G. W. Hynd & J. E. Obrzut (Eds.), Neuropsychological assessment and the school-age child (pp. 195-235). New York: Grune & Stratton. Selz, M., & Reitan, R. M. (1979). Rules for neuropsychological diagnosis: Classification of brain function in older children. Journal of Consulting and Clinical Psychology; 47, 258264. Shaffer, D., O'Connor, P. A., Shafer, S. Q., & Prupis, S. (1983). Neurological "soft signs": Their origins and significance for behavior. In M. Rutter (Ed.), Developmental neuropsychiatry (pp. 144-163). New York: Guilford Press. Shah, A., & Holmes, N. (1985). Brief report: The useoftbe Leiter International Performance Scale with autistic children. Journal of Autism and Developmental Disorders, 15, 195-203. Slomka, G. T., & Tarter, R. E. (1984). Mental retardation. In R. E. Tarter & G. Goldstein (Eds.), Advances in clinical neuropsychology(Vol. 2, pp. 109-137). New York: P1enumPress. Slomka, G., Tarter, R., & Hegedus, A. (1984). Agenesis of the frontal lobes: Neuropsychological sequelae. International Journal of Clinical Neuropsychology, 6, 12-16. Sparrow, S. S., Balla, D. A., &Cicchetti, D. V. (1984). Vineland Adaptive Behavior Scales. Circle Pines, MN: American Guidance Services. Spiegler, B., Baron, I. S., Hammock, M. K., &McCullough, D. C. (1985, February). Verbal and spatial recall in children with hydrocephalus. Paper presented at the meeting of the International Neuropsychological Society, San Diego. Spreen, 0., Tupper, D., Risser, A., Tuokko, H., & Edgell, D. ( 1984). Human developmental neuropsychology. New York: Oxford University Press. Sternberg, R. J. (1984). The Kaufman Assessment Battery for Children: An information-processing analysis and critique. The Journal of Special Education, 18, 269-279. Stoiber, K. C., Bracken, B. A., & Gissal, T. J. (1983). Cognitive
245
processing styles in reading-disabled and a matched group of normal children. Journal of Psychoeducational Assessment, 1, 219-233. Sutter, E. G., &Battin, R. R. (1984). Using traditional psychological tests to obtain neuropsychological information on children. international Journal of Clinical Neuropsychology, 6, 115-119. Taylor, H. G., Fletcher, J. M., & Satz, P. (1984). Neuropsychological assessment of children. In G. Goldstein & M. Hersen (Eds.), Handbook of psychological assessment (pp. 211-234). Elmsford, NY: Pergamon Press. Telzrow, C. F. ( 1984). Practical applications of the K-ABC in the identification of handicapped preschoolers. The Journal of Special Education, 18, 311-324. Telzrow, C. F. ( 1985a). The science and speculation of rehabilitation in developmental neuropsychological disorders. In L. C. Hartlage & C. F. Telzrow (Eds.), Neuropsychological aspects of individual differences: A developmental perspective (pp. 271-307). New York: Plenum Press. Telzrow, C. F. (1985b, October). Use of the Leiter, the K-ABC, and interactive observation to differentiate pure hyperlexic syndromes from autism in preschool children. Paper presented to the meeting of the National Academy of Neuropsychologists, Philadelphia. Telzrow, C. F., Century, E., Harris, B., & Redmond, C. (1985, April). Relationship between neuropsychological processing models and dyslexic subtypes. Paper presented to the meeting of the National Association of School Psychologists, Las Vegas. Telzrow, C. F., Century, E., Redmond, C., Whitaker, B., & Zimmerman, B. (1983). The Boder Test: Neuropsychological and demographic features of dyslexic subtypes. Psychology in the Schools, 20, 427-432. Thorndike, R. L., Hagen, E. P., & Sattler, J. M. (1986). Sranford-Binet Intelligence Scale: Guide for administering and scoring the fourth edition. Chicago: Riverside. Wechsler, D. (1974). Manual for the Wechsler Intelligence Scale for Children-Revised. New York: Psychological Corporation. Wilkening, G. N., & Berg, R. A. (1985, October). Neuropsychological performance patterns of early and late acquired brain damage. Paper presented at the meeting of tbe National Academy of Neuropsychologists, Philadelphia. Willis, D. J., Culbertson, J. L., & Martens, R. A. (1984). Considerations in physical and health-related disorders. In S. J. Weaver (Ed.), Testing children (pp. 185-196). Kansas City, MO: Test Corporation of America. Woodcock, R. W., & Johnson, M. B. (1977). Woodcock-Johnson Psycho-educational Battery. Hingham, MA: Teaching Resources. Zingale, S. A., & Smith, M. D. (1978). WISC-R patterns for learning disabled children at three SES levels. Psychology in the Schools, 15, 199-204.
14 Radiological Techniques in Neuropsychological Assessment ERIN D. BIGLER
Introduction Since the introduction (in the early 1970s) of computerized axial tomography (CT) scanning techniques (see review by Oldendorf, 1980), imaging of the brain has been revolutionized. Current CT imaging is capable of generating images that offer detailed depictions of the major anatomical structures of the brain. Prior to the advent of the CT, only inferential imaging methods were available. These procedures were not universally utilized because of their limited clinical efficacy (e.g., radioisotope scans could not detect anything but the largest type of pathology such as a tumor) and morbidity risks with invasive procedures (e.g., pneumoencephalography). Accordingly, for a patient to undergo one of these procedures, there had to be clear clinical justification, and the neuroimages obtained were crude facsimiles in comparison to today's standards. However, with the current status of CT scanning the patient is at no greater risk than for a routine X-ray procedure, the CT scan provides an excellent image of the brain, and as such has become a routine procedure performed on most patients, including infants and children, presenting with neurologic symptoms. Clinical neuropsychology had its beginning in the pre-CT era. Prior to 1975, lesion-localization studies were primarily based on neuropsychologic inference, were dependent on findings from neurologic exam or direct inspection of the brain during neurosurgery, or on the rather primitive imaging
ERIN D. BIGLER • Department of Psychology, University of Texas at Austin, Austin, Texas 78712; and Austin Neurological Clinic, Austin, Texas 78705.
techniques of that time (e.g., technetium brain scan, pneumoencephalography, arteriography). Currently, in the post-CT scan era, where CT imaging is now commonplace, more precise lesion-localization relationships have been and are being established (Bigler, 1984; Kertesz, 1983). In fact, an era has now been reached in which CT information, when available, should be routinely used in neuropsychologic diagnostics. Historically, neuropsychologic diagnostic practice prided itself in its independence from other neurologic tests in the assessment of brain functioning (see Lezak, 1983). In current practice, however, ignoring such CT information when available could greatly detract from the completeness of an evaluation. This will become an even greater factor with further improvements in neuroimaging. For example, the new technique of magnetic resonance imaging (MRI) poses no radiation risks, provides an exceptional image of anatomy (see Figure 1), and is superior to CT images in depicting a number of underlying pathologic disorders. Thus, there will be a greater neuropsychology-brain imaging interface as brain-behavior relationships are pursued further. With these factors in mind, this chapter has been organized to present an overview of radiologic features of CT and MRI across the major pathologic categories and the relationship of such findings with neuropsychologic assessment. Normal brain anatomy will be presented first, followed by a review of the major pathologic categories of childhood neurologic disease and disorder. For further details concerning the techniques behind CT and MRI, see Buonanno (1984), Laffey, Mitchell, Teplick, and Haskin (1976), Laffey, Oaks, Swami, Teplick, and Haskin (1976), New and Scott (1975), and Oldendorf (1980).
247
248
CHAPTER 14
FIGURE 1. (Left) MRI scan views in the sagittal plane; (right) anatomic sections taken at a similar plane. Note the clarity of the MRI sections and the precise anatomic detail that can be achieved with this imaging technique.
RADIOLOGICAL TECHNIQUES IN NEUROPSYCHOLOGICAL ASSESSMENT
Normal Anatomy Early in CT technology, out Of convention and standardization, increasing shades of white corresponded with increased tissue density and increasing shades of darkness corresponded with decreased tissue density. According! y, as can be seen in the CT scans in Figure 2, the bone is white and the ventricular system is dark, representing the range of density seen in the cerebrum/cranial vault. Outside of the brain and embedded in sinus cavities of the skull are air pockets and these show an even Jesser degree of density (i.e., the darkest area seen represents the density of air on an image; see Figure 2). As gray matter is comprised of densely compacted cell bodies, the outer rim or mantle of the brain and internal nuclei have a lighter appearance (see Figure 2). In contrast, white matter structures, which are less dense and typically represent white matter myelinated pathways, have a darker appearance (see Figure 2). A key to interpretation of images is to examine for the normal distribution of white versus gray matter and the general symmetry of the brain as depicted in Figure 2. Sites of abnormal densities (whether hypo- or hyperdense) as well as marked asymmetries of brain development, typically signify underlying pathology. The nature and type of cerebral pathology seen with the major neurologic disorders of childhood will be presented in the following sections.
cr
249
such as progressive leukodystrophy (see Figure 4) and Huntington's chorea (see Figure 5), the degenerative pattern may appear initially to be focal.
Neuropsychologic Findings Although the correspondence between cortical degeneration seen on the CT and neuropsychologic outcome is not a one-to-one relationship (see Bigler, Hubler, Cullum, & Turkheimer, 1985), there may be some relationship between areas of more focal cerebral atrophy and corresponding neuropsychologic deficits (e.g., left perisylvian atrophy may be associated with a greater number of language symptoms). However, with degenerative diseases, there will typically be nonspecific findings of neuropsychologic impairment including decreasing intellectual ability, diminishing motor function and capacity, and a variety of problems in higher-order language and integrative sensory-percept ual-motor functioning. It should also be noted that neuropsychologic findings appear to be more sensitive to the progression of·degenerative diseases than are many CT or MRI techniques (Bigler, 1988). Typically, detectable changes in progressive atrophy in degenerative disorders as seen by scan are met with marked changes in neuropsychologic functioning.
cr
Neoplasms Degenerative Syndromes Neuroimaging Findings
Neuroimaging Findings Most tumors of the central nervous system can be readily detected by a variety of neuroimaging techniques. With CT scanning there is typically a noticeable density difference in and around the tumor site as well as some displacement of surrounding structures (see Figures 6-10). Tumors can be characterized as intrinsic (deriving from brain cells, typically glial in nature) or extrinsic (originating in bone or meninges). Typically, intrinsic tumors have a more destructive effect. Some tumors such as metastasizing neoplasms may have multiple sites.
CT imaging of the brain in the normal individual, and particularly in children, shows either no sulcal cleft or only a slight indentation (see Figure 2). This is due to the closeness of the gyri in the normal state, and in the absence of any significant sulcal distance a density change cannot be detected, and thus the image shows no density difference across the surface of the brain. However, in degenerative diseases affecting cortical neurons; there is a loss of neuronal density resulting in gyrai shrinking and sulcal cleft widening, which can be readily visualized in Neuropsychologic Findings the CT image (see Figure 3). With cortical degeneration, there is an actual volume loss in the amount of The nature and type of neuropsychologic deficit neural tissue and a frequent secondary effect of this is associated with brain tumor depend on the location, an enlargement of the ventricular system (hydro- size, and type of tumor. As depicted in Figure 6, the cephalus ex vacuo). With progressive degenerative glioblastoma multiforme (grade IV astrocytoma) diseases, the cortical atrophy pattern is typically dif- may develop into a very large tumor that will produce fuse and generalized. However, in some disorders, widespread deficits and frequently hemisyndromes.
250
CHAPTER 14
S#l(~IUM 0' CQIIIII'V$
CALLOSUM
fiGURE 2. Cf scan views in the horizontal plane with corresponding diagrammatic representation of major anatomic structures. Note the symmetry present throughout the major cerebral strucrures.
l"tTE,.HE-..[$#HfRIC
.-t----- !....:1..-'lt----- - - FISSUAE
'OSTt:RIOA FRO"tfA4.
~--~--T------- COAn•
FIGURE 2.
(Continued)
252
CHAPTER 14
FIGURE 3. Cortical atrophy in a case of progressive degeneration. Note the excessive prominence of the sulci, which is a sign of gyral shrinkage.
In addition to the direct effects of the tumor on brain tissue, secondary effects due to cerebral edema and vascular compression as well as radiation and surgical effects may produce widespread damage (see Figure 7).
Vascular Disorders Neuroimaging Findings The majority of childhood vascular disorders are associated with congenital anomalies or secondary to some disease or traumatic process. The common vascular disorders of adulthood (e.g., cerebral arteriosclerosis, multi-infarct dementia) are only rarely seen in childhood. The most common anomalies are in the form of aneurysms and arteriovenous malformations (AVMs). Aneurysms and A VMs have a characteristic pattern on cr, particularly following the injection of contrast medium (a dye that assists in visualizing the vascular components of the brain; see Figure 11). A cerebrovascular accident may take the form of hemorrhage or occlusion. The pathology that may result from this is an area of focal tissue loss associated with areas subserved by the affected vascular system and frequently secondary effects due to the edema produced by the hemorrhage or infarction (see Figure 12). Hemorrhage is not an
FIGURE 4. cr scan of focal white matter degeneration (leukodystrophy) in a 9-year-old male child. The patient presented initially with "learning problems," but rapid progression ensued. At the time of the Cf scan, pronounced deficits in all areas of higher cortical functioning were present.
uncommon sequela in (closed head injury) CHI and will be discussed elsewhere. Infectious disease may also produce occlusion due to the space-occupying effects of the infection and/ or compression secondary to edema .. There is an increased risk of cerebral hemorrhage with premature delivery and this may
FIGURE 5. cr scan in a patient with Huntington's chorea. Note the ventricular dilation of the anterior horns, which represents degeneration of the basal ganglia, and the presence of scattered sulcal enlargement indicative of cortical atrophy.
b
. ·...... I
..
• • '
FIGURE 6. (a) cr scan of a glioblastoma multiforme (grade IV astrocytoma) of the right hemisphere. Note the extent of the tumor and its spread into the left hemisphere. (b) This patient had classic features of a right hemisphere syndrome with left side neglect as depicted by her copying (and absence of the left aspect) and positional placement of the Bender Visual Motor Gestalt figures.
FIGURE 7. cr scan of a 9-year-old female who had an astrocytoma removed from the mid-right temporal lobe. However, she has had to undergo extensive chemotherapy and radiation therapy. The cr shows generalized cortical atrophy and density loss in white matter regions. Such effects represent the secondary iatrogenic effects of the cancer treatment.
FIGURE 8. MRI scan in the coronal position depicting the presence of cerebellopontine angle tumor (dark, high-density mass about the size of a golf ball located in the cerebellum compressing the brain stem). Note the anatomic detail that can be visualized with MRI. Note also the distortion present in surrounding brain structures as a result of the expanding tumor.
254
CHAPTER 14
FIGURE 9. CT reveals an area of increased density in the posterior left hemisphere, which turned out to be an ependymoma. This is a tumor that develops from the lining of the ventricle (ependyma) and occurs more frequently in children and adolescents than adults. This child was thought to have developmental dyslexia, but began having seizures, which prompted CT evaluation and the discovery of the tumor.
FIGURE 10. MRI scan showing the location of a cerebral neoplasm (oligodendroglioma) in the left temporal lobe in a 7-year-old child. A partial left temporal lobectomy was undertaken to excise the tumor. Presurgery, the child's WISC-R scores in intellectual functioning were: VIQ = 87, PIQ = 93, FSIQ = 89. One year postsurgery, they were: VIQ =54, PIQ = 73, FSIQ = 61. Additionally, a marked receptive aphasia, ADD, and hyperactivity resulted from the left temporal lobe damage.
take the form of an intraventricular bleed, which may produce widespread damage.
Neuropsychologic Findings With childhood cerebrovascular diseases in which some anomaly is present coincidental with brain development, the tissue around such an AVM or aneurysm may be damaged in development, ancf there may be a functional realignment or reorganization (see Bigler & Naugle, 1984) by intact surrounding or homologous areas of the brain (see Figure 11). Thus, although focal damage may be seen on the cr' if this damage was due to congenital vascular anomaly or accident, damage in this area may not fully correspond with what would be expected. Thus, from the neuropsychologic perspective, one has to be careful in such cases because of a tendency to overestimate the damaging effects of such an early lesion by the degree of damage seen on the CT scan. In cases of vascular damage where a preexisting anomaly was not present, the correspondence between areas of damage and neuropsychologic outcom~ is more clear (see Figure 12). The areas of damage and their corresponding neuropsychologic features are presented in Figure 13.
FIGURE 11. CT scan of a congenital AVM in the posterior left hemisphere. The AVM has been "enhanced" using a dye procedure. Note the surrounding tissue loss seen as a decreased (i.e., dark) density. There is a bone defect present that represents an old surgical incision following a spontaneous rupture and subsequent surgical repair. This case also demonstrates the cerebral plasticity that may occur from congenital vascular disorders in that even though the entire posterior left hemisphere has been affected, the patient displayed only minimal language impairment and that mainly in the area of reading. Otherwise, the patient had normal verbal intelligence (VIQ = 103) and language functioning.
RADIOLOGICAL TECHNIQUES IN NEUROPSYCHOLOGICAL ASSESSMENT
255
FIGURE 12. The cr on the left depicts the acute effects of an intracerebral hemorrhage in the right frontal-temporal region secondary to the rupture of a congenital aneurysm. The high-density (white) mass represents coagulated blood surrounded by edema (dark area). Also note the compression of the ventricular system in the right hemisphere and the distortion across mid-line. The cr on the right shows the chronic effects of such a hemorrhage. Note a large area of infarction (dark area) in the anterior temporal-posterior frontal region. Also note the increased size of the right anterior hom of the lateral ventricular system. This is a sign of tissue loss in the adjacent frontal region.
Cranial Cerebral Trauma Neuroimaging Findings Cranial cerebral trauma results in numerous damaging effects to the brain including contusions, white matter shearing, traumatic infarction, edema, hematoma, and subsequent compression effects, as well as a variety of biochemical derangements (Kingston, 1985). CT scanning quite accurately depicts the presence of most of these effects (see Figures 1418). However, the relationship between initial CT findings and outcome is quite specious and nonpredictive. Thus, for determining the stable effects of CHI on cognitive functioning, the CT or MRI scan that is done some 3 or more months following brain injury will correspond best with the static, stable residual lesions seen with cerebral traumas. Impact contusions, the area of previous hemorrhage, or localized edema may produce focal damaging effects (see Figure 18). Severe edema may also produce
widespread tissue loss through compression and generalized infarction effects (see Figure 15). Also in cases of severe CHI, even without significant hemorrhage, a diffuse atrophy may occur as a result of cortical contusion effects, white matter shearing, and subsequent neuronal loss (see Figure 17).
Neuropsychologic Findings Neuropsychologic sequelae associated with cerebral trauma are related to both the focal as well as diffuse findings on CT. It should be kept in mind that the postmortem studies have demonstrated that even in mild-moderate CHI cases, some degree of neuronal derangement may occur at the microscopic level that may not be detected by CT (Kingston, 1985). Thus, the correspondence between CT findings and neuropsychologic outcome is not a one-toone relationship, yet the findings offocal atrophy or encephalomalacia do tend to correspond more specifically with neuropsychologic outcome.
256
CHAPTER 14 FunctiOn
Antenor Cerebral
M1ddle Cerebral
Pos.tenor Cerebral
IIIIotor
Contralateral weakness or paralySIS. typically most affecting the distal lower extremity. Rapid ahernating movements may be impaired or an inability to maintain them.
Contralateral paralysis w1th face and upper extrem1ty more affected than lower.
Typically motor involvement not present except when proximal artenes are affected; then hemiba.llismus may result be· cause of subthalamic nucleus lesion. cranial nerve palsy. or hemiparesis owing to involvement of the corticospinal tract as it passes down the brain stem.
Sensory
Absent or mild contralateral tactile loss.
Depending on extent of involvement. there may be contralateral auditory, visual. and tactile sensory disturbance.
If unilateral, then contralateral hemianopsia typically develops. Cortical blindness occurs if bilateral. Proximal occlusion may affect the thalamus. producing sensory disturbance. pain, or both.
Language
Impaired articulation or disturbance in motor inertia may be present.
With left hemisphere involvement may develop a Broca's, Wernicke's, global, or conduction aphasia depending on site and extent of involvement.
If involves the left hemisphere. alexia without agraphia or other aphasic symptoms may
Praxis
If anterior corpus callosum is affected, there may be left arm apraxis.
Apraxia may occur with lesions in either hemisphere. Constructional apraxia (in the absence of aphasia) and dressing apraxia are common with right hemisphere involvement.
Constructional apraxia with right hemisphere involvement may occur.
Spatial-Perceptual
Typically not affected.
If right hemisphere is affected. varying degrees of impairment may be present.
If right hemisphere is affected. primarily visuospatial functions will be impaired.
Memory
Some disturbance in new memory may be present.
With left hemisphere involvement greater tendency for dis· turbance in verbal memory: with right hemisphere greater tendency for v1sual memory disturbance.
Global amnesia may occur that is either transient or permanent. Permanent short-term memory deficits may also occur.
Gnosis
Infrequently may display features of anosognosia (failure to appreciate loss of function).
With right hemisphere involvement. may have greater tendency to develop topographagnosia (loss of direction) as well as anosognosia.
Typically not affected.
Behavior
Frontal lobe syndrome may develop with right hemisphere. prosody may be affected.
Impaired prosody if right hemi· sphere involved.
No particular behavioral syndrome is characteristic.
FIGURE 13.
occur.
cr depiction of functional deficit associated with locus of cerebral infarction.
FIGURE 14. Posttraumatic effects secondary to gunshot wound to the posterior left cerebral hemisphere. Generalized encephalomalacia and infarction are present in the entire posterior pole of the left cerebral hemisphere including the entire left occipital lobe and posterior aspects of both the temporal and parietal lobes. Neuropsychologically, the patient had the expected features of a right visual field defect, dyslexia, and deficits in verbal comprehension.
FIGURE IS. Pronounced ventricular dilation (ventriculomegaly) secondary to diffuse cerebral edema produced by cerebral trauma. The top two scans depict diffuse edema as demonstrated by reduced ventricular size (Cf taken 2 days posttrauma). The bottom two scans were taken 2 months posttrauma and show severe dilation of the entire ventricular system as well as infarction in the left frontal region.
FIGURE 16. Self-inflicted gunshot wound to the frontal region. Note the point of entry in the right frontal area with the bullet trajectory passing obliquely from right to left. In this patient, a classic frontal lobe syndrome evolved with features of marked impulsivity, emotional lability, and traumatic dementia (WISC-R results: VIQ = 73, PIQ =57, FSIQ = 63).
258
CHAPTER 14
Infectious Disorders Neuroimaging Findings
FIGURE 17. Cortical atrophy as a result of severe closed head injury. Note the prominence of cortical sulci. Such findings are typically associated with some degree of nonspecific cognitive dysfunction. This patient was an 18-year-old college freshman at the time of the injury. The Cf was taken 2 years postinjury and neuropsychological studies indicated the expected nonspecific dysfunction. Intellectual scores on the WAIS-R were: VIQ = 86, PIQ = 81, FSIQ = 82. Based on premorbidestimates these results support a reduction of 15-20 points.
The major problem with infection appears to center on whether the damaging effects are focal versus generalized. Certain types of infectious processes (i.e., abscesses) may produce focal cerebral effects due to the direct invasion of the abscess and focal effects of edema. This typically is well visualized on CT as an increase in density associated with a mass effect during the purulent stage and focal, necrotic effects seen in the chronic. With meningitis and encephalitis, the damaging effects tend to be dependent upon how advanced the infection develops and whether it is localized or generalized. The most damaging effects, whether it be in meningitis or encephalitis, occur with excessive cerebral swelling, which may result in vascular occlusion and parenchymal compression (see Figure 19).
Neuropsychologic Findings Neuropsychologic sequelae associated with cerebral infection depend upon whether the effects are focal or generalized and the effects of cerebral swelling. The outline presented under Vascular Disorders
FIGURE Ill. Local frontal pole atrophy in a case of traumatic encephalopathy secondary to closed head injury. The dark area that is present across the bifrontal region represents cortical atrophy. Note also the greater prominence of the sylvian fissure bilaterally. This patient was a 16-year-old honor student at the time of the accident. After the head injury, a traumatic change in personality and intellect evolved (WISC-R results 9 months postinjury: VIQ = 87, PIQ = 78, FSIQ = 81).
RADIOLOGICAL TECHNIQUES IN NEUROPSYCHOLOGICAL ASSESSMENT
holds true with infectious disorders in terms of focal effects. When the infectious process has been generalized and diffuse, then the neuropsychologic impairment is correspondingly nonspecific.
Anoxia Neuroimaging Findings In terms of neuropsychologic assessment, CT findings obtained approximately 3 or more months following the acute anoxic injury are necessary to establish the relationship between chronic structural effects of this type of brain injury and cognitive impairment. Typically, anoxic injury results in generalized, nonspecific damage that is typically depicted on the CT image in the form of generalized cortical atrophy and secondary ventricular dilation (see Figure 20).
Neuropsychologic Findings
259
correspond with what would be expected given the physical examination findings (see Figure 21). In mariy cases of congenital hemiplegia, the loss of neuronal tissue corresponding with the contralateral motor area is evident, but frequently, there is good compensatory recovery of function by the· intact hemisphere. Thus, one may overpredict the sequelae given the size and extent of a lesion seen on CT examination in the patient with congenital hemiplegia syndrome. This underscores the importance of neuropsychologic testing in establishing the level of functioning.
Hydrocephalus On CT scanning the common feature of hydrocephalus is the enlargement of the ventricular system (see Figure 22). The key in interpreting the neuropsychologic effects of hydrocephalus tends to be associated with the degree of intactness of cortical tissue. However, there are now several cases in the literature of individuals with relatively normal intelligence despite having severe hydrocephalus. Therefore, clinical interpretation based only on CT findings needs to be done with caution.
The typical neuropsychologic findings in children with anoxic brain damage usually reflect some degree of generalized impairment. Accordingly, intellect and cognitive abilities are usually diminished Cortical Dysplasias and may actually reflect marked declines in functioning. Language, motor, and sensory deficits may also In cortical dysplasia, there is a failure of neube marked. Although CT scan abnormalities usually ronal growth or aberrant development. A host of abstabilize by approximately 3-6 months after the normalities can be visualized on the CT, but the most anoxic insult, cognitive improvement, particularly in common tends to be some type of asymmetry in corchildren, may continue for up to 2 or more years after tical development or a significant density difference the injury. or a gyral pattern irregularity (see Figure 23).
Congenital Anomalies/ Developmental Disorders Because of the complexity of the nervous system and its vulnerability during embryogenesis, numerous congenital malformations exist (Behan & Geschwind, 1985). The voluminous number of congenital anomalies far exceed the scope of this chapter, but the more common cerebral abnormalities of development are those seen in infantile hemiplegia, hydrocephalus, and dysplasic cortical development. Idiopathic mental retardation syndromes may also be included in this category.
Infantile Hemiplegia Frequently with infantile hemiplegia, there is an area of focal damage seen on the CT scan that does
Mental Retardation In Down's syndrome, CT scan analysis typically demonstrates frontal and temporal lobe regions that are evidently small (see Figure 24). The remainder of the gross anatomy is typically normal in appearance. In cases of idiopathic mental retardation, frequently no specific abnormality may be evident (see Figure 25).
Autism/Schizophrenia Some children with autism · or schizophrenia may display CT scan abnormalities, typically in the form of ventricular dilation and/or presence of cortical atrophy (see Figure 26). These anatomic irregularities do not appear to represent any specific diagnostic category for these disorders, but it does appear that the greater the degree of structural abnormality
260
CHAPTER 14
FIGURE 19. Acute-stage encephalitis is depicted in the cr films on the left. Note the lack of definition in the ventricular system, which is due to massive cerebral edema. Chronic postencephalitis effects are shown in the cr films on the right with evidence of generalized atrophy and ventricular enlargement. Prior to the brain infection, this 13-year-old male was in a regular classroom setting maintaining an average to aboveaverage GPA and had no history of behavior disorder. However. following this brain infection, the patient developed pronounced learning and behavior deficits. Intellectual studies indicated the following WISC-R results: VIQ = 62, PIQ = 71, FSIQ = 64. Behaviorally, the child had prominent attentional deficits associated with poor impulse control.
RADIOLOGICAL TECHNIQUES IN NEUROPSYCHOL OGICAL ASSESSMENT
FIGURE 20. Generalized cortical atrophy and ventricular dilation secondary to anoxic brain injury.
261
FIGURE 22. Severe hydrocephalus in a 6-year-old child with spina bifida. Note the marked destruction of the cortical mantle, particularly in posterior regions. Although mental retardation typically accompanies such cases of hydrocephalus, in some children the mental retardation is not in proportion with what would be expected given the degree of hydrocephalus (see Bigler & Naugle, 1984). In this case, the child's Weschler Preschool and Primary Scales of Intelligence scores were as follows: VIQ = 75, PIQ = 66, FSIQ = 68. Also, despite essentially no remaining vital tissue in the occipital region, the child could still visually process information. This demonstrates the marked degree of cerebral plasticity that may accompany some early brain insults.
present, the greater the degree of cognitive impairment and the more resistant are the symptoms to any treatment regimen (see Bigler, 1984). Childhood Epilepsy
In some cases of epilepsy, the structural basis may be visualized by CT scanning (see figure 27). These are typically areas of focal abnormality and likewise may correspond to focal neuropsychologic findings.
Learning Disability FIGURE 21. cr scan of a patient with idiopathic congenital hemiplegic syndrome. As is readily seen, a large cystic area is present adjacent to and communicating with the left lateral ventricular system. This represents focal damage in the mid-frontal region, consistent with contralateral paralysis.
Although cr irregularities are not universally seen in learning-disabled children (see Denckla, LeMay, & Chapman, 1985), nor are there any CT abnormalities that are diagnostic of learning disability, there are cases in which a structural basis appears to be present. These typically take the fonn
262
CHAPTER 14
FIGURE 23. Numerous congenital anomalies are present in this
cr scan of a 10-year-old female who presented with a history of idiopathic mental retardation (WISC-R: VIQ = 59, PIQ = 55, FSIQ = 53). No physical stigmata were present. cr demonstrates an agenesis of the corpus callosum, hemispheric asymmetry·, and dilated posterior horns of the lateral ventricles, particularly on the left. Behaviorally, the child was hyperactive, impulsive, and with shortened attention span in addition to the mild level of mental retardation present.
FIGURE 25. CT scan in a patient with idiopathic mental retardation. The scan is generally within normal limits, although the ventricular system may be considered somewhat generous for a 17year-old. Despite the profound level of retardation, no gross abnormality is noted in general brain morphology.
of cerebral asymmetries (see Figure 28) or more specific pathology in the posterior association cortical areas (see Figure 29).
Birth Injury A variety of cerebral damaging events may accompany a traumatic or problematic birth. These deficits may vary from focal to generalized effects (see Figure 30).
FIGURE 24. cr scan of an adult (24-year-old) with Down's syndrome. The irregularity noted on cr scanning in the Down's syndrome patient is the diminished size of the frontal and temporal lobes. (Note the short distance from the rostral aspect of the anterior horns to the tip of the frontal poles, in comparison to the normal configuration shown in Figure 2.)
FIGURE 26. cr scan of an 11 -year-old autistic child. Note the general enlargement of the ventricular system with particular enlargement on the left. Such findings suggest that significant structural abnormalities occur in some children with autism.
RADIOLOGICAL TECHNIQUES IN NEUROPSYCHOLOGICAL ASSESSMENT
263
cr
FIGURE 27. scan in a 14-year-old female with temporal lobe epilepsy. Note the area of focal tissue loss in the right anterior temporal region. This child had a history of significant learning disability.
FIGURE 29. Posterior hom dilation in a young adult patient with a history of pronounced dyslexia. Such findings implicate probable dysplastic development in parietal and occipital areas typically considered crucial for aspects of visual-verbal processing.
cr
FIGURE 30. scan in a child who had undergone a difficult labor and who suffered respiratory and cardiac failure. The shows multiple areas of infarction, namely in the occipital region, bilaterally, agd in the left frontal region. The child presented with hyperactivity, poor attention, mild mental retardation, and cortical blindness.
cr
FIGURE 28. Cerebral asymmetry that is sometimes observed in certain developmental and learning disorders. Note that the left cerebral hemisphere is distinctly smaller than the right. Also note the ventricular asymmetry.
264
CHAPTER 14
References Behan, P., & Geschwind, N. ( 1985). Dyslexia, congenital anomalies and immune disorders: The role of the fetal environment. Annals of the New York Academy of Sciences, 457, 13-18. Bigler, E. D. (1984). Diagnostic clinical neuropsychology. Austin: University of Texas Press. Bigler, E. D. (1988). Neuropsychological and identification in dementia. In H. A. Whitaker (Ed.), Neuropsychological Studies of Nonjocal Brain Injury. New York: SpringerVerlag. Bigler, E. D., Hubler, D. W., Cullum, C. M., & Turkheimer, E. ( 1985). Intellectual and memory impairment in dementia: Cf volume correlations. Journal of Nervous and Mental Disease, 173, 347-352. Bigler, E. D., & Naugle, R. I. (1984). Case studies in cerebral plasticity. International Journal of Clinical Neuropsychology, 7, 12-23. Buonanno, F. S. (1984). Neurologic clinics (Vol. 2). Philadelphia: Saunders.
cr
Denckla, M. B., LeMay, M., &Chapman, C. A. (1985). Few CT scan abnormalities found even in neurologically impaired learning disabled children. Journal of Learning Disabilities,
/8, 132-135. Kertesz, A. (1983). Localization in neuropsychology. New York: Oxford University Press. Kingston, W. J. (1985). Head injury. Seminars in Neurology. 5, 197-270. Laffey, P. D., Mitchell, M. H., Teplick, J. G., & Haskin, M. E. (1976). Computerized tomography in clinical pediatrics. Philadelphia: Medical Directions. Laffey, P. D., Oaks, W. W., Swami, R. K., Teplick, J. G., & Haskin, M. E. (1976). Computerized tomography in clinical medicine. Philadelphia: Medical Directions. Lezak, M. (1983). Neuropsychological assessment. New York: Oxford University Press. New, P. F. J., & Scott, W. R. (1975). Computed tomography of the brain and orbit. Baltimore: Williams & Wilkins. Oldendorf, W. H. (1980). TheQuestforanlmageofBrain. New York: Raven Press.
15 Psychophysiological Evaluation of Children's Neuropsychological Disorders THALIA HARMONY
The objective of this chapter is to discuss those procedures that explore psychological processes by measuring different physiological variates. Study of brain electrical activity in humans allows the exploration of two fundamental levels of function, a very basic one, which gives information about the functional and anatomical integrity of the nervous system, and a second one, which explores higher cognitive activity. At the first level, it is possible to detect such localized lesions as an epileptogenic focus or functional alterations of the sensory pathways. Here, accumulated knowledge of brain electrical activity is strong enough to have practical clinical applications leading to differential diagnosis and differential treatment. In this chapter, studies of spontaneous EEG activity and of evoked sensory potentials are reviewed. With regard to spontaneous EEG activity, two main procedures will be analyzed: routine EEG visual interpretation and frequency analysis of the EEG. Routine EEG visual interpretation is a powerful tool for the detection of paroxysmal activity and of special types of patterns described in children. Frequency analysis of the EEG has been shown to be a precise method for evaluation of EEG maturation and EEG background activity. In neuropsychological disorders it is very important to emphasize the following paradox: these disorders are studied using new procedures the intention of which is to obtain better methods for the diagnosis of such disorders, owing to the fact that the
THALIA HARMONY • Neurosciences Research Program, Iztacala School, National Autonomous University of Mexico, 54030 Tlalnepantla, Mexico City, Mexico.
clinical diagnostic procedure is not accurate in many cases. Those investigators attempting to develop new procedures of evaluation with regard to patients with well-defined neurological lesions have a fundamental reference based in anatomopathological diagnosis and surgical observation. However, in trying to develop procedures for the study of the neuropsychological disorders of the children, clinical diagnosis (which we know is not accurate in many cases) is typically a very general symptomatic diagnosis. Many times we have no globally accepted reference at all and thus do not know if the new procedure is any better than the usual diagnostic method. When developing new diagnostic procedures, a specific series of steps must be followed (Harmony, 1984a). If it is an electrophysiological procedure, we first identify and extract those electrophysiological features that are most directly related to the phenomenon we want to evaluate. This may be obtained by usual statistical procedures. However, diagnostic classification requires more than differentiating at a statistically significant level. The second step is the establishment of a decision rule for assigning a given subject to a diagnostic set. These sets may be known a priori or as a result of data analysis. A significant problem in this process is that although we are looking for procedures to provide more objective and more accurate diagnoses, the only means we have to define the groups among whom we look for electrophysiological differences are traditional procedures known to be inadequate. Our answer to this problem has been to examine the structure of the data derived from our diagnostic procedures to see if there are any statistically homogeneous groups and to detect deviant values or outliers. If it is possible to relate new groups to some clinical characteristic, the prob265
266
CHAPTER 15
lem may be solved. In grouping analyses, many variables must be considered. In the case of children, age is the most important independent variable, and agematched groups do not solve the problem. Satterfield and Braley (1977) were able to show that some measures of auditory evoked responses have a trend of development in normal children that is different (opposite) from that in hyperkinetic children, and when the total samples matched by age were compared no significant differences were found. Age effect may be removed by regression procedures as shown later. Sex is also an important variable. Once the homogeneous groups have been determined, it is possible to proceed to the establishment of a decision rule. If homogeneity is observed only for normal children, then what proceeds is the creation of norms. Cross-validation of norms with new subjects will define the nature of the procedure. If many groups appear and they correspond to the original groups, then a priori criteria and the new procedure are equivalent. If many groups appear and have no relation to the original groups, then it is necessary to analyze other converging variables such as neuropsychological tests, pharmacological manipulations, and the like. The state of the art in psychophysiological diagnosis varies greatly depending on the procedure. For variates derived from frequency analysis of the EEG and for sensory evoked responses, norms exist and deviant values can be identified. In relation to ERPs, only statistically significant differences between healthy children and children with different types of neuropsychological disorders are available. In this chapter, consistent results obtained in homogeneous ·samples of children will be presented. Reviews of the field frequently mix papers on different topics under one clinical item and therefore many contradictory results emerge. We discuss herein only those papers in which selection of the sample was well explained and it was possible to find homogeneity in relation to a sign or a deficit.
EEG Maturation and Abnormal EEG Activity Visual Interpretation EEG background activity is classified in four frequency bands: delta(< 3.5 Hz), theta (4-7Hz), alpha (8-13Hz), and beta(> 14Hz). Delta rhythm is the dominant activity during the first 2 years of life but then tends to disappear. Delta activity detected by
visual inspection in the awake adult is abnormal. Theta rhythm is the most frequent activity during childhood. In children from 1 to 5 years old, theta rhythm of high amplitude is observed in posterior regions and is the dominant activity from age 4 to 6 years in central, temporal, and parieto-occipital regions. Alpha rhythm is most prominent in posterior regions under conditions of relaxation, with eyes closed in awake adults. Infants between I and 2 years old show brief periods of activity within the alpha band. By age 5 years, a child's alpha rhythm becomes clearly discernible. From age 6 to 9, the EEG shows a mixture of frequencies within the alpha range and lower. From age 10 to 13, the alpha rhythm is almost stabilized, with characteristics similar to those observed in the adult. Beta activity is observed in 25% of awake children. Some special rhythms appear in normal children, although they are more frequently seen in children with behavioral or cognitive alterations (Petersen, Sellden, & Eeg-Oiofsson, 1975). Among the rhythms that have these characteristics are the following: 1. Slow posterior waves (Aird & Gastaut, 1959), slow fused transients, or polyphasic potentials (Petersen & Eeg-01ofsson, 1971) are considered normal by several authors, although Cohn and Nardini ( 1958) reported this pattern to be related to personality disorders or aggressive social behavior, and to disturbances in reading, writing, and adaptive behavior (Cohn, 1958, 1961). 2. Diffuse theta activity in drowsiness (hypnagogic hypersynchrony or drowsy waves; Kellaway & Fox, 1952) declines rapidly from a maximal value at age 1 year to age 89 years, at which it appears only rarely. 3. Diffuse rhythmic 4- to 5-Hz activity with parieto-occipital maximum, is considered normal in young children, although it has been associated with febrile convulsions and may be considered a sign of predisposition to convulsions (Doose, Gerken, & Volzke, 1968). 4. 2.5- to 4.5-Hz slow posterior parieto-temporal-occipital rhythm is typical of childhood, but after age 4 is rare. At high amplitude, this type of slow posterior rhythm has been correlated to paroxysmal activity at rest (Eeg-Oiofsson, Petersen, & Sellden, 1971). Paroxysmal activity is defined as a group of waves that appear and disappear abruptly and is clearly distinguished from background activity by
PSYCHOPHYSIOLOGICAL EVALUATION
different frequency, morphology, and amplitude. Spikes, multispike complexes, sharp waves, spike and slow wave complexes, and paroxysmal slow waves are recognized as the fundamental paroxysmal patterns. In normal children, such activity is rarely observed (1.5-6%; Petersen et al., 1975). The interpretation of Rolandic spikes has been controversial, for it has been observed that subjects with this type of focus have a higher incidence of seizures and cerebral lesions. Other important and controversial aspects of EEG interpretation in children are the effects of hyperventilation. In normal children, spike and sharp wave paroxysmal activity has been reported in 0.3% of cases by Petersen and Eeg-Olofsson (1971), 2.3% by Doose et al. (1968), and 0.5% by Gibbs, Gibbs, and Lennox (1943). These are very low rates, but many electroencephalographists consider the observation of paroxysmal activity only during hyperventilation to be normal. Results for mental retardation, minimal brain dysfunction, and specific learning disabilities showed a significantly higher proportion of abnormal EEGs than for normal children (Netchine & Lairy, 1975). The most frequent abnormalities are: high amount of slow EEG activity, positive spikes, and paroxysmal spike and wave activities (Hughes, 1971; Lezny, Provasnik, Jirasek, & Komarek, 1977; Murdoch, 1974; Becker, Velasco, Harmony, Marosi, & Landazuri (1987).
Frequency Analysis If the EEG is considered as a mathematical function, it is possible to express its decomposition in a series of sine and cosine waves by Fourier analysis. These sine and cosine waves are harmonically related, beginning with the fundamental frequency that has a period equal to the sample length (if EEG epochs of 2 sec are analyzed, the fundamental frequency will be 0.5 Hz with harmonics at 1.0, 1.5, 2.0, ... Hz). The component at each frequency can be represented as the sum of a sine and a cosine wave of that particular frequency. For each wave, three measures will explain it completely: the frequency, the amplitude, and the phase shift. If the sine and cosine components of a particular frequency are squared and added, the amount of power at that frequency is obtained. Thus, the phase relationships are lost and only the frequency components are obtained. The most frequent application of Fourier analysis has been the computation of the power spectrum and the calculation of the power for each frequency band, as shown in Figure l. Pascual, Valdes, and Alvarez
267
fogS .: ·. ~
...
• .
-~r.-
::
·.
• - --.
:
I
~
:e
a.
.----- -,
:~· .· . ·.;
L-~~::---:;,!;:-;=----'
3.5
7.5
12.5
Hz
19
FIGURE 1. Broadband analysis. Points represent the logarithm of the power spectrum for different frequencies. Bands delta, theta, alpha, and beta are shown. It is possible to observe that the approximation in some bands is not adequate.
(1985) developed an EEG model derived from frequency analysis. In this model two basic processes are identified: the xi process characterized by a spectral peak with its maximum located at zero frequency and with a slow decrease in power at increasing frequencies, and the alpha process characterized by a superimposed spectral peak on the xi process, which is usually centered in the classical frequency range circa 6-13 Hz; it may be absent, especially at lower ages and in frontal derivations. According to Pascual et al. (1985), the appearance of the alpha process, its growth in height and in peak frequency are characteristic of EEG maturation. Figure 2 presents the different measures derived from this model. Other models have been reviewed by Harmony (1984a).
Maturation Matousek and Petersen ( 1973) performed a frequency analysis of a large sample of people from 1 to
fogS
FIGURE 2. The xi-alpha model of the EEG. Points represent the same spectrum as in Figure I. The solid line shows the approximation by the model. Discontinuous lines illustrate the xi process (~) and the alpha process (a). Five measures derived from the model are also shown: amplitude (A) and semibandwidth (B) of the xi and alpha processes and position alpha (ll").
268
CHAPTER 15
21 years old. The analysis divided the EEG activity into six frequency bands: &(1.5-3.5 Hz), 6 (3.5-7.5 Hz), a 1 (7.5-9.5 Hz), a 2 (9.5-12.5 Hz), !3 1 (12.517.5 Hz), and !32 (17.5-25 Hz). The authors also obtained norms according to age by a multiple regression analysis of several band parameters in order to reach the highest possible correlation with age (Matousek & Petersen, 1971). These regression equations could then be applied to any other EEG and the hypothetical age calculated (Friberg, Matousek, & Petersen, 1980). In cross-validation, John, Ahn, Prichep, Trepetin, Brown, and Kaye (1980) computed linear regression equations for each EEG band and for each derivation in 306 healthy children from 6 to 16 years old and compared the results obtained with those previously described by Matousek and Petersen (1973). They found that the same equations described the EEG according to age in American and Swedish normal subjects. In a subsequent paper, Ahn, Prichep, John, Baird, Trepetin, and Kaye (1980) demonstrated that nonnally functioning children from Barbados have an EEG that may be accurately predicted by previously described equations for American children. In a rural population from Mexico, Harmony (1984b) observed that such equations adequately described the EEG in central and frontotemporal derivations but that in parieto-occipital regions the slope of the regression equations did not fit. As she had recorded all the children from a small town and many of them had histories of brain damage, she attributed such a finding to the presence of slight brain dysfunction in some of the subjects, which may produce less pronounced slopes or no correlation with age of the different EEG bands in parieto-occipital derivations. Alvarez, Valdes, and Pascual (1987) validated the equations of John et al. in a sample of 96 nonnal Cuban children. All of the above-mentioned results were obtained with the same bipolar derivations originally used by Matousek and Petersen: C3Cz, C4Cz, F7T3, F8T4, T3T5, T4T6, P301, and P402. However, the equations described by John et al. ( 1980) are useful only for children selected with strict criteria of ''normality. '' Harmony, Alvarez, Pascual, Ramos, Marosi, Dfaz de Le6n, Valdes, and Becker ( 1988) compared the slopes of the regression equations for delta, theta, alpha, and beta relative power in six groups of children having different economic and psychosocial characteristics with the slopes previously published by John et al. (1980). The groups studied were: (1) 96 cuban children selected with strict criteria of normality, (2) 28 children from a middle class school of Mexico City selected with strict criteria of normality, (3) 28 children from
a marginal urban area of Mexico City with clear socioeconomic and cultural handicaps, (4) 30 children from a marginal urban area of the city of Toluca in Mexico, (5) 55 children from Caracas who belonged to marginal and low socioeconomic classes, and (6) 26 children from a middle class school of Caracas. All children were healthy at the moment of the study, had a normal IQ, and attended school regularly without academic or behavioral problems. In groups (3), (4), and (5) children very frequently presented antecedents of risk factors. Those children who had grown up with adequate nutritional, sanitary, and environmental conditions showed the same slopes as children of the United States. Children nourished in poor socioeconomic and sanitary environments, who frequently had pathological personal antecedents of risk factors associated to brain damage, showed either a slow maturation of the EEG characterized by smaller slopes of theta relative power or showed a great variance of EEG parameters with these parameters having no relation to age. We do not know if these significant differences in EEG maturation have a functional correlate in relation to cognitive activity at present or in the future, or if such deviations are a trace of the event from which the brain has recuperated without any physiological meaning at the moment and in the future. We have moved a step forward in the neurometric approach demonstrating that healthy children that grew up in adequate environments in various countries have similar patterns of EEG maturation. However, there are other concurrent factors that should be taken into consideration. In a subsequent paper, it was reported that besides age, sex had an important effect in all bands in the posterior areas and in almost all derivations in the theta and alpha bands. Risk alone had no effect, but important interactions between risk and sex were observed in the alpha band (Dfaz de Le6n, Harmony, Marosi, Becker, and Alvarez, 1988). In the computerized evaluation of EEG maturation it has been very important to the introduction of the concepts of maturational lag and developmental deviation (John, Prichep, Ahn, Easton, Fridman, & Kaye, 1983). The age at which the observed spectrum is within nonnal limits can be considered the apparent physiological age of the subject. The difference between the apparent physiological age and the chronological age is defined as the maturational lag. If the observed values of the spectrum do not lie within normal limits at any age, then it is defined as displaying a developmental deviation. In other words, developmental deviation is present when the
EEG pattern is not similar to an EEG at any age. Valdes, Biscay,. Pascual, Jimenez, and Alvarez
PSYCHOPHYSIOLOGICAL EVALUATION
( 1985) introduced a more formal mathematical description of the developmental deviation. They defined the space of normal variation as the feature space spanned by normal individuals and the pathological space as the complement of the normal space. They measured the Mahalanobis distance for each subject within the normal space and within the pathological space. Concepts like these are very important from a theoretical point of view, because they try to discriminate between immaturity and abnormality, two aspects that have long been debated in an attempt to explain children's neuropsychological disorders. Clinical and related treatment implications are manifold.
Children's Neuropsychological Disorders Satterfield, Schell, Backs, and Hidaka (1984), in a cross-sectional and longitudinal study of hyperactive children, found that interactions between diagnostic group and age were highly significant for all EEG frequency bands. Such interactions occurred chiefly between the youngest group (6.8 years) and the middle (8.3 years) group. Younger hyperactive children had lower power in all bands than did normal controls. The middle and oldest hyperactive children had higher power in ~ bands than did controls. ~ differences in hyperkinetics have also been described by Callaway, Halliday, and Naylor (1983). Children with minimal brain damage (attention deficits, hyperkinesis, and learning disorders) showed less a attenuation during active performance of mental tasks (Fuller, 1977). In dyslexic children, although visual scanning of the EEG showed no detectable abnormalities, differences in left parieto-occipital spectral values made possible a correct classification of 87% of the children. Dyslexics had higher relative 6 power than did normals and greater coherence values within hemispheres during reading, whereas normal children showed greatest coherence between hemispheric homologous leads during reading (Sklar, Hanley, & Simmons, 1973; Hanley & Sklar, 1976). Abundance of 6 activity in the left hemisphere during several eyes-open conditions in dyslexics has been described by Rebert, Wexler, and Sproul (1978). John eta/. ( 1983) published results obtained for a large sample of normal and learning-disabled children following a strict evaluation and validation procedure of different variates derived from EEG frequency analysis. From a methodological point of view, several aspects were taken into account, but
269
only two will be emphasized: (I) Age regression equations were computed for univariate and multivariate data in half of the control group using the second half for replication. (2) From univariate and multivariate data, the z score using the normative regression equations was computed for each measurement of each child. This procedure ensures removal of the age effect and establishes a common metric of relative probability for each measurement across age. Children with specific learning disabilities showed greater 8 activity and lower a activity in parieto-occipital derivations than did normal subjects. A discriminant function was computed, and the independent replication of such function demonstrated that 85% of normals and 65% of learningdisabled children were adequately classified. In the sample of children from suburban areas of Mexico, in which all children had antecedents of risk factors, Diaz de Le6n, Harmony, Marosi, Landazuri, Becker, and Banuelos ( 1985) observed significant differences between learning-disabled children and control children, the former having more 6 and less a in parietal and occipital monopolar leads. Fein, Galin, Yingling, Johnstone, Davenport, and Heron (1986) studied a selected group of 34 boys (ages 9-13) with pure dyslexia and a sample of normal children matched by age and sex. They found that dyslexics had lower power in the high f3 band, but were unable to demonstrate lower a and greater 8 or 6 levels in those children. As they studied agematched groups, it is possible that they did not find significant differences in 8, 6, and a bands as have other authors because they did not remove the age effect. As Satterfield eta/. (1984) emphasized, if the developmental curves have different slopes in control and experimental groups, differential effects may be observed according to age, but if the whole sample is analyzed and compared, then significant differences disappear. Gasser, Mocks, Lenard, Bacher, and Verleger (1983) reported that spectral parameters differentiated between a group of 10- to 13-year-old mentally retarded children and a matched (sex, age, and social class) normal group. Larger 8 and 6 activities were observed in the mentally retarded. Alvarez, Ricardo, Valdes, and Pascual (1985) compared a group of 63 mentally retarded children (WISC IQ in the range 50-70) with an independent normal sample not used for the computation of EEG norms. All values were z scores to remove the age effect. Impressive differences were observed for broadband frequency analysis and for the xi-alpha model. Mentally retarded children were characterized by lower power in all bands compared to nor-
270
CHAPTER 15
mal children and higher 8 and 6 power. The most important differences were observed in e left-right coherence. Computation of the maturational lag and the distances in the normal and the pathological space between the independent normal sample and the mentally retarded one showed that a great discrimination may be obtained by the computation of the maturational lag. Figure 3 illustrates the ROC (receiver operating characteristics) curves for these three global measures in the mentally retarded. The x axis corresponds to the accumulated percentage of falsepositives, and they axis to the accumulated percentage of hits. For 2.4% false-positives, 80.95% of hits were observed in the maturational lag. Such results show how, on the basis of the EEG, it is possible to obtain accurate procedures for the evaluation of many neuropsychological disorders of childhood.
stimulus, but rather are associated with the subject's prior experience and current intentions and decisions and are more reflective of higher cognitive processes. Whereas exogenous components are always elicited by external stimuli, endogenous components are nonobligatory responses to stimuli. The same physical stimulus sometimes will and sometimes will not elicit the component depending on the experimental instructions. Moreover, endogenous components could be elicited in the absence of a stimulus, if such absence has an appropriate role in the subject's task (Sutton, Tueting, Zubin, & John, 1967). Endogenous components may be elicited with the same characteristics even if the stimuli belong to different sensory modalities but the role in the task is the same. Variations in the task produce variations in the endogenous components. Research on ERP has been directed toward identifying particular ERP components as markers of specific stages of information processing such as encoding, selecting, memorizing, Electrophysiological Assessment of decision making, and so on (Donchin, Ritter, & McCognitive Processing Callum, 1978; Hillyard & Kutas, 1983). Some of the most widely studied endogenous components are the Introduction Contingent Negative Variation (Walter, Cooper, Aldridge, McCallum, & Winter, 1964), the "processThe waveform of ERPs is complex and extends ing negativity" (Naatiinen & Michie, 1979) or several hundred milliseconds after stimulus presenta- "negative difference wave" (Hansen & Hillyard, tion. The ''late'' potentials correspond to a variety of 1980), P300 (Sutton et al., 1967), and N400 (Kutas processes that are invoked by the psychological de- & Hillyard, 1980). Before presenting the results obmands of the situation rather than evoked by the pre- tained for various disorders, we shall describe briefly sentation of the stimulus. These are the endogenous those endogenous components that have been studied components. Their characteristics are related only in children's neuropsychological problems. partially to the physical parameters of the evoking
Contingent Negative Variation (CNV) hits
ml ns.
.5:
ps
.5
f.p
FIGURE 3. ROC curves derived from the comparison of the independent normal sample and the group of mentally retarded children by broadband analysis of the EEG. The x axis shows the accumulated percentage values of false-positives (f.p) and they axis the accumulated percentage value of hits. ml, maturational lag; ns, the Mahalanobis distance in the normal space; ps, the Mahalonogis distance in the pathological space. (Courtesy of Alvarez et at.. 1985.)
In assessing CNV, a conditional signal (S I) is given, followed by a constant delay of l sec or more before the second or imperative stimulus (S2) to which the subject has to respond. After S 1, a slow negative potential in the interstimulus interval appears. ThisistheCNV. Walteretal. (1964)observed that the buildup of CNV coincided with the learning of stimulus-response association. The CNV thus appeared to represent a fronto-cortical system involved in the formation of sensorimotor expectancies. The relation of CNV amplitude to learning varies as a function of what is being learned. CNV amplitude increases during early acquisition; its duration and sustained amplitude depend upon the nature of the task. The tasks imposing the greatest cognitive demands appear to sustain the CNV for the longest period after initial acquisition. There is evidence suggesting that CNV amplitude is not simply propor-
PSYCHOPHYSIOLOGICAL EVALUATION
tional to task difficulty, and this has been attributed to the influence of stress and anxiety on the CNV. Peak CNV amplitude is not related to task difficulty; amplitude is in the same range whether for Pavlovian conditioning or cognitive learning. These results are compatible with either a motivation- or attentionbased model of the role of CNV in learning. The interest a subject has in a task might be unrelated to its difficulty or complexity at the start of learning. In later trials the CNV would tend to decrease in amplitude as the task became less engaging or demanding, i.e., as the interest and motivation of the subject waned. CNV has been associated with motivation, incentive, intentionality, expectancy, readiness, and intense concentration (Donald, 1980; Glanzmann & Froelich, 1984; Proulx & Picton, 1980; Tecce, Savignano-Bowman, &·Meinbresse, 1976). In a most general sense the CNV can be considered as a cortical change that occurs when an individual's behavior is directed toward a planned action in relation to a sequence of two or more events. The action can be an overt motor response, the inhibition of a motor response, or a decision. According to Marczynski (1978), the slow negative potentials are mediated by a concerted action of the cholinergic and catecholaminergic components of the ascending reticular activating system. The cholinergic component depolarizes neuronal dendrites, reduces the firing threshold of large populations of neurons in specific thalamic nuclei and cortex, and prolongs the discharge to increasing volleys. Simultaneously the catecholaminergic component appears to block the function of the GABAergic recurrent inhibitory circuits responsible for hyperpolarization of large populations of neurons. Somjen ( 1978) proposed that cholinergic input may activate the sources of potassium, which in tum produces depolarization of neuroglial cells, thus causing sustained extracellular currents.
Processing Negativity Selective attention to auditory stimuli elicits a broad negative ERP that begins 60 to 80 msec after the stimulus. This negative wave increases the amplitude of the NIOO wave (Hansen & Hillyard, 1980). For this reason the early attention effect was viewed as an augmentation of the evoked NIOO wave to attended channel stimuli (Hillyard, Hink, Schwent, & Picton, 1973). This attention-related component can be visualized and quantified as the negative "difference wave" of the ERP to stimuli in an attended channel minus the ERP to the same stimuli when they
271
are not attended (Hillyard & Kutas; 1983). The term channel has been used to refer to stimuli that share some simple cue characteristic such as position in space or tone. Because of the short latency of the so-called processing negativity, Hillyard et al. (1973) proposed that it was a sign of an early stage of selection that gives preferential access to stimuli belonging to an attended "channel." According to this view, stimuli in unattended channels are rejected after a rapid initial analysis of their channel cues by a hypothesized "filter" or "attenuator" mechanism, also known as "stimulus set" (Broadbent, 1971). Following this initial filtering, material in the attended channel is analyzed along more complex cognitive or semantic dimensions and is compared with stored information in memory. This subsequent mode of selection is also known as ''response set'' or "cognitive selection." Hillyard, Picton, and Regan ( 1978) associated response set selections with the P300 component. In patients with lesions of the prefrontal cortex, Knight, Hillyard, Woods, and Neville (1981) showed that although the cortex is not the primary generator of the prominent auditory Nl 00 wave or the auditory processing negativity, it modulates the generators of those ERP components. Nd and N I 00 are independent components: at a low rate of stimulation, Nd may begin after NIOO. Increasing stimulation rate produced processing negativity with shorter onset latencies while P300 latency is increased (Hansen & Hillyard, 1984).
P300 P300 is an endogenous positive wave with a latency of 300 msec or greater and is typically elicited by rare target stimuli in a detection task. In selective attention tasks, when the signal is correctly detected the ERP shows a large P300 component that does not appear when the targets are missed. P300 also may be elicited when the signal is absent if this is a correct report. That is, whether the decision is based on signal presence or absence, P300 waves are enlarged for more confident decisions and for Jess expected outcomes. Because the auditory NIOO and P300 components show parallel increases in amplitude and decreases in latency as a function of rated confidence of signal detections, but P300 is also sensitive to whether or not the signal was correctly identified, it has been suggested that processing negativity is highly correlated with "stimulus set" and P300 with "response set" (Hillyard et al., 1978). Thus, P300 seems to be related to stimulus evaluation processes
272
CHAPTER 15
and independent of response selection and execution processes. P300 can be elicited in a variety of experimental paradigms if the experimenter uses random alterations in stimulus parameters. However, P300 is not evoked solely by physical changes in stimulation: undetected targets in a vigilance task do not elicit P300, and predictable changes in stimulus parameters do not elicit P300. In other words, occasional changes in a physical aspect of an ongoing train of stimuli may or may not be associated with P300 depending on whether the subject knows in advance when the change will take place. It is the activation of a unique processor, a subroutine, with a specific function that is indicated by the appearance of P300. This activity may be invoked for a variety of reasons during the performance of different tasks but its role in the sequence of information processing activities invoked by the task is always the same (Donchin et al., 1978). Any stimulus, once identified and categorized, is thought to initiate two processes, one concerned with response selection, the other with what might be called ''context updating,'' which depends in tum on stimulus evaluation. If the unexpected happens, the model of the operating context must be revised. Donchin et al. (1978) hypothesized that this context revision process is manifested by P300. Johnson and Fedio (1984) considered that P300 amplitude is affected by three variables: probability of target stimuli, task information, and equivocation. The first two are independent and may be combined in a linear way. Both are modulated by equivocation, which is equal to the amount of missed information that occurs during the presentation of one stimulus. P300 has a broad centroparietal scalp distribution, which has led to inferences that it may be generated by a diffuse cortical source. However, a broad scalp distribution can also be produced by a subcortical source: Halgren, Squires, Wilson, Rohrbaugh, Babb, and Crandall (1980) proposed that P300 was generated in the hippocampus. However, Yingling and Hosobuchi (1984) observed that rare auditory stimuli that were detected by the subject were accompanied by a prominent P300 component at the scalp, and by a negative activity at the subcortical sites with the same latency as the scalp positivity. This is inconsistent with a hippocampal generator for P300 but is consistent with a generator in the thalamus or more dorsally located structures. Johnson and Fedio ( 1984) did not tind consistent asymmetries in the distribution of P300 that distinguish left from right temporal lobectomy in patients, proposing multiple P300 generators owing to the fact that the distribution of P300 has repeatedly been found to be task dependent. For example, unexpected novel stimuli in the
visual or auditory modality generate a frontally distributed P300 complex that has been related to the orienting reflex (Courchesne, 1977). Patients with prefrontal lesions showed no enhancement of the frontocentral P300 to unexpected novel stimuli, while the parietal P300 component to target stimuli was normal (Knight, 1984).
Developmental Changes of the Endogenous Components At age 7 years, some endogenous components have characteristics similar to those observed in adults; however, in general, those of children are characterized by larger peak latencies and longer duration of ERPs. This observation has been related to a delay in cognitive process in childhood compared with adulthood. P300 to target visual or auditory stimuli is of greater latency in smaller children (Courchesne, 1977; Goodin, Squires, Henderson, & Starr, 1978). However, there are also differences in waveform and topographic distribution in some components depending on task: Courchesne ( 1977) reported that rare nontarget stimuli elicited different waves in children and in adults. In children, long latency, both negative (N410) and positive (P900), was observed in frontal regions; in adults, rare stimuli elicited N200 and P300. Differences in waveform and topographic distribution have been related to age differences in the quality or mode of information processing associated with differences in either brain generator geometry or brain generator location (Friedman, Brown, Vaughan, Comblatt, & Erlenmeyer-Kimling, 1984a,b; Kok & Rooijakkers, 1985). Neville, Kotas, and Schmidt (1984) also observed that ERPs from children aged 9-13 years were very different morphologically from ERPs from adults, even though they were recorded using the same physical stimuli under the same task conditions. Children's ERPs were characterized by a large negativity around 400 msec, whereas adults were characterized by a positivity around the same time. As they pointed out, synaptic development continues at least until age 16, and it is likely that developmental changes in ERPs reflect both structural maturation and functional organization of neural systems during development. Moreover, ERPs apparently index the alterations of such neural development that occur when early experiences are abnormal. They investigated deaf and hearing subjects with a paradigm that would produce reliable language-related patterns of intra- and interhemispheric specialization in hearing subjects, by visual presentation in different visual
PSYCHOPHYSIOLOGICAL EVALUATION
fields of words. A positive shift observed in ERPs from the anterior left hemisphere in hearing subjects was not seen in the deaf subjects. Additionally, the negative peak in the right hemisphere ERPs from deaf subjects was less pronounced in ERPs from hearing subjects. From these results they concluded that altered early sensory and language experience of the deaf subjects has changed the normal trajectory of sensory and language-related processes, and that . these changes can be studied using the ERP technique. Neville et al. (1984) proposed that, because our present understanding of the functional significance and neural origins ofERPs is limited, developmental ERP studies should proceed in parallel with studies of normal adults. Studies of development hold the promise of clarifying our understanding of those neural systems that subserve cognitive processes in normal children and adults, and the ways in which these systems might be altered by early experience.
Mental Retardation The possibility that ERPs may reflect neural substrates of intellectual capacity has engaged the attention of many investigators. Chalke and Ertl ( 1965) were the first to report ERP correlates of intelligence. They recorded visual evoked responses (VERs) and found that less-intelligent subjects had longer latencies. Rhodes, Dustman, and Beck ( 1969) compared VERs in 10- and 11-year-old children with IQs ranging from 70 to 90 with those of children whose IQs were between 120 and 140. They observed larger responses and higher ratios of right central/left central VERs in bright children. The finding of lower amplitude in dull children is not consistent with the results for Down's syndrome: these children have been reported to have a larger than normal amplitude of late components in visual and somatosensory responses (Bigum, Dustman, & Beck, 1970) and in auditory evoked responses (AERs) (Barnet & Lodge, 1967). This discrepancy suggests that ERP deviations in Down's syndrome reflect processes other than those associated with lower intelligence. Barnet and Ohlrich ( 1971) described a lack of evoked response decrement in Down's infants when studying the effect of habituation on the AERs. Squires, Aine, Buchwald, Norman, and Galbraith (1980) reported a surprising result in their study of 16 adult patients with Down's syndrome, 15 adult retarded subjects of unknown etiology, and 15 normal adults: patients with Down's syndrome had shorter 1-11, III-IV intervals and long-
273
er IV-V intervals than did controls. Wave V also showed abnormally small latency increases at fast click rates. The group of mentally retarded of unknown origin. displayed shorter III-IV intervals and longer IV- V intervals compared with controls. The authors concluded that Down's syndrome is characterized by an abnormal functioning of the auditory brain stem pathway. Sohmer and Student (1978), studying a group of 10 children with psychomotor retardation, found longer latencies of wave I and longer brain stem transmission times. In two children, waves IV and V were absent. Goldman, Sohmer, Godfrey, and Manheim ( 1981) compared normal children of normal IQ values with gifted children and with children with low IQs. They observed that retarded children had lower amplitudes of the BAERs and higher amplitudes of the long-latency AERs than did gifted children. The unique common finding for those children with low IQs was that they were of low social class. The authors concluded that sensory deprivation might be the reason for such abnormalities of the auditory responses. Finley, Faux, Hutcheson, and Amstutz (1985) studied the correlation between P300 latency and cognitive impairment determined by the HalsteadReitan neuropsychological battery test. They found that children with abnormal evaluations, on the Halstead-Reitan showed greater P300 latencies than did those with normal values on the neuropsychological test. They proposed that P300 might be used to differentiate functional and organic cognitive disorders. In order to evaluate possible influences of differences in intelligence on ERP' s relations to psychopathology, Shagass, Roemer, Straumanis, and Josiassen (1981) compared high- and low-IQ psychiatric patients with respect to different measures of VERs, AERs, and somatosensory evoked responses (SSERs). Low-IQ patients had lower VER and AER amplitudes. For SSERs, low-IQ patients had higher early (N60) and lower later (Pl85) waves than did high-IQ patients. The authors concluded that these characteristics, although resembling deviations from normality reported for psychotics, could account for part, but by no means all, of the psychopathologyrelated ERP differences. Motor potentials to non cued movements showed a very different pattern in mentally retarded adolescents and normals: a very prominent positive slow wave, which according to Karrer, Warren, and Ruth (1978) may reflect immaturity or impairment of sensorimotor development. The aforementioned results are difficult to in-
274
CHAPTER 15
terpret only on the basis of intelligence: the influence of brain damage at the lower end of the IQ scale is possible. Low IQ scores and ERP abnormalities may both be the consequences of other causes not yet considered.
Childhood Autism From a neurophysiological point of view, studies on infantile autism have been conducted in order to show defects in information processing, defects in mechanisms of orienting to novel stimulation, failure in the devdopment of hemispheric specialization, and maturational delay. Due to serious problems in obtaining the cooperation of these children, technical difficulties are frequent and for this reason some studies have been carried out during unmedicated sleep. As in severely retarded children, it is often difficult to state the role played respectively by autism and by mental deficit. ERPs have been used in an attempt to discover whether electrophysiological data differ for emotional troubles and for cognitive disturbances. The study of BAERs in autistic children have yielded contradictory results. Sohmer and Student (1978) studied autistic children, children with minimal brain dysfunction, and children with psychomotor retardation. They observed an increased latency of all waves, including wave I, in all groups. In order to avoid the effects of the possible presence of any conductive loss, they measured the transmission time and obtained even more significantly increased differences between normal children and patient groups than with latencies alone. Although the results per se do not clarify the question as to whether this functional deviation was due to a specific lesion of the auditory pathways or was a sign of diffuse brain damage, they concluded that a diffuse lesion was the most probable explanation. Tanguay, Edwards, Buchwald, Schwafel, and Allen (1982) compared a group of autistic children with a control group matched by age and sex and found only increased latency of wave I to right ear stimulation with clicks of 42 db in autistic children. Skoff, Mirsky, and Turner (1980) and Rosenblum, Arick, King, Stubb, Young, & Pelson (1980) reported increased transmission times in autistic children, whereas Tanguay et al. (1982) did not find modifications in the transmission times. Rumsey, Grimes, Pikus, Duara, and Ismond (1984) did not find differences in BAER characteristics in their study of 25 children and adults with pervasive developmental disorders, including autism, versus 25 sex- and age-matched controls.
Many variables may explain such contradictory results: the intensity and frequency of stimulation used vary from laboratory to laboratory, and some studies have not taken into account the sex effect in their samples nor risk factor effects that have been shown to have a direct influence on BAERs. Therefore, more experimental data are needed in order to determine if autistic children have deficiencies in auditory sensory processing. Cortical evoked responses in most studies have shown a smaller amplitude in autistic children than in normal children (Lelord, Laffont, Jusseaume, & Stephant, 1973; Omitz, Tanguay, Lee, Ritvo, Sivertsen, & Wilson, 1972; Small, De Myer, & Milstein, 1971). Martineau, Laffont, Bruneau, Roux, and Lelord (1980) studied the ERP of 82 children with autistic behavior and/ or mental retardation and 43 normally adapted children. Subjects were recorded during two sessions of sound (S) and light (L) conditioning in three stages: (1) habituation: sound alone; (2) conditioning: coupling SL; and (3) extinction: S alone after SL series. Conditioning of ERP after coupling SL, which is expressed as an increase in the amplitude of sound ERP in the occipital area, was absent in mentally retarded and weak in autistic children. In the same group of children, subgrouping on the basis of clinical characteristics yielded three major groups: autism, mental retardation, and normal adaptation. From an electrophysiological point of view, only two major groups were clearly defined: (l) few conditioned ERPs with small amplitude, small unconditioned ERPs, and generalized conditioned and unconditioned slow potentials, and (2) many conditioned ERPs, absence of generalized slow potentials, and many vertex slow potentials or CNV. From these results the authors concluded that ERP helps to differentiate pathological groups from a normal group but is insufficient to distinguish the autistic from the mentally retarded. In a subsequent paper, Martineau, Garreau, Barthelemy, and Lelord (1984) measured amplitude and latencies of peaks NlOO, P200, and P300 to S alone and to paired stimuli in 18 normal and 15 autistic children. Latencies of AERs to S alone were smaller in autistic children than in normal controls. P300 amplitude was also larger in the autistic group to S alone. Pairing S with L, latencies had few modifications but the amplitude of all peaks increased in controls and diminished in autistic children. P300 showed greater variability in autistic children. This finding agrees with that of Novick, Vaughan, Kurtzberg, and Simon (1980), who examined the standard deviation of the mean ERP in autistic chi!-
PSYCHOPHYSIOLOGICAL EVALUATION
dren and found considerable variability. Callaway ( 1975) related variability to thought disorders and variable performance in schizophrenia. It also has been found that the urinary level of homovanillic acid, one of the main derivatives of dopamine, is higher in autistic children than in controls, and that there is a negative correlation between homovanillic acid levels and amplitude of AERs and VERs in autistic and normal children. According to Garreau, Martineau, Barthelemy, and Lelord (1984), this suggests a relationship between evoked potential characteristics and dopamine metabolism. Another approach to the study of infantile autism has been that followed by Courchesne, Lincoln, Kilman, and Galambos (1985) based on the clinical observation that autistic individuals do not orient to novel information in a normal fashion. VERs and AERs to stimuli requiring simple classification decisions and ERPs to unexpected, novel information, presented without forewarning to subjects, were analyzed in a group of I 0 nonretarded autistic subjects between 13 and 25 years old. Two conditions were studied in each session of AERs and VERs: no task condition, where children simply looked at or listened to these stimuli, and a task or performance condition where they pressed a button at the occurrence of target stimuli intermixed with unexpected, novel stimuli and also with expected, familiar stimuli. In the task condition, auditory stimuli evoked AERs of smaller amplitudes in autistic subjects to novel sounds in vertex, to target (NHX>, P300), and nontarget sounds (NIOO, P300) in frontal regions. In the visual modality the autistic group had smaller VER amplitudes at the frontal sites to novels and targets. The results suggested: (1) nonretarded autistic subjects may have a limited capacity to process novel information-they are neither hypersensitive to novel information nor misperceive it as nonnovel and insignificant; (2) classification of simple visual information may be less impaired than auditory; and (3) with the exception of only one latency difference, visual and auditory ERP abnormalities do not seem to reflect maturational delay. This study was performed in high-functioning autistic subjects to permit the analysis of the relationship between neurophysiological variables and information processing not confounded by mental retardation, poor attention, poor cooperation, and low performance capacity. The results supported the findings of Martineau et al. ( 1984) and Novick et al. ( 1980) in relation to a decreased P300 to auditory stimuli in autistic subjects, which reflects that some particular aspects of information processing are abnormal in autism. Such replicability across laborato-
275
ries, designs, and quantification approaches gives support to the hypothesis that long-latency ERP components associated with cognitive processing could prove valuable in determining the neurobiological dysfunction or autism (Courchesne et al .. 1985). On the other hand, Courchesne et al. did not find the striking changes, in the AERs and VERs during nontask conditions, that have been reported by others. Such differences may be due to the fact that autistic children may display a pattern of ERP abnormalities different from that of autistic adolescents and young adults.
Childhood Schizophrenia Strandburg, Marsh, Brown, Asarnow, and Guthrie ( 1984) studied 10 schizophrenic children and 13 age-matched normal children lacking any positive antecedent in their clinical history. ERPs were recorded at midline frontal, central, and parietal regions and left and right frontotemporal and posterotemporal leads using short-circuited ear lobes as reference during the performance of the Span Apprehension task (Span). This task was chosen because schizophrenic children and adults showed impairment with regard to it (Asarnow & MacCrimmon, 1981). Subjects had to discriminate one or two letters from four different arrays of increasing difficulty. A warning tone was presented 500 msec prior to each array. The averaged epoch included this 500msec interval and I sec of activity following presentation of the Span arrays. Schizophrenic children performed worse than the normal children on each of the array types, but better than chance. Normal children exhibited a long reaction time during the most difficult array relative to their performance in the easier condition. Schizophrenic children do not differ on either condition. However, there were no significant differences in the reaction time between the two groups in any condition. The sequence of ERPs was: an initial auditory complex to the warning stimulus, followed by a preparatory CNV and then by the visual waves PIOO, Nl40, P200, and finally by several late positive components. The following differences were found between the groups: The schizophrenic group produced a small CNV that was slow to develop and resolve, as well as diminished amplitudes for N100, P300, and slow components. This suggests that such children are impaired in their ability to regulate processes involved in the mobilization and direction of attention and the discrimination of target stimuli. Significantly, the schizophrenic children did not show progressive increases in N100 and slow-wave
276
CHAPTER 15
amplitudes in response to increases in array difficulty, as was the case in normal children. ERPs of schizophrenic children were most aberrant in frontal leads, but lateralized deficits in the posterotemporal region were also seen: normal children had higher VER amplitudes in the right hemisphere than the left. The schizophrenic children did not show this lateralization and produced smaller responses at both leads. The results of this study were very similar to what have been found in schizophrenic adults: the CNV amplitude is lower in schizophrenic patients than in normal subjects, as has been reported by many authors (Bachneff & Engelsmann, 1983; Rizzo, Spadoro, Albani, & Morocutti, 1984; TimsitBerthier, Mantanus, Ansseau, Doumont, & Legros, 1983; Van den Bosch, 1984). The second important difference observed in schizophrenic children was the low amplitude of NIOO and P300. In relation to these components, it has been mentioned that they reflect stimulus set and response set, according to the theory of Broadbent (1971 , 1977) who differentiated two levels of attentional selectivity: stimulus set-selecting a channel of information characterized by the same physical attribute; and response set-selecting those stimuli requiring a particular response. In dichotic listening tasks in which the subject has to focus attention and detect occasional targets among standard tones in one ear and ignore all tones presented to the other ear, the N 100 potential in normal subjects is larger for all tones in the attended ear, and P300 is larger only for the detected targets. The relative amplitude difference between the responses to attended and ignored tones provides the measure of selective focusing on one auditory channel (stimulus set) and between the attended standard tones and targets, the measure of response set selection. Using this paradigm, Baribeau-Braun, Picton, and Gosselin (1982) found that theN 100 potential was significantly larger for the controls than for the schizophrenics for the attended and for the ignored stimuli, but the NIOO amplitude was significantly modulated by attention only at fast rates of stimulation. The P300 component was significantly larger for controls than for schizophrenics even when targets were detected accurately and throughout the experimental manipulations. These results suggest that schizophrenics suffer from a general inefficiency in obtaining information from significant stimuli. They are unable to organize and maintain an effective strategy for processing information, and this could result from an inability to organize the processes in an optimal manner. Other authors have also observed decreased amplitudes of
NIOO and P300 (Hiramatsu, Kameyama, Niwa, Saitoh, Rymar, & ltah, 1983; Roth, Pfefferbaum, Horvath, Berger, & Berts, 1980). At present, it appears that all investigators agree that N 100 and P300 are of lower amplitude in schizophrenics, both to auditory and to visual stimuli. Lower amplitude of P300 might be a genetic characteristic: In a study employing syllable discrimination tasks in siblings of schizophrenic probands, Saitoh, Niwa, Hiramatsu, Kameyama, Rymar, & Itoh (1984) reported that although siblings displayed increased amplitude of N100 according to the allocation of their attention between two different channels, there was no augmentation of P300 to target stimuli. Mean amplitudes of P300 for siblings were nearly equal to those for unmedicated schizophrenics, with these values in siblings being significantly smaller compared to those of normal controls. Based on these results, it was concluded that abnormalities of P300 in siblings may reflect a genetic predisposition to schizophrenia. From these studies it is possible to conclude on the basis of ERPs that .schizophrenics are characterized by: 1. Deficient ability to mobilize those processes that underlie preparatory set demonstrated by the low amplitude of their CNV 2. Deficit in focusing attention or stimulus set deficiency demonstrated by the failure to enhance N l 00 during selective attention 3. Deficits in the processes involved in the identification of significant stimuli or response set deficiency demonstrated by the low amplitude and insensitiveness to change of P300 In relation to the abnormalities described by Strandburg et al. in ERP hemispheric lateralization in schizophrenic children, the study of Hiramatsu et al. (1983) using auditory syllable discrimination tasks also supported this finding. Jutai, Gruzelier, Connolly, Manchanda, and Hirsch (1984) found abnormal patterns of hemispheric activation in schizophrenics. Such abnormalities may be capable of altering both the normal reception and elaboration of visual and auditory information. Location of such abnormal activation in temporal areas may also have implications for possible temporal lobe abnormalities in schizophrenia.
Hyperkinesis The goals in studying ERPs in hyperkinetic children were to demonstrate a possible delay of the rna-
PSYCHOPHYSIOLOGICAL EVALUATION
turation of the central nervous system or a low arousal level, which have been the hypotheses proposed to explain the behavioral disturbance and the improvement in behavior following drug treatment. Demonstration of homogeneous effects of ERPs in hyperkinetic children is difficult, as many apparently contradictory results have been obtained. However, upon detailed analysis of the .samples of children studied, it is possible to note that in almost all reviews on this aspect, the authors mixed the results obtained for learning disabilities with those for hyperkinesis. Thus, in this chapter both developmental abnormalities have been separated in an attempt to classify the results obtained. The term attentional deficit disorders has not been used because this classification appeared after the papers that will be quoted. From a clinical approach, if hyperkinesis is the more relevant sign, pharmacological treatment will be recommended. If the learning deficit is predominant, specific pedagogic rehabilitation will be indicated. However, in some children both symptoms are of equal relevance, and in these cases almost all psychiatrists would recommend pharmacological treatment and pedagogic training. With this in mind, papers that emphasized hyperkinesis or drug treatment have been grouped in this section. In next section, only studies that focused on learning disabilities will be presented. Buchsbaum and Wender (1973) recorded VERs to flashes of four different intensities in 24 children aged 6 to 12 years with minimal brain dysfunction (MBD) and a matched control group of 48 children. The mean amplitude across intensities for N 140P200 was higher in the experimental group. In addition, the normal group had a slower rate of increase of amplitude with increasing stimulus intensity. The MBD group had shorter latencies for PIOO-Nl40 and P200. The normal maturation process appears to be an increase in latency at low intensities and a decrease at high intensities. The amphetamine responder group showed little change with age in either overall latency or latency-intensity change. In the responder group, amphetamine produced larger latencies at low intensities. In contrast, in nonresponders the amphetamine effect was to reduce latencies at the lowest intensities and to prolong them at the highest intensities, thus moving further in an abnormal direction. The authors concluded that the data supported the concept of a "maturational lag" in MBD children. These results were not replicated by Hall, Griffin, Moyer, Hopkins, and Rappaport (1976), who did not find statistically significant differences between groups. However, they found significant negative correlations between degree of hy-
277
peractivity and peak V-VI (140-200) amplitude, and between therapeutic response and the attentioninattention differences in mean peak latency of wave V. This is the only report in the literature in which smaller amplitude of VERs has been correlated with hyperkinesis. Callaway et al. ( 1983) recorded VERs to flashes during passive and active conditions in hyperkinetic and control children. In the active case, the subject pressed a microswitch for rare dim flashes embedded in a series of brighter flashes to which VERs were averaged. In the passive case, there were no dim flashes. The authors reported group effects as well as interactions with age and condition. In relation to age, they observed greater amplitude of Nl60 with increasing age, contrary to what has been reported elsewhere in the literature. Hyperkinetic children had larger VERs, appearing more mature in this measure according to these authors. In normal children, N200 latency increased with age, whereas hyperkinetics showed a decrease with age. Nl60, N200, and P230 amplitudes were higher during the active task and in all three measures hyperkinetics had larger amplitudes than did normals. Michael, Klorman, Salzman, Borgstedt, and Dainer ( 1981) measured the late positive components (maximum amplitude in the 250-600 msec range) of VERs to the X and BX version of the Continuous Performance Test: the child was asked to watch a screen on which one of six letters was displayed. The child had to press a microswitch detecting the letter X first, and in the second part of the study the child had to press the button upon presentation of the letter X if the preceding cue was B. Late positive components reflected the attentional demands of the task, as indicated by larger amplitudes evoked by targets than nontargets. The hyperactive children exhibited smaller late positive components than did the normal peers, but the diagnostic differences were sharper among younger children. Prichep, Sutton, and Hakerem (1976) recorded long-latency AERs to clicks during a guessing paradigm under conditions of certainty and uncertainty: single and double clicks were presented randomly. In the uncertain condition the subject guessed, prior to each trial, whether a single or a double click would be presented. In the certain condition the subject was told, prior to each trial, whether there would be a single or a double click. Differences between the normal and the hyperkinetic children were only observed during the uncertain condition to the second click. Hyperkinetics were characterized by: lower amplitude of P 186, higher amplitude of N250, and an extremely large P300 during the certain condition.
278
CHAPTER 15
Because P300 would be expected to be relatively small under conditions of certainty (Sutton et al., 1967), the fact that hyperkinetics have larger P300 in the certain condition suggested that they were responding inappropriately to task demands. Another important observation was that there were no changes in the AERs of hyperkinetics during either certain or uncertain conditions. The latter result is supported by evidence presented by Loiselle, Stamm, Maitinsky, and Whipple (1980), who studied the AERs to tone pips during a selective attention task. Two independent dichotically presented series of pips at 800Hz to one ear and 1500Hz to the other ear with interspersed signal pips of 840 or 1150 Hz were delivered. The subject was instructed to listen to and count the number of signal pips to one ear and ignore those to the other ear. For both groups the amplitudes of N100 were higher to the attended than to the nonattended channels. However, the amplitude difference between these channels was substantial and significant only for the control group. P300 waves were evaluated to the attended signal tones and to the concurrent signals from the nonattended channel. Mean P300 amplitudes to the nonattended signals did not differ between groups. However, although both groups had higher mean amplitudes to the attended than to the nonattended signals, the amplitude difference was only significant for the controls. Also, P300 to attended signals was significantly greater for the control boys. Using long-latency AERs to unattended clicks, Satterfield and Braley ( 1977) demonstrated that age was an important factor when comparing normal and hyperkinetic children: younger hyperactive children were found to have significantly smaller AER amplitudes, and older hyperactive children had AER amplitudes larger than those of age-matched controls. Considering the total samples, there was no overall group difference. Recently, Satterfield et al. (1984) reported a cross-sectional and longitudinal study of age effects on a set of electrophysiological measures, including AERs in hyperactive and normal children. They showed that the amplitude decreased with age among the controls, but increased among the hyperactive subjects. The age by diagnostic group (hyperactive/nonnal) interaction was significant for all amplitude measures, and the youngest children (6.8 to 8.3 years) contributed most to the overall interaction. The authors concluded that their data did not fit a delayed maturational model, because the youngest hyperactive group had AER5 and BEG measures similar to those of the older nonnal group, whereas the oldest hyperactive group had measures more like
those of younger normal subjects. Such findings are more consistent with an aberrant than with a delayed maturational hypothesis. In Table I these results have been summarized in order to assist the following discussion. Taking into consideration the opposite development of longlatency AERs children, it is possibly to understand why, in the study of sensory evoked responses (visual or auditory) during passive conditions, the most frequent observation was the absence of significant differences (Hall et al., 1976; Loiselle et al., 1980; Michael et at., 1981). If the range of age and the number of subjects are carefully observed, it clearly appears that as age has not been taken into account, the overall total differences between groups were not significant possibly due to a masking of differential effects of age. However, two papers did show higher amplitudes during passive conditions in hyperkinetic children (Buchsbaum & Wender, 1973; Callaway et al., 1983). Experiments done during active conditions may be classified as to those that emphasized ( l) the amplitude of the first waves or (2) P300 or late positive components. Callaway et al. (1983) and Prichep et al. (1976) found higher amplitudes of the negative waves in hyperkinetic children. But what is more remarkable is the fact that there were no significant changes of amplitude during different conditions. P300 followed a contrary effect in normal and hyperkinetic children depending on the paradigm: it was abnormally high in certain conditions in hyperkinetics (Prichep et al., 1976) and of lower amplitude to targets (Loiselle et al., 1980), this effect being more clearly defined in younger hyperkinetics (Michael et al., 1981). From the above discussion, it is possible to conclude that hyperkinetic children are characterized by an aberrant development of the central nervous system and by severe attentional defects at both the stimulus and response sets.
Effect of Medication on the ERPs of Hyperkinetic Children Table 2 summarizes the most important effects of drug treatments in hyperkinetic children. It is obvious that in ''responders,'' amphetamine and methylphenidate produced an improvement in ERP measures toward normalization. During passive conditions, modifications by drugs are not as clear as dur· ing active conditions. These results have great practical importance, because for the evaluation of such drugs it is not necessary to wait for long periods. In
PSYCHOPHYSIOLOGICAL EVALUATION
279
TABLE 1. Event-Related Potentials in Hyperkinetics Reference
No. of No. of control hyperkinetic
Age
Stimuli
Condition
Lead
Results
Buchsbaum & Wender (1973)
48
24
6-12
Photic
Passive
~
MBD with higher amplitudes N140 P200.
Hallet al. (1976)
9 10
10 12
6-ll
Photic
Passive
cz
No differences between groups.
Callaway er al. (1983)
18
18
6-13
Photic
Passive Active
Cz, Pz
Passive: hyperkinetics with higher Nl amplitude. Active: hypeddnetics with higher Nl N2 P2. Age interaction.
Michael et al. (1981)
21
21
6-13
Photic CPT-X BX
Passive Active
~,Pz
Active: hyperkinetics with smaller late positive components. Younger highest effeet. Age interaction.
Prichep et al. (1976)
8
16
8-11
Single or double clicks
Certain Uncertain
cz
Certain: P300 higher in hyperlcinetics. Uncertain: hyperkinetics with smaller P186 and higher N2SO. Hyperkinetics: No change certain versus uncertain.
Loiselle et al. (1980)
15
12
12-14
Tone pips
Unattended Attended
4
Attended: hyperkinetics no significant amplitude increment of NIOO and P300.
Satterfield er al. (1984)
52
127
6-12
Clicks
Unattended
Differential effect according to age. Younger hypeddnetics had smaller amplitudes and older hyperkinetics had larger amplitudes than nonnals.
Follow-up
48
51
Clicks
Unattended
Follow-up: Nonnals decreased amplitudes with age, hyperkinetics increased amplitudes with age.
two different sessions, one before treatment and the second one after a dose of the drug, it is possible to reach conclusions about whether a particular child will respond to pharmacological treatment.
Dyslexia and Learning Disabilities Many papers have been dedicated to the study of ERPs in learning disabilities (LD). Such studies have tried to demonstrate perceptual deficits, abnormalities on hemispheric specialization, and deficiencies in temporal processing of verbal stimuli. The first publications examined VERs. Conners (1971) recorded in parietal and occipital derivations VERs to flashes in twenty-seven 8- to 12-year-old "poor readers." He found a strong correlation between amplitude attenuation of wave N200 from left parietal derivation and severity of reading disability.
This finding was replicated in a second experiment on twenty 9- to 15-year-old subjects. Preston, Guthrie, and Childs (1974) recorded VERs to flashes and the word "cat" in nine disabled readers aged 10 and in two control groups, one matched for age and IQ and the other for reading level and IQ. They also found that N200 in the left parietal region was significantly smaller in LD than in controls. They concluded that as this finding could not be explained on the basis of age, IQ, or reading level, it probably represents a neurological concomitant of the disorder. They were not able to exclude the possibility that disabled readers were simply less attentive to the stimuli. Cohen ( 1980) studied 11 children with primary dyslexia and a control group matched by sex, age, and intelligence. He used a CNV paradigm on which the strobe light was the warning signal followed by a 500-msec-duration tone that was terminated by the
280
CHAPTER 15
TABLE 2. Effect of Treatment on Event-Related Potentials Reference
Drug
Stimuli
Results
Buchsbaum & Wender (1973)
Amphetamine
Photic
Responders: Amphetamine prolonged latencies at lower intensities
Greenhill, Rieder, Wender, Buchsbaum, & Zahn (1973)
Lithium carbonate
Photic
Nonresponders: LC reduced both amplitude and latencies at lower intensity
Saletu, Saletu, Simeon, Viamontes, & Itil (1975)
Thioridazine
Photic
Responders to Td: Short latencies and smaller amplitudes 200-300 msec before treatment. Td improves attention. Responders to A: short latencies and higher amplitudes 80-100 msec before treatment. A decreases hyperactivity.
Hall et al. (1976)
Amphetamine
Photic
Responders: Lower amplitude and longer latency in inattention.
Prichep et al. (1976)
Methylphenidate
Clicks
Responders: M produced higher differences in Pl86 between certain and uncertain, higher amplitude of P186 and lower amplitude of N250
Michael et al. (1981)
Methylphenidate
Photic CPT-X BX
Amplitude of parietal late positive components enhanced by M
Klorman, Salzman, Bauer, Coons, Borgstedt, & Halpern (1983)
Methylphenidate
Photic CPT-X BX
Photic: M reduced amplitudes and prolonged latencies CPT: M produced higher amplitudes of late positive components
Halliday, Callaway, & Naylor (1983)
Methylphenidate
Photic
Effect of M varied according to age and experimental paradigm. Two types
Halliday, Callaway, & Rosenthal (1984)
Methylphenidate
Photic
Amphetamine
of effects on ERP amplitude: monotonically changes with dose, interaclion dose-attention nonmonotonically. M speeds reaction time with011t shortening ERP latencies. Responders: Amplitude NlP2 difference between M and placebo during attending condition greater than 3 f.1.V
subject pressing a button. The two groups showed differences between both amplitude and latency of VERs. The latencies of NlOO and P200 were longer in the dyslexic group. Cohen and Breslin (1984) compared the VERs to flashes and to verbal stimuli in 16 reading-impaired boys and 16 normal control boys aged ll. The two groups had similar IQs and came from the same school classes. The authors corroborated the finding of P200 increased latencies in the dyslexic group both to flashes and to words. They also computed correlation coefficients between left and right early, middle, and late components. Correlation coefficients were very high for flash stimuli in both groups in middle and late latency components. Dyslexics have higher correlation coefficients to words than to controls. According to these authors, high interhemispheric correlation would imply that the hemispheres do the same things and low correlations would mean that contralateral positions on the two hemispheres reflect different processes. To analyze the intrahemispheric specialization, they used
intercorrelations among the leads of each hemisphere. The lowest correlations were obtained for the early and late components of the control group to word stimuli in the left hemisphere. Left hemisphere intercorrelations were significantly lower than right hemisphere intercorrelations for VERs to words in the control group. Dyslexic children showed no significant differences between both hemispheres either to flashes or to words. Cohen and Breslin proposed that the early phase of information reception and processing is represented bilaterally in the visual cortex with bilateral stimulation. This is followed by a period of information transfer to all of the association areas of each hemisphere, lasting between 100 and 200 msec. Finally the information is recognized and registered by a brain area that is specialized for that verbal process of template matching, usually a left hemisphere function. The failure of hemispheric specialization in even the early phase of target registration and recognition is thought to be the functional disability in the specific dyslexic individual.
PSYCHOPHYSIOLOGICAL EVALUATION
Shelburne ( 1978) studied nine boys aged 9 to 14 with relatively pure reading disabilities. Visual stimuli consisted of letters presented sequentially to form consonant-vowel-consonant (CVC) trigrams. Each trial consisted of the presentation of blank-CVCblank. eve formed either words or paired nonsense syllables with the same first two letters as the word. The child was instructed to observe the letters during each trial and to decide whether or not the eve was a word. After the second blank the child pushed a toggle switch to the right (word) or left (nonsense syllable). Immediate auditory feedback indicated whether the response was correct (tone) or incorrect (buzzer). VERs for each position in the CVC trigram were averaged separately in central and parietal leads. In previous experiments (Shelburne, 1973) normal children who performed well on this problem-solving task showed greater positive amplitude of VERs from the third position stimuli than VERs from first and second position stimuli. With the exception of one reading-disabled child, the reading disability children showed no significant VER differences. In contrast, 17 out of 20 normals tested previously showed significant VER differences. VERs may also be obtained by presenting taskirrelevant visual stimuli while subjects perform some task. This procedure is known as probe-ERP paradigm. In these experiments it is assumed that a brain region is less responsive to the probe stimulus when that region is engaged by the concurrent task. With this procedure, Johnstone, Galin, Fein, Yingling, Herron, and Marcus (1984) studied 34 controls and 32 dyslexic boys aged 10 to 12. VERs to flashed checkerboards were recorded while the children performed silent and oral reading at two levels of difficulty. In all children for all four reading tasks, the right midtemporal VER was larger than at any other lead and the left midtemporal VER was smaller than at any other lead. Principal component analysis for both groups in the seven leads was computed. Eleven factors accounted for 91% of total variance. Dyslexics showed a significant amplitude decrease in the positive components of 250-350 msec range while reading difficult material at bilateral central and parietal derivations. Normal readers showed no such effect. Ahn ( 1977) sought to identify differential characteristics of groups of children with specific reading disabilities (verbal underachievers), specific arithmetic disabilities (arithmetic underachievers) and generalized learning disability (mixed underachievers), all with normal IQ. VERs to II different visual stimuli-a blank slide, and 7 lines/inch grid, a 27 lines/inch grid, large and small squares and dia-
281
monds, and the letters b, d, p, and q-were recorded in all leads of the I 0/20 system. Similar results were obtained with all visual stimuli. Significant differences between normals and verbal underachievers were more marked on the left hemisphere in central and parietal regions in the latency domains of 280340 and 410-480 msec. Arithmetic underachievers showed more marked differences in right central and parietal leads in the latency domains of 220-270 and 310-380 msec. Mixed underachievers and normals were significantly different in the VERs in central, parietal, occipital, and posterotemporal left hemisphere around 220-260 msec of latency. As the reading- and arithmetic-specific disabilities showed most significant differences with respect to late components in the latency domain of 300 msec or later, and the mixed underachievers showed most significant differences markedly earlier, at about 200 msec, Ahn suggested that the first two groups suffered from a deficit largely related to cognitive processes and that the mixed group suffered from a deficit more perceptual in nature. Harmony and Di'az de Leon (1982, 1983) compared the VERs in three groups of children from a rural area: those with no academic problems, academic underachievers, and illiterate. Visual stimuli were a blank slide, a 7lines/inch grid, a 27lines/inch grid, and the letters b and d. Values of waveform and amplitude symmetry were very similar to those previously reported in adults (Harmony, Ricardo, Otero, Fernandez, & Valdes, 1973). Children with no academic problems showed a good response to the change of stimulation: VERs to patterns were of higher amplitude in occipital areas compared to VERs to flashes (Figure 4). VERs to the letters b and d were different in parietal leads (Figure 5). In iiliterate children and in academic underachievers, the changes of VERs to different stimuli were not so apparent. However, the most important differences between the three groups were in the measurement of variability within the evoked responses to the same stimuli. Academic underachievers have greater variability of VERs in parietal leads than controls, and illiterate children showed higher variability of VERs in central and parietal areas. These results were interpreted as a dysfunction of the association parietal areas in academic underachievers and illiterate children. Failure to have consistent VERs in central areas in illiterate children may be related to a more severe dysfunction of the unspecific systems related to attention. Such dysfunction may be produced by deficits in early experience or sensory deprivation, for these children belonged to severely culturally deprived families.
282
CHAPTER 15 8
c
f
71
~ ~
0~
~
~J
FIGURE 4. Visual evoked responses to flash and to 7 lines/inch grid in a nonnal child. Observe the change in occipital (0) regions.
In another study, Harmony, Marosi, Diaz de Le6n, Becker, and Landazuri (1984) recorded VERs to blank and to a checkerboard pattern in left and right central, parietal, occipital, and posterotemporal leads using linked ear lobes as reference. Children between 7 and 13 years old were selected from a large sample according to a test developed to analyze reading and spelling difficulties. Those children who read and write according to their age and grade at school composed the control group. Children with great difficulties in reading and spelling were assigned to the LD group. Children with specific sensory disturbances such as visual refraction problems or hypoacusia, behavioral problems, or borderline or low IQ were omitted. All children had a normal IQ (WISC). Socioeconomic level was evaluated according to type
b
p~
d
~
0~ FIGURE 5. Visual evoked responses to letters band din a nonnal child. Occipital (0) VERs do not change, but parietal (P) VERs are of different wavefonn.
?f work
and scholarship of the parents, economic mcome per capita, and conditions of the house. Pediatric and neurological examinations showed that children were healthy at the time of the study, although a large group had personal antecedents of risk factors associated with brain damage in both groups. The power calculated as the sum of the square of the amplitude values taking as reference the mean value along the 512-msec time epoch was measured for each VER. Also the left power/right power ratio and the correlation coefficient were computed between VERs from homologous leads. Power values were transformed to logarithmic values and correlation co~fficie~ts to Fisher's Zr transform. Regression equations wtth age showed that left occipital power values to both stimuli, right occipital and left posterotemporal VERs to pattern have a significant negative correlation with age in control subjects. Older children have smaller power values. This result agrees with other observations (Celesia & Daly, 1977). For measurements correlated with age, z values for each child were computed eliminating age effects. Those not directly correlated with age were directly compared by Student's t test. LD children had higher power values in right posterotemporal lead to both stimuli than did control subjects, and a smaller left/right power ratio in posterotemporal areas (Figures 6,7). The smallest VERs were recorded in left posterotemporal areas in both controls and LD to either fl~h or checkerboard. Therefore, we have observed, as did Johnstone et al. (1984), larger VERs in right than in left posterotemporal areas in passive conditions. LD children had lower correlation coefficients than controls in parietal leads. Figure 8 presents an example. A discriminant analysis using the standardized values (corrected by age) of power, power ratio, and correlation coefficients for pattern VERs showed a significant discrimination between controls and LD. However, when risk factors were taken into account, controls with antecedents and LD without antecedents were more abundant among the misclassified subjects. Therefore, in order to determine whether dependent variables other than learning disabilities were affecting the results, we did a multiple regression analysis. Independent variates considered were sex, age, risk factors, socioeconomic status, and pedagogic evaluation. It was corroborated that power of VERs clearly decreased with age. It was also observed that girls had lower power than boys. Low socioeconomic status, antecedents of risk factors, and the presence of LD were associated with a decrement of VERs power in parietal areas. These results showed the complex interactions that exist between
PSYCHOPHYSIOLOGICAL EVALUATION
4000
FLASH
283
PATTERN -CONTROLS [')ilL D
2000
O.,L----
FIGURE 6. Mean power of left (T5) and right (T6) posterotemporal leads ofVERs to flash and to a checkerboard in control and learningdisabled (LD) children.
15 o1
15 o2
15 tS
15
t6
FIGURE 7. VERs to a checkerboard pattern in an LD child. Observe the high amplitude of the VER in posterotemporal right (T6) region.
different factors on simple measures of ERPs (Harmony, Becker, Marosi, Landazuri, Baiiuelos, Diaz de Leon, and Hinojosa, 1985). They also put a note of caution in interpreting results provided by ERPs. At the present moment we are proceeding the analysis in order to know more about the interactions of such factors. Depression of VERs in parietal areas has been considered a sign of visual processing deftcits, as mentioned earlier. The fact that socioeconomic and risk factors produce a decrement in these areas poses a new question in relation to the possibility that such factors may have a causal relationship with visual processing deficits. The observation that in all children right posterotemporal VERs were larger than left posterotemporal responses, and that in LD children this difference is still more apparent due to larger right temporal responses, is very difficult to explain. Strandburg et al. (1984) using the same leads found the same in normal children, whereas in schizophrenics the contrary was observed. Johnstone et al. ( 1984) also described this asymmetry in normal and dyslexic children. These authors reported that right temporal predominance was not observed when using checkerboards covering the central fields out to 6° even in task-relevant experiments. Therefore, the
284
CHAPTER 15
IO
1168
1336
fJ 68
13 3 6
FIGURE 8. (Left) VERs to flash and (right) VERs to a checkerboard pattern in an LD child. It is possible to observe the wavef.onn asymmetry in parietal leads (p3, p4).
characteristic to find larger right temporal VERs is that stimulation should be given to the periphery of the retina. Such large potentials may be related to those described by Srebro ( 1985) using the Laplacian visual potential when upper field stimuli were used. He associated this large potential to functions like spatial orientation and identification of some attributes of the stimuli. If correct, this finding may be important from a physiopathological point of view. Lower power of VERs in parietal and occipital regions was observed in children with more severe antecedents of risk factors. It has been reported that some risk factors such as perinatal asphyxia or higher concentrations of heavy metals in children produce smaller VERs, though it is likely that this factor is influencing the VERs. Long-latency AERs to pairs of tone pips used as probe stimuli during two visual conditions were recorded by Shucard, Cummins, and McGee (1984) in a selected sample of 30 strongly right-handed male disabled readers and 30 normal subjects. The entire alphabet was displayed; during one condition the subjects have to press a button upon viewing a letter with a long "e" sound (letter sounds condition), and during the second condition subjects have to identify letters having either closed circular shape or only straight lines (letter shapes condition). Compared
with normal readers, disabled readers showed significantly lower amplitude of AERs recorded in the right hemisphere (T4-Pz) during both tasks. Disabled readers also showed significantly higher amplitude of left responses (T3-Pz) during the letter sounds condition. For both conditions the reading-disabled subjects showed significantly lower amplitudes of right than left hemisphere AERs. Task-related strategies did not differ between groups. The pattern of AER amplitude asymmetry found for disabled readers, which was opposite to that found for normal readers, suggests that the same reading-related task activated different cerebral processes in the two groups studied. Fried, Tanguay, Boder, Doubleday, and Greensite (1981) also were interested in lateralization of AERs in dyslexic children. Stimuli were voiced words do and go and strummed chords A7 and D7, randomly arranged. Control subjects showed greater waveform differences between AERs to words and chords in the left hemisphere. Developmental dyslexic children were divided into dysphonetic (great difficulty in reading and spelling words phonetically) and dyseidetic (difficulty in recognizing written words) groups. The former group showed greater differences to words and chords in the AERs of the right hemisphere; in the latter group the greatest dif-
PSYCHOPHYSIOLOGICAL EVALUATION
ferences were found in the left hemisphere, as in control subjects. The latter two studies are coincident in the demonstration that the right hemisphere of disabled readers (or at least a subgroup of them) is activated during language processing, either if stimulus is presented as speech stimulus or in a condition that involves visual-phonemic transfer. As we have already reviewed for VERs, it has been possible to find a lack of activation in the left hemisphere of disabled readers when they have to process words (Cohen & Breslin, 1984; Shelburne, 1978). Deficits in selective and sustained attention have also been studied with ERPs in LD children. Sobotka and May ( 1977) recorded VERs during a detection task on which subjects were required to respond to dim flashes occurring in a train of brighter flashes. VERs were averaged for brighter flashes. Dyslexics showed higher amplitudes ofPlNl waves than did normal children, and a slower reaction time to attended stimuli. The authors concluded that because dsylexic boys showed higher amplitudes of VERs to unattended stimuli, this may be due to an attentional deficit. The results obtained were exactly contrary to those of Conners ( 1971 ), Johnstone et al. (1984), and Preston et al. (1974). VERs to the X and BX version of the Continuous Performance Test had smaller late positive components (250-600 msec) in the dyslexic children as compared with a normal group (Dainer et al., 1981). Musso and Harter ( 1978) compared normal children and children with reading disabilities attributed to visual and auditory perceptual problems. The subjects' abilities were assessed in a visual discrimination attentional task that consisted of flashing a warning stimulus followed after 1100 msec by one or two randomly presented flashes, a relevant or an irrelevant stimulus. For a correct response, the subject was given a token; for a mistake, the subject lost a token and negative feedback was given immediately in the form of a loud click. The effect of selective attention to relevant and irrelevant stimuli was measuted by finding the difference in amplitude of the VERs to the second stimulus when it was relevant or irrelevant. The reading-disabled children with visual problems showed greater relevant-irrelevant differentiation in their VERs than did normals, which suggested that they were selectively attending more than the other group in order to compensate for their deficiency. However, latency of P300 component was larger in children with visual than with auditory problems, who in tum had latencies longer than those of normal children. These latency differences were interpreted
285
as suggesting that the reading-disabled child processes sensory information at a slower rate than does the normal child, which may be indicative of a neural deficiency. Cohen (1980) reported larger CNV in controls than in dyslexics. However, the reaction time was the same for both groups and for this reason he concluded that the deficit observed in VERs and CNV was more related to a neurophysiological dysfunction in visual information processing. Fenelon ( 1978) recorded CNV s from frontal and parietal sites over both hemispheres of problem and normal readers during several stimulation conditions. Reading-disabled children showed CNVs of smaller amplitude in left parietal area. Assessment of intra- and interhemispheric symmetry of waveforms was done by computation of correlation coefficients. Right hemisphere responses were more highly correlated in the disabled readers than in the normal group in the visual stimulation conditions, but not when auditory stimuli were used. A brief review of BAERs in children with learning problems will now be presented. It was mentioned in relation to mental retardation and autism that Sohmer and Student ( 1978) recorded BAERs in several groups of children including a group with MBD. They observed longer latencies and transmission time in MBD children than in the normal group. However, the authors did not give detailed characteristics for this group. Goldman et al. (1981) observed smaller amplitudes of waves I to V of BAERs in borderline children or children with poor scholastic achievement. The decreased response amplitudes of compound action potentials and evoked auditory responses were interpreted as a decrease in the number of responding units. The common factor shared by these children was lower socioeconomic-cultural status. Because this background is often associated with such factors as slight undernourishment of the mother during her childhood, early marriage and pregnancy, high frequency of pregnancy, and limited prenatal and postnatal care of the children, the authors proposed that these factors could lead to mild forms of MBD. They also considered that the retarded children having been raised in restricted, culturally deprived, and nonstimulating environments may also develop similar changes in neuronal function expressed as decreased response amplitudes. Children with speech and language disorders exhibited amplitudes for waves I, III, and V smaller than those of the normal group, with no change in latency (Mason & Mellor, 1984). Two explanations were suggested for this
286
CHAPTER 15
finding: an abnormal functioning of the auditory system due to deprivation of normal speech and language development, or differences in the electrical conductivity of the tissue in children with language disorders. From this review on BAERs it appears clear that many factors may modify the development of such potentials. However, the alterations observed suggest dysfunction of brain stem auditory centers in some children with LD. The brain stem auditory system is especially vulnerable to perinatal injury (Griffiths & Laurence, 1974; Myers, 1972). The tectum is the most likely to be affected by a brief episode of asphyxia at birth because it has a high oxygen requirement and as its myelinization is complete at the time of birth, this damage is irreversible. As is well known, the tectum is a major center for selective attention and orientation to sounds and to visual stimuli. Also, extraction of major morphemic word stems and syntactic features from environmental speech may be a primitive perceptual function of brain stem auditory centers (Simon, 1975). Therefore, objective demonstration by recording BAERs of some alteration at this level is extremely important for children's neuropsychological disorders. Not all children with these problems will show BAER abnormalities, but in those who do, specialized rehabilitation programs should be planned to improve language skills. To summarize what has been found in selected samples of children with specific learning disorders and the study of ERPs, such children are characterized by:
important factor to be taken into account when defining norms.
References Ahn, H. ( 1977). Electroencephalographic evoked potentials com· parisons of normal children with different modes of underachievement. Doctoral dissertation, University of Iowa. Ahn, H., Prichep, L., John, E. R., Baird, H., Trepetin, M., & Kaye, H. (1980). Developmental equations reflect brain dysfunctions. Science, 210, 1259-1262. Aird, R. B., & Gastaut, Y. (1959). Occipital and posterior electroencephalograpbic rhythms. Electroencephalography and Clinical Neurophysiology, 11, 637-656. Alvarez, A., Val~s. P. & Pascual, R. D. (1987). US EEG developmental equations confirmed for Cuban school children. Electroencephalography and Clinical Neurophysiology, 67, 330-332.
Alvarez, A., Ricardo, J., Valdes, P. A., & Pascual, R. D. (1985). Clinical evaluation of the xi-alpha model for the EEG: Comparison with broad band spectral features. Neuroscience Branch Technical Report No. NH-0, National Research Center of Cuba. Havana. Asarnow, R. F., & MacCrimmon, D. (1981). Span of apprehension deficits during the post-psychotic stages of schizophrenia: A replication and extension. Archives of General Psychiatry, 38, 1006-1011. Bachneff, S. A., &Engelsmann, F. (1983). Correlates of cerebral event-related slow potentials and psychopathology. Psychological Medicine, 13, 763-770. Baribeau-Braun, J., Picton, T. W., & Gosselin, J. I. (1982). Schizophrenia: A neurophysiological evaluation of abnormal information processing. Science, 219, 874-876. Barnet, A. B., & Lodge, A. (1967). Clicked evoked EEG responses in normal and developmentally retarded infants. Nature, 214, 252-255. Barnet, A. B., & Ohlrich, E. S. (1971). Evoked response decrement during auditory stimulation in normal and mongoloid infants. Electroencephalography and Clinical Neurophysiology, 31, 290-291. Becker, I., Velasco, M., Harmony, T., Marosi, E., &Landazuri, M. (1987). Electroencephalographic characteristics of children with learning disabilities. Clinical Electroencephalography, 18, 93-101. Bigum, H. B.,Dustman, R. E.,&Beck, E. C. (1970). Visual and somatosensory evoked responses from mongoloid and normal children. Electroencephalography and Clinical Neurophysiology, 28, 576-585. Broadbent, D. E. (1971). Decision and stress. New York: Academic Press. Broadbent, D. E. (1977). Levels, hierarchies and the locus of control. Quarterly Jou1111.1l of Experimental Psychology. 19,
1. Deficiencies in visual information processing even to simple visual stimuli. 2. Abnormal lateralization in those tasks related to language processing, either to visual or to auditory stimuli. 3. Attentional deficits appear not to be fundamental in these children, for almost all studies report normal performance. However, depressed CNV, late positive components, and delayed P300 are observed. These findings have been interpreted as more related to slower rate of information processing and not to the stimulus set phase of selective attention. 4. Other factors have to be considered in evaluating LD children besides age, and these are pathological antecedents, sex, poor cul181-201. tural environments, and so forth. Buchsbaum, M., & Wender, P. (1973). Averaged evoked re5. The study of BAERs in these children seems sponses in normal and minimally brain dysfunctioned chilto be very important to detect those who may dren treated with amphetamine. Archives ofGeneral Psychiahave a brain stem dysfunction. Sex is an try. 19. 764-770.
PSYCHOPHYSIOLOGICAL EVALUATION Callaway, E. (1975). Brain electrical potentials and individual psychological differences. New York: Grone & Stratton. Callaway, E., Halliday, R., & Naylor, H. (1983). Hyperactive children's ERP fail to support underarousal and maturational lag theories. Archives of General Psychiatry, 40, 12431248. Celesia, G. G. & Daly, R. F. (1977). Effects of aging on visual evoked responses. Archives Neurology, 44, 403-407. Chalke, F. C. R., & Ertl, J. (1965). Evoked potentials and intelligence. Life Sciences, 4, 1319-1322. Cohen, J. (1980). Cerebral evoked potentials in dyslexic children. In D. A. Otto (Ed.), Multidisciplinary perspectives in eventrelated brain potential research. Washington, DC: EPA-600/9-77 -043. Cohen, J., & Breslin, P. W. (1984). Visual evoked responses in dyslexic children. Annals of the New York Academy of Sciences, 425, 338-343. Cohn, R. (1958). On the significance of bioccipital slow wave activity in the electroencephalogram of children. Electroencephalography and Clinical Neurophysiology, 10, 766-768. Cohn, R. ( 1961). Delayed acquisition of reading and writing abilities in children. Archives of Neurology (Chicago), 4, 153164. Cohn, R., & Nardini, J. E. (1958). The correlation of bilateral occipital slow activily in the human EEG with certain disorders of behavior. American Journal ofPhysiology, 115, 44--54. Conners, C. K. (1971). Cortical visual evoked response in children with learning disorders. Psychophysiology, 7, 418-428. Courchesne, E. (1977). Event related brain potentials: Comparison between children and adults. Science, 197, 589-592. Courchesne, E., Lincoln, A., Kilman, B. A., & Galambos, R. (1985). Event-related brain potential correlates of the processing of novel visual and auditory information in autism. Journal ofAutism and Developmental Disorders, 15, 55-76. Dainer, K. B., Klorman, R., Salzman, L. F., Hess, D. W., Davidson, P. W., & Michael, R. L. (1981). Learning disordered children's evoked potentials during sustained attention. Journal of Abnormal Child Psychology, 9, 79-94. DiazdeLeon,A. E., Harmony, T.,Marosi,E.,Landazuri,A. M., Becker, J., & Banuelos, E. (1985). EEG frequency analysis in academic underachievers. Electroencephalography and Clinical Neurophysiology, 61, Sl30. Diaz de LeOn, A. E., Harmony, T., Marosi, E., Becker, J., & Alvarez, A. (1988). Effect of different factors on EEG maturation. International Journal of Neurosciences (accepted for publication). Donald, M. W. (1980). Memory, learning and event-related potentials. In H. H. Komhuber & L. Deecke (Eds.), Motivation, motor and sensory processes of the brain: Electrical potentials, behavior and clinical use. Amsterdam: Elsevier. Donchin, E., Ritter, W., & McCallum, W. C. (1978). Cognitive psychophysiology: The endogenous components of the ERP. In E. Callaway, P. Tueting, & S. Koslow (Eds.), Eventrelated brain potentials in man. New York: Academic Press. Doose, H., Gerken, H., & Volzke, E. (1968). Geneticsofcentrencephalic epilepsy in childhood. Epilepsia, 9, 107-115. Eeg-Oiofsson, 0., Petersen, 1., & Sellden, U. (1971). The development of EEG in normal children from the age of I to 15 years: Paroxysmal activity. Neuropiidiatrie, 4, 375-404.
287
Fein, G., Galin, D., Yingling, C. D., Johnstone, J., Davenport, L., & Heron, J. (I 986). EEG spectra in dyslexic and control boys during resting conditions. Electroencephalography and Clinical Neurophysiology, 63, 87-97. Fenelon, B. (1978). Hemispheric effects of stimulus sequence and side of stimulation on slow potentials in children with learn· ing problems. InD. A. Otto (Ed.), Multidisciplinary perspectives in event-related brain potential research. Washington, DC: EPA-600/9-77-043. Finley, W. W., Faux, S. F., Hutcheson. J., &Amstutz, L. (1985). Long latency event-related potentials in the evaluation of cognitive functions in children. Neurology, 35, 323-327. Friberg, S., Matousek, M., &Petersen, I. (1980). A mathematical model for the age development of the background activity in the human electroencephalogram. Research Laboratory of Medical Electronics, Chalmers University of Technology, Gothenburg, ISBN 91-7546-011-4. Fried, I., Tanguay, P. E., Boder, E., Doubleday, C., &Greensite, M. (1981). Developmental dyslexia: Electrophysiological evidence of clinical subgroups. Brain and Language, 12, 1422. Friedman, D., Brown, C., Comblatt, B., Vaughan, H. G., & Erlenmeyer-Kjm)ing, L. (1984a). Changes in the late taskrelated brain potentials during adolescence. Annals of the New York Academy of Sciences, 425, 344-352. Friedman, D., Brown, C., Vaughan, H. G., Comblatt, B., & Erlenmeyer-Kimling, L. (1984b). Cognitive brain potential components in adolescents. Psychophysiology, 21, 83-96. Fuller, P. W. (1977). Computer estimated alpha attenuation during problem solving in children with learning disabilities. Electroencephalography and Clinical Neurophysiology, 42, 149-166. Garreau, B., Martineau, J., Barthelemy, C., &Lelord, G. (1984). Electrocortical activity (evoked potentials) and dopamine metabolism (homovanillic acid) in normal and emotionally disturbed children. Annals of the New York Academy of Sciences, 425, 353-356. Gasser, T., Mocks, J., Lenard, H. G., Bacher, P., & Verleger, R. (1983). The EEG of mildly retarded children: Developmental, classificatory and topographic aspects. Electroencephalography and Clinical Neurophysiology, 55, 131-144. Gibbs, F. A., Gibbs, E. L., & Lennox, W. G. (1943). Electroencephalographic response to hyperventilation and its relation to age. Journal of Pediatrics, 23, 497-505. Glanzmann, P., & Froelich, W. D. (1984). Anxiety, stress, and contingent negative variation considered. Annals of the New York Academy of Sciences, 425, 578-583. Goldman, Z., Sohmer, H., Godfrey, C., & Manheim, A. (1981). Auditory nerve, brainstem and cortical response correlates of learning capacity. Physiology and Behavior, 26, 637-645. Goodin, D., Squires, K., Henderson, B., & Starr, A. (1978). Age related variations in evoked potentials to auditory stimuli in normal subjects. Electroencephalography and Clinical Neurophysiology, 44, 447-458. Greenhill, L. L., Rieder, R. 0., Wender, P. H., Buchsbaum, M., & Zahn, T. P. (1973). Lithium carbonate in the treatment of hyperactive children. Archives of General Psychiatry, 28, 636-640. Griffiths, A. D., & Laurence, K. M. (1974). Theeffectofhypoxia
288
CHAPTER 15
and hypoglycemia in the brain of the newborn infant. Developmental Medicine and Child Neurology, 16, 308-319. Halgren, E., Squires, N. K., Wilson, C. L. Rohrbaugh, J. W., Babb, T. L., &Crandall P. H. (1980). Endogenous potentials generated in the human hippocampal formation and amygdala by infrequent events. Science, 210, 803-805. Hall, R. A., Griffm, R. B., Moyer, D. L., Hopkins, K. H., & Rappaport, M. (1976). Evoked potential stimulus intensity and drug treatment in hyperkinesis. Psychophysiology, /3, 405-418. Halliday, R., Callaway, E., & Naylor, H. (1983). Visual evoked potential changes induced by methylphenidate in hyperactive children: Dose/response effects. Electroencephalography and Clinical Neurophysiology, 55, 258-267. Halliday, R., Callaway, E., & Rosenthal, J. H. (1984). The visual ERP predicts clinical respOnse to methylphenidate in hyperactive children. Psychophysiology, 21, 114-121. Hanley, J., & Sklar, B. (1976). Electroencephalographic correlates of developmental reading dyslexias: Computer analysis of recordings from normal and· dyslexic children. In G. Leisman (Ed.), Basic visual processes and learning disability. Springfield, IL: Thomas. Hansen, J. C., & Hillyard, S. A. (1980). Endogenous brain potentials associated with selective auditory attention. Electroencephalography and Clinical Neurophysiology, 49, 277-290. Hansen, J. C., & Hillyard, S. A. (1984). Effects of stimulation rate and attribute cuing on event-related potentials during selective auditory attention. Psychophysiology, 21, 394405. Harmony, T. ( 1984a). Neurometric assessment of brain dysfunction in neurological patients. Hillsdale, NJ: Etlbaum. Harmony, T. (1984b). Neurometria y Maduracion Cerebral. Neurologfa, Neurocirugfa y Psiquiatr(a, 25, 7-12. Harmony, T., Alvarez, A., Pascual, R., Ramos, A., Marosi, E., Dfaz de LeOn, A. E., Valdes, P., & Becker, J. (1988). EEG maturation on children with different economic and psychosocial characteristics. International Journal of Neurosciences (accepted for publication). Harmony, T., Becker, J., Marosi, E., Landazuri, A. M., Baiiuelos, E., Dfaz de LeOn, A. E., & Hinojosa, G. (1985). EEG, visual evoked responses and brain stem evoked responses in children with risk factors of brain damage. Electroencephalography and Clinical Neurophysiology, 61,
Sl28. Harmony, T., & Dfaz de LeOn, A. E. (1982). Visual evoked responses in children with learning disorders. Second International Evoked Potential Symposium, Cleveland, Ohio. Harmony, T., & Diaz de LeOn, A. E. (1983). La variabilidad de los potenciales relacionados a eventos como un signo de disfunci6n cerebral. Tercer Coloquio 1nterno de 1nvestigaci6n de la ENEP UNAM, lztacala, Mexico. Harmony, T., Marosi, E., Diaz de LeOn, A. E., Becker, J., & Landazuri, A. M. (1984). Estudio de los potenciales e11ocados visuales en niiios con problemas del aprendizaje. XXIII International Congress of Psychology, Acapulco, Mexico. Harmony, T., Ricardo, J., Otero, G., Fernandez, G., & Valdes, P. (1973). Symmetry of the visual evoked potentials in normal subjects. Electroencephalography and Clinical Neurophysiology, 35, 237-240. Hillyard, S. A., Hink, R. F., Schwent, V. L., & Picton, T. W.
(1973). Electrical signs of selective attention in the human brain. Science, 182, 177-180. Hillyard, S. A., & Kotas, M. (1983). Electrophysiology of cognitive processing. Annual Review ofPsychology. 34. 33-61 . Hillyard, S. A., Picton, T. W., & Regan, D. M. (1978). Sensation, perception and attention: Analysis using ERPs. In E. Callaway, P. Tueting, & S. H. Koslow (Eds.), Event-related potentials in man. New York: Academic Press. Hiramatsu, K., Kameyama, T., Niwa, S., Saitoh, 0., Rymar, K., & Itoh, K. (1983). Schizophrenic deficits in information processing as reflected in event-related potential abnormalities during syllable discrimination tasks. Advances in Biological Psychiatry, 13, 63-74. Hughes, J. R. (1971). Electroencephalography and learning disabilities. In H. R. Myklebust (Ed.), Progress in learning disabilities (Vol. 2). New York: Grone & Stratton. John, E. R., Ahn, H., Prichep, L., Trepetin, M., Brown, D., & Kaye, H. (1980). Developmental equations for the electroencephalogram. Science, 210, 1255-1258. John, E. R., Prichep, L., Ahn, H., Easton, P., Fridman, J., & Kaye, H. (1983). Neurometric evaluation of cognitive dysfunctions and neurological disorders in children. Progress in Neurobiology, 21, 239-290. Johnson, R., & Fedio, P. (1984). ERP and P300 activity in patients following unilateral temporal lobectomy. Society Neuroscience Abstracts, 10, 847. Johnstone, J., Galin, D., Fein, G., Yingling, G., Herron, J., & Marcus, M. (1984). Regional brain activity in dyslexic and control children during reading tasks: Visual probe event related potentials. Brain and Language, 21, 233-254. Jutai, J. W., Gruzelier, J. H., Connolly, J. F., Manchanda, R., & Hirsch, S. R. (1984). Schizophrenia and spectral analysis of the visual evoked potential. British Journal of Psychiatry. 145, 496-501. Karrer, R., Warren, C., & Ruth, R. (1978). Slow potentials of the brain preceding cued and noncued movement: Effects of development and mental retardation. In D. A. Otto (Ed.), Multidisciplinary perspectives in event-related brain potential research. Washington, DC: EPA-600/9-77-043. Kellaway, P., & Fox, B. J. (1952). Electroencephalographic diagnosis of cerebral pathology in infants during sleep. I. Ra-
tionale, technique and the characteristics of normal sleep in infants. Journal of Pediatrics, 4, 262-287. Klorman, R., Salzman, L., Bauer, L., Coons, H., Borgstedt, A., & Halpern, W. (1983). Effects of two doses of methylphenidate in cross-situational and borderline hyperactive children's evoked potentials. Electroencephalography and Clinical Neurophysiology, 56, 169-185. Knight, R. T. (1984). Decreased response to novel stimuli after prefrontal lesions in man. Electroencephalography and Clinical Neurophysiology, 59, 9-20. Knight, R. T., Hillyard, S. A., Woods, D. L., & Neville, H. J. (1981). The effects of frontal cortex lesions in event-related potentials during auditory selective attention. Electroencephalography and Clinical Neurophysiology, 52. 571-582. Kok, A., & Rooijakkers, J. A. J. (1985). Comparison of eventrelated potentials of young children and adults in a visual recognition and word reading task. Psychophysiology, 22. 11-23. Kotas, M., & Hillyard, S. A. (1980). Reading senseless sentences:
PSYCHOPHYSIOLOGICAL EVALUATION Brain potentials reflect semantic incongruity. Science. 207, 203-205. Lelord, G., Laffont, F., Jusseaume, P., & Stephan!, J. (1973). Comparative study of conditioning of AER by coupling sound and light in normal and autistic children. Psychophysiology, 10, 415-425. Lezny, L., Provasnik, K., Jirasek, J., & Komarek, L. (1977). The value of EEG, especially of hyperventilation test in learning disabilities. Activitas Nervosa Superior. 19, 263-264. Loiselle, D. L., Stamm, J. S., Maitinsky, S., & Whipple, S.C. (1980). Evoked potentials and behavioral signs of attentive dysfunctions in hyperactive boys. Psychophysiology, 17, 193-201. Marczynski, T. J. (1978). Neurochemical mechanisms in the genesis of slow potentials: A review and some clinical implications. In D. A. Otto (Ed.), Multidisciplinary perspectives in event-related brain potential research. Washington, DC: EPA-600/9-77-043. Martineau, J., Garreau, B., Barthelemy, C., & Lelord, G. (1984). Evoked potentials and P300 during sensory conditioning in autistic children. Annals of the New York Academy of Sciences, 425, 362-369. Martineau, J., Laffont, F., Bruneau, N., Roux, S., & Lelord, G. (1980). Event-related potentials evoked by sensory stimulation in normal, mentally retarded and autistic children. Electroencephalography and Clinical Neurophysiology, 48, 140153. Mason, S.M., & Mellor, D. H. (1984). Brain-stem, middle latency and late cortical evoked potentials in children with speech and language disorders. Electroencephalography and Clinical Neurophysiology, 59, 297-309. Matousek, M., & Petersen, I. (1971). Objective measurement of maturation defects and other EEG abnormalities by means of frequency analysis. V WorldCongressofPsychiatry, Mexico. Matousek, M., & Petersen, I. (1973). Frequency analysis of the EEG in normal children and adolescents. In P. Kellaway & I. Petersen (Eds.), Automation of clinical electroencephalography. New York: Raven Press. Michael, R. L., Klorman, R., Salzman. L. F., Borgstedt, A. D., & Dainer, K. B. (1981 ). Normalizing effects of methylphenidate on hyperactive children's vigilance performance and evoked potentials. Psychophysiology, 18, 665-671. Murdoch, B. D. (1974). Changes in the electroencephalogram in minimal cerebral dysfunction: A controlled study over 8 months. South African Medical Journal, 48, 606-610. Musso, M. F., & Harter, M. R. (1978). Contingent negative variation, evoked potential and psychophysical measures of selective attention in children with learning disabilities. In D. A. Otto (Ed.), Multidisciplinary perspectives in event-related brain potential research. Washington, DC: EPA-600/9-77043. Myers, R. E. (1972). Two pattel;lls of perinatal brain damage and their conditions of occurrence. American Journal of Obstetrics and Gynecology, 112, 246-276. Niiiitiinen, R., & Michie, P. T. (1979). Early selective attention effects on the evoked potential: A critical review and interpretation. Biological Psychology, 8, 81-136. Netchine, S., & Lairy, G. C. (1975). The EEG and psychology of the child. Handbook ofElectroencephalography and Clinical Neurophysiology (Vol. 6B) 69-104.
289
Neville, H. J., Kutas, M., & Schmidt, A. (1984). Event-related potential studies of cerebral specialization during reading. Annals of the New York Academy ofSciences, 425, 370-376. Novick, 8., Vaughan, H. G., Kurtzberg, D., &Simon, R. (1980). An electrophysiological indication of auditory processing defects in autism. Psychiatric Research, 3, 107-I 15. Omitz, E. M., Tanguay, P. E., Lee, J. C. M., Ritvo, E. R., Sivertsen, B., & Wilson, C. (1972). The effect of stimulus interval on the auditory evoked response during sleep in autistic children. Journal of Autism and Childhood Schizophrenia, 2, 140-150. Pascual, R. D., Valdes, P., & Alvarez, A. (1985). Multivariate spectral modelling of the EEG: The xi-alpha model. Neuroscience Branch Technical Report No. Nc-()()1, National Research Center of Cuba, Havana. Petersen,!., & Eeg-Oiofsson, 0. (1971). The development of the EEG in normal children from the age of I through 15 years: Non-paroxysmal activity. Neuropiidiatrie, 3, 277-304. Petersen, I., Sellden, U., & Eeg-Oiofsson, 0. (1975). The evolution of the EEG in normal children and adolescents from I to 21 years. Handbook of Electroencephalography and Clinical NeJlrophysiology (Vol. 68) 31-68. Preston, M.S., Guthrie,J. T., &Childs, B. (1974). Visual evoked responses in normal and disabled readers. Psychophysiology, ll. 452-457. Prichep, L., Sutton, S., & Hakerem, G. (1976). Evoked potentials in hyperkinetic and normal children under certainty and uncertainty: A placebo and methylphenidate study. Psychophysiology, 13, 419-428. Proulx, G. B., & Picton, T. W. (1980). The CNV during cognitive learning and extinction. In H. H. Komhuber & L. Deecke (Eds.), Motivation, motor and sensory processes ofthe brain: Potentials, behavior, and clinical use. Amsterdam: Elsevier. Rebert, C. S., Wexler, B. N.,&Sproul,A. (1978). EEGasymmetry in educationally handicapped children. Electroencephalography and Clinical Neurophysiology, 45, 436-442. Rhodes, L. E., Dustman, R. E., &Beck, E. C. (1969). The visual evoked response: A comparison of bright and dull children. Electroencephalography and Clinical Neurophysiology, 27, 364-372. Rizzo, P. A., Spadoro, M., Albani, G. F., & Morocutti, C. (1984). Contingent negative variation and schizophrenia: A long term follow-up study. Biological Psychiatry, /9, 17191724. Rosenblum, S. M., Arick, J. R., King, D. A., Stubbs, E. G., Young, N. B., & Pelson, R. 0. (1980). Auditory brainstem evoked responses in autistic children. Journal of Autism and Developmental Disorders, 10. 215-225. Roth, W. T., Pfefferbaum, A., Horvath, T. B., Berger, P. A., & Berts, K. (1980). P3 reduction in auditory evoked potentials of schizophrenics. Electroencephalography and Clinical Neurophysiology, 49, 497-505. Rumsey, J. M., Grimes, A. M., Pikus, A. M., Duara, R., & Ismond, D. R. (1984). Auditory brainstem responses in pervasive developmental disorders. Biological Psychiatry, 19. 1403-1418. Saitoh, 0., Niwa, S., Hiramatsu, K., Kameyama, T., Rymar, K., & Itoh, K. ( 1984). Abnormalities in late positive components of event related potentials may reflect a genetic predisposition to schizophrenia. Biological Psychiatry, 19. 293-303.
290
CHAPTER 15
Saletu, B., Saletu, M., Simeon, J., Viamontes, G., & Itil, T. M. sustained potential shifts. In D. A. Otto (Ed.), Multidisciplin(1975). Comparative symptomatological and evoked potenary perspectives in event-related brain potential research. tial studies with d-amphetamine, thioridazine and placebo in Washington, DC: EPA-600/9-77-043. hyperkinetic children. Biological Psychiatry, 10, 253-275. Squires, N., Aine, C., Buchwald, J., Norman, R., & Galbraith, Satterfield, J., & Braley, B. (1977). Evoked potentials and brain G. (1980). Auditory brain stem response abnormalities in maturation in hyperactive and normal children. Psychoseverely and profoundly retarded adults. Electroenphysiology. 43, 43-51. cephalography and Clinical Neurophysiology, 50, 172-185. Satterfield, J. H., Schell, A.M., Backs, R. W., & Hidaka, K. C. Srebro, R. (1985). Localization of visually evoked cortical activity (1984). A cross sectional and longitudinal study of age effects in humans. Journal of Physiology (London), 360, 233-246. of electrophysiological measures in hyperactive and normal Strandburg, R. J., Marsh, J. T., Brown, W. S., Asamow, R. F., & children. Biological Psychiatry, 19, 973-990. Guthrie, D. (1984). Event-related potential concomitants of Shagass, C., Roemer, R. A., Straumanis,J. J., &Josiassen, R. C. information processing dysfunction in schizophrenic chil(1981). Intelligence as a factor in evoked potential studies of dren. Electroencephalography and Clinical Neurophysipsychopathology. I. Comparison oflow and high IQ subjects. ology, 57, 236-253. Biological Psychiatry, 16, 1007-1029. Sutton, S., Tueting, P., Zubin, J., &John, E. R. (1967). InformaShelburne, S. A. (1973). Visual evoked responses to language tion delivery and the sensory evoked potentials. Science, /55, stimuli in normal children. Electroencephalography and 1436-1439. Clinical Neurophysiology, 34, 135-143. Tanguay, P. E., Edwards, R, M., Buchwald, J., Schwafel, J., & Shelburne, S. A. (1978). Visual evoked potentials to language Allen, V. (1982). Auditory brainstem evoked responses in stimuli in children with reading disabilities. In D. A. Otto autistic children. Archives of General Psychiatry, 39, 174(Ed.), Multidisciplinary perspectives in event-related brain 180. potential research. Washington, DC: EPA-600/9-77-043. Tecce, J. J., Savignano-Bowman, J., & Meinbresse, D. (1976). Shucard, D. W., Cummins, K. R., & McGee, M. G. (1984). Contingent negative variation and the distraction-arousal hyEvent-related brain potentials differentiate normal and dispothesis. Electroencephalography and Clinical Neuabled readers. Brain and Language, 21, 318-334. rophysiology, 41, 277-286. Simon, N. ( 1975). Echolalic speech in childhood autism. Archives Timsit-Berthier, M., Mantanus, H., Ansseau, M., Doumont, A., of General Psychiatry, 32, 1437-1442. & Legros, J. J. (1983). Methodological problems raised by Sklar, B., Hanley, J., & Simmons, W. W. (1973). A computer contingent negative variation interpretation in psychoanalysis of EEG spectral signatures from normal and dyslexic pathological conditions. Advances in Biological Psychiatry, children./£££ Transactions on Biomedical Engineering, 20, 13, 80-92. 20-26. Valdes, P., Biscay, R., Pascual, R., Jimenez, J. C., & Alvarez, Skoff, B. F., Mirsky, A. F., & Turner, D. (1980). Prolonged A. (1985). A quantitative description of the developmental brainstem transmission time in autism. Psychiatric Research, neurornetric parameters. Neuroscience Branch Technical 2, 157-166. Report No. Nc-003, National Research Center of Cuba, Havana. Small, I. G., De Myer, M. K., & Milstein, V. (1971). CNV responses of autistic and normal children. Journal of Autism Van den Bosch, R. J. ( 1984). Contingent negative variation: Comand Childhood Schizophrenia, 1, 215-231. ponents and scalp distribution in psychiatric patients. Biological Psychiatry, 19, %3-912. Sobotka, K. R., & May, J. G. (1977). Visual evoked potentials and reaction time in normal and dyslexic children. Psycho- Walter, W. G., Cooper, R., Aldridge, V. J., McCallum, W. C., & Winter, A. L. (1964). Contingent negative variation: Aa physiology, 14, 18-24. electric sign of sensorimotor association and expectancy in Sohmer, H., & Student, M. (1978). Auditory nerve and brain stem the human brain. Nature, 203, 380-384. evoked response in normal, autistic, minimal brain dysfunction and psychomotor retarded children. Electroen- Yingling, C. D., & Hosobuchi, Y. (1984). A subcortical correlate of P300 in man. Electroencephalography and Clinical Neucephalography and Clinical Neurophysiology, 44, 380-388. rophysiology, 59, 12-16. Somjen, G. G. ( 1978). Contribution of neuroglia to extracellular
16 Techniques of Localization in Child Neuropsychology GRETA N. WILKENING
The ability to relate constellations of neuropsychological deficit and strength to localized structural brain damage has been a major achievement in the endeavor to understand how the brain functions and to increase the utility of behavioral tools for understanding clinical problems. The majority of research dealing with localization has involved the study of adults with pathological changes of the brain. Less information is available regarding localized deficits in children. Satz and Bullard-Bates (1981), for example, point to the relative paucity of studies concerning acquired childhood aphasia in contrast to numerous studies examining aphasia, and the effect of varying lesion localizations, in the adult population. This relative inequality reflects a number of factors including the types of lesions typically acquired by children, how they differ from lesions acquired by adults, and the relatively greater difficulty involved in assessing and understanding a developing organism (Boll & Barth, 1981). Given the limited data regarding localization of function in children, it is tempting to assume that the integration of structure and function in the child's brain is identical to what has been described in the adult brain, and to utilize similar techniques for aiding in diagnosis, localization, understanding disabilities, and suggesting remediation. There is, however, abundant evidence that functional localization within the child's brain may be different from the relationship seen in the adult brain although the magnitude of this difference is disputed. Techniques for localization in children must be based not only on an awareness of lateralized and intrahemispheric patterns of presentation, such as those so important in understanding the effects of structural brain damage GRETAN. WILKENING • DepartmentofNeurology, The Children's Hospital, Denver, Colorado 80218.
in the adult population, but must also account for the research suggesting that there is an interaction between lesion localization and age at time of injury that mediates where in the brain cognitive and perceptual processes occur subsequent to early brain damage (Fedio & Mir8ky, 1969). Techniques for lesion localization in children are intimately related to the research addressing the possibility of plasticity and equipotentiality in the young brain. For example, if children who sustain early lateralized brain damage rarely show long-term unilateral sensory deficits (e.g., plasticity of function), then the failure to identify lateralized sensory findings cannot be interpreted as consistent with lack of lateralized damage. Similarly, if language disorders are as common secondary to early right as early left hemisphere lesions (e.g. , equipotentiality), then the presence of a language disorder cannot be interpreted as strongly indicative of a left hemisphere lesion, as would be interpreted in an adult. Adequate appreciation of the current research on equipotentiality and plasticity is essential in understanding both the limitations and the possibilities for lesion localization in the younger age group.
Equipotentiality and Plasticity in the Young Brain There has been abundant discussion and debate about whether the young brain is equipotential (i.e., are all parts of the brain uniformly competent to support specific functions, such as language?) and whether it is more plastic than the adult brain (does it recover more readily than the adult brain from similar injuries?). Perhaps the easiest group of patients to use to discuss the equipotentiality hypothesis are those who have had hemispherectomies (St. James-Rob291
292
CHAPTER 16
erts, 1979, 1981). This group of patients allows us to assess whether either hemisphere can support, with equal success, language and other functions thought to have lateralized representation within the adult brain. In a series of studies evaluating individuals who received hemispherectomies prior to 6 months of age, who have normal intelligence and eventually adequate school performance, Dennis and her coworkers convincingly demonstrated that language can be supported in the isolated right hemisphere (e.g., Dennis & Kohn, 1975). Children who received early left hemispherectomies, so that we know that language is supported by an isolated right hemisphere, learn to speak and understand language at a level sufficient for everyday discourse. Nevertheless, when compared to individuals who had early right hemispherectomies, those with early left hemispherectomies show subtle language disorders as evidenced by differences in the rate of development of certain language skills, the time taken to complete tasks of syntactic understanding (Dennis & Kohn, 1975), and differences in proficiency in understanding complex syntax (Dennis & Whitaker, 1977). Syntactic deficits are evident in a variety of contexts including production, repetition, comprehension, awareness of anomalies, and judgment of word relationships (Dennis & Whitaker, 1976). The written language of patients with an isolated right hemisphere reflects a similar configuration (Dennis, Lovett, & Weigel-Crump, 1981). Individuals with early left hemidecortectomies learn to read and spell, however, the type of errors they make suggests that the isolated right hemisphere does not have access to higher-order linguistic rules. For example, when asked to assess if a string ofletters could be a word (e.g. , is the order and grouping of letters in nonwords permissible in the English language?) the left hemidecorticate subjects are less sensitive to the linguistically probable sequence ofletters in English. Less information is available about the development of ''right hemisphere skills'' in the isolated left hemisphere. Kohn and Dennis (1974) reported that both right and left hemidecorticate subjects perform competently on tasks of personal spatial organization, and identification of fragmented visual stimuli in overlapping figures. However, the right hemidecorticates (e.g., an isolated left hemisphere) have significant difficulties on tasks requiring the use of spatial organization skills in more complex test situations. For example, the isolated left hemisphere performs less competently than the isolated right hemisphere on tasks such as mazes. Kohn and Dennis concluded that although the isolated left hemisphere
supports basic visual spatial skills, it is unable to sustain right hemisphere functions that generally develop in the second decade of life. In summary, neuropsychological test data suggest that the two hemispheres are not equipotential. Although language develops in the isolated right hemisphere, comprehension and production, both written and oral, show subtle disturbances. Similarly, the isolated left hemisphere, although capable of basic directional and visual spatial perception, cannot as easily mediate complex visual spatial decisions. The pattern is similar to what is seen in adults who sustain later focal brain damage, although it clearly represents not only an attenuated pattern, but a different pattern (Dennis & Kohn, 1974). Other data supporting the idea that the two hemispheres are not equipotential even at birth included structural (Geschwind & Galaburda, 1985) and physiological (Witelson, 1977) differences between the hemispheres that exist from birth, or from very early in life. If the brain is not equipotential, i.e., specific areas are neuroanatomically designed to support specific functions, how do children with early lesions acquire functions that should have been mediated by damaged areas? It has been suggested that the young brain has greater plasticity, that functional reorganization is possible, with recommitment of neural tissue to .compensate for damaged brain (Chelune & Edwards, 1981). Recommitment is thought possible only when the neural substrate that is to substitute for damaged brain is not yet consistently supporting other cortical functions (Finger & Stein, 1982). If recommitment can occur, there must be a timeline that would help us to understand at what ages the process of anomalous localization of function is possible, and when these changes are no longer viable, with damage that may occur producing behavioral patterns more similar to what is seen in the adult population. Unfortunately, there is no firmly demonstrated or consistently agreed upon timeline to describe when functional reorganization in humans is possible (Hecaen, 1976). The largest set of data pertains to recovery of function in childhood aphasia, but even here the time constraints are unclear with authors variously citing from 2 years to puberty as the point beyond which reorganization is impossible (Satz & Bullard-Bates, 1981). In considering the issue of plasticity and recommitment and the research regarding children's recovery, it must be remembered that there are alternative hypotheses explaining why children recover from brain damage differently than do adults (Finger &
TECHNIQUES OF LOCALIZATION
Stein, 1982; Chelune & Edwards, 1981). In fact, there is some evidence that functional reorganization, such as displacement of language to the right hemisphere, is relatively rare (Milner, 1974 ). Additionally, there are other researchers who maintain that differential age effects reflect only methodological error, not biological reality (St. JamesRoberts, 1979, 1981). An important issue in considering localization of function in children are the data suggesting that the most pronounced and consistent sequela of early brain damage is a general decline in intellectual function (Milner, 1969; Reitan, 1974; Rutter, 1982). Even in those studies that demonstrate that the pattern of perceptual and cognitive disabilities in early focal acquired lesions is similar to what is seen in adults, there is an overall decrement in measured intelligence when compared to controls (Aram, Kelman, Rose, & Whitaker, 1985). Woods (1980), for example, compared children with early unilateral (evident prior to 1 year of age) or late unilateral (evident after the first year of age; right hemisphere damage mean age of onset 6.4 years, left hemisphere damage mean age of onset 5. 7 years) damage to the performance of their closest in age sibling. Both the early right and early left hemisphere-damaged groups differed from their sibling controls on verbal and performance IQ. The late left group differed significantly from their sibling controls on both verbal and performance IQ. The late right group differed at a significant level only on performance IQ. Rutter (1982) similarly found an age effect, with early localized lesions more likely to produce a decrement on tests of intellectual functioning. Despite the paucity of consistent data, and the many questions that remain unanswered, some conclusions are possible that may serve as a guide in clinical practice. It seems that there are consistent patterns that emerge, even when young children sustain brain damage, with left hemisphere lesions producing disturbances in higher-level linguistic func-tions including language-based academic skills, and right hemisphere lesions producing subtle difficulties in the organization of extrapersonal space. These deficits are both le:;;s striking and less consistent than what is seen in adults who sustain similar injuries (Rutter, 1982; McFie, 1961). The specific deficits are often superimposed on a pattern of an overall decrement in measured intelligence. The patterns that emerge seem to change with age so that the older a child when sustaining a focal injury, the more likely he or she is to present with a pattern that is similar to brain-behavior relationships as they are manifested in adults. In terms of clinical utility, it must also be
293
remembered that children more rarely than adults, present with acute, progressive, and focal cortical disorders such as those disorders that produce the specific patterns seen in adulthood (Reitan, 1972). This reflects both the type of injuries and illnesses most common in children, and the fact that the period between injury and evaluation is often much longer in children. For all these reasons, localization of function in children can be a perilous and difficult enterprise. Given that children with early focal brain injury sometimes manifest a pattern of focal-looking deficits, what can be said about how to make sense of the data regarding a specific child? A review of the literature clearly demonstrates that there have been few attempts to discuss techniques for localization in child neuropsychology and only rare descriptions of specific patterns of deficit that are secondary to focal injuries in children. There are numerous studies (e.g., Dennis, 1977;Pennington, VanDoornink, McCabe, & McCabe, 1985) that have looked at groups of children with nonfocal brain damage (at least at the structural level) and suggested that the patterns of cognitive and perceptual deficiencies are similar to adults with lesions in specific areas. Similarly, many studies have discussed children with no known damage, but who have learning disabilities, and related their pattern of disturbance to what is seen in adults with focal injuries (e.g., developmental Gerstmann's syndrome). There are far less data, however, discussing specific syndromes in children with known focal lesions. In the next section, the data that are available will be discussed in terms of functions rather than in terms of specific brain areas. Subsequent to that, several cases including analysis of localized deficits will be presented.
Motor Functions Assessment of motor skills has been a standard part of neuropsychological evaluation. Traditional forms of assessment include evaluation of motor speed (e.g., the Finger Tapping or Finger Oscillations Test), dexterity (e.g., the Grooved Pegboard Test), and strength. These measures are useful in localization if they are suggestive of unusual left/right differences. There have been a number of studies evaluating the normal level of difference between hand perfor~ mance at different ages (Spreen & Gaddes, 1969; Denckla, 1973). Denckla (1973) noted that there appears to be a normal developmental sequence de-
294
CHAPTER 16
scribing the difference between hand performance on repetitive tasks. In very young children, right-sided (most often) functions are established first, with a wide gap between left- and right-sided performance. Left-sided performance then increases rapidly so that although right-sided performance continues to be superior, the asymmetry is not as great. Annett (1970), however, found no systematic change in the difference in performance between hands across ages. Typically, the difference between hands in normal children between 5 and 13 years of age has been reported to be 10-20% (Spreen & Gaddes, 1969; Denckla, 1973). Differences between dominant and nondominant hand performance greater than 20% suggest a relative weakness in the nondominant hand. This would be consistent with damage to the hemisphere contralateral to the nondominant hand. Differences between hands that are less than expected suggest poorer than normal performance of the dominant hand. This could be indicative of damage to the dominant, most often left, hemisphere. Conclusions must be cautious, within the context of other data. There is some suggestion that in a normal population, the disparity between left and right hand performance is less for young males than for young females (Annett, 1970). There have been a number of studies indicating congruence between hand performance in children and other neuropsychological deficits presumed to accompany lateralized dysfunction. When locus is defined by the presence of hemiplegia (Annett, 1973; Kiessling, Denckla, & Carlton, 1983; Kershner & King, 1974), numerous studies have reported poorer speech and language skills in children with rightsided motor deficits than with left-sided motor dysfunction. Rourke, Yanni, MacDonald, and Young (1973), studying learning-disabled children with presumed, but nondemonstrable, "cortical dysfunction,'' found poorer right-handed motor performance in children with cognitive deficits presumed mediated by the left hemisphere (e.g., reading and spelling). The differential hand performance was seen only on more complex measures of motor performance (Grooved Pegboard), but not on measures of simple motor function (as in finger tapping) (Rourke & Strang, 1978). This finding must be used with circumspection as it is based on consistency of neuropsychological deficits, not structural or physiological evidence of brain lesions. When interpreting lateralizing motor deficits or the lack of such in clinical practice, several caveats are warranted. Differences are interpretable only if they are consistent across measures. Though neuropsychologists emphasize (Boll, 1981) that left-right
difference are one of the four means of inference used in neuropsychological diagnosis, lack of lateralized motor deficits is not always consistent with the absence of brain damage, nor with the absence of lateralized brain damage, especially when a child is seen after a period of recovery. Certainly when lateralized motor deficits are clearly demonstrable, one's confidence in understanding the data is increased. Nevertheless, the pattern of recovery seen in young children must be remembered. Simple, more basic functions recover better than more complex skills (Wilkening & Golden, 1982), so that often motor recovery will be quite good. Cortical lesions recover better than subcortical lesions (Goldman, 1974) so that lateralized motor difficulties do not necessarily suggest that consistent cognitive deficiencies will be seen. During evaluation of motor functions, qualitative observations are possible, and important. Mirror movements are normal for the first decade of life. Nevertheless, the consistent and lateralized presence of mirror movements (e.g., unintentional, unconscious movements of the hand that the child is not trying to move) may be suggestive of damage in the hemisphere ipsilateral to the hand unintentionally moving (Woods & Teuber, 1978b).
Deficits Tactile Pathognomonic signs of brain damage such as consistent defects of simple sensory sensation, and tactile extinctions are important diagnostically, and may be indicative of localized lesions. When demonstrated, the localization indicated is the same as that suggested by these symptoms in the adult population, for example, a lesion of the parietal lobe contralateral to the neglected or insensitive side. Such signs are rarely seen in children, however, probably because children less often sustain focal lesions, and are most often evaluated when the injury has become chronic. A sensory exam should be included in all neuropsychological batteries. Standard evaluations often include assessment of tactile sensitivity, the presence of finger agnosia, attention to double simultaneous stimulation, graphesthesia, and stereognosis. The stimuli used and the response mode required depend upon the developmental level of the child being assessed, and the child's status insofar as his or her ability to respond. Kinsbourne and Warrington (1962, 1963b) described a means of assessing "finger sense" (topographic organization of tactile stim-
TECHNIQUES OF LOCALIZATION
uli) so that adequate performance is not dependent upon naming of fingers, a task often too difficult for younger children. They found that by 6 years of age, most of the children assessed were able to successfully complete their tasks of finger differentiation and could assess the spatial relationship between fingers. "Finger sense" is not the same as finger localization as described by Benton (1955), but also may be related to adequate function of the tertiary parietal areas. In completing an assessment of tactile functions in children, extreme care must be taken not to confuse inattention with poor sensory performance. Unilateral difficulties of simple tactile sensation (touch sensation, increased two-point discrimination, finger agnosia) are indicative of injury to the posterior portion of the contralateral hemisphere. Trouble with the more complex integrative tactile task of stereognosis is more difficult to lateralize. Unilateral deficits suggest injury to the parietal lobe of the contralateral hemisphere. Bilateral difficulties may, however, be indicative of severe right hemisphere dysfunction, as the right hemisphere appears dominant for tasks necessitating spatial problem solving in adults (Benton, Hamsher, Varney, & Spreen, 1983; Rapin, 1982). There is some suggestion that the same pattern is true for children (Rudel, Teuber, & Twitchell, 1974). The adult literature also suggests that bilateral disorders of topographic organization of tactile stimuli (e.g., difficulties with bilateral finger agnosia) seen as part of Gerstmann' s syndrome are a consequence of lesions of the left angular gyrus (Strub & Geschwind, 1983). Kinsboume and Warrington ( 1962) and Benson and Geschwind ( 1970) feel that although a developmental Gerstmann's variant exists, no known localized pathology can be posited, despite the similarity to the adult symptoms. Interpretation of localization based on the results of a sensory exam must take into account the chronicity of the condition, and whether the child has sustained a known injury or is presumed to have a developmental anomaly. The tactile consequences, for example, of congenital absence of the corpus callosum, a developmental disorder in which interhemispheric transfer of information is limited, are different from what is seen in an adult whose corpus callosum is surgically divided. Such children can complete tasks of intramanual transfer of information (Dennis, 1976). Similarly, early brain damage does not have the effect of increasing the threshold for tactile stimuli so that errors by children with known, sometimes lateralized, brain damage on tasks assessing elementary sensory functions (two-point discrimination, position sense, localization with double si-
295
multaneous stimulation) are rare (Rudel et at .• 1974). Nici and Reitan ( 1986) reported that sensory tasks are ineffective in discriminating brain-damaged from normal children. Unlike adults with brain injury who often have deficits in sensitivity, children rarely make errors of detection, though they do have comparable difficulty on higher-level integrative tasks.
Visual Perception As in the other modalities, pathognomonic signs of brain damage such as visual field defects are useful for diagnosis and topographic localization, but are infrequently seen in children, especially those with chronic lesions. Primary disturbances of visual functioning have the same localizing significance in children as they would in adults. A left homonymous hemianopsia is characteristic of pathological changes affecting the visual fibers of the right hemisphere, posterior to the optic chiasm. A right homonymous hemianopsia suggests pathological changes in the left hemisphere. The quality of the hemianopsia, and the accompanying neuropsychological changes vary according to whether the lesion is more anterior (e.g. , temporal) or more posterior (parietal or occipital) (Martyn, 1975). A binasal hemianopsia requires involvement of bilateral temporal visual fibers simultaneously, and can be secondary to papilledema. Bitemporal hemianopsias occur secondary to lesions compromising the optic chiasm and are seen in children with pituitary tumors, though they occur most often as a consequence of a craniopharyngioma or optic glioma (Martyn, 1975). Lesions of the primary occipital cortex are less common in children than in adults (The Ophthalmologic Staff of the Hospital for Sick Children, 1967), though they are seen. A recently described, presumably genetic disorder, MELAS (mitochondrial encephalopathy, lactic acidosis, and stroke-like episodes) (Pavlakis, Phillips, Mauro, DeVivo, & Rowland, 1984), is notable for episodes of cortical blindness secondary to change in the primary visual cortex. Visual extinctions are rarer in children than in adults. When they are seen, visual extinctions are suggestive of parietal neglect. Visual neglect may sometimes be observed when children are completing the picture arrangement section of the WISC-R, the reading section of the WRAT, or the Peabody Picture Vocabulary Test. Neglect is manifest by consistently missing items on one side of the page or table. Informal cancellation tasks, for instance, having the child circle all the x 's on the page, can be used
296
CHAPTER 16
to identify visual field neglect. Obviously, such patterns are important to identify for safety reasons, even more so than for aiding in localization. Visual agnosia in children has not been described in the literature, though it is seen clinically. In my practice, two cases of associative visual agnosia secondary to confirmed damage have been identified. Associative visual agnosia is typified by inability to assign meaning to what one observes despite adequate visual perception (Alexander & Albert, 1983). In one case the bilateral occipital lesions that have been described as producing agnosia in adults were demonstrable. In this child the lesion was secondary to bilateral strokes, the pathological process often responsible for visual agnosia in adults. Only unilateral (right hemispheric) structural damage was demonstrable on Cf scan in the second child. The child had had herpes encephalitis, so that it is possible that bilateral lesions were present, though not seen on CT scan. Both children had accompanying prosopagnosia, and central achromatopsia. They were able to name colors in order to answer questions (e.g., "What color is grass?") and performed adequately on tasks such as the Ishihara plates, but were unable to match colors or name the color of an object they were shown. Associative visual agnosia is demonstrable on the visual section of the Luria-Nebraska Neuropsychological Battery (Golden, Hammeke, & Purisch, 1980). Such children cannot easily identify pictured objects, especially when they are presented with line drawings. They more easily identify real objects or photographs. Their performance on the Boston Naming Test may be disturbed. The naming errors they make reflect their visual deficits. For example, a pretzel may be called a worm, with correct naming possible once the function (e.g., "It's something you eat'') is described. I have found the Gestalt Closure Section of the Kaufman Assessment Battery for Children (K-ABC) (Kaufman & Kaufman, 1983) to be helpful in understanding such children. Qualitatively, the examiner may also observe errors on the Picture Completion and Picture Arrangement sections of the WISC-R. These children are unable to identify what is missing on Picture Completion because they are not sure what the picture represents, nor can they make sense of the order in Picture Arrangement because they have no idea what the story is supposed to be about. Sometimes their inability to recognize objects despite adequate perception is demonstrable on the Peabody Picture Vocabulary Test. Here their errors are secondary to visual errors, not receptive language difficulties. Adequate perfor-
mance on construction tasks, judging line orientation, and block design is seen. Prosopagnosia can be informally evaluated using family snapshots, and asking the children to identify family members. This will clearly demonstrate the deficit. The Faces and Places section of the K-ABC can also be used. Colored chips often used in Montessori classrooms, or paint chips may be used to assess color matching skills. In our Neuropsychology Lab, one case of apperceptive agnosia has been identified. Apperceptive visual agnosia is identified when the inability to recognize objects is a consequence of deficits in basic visual perception (Alexander & Albert, 1983). The child had aCT-documented lesion in the right parieto-occipital area. With this type of agnosia more basic deficits in perception are seen. The child was not able to name pictured objects, but also had difficulty judging the orientation of lines, or matching faces or pictures. Performance on the Gestalt Closure Section of the K-ABC was notably deficient. Identification of objects was attempted by picking out a detail, and then guessing at what the whole might be. Though patterns of deficit similar to what is seen in adults with right parieto-occipital and bilateral occipital damage are demonstrable, all the cases I have reported reflect lesions acquired after 5 years of age. Children with earlier acquired lesions do not generally demonstrate such circumscribed, specific deficits. Rudel and Teuber (1971), for example, found that children with early lesions had difficulty with a route-finding task of spatial orientation, regardless of the site of their lesion. The same task is differentially sensitive to parietal lesions, particularly right-parietal lesions in the adult population. All children with early brain damage performed poorly on this task, their mean performance being poorer than children whose mean chronological age was less than the mean mental age of the brain-injured children. Of interest is that when the brain-injured group is divided into two groups-those whose WISC Performance IQ was less than their Verbal IQ, and those whose Verbal IQ was less than their Performance IQ-the pattern is as one might anticipate based on adult performance. Poorer performance on the routefinding task is more clearly associated with poorer performance on Performance than on Verbal IQ. This pattern of performance-a relative impoverishment of Performance IQ compared to Verbal IQ-has frequently been reported in children with right-sided brain injuries (Woods & Teuber, 1973; Kershner & King, 1974; Annett, 1973; Fedio & Mirsky, 1969). It may be that what this represents is
TECHNIQUES OF LOCALIZATION
297
aphasia literature. There have been some recent attempts to describe specific types of aphasia in children, and to relate these to specific areas of damage. This represents not only an attempt to increase our understanding of localization of function within the immature brain, but also an alteration in thinking about acquired childhood aphasia. In the past the disorder has typically been described as homogeneous, regardless of intrahemispheric localization. The traditional description of childhood aphasia included rapid onset followed by a period of mutism or markedly decreased spontaneous language. Receptive language was said to be relatively more intact. Jargon aphasia with increased fluency was said to not occur in children. Resolution of the aphasia typically has been reported to be good, with rapid recovery. More recent reports suggest that there may be differences between aphasias associated with damage to specific brain areas even in children, and that recovery may not be as rapid nor as complete as has been previously reported, producing chronic patterns of disability. Language Skills Hecaen ( 1983) described differing presentations of aphasia in children on the basis of more anterior The study of language disorders in the adult versus more posterior localization. He found mutism population has had a major impact on our understand- as the presenting symptom to be more common with ing of how specific patterns of behavioral disorder frontal-Rolandic than temporal lesions, similar to the reflect localized brain damage. There are multiple pattern seen in adults. Similarly, dysarthria was schemes for classifying aphasia in adults and ongoing found to accompany 81% of the anterior lesions causcontention about the specific localization of various ing aphasia, but only 20% of the temporal lesions. In forms of aphasia. Nevertheless, in adult neuropsy- general, within Hecaen's (1983) population, all lanchology one is fairly confident that acute acquired guage functions were more disturbed in the presence language disorders suggest left hemisphere pa- of anterior left hemisphere lesions. When comthology. Similar conclusions may be possible in chil- prehension deficits were seen in aphasic children, dren, though the pattern may be attenuated and less they were more likely to be associated with temporal consistent. It certainly is more widely disputed. lobe disorders (Hecaen, 1976). Naming difficulties Historically, there has been the contention that were found to have no localizing significance, simacquired aphasia in children occurs as a consequence ilar to what is reported in the adult population. of either left or right hemisphere lesions (Basser, Aram, Rose, Rekate, and Whitaker (1983) de1962). Recently, this notion has been strenuously scribed an acquired aphasia occurring secondary to a disputed, and a number of reviews (Satz & Bullard- vascular lesion in the putamen, the anterior limb of Bates, 1981; Carter, Hohenegger, & Satz, 1982; the internal capsule, and the lateral aspect of the head Woods & Teuber, 1978a) rigorously demonstrate of the caudate. During the acute phase, the child had that when only clearly confirmed localized cases are a nonfluent aphasia, marked by both phonemic and considered, acquired aphasia in children, like apha- semantic paraphasias. (This is in contrast to Hecaen' s sia in adults, is most likely to occur secondary to a left report (1976) that paraphasias are rare in· childhood hemisphere lesion. aphasia.) In the acute phase, difficulty with comAcquired aphasia in children would then sug- prehension was demonstrable, but no dysarthria was gest that a pathological change to the left hemisphere present. Six and one-half months after onset, verbal has occurred. This makes the assessment of acute language abilities had recovered completely, reading problems relatively straightforward in terms of later- was normal, but spelling continued to be depressed. alization, but does not address intrahemispheric disFerro, Martins, Pinto, and Castro-Caldas ( 1982) tinctions. These distinctions are pursued in the adult also reported on an aphasia that occurred secondary to an attenuated presentation, a less differentiated though similar version, of what is seen in the adult population of individuals who have sustained right hemisphere lesions. Again, this conclusion must be tempered by other research that has found little association between Verbal and Performance discrepancies and site of lesion (Chadwick, Rutter, Thompson, & Shaffer, 1981). Care also must be taken in interpretation because a number of studies have shown that Performance skills appear not to recover as well as Verbal skills secondary to diffuse lesions (Levin, Benton, & Grossman, 1982; Chadwick et al., 1981). What is useful clinically from this array of data is that when Performance IQ is lower than Verbal IQ, one might hypothesize that there are signs of right hemisphere damage, and the rest of the data must be considered with this in mind. Other evidence is clearly required, however, to corroborate and support such a contention.
298
CHAPTER 16
a subcortical infarct (involving the internal capsule, lenticular nucleus, and the insula). However, this report is less useful because the child was left-handed, and the lesion right-sided. Though the hallmark of acquired childhood aphasia has been thought to be sparseness of overall output, Van Dongen, Loonen, and Van Dongen ( 1985) reported the occurrence of fluent aphasias associated with cr-documented lesions of the left temporal areas. In the three patients described, articulation was normal, with rapid rate of speech and normal prosody. Naming and repetition were disturbed, as was comprehension. Paraphasias and neologisms were prominent. Woods and Teuber (1978a) also describe a case of jargon aphasia secondary to a left hemisphere lesion, though the intrahemispheric localization is unknown. In my practice I have seen one child with a clear disturbance in phonemic discrimination and comprehension similar to what is seen in adults with fluent aphasias. cr and BEG results both suggested a left anterior temporal disruption. In contradistinction to Van Dongen and colleagues' case, the lesion was more anterior, and the child was dysfluent. Frequent paraphasias were present, however. There is some suggestion that the specificity of the language disorders seen in children secondary to focal damage varies as a function of age of onset (Hecaen, 1983). Some authors contend that younger children, those .less than 10 years of age or so, are more likely to have the classic pattern (e.g., initial mutism with little comprehension deficit and rare paraphasias), with older children demonstrating patterns that are more consistent with aphasias seen in the adult population (Satz & Bullard-Bates, 1981; Alajouanine & Lhermitte, 1965). This pattern is not reported by all authors. Similarly, there has been the suggestion that the pattern of recovery differs across ages, though again not all authors report this pattern (Woods & Teuber, 1978a). Most authors agree that even in the presence of good recovery of oral language, children who have had left hemisphere lesions with consequent aphasia are at great risk for ongoing academic deficits in mathematics and spelling, with reading difficulties often less extensive, but frequently seen (Hecaen, 1976, 1983; Alajouanine & Lhermitte, 1965). Given more recent reports, it is clear that when an evaluation reveals an acute language disturbance in a right-handed child, the likelihood of a left hemisphere disturbance is high. In children with known acquired lesions, it is often helpful, in taking the history, to ask about the child's pattern of recovery
and whether the child experienced a period of mutism or sparse language. Often this has been observed by the parents as the child recovers, without it ever being labeled "aphasia." Such information is useful in making a topical diagnosis. The evaluation of language in children requires assessment of both its receptive and expressive components. Adequate verbal IQ is not an indication of normal language functioning, though the presence of a significant Verbal-Performance discrepancy with Verbal lower than Performance is suggestive of language difficulties. This is especially true when the Full Scale IQ is within the above-average range. Evaluation of expressive language includes assessment of repetition, naming, and complex verbal formulation. Naming deficits can be assessed in young children using the Stanford-Binet Intelligence Scale or the K-ABC. There are norms available for the Boston Naming Test as well. Norms for agerelated changes in repetition of words and sentences are widely available. Complex verbal formulation can be observed on the WISC-R and in informal assessment. Evaluation of receptive language must not be confined to understanding of words or simple sentences. Informal assessment of receptive skills may fail to reveal profound difficulties in understanding complex syntactic structures, for in most informal situations multiple cues are provided to aid in understanding. The Token Test, the Receptive Language Scale of the Luna-Nebraska Neuropsychological Battery, and other formal measures of receptive language functions are available. Care must be taken in interpreting localized damage when the only indication of left hemisphere disturbances is a language disorder.
Academic Skills As previously noted, difficulties with reading, spelling, and arithmetic are frequent sequelae of left hemisphere damage that is acquired in childhood. This is true even in the absence of ongoing overt language disorders. Hecaen (1976) reported that arithmetic difficulties are the most frequent deficit, occurring in II of 15 cases in a population of children who had acquired aphasia earlier in their childhood. Woods and Carey (1978) reported ongoing spelling difficulties in children with a clinically good recovery from aphasia. Spelling difficulties were found even in the group of children who sustained left hemisphere lesions prior to l year of age, and who as a
TECHNIQUES OF LOCALIZATION
group had no other signs of an ongoing aphasic disturbance. Difficulties in acquiring spelling skills subsequent to early left hemispherectomy were also reported by Dennis et al. (1981). In comparing individuals with early right and left hemidecortectomies, they found that those with an isolated right hemisphere did not attend as automatically to grapheme-phoneme relationships, were Jess aware of the valid letter sequences occurring within words, and more frequently violated orthographic rules. The isolated right hemisphere, though able to read, acquires this skill more slowly, does not perform as automatically, and is less able to exploit the linguistic structure of a sentence in order to aid in fluent reading. Reading and spelling deficits were observed by Chadwick et al. (1981) in a group of children with focal injuries to the left hemisphere occurring secondary to depressed skull fractures with dural tears. As in other studies, academic deficits discriminated children with left hemisphere injuries from those with right hemisphere injuries, whereas other measures, e.g., WISC scores, measures of verbal fluency, Matching Familiar Figures, left/right disorientation, did not. Difficulty in acquiring academic skills secondary to earlier left hemisphere injuries is regularly reported. Acquired alexia in children is rarely reported. In adults, acquired alexia occurs secondary to a variety of lesions. Alexia with agraphia, marked by difficulty reading aloud as well as silently, and spelling deficits marked by better copying than writing to dictation, reflects damage to the parietal temporal area of the left hemisphere (Benson, 1979; Greenblatt, 1983). Some degree of fluent aphasia is almost invariably associated with alexia with agraphia. Alexia without agraphia is notable though rare. This syndrome is marked by sustained ability to write in the presence of very impaired reading skills. In fact, the patient is unable to read what he or she has just written. A right homonymous hemianopsia is an almost constant feature (Benson, 1979). This syndrome occurs secondary to a lesion (most often a vascular accident) in the left medial inferior occipital region with concurrent pathological changes to the splenum of the corpus callosum. Alexia secondary to frontal lesions is also reported in the adult population. Frontal alexia is distinguishable by the patient's sustained ability to understand some written material. Patients can read many of the substantive words in a sentence aloud or for comprehension but are unable to identify individual letters. Severe dysgraphia is seen, with difficulty in copying as well as writing to dictation. It is most often accompanied by
299
a nonfluent aphasia. Frontal alexia is thought to occur with lesions of the anterior portion of the left hemisphere, with both cortical and subcortical involvement (Benson, 1979). There is some evidence that children's failure to acquire reading reflects pathology in the same localizations implicated in adults with acquired aphasia. Duffy, Denckla, Bartels, and Sandini (1980) reported group differences between dyslexic and normal boys on topographically transformed evoked potential and EEG data. Differences were prominent in the classical left temporal and parietal speech regions, bilateral frontal areas, and the left anterior lateral cortex. Similarly, Galaburda, Sherman, Rosen, Aboitiz, and Geschwind ( 1985) reported autopsy findings of four men with developmental dyslexia. All four revealed alterations in the usual asymmetry of the brain (notably symmetric planum temporale rather than left larger than right) and developmental abnormalities consisting of architectonic dysplasias and neuronal ectopias. Though the developmental abnormalities were bilaterally distributed, they were predominantly located in the peri-Sylvian classical speech areas of the left hemisphere. Though the lack of normal asymmetry of the planum temporale occurs in 25% of the population, and developmental anomalies occur in l 0% of the population, the authors felt that it is Jess likely for both abnormalities to occur simultaneously by chance. They suggested that the pathogenetic process produces dyslexia only when it is significant enough to produce both types of developmental changes. Levine, Hier, and Calvanio (1981) reported one child who was unable to learn to read or write subsequent to an arteriovenous malformation producing a hemorrhage in, and subsequent removal of, the left temporal lobe. No major disorder of intelligence or speech was present. The authors suggested that the left temporal lobe may be involved in the acquisition of written language skills. Galaburda et al. (1985) reported five cases of dyslexia also associated with angiomatous abnormalities of the left posterior cortex, three left temporoparietal and two left temporal. Again, these cases represent failure to acquire written language skills rather than loss of skills subsequent to focal damage. I have seen one clear case of isolated acquired alexia in a child in my clinical population. The young man, who was 14 years old when he lost his reading skills, had recurrent abscesses of the left frontal and left parietal occipital cortex. Alexia and agraphia were present, unaccompanied by a right hemiparesis. Language was notable for a marked dysnomia and a
300
CHAPTER 16
repetition disturbance. His speech was fluent with periodic paraphasias, but no neologisms. Understanding was disturbed. His alexia was profound, including inability to match letters written in different types. He was unable to write to dictation. His alexia was secondary to the more posterior lesion, and there has been little recovery. This case is reported in detail later in this chapter. The fact that we most often see failure to develop rather than loss of written language skills in children reflects our inability to assess skills that are not yet firmly established. Young children are not supposed to know how to read or spell so we cannot possibly answer questions regarding the integrity of these functions. Therefore, children only rarely can be demonstrated to lose writing or reading skills; more frequently, they simply fail to develop. The implication of this (e.g., we are unable to assess skills not yet present) is apparent throughout child neuropsychology, and is particularly problematic in thinking about localization of function. We can easily establish that children who have had left hemisphere damage do not as easily learn to read as those with no known damage or those with right hemisphere involvement. But because children rarely have reading skills to lose, we cannot directly relate the damage they sustain and subsequent loss of skill. Only rarely do we see disruption as an immediate consequence rather than a long-term sequela. Despite these limitations, the evidence that reading and spelling disorders are a consequence of left hemisphere dysfunction is compelling. There is no evidence clearly implicating one area within the left hemisphere. At a case report level, it appears likely that the same areas that produce alexia in adults may produce loss of reading skills or the failure to acquire reading skills in children. Rudel ( 1985) perhaps articulated this most clearly, noting that more recent evidence suggests that children with developmental dyslexia appear to have a basic disturbance in language processes suggestive of left hemisphere dysfunction. Arithmetic skills, perhaps because they are so complex, may be disrupted in adults as a consequence of either right or left hemisphere damage, or diffuse involvement. Calculation skills may be disrupted after damage to the tertiary areas of the dominant hemisphere (Luria, 1973). Written arithmetic skills may be disrupted subsequent to posterior right hemisphere damage presumably as a consequence of pervasive spatial confusion. There is little evidence regarding specific localizing implications of dyscalculia in children. As was the case with reading and writing difficulties, deficits in arithmetic are reported
. in terms of failure to acquire skills rather than in terms of loss of previously acquired arithmetic functions. As previously noted, dyscalculia is frequently reported as a long-term sequela of childhood aphasia, presumably reflecting left hemisphere damage. Dyscalculia is one of the symptoms defining Developmental Gerstmann's syndrome with acquired Gerstruano's syndrome, which is thought to be a consequence of dominant angular gyrus deficits, seen in adults. The localizing significance of arithmetic deficits is confused, however, by reports that developmental dyscalculia has been associated with other indications of right hemisphere disturbances. Rourke and his co-workers (Rourke Finlayson, 1975; Rourke & Strang, 1978; Strang & Rourke, 1985a,b) presented evidence that specific difficulty in developing arithmetic skills with relatively better performance in reading and spelling exists concurrently with a constellation of deficits that distinguishes them from children who perform relatively competently in arithmetic but poorly in reading and spelling. Children with specific arithmetic deficits have particular difficulties with bilateral tactile perceptual and psychomotor impairment (with left-sided performance more impaired than right-sided performance) and impaired visual spatial organizational skills. These children often have an associated defect in social perception and judgment. This pattern is suggestive of difficulties with tasks mediated by the right hemisphere. Rourke and his colleagues postulated that the failure to have adequate sensory motor experience secondary to perceptual difficulties may account for the specific difficulty in acquiring arithmetic skills. It should be noted that although this research suggests relative right hemisphere dysfunction in developmental dyscalculia, there is no evidence to suggest a structural or focal lesion. Weintraub and Mesulam (1983) reported similar neuropsychological findings, however, in adults with known right hemisphere pathology. Clearly, dyscalculia or acalculia, because it represents difficulty with a task that is so multifaceted and complex, has limited localizing significance. The pattern of arithmetic, reading, and spelling deficits can be suggestive of laterality with all three skills often impaired in left hemisphere dysfunction, and arithmetic incongruously impaired with right hemisphere dysfunction. Qualitative analysis of the errors made can be useful in understanding what part of the arithmetic process is disordered, and hence suggestive of localization. For example, if the child's performance is impaired because he or she cannot
TECHNIQUES OF LOCALIZATION
read numbers, one might wonder about left hemisphere disruption. This is particularly true if the dyscalculia is seen in the context of dyslexia and dysnomia. It should be emphasized, however, that dyscalculia has only limited localizing significance, and must be interpreted in the context of the remaining battery (Strub & Geschwind, 1983).
Executive Functions Executive functions-the ability to evaluate a problem, plan a response, carry out that plan, and assess the adequacy of the response within the context of the ongoing environment-are thought to be subserved by the prefrontal cortex (Luria, 1973). These functions are complex, rather subtle when operating effectively, and appear to mature with age. There have been vigorous disputes as to what constitutes a cognitive indication of prefrontal damage in adults. Recent reviews suggest that formal neuropsychological testing, including performance on standard tests of intelligence, is generally unaffected (Stuss & Benson, 1983). Performan~e on Trails B may be somewhat slow (Jarvis & Barth, 1984) secondary to mental inflexibility and difficulties with attention. Performance on the Wisconsin Card Sorting Test is grossly impaired with multiple perseverations predominating (Robinson, Heaton, Lehman, & Stilson, 1980). There may be decreased verbal fluency, and increased disinhibition on the Stroop Word Color Test (Stuss & Benson, 1983). Memory may be disrupted by retroactive and proactive inhibition. Qualitatively, patients with damage to the prefrontal cortex demonstrate changes in personality functioning with both overexaggerated and apathetic responses reported, often both seen in series (Mesulam, 1986). Echopraxic responses to the environment, i.e., responses that are a direct imitation onmediated by verbal knowledge (Luria, 1973), are observed. Lhermitte (1986) and Lhermitte, Dillon, and Serdaru ( 1986) described patients with damage to the prefrontal cortex as being overly responsive to sensory stimuli (as processed by the parietal cortex) because they lack the inhibition of inappropriate responses, which is normally provided by the frontal cortex. Consequently, imitation of gestures even when the patient is told that imitation is inappropriate is seen. Behavior increasingly is under control of perceptions rather than autonomous self-regulation. Self-appraisal is thought to be limited. There has been much discussion as to whether children can develop "frontal lobe" deficits (Gold-
301
en, 1981). As a later-myelinating area of the cortex (Yakolev & Lecours, 1967), it has been suggested that it is impossible to observe functional deficits of the frontal cortex until the injured child has reached adolescence. Some recent evidence suggests, however, that performance on tasks presumably sensitive to frontal lobe functions in adults shows rather dramatic developmental changes prior to adolescence. For example, on the Wisconsin Card Sorting Test normal adult levels of performance are reached by children by 10-12 years of age (Chelune & Baer, 1986). Passier, Isaac, and Hynd ( 1985) demonstrated that the verbal conflict and perseveration-eliciting maneuvers used to demonstrate deficits in adults with frontal damage fail to produce differential responses from control situations once children are about 10 years of age. This again suggests that "frontal skills'' mature through latency, but are demonstrable by about 10-12 years of age. There is some suggestion then that in children who acquire specific focal frontal damage subsequent to age 10, one should be able to demonstrate impairments similar to what is seen in the adult population. It should be emphasized, however, that there currently is no group empirical evidence indicating that disproportional difficulties on, for example, the Wisconsin Card Sorting Test suggest structural damage to the frontal cortices of children. The same statement is true for other tests thought to be sensitive to frontal disorders in adults. Clinically, the Wisconsin Card Sorting Test has been sensitive to damage to the prefrontal cortex in children even when such impairment is not seen on other measures. Interpretation is greatly assisted and strengthened by concomitant findings, especially ongoing motor deficits. A note of caution, however: difficulties on "frontal lobe tasks" are often seen with more diffuse disorders in adults (Robinson et al., 1980) and I would anticipate the same might be true in children.
Memory Research into memory deficits specific to a given modality has often emerged from the study of those with temporal lobe lesions and/ or temporal resections for amelioration of uncontrolled seizures (Milner, 1968). Current evidence suggests dissociation between verbal and nonverbal memory with patients with left temporal damage evidencing difficulties remembering verbal material and those with right temporal damage experiencing difficulty in remem-
302
CHAPTER 16
bering visual information that cannot be easily verbally encoded. These results have been repeatedly demonstrated in adults. Butters (1984) and Squire ( 1981, 1982) also demonstrated a differential pattern of amnesia related to temporal versus diencephalic damage. One such pattern is that individuals with temporal lobe damage demonstrate an abnormal rate of forgetting. Patients with diencephalic damage appear to forget at a more normal rate. Other dimensions of memory as they relate to specific localized deficits in the adult population have been addressed. Not surprisingly, however, there has been little work looking at specific memory deficits in children and relating these to focal impairment. One exception is the study by Fedio and Mirsky (1969). They compared the performance of four groups of children: normals, children with generalized three cycle per second spike and wave activity on EEG, children with unilateral right hemispheric epileptiform discharges on EEG, and children with unilateral left hemispheric epileptiform discharges on EEG. In addition to an assessment of intellectual functions and attention, assessment of memory, both verbal and nonverbal and immediate as well as delayed, was completed. Although there were no significant differences between the groups on length of memory span, there were moderately specific effects on measures of memory exceeding the subjects' basal memory span (supra-span). Children with left temporal involvement demonstrated a flatter learning curve on a task of verbal learning than the remaining groups. Children with right temporal involvement had the lowest nonverbal supra-span. Delayed memory was also differentially impaired, with children with right temporal EEG abnormalities demonstrating greater memory loss on a delayed reproduction of the Rey-Osterrieth Figure than the remaining groups, despite equivalent periods of exposure. These specific memory deficits were seen despite normal performance on tasks of sustained attention.
Reasons for Attempting Neuropsychological Localization Recent advances in neuroimaging (e.g., EEG, CAT scans, magnetic resonance imaging) provide noninvasive, relatively risk-free approaches to the localization of structural brain lesions. Other measures (Brain Electrical Activity Mapping, Positron Emission Tomography, regional cerebral blood flow) provide a means of assessing physiological activity portrayed in a manner that reflects topograph-
ically circumscribed patterns. As previously mentioned, many of the disorders affecting children (e.g., head injuries) damage the brain in a manner more likely to produce diffuse rather than focal injuries. All this being the case, why worry about localization of function in children? At a heuristic level there are a number of reasons why pursuing the possibility that there are circumscribed psychological effects of focal injury in children is important. Understanding the long-term implications of early brain damage in terms of the issue of plasticity depends upon further elaboration of our understanding of focal damage and the factors mediating its expression. Increased understanding of acquired focal effects may ultimately help us to understand developmental brain pathology. For example, if specific right hemisphere damage produces visual/spatial deficits, and patients with Turner's syndrome have a similar-appearing neuropsychological profile, we may be able to begin to understand how intrauterine and genetic disturbances affect brain development. Using such an approach, Geschwind and Galaburda ( 1985) theorized upon the emergence of developmental dyslexia. Their theory is based upon a model of intrauterine differences with these differences reflected in later lateralized effects. Localization of function studies in children may also be useful because these should allow us to describe mutually exclusive patterns of performance that typify different disease states. A model for this in the adult literature would be the dissociable memory deficits seen in diencephalic and temporal lobe patients. Clinically, understanding focal brain-behavior relationships in children is of importance in structuring an evaluation, providing counseling to families, and assisting in the development of remediation strategies. Understanding of focal effects is helpful in the process of evaluation. If we know a child has a specific lesion, we can anticipate other associated deficits and use this to structure the evaluation (though it should be clear that children often do not fit the anticipated pattern, and a thorough, global evaluation followed by subsequent finer testing of problematic areas is always recommended!). Knowledge of topographic brain-behavior relationships allows us to anticipate associated deficits that should be looked for (e.g., if visual agnosia is identified, prosopagnosia should be anticipated). Counseling based on focal brain-behavior relationships also utilizes the anticipated concurrence of patterns of disability. For example, children with visual-spatial disorders often struggle terribly with learning long division. Counseling to help families and children anticipate and plan for such difficulties is helpful.
TECHNIQUES OF LOCALIZATION
My focus during parent counseling is to help parents see the difficulties their child has as part of one pattern that makes sense, not as a myriad of distinct symptoms. For example, the hemi-neglect, difficulty recognizing familiar people, and difficulty drawing one mother identified are not three problems. Rather they are three facets of a unitary disorder. Such an approach supports families by decreasing anxiety, allowing them to plan for the future, and helping them to make sense of the diversity of home observations. Remediation strategies can be based upon a knowledge of topographic brain-behavior relationships. This is true both for acquired focal lesions and for deficits with no apparent etiological cause. Children with acquired lesions will have some recovery of function but are often left with subtle deficits reflecting the focality of the initial damage. Remedial strategies must be based upon compensatory approaches. Such strategies can be developed using those functions known to be mediated by intact areas. For example, a child with a known acquired frontal lesion who is unable to internally mediate responses to external events may have intact verbal skills. Such a child can be taught to ''self-talk,'' to use intact verbal skills to structure his or her own behavior. The use of known topographic brain-behavior relationships to aid in the remediation of developmental dysfunctions that present in a manner suggesting focal pathology (e.g., language deficits seen in the presence of inability to acquire reading skills) is . recommended by most authors, though the best strategy given this information is controversial (Rudel, 1985). Most special educators advocate use of a strength-based approach to remediation. Compensatory strategies can be derived based upon using intact skills, or brain areas that are thought to be intact. Clearly, the data 'available in child neuropsychology pertaining to topographic relationships are scanty, and the factors mediating the long-term presentation of focal injuries are not clear. Statements regarding localization of function in children should be made with circumspection. Often the best use of the data is to guide the psychologist in his or her thinking, planning, and guidance of the family and other professionals, rather than attempting to clearly say "where" the damage is located.
Case Studies The following cases are included to demonstrate how considering deficits as representative of a focal
303
pattern may be useful in planning for a specific child. The cases have been chosen to illustrate how the clinician might treat the data and are not representative of all patients seen in the practice of child neuropsychology.
Patient 1 F. U. is an 8-year 10-month-old female who was seen for a neuropsychological evaluation during hospitalization in an inpatient psychiatric unit. The hospitalization was evaluative in nature, with F. U. coming from and returning to a residential child care facility. There were specific concerns regarding F.U. 's clinging, dependent behavior and her difficulty organizing herself in self-care routines. She was seen as hyperactive. F. U. 's presentation was remarkable for her eager, enthusiastic approach to the evaluation. She remembered the examiner's name, having read it from the posted listing of appointments. On the walk to the exam room she stopped and checked printed signs, reading names to monitor her progress. The results of her evaluation are presented in Table 1. F. U. 's level of intellectual performance is at the lower end of the normal range. There are suggestions of both right and left hemisphere disruption, with right-sided sensory and motor difficulties clearly present. Perhaps more outstanding, however, is her profound spatial disorganization and construction dyspraxia. Though one might wonder if the spatial difficulties seen were another manifestation of her left hemisphere impairment, the quality of her productions (see Figure 1) and her skillful language (good reading and spelling, no naming problems) does not suggest that the left hemisphere difficulties explain the majority of her perceptual problems. Qualitative and item assessment is revealing. F. U. misses memory items requiring recollection of visual material; however, she learns verbal material readily. Performance on tasks requiring recognition of degraded pictures is notable for attention to details, with F.U. trying to identify the whole from observation of a part. Assessment of her performance reveals that in addition to the left hemisphere damage, evident in terms of right-sided sensory and motor errors, there is also an indication of right posterior dysfunction. The clear identification of neuropsychological symptoms is particularly helpful in a child who has a primary psychological or psychiatric diagnosis. It is difficult to attribute F. U. 's lateralized sensory motor difficulties or the specific visual perceptual deficits to
304
CHAPTER 16
TABLE 1. Evaluation Results for F.U. (RightHanded Female Aged 8 Years 10 Months) Stanford-Binet Intelligence Scale: 4th Edition Verbal Reasoning Standard Age Score Abstract-Visual Reasoning Standard Age Score Quantitative Standard Age Score Short-Tenn Memory Standard Age Score Composite Standard Age Score Wide Range Achievement Test Reading Standard Score Spelling Standard Score Arithmetic Standard Score
74
!
i
80
I
90
}
91
81
i
98 94 65
Gestalt Closure (Kaufman Assessment Battery for Children) Scaled Score = 1 Judgement of Line Orientation, Fonn V 2130 correct Boston Naming Test 40 correct responses (X for age
FlGURE 1. Rey-Osterrieth •mme
Given F.U.'s perceptual problems, a reconsideration of her behavior is necessary. Might her Trail-Making Test Fmger-Tapping Test clinging, dependent behavior relate to confusion RX = 20.2 A = 19" about where she is, fear of getting lost, and difficulty LX= 26.2 analyzing new and unusual situations? Indeed, interRey-Osterrieth Figure ventions aimed at helping F. U. to remain spatially See Figure 1 organized; discussing where the group is going, lookLuna-Nebraska Neuropsychological Battery-Children's Revised ing at maps, discussing what would be happening T-Score prior to leaving the hospital, and allowing F.U. to ~ utilize her relatively more intact verbal skills (talking Motor 72 about what to do if you get lost, rehearsing finding Rhythm Tactile 58 the store manager, and so on) decreased F.U. 's overVisual 79 reliance on adults. Similarly, she was aided in social Receptive Language 71 relationships by establishing a photo album of imporExpressive Language 83 tant people that contained verbal labels. F.U. would Writing 53 use the book to help prepare herself for visits with Reading hospital staff (whose names she could almost always Arithmetic remember, but whose faces were less clear in her 65 Memory recollection). Labeling her dresser drawers with the 81 Intellectual Processes contentshelpedF.U. to remain organized. Similarly, 72 Pathognomonic 76 making a daily schedule dependent upon reading Left Sensory Motor 47 words, not pictures of activities or reading a clock Right Sensory Motor - - - - - - - - - - - - - - - - - - - face, has assisted F.U. in maintaining independent activity. F. U. 's difficulties suggest that recognition and expression of affect may be an area of vulpsychiatric dysfunction. Though F.U. has ongoing nerability for her at a neurological as well as at a psychological difficulties, some of her ''disturbed'' psychological level. Therapeutic staff members have behavior relates to her focal-appearing organic defi- worked with F. U. to assist her with affective recognicits. These deficits appear to be chronic in nature. tion and expression. There is no report of a recent change in performance. Review of records reveals that F.U. was thought to Patient 2 have "mild cerebral palsy" as a toddler. A follow-up K.N. was 15 years old at the time of evaluation. neurological exam completed subsequent to the neuropsychological exam revealed some hyperactive During this evaluation he was hospitalized in a mediright-sided reflexes. ACT and MRI completed sub- cal inpatient unit for treatment of left hemisphere abscesses that on CT scan were localized to the left sequent to this exam were both normal.
= 38)
:!
TECHNIQUES OF LOCALIZATION
305
solving abscesses, rather than to any secondary depression. K.N.'s protocol suggests that the left parietal occipital lesion has produced a Wemicke's-like aphasia. There are signs of angular gyrus involvement. The functional correlates included a marked repetition and comprehension disturbance. K.N. was unable to understand complex grammatical structures. Anomia was present and though some paraphasic errors were seen, K.N.'s most common response was to engage in circumlocution. Most marked was the profound alexia with agraphia. K.N. was unable to easily match letters or words written in TABLE 2. Evaluation Results for K.N. (Right- different styles. Numbers were misread as well. Spelling errors included difficulty writing to dictaHanded Male Aged 15 Years) tion (e.g., wonche for once, nive for knife), as well as Luna-Nebraska Neuropsychological errors in spontaneously generated material (e.g., Batterypebl for people, wate for want). Fonn 2 Left angular gyrus lesions often produce dyscalculia. K.N. demonstrates clear calculation diffiT-Score Motor culties. Mental arithmetic was mildly disturbed. 37 Rhythm 32 Written arithmetic was disrupted as well, with K.N. Tactile 49 episodically confused in number recognition, reflectVisual 48 ing his left hemisphere deficits. Receptive 78 Evidence of a language disturbance secondary Expressive 86 to the posterior lesion was accompanied by both Writing 79 qualitative and quantitative evidence of changes conReading 85 sistent with left frontal damage. Perseveration both Arithmetic 87 within and between tasks was seen, despite a relaMemory 61 tively low level of perseverative responses on the Intellectual Processes 65 Left Frontal Wisconsin Card Sorting Test. Some echopraxic re75 Left Sensory Motor 56 sponding was seen. K.N. was unable to persist at a Left Parietal Occipital 76 task for a 10-second interval. He demonstrated clear Left Temporal 64 problems in making accurate observations about abRight Frontal 43 stract situations and was unable to easily generalize Right Sensory Motor 57 from one situation to aid in understanding another. Right Parietal Occipital 56 His deficits are consistent with what has been deRight Temporal 67 scribed in frontal lobe disorders (Luria, 1973). One Category Test would anticipate these cognitive changes to be ac39 errors companied by the personality changes described in frontal lobe disorders. Indeed, K.N. vacillated beWisconsin Card Sorting 13 perseverati ve responses tween irritable, hyperactive impulsive behavior and withdrawn apathy. Given the congruence of his cogPeabody Individual Achievement Test nitive and emotional presentation, and the locus of Reading Recognition 2.6 grade level his lesions, one might best interpret K.N. 's presentaMath 5. 3 grade level tion as secondary to the organic damage he has Spelling 2.7 grade level sustained. Boston Diagnostic Aphasia Exam Discharge planning required careful coordinaReading Comprehension 2 correct/7 tion with K.N. 's school. Frequent and intense speech Matching words and and language services were arranged. An individual letters 8 correct/ 10 aide, to read for K.N. and to help him remain on task, WISC-R (I month after current evaluation) was provided. K.N.'s frontal symptomatology sugVerbal IQ 81 gests that social relationships might be endangered. Perfonnance IQ 91 Consequently, psychotherapy to help K.N. cope with Full-Scale IQ 85 this changed status and to learn new skills was arfrontal and left parietal occipital areas. Despite known localized pathology, there were ongoing questions regarding cognitive and behavioral problems seen on the ward (e.g., depression versus organicity, and questions about discharge planning). K.N. 's evaluation results are presented in Table 2. His presentation clearly demonstrates focal neuropsychological deficits, with K.N. 's perceptual and cognitive difficulties exactly what one would have predicted on the basis of anatomic localization. This would suggest that the disturbances identified in his day-to-day functioning are closely related to the re-
306
CHAPTER 16
ranged. Intense education of school staff and family was provided. On follow-up exam, K.N.'s frontal symptoms have resolved quite nicely though he continues to be an impulsive young man with poor planning skills and an unusually high frequency of facetious comments. His attention remains poor. He has reacquired minimal reading and spelling skills. He is quite dysnomic but his comprehension and repetition are far better. We are now working with his school to begin vocational assessment and training.
Conclusions Localization of function in child neuropsychology is difficult at best, and perhaps impossible (Wilson, 1986). The obstacles to using such techniques with children include the lack of understanding of how functions are localized in the nonnally developing brain, with even greater confusion regarding the interaction between injury and functional organization in an atypical cortex. It is tempting, but often dangerous and spurious, to use data available about adults in an attempt to understand the neuropsychological presentations of children. Nevertheless, there are some reasons to at times consider whether the data one obtains fit a known neurological pattern, and whether the results might be attributable to a focal injury. When there is no need to suggest structural localization, a consideration of topographic brain-behavior relationships aids in providing a fuller understanding of the syndrome, and allows for more effective and thorough evaluation as well as directing remediation. Demonstration of known patterns of neuropsychological deficit can be useful in discriminating pure psychiatric from neuropsychological syndromes. The development of techniques in child neuropsychology including techniques of localization is clearly in its nascency. Careful research coupled with meticulous self-appraisal of cautious clinical work will be required to further refine and develop techniques. ACKNOWLEDGMENTS
The author gratefully acknowledges the secretarial support of Beverly Hill and Jo-Anne Schilling. The always rapid, cheerful, and conscientious library assistance of Anne Klenk, Carol Morgan, and Susan Osborn of The Children's Hospital Medical Library
was invaluable in this project, as well as so many others.
References Alajouanine, T., & Lhermitte, F. (1965). Acquired aphasia in children. Brain, 88(4), 614-662. Alexander, M.P., & Albert, M. L. (1983). The anatomical basis of visual agnosia. In A. Kertesz (Ed.), Localization in neuropsychology (pp. 393-418). New York: Academic Press. Annett, M. ( 1970). The growth of manual preference and speed. British Journal of Psychology, 6/(4), 545-558. Annett, M. (1973). Laterality of childhood hemiplegia and the growth of speech and intelligence. Cortex, 9, 4-33. Aram, D. M., Kelman, B. L., Rose, D. F., & Whitaker, H. A. (1985). Verbal and cognitive sequelae following unilateral lesions acquired in early childhood. Journal of Clinical and Experimental Neuropsychology, 7(1), 55-78. Aram, D. M., Rose, D. F., Rekate, H. L., & Whitaker, H. A. (1983). Acquired capsular striatal aphasia in childhood. Archives of Neurology, 40, 614-617. Basser, L. S. (1962). Hemiplegia of early onset and the faculty of speech with special reference to the effects of hemispherectomy. Brain, 85, 427-460. Benson, D. F. (1919).Aphasia, alexia, and agraphia. Edinburgh: Churchill Livingstone. Benson, F. D., & Geschwind, N. (1970). Developmental Gerstmann syndrome. Neurology, 20, 293-298. Benton, A. (1955). Development of finger localization capacity in school children. Child Development, 26. 225-230. Benton, A. L., Hamsher, K. D., Varney, N. R., & Spreen, 0. (1983). Contributions to neuropsychological assessment. New York: Oxford University Press. Boll, T. J. (1981). The Halstead-Reitan Neuropsychological Battery. In S. B. Filskov & T. J. Boll (Eds.), Handbook of clinical neuropsychology (pp. 577-607). New York: Wiley. Boll, T. J., & Barth, J. T. (1981). Neuropsychology of brain damage in children. In S. B. Filskov & T. J. Boll (Eds.), Handbook of clinical neuropsychology (pp. 418-452). New York: Wiley. Butters, N. (1984). The clinical aspects of memory disorders: Contributions from experimental studies in amnesia and dementia. Journal of Clinical Neuropsychology, 6(1), 17-36. Carter, R. L. K., Hohenegger, M. L., & Satz, P. (1982). Aphasia and speech organization in children. Science, 218(19), 797799. Chadwick, 0., Rutter, M., Thompson, J., & Shaffer, D. (1981). Intellectual performance and reading skiils after localized head injury in childhood. Journal of Child Psychology and Psychiatry, 22, 117-137. Chelune, C. J., & Baer, R. A. (1986). Developmental nonns for the Wisconsin Card Sorting Test. Journal of Clinical and Experimental Neuropsychology, 8(3), 219-228. Chelune, C. J., & Edwards, P. (1981). Early brain lesions: Ontogenetic environmental and considerations. Journal of Consulting and Clinical Psychology. 49(6), 7n-790. Denckla, M. (1973). Development of speed in repetitive and sue-
TECHNIQUES OF LOCALIZATION
cessive finger movements in normal children. Developmental Medicine and Child Neurology. /5, 635-645. Dennis, M. (1977). Cerebral dominance in three forms of early brain disorder. In E. Blaw, I. Rapin, & M. Kinsbourne (Eds.), Topics in child neurology (Vol. I, pp. 189-212). New York: SP Books Division of Spectrum Publications. Dennis, M. (1976). Impaired sensory and motor differentiation with corpus callosum agenesis: A lack of callosal inhibition during ontogeny? Neuropsychologia, 14, 455-469. Dennis, M., & Kohn, B. (1975). Comprehension of syntax in infantile hemiplegics after cerebral hemidecortication. Left hemisphere superiority. Brain & Language, 2, 472-482. Dennis, M., Lovett, M., & Weigel-Crump, C. A. (1981). Written language acquisition after left or right hemicortication in infancy. Brain and Language, 12, 54-91. Dennis, M., & Whitaker, H. A. (1976). Language acquisition following hemidecortication: Linguistic superiority of the left over the right hemisphere. Brain and Language, 3, 404-433. Dennis, M., & Whitaker, H. A. (1977). Hemispheric equipotentiality and language acquisition. InS. S. Segalowitz & F. A. Gruber (Eds.), Language development and neurological theory (pp. 93-106). New York: Academic Press. Duffy, F. H., Denckla, M. P., Bartels, P. H., & Sandini, G. ( 1980). Dyslexia: Regional difference in brain electrical activity by topographic mapping. Annals of Neurology, 7(5), 412-420. Fedio, P., & Mirsky, A. F. (1969). Selective intellectual deficits in children with temporal lobe or centrencephalic epilepsy. Neuropsychologia, 7, 287-300. Ferro, J. M., Martins, I. P., Pinto, P., & Castro-Caldas, A. (1982). Aphasia following right striato-insular infarction in a left-handed child: A clinico-radiological study. DevelopmentalMedicineandChildNeurology. 24, 173-182. Finger, S., & Stein, D. G. (1982). Brain damage recovery: Research and clinical perspective. New York: Academic Press. Galaburda, A.M., Sherman, G. F., Rosen, G. D., Aboitiz, F., & Geschwind, N. (1985). Developmental dyslexia: Four consecutive patients with cortical anomalies. Annals of Neurology, 18(2), 222-223. Geschwind, N., & Galaburda, A.M. (1985). Cerebrallateralization: Biological mechanisms, associations and pathology. I. Hypothesis and a program for research. Archives of Neurology, 42, 428-459. Golden, C. J. (1981). The Luna-Nebraska Children's Battery: Initial theory and formulation. In G. W. Hynd & 1. E. Obrzut (Eds.), Neuropsychological assessment and the school-aged child (pp. 277-302). New York: Grune & Stratton. Golden, C. J., Hammeke, T. A., & Purisch, A. D. (1980). The Luria-Nebraska Neuropsychological Battery. Los Angeles: Western Psychological Services. Goldman, R. S. (1974). An alternative to developmental plasticity: Heterology of CNS structures in infants and adults. In D. G. Stein, J. J. Rosen, & N. Butters (Eds.), Plasticity and recovery offunction in the central nervous system (pp. 149174). New York: Academic Press. Greenblatt, S. H. (1983). Localization of lesions in alexia. In A. Kertesz (Ed.), Localization in neuropsychology (pp. 324356). New York: Academic Press. Hecaen, H. (1976). Acquired aphasia in children and the on-
307
togenesis of hemispheric functional specialization. Brain and Language, 3, 114-134. Hecaen, H. (1983). Acquired aphasia in children: Revisited. Neuropsychologia, 21(6), 581-587. Jarvis, P. E., &Barth, J. 1. (1984). Halstead-Reitan Test Battery: An interpretive guide. Odessa, FL: Psychological Assessment Resources. Kaufman, A. S., &Kaufman, N. L. (1983). Kaufman Assessment Battery for Children. Circle Pines, M American Guidance Service. Kershner, 1. R., & King, A. 1. (1974). Laterality of cognitive functions in achieving hemiplegic children. Perceptual and Motor Skills, 39, 1283-1289. Kertesz, A. (1983). Issues in localization. In A. Kertesz (Ed.), Localization in neuropsychology (pp. 1-19). New York: Academic Press. Kiessling, L. S., Denckla, M. B., & Carlton, M. (1983). Evidence for differential hemispheric function in children with hemiplegic cerebral palsy. Developmental Medicine and Child Neurology, 25, 727-734. Kinsbourne, M., & Warrington, E. K. (1962). A study of finger agnosia. Brain, 85, 47-66. Kinsboume, M., & Warrington, E. K. (1963a). The developmental Gerstmann syndrome. Archives of Neurology, 8, 490501. Kinsbourne, M., & Warrington, E. K. (1963b). The development of finger differentiation. Quanerly Journal of Experimental Psychology, 15, 132-137. Kohn, B., & Dennis, M. (1974). Selective impairments ofvisuospatial abilities in infantile hemiplegics after right cerebral hemidecortication. Neuropsychologia. 12, 505-512. Levin, H. S., Benton, A. L., & Grossman, R. G. (1982). Neurobehavioral consequences of closed head injury. New York: Oxford University Press. Levine, D. N., Hier, D. B., & Calvanio, R. (1981). Acquired learning disability for reading after left temporal lobe damage in childhood. Neurology, 31, 257-264. Lhermitte, F. (1986). Human autonomy and the frontal lobes. Part II: Patient behavior in complex and social situations. The "environmental dependency syndrome." Annals of Neurology, 19(4), 335-343. Lhermitte, F., Dillon, B., & Serdaru, M. (1986). Human autonomy and the frontal lobes. Part I: Imitation and utilization behavior. A neuropsychological study of 75 patients. Annals of Neurowgy, 19(4), 326-334. Luria, A. R. (1973). The working brain. New York: Basic Books. Martyn, L. (1975). Pediatric neuro-ophthalmology. In R. D. Harley (Ed.), Pediatric ophthalmology. Philadelphia: Saunders. McFie, J. (1961). Intellectual impairment in children with localized post-infantile cerebral lesions. Journal ofNeurology, Neurosurgery and Psychiatry, 24, 361-365. Mesulam, M. M. (1986). Frontal cortex and behavior. Annals of Neurology, 19(4), 320-325. Milner, B. (1968). Visual recognition and recall after right temporal lobe excision in man. Neuropsychologia, 6, 191-209. Milner, B. (1969). Residual intellectual and memory deficits after head injury. In E. Walker, W. Caveness, & M. Critchley (Eds.), The late effects of head injury (pp. 84-97). Springfield, IL: Thomas.
308
CHAPTER 16
Milner, B. (1974). Hemispheric specialization: Scope and limits. In F. 0. Schmitt & F. G. Worden (Eds.), The neurosciences: Third study program (pp. 75-89). Cambridge, MA: MIT Press. Nici, J., & Reitan, R. M. (1986). Patterns of neuropsychological ability in brain-disordered versus normal children. Journal of Consulting and Clinical Psychology, 54(4), 542-545. The Ophthalmologic Staff of the Hospital for Sick Children, Toronto. (1967). The eye in childhood. Chicago: Yearbook Medical. Passier, M.A., Isaac, W., & Hynd, G. A. (1985). Neuropsychological development of behavior attributed to frontal lobe functioning in children. Developmental Neuropsychology, 1(4), 349-370. Pavlakis, S. G., Phillips, P. C. D., Mauro, S., DeVivo, D. C., & Rowland, L. P. (1984). Mitochondrial myopathy, encephalopathy, lactic acidosis and stroke-like episodes: A distinctive clinical syndrome. Annals of Neurology, 16(4), 481-488. Pennington, B. F., VanDoomink, W. J., McCabe, L. L., & McCabe, E. R. (1985). Neuropsychological deficits in earlytreated phenylketonurics. American Journal of Mental Deficiency, 88(5), 467-474. Rapin, I. (1982). Children with brain dysfunction: Neurology, cognition, language and behavior. New York: Raven Press. Reitan, R. M. (1972). Verbal problem solving related to cerebral damage. Perceptual and Motor Skills, 34, 515-524. Reitan, R. M. (1974). Psychological effects of cerebral lesions in children of early school age. In R. M. Reitan & L. A. Davison (Eds.), Clinical neuropsychology: Current status and applications (pp. 53-89). New York: Wiley. Robinson, A. L., Heaton, R. K., Lehman, A. W., & Stilson, D. W. (1980). The utility of the Wisconsin Card Sorting Test in detecting and localizing frontal lobe lesions. Journal of Consulting and Clinical Psychology, 48(5), 605-614. Rourke, B. P., & Finlayson, A. J. (1975). Neuropsychological significance of variations in patterns of performance on the Trail-Making Test for older children with learning disabilities. Journal of Abnormal Psychology, 4, 412-421. Rourke, B. P., & Strang, J.D. (1978). Neuropsychological significance of variations in patterns of academic performance: Motor, psychomotor, and tactile-perceptual abilities. Journal of Pediatric Psychology, 3(2), 62-66. Rourke, B. P., Yanni, D. W., MacDonald, G. W., & Young, G. C. (1973). Neuropsychological significance of Jateralized deficits on the Grooved Pegboard Test for older children with learning disabilities. Journal ofConsulting and Clinical Psychology, 41(1), 128-134. Rudel, R. (1985). The definition of dyslexia: Language and motor deficits. In F. H. Duffy & N. Geschwind (Eds.), Dyslexia: A neuroscientific approach to clinical evaluation (pp. 33-54). Boston: Little, Brown. Rudel, R. G., & Teuber, H. L. (1971). Spatial orientation in normal children and in children with early brain injury. Neuropsychologia, 9, 401-407. Rudel, R. G., Teuber, H. L., & Twitchell, T. E. (1974). Levels of impairment of sensori-motor functions in children with early brain damage. Neuropsychologia, 12, 95-108. Rutter, M. (1982). Developmental neuropsychiatry: Concepts, is-
sues, and prospects. Journal ofClinical Neuropsychology, 4, 91-115. Satz, P., & Bullard-Bates, I. C. (1981). Acquired aphasia in children. In M. T. Sarno(Ed.),Acquiredaphasia (pp. 399-426). New York: Academic Press. Spreen, 0., & Gaddes, W. H. (1969). Developmental norms for 15 neuropsychological tests ages 6-15. Cortex, 5, 171-191. Squire, L. R. (1981). Two forms of human amnesia: An analysis of forgetting. The Journal of Neuroscience, /(6), 635-640. Squires, L. R. (1982). The neuropsychology of human memory. Annual Review of Neuroscience, 5, 241-273. St. James- Roberts, I. (1979). Neurological plasticity. recovery from brain insult, and child development. In H. W. Reese & L. P. Lipsitt (Eds.), Advances in child development and behavior (pp. 254-319). New York: Academic Press. St. James-Roberts, I. (1981). A reinterpretation of hemispherectomy data without functional plasticity of the brain. Brain and Language, 13, 31-53. Strang, J. D., & Rourke, B. P. (1985a). Adaptive behavior of children who exhibit specific arithmetic disabilities and associated neuropsychological abilities and deficits. In B. P. Rourke (Ed.), Neuropsychology of learning disabilities: Essentials of subtype analysis (pp. 302-328). New York: Guilford Press. Strang, J. D., & Rourke, B. P. (1985b). Arithmetic disability subtypes: The neuropsychological significance of specific arithmetical impairment in childhood. In B. P. Rourke (Ed.), Neuropsychology of learning disabilities: Essentials of subtype analysis (pp. 167-183). New York: Guilford Press. Strub, R. L., & Geschwind, N. (1983). Localization in Gerstmann syndrome. In A. Kertesz (Ed.), Localization in neuropsychology. New York: Academic Press. Stuss, D. T., & Benson, F. D. (1983). Frontal lobe lesions and behavior. In A. Kertesz (Ed.), Localization in neuropsychology (pp. 429-454). New York: Academic Press. Van Dongen, H. R., Loonen, C. B., & Van Dongen, K. J. (1985). Anatomical basis for acquired fluent aphasia in children. Annals of Neurology, 17(3), 306-309. Weintraub, S., & Mesulam, M. M. (1983). Developmental learning disabilities of the right hemisphere. Archives of Neurology, 40, 463-468. Wilkening, G., & Golden, C. J. (1982). Pediatric neuropsychology: Status, theory, and research. In P. Karoly, J. J. Steffen, & D. J. O'Grady (Eds.), Child health psychology: Concepts and issues (pp. 119-153). New York: Pergamon Press. Wilson, B. C. (1986). An approach to the neuropsychological assessment of the preschool child with developmental deficits. In S. B. Filskov & T. J. Boll (Eds.), Handbook of clinical neuropsychology (Vol. 2, pp. 121-171). New York: Wiley. Witelson, S. F. (1977). Early hemispheric specialization and interhemispheric plasticity: An empirical and theoretical review. InS. J. Segalowitz & F. A. Gruber (Eds.), Language development and neurological theory (pp. 213-287). New York: Academic Press. Woods, B. (1980). The restricted effects of right hemisphere lesions after age one: Wechsler test data. Neuropsychologia, 18, 65-70.
TECHNIQUES OF LOCALIZATION
309
Woods, B. T., & Carey, S. (1978). Language deficits after appar-
Woods, B. T., & Teuber, H. L. (1978a). Changing patterns of
rology. 6(5), 405-409. Woods, B. T., & Teuber, H. L. (1973). Early onset of comple-
Woods, B. T., & Teuber, H. L. (1978b). Mirror movements after childhood hemiparesis. Neurology, 28, 1152-1158.
Transactions of the American Neurological Association. 98. 113-117.
cles of regional maturation of the brain. In A. Minkoski (Ed.), Regional development of the brain in early life (pp. 3-
ent clinical recovery from childhood aphasia. Annals of Neumentary specialization of cerebral hemispheres in man.
childhood aphasia. Annals of Neurology, 3(3), 273-280. Yakolev, P. T., & Lecours, A. R. (1967). The myelogenetic cy70). Oxford: Blackwell.
17 Neuropsychological Sequelae of Substance Abuse by Youths ROBERT W. ELLIOTT
Drug- and alcohol abuse-related problems have been ior relationships. When drugs have been introduced part of civilization since the ancient days, but it was into the brain, the specific purpose of the neuronot until the 1960s, when the counterculture became psychological evaluation can be to describe and meaa guiding force for the teen population, that society sure changes associated with altered cognitive funcbegan to express widespread concern about adoles- tioning by investigating higher cortical functions, cent use of drugs and alcohol. In 1975, a group of such as reasoning, memory, language, perception, University of Michigan researchers began a longitu- and sensory-motor skills. dinal series of studies to investigate the prevalence of Assessing youth is more complicated than asdrug use by high school students (Johnston, O'Mal- sessing adults because of differences in emotional ley, & Bachman, 1985). Although, by I985, John- development and brain development at different ages ston et al. found that there had been a gradual decline and stages of development. During adolescence, in the overall use of illicit drugs by the surveyed high there are spurts and plateaus in cognitive developschool seniors, nearly two-thirds had experimented ment; therefore, assessment of this dynamic and unwith an illicit drug before graduation, and nearly half stable system should involve procedures and methhad used a drug other than marijuana. ods designed with these issues addressed (Rourke, Children and adolescents, as well as young Bakker, Fish, & Strang, I983a-c). adults, are populations rarely studied in drug abuse research (Ivnik, I 986; Grant & Reed, I 985) although this particular population is important because of is- Definitions sues in brain development. It is during adolescence that the higher-level cognitive attributes of planning, This chapter will focus primarily on the chronic evaluation, flexibility, internalized behavioral con- effects of drugs. Chronic use implies the possibility trols, higher-level abstracting skills, and ethical of long-tenn consequences after ingestion of the drug awareness are developing (Golden, 1985). Use of has ceased (Ware, 1979). Acute effects are mendrugs during this period could interfere with develop- tioned only when very few or no chronic effect studment of one or more of these attributes and have long- ies have been reported. tenn ramifications for personality functions sub"Drug dependence" and "addiction" are used served by the frontal and prefrontal regions of the synonymously in this chapter. Both terms indicate a neocortex. psychological and/or physical "need" for a particular substance (Parsons & Farr, 1981). "Tolerance" Assessment of Youth indicates that increasingly larger doses are necessary to achieve the same effects obtained with earlier use The general purpose of a neuropsychological of the drug (Jaffe, I 975). "Withdrawal syndrome" assessment is to describe and evaluate brain-behav- refers to the physical and psychological characteristics observed in an individual when access to an addictive drug is reduced or terminated. ROBERT W. ELLIOTT • Department of Special Education, The research reviewed here has been. limited to South Bay Union High School District, Redondo Beach, Califorthose studies investigating children, adolescents, and nia 90277. 311
312
CHAPTER 17
young adults, which herein will refer to individuals from birth through age 25 years.
Drug Classification There are a variety of classification schemes available that categorize drugs based on different criteria. Kiss in's ( 1977) system is one of only a few that are based on behavioral and central nervous system consequences. He grossly divides psychoactive drugs into depressants, stimulants, and hallucinogens (Cepeda's chapter in this volume uses a more detailed system). Parsons and Adams (1983) added inhalants to Kissin's classification system. No specific classification scheme was adopted for this review because most of the specific drugs listed in these systems have never been researched for long-term neuropsychological effects. Only drugs that have been the subject of more than one neuropsychological investigation are discussed.
Marijuana/Hashish Marijuana today is about 10 times more potent than the marijuana available 10 years ago. Hashish is a refined form of marijuana, containing about 8% marijuana and about eight times the amount of A-9tetrahydrocannabinol (THC) found in most marijuana (Bell, 1985). Marijuana acts on the central nervous system as a depressant, although some researchers (Kissin, 1977) treat it as a hallucinogen. Over one-half of surveyed 1985 high school seniors admit to use of marijuana and/or hashish at some time in theirlives (Johnston etal., 1985). Peak use is most prevalent in late adolescence (Schuckit, 1984, p. 123). Users report euphoria, a relaxed state, sleepiness, heightened sexual arousal, hunger, decreased social interaction, short-term memory deficits, and difficulty completing multiple-step tasks (Schuckit, 1984). Few studies report development of tolerance or physical dependence. Kolansky and Moore (1971) examined the psychological and neurological performance of 38 psychiatric patients, aged 13 to 24 years, all of whom smoked marijuana two or more times a week for periods of more than 1 year. Although no standardized neurological exam was conducted, all of the patients displayed limited attention span and impaired cognition. Individuals who smoked marijuana four or five times a week displayed slurred speech,
tremors, and deficits in gait and depth perception. The researchers failed to establish a control group, and no statistical data, description of the measures, or information about the last use of marijuana was mentioned in this study. Evidence of cerebral atrophy, including enlarged left lateral ventricles, temporal lobe dilatations, and damage to the region of the caudate nuclei and basal ganglia, was found by air encephalography in lO male patients, average age 22 years, who had smoked marijuana over a period of 3 to 11 years (Campbell, Evans, Thompson, & Williams, 1971). Memory and concentration problems were evident in most of the subjects. A serious flaw was introduced into this study when the authors failed to control for use of other drugs, because most of the subjects admitted to using other drugs. There was also no indication as to when each subject had last used a drug. Gleaton and Gowen (1985) hypothesized that because the marijuana cannabinoid chemicals remained in the body for weeks, there were subtle, long-term neurological effects, such as loss of shortterm memory, speech deficits, and symptoms consistent with pre-senile dementia. They failed to substantiate their position with research findings. A joint ARE/WHO scientific meeting report (Addiction Research Foundation, 1981) summarized many studies focused on the effects of cannabis. The report noted that there was no evidence of residual brain damage, because computer-assisted tomographic (CAT) scans of youthful cannabis users were asymptomatic. Grant, Rochford, Fleming, and Stunkard ( 1973) studied 29 male medical students who had smoked marijuana an average of three times a month regularly for at least 3 years, and compared them against a non-drug-using control group. The median age of both the users and the controls was 23 years. The Category Test, Tactual Performance Test (TPT), Raven's Progressive Matrices, the Goal-Directed Serial Alternation Test, and a special neurological exam were administered. The only significant difference between theusers and non users was on the TPT, localization score. Although the researchers failed to find a number of significant differences, their sample was limited to bright, light to moderate users, who may have been well compensated. Carlin and Trupin ( 1977) recruited a group of I0 well-educated subjects with an average age of 24 years. Candidates who had used any other recreational drug I0 or more times, or who had suffered any neurological illness or injury, were excluded. The subjects were asked to abstain from use of marijuana for 24 hours prior to administration of the Halstead
NEUROPSYCHOLOGICAL SEQUELAE OF SUBSTANCE ABUSE BY YOUTHS
Neuropsychological Battery. The only significant (p < 0.05) score difference between the users and nonusers was on the Trail Making Test, Part B, but surprisingly, the user group performed better than the nonuser group. Culver and King (1974) administered the Halstead Neuropsychological Battery, WAIS, Trail Making Test, Laterality Discrimination Test, and selected tests from the Kit of Factor-Referenced Cognitive Tests, to three groups of intact college seniors (n = 84), aged 20 to 25 years. Group one were LSD/mescaline users, the second group were marijuana/hashish users, and the third group served as controls. The marijuana/hashish users had used marijuana and/or hashish at least twice a month for at least 12 months, and could be classified as light to moderate users. The drug users agreed to abstain from all drug use for the 7 days preceding testing. There were no significant differences between the marijuana/hashish users and the controls. One study yielding positive physiological and neuropsychological results with youthful marijuana users (Campbell et al., 1971) was seriously flawed because most of the subjects used other drugs and also may have been manifesting acute rather than chronic effects. Still, this study cannot be ignored. Heath, Fitzgarrell, Fontana, and Garey ( 1980) found that exposure to THC, at doses commensurate with those used by human marijuana smokers, could result in changes in the limbic structure and endoplasmic reticulum and possibly produce nuclear inclusion bodies in rhesus monkeys. As suggested by Carlin (1986), deficits associated with marijuana use may be more subtle than can be detected by current methods of assessment. Final conclusions cannot be drawn because most of the research generated to date on marijuana has been flavored by methodological problems and biased sampling, but long-term effects with chronic use seem highly probable (Fisher, 1987).
Cocaine Cocaine has been separated from the other stimulants in this chapter owing to its unique history, properties, increasing prevalence, public health threat, and addictive nature. Cocaine is defined as a potent local anesthetic but unlike other local anesthetics such as procaine or lidocaine, cocaine is a powerful psychomotor stimulant that shares many of the behavioral and biochemical properties found in the psychomotor stimulants amphetamine and meth-
313
ylphenidate (Fischman et al., 1976; Martin, Sloan, Sapira, & Jasinski, 1976; Post, Weiss, Pert, & Uhde, 1987; Trites, Sub, Offord, Neiman, & Preston, 1974). Thus, cocaine's powerful combination as a local anesthetic and psychomotor stimulant makes its effects difficult to interpret. The 11th annual University of Michigan high school survey (U.S. Journal of Drug and Alcohol Dependence, 1985, p. 4) revealed that adolescent cocaine use was up in 1985 among males and females in most regions of the United States. The survey found that cocaine had been tried by 17.3% of the seniors in the class of 1985, up from 16 .I% the previous year. This increase is alarming and has caused a great deal of renewed search for preventive programs (Elliott, 1987). An additional concern is the high incidence of habitual use. Siegel (1984) completed a longitudinal study of 99 recreational users. He reported that 50% of these users maintained recreational use over a 10-year follow-up period. Of the remaining 50% who developed some degree of habitual use, 40% were "mild abusers" and 10% were severe freebase users. Cocaine, an alkaloid derived from the mountain shrub Erythroxylon coca, is considered the most potent stimulant of natural origin. It can both incapacitate and excite neurons in the central and sympathetic nervous systems (Maranto, 1985). Rosecran and Spitz ( 1987) noted that cocaine could produce subtle imbalances in brain chemistry increasing the need for more of the drug. Because cocaine is lipid soluble, it reaches the brain quickly and stimulates production of large amounts of dopamine, which may explain why use is so pleasurable (Estroff, 1985). The resultant high is short-lived, lasting only 10-30 minutes (Elliott, 1987). The issue of its addictiveness has been controversial since the 19th century (Rosecran & Spitz, 1987). As recently as 1982, Van Dyke and Byck noted in a Scientific American article that "The pattern of behavior is comparable to that experienced by many people with peanuts or potato chips. It may interfere with other activities of the individual, but it may be a source of enjoyment as well" (p. 138). Kaplan, Freedman, and Sadock (1980) noted that if " ... used no more than two or three times a week, cocaine creates no serious problems" (p. 1621). Even in the American Psychiatric Association's diagnostic manual (DSM; APA, 1980), cocaine has a category for abuse but not for dependence (p. 173). The implication is that cocaine is a less serious addictive drug. The DSM-III Advisory Committee on Substance Abuse Disorders concluded that there was no evidence of withdrawal or tolerance, and thus no
314
CHAPTER 17
evidence of physiological addiction (Spitzer, Williams, & Skodal, 1980). Clinicians are now discovering tolerance and withdrawal features in freebase and intravenous users (Gawin & Kleber, 1987; Spitz & Rosecran, 1987). Although the long-term neuropsychological effects of both occasional and chronic cocaine use are unknown (Grant & Mohns, 1975), some data on neurological, physiological, and affective consequences have been published (Adams & Durrell, 1984). Nunes and Rosecran (1987) noted many effects from chronic use including insomnia, hallucinations, paranoia, depression, lethargy, muscular twitches, tremors, and muscular weakness. Elliott noted all of these effects and also convulsions in his clinical work with recovering adolescent cocaine users (Elliott, 1987). Gawin and Kleber followed a group of 30 cocaine abusers seeking treatment over a period of months. They found that many aftereffects of cocaine use persisted for months and speculated that many subtle effects might persist indefinitely (Gawin & Kleber, 1986). Washton and Gold (1987) reported that 39% of 500 cocaine users who had contacted the BOOCOCAINE hotline claimed that they had suffered cocaine-induced brain seizures with loss of consciousness. Reports of acute and short-term effects are much more prevalent (Fischran & Schuster, 1981; Jones, 1984). Myers and Earnest (1984) noted that after three young adults (aged 20 to 25 years) injected 140-160 mg of cocaine, each sustained a generalized convulsion. A neurological exam completed shortly thereafter, which included EEG and CT evaluations, was normal (Myers & Earnest, 1984). A few investigators have looked at the longterm consequences of chewing coca leaves by South American Indians (Cagliotti, 1980; Negrete & Murphy, 1967; Zapata-Ortiz, 1970). Although there were indications of deficits in attention, speed of responding, auditory and visual memory, and accuracy, the deficits were subtle and the studies not well controlled. Occasionally, one comes across a statement that cocaine may cause "brain damage" but these statements have not been validated in human studies (Maranto, 1985). Although the issue of addictiveness remains controversial, the preponderance of findings from recent research suggests that tolerance does exist. Withdrawal signals are detectable on the EEG and in sleep patterns, but the signals are weak compared to the signals associated with withdrawal from barbiturates, alcohol, or opiates (Van Dyke & Byck, 1982). It may be that what we have come to believe are traditional withdrawal symptoms with most drugs do
not apply to cocaine. A new model of addictiveness is necessary with cocaine. The physical dangers connected with even a single use of cocaine have been given wide coverage by the media during the last few years. Several wellknown athletes have died suddenly after using cocaine. Athletes, in particular, may use this stimulant to mask fatigue, which can lead to injury and a gradual deterioration of performance .. Neuropsychological studies of chronic cocaine use by young adults or adolescents have not been attempted. Careful, controlled long~term studies of cocaine's effects on neuropsychological performance are much needed at present, especially due to the widespread mythology surrounding use of cocaine and beliefs in its enhancement of physical performance, alertness, and sexual response (Pope, 1987). EEG, brain imaging, and neuroendocrine techniques are available to document such changes and have been used in much of the research focusing on drug addictions.
Stimulants After marijuana, the group of illicit drugs most widely used by teenagers are stimulants. Johnston et al. (1985), in their annual survey of high school seniors, reported a use rate of 28%, the rate being slightly higher for females than males. Stimulants can be grouped in three categories: amphetamines, dextroamphetamines, and methamphetamines. Although the structure of each of these synthetic drugs is different, each can produce similar, long-lasting cocaine-like effects. They differ primarily in their strength and duration of effect. Stimulants act on the CNS by potentiating the effects of norepinephrine, activating areas of the sympathetic nervous system (Young, Young, Klein, Klein, Klein, & Beyer, 1977). In toxic doses, stimulants produce stereotypic behaviors called punding (Rylander, 1972), irritability, slurred speech, ataxia, tremor, paranoia, hallucinations, and death (Louria, 1969; Holbrook, 1983a). Young students are frequently introduced to stimulants to fight fatigue and increase alertness during active periods. Although physical withdrawal symptoms are not evident when stimulants are suddenly discontinued, residual psychological effects may continue for several months (Holbrook, 1983a). Physical tolerance can develop in hours to days with continued use, especially with .. speed" or methamphetamine (Schuckit, 1984, p. 86).
NEUROPSYCHOLOGICAL SEQUELAE OF SUBSTANCE ABUSE BY YOUTHS
Trites et al. (1974) evaluated the neuropsychological functioning of a group of average-IQ adolescents who were amphetamine users. Results of the Halstead-Reitan Neuropsychological Battery failed to indicate brain dysfunction, the impairment index being 0.30. In a well-controlled study of chronic stimulant users, Grant, Adams, Carlin, and Rennick ( 1977) were unable to find evidence of neuropsychological impairment. A few researchers (Rylander, 1972; Connell, 1966; Schuckit, 1984; Young et al., 1977; Louria, 1969) have suggested that the use of stimulants may damage the CNS, disrupting memory, concentration, and abstract reasoning skills, but none of the researchers presented corroborating evidence.
315
other abused substances, such as glues, solvents, and paint sprays (Callen, 1984), does not target a single organ system. Comstock and Comstock ( 1977) reported that nervous system, liver, and kidney injuries have been associated with toluene use. N-Hexane use is clearly associated with polyneuropathy (Grant & Reed, 1985). Gasoline vapor inhalation can cause motor neuronal degeneration, disorders of the hematopoietic system, and encephalopathy owing to the various additives such as TCP, lead, and benzene. Aerosols frequently contained Freon, which can cause hypercapnia, cardiac arrhythmia, renal tubular necrosis, uremia, and sudden death (Wilde, 1975).
Glue Sniffing
Inhalants Inhalant abuse involves the voluntary inhalation of the fumes from aerosols, anesthetics, or other substances in an effort to achieve an intoxicated state. The "high" is dose-related, cumulative over a short period of time, and may persist for periods ranging from a few minutes to a few hours. Depression, sedation, disorientation, coma, and death may result. Inhaling substances to enhance religious or mystical experiences goes back centuries (Berry, Heaton, & Kirby, 1978). Sniffing glue to achieve intoxication was virtually unknown in the United States until about 1960 (Korman, Trimboli, & Semler, 1978). Soon thereafter, glue sniffing became epidemic in many other countries. Solvents were attractive because they were widely available, inexpensive, rapid in their ability to produce a "high," and they left the user with only a mild hangover (Cohen, 1975). Since 1981, inhalant use has been on the rise (Sharp & Korman, 1980). Johnston et al. (1985) found that 19% of their high school seniors had used inhalants sometime during their lives, with II% having used the drug in the previous year. In working with inhalant-abusing youths, whom courts have placed in treatment facilities, it has been reported that Hispanics and blacks favored spray paint and Anglos favored gasoline. Young et a/. ( 1977) reported that inhalants seem to be used primarily by those between 8 and 16 years. Inhalants are diverse in their physiological, pharmacological, and behavioral effects, yet most researchers tend to treat inhalants as a single class of drugs. Toluene, which may be the most widely abused inhalant, either by itself or as an ingredient in
Grant and Reed ( 1985) reviewed the available literature on the neuropsychological consequences of toluene inhalation. They felt that although toluene had a fairly high margin of safety, evidence indicated that long-term heavy use produced generalized cerebral atrophy and neuropsychological deficits. Allison and Jerrom (1984) compared the psychological test performance of I 0 Scottish adolescents (mean age 15 years) who had inhaled solvents for an average of 4i years against 10 matched control subjects who had never inhaled solvents. The major toxic component in the glue was toluene/acetone. Intelligence, memory, and attention were assessed in both groups by means of the Vocabulary and Block Design subtests of the WISC, the Wechsler Memory Scale (WMS), and the Paced Auditory Serial-Addition Task (PASAT). The results of this study indicated weaker attention, verbal and visual memory, and visuospatial problem-solving skills in long-term users. Verbal abilities remained intact. Two criticisms were made regarding the design of this study. First, the test administrator was aware of the group assignment of each subject. Second, because of the recency of the last use of solvents ( lO days to 6 weeks), acute rather than long-term consequences may have been identified. More clear-cut findings were reported by Channer and Stanley (1983) in their examination of a 16year-old male who had begun to experience visual hallucinations after inhaling a toluene/ acetone-based glue for 3 months. Four months after he had stopped sniffing glue, a normal cr brain scan was recorded with a diffusely abnormal EEG and delayed visual evoked responses (VER). After 8 months, the hallucinations continued, the EEG was unchanged, and the VERs were only minimally improved. Tsushima and Towne (1977) carried out an ex-
316
CHAPTER 17
tensive neuropsychological evaluation on a group of 20 ethnically mixed, paint-sniffing (primarily toluene-based) individuals (mean age of 18.5 years, range 11 to 24 years) from low-income housing projects. Although other drugs were used, paint fume inhalation was primary. They reported an average use of 2. 3 cans of paint per day, for periods ranging from 2 to 13 years. A group of20 non-paint-sniffing individuals from the same neighborhood, matched for age, education, and SES, served as a control group. The test battery, administered to both the experimental and the control group, included the Finger Tapping Test, Seashore Rhythm Test, Trail Making Test, Grooved Pegboard Test, and Coding subtest of the WISC-R, Stroop Color and Word Test, MemoryFor-Designs Test, and the Peabody Picture Vocabulary Test. The data revealed significant differences (2:: 0.05) between the sniffers and the controls on 11 of the 13 measured variables, with sniffers exhibiting lower levels of performance on measures of motor speed, auditory discrimination, visuomotor function, and memory. Limitations included the possibility of preexisting deficits in the sniffers, the admission by all of the sniffers that they had sniffed paint the day before the assessment, imprecise paint-sniffing information, and lack of information about unknown elements in the paint. In one of the earliest studies of glue sniffing, Massengale, Glaser, LeLievre, Dodds, and Klock (1963), evaluated 12 boys (aged lO to 16 years, median age 13 years)for cognitive function 12 hours to 7 days after their hist contact with glue. Toluene was the major volatile component. Five tests were administered, measuring attention, fine motor performances, detection of changes, design integration, and design recall. Despite inhalation of glue for periods ranging from 1 to 42 months, there were no significant differences between the performance or physical characteristics of the glue sniffers and a matched control group.
Gasoline Inhalation The inhalation of gasoline fumes for the purpose of intoxication became a widespread problem following World War I (Lewis & Patterson, 1974). This fad quickly passed and was largely forgotten for a time. In the 1950s, literature on the effects of gas sniffing began to reappear (Clinger & Johnson, 1951; Faucett & Jensen, 1952), but the focus had changed from the effects on adults to the effects on children. During the last few years, research on gas sniffing has focused on children (Lewis & Patterson, 1974). Researchers have also been concerned about the effects of leaded gasoline (Remington & Hoffman, 1984) because
lead, as a causative agent, has been implicated in irreversible changes in cogntion (Ross, 1982). Valpey, Sumi, Compass, and Goble (1978) discussed an individual who began to inhale gasoline and glue at the age of 13 years. Treatment for acute symptoms of gas sniffing began at the age of 17 years. One week after discontinuation of gas sniffing, some skills improved. Nine months later, he was readmitted after returning to gas sniffing. Tremors, bizarre behavior, poor time orientation, and deficits in recall of immediate and recent events were evident. Two weeks after entering treatment, the tremors subsided. Two years later, at the age of20 years, he was again treated for gasoline inhalation. The earlier symptoms returned and he became violent, suffered a convulsion, and developed symptoms of severe dementia. Nine months later, he was again admitted to the hospital, but died three days later. An autopsy revealed mild cerebral cortical atrophy with ventricular dilation and patchy cerebellar atrophy. The hippocampus, pons, thalamus, and pallidum were all adversely impacted. As his serum lead content was markedly elevated on every admission, it was suspected that tetraethyllead caused the dementia, ataxia, dysmetria, and dysarthria. Although the researchers indicated that the patient had suffered chronic encephalopathy from sniffing gas, it was quite apparent that the patient did not stop sniffing gas for more than short periods of time following each of his hospital discharges. Therefore, the effects could be more properly classified as acute rather than chronic. Seshia, Rajani, Boeckx, and Chow's (1978) study supported the Valpey et al. (1978) study. They completed standard neurological examinations (EEG, Nerve Conduction Study, ECG, blood profile, X ray, and urine assay) on 50 Native Americans, aged 4 to 20 years, who had _been referred to a pediatric neurology service for evaluation and treatment following inhalation of leaded gasoline. Inhalation of gas vapors occurred for periods from 6 months in the youngest child to over 5 years in those older than 12 years. Acute, but abnormal neurological signs were apparent in 92% of the patients at day one. At 8 weeks, only one patient manifested residual neurological signs. This study suggested that although the acute neurological effects of gasoline vapor inhalation are acutely significant, over time the adverse effects dissipate.
Multisolvent Inhalation Many investigators examining the effects of inhalants have attempted to limit their research population to youths who abuse only one type of solvent.
NEUROPSYCHOLOGICAL SEQUELAE OF SUBSTANCE ABUSE BY YOUTHS
Other investigators have grouped their subjects together, regardless of the type of inhalants used. Comstock (1977) evaluated 22 patients (mean age 17 years, range 13-26 years) who sought hospital treatment for solvent (toluene primarily) abuse. The patients had used inhalants an average of 4 years (range 1-11 years). Twenty of the patients reported abuse of drugs as well as inhalants. The evaluation process included EEG, EMG, nerve-muscle biopsy, and a general medical lab exam. The Halstead-Reitan was administered to two patients. Despite heavy exposure to volatile solvents, the medical exam did not find CNS abnormalities. A mental status exam identified "mental-grasp deficiencies'' in 55% of the patients soon after their admission. Two weeks later, there was a clearing of these deficiencies. Of the two patients administered the Halstead-Reitan, one scored in the pathological range on a number of the subtests. The author hinted at the possibility of residual impairment adversely affecting the ability of the patient to benefit from ''talking psychotherapy.'' The individual impaired on the Halstead-Reitan remained impaired after 2 weeks. Limitations in this study included lack of a control group and failure to report on preexisting or contributing conditions. In a well-controlled study, Bigler ( 1979) performed extensive neuropsychological evaluations on ten 16- to 19-year-olds (mean age 17.8 years) with a history of chronic inhalant abuse (range 2-6 years). The same testing was performed on three control groups matched for age, sex, and education. The respective control groups ( 10 patients each) were: (a) brain-damaged, (b) psychotic non-brain-damaged, and (c) non-brain-damaged/nonpsychotic patients. Neuropsychological evaluations were begun about 48 days after admission. They included administration of theWAIS, Category Test, TPT, Rhythm Test, Speech-Sounds Perception Test, Finger Oscillation, and Trail Making Test. The inhalant abuse group scored within the impaired range on most neuropsychological measures and was similar to the braindamaged group in neuropsychological functioning. The nature of the neuropsychological deficits suggested diffuse cerebral dysfunctioning. A factor that may have influenced the findings was medication effects-7 of the I0 inhalant abusers were on a neuroleptic medication (Thorazine, Haldol, Cogentin, Mellaril, or Navane) of unreported dosage. In one of the most widely cited and carefully designed studies of inhalant abuse, Berry et al. (1978) evaluated 37 inhalant-abusing youths (average age 18 years, range 14-29 years) who were referred to a drug abuse treatment program. Sixty-two percent were Hispanic, and they averaged five arrests
317
in the preceding 2 years. The inhalant-abusing group used an inhalant an average of over three times a day for periods of time ranging from 1i to 17 years (average 5.5 years). The control group was comprised of II subjects matched for age, sex. ethnicity. education, background, and use of substances other than inhalants. The MMPI, WAIS, Category Test, TPT, Speech-Sounds Perception Test, Seashore Rhythm Test, Finger Tapping Test, Trail Making Test, Aphasic Screening Test, Spatial Relations, Reitan-Kl~ve Sensory-Perceptual Examination, Tactual Form Recognition, Grip Strength, Motor Steadiness Battery, Grooved Pegboard, Hole-Type Steadiness Test, Maze Coordination Test, and a modified Reitan Story Memory Test were administered to all subjects. The neuropsychological test scores were generally lower for the inhalant group than for the controls. The TPT, Tactual Form Recognition, Grip Strength, Maze Coordination, measures of learning efficiency, memory score on the Story Memory Test, and both neuropsychological summary scores reached levels of significance. Berry et al. concluded that inhalant abuse caused impairment. Korman et al. (1978) assessed 273 individuals (average age 21 years) seen in a psychiatric emergency room over a 12-month period. With information obtained in a standardized psychiatric interview, the researchers were able to establish three experimental groups and one control group, all matched for sex, age, and ethnicity. The experimental groups consisted of 37 inhalant users who used no other drugs, 54 inhalers who used other drugs, and 91 noninhalant multidrug users. Interview process ratings between the four groups on abstraction, insight, judgment, and other cognitive functions were more severely impaired for the inhalant groups than for the other groups. Noting the conflicting results with well-designed and -controlled studies, Korman, Matthews, and Lovitt ( 1981) gathered 109 volunteer teenage subjects for extensive neuropsychological testing. All of the subjects had used drugs but the range of use was unreported. Sixty-eight reported inhalants as their major drug of abuse, and 41 reported drugs other than inhalants as their major drug of abuse. The Wechsler IQ test, most of the Halstead-Reitan Neuropsychological Battery, grip strength, motor speed, and academic achivement measures were administered to all. Approximately 30% of the performance measures were completed with significantly lower levels of performance by the inhalant abusers than by the other drug abusers. Swprisingly, the more global and complex measures, such as the Wechsler IQ and academic measures, were most clearly implicated, although both groups scored well
318
CHAPTER 17
below average. The authors concluded that inhalant abusers developed deficits in a wider range of cognitive areas than was previously known. Some caution should be taken in interpreting the results of this study as there was no drug-using control group, all of the abusers were multidrug users, the performances of both groups were weak, and there was no mention of the time elapsed since the last use of a drug.
Other Inhalants Acute physiological data and industrial exposure results have been reported and compiled by many different investigators for typewriter correction fluid, deodorants, Pam, polishes, amyl nitrate, nutmeg, chloroform, and ether (Chenoweth, 1977; Lowry, 1979; Akesson, 1965; Lewis & Patterson, 1974; Smialek, 1985), but none of these substances have been investigated neuropsychologically. The evidence is clear that inhalation of most, if not all, solvent fumes produces obvious acute CNS effects, including peripheral neuropathy. The issue is not as clear regarding long-term chronic effects. Most of the available research indicates a clearing of the nervous system over time when the primary abused solvent is gasoline or glue. The evidence is much more suggestive of permanent CNS impairment when multiple solvents are inhaled. Such impairment is apt to be general and most likely will affect the complex, higher-order neuropsychological processes. Methodological problems and experimental artifacts complicate conclusions. Very few studies are longitudinal and rarely is the issue of "presniffing" differences between experimental and control groups addressed. Although multidrug users are identified as such, rarely are the types or extent of secondary drugs used by subjects fully identified. Although radiological investigations could add valuable diagnostic information, such techniques are rarely used in combination with neuropsychological investigations to confrrm or refute the presence of organic changes.
Phencyclidine (PCP) PCP was created by pharmacological researchers as an animal anesthetic in 1956. In 1957, human clinical trials were initiated, but the discovery of adverse side effects forced discontinuation of all human research by 1965. PCP made its appearance as a "street drug" in 1966 in San Francisco, and was marketed as the "PeCe Pill" (Reed & Kane, 1972).
B~ause of unexpected and unpleasant effects, it soon began to appear to new guises and in combination with other drugs (Schuckit, 1984; Tong, Benowitz, Becker, Forni, & Boerner, 1975). Although PCP is generally identified by users as a hallucinogen, it cannot be accurately placed in any single drug category. Most experts identify PCP as a tranquilizer-anesthetic with hallucinogenic properties (Young et al., 1977).lts effects vary, depending on the dosage, other drug involvements, age of the user, personality traits, route of administration, and the physical setting. PCP increases the activity of the brain dopamine system and interferes with the synaptic transmission between cells (Institute of Medicine, National Academy of Sciences, 1985; Schuckit, 1984). The use of PCP among adolescents declined between 1979 and 1981, but there has been little change since then. In the eighth annual High School Senior Survey (Johnston et al., 1985), 5% of high scllool seniors admitted to use of PCP at some time in their lives, with 1% admitting to use during the preceding month. Little is known about the neuropsychological effects of PCP, although clinical observations have indicated a progression from irritability to violence as PCP use is continued (Fauman & Fauman, 1979). Mental health workers, friends of users, and users themselves report decreased intellectual functioning and confusion long after use of the drug has been discontinued (Ware, 1979; Schuckit, 1984). PCP produces dramatic changes in behavior, memory, perception, and orientation. Because of these effects, researchers have been concerned that PCP use may result in permanent damage to the CNS, yet there has been very little formalized research completed on neuropsychological sequelae of PCP abuse (Grant & Reed, 1985; Light, 1984). Ware (1979) compared the performance of eight chronic PCP users to that of a group of eight multidrug non-PCP users. The individuals [mean ages 19 and 20 years (range 16-35 years), respectively] were inpatients at a state hospital, and were matched for age, education, race, sex, current medication and dosage level, and handedness. To ensure that they were not showing acute effects of their drug abuse, testing was not started until all of the subjects had been in the hospital a minimum of 3 weeks. The Halstead-Reitan Neuropsychological Battery and WAIS were administered to all subjects. The PCP users performed worse than the multidrug users on theWAIS Verbal IQ, Performance IQ, and Full Scale IQ, on the Information, Comprehension, Picture Completion, Block Design, and Object Assembly
NEUROPSYCHOLOGICAL SEQUELAE OF SUBSTANCE ABUSE BY YOUTHS
subtests of the WAIS, and on the Speech-Sounds Perception test of the Halstead-Reitan Battery. The raters considered the PCP group to be more neuropsychologically impaired than the multidrug group, but the differences were not statistically significant. The researchers hinted at the possibility that chronic users may experience more impairment in right than left hemispheric functions. Ware considered theresults equivocal because the predrug intelligence levels were unknown and, though the data suggested poorer performance by the PCP user group, much of the data did not reach levels of statistical significance. Additionally, the study lacked a non-drugusing control group, the sample sizes were small, half of the patients were on psychotropic medication, the two groups differed in IQ, a history of significant alcohol consumption was apparent in most of the patients, and most of the patients had a non-drugrelated psychiatric disorder. Such problems are common in this field, however. Additionally, the power of random assignment and careful control of the abused drug are control methods obviously unavailable to the researcher. Carlin, Grant, Adams, and Reed (1979) addressed the issue of chronic use of PCP by employing a non-drug-using control group in their study. They compared three groups of 13 recruited subjects on the Halstead-Reitan Battery, W AIS, and MMPl. Group I were persons who had taken PCP an average of 27 months (range 1 month to 27 years); Group II, persons who had used a variety of drugs other than PCP, but who had been drug-free for at least 3 weeks; Group III, a non-drug-using comparison group. The three groups (average age 25.9 years) were matched for education, sex, and race. Both drug-using groups displayed deficits in abstracting abilities and complex perceptual-motor skills, but these groups could not be distinguished on the basis of the test results. Weaknesses in the study, including a small sample size, reliance on self-reports for medical and drug use history, lack of external neurological criteria, higher IQ for the control group than the multidrug group, and a failure to control for the considerable amount of ethanol ingested by both drug groups, militate against much of the results. Crane (1984) evaluated the effects of chronic, long-term PCP ingestion on CNS functioning in adolescents. Three groups of individuals were selected from a residential treatment population and matched for age (average age 16 years, range 15-20 years), sex, ethnicity, and education. Group I comprised 10 subjects with a history of multidrug and PCP use; Group II, I 0 multidrug users who had not used PCP; Group Ill, 10 non-drug-using controls. Groups I and
319
II had used drugs at least twice a week for at least 6 consecutive months but, on average, had not used any drug during the 3-month period prior to testing. Each subject was given a standard neurological examination, EEG, and the WAIS and HalsteadReitan Battery. The only significant difference between the groups was that the PCP/multidrug group displayed more abnormal findings on Cranial Nerve VIII measures. Crane concluded that PCP's longterm effects may not impair cortical functions, but may impair subcortical functions regulated by the brain stem. In a five-part study of the effects of PCP on adolescent cognitive functioning, Light (1984) investigated preexisting medical records and test results to determine whether multidrug/non-PCP, moderate PCP, and heavy PCP use correlated with neuropsychological test results. In Study I, Light reviewed medical records from 116 middle- to uppersocioeconomic-class 17-year-olds who had been assessed an average of 12 days after admission to a drug treatment facility. Each subject was administered the Shipley Institute of Living Scale, Reading subtest of the Wide Range Achievement Test, Benton Visual Retention Test, and MMPI. Although heavy PCP users scored lower than the other groups on the Shipley, there was some question that the group had not fully recovered from the acute effects of PCP by the time they were tested. In Study II, Light (1984) administered a neuropsychological battery to fifty-two 18- to 21-year-olds who were classified as multidrug/non-PCP users, moderate PCP users, or heavy PCP users, matched on age, race, sex, and medication. Multidrug/nonPCP and moderate PCP users were assessed after 4-6 weeks of sobriety; heavy PCP users, after 149 days. Portions of the Halstead-Reitan Battery, two memory items from the Luria-Nebraska Neuropsychological Battery, the Arithmetic subtest of the WRAT, and five subtests (Information, Comprehension, Picture Arrangement, Block Design, and Object Assembly) from the WAIS were administered. •'Only one (Trails A) of the 27 neuropsychological variables was significantly different across the three groups, and this variable did not vary systematically with PCP use" (p. 82). Light was unable to match the subjects on a number of demographic variables, length of sobriety was much longer for the heavy PCP users, and the non-PCP group may have been more impaired due to problems associated with use of other drugs. The issue of the contribution of preexisting conditions to current levels of cognitive functioning in PCP/multidrug-using individuals was addressed by
320
CHAPTER 17
Light (1984) after he reviewed the school records of 37 of the 52 subjects in Study II. Light was able to obtain four or five achievement/ intellectual test scores on most subjects. The school test scores indicated average or below-average functioning for all three groups compared to the general population. There were no significant differences among the three groups. All researchers acknowledge that PCP has dramatic acute effects on the CNS, but the effects vary greatly among individuals. Dosage level appears to be a critical element, heavy users developing more pathology than light users. Because the acute effects of PCP linger for weeks, study of the chronic effects of PCP has been difficult. Still, most of the available research with adolescents and young adults does indicate neuropsychological impairment with chronic, long-term use. Motor performance skills appear to be more adversely affected than verbal skills.
Lysergic Acid Diethylamide (LSD) The most recent data on the use of LSD by adolescents indicate a decline over the last few years. In 1984, 8% of high school seniors reported use of LSD at some time in their lives. About 4% reported use during the previous year (Johnston et al., 1985). Although LSD is classified as a hallucinogen, it does not cause the user to hallucinate per se. Instead, the user views the world in a distorted fashion or misinterprets sensory experiences. LSD's acute effects are variable and unpredictable. Flashbacks, reported to occur days or months after the initial dose, are rare. LSD alters sensations, reasoning, perception, and mood states. The drug primarily affects the visual cortex, limbic system, and reticular formation (Holbrook, 1983b). Physical dependence and withdrawal do not occur, but tolerance quickly develops and abates. Of all the hallucinogens, LSD has generated the more neuropsychological research. Because of LSD' s extreme psychopharmacological potency, various researchers have speculated that CNS damage might result from repeated exposure. Cohen and Edwards (1969) completed one of the first studies of the effects of LSD on neuropsychological functioning. They administered the Halstead-Reitan Battery, Raven's Progressive Matrices, and a spatial orientation test to two groups of 21-year-old volunteers matched for age, sex, and education. The LSD user group had an average of 70 LSD experiences and were known to use other drugs.
The control group included inexperienced LSD users. If they used other drugs was not documented. Of the 15 measures obtained, the LSD user group displayed impaired functioning on the visuospatial orientation and Trails A tests. There was also a positive correlation between the number of LSD experiences and performance on Trails A and Raven tests. There was no evidence of generalized neuropsychological dysfunctioning. Biases in this study that may have distorted the findings include a lack of knowledge of premorbid levels of functioning and lack of awareness of the subject's alcohol and other drug experiences. Wright and Hogan ( 1972) replicated the Cohen and Edwards (1969) study with forty 20-year-old (age range 17-24 years) LSD users. The subjects had used LSD an average of 29 times over a period of from 4! to 27 months. None of the LSD users were judged acutely toxic at the time of testing, although 4 had used LSD the previous day. The HalsteadReitan Battery, Trail Making Test, WAIS, and an aphasia test were administered. The LSD group performed better than the non-LSD group on the Information subtest and worse on the Comprehension subtest of the WAIS. There were no other significant differences between the two groups. One reason why these findings may be at variance with those of Cohen and Edwards (1969) may be that the Wright and Hogan LSD users used LSD, on average, less than half as many times as the Cohen and Edwards LSD users. Cognitive and perceptual testing and EEG studies were completed on 21 volunteers (average age 20 years, range 15-27 years) who had ingested LSD an average of 65 times. None had ingested LSD for 48 hours prior to testing. An auditory two-tone evoked potential showed no abnormalities, but the LSD users were uniquely sensitive to low-intensity stimulation on visually evoked potential procedures. Although Blacker, Jones, Stone, and Pfefferbaum(1968) felt that some of the EEG findings and behaviors displayed by the LSD users were "suggestive of minimal brain damage" (p. 348), the evidence was not conclusive. Several methodological problems were apparent in this study. There was no matched control group, all of the LSD users used other drugs as well, some of the subjects may have been experiencing an acute reaction to some drug at the time of testing, and the cognitive measures were nonstandardized. Acord (1972) gathered a group of forty 17- to 24-year-old (average age 20 years) military hospital inpatients and outpatients, of average intelligence, who had ingested LSD, MTA, STP, mescaline, or psilocybin at least once. The Indiana Neuro-
NEUROPSYCHOLOGICAL SEQUELAE OF SUBSTANCE ABUSE BY YOUTHS
psychological Battery and WAIS were administered. Using Halstead's cutoff scores (Halstead, 1947), Acord reported that scores on the Category Test and the Tactual Performance Test were in the braindamaged range. There was no control group, no information regarding the subjects' last exposure to drugs, past neurological and psychiatric history were unknown, and, although different drugs were used, all of the drugs were grouped together as hallucinogens. Acord and Barker (1973) expanded the 1972 study by adding a control group. Subjects were drawn from the same military hospital and matched for intelligence and education. The experimental group of fifteen 21-year-old subjects had a history of ingestion of LSD, MTA, STP, DMT, mescaline, or psilocybin on at least one occasion. The fifteen 22year-old control had not used drugs. The Tactual Performance Test (TPT) Localization Component, Trails B, and Category Test were administed to all subjects. The drug-using group performed significantly less well than the non-drug-using group on the Category Test and the TPT Localization Component. The authors suggested the possibility of a causal relationship between use of hallucinogens and brain damage. In a study comparing LSD I mescaline users and marijuana users against a control group with no history of marijuana or LSD use, Culver and King (1974) administered the Halstead-Reitan Battery, WAIS, Laterality Discrimination Test, three spatial-perceptual tests from the Kit of Factor-Referenced Cognitive Tests, and the MMPI. Matched triads of 28 subjects each were established from a group of bright 20- to 25-year-old college seniors. The LSD and marijuana users were asked to stop using any drugs at least 7 days before they were scheduled to be tested. LSD users scored significantly lower than the marijuana and control groups on the WAIS Performance and Full Scale IQ, Picture Completion subtest, Cube Comparison Test, and the Trails A and B time scores. A year later, Culver and King repeated the same testing on new subjects who were similar, in some respects, to those in the first study. The second group of triads were more closely matched on the WAIS. The only measures on which the LSD user group was significantly weaker than the marijuana and control groups were the Trails Band A plus B time scores. For the LSD user group, the only measure that reached statistical significance for both studies was the Trail Making Test. Although the LSD user's Trail Making Test time was slower than the marijuana and control groups' time, their performances were within the normal range.
321
Neuropsychologists investigating the effects of LSD on adolescent and young adult development have generally used the same instrumentation. Although most studies reported impaired neuropsychological performance with chronic use of LSD, the significant findings were inconsistent from study to study probably because of preexisting differences among the populations studied, length of use of LSD, and time since last use of LSD. Most studies reporting significant findings suggest impairment in abstracting and speeded sequencing tasks, with overall levels of functioning remaining intact. Thus, it appears that LSD has subtle, dosage-dependent effects on the CNS when it is used chronically.
Heroin In the 1960s and 1970s, a young population of addicts came to be recognized. Opium-rich Southeast Asia had produced many addicts-American soldiers returning home from Vietnam (Woolf, l983a). Between 1975 and 1981, the use of opiates (other than heroin) by high school seniors remained stable, as has opium use by high school seniors since 1980. Of surveyed seniors, 9.7% report using an opiate (other than heroin) at least once during their life, whereas only 1.3% report having ever used heroin (Johnston et al., 1985). Heroin is a narcotic analgesic of the opiate class. It is a semisynthetic derivative of morphine but can be three times as potent as morphine (Young et al., 1977). Other opiate analgesics include opium, codeine, hydromorphone (Dilaudid), oxycodone (Percodan), propoxyphene (Darvon), meperidine (Demerol), diphenoxylate (Lomotil), and pentazocine (Talwin) (Schuckit, 1984). The actions of these drugs are homogeneous, tolerance develops rapidly, and all are extremely addictive. Heroin is generally injected into the body, but can also be administered orally, snorted, or smoked. Some addicts mix cocaine or amphetamines with heroin, producing a mixture called "speedball." Concerning autopsies on persons who have died of a heroin overdose, 60% show significant cerebral edema (Parsons & Adams, 1983). Hill and Mikhad (1979) found significantly smaller sulci and ventricle/brain indexes in their study of adult heroin abusers. The only study examining the long-term cerebral effects of heroin addiction on younger individuals assessed performance on the Halstead-Reitan Battery. Fields and Fullerton (1975) compared
322
CHAPTER 17
the performance levels of a group of heroin-addicted veterans with a brain-damaged group and a control group without a history of brain damage or drug use. The three groups of 25 hospitalized subjects each were matched for age, sex, and education. The heroin and control groups were also matched on IQ. The heroin addicts had used heroin an average of almost 5 years (range 1-10 years). The brain-damaged group performed significantly worse on the Halstead-Reitan Battery than the heroin and control groups, whose scores were equivalent. Information regarding premorbid history, use of alcohol, and extent of use of heroin was not provided. Platt, Scura, and Hannon (1973) compared a group of youthful incarcerated heroin addicts with a group of nonaddicted controls with the Means-Ends Problem Solving (MEPS) story completion technique. The MEPS measures how appropriately and effectively an individual can resolve typical real-life problem situations. The 28 heroin addicts had a minimum of a 2-year period of consistent addictive behavior (mean 3.77 years). There were 31 nonaddicted controls with no history of use of heroin. The heroin addicts were about 22 years old and had 10.6 years of education. The control group averaged about 20 years old and had 9.2 years of education. Results indicated that the heroin addicts were less able to generate solutions to problematic life situations. Because of the absence of research on the longterm effects of heroin addicition on brain-behavior relationships, it is not clear, at present, what effects chronic heroin use has on neuropsychological functions. Further study is required, particularly with heroin-using individuals who are not using other drugs.
environmental features all affect the individual's reaction to the drug. Tolerance can quickly develop, which makes barbiturates highly dangerous. Judd and Grant (1975) studied a group of 50 multidrug users who were heavy users of depressants, a group of 19 neurologically intact medical patients, and a group of 19 brain-damaged patients (average ages 25, 24, and 22 years, respectively). Two to three weeks after admission to the hospital, the three groups were administered the HalsteadReitan Battery, WAIS, and MMPI. Of the 18 multidrug users who also had a history of head injury, 9 manifested neuropsychological abnormalities. The drug users performed significantly worse than the medical patients on 13 of 29 measures. Abstracting ability, accuracy of perception, motor speed, nonverbal learning, and accuracy of perception were impaired functions in the drug user group. Although the investigators noted some limitations to their study, they failed to address the issue of the age range of the drug-using group (14 to 54 years). With such a wide range of ages, it is possible that more significant results were canceled out, as it is known that there is considerably more variability in performance levels of older adults compared to younger adults (Elliott, 1985). In summary, it appears that for some young people, use of sedatives and barbiturates over a long period can lead to impaired neuropsychological functioning. Available research continues to use poor research methods. Sample sizes are small, use of multipie drugs is not controlled in the research designs, premorbid abilities are unknown, and follow-up studies rarely track a subject for more than 4 weeks.
Sedatives/Barbiturates
Alcohol
Since the NIDA High School Senior Survey began in 1975, the use of barbiturates and other sedatives has been on the decline. In 1984, 9. 9% of high school seniors reported using barbiturates and 13% reported use of sedatives at some point in their lives (Johnson et al., 1985). Barbiturates are CNS depressants that vary in duration of action. The fast-acting subclass is used to induce anesthesia (sodium pentothal), the short- to immediate-acting subclass aids in sleep induction (pentobaribtal), and the long-acting subclass is used to treat chronic conditions such as epilepsy (phenobarbital). Barbiturates depress the CNS and interfere with synaptic functioning. Dosage, type of barbiturate, metabolism, use of other drugs, and
Alcohol is the most misused drug in the United States today. It is by far the most widely used drug by 12- to 17-year-olds (Nathan, 1983; Beyette, 1983).lt is estimated that 6% of high school students drink on a daily basis (Beyette, 1983; Cohen, 1985). Johnston et al. (1985) noted in their annual high school senior survey that 92.6% of the seniors had used alcohol at some time in their lives, and 67.2% had consumed alcohol in the previous month. Since 1979, there has been no change in lifetime prevalence, although daily use has declined. Beer, wine, and distillates share the same active ingredient-ethyl alcohol or ethanol, which is the intoxicant. Alcohol is a depressant that adversely affects most body systems, especially the CNS. De-
NEUROPSYCHOLOGICAL SEQUELAE OF SUBSTANCE ABUSE BY YOUTHS
pressed reaction time, muscular incoordination (American Academy of Pediatrics, 1984), agitation or depression, sensory impairment, and disinhibition (NIAAA, 1982) coma, and death occur with increasing dosage levels. Tolerance and physical and psychological dependence develop over time, with diagnosable alcoholism becoming established after 3 to 15 years of prolonged use (Young eta/., 1977). Damage to the CNS upon chronic use of alcohol has been well established. Nerve centers are destroyed (Board on Mental Health and Behavioral Medicine of the Institute of Medicine, 1985), the lateral and third ventricles enlarge (Wilkinson, 1985), there may be a marked reduction of cerebral blood flow (Tarter, 1980), sulci widen, the gyri are narrowed (Callen, 1984), peripheral neuropathy develops (Schuckit, 1984), there can be a loss ofNissl substance (RNA), the substantia nigra can be damaged, lesions can develop anywhere in the cerebral cortex or in the central brain structures, and generalized diffuse, nonspecific, atrophy can take place (Freund, 1985; Grant & Reed, 1985; Eckhardt & Ryback, 1981). Functional deficits accompany CNS damage attributed to alcohol abuse, although age, drinking history, use of other drugs, educational background, socioeconomic status, ethnic background, and perhaps sex may modify behavioral and cognitive deficits on neuropsychological instruments (Goldman, 1983). The majority of research studies of alcoholics have been done with inpatients, abstinent 1 to 4 weeks (Grant & Reed, 1985). Investigations have revealed disturbances in abstracting ability, complex perceptual motor skills, visual perceptual capabilities, learning capacity, visuospatial integration, short- and long-term verbal and nonverbal memory, information processing rate, sequential organization capabilities, visual search and scanning skills, concentration, perceptual analysis, synthesis, orientation, and sequencing skills, manual dexterity, memory functions, and acquisition of new information (Wilkinson, 1985; Tarter & Edwards, 1985; Eckhardt & Ryback, 1981; Grant & Reed, 1985). Global intellectual capabilities and language skills have not been implicated (Tarter & Ryan, 1983; Eckhardt, Parker, Noble, Feldman, & Gottschalk, 1978; Parsons & Leber, 1982; Goldman, 1983). Although the study of the social and personality effects of alcohol on children and adolescents has received considerable attention in recent years, there have been very few studies of the neuropsychological status of drinking children and teens (Tarter & Edwards, 1985). Defining alcohol abuse in these studies of youthful drinkers was difficult because many of
323
the characteristic medical problems, such as cirrhosis and pancreatitis associated with alcohol abuse, had not had enough time to develop. The cognitive performance of young sober social drinkers was investigated by Hannon, Day, Butler, Larson, and Casey (1983). The Shipley Institute of Living Scale, Wisconsin Card Sorting Test, Digit Symbol Test, Trail Making Test, and TPT were administered to a group of 52 female and 40 male college students (mean age 20.3 years). They were asked to refrain from taking any drug or consuming alcohol for 24 hours prior to testing. The results indicated decreased neuropsychological performance with increased quantity of alcohol per occasion and total lifetime consumption by both sexes. Women appeared to be more adversely affected, neuropsychologically, by social drinking than men. Hannon et al. ( 1985) replicated their 1983 study with a larger sample, a longer period of no alcohol use, and a few changes in the tests administered. There were I 03 female and 67 male co11ege students (mean age 20.8 years). The Trail Making Test and TPT were dropped from the test battery and Raven's Advanced Progressive Matrices was added. Subjects were asked to refrain from drinking any alcohol for 2 weeks prior to the testing. Although there was a predicted negative relationship between quantity and cognitive variables for women, and three unpredicted relationships for men, the correlations were weak and conclusions speculative. In the only published study investigating the role of age of drinking onset upon neuropsychological functioning, Portnoff (1982) administered the WAIS, Shipley Institute of Living Scale, Wechsler Memory Scale, and an abbreviated version of the Halstead-Reitan Battery to two groups of 10 chronic alcoholics dichotomized by age of drinking onset. Portnoffhypothesized that differential onset of drinking would be related to brain maturation and would be reflected in neuropsychological test results. The early onset group started drinking steadily at 14.1 years (range 12-18 years) and drank for an average of 19.4 years. The later-onset group started drinking steadily at 23.4 years (range 20-30 years) and drank for an average of 16.5 years. All of the subjects discontinued using all medication and had been alcoholfree for an average of 3. 8 weeks prior to testing. The two groups were essentially similar in sex, education, race, and handedness. Significant differences on the Shipley, Wechsler Memory Scale, Category Test, Trails B, and Rhythm Test suggested that early onset alcoholics developed more impairment on measures of abstraction, rhythm perception, visuospatial sequencing, verbal memory, and figural memory. It
324
CHAPTER 17
appeared that individuals who started drinking at an early age may have been more vulnerable to the adverse neuropsychological effects of alcohol than individuals who started drinking at a later age. The researchers did not review early school or medical records to control for premorbid developmental differences between the two groups, the early onset drinkers had been drinking 3 years longer than the later-onset drinkers, and the sample sizes were very small. Thus, the findings of this study, although strongly suggestive, were not conclusive. The very few studies focusing on the neuropsychological effects of alcohol on children have suggested deficits in abstracting and problem-solving skills. Memory deficits have not been implicated, although very few child studies have adequately tested memory functions. In child and adolescent alcohol use studies, longitudinal investigations are needed to pinpoint specific neuropsychological functions affected by alcohol use correlated with chronological age.
may be the most critical drug issue facing our youth today (Gold eta/., 1985) because of the synergistic effect of certain drug combinations. For instance, use of alcohol and depressants such as sleeping pills, hallucinogens, and stimulants may lead to a potentiation of side effects (Schuckit, 1984). It has been shown that in today's Western society, there is a progression with respect to drug use. If a young person who drinks and smokes elects to use other substances, the next choice will probably be marijuana, then a hallucinogen, depressant, or stimulant (Hamburg, Kraemer, & Jahnke, 1975). The most extensive study of multidrug abuse (Grant et al., 1978) investigated the neuropsychological functioning of 151 25.5-year-old multidrug users, 66 28.8-year-old psychiatric patients, and 59 26.1-year-old non-drug-using, nonpatient volunteers. Each subject was given the Halstead-Reitan Battery, W AIS, a grooved pegboard test, and an MMPI. The multidrug users were evaluated between 21 and 30 days after they enrolled in treatment. Clinical evaluations of the protocols results in 37% of the multidrug users, 26% of the psychiatric patients, and 8% of the nonpatient group falling within the Multidrug Abuse neuropsychologically impaired category. Age, education, premorbid medical factors, and extensive use Most members of today's society who misuse of depressants and opiates in the multidrug users, and one drug tend to misuse at least one other drug as well extensive use of antipsychotic drugs in the psychi(Maddux, Hoppe, & Costello, 1986; Schuckit, 1984; atric group appeared to be related to neuropsychologParsons & Parr, 1981; Gold, Verebey, & Dackis, ical impairment. A factor analysis of the data re1985). Maddux et al. (1986) surveyed 133 senior vealed significantly weaker performance by the medical students and found that most had abused multidrug users on general verbal intelligence, visusome type of psychoactive substance sometime be- al-motor, tactile-motor, and perceptual skills. fore entering medical school. Most of those who had Three months later, 61% of the multi drug users, used a drug reported use of more than one drug. In a 77% of the psychiatric patients, and 86% of the nonsurvey of Army patients, over half of those reporting patient group were reevaluated. Only mild improveto a multidrug clinic reported use of three or more ment was noted in the multidrug group. Neurodifferent substances (Cook, Hostetter, & Ramsay, psychological impairment was found in 34% of the multidrug users, 27% of the psychiatric group, and 1975). When two or more drugs are introduced into the 4% of the nonpatient group. Again, heavier use of body, the potential for an interactive relationship ex- depressants and opiates, as well as premorbid illness ists. This interactive effect may be additive, potenti- history and recent drug taking, was related to neuroated, or inhibiting. An additive effect occurs when psychological impairment. The authors speculated the effect of two or more drugs is greater than could that impairment associated with depressant and be achieved by one drug, but no greater than addition opiate use may be enduring and that multidrug users of the drug responses. Potentiation occurs when one may be deficient in verbally mediated problem-solvdrug enhances the effect of another drug when the ing skills. Ivnik (personal communication, February two are taken together. The combination of the drugs may result in an effect that is greater than could be 1986) administered the WAIS-R or WISC-R, the achieved by simple addition of the drugs. Inhibition arithmetic subtest of the WRAT, the reading section occurs when one drug diminishes or reverses the ef- of the Woodcock-Johnson Psychoeducational Test fect of another drug when the two drugs are taken Battery, the A YLT, and the MMPI to II 0 middleclass, 13- to 19-year-olds (average 15.9 years) who together (Woolf, 1983b). Some researchers believe that multidrug abuse were in treatment for drug abuse. The primary drugs
NEUROPSYCHOLOGICAL SEQUELAE OF SUBSTANCE ABUSE BY YOUTHS
of abuse were marijuana and alcohol. Multidrug abuse ranged from I to 4 years, with wide variations in individual use. All testing was completed within I0 days after initial presentation. Test results failed to identify evidence of significant cognitive dysfunction. Complications in this study included an absence of a control group and a wide variation between subjects in extent of drug use. In an investigation of the neurological and neuropsychological consequences of PCP use by adolescents, Crane (1984) compared 20 PCP/multidrug (except inhalants) users, 10 multidrug users who had not used PCP, and 10 nondrug users. The subjects, who were in a residential treatment facility, we~ between 15 and 20 years of age, and were matched for age, sex, and education. A neurological examination, neuropsychological evaluation, and EEG were administered to each subject. The subjects had been free of drugs an average of about 3 months at the time of testing. Although Crane was unable to identify any deficits in the multidrug user group, that group differed from the other two with respect to ethnicity and may have experienced a greater degree of social, psychological, and intellectual development. Neuropsychological impairment was found in a group of young multidrug users who had been free of drugs an average of 2 months. Grant, Mohns, Miller, and Reitan (1976) compared the performance levels of a group of 22 male multidrug users (mean age 22 years) in a residential drug treatment program against the performance levels of a group of 19 medical and 19 neurological patients matched for age, education, and sex. At the time of testing, most subjects had been drug-free for 2 months. The evaluation included the Halstead-Reitan Battery, WAIS, and MMPI. About half of the multidrug users showed mild, generalized neuropsychological impairment; 11 to 26% of the medical patients and 84 to 89% of the neurological patients were judged as impaired. There was no apparent association between impairment and the use of any specific drug. In contrast, Bruhn and Maage (1975) were unable to identify any neuropsychological deficits associated with heavy use of combinations of marijuana, hallucinogens, amphetamines, and opiates. Eightyseven Danish male prisoners (average age 23.1 years), matched for age and educatjon, were divided into four groups, ranging from no drug use to heavy use of four or more drugs. Each subject was given the WAIS, Category Test, learning and memory tests, Seashore Rhythm Test, Hidden Patterns Test, and a reaction time test. There were no apparent differences between the groups in abstract reasoning, concentration, perception, learning capacity, or any
325
other area of cognitive functioning. The researchers assumed that the subjects were not using drugs during testing because they were prisoners. Today, with our knowledge of the extensive use of drugs in the prison system, such an assumption seems naive. The research findings on the effects of multidrug abuse in adolescents are mixed. A common finding in studies of drug use in adolescent delinquents is deficient verbal IQ on the Wechsler scales and a higher frequency of multidrug use than for nondelinquent drug users. The directionality of causation in this relationship remains open. Studies that have failed to find neuropsychological deficits seem to involve the youngest subjects, lack a control group or, if there is a control group, the control group is poorly matched with the experimental group. Studies that have found a positive relationship between multidrug use and neuropsychological impairment have tended to associate the impairment with use of depressants and opiates. When impairmel)t was evidenced, the primary systems affected included visual, tactile, motor, perceptual skills, and verbally mediated functions.
Designer Drugs This short section deals with a group of substances that have been created in the streets in an attempt to circumvent existing laws against production and distribution of certain controlled drugs. The term "designer drug" has been adopted by drug regulation agencies and the media to describe a mindaltering substance that is a chemical variation of another substance already regulated under the federal controlled substances act. The chemical variation is achieved by changing a few molecules of an existing illicit drug so that the resultant drug is modified enough molecularly to fall outside of existing regulations. Generally, designer drugs are hallucinogens or opiates. Because the quality controls in the labs that create these drugs are highly variable, the end product may be ten times weaker or ten times stronger than the drug being copied. One designer drug that has become popular with adolescents is known as Ecstasy, which is MDMA. The National Institute on Drug Abuse (Rusche, 1986) reported that MDMA was closely related to methamphetamine and methylene deoxyamphetamine sulfate (MDA). Methamphetamine causes brain degeneration resulting in symptoms similar to those seen in Parkinson's disease. MDA "induces prolonged serotonin neurochemical deficits by de-
326
CHAPTER 17
stroying serotonin nerve terminals'' (Leary, 1985). Such chemical reactions can adversely impact sleep, mood states, sensitivity to pain, and induce aggressiveness. Animal lab studies suggest that even after a single dose of Ecstasy, serotonin cells in the brain may be harmed (Turkington, 1986). No human studies have been reported, but the widespread popularity of designer drugs has caught the attention of the Drug Enforcement Administration(DEA). On July 1, 1985, theDEA madeMDMA a Schedule-One illicit drug like heroin, making it illegal to prescribe or possess it for other than authorized research.
Methodological Issues in Substance Abuse Research When research findings from the field of alcoholism and drug abuse research are evaluated, several methodological considerations and pitfalls must be taken into account. One of the most important issues deals with selection of subjects. Those who use alcohol and drugs are apt to differ from the general population in ways that may have neuropsychological implications. Some of these differences may predate use of any mind-altering substance, and therefore would have nothing to do with the adverse effects of any substance (Grant & Reed, 1985). Such differences could involve education, sociocultural and ethnic influences, nutrition, personality differences, and history of medically related neurological damage. Many research studies fail to match subjects along these lines, thereby compromising the interpretation of the findings. Other methodological flaws in drug and alcohol research literature include a reliance upon retrospective studies, use of inappropriate and unmatched control groups, lack of adequately sized experimental and control groups, reliance upon neurological or neuropsychological data-gathering methods that have not yet been proved or reliable, and use of measures of central tendency, which assume a uniformity of impairment that may not exist (Carlin, 1986). There are two critical design features that, when ignored, can invalidate findings. First, many researchers do not take into account how much time has elapsed since the last use of drugs. In some studies, subjects are tested just days after they last used a drug, yet the researcher states, or implies, that chronic drug use issues were addressed. A second issue deals with the extent and nature of self-reported drug
use. Individuals who use a drug tend to use more than one drug (Parsons & Adams, 1983) and many of the drugs purchased on the streets by adolescents often are impure and contain ingredients other than the presumed drug. Yet many researchers rely exclusively on the drug users' self-report of their drug use history; self-reports by drug users as to type and purity of drug, although convenient to researchers, makes interpretation of results extremely difficult and inaccurate and may be, in large part, responsible for the many seemingly contradictory findings in the area. The correct amount of drug used and duration of drug usage need to be accurately recorded. Research has shown that the type of drug, amount, and duration of use are all important elements when conducting drug effect research (Korman, 1977). The type of solvent involved, the medium or host agent for the drug, and presence of toxic impurities further confound research on the effects of street drugs. Another methodological issue deals with alternative explanations for significant research findings. At times, positive research results on motor function tests may be a result of peripheral nerve damage rather than CNS neuropathology. This is particularly apparent in research with solvents. Yet few researchers undertake nerve conduction studies, which could differentiate peripheral from CNS impairments. Sometimes, research findings may be significant but the effect is obscured because the researcher has combined all drug-using group information under one category. Researchers who take drug use histories find that most drug users use two or more drugs. Although they continue to identify the subject by a primary drug, they fail to identify the secondary drug and possible interactive effects.
Summary Although there has been evidence of a general decline in overall level of adolescent drug use during the last few years, the decline has been gradual and does not involve all drug classes. Use of marijuana, amphetamines, methaqualone, barbiturates, tranquilizers, and LSD has declined. Use of opiates, cocaine, PCP, and alcohol has remained unchanged, whereas that of inhalants has increased in the last few years. Johnston et al. (1985) reported that 62% of America's adolescents have used an illicit drug before leaving high school. Nearly half have used a drug other than marijuana. One in twenty is consum-
NEUROPSYCHOLOGICAL SEQUELAE OF SUBSTANCE ABUSE BY YOUTHS
ing alcohol on a daily basis, and almost 40% have had five or more drinks on one occasion during the preceding 2 weeks. This level of drug use remains disturbingly high and has implications for drug education and rehabilitation programs. If research can demonstrate that use of a drug adversely affects abstracting and conceptualization capabilities, then verbal, highly intellectualized therapeutic techniques may be useless or less appropriate than behavioral conditioning techniques. If perceptual motor, spatial, and sequencing skills are compromised, then a more didactic, verbal therapeutic approach may be appropriate. Knowledge of the specific effects of certain drugs will allow the therapeutic team to intercede in family and environmental issues in a knowledgeable manner. Definitive statements, offered by different investigators, about the effects of different drugs or combinations of drugs upon the CNS of children, adolescents, and young adults are difficult to summarize, as there are substantial differences in outcomes, methodology, and implications from study to study. A major problem in all drug research concerns causality. Did the drug cause the effect or did the existence of the effect cause use of the drug? Another major issue is the fact that nearly all drug users use more than one drug (especially alcohol). This makes it very difficult for an investigator to research the effects of a specific drug, as most users have not used one drug exclusively. In research with adults, multidrug studies have indicated that up to half of all drug users exhibit neuropsychological deficits during the first few weeks after discontinuation of drug taking. Although many users show signs of recovery 3 to 6 months later, residual impairment remains (Grant & Reed, 1985). Research with adolescents is less damaging in its implications than research with adults. Although it is recognized by most researchers that extent of drug use is an important factor, very few researchers have attempted to assess the reliability of drug users' selfreports. When severe use has been documented, even with young individuals (Grant et al., 1976, 1977), some CNS impairment is apparent. The younger brain appears to be more resilient to the adverse effects of mild to moderate drug use, even over a period of years. As the brain ages, it becomes increasingly sensitive to drugs, especiaUy depressants (Greenblatt, Sellers, & Shader, 1982). Although the majority of weU-controlled studies examining the neuropsychological implications of adolescent drug abuse do not report consistent tind-
327
ings of impaired functioning, long-term consequences are clearly evidenced for some and implicated for many others. As Parsons and Farr (1981) noted, clinicians can always recall the "burned-out" teen referral who was ''obviously'' brain-damaged and the literature is full of individual case studies of severely impaired individuals whose diminished functioning was linked to drug usage. When groups of abusing individuals are studied, such dramatic findings are not as apparent. Severely impaired individuals may represent the exception rather than the rule. This review of the studies of the chronic effects of substance abuse on children, adolescents, and young adults does not indicate consistent neuropsychological deficits associated with use of most substances. Those substances most consistent in demonstrating chronic effects with the young include alcohol, depressants, and multiple drug use. Most of the available research on the neuropsychological effects of substance abuse has been completed on adult populations. Research with children and adolescents involves controlling for a number of confounding factors that are not as important in adult research. Developmental changes involving brain functioning, emotional development, age of onset of drug use, and various environmental factors contribute to the relatively greater complexity involved in studying children and adolescents (Tramontana, 1983). Future investigations of the neuropsychological sequelae of child and adolescent substance abuse need to address the methodological issues noted earlier. The extent of drug use needs to be more systemically documented and verified by outside objective sources. Sample sizes need to be enlarged and careful matching is required in developing control groups. Longitudinal studies need to be developed that follow large groups of children before they start taking drugs and monitor them through high school and possibly beyond. Only in this way can at-risk factors be identified. Last, the extent of alcohol use accompanying most drug abuse needs to be fully evaluated, as it is known that alcohol is usually the first substance used by most individuals who have a history of drug abuse.
References Acord, L. D. (1972). Hallucinogen drugs and brain damage. Military Medicine, 137, 18-19. Acord, L. D., & Barker, D. D. (1973). Hallucinogenic drugs and
328
CHAYTER 17
cerebral deficits. Journal of Nervous and Mental Disease, 156, 281-283. Adams, E. H., & Durell, J. (1984). Cocaine: A growing public health problem. In J. Grabowski (Ed.), Cocaine: Pharmacology, effects, and treatment of abuse (pp. 9-14). Washington, DC: U.S. Government Printing Office. Addiction Research Foundation. (1981). Report of an ARFIWHO scientific meeting on adverse health and behavioral consequences of cannabis use. Toronto; Author. Akesson, H. 0. (1965). Nutmeg intoxication. Lancet, 1, 12711272. Allison, W. M., & Jerrom, D. W. (1984). Glue sniffing; A pilot study of the cognitive effects of long-term use. 1nternational Journal of the Addictions, 19, 453-458. American Academy of Pediatrics. (1984). Ethanol in liquid preparations intended for children. Pediatrics, 73, 405-407. American Psychiatric Association. (1980). Diagrwstic and Statistical Manual of Mental Disorders (3rd ed.). Washington, DC: Author. Bell, C. S. (1985, December I). Potent pot: Today's stronger marijuana poses new dangers for adolescents. Daily Breeze, p. B-ll. Berry, G. J., Heaton, R., & Kirby, M. (1978). Neuropsychological assessment of chronic inhalant abusers; A preliminary report. In C. W. Sharp & L. T. Carroll (Eds.), Voluntary inhalation of industrial solvents (pp. 111-136). Rockville, MD: National Institute on Drug Abuse. Beyette, B. (1983, September 18). Teenager's No. I drug: Alcohol. Los Angeles Times, Part VI, pp. 1, 22-25. Bigler, E. D. (1979). Neuropsychological evaluation of adolescent patients hospitalized with chronic inhalant abuse. Clinical Neuropsychology, 1, 8-12. Blacker, K. H., Jones, R. T., Stone, G. C., & Pfefferbaum, D. (1968). Chronic users of LSD: The "acid heads." American Journal of Psychiatry, 125, 341-351. Board on Mental Health and Behavioral Medicine of the Institute of Medicine, National Academy of Sciences. (1985). Research on the major disorders; Drug abuse and alcoholism. American Journal of Psychiatry, 142 (Supplement, July), 20-24. Bruhn, P., & Maage, N. (1975). Intellectual and neuropsychological functionsinyoungmenwithheavyandlong-termpattemsofdrug abuse. AmericanJournalofPsychiatry,132, 397-401. Cagliotti, C. (1980). Some considerations about the chewing of coca leaf in the Argentina Republic. In F. R. Jeri (Ed.), Cocaine, 1980 (pp. 137-144). Lima; Pacific Press. Callen, K. (1984). Toluene+ alcohol= danger.ADAMHANews, 10 (Supplement, September), 211-216. Campbell, A. M., Evans, M., Thompson, J. L., & Williams, M. J. (1971). Cerebral atrophy in young cannabis smokers. Lancet, 2, 1219-1224. Carlin, A. (1986). Neuropsychological consequences of drug abuse. In I. Grant & K. M. Adams (Eds.), Neuropsychological asssessment of neuropsychiatric disorders (pp. 478497). New York: Oxford University Press. Carlin, A. S., Grant, I., Adams, K. M., & Reed, R. (1979). Is phencyclidine (PCP) abuse associated with organic mental impairment? American Journal of Drug and Alcohol Abuse, 6, 273-281.
Carlin, A. S., & Trupin, E. W. (1977). The effects of long-term chronic marijuana use on neuropsychological functioning. International Journal of the Addictkms, 12. 617-624. Channer, K. S., & Stanley, S. (1983). Persistent visual hallucinations secondary to chronic solvent encephalopathy: Case report and review of the literature. Journal ofNeurology, Neurosurgery & Psychiatry, 46, 83-86. Chenoweth, M. B. (1977). Abuse of inhalation anesthetic drugs. In C. W. Sharp & M. L. Brehm (Eds.), Review of inhalants: Euphoria to dysfunction (pp. 102-111). Washington, DC: U.S. Government Printing Office. Clinger, 0. W., & Johnson, N. A. (1951). Purposeful inhalation of gasoline vapors. Psychiatry Quaner/y, 25, 557-567. Cohen, S. (1975). Inhalant abuse. Drug Abuse Newsletter, 4, San Diego. Cohen, S. (1985). Teenage drinking; The bottle babies. In The alcoholism problems: Selected issues (pp. 26-32). New York: Haworth Press. Cohen, S., & Edwards, A. E. (1969). LSD and organic brain impairment. Drug Dependence, 2, 1-4. Comstock, B. S. (1977). A review of psychological measures relevant to central nervous system toxicity, with specific reference to solvent inhalation. Clinical Toxicology, 11, 317324. Comstock, E. G., & Comstock, B.S. (1977). Medical evaluatioo of inhalant abusers. In C. W. Sharp & M. L. Brehm (Eds.), Review of inhalants: Euphoria to dysfunction (pp. 54-80). Washington, DC: U.S. Government Printing Office. Connell, P. H. (1966). Clinical manifestations and treatment of amphetamine type of dependence. Journal of the American Medical Association, 196, 718-723. Cook, R. F., Hostetter, R. S., & Ramsay, D. A. (1975). Patterns of illicit drug use in the Army. American Journal ofPsychiatry, 132, 1013-1017. Crane, J. A. (1984). The neurological and neuropsychological consequences of chronic, long-term phencyclidine (PCP) ingestion in adolescents. Dissertation Abstracts International, 45, 3-B 1010. (University Microfilms No. DA 841 3715) Culver, C. M., & King, F. W. (1974 ). Neuropsychological assessment of undergraduate marijuana and LSD users. Archives of General Psychiatry, 31, 707-711. Eckhardt, M. J., Parker, E. S., Noble, E. P., Feldman, D. J., & Gottschalk, L. A. (1978). Relationship between neuropsychological performance and alcohol consumption in alcoholics. Biological Psychiatry, 13, 551-565. Eckhardt, M. J., & Ryback, R. S. (1981). Neuropsychological concomitants of alcoholism. In M. Galanter (Ed.), Currents in alcholism (Vol. III, pp. 5-27). New York: Grone & Stratton. Elliott, R. W. (1985). Aging effects and the professional pilot. Hearings on age discrimination and the FAA age 60 rule. Select Committee on Aging, House of Representatives, 97th Congress (Committee Publication No. 99-533, pp. 375387). Washington, DC: U.S. Government Printing Office. Eliott, R. W. (1987, October). School consultation in clinical neuropsychology. Paper presented at the meeting of the National Academy of Neuropsychologists, Chicago. Elliott, R. W. (1988). The nature of adolescent drug use in an inpatient population. (Unpublished raw data.)
NEUROPSYCHOLOGICAL SEQUELAE OF SUBSTANCE ABUSE BY YOUTHS Estroff, T. (1985, October 22). Craving for cocaine by ex-users eased by prescription drug. The Times, Trenton, NJ, p. 83. Faucett, R. L., & Jensen, R. A. (1952). Addiction of the inhalation of gasoline fumes in a child. Journal of Pediatrics. 41. 364-368. Fauman, M. A., & Fauman, B. J. (1979). Violence associated with phencylidine abuse. American Journal of Psychiatry, 136, 1584-1586. Fields, F. R., & Fullerton, J. R. (1975). The influence of heroin addiction on neuropsychological functioning. Journal of Consulting and Clinical Psychology, 43, 114. Fischman, M. W., & Schuster, C. R. (1981). Acute tolerance to cocaine in humans. In L. S. Harris (Ed.), Problems of drug dependence, 1980: Proceedings of the 4Znd annual scientific meeting, the committee on problems of drug dependence, (pp. 241-242). Washington, DC: U.S. Government Printing
l29
neuropsychological study of polydrug users. Archives of General Psychiatry, 33, 973-978. Grant, I., & Reed, R. (1985). Neuropsychology of alcohol and drug abuse. In A. I. Alterman (Ed.), Substance abuse and psychopathology (pp. 289-341). New York: Plenum Press. Grant, 1., Rochford, J., Fleming, T., & Stunkard, A. (1973). A neuropsychological assessment of the effects of moderate marijuana use. Journal ofNervous and Mental Disease, 156, 278-280. Greenblatt, D. J., Sellers, E. M., & Shader, R.I. (1982). Drug disposition in old age. Medical Intelligence, 306, 10811088. Halstead, W. C. (1947). Brain and intelligence. Chicago: University of Chicago Press. Hamburg, B. A., Kraemer, H. C., & Jahnke, W. (1975). A hierarchy of drug use in adolescence: Behavioral and attitudinal. Office. American Journal of Psychiatry, 132, 1155-1164. Fischman, M. W., Schuster, C. R., Resnikov, L., Shick, J. F. E., Hannon, R., Butler, C. P., Day, C. L., Kahn, S. A., Quitoriano, Krasnegor, N. A., Fennell, W., & Freeman, D. X. (1976). L.A., Butler, A.M., & Meredith, L.A. (1985). Alcohol use Cardiovascular and subjective effects of intravenous cocaine and cognitive functioning in men and women college stuadministration in humans. Archives of General Psychiatry, dents. In M. Galante (Ed.), Recent developments in alco33, 983-989. holism (Vol. 3, pp. 241-252). New York: Plenum Press. Fisher, K. (1987, November). First hint of harm from mari- Hannon, R., Day, C. L., Butler, A.M., Larson, A. J., & Casey, juana use. American Psychological Association Monitor, M. (1983). Alcohol consumption and cognitive functioning p.21. in college students. Journal of Studies on Alcohol, 44, 283Freund, G. (1985). Neuropathology of alcohol abuse. In R. E. 298. Tarter & D. H. Van Thiel (Eds.), Alcohol and the brain: Heath, R. G., Fitzgarrell, A. T., Fontana, C. J., & Garey, R. E. Chronic effects (pp. 3-17). New York: Plenum Press. (1980). Cannabis sativa: Effect on brain function and ultraGawin, F. H., & Kleber, H. D. (1986). Abstinence symptomology structure in rhesus monkeys. Biological Psychiatry, 15, 657and psychiatric diagnosis in cocaine abusers: Clinical obser690. vations. Archives of General Psychiatry, 43, 107-113. Hill, S. Y., & Mikhad, M. A. (1979). Computerized transaxial Gawin, F. H., & Kleber, H. (1987). Issues in cocaine abuse: tomographic and neuropsychological evaluations in chronic Treatment research. In S. Fisher, A. Raskin, & E. H. alcoholics and heroin abusers. American Journal of PsychiaUhlenhuth (Eds.), Cocaine: Clinical and Biobehavioral Astry, 136, 598-602. pects. New York: Oxford University Press. Holbrook, J. M. (1983a). CNS stimulants. In G. Bennett, C. Gleaton, T. J., & Gowen, S. (1985). The adolescent drug epiVourakis, & D. S. Woolf (Eds.), Substance abuse: Phardemic and the chronic young adult patients: Is there a link? macologic, developmental, and clinical perspectives (pp. Psychiatry Letter, 3, 1985. 57-70). New York: Wiley. Gold, M.S., Verebey, K., & Dackis, C. S. (1985). Diagnosis of Holbrook, J. M. (1983b). Hallucinogens. In G. Bennett, C. drug abuse, drug intoxication and withdrawal states. Fair Vourakis, & D. S. Woolf (Eds.), Substance abuse: PharOaks Hospital Psychiatry Letter, 3, 23-24. macologic, developmental, and clinical perspectives (pp. Golden, C. J. (1985). Computer models and the human brain. 86-99). New York: Wiley. Computers in Human Behavior, 1, 35-48. Institute of Medicine, National Academy of Sciences. (1985). Goldman, M. S. (1983). Cognitive impairment in chronic alcoResearch on mental illness and addictive disorders: Progress holics: Some cause for optimism. American Psychologist, and prospects. A report of the Board of Mental Health and 38, 1045-1054. Behavioral Medicine, Institute of Medicine. American JourGrant, 1., Adams, K. M., Carlin, A. S., & Rennick, P.M. (1977). nal of Psychiatry, 142 (Supplement 7). Neuropsychological deficit in polydrug users: A preliminary Ivnik, R. J. (1986). Neuropsychological aspects of adolescent report of the findings of the collaborative neuropsychological chemical abuse. Journal ofClinical and Experimental Neurostudy of polydrug users. Drug and Alcohol Dependence, Z, psychology, 7, 605. 91-108. Jaffe, J. H. (I 975). Drug addicition and drug abuse. In G. GoodGrant, 1., Adams, K. M., Carlin, A. S., Rennick, P.M., Judd, L. man & A. Gilman (Eds.), The pharmacological basis of L., & Schooff, K. (1978). The collaborative neurotherapeutics (5th ed.). New York: Macmillan. psychological study of polydrug users. Archives of General Johnston, L. D., O'Malley, P.M., & Bachman, J. G. (1985). Use Psychiatry, 35, 1063-1074. of licit and illicit drugs by America's high school students: Grant,!., & Mohns, L. (1975). Chronic cerebral effects of alcohol 1975-1984. Rockville, MD: U.S. Government Printing and drug abuse. International Journal of the Addictions, 10, Office. 883-920. Jones, R. T. (1984). The pharmacology of cocaine. In J. GraGrant, 1., Mohns, L., Miller, M., & Reitan, R. M. (1976). A bowski (Ed.), Cocaine: Pharmacology, effects, and treat-
330
CHAPTER 17
ment of abuse (pp. 34-53). Washington, DC: U.S. Government Printing Office. Judd, L. L., & Grant, I. (1975). Brain dysfunction in chronic sedative users. Journal of Psychedelic Drugs, 7, 143-149. Kaplan, H. 1., Freedman, A.M., & Sadock, B. J. (1980). Comprehensive textbook ofpsychiatry III. Baltimore: Williams & Wilkins. Kissin, B. (1977). Alcoholism and drug dependence. In R. C. Simons & H. Pardes (Eds.), Understanding human behavior in health and illness. Baltimore: Williams & Wilkins. Kolansky, H., & Moore, W. T. (1971). Effects of marijuana on adolescents and young adults. Journal ofthe American Medical Association, 216, 486-492. Korman, M. (1977). Clinical evaluation of psychological factors. InC. W. Sharp&M. L. Brehm(Eds.),Reviewofinhalants: Euphoria to dysfunction. Rockville, MD: U.S. Government Printing Office. Korman, M., Matthews, R. W., & Lovitt, R. (1981). Neuropsychological effects of abuse of inhalants. Perceptual and Motor Skills, 53, 547-553. Korman, M., Trimboli, F., & Semler, I. (1978). A psychiatric emergency room study of inhalant use. In C. W. Sharp & L. T. Carroll (Eds.), Voluntary inhalation of industrial solvents (pp. 137-158). Rockville, MD: National Institute on Drug Abuse. Leary, W. E. (1985, September 2). 'Ecstasy' drug may destroy nerve fibers. Daily Breeze, Torrance, CA. Lewis, P. W., & Patterson, D. W. (1974). Acute and chronic effects of the voluntary inhalation of certain commercial volatile solvents by juveniles. Journal of Drug Issues, 4, 162175. Light, R. (1984). Neuropsychological, intellectual, and personality functioning in phencyclidine abusing adolescents and young adults. Unpublished doctoral dissertation, Indiana University, Bloomington. Louria, D. B. (1969). Medical complications of pleasure giving drugs. Archives of Internal Medicine, 123, 82. Lowry, T. P. (1979). Amyl nitrite: An old high comes back to visit. Behavioral Medicine, 6, 19. Maddux, J. F., Hoppe, S. K., & Costello, R. M. (1986). Psychoactive substance use among medical students. American Journal of Psychiatry, 143, 187-191. Maranto, G. (1985). Coke: The random killer. The Care Medic. (Available from CareUnit Hospital, 401 South Tustin Avenue, Orange, CA 92666) Martin, W. R., Sloan, J. W., Sapira, J. D., & Jasinski, D. R. (1976). Physiological, subjective, and behavioral effects of amphetamine, methamphetamine, ephedrine, phenometrazine, and methylphenidate in man. Clinical Pharmacological Therapy, 12, 245-258. Massengale, 0. N., Glaser, H. H., LeLievre, R. E., Dodds, 1. B., & Klock, M. E. (1963). Physical and psychologic factors in glue sniffing. New England Journal ofMedicine, 269, 13401344. Myers, J. A., & Earnest, M.P. (1984). Generalized seizures and cocaine abuse. Neurology, 34, 675-676. Nathan, P. E. (1983). Failures in prevention-Why we can't prevent the devastating effect of alcoholism and drug abuse. American Psychologist, 38, 459-467.
National Institute on Alcohol Abuse and Alcoholism (NIAAA). (1982). Physiological effects of alcohol. Alcohol topics in brief. Rockville, MD: Author. Negrete, J. C., & Murphy, H. B. M. (1967). Psychological deficit in chewers of coca leaf. Bulletin on Narcotics, /9, ll-18. Nunes, E. V., & Rosecran, J. S. (1987). Human neurobiology of cocaine. In J. S. Rosecran & H. I. Spitz (Eds.), Cocaine abuse: New directions in treatment and research. New York: Brunner/Mazel. Parsons, 0. A., & Adams, R. L. (1983). The neuropsychological examination of alcohol and drug abuse patients. In C. J. Golden & P. J. Vincenti (Eds.), Foundations of clinical neuropsychology (pp. 215-248). New York: Plenum Press. Parsons, 0. A., & Farr, S. P. (1981). The neuropsychology of alcohol and drug use. InS. B. Filskov & T. J. Boll (Eds.), Handbook of clinical neuropsychology (pp. 320-365). New York: Wiley. Parsons, 0. A., & Leber, W. R. (1982). Alcohol, cognitive dysfunction and brain damage. In Biomedical processes and consequences of alcohol use (pp. 213-253). Washington, DC: U.S. Government Printing Office. Platt, J. J., Scura, W. C., & Hannon, J. R. (1973). Problemsolving thinking of youthful incarcerated heroin addicts. Journal of Community Psychology, I. 278-281. Pope, K. S. (1987). Cocaine: Principals of assessment and treatment. The Independent Practitioner. Bulletin of the Division of Psychologists in Independent Practice, Division 42 of the American Psychological Association, 7(4), 11-16. Portnoff, L. A. (1982). Halstead-Reitan impairment in chronic alcoholics as a function of age of drinking onset. Clinica/ Neuropsychology, 4, ll5-119. Post, R. M., Weiss, S. R. B., Pert, A., & Uhde, T. W. (1987). Chronic cocaine administration: Sensitization and kindling effects. InS. Fisher, A. Raskin, & E. H. Uhlenhuth (Eds.), Cocaine: Clinical and biobehavioral aspects. New York: Oxford University Press. Reed, A., & Kane, A. W. (1972). Phencyclidine (PCP): Another illicit psychedelic drug. Journal of Psychedelic Drugs, 5. 812. Remington, G., & Hoffman, B. F. (1984). Gas sniffing as a form of substance abuse. Canadian Journal ofPsychiatry. 29, 3135. Rosecran, J. S., & Spitz, H. I. (1987). Cocaine reconceptualized: Historical overview. In J. S. Rosecran & HI. Spitz (Eds.), Cocaine abuse: New directions in treatment and research. New York: Brunner/Mazel. Ross, C. A. (1982). Gasoline sniffing and encephalopathy. Canadian Medical Association Journal. 127. 1195-1197. Rourke, B. P., Bakker, D. J., Fish, J. L., & Strang, J.D. (1983a). Elements of brain development. In Child neuropsychology, an introduction to theory. research, and clinical practice (pp. 9-35). New York: Guilford Press. Rourke, B. P., Bakker, D. J., Fish, J. L., & Strang, J.D. (1983b). Neuropsychological assessment: Aims, content, and style. In Child neuropsychology, an introduction to theory, research, and clinical practice (pp. 112-152). New York: Guilford Press. Rourke, B. P., Bakker, D. J., Fish, J. L., & Strang, J.D. (1983c). Plasticity of the developing brain and behavior. In Child neu-
NEUROPSYCHOLOGICAL SEQUELAE OF SUBSTANCE ABUSE BY YOUTHS
331
clinical neuropsychology (pp. 309-340). New York: Plenum ropsychology, an introduction to theory, research, and clinical practice (pp. 81-11 I). New York: Guilford Press. Press. Rusche, S. (1986, AprilS). The drug scene. Los Angeles Times, Trites, R. L., Sub, M., Offord, D., Neiman, G., & Preston, D. (1974, October). Neuropsychologic and psychosocial anteLos Angeles. Rylander, G. (1972). Psychoses and the punding and choreiform cedents and chronic effects of prolonged use of solvent and syndromes in addiction to central stimulant drugs. Psychimethamphetamine. Paper presented at the meeting of the Inatria, Neurologia, Neurochirugia, 75, 203-212. ternational Psychiatric Research Society, Ottawa. Schuckit, M. A. (1984). Drug and alcohol abuse: A clinical guide Tsushima, W. T., & Towne, W. S. (1977). Effects of paint sniffing on neuropsychological test performance. Journal of Abto diagnosis and treatment (2nd ed.). New York: Plenum normal Psychology, 86, 402-407. Press. Seshia, S. S., Rajani, K. R., Boeckx, R. L., & Chow, P. N. Turkington, C. (1986, March). Brain damage found with designer ( 1978). The neurological manifestations of chronic inhalation drugs. APA Monitor, p. 18. of leaded gasoline. Developmental Medicine and Child Neu- U. S. Journal ofDrug and Alcohol Dependence. (1985). One in six rology, 20, 323-334. seniors has tried cocaine. U.S. Journal of Drug and Alcohol Sharp, C. W., &Korman, M. (1980). Volatile substances. InJ. H. Dependence, 9, pp. 2, 4. Lowinson & P. Ruiz (Eds. ), Substance abuse: Clinical prob- Valpey, R., Sumi, S. M., Compass, M. K., & Goble, G. J. lems and perspectives (pp. 223-254). Baltimore: Williams & (1978). Acute and chronic progressive encephalopathy due to Wilkins. gasoline sniffing. Neurology, 28, 507-510. Siegel, R. K. (1984). Changing patterns of cocaine use: Longitudi- Van Dyke, C., & Byck, R. (1982). Cocaine. Scientific American, nal observations, consequences, and treatment. In J. Gra246, 128-141. bowski (Ed.), Cocaine: Pharmacology, effects, and treat- Ware, L.A. (1979). Neuropsychological functioning in users and ment of abuse (pp. 92-110). Washington, DC: U.S. nonusers of phencyclidine. Dissertation Abstracts InternaGovernment Printing Office. tional, 39, 5126-5127. (University Microfilms No. Smialek, J. E. (1985, March 7). Teen sniffing typewriter correc7908094) tion fluid. Daily Breeze, Torrance, CA. Washton, A. M., & Gold, M. S. ( 1987). Recent trends in cocaine Spitz, H. I., & Rosecran, J. S. (1987). Overview of cocaine abuse abuse as seen from the "800-Cocaine" hotline. In A. M. treatment. In J. S. Rosecran & H. I. Spitz (Eds.), Cocaine Washton & M.S. Gold (eds.), Cocaine: A clinician's handabuse: New directions in treatment and research. New York: book. New York: Guilford Press. Brunner/Mazel. Wilde, C. (1975). Aerosol metallic paints: Deliberate inhalation. Spitzer, R. L., Williams,J. B. W., &Skoda!, A. E. (1980). DSMA study of inhalation and/or ingestion of copper and zinc 111: The major achivements and an overview. American Jourparticles. International Journal of Addiction, 10, 127-134. nal of Psychiatry, 143, 463-468. Wilkinson, D. A. (1985). Neuroradiologic investiations of alcoTarter, R. E. (1980). Brain-damage in chronic alcoholics: Areholism. In R. E. Tarter & D. H. Van Thiel (Eds.), Alcohol view of the psychological evidence. In D. Richter (Ed.), and the brain (pp. 183-215). New York: Plenum Press. Addiction and brain damage (pp. 267-295). London: Croom Woolf, D. S. (1983a). Opioids. In G. Bennett, C. Vourakis, & D. Helm. Woolf (Eds.), Substance abuse: Pharmacologic, developTarter, R. E., & Edwards, K. L. (1985). Neuropsychology of mental, and clinical perspectives (pp. 70-80). New York: alcoholism. In R. E. Tarter & D. H. Van Thiel (Eds.), AlcoWiley. hol and the brain: Chroniceffects(pp. 217-242). New York: Woolf, D. S. (1983b). Polypharmacy and the addict. In G. BenPlenum Press. nett, C. Vourakis, & D. Woolf (Eds.), Substance abuse: Tarter, R. E., & Ryan, C. M. (1983). Neuropsychology of alcoPharmacologic, developmental, and clinical perspectives holism: Etiology, phenomenology, process, and outcome. In (pp. 110-120). New York: Wiley. M. Galanter (Ed.), Recent developments in alcoholism (Vol. Wright, M., & Hogan, T. P. (1972). Repeated LSD ingestion and I, pp. 449-469). New York: Plenum Press. performance on neuropsychological tests. Journal ofNervous Tong, T. G., Benowitz, N. L., Becker, C. E., Forni, P.J., & and Mental Disease, 154, 432-438. Boerner, U. (1975). Phencyclidine poisoning. Journal of the Young, L. A., Young, L. G., Klein, M. M., Klein, D. M., & American Medical Association, 234, 512-513. Beyer, D. (1977). Recreational drugs. New York: Collier Tramontana, M. G. (1983). Neuropsychological evaluation of Books. children and adolescents with psychopathological disorders. Zapata-Ortiz, V. (1970). The chewing of coca leaves in Peru. In C. J. Golden & P. J. Vincente (Eds.), Foundations of International Journal of Addictions, 5, 287-294.
III Techniques of Intervention
18 Neuropsychological Models of Learning Disabilities Contribution to Remediation CHE KAN LEONG
Introduction
(Downing & Leong, 1982; Rutter, 1978). The question of what constitutes significant discrepancy is a vexed one, although a lag of 24 months or more is usually taken as a yardstick to gauge "severe performance deficit'' for these children. Neuropsychologically, the definition by Mattis (1978) is a reasonable one:
The term model is defined by Kuhn (1970) as "the entire constellation of beliefs, values, techniques, and so on shared by the members of a given community. . . . It denotes one sort of element in that constellation, the concrete puzzle-solutions which . . . can replace explicit rules as a basis for Dyslexia is a diagnosis of atypical reading developthe solution of the remaining puzzles of normal sciment as compared to other children of similar age, ence" (p. 175). Furthermore, a paradigm or a model intelligence, instructional program, and socio-cultural "need not, and in fact never does, explain all the opportunity which, without intervention, is expected to persist and is due to a well-defined defect in any one of facts with which it can be confronted" (p. 18). Based sever.d specific higher cortical functions. (p. 54) on these premises, the present chapter examines several neuropsychological models of learning disThe differentiation between a given observable disorabilities and assesses their contribution to the re- der in dyslexics and in brain-damaged individuals is mediation of learning-disabled children. In particuan important one but one not easily made. What must lar, the emphasis is on neuropsychological models as be delineated are: "atypical reading development," problem-solving approaches that may explain specifimpairment in ''specific higher cortical functions,'' ic reading disabilities in children. These individuals and whether or not such an impairment is a deficiency or children with developmental dyslexia are those amenable to remediation and not just a defect per se. with severe difficulties with reading conceptualized Although the presence of a specific disorder in a as an internalized form of language. critical process implies the presence of dyslexia, we should also note the qualification by Mattis ( 1978) that ''the absence of this specific defect does not Focus on Developmental Dyslexia imply the absence of disordered reading" (p. 56). Psychometrically, developmental dyslexia refers to a heterogeneous group of reading disabilities characterized by reading/spelling attainment significantly below the level predicted on the basis of the child's chronological age or measured intelligence CHE KAN LEONG • Department for the Education of Exceptional Children, University of Saskatchewan, Saskatoon, Saskatchewan S7N OWO, Canada.
Dissociable Academic Skills There are several reasons for the narrower focus in this chapter on specific reading disabilities or developmental dyslexia rather than on the umbrella concept learning disabilities per se. One reason is that the connotation of learning disabilities is becoming too broad, too diffuse, and that the children so designated are quite heterogeneous. Although the 335
336
CHAYTER 18
tenn may be a convenient label for the delivery of habilitation services, it needs to be redefined for rigorous scientific studies (Leong, 1982, 1987). Another reason is that related disabilities in reading and spelling, which are often subsumed under the same omnibus term learning disabilities, are dissociable (Frith, 1978, 1979, 1980). Whereas there are obvious parallels between the processes of reading and spelling, proficiency or impairment in the one subskill does not necessarily imply corresponding proficiency or impairment in the other area. Reading and spelling are not reverse processes. In children the use of rules for reading and for spelling is distinguishable (Baron, Treiman, Wilf, & Kellman, 1980) and can be markedly different (Bradley & Bryant, 1985). In developmental dyslexia a speech-based deficiency may explain certain aspects of reading and spelling difficulties, but not the full range of results in a number of empirical studies on reading and spelling disabilities. The 11- to 13year-old dyslexics studied by Frith (1978, 1979, 1980), who were poor in producing the conventional spelling of words, were comparable to good spellers in the spelling of dictated non words, but were poor at reading written nonwords. Parallel findings come from the patient R.G. observed by Beauvois and Derouesne (1979, 1981). R.G. spelled each word by using phoneme-to-grapheme rules but his reading showed the opposite patterns. He read lexically and was very poor at reading nonlexically with grapheme-phoneme correspondence rules. These various findings of the differential use of linguistic structural codes and a general delay in accessing these codes suggest that ''spelling production might be viewed as being functionally distinct from reading" (Seymour & Porpodas, 1980, p. 471). What needs to be further explored is the nature and degree of these functional distinctiveness and relatedness of reading and spelling and their effects on children with specific reading disabilities.
Neuropsychological Models The focus of this chapter is thus on the neuropsychological analysis of severe reading disabilities both for research and for habilitation purposes. The analysis is attempted not only in terms of brain-behavior relationship but also more in tenns of component processes of reading and the breakdown of these processes. In this regard, we are reminded by Spreen ( 1976) that neuropsychological models of learning disorders must encompass both "normal" and "ab-
nonnal" learners and must add to our understanding of learning processes and remediation practices. Similar views are evident in the appraisal of current (to the late 1970s) advances in theory and research of dyslexia (Benton & Pearl, 1978). The following sections discuss indirect and direct approaches to ''training the brain"; a model' based on Luria's (1966a,b, 1973, 1977) simultaneous-successive syntheses and planning; and an emphasis on language access as an effective approach to remediation.
"Training the Brain": Indirect Approaches Indirect approaches to ·'training the brain'' generally involve some form of sensory-motor integration as a putative integration of the CNS at the subcortical level. In her reeducation program, Ayres ( 1972) suggested that learning is a fonn of movement in response to a stimulus and that body-image development relates to the assimilation of tactile-kinesthetic stimuli. In other words, psychoeducational development of children with learning disorders must be seen as a successive integration of perception of stimulus, movement, conscious body-image and culminating in learning. The assumption that bodyimage concepts are the resultant of the different stimuli received in the brain via the sensory systems will need to be further researched. It is interesting to note that Ayres's reeducation principle with its putative link to the CNS still has considerable following in a number of European countries. This is shown by the habilitative work of the Swiss team of Affolter and Stricker (1980), and the Thea Bugnet or the Bon Depart (good start) approach followed in Poland (see Duane & Leong, 1985). The Bon Depart approach emphasizes visualaural-motor activities involving the visual element (graphic symbols), the aural element (songs), and the motor element (harmonizing rhythmic movements with graphic symbols and songs). The simultaneous integration of visual-aural-motor activities aims at making more efficient the visual, aural, and tactilekinesthetic analyzers. In essence, the Bon Depart approach is one of psychomotor rehabilitation combined with stimulation of psychomotor development. The actual habilitation involves: (l) motoric exercises emphasizing relaxation and agility, (2) motorauditory exercises carried out to the tune of songs sung by the children, and (3) motor-visual-auditory exercises that form the bulk of the activities. It is claimed in these European countries that this good-start approach appears to be effective not only with preschool children, but also with children
REMEDIATION
with dyslexia and dysgraphia. The eagerness with which the Ayres approach and its extension is followed probably illustrates the goodness of fit between the perceptual-motor match and what Cruickshank (1975, p. 272) termed the "psychoeducational match.'' He also cautioned that these are instances where clinical practice coupled with the logic of the concept should provide a base for research and for linking theory with practice.
"Training the Brain": Direct Approaches
"Habilitation Movement" Direct "brain training" includes the wellplanned and systematic stimulation through the visual, auditory, and kinesthetic senses. One such approach is that of the "rehabilitation movement" of Diller and associates (Weinberg, Piasetski, Diller, & Gordon, 1982). These researchers use visual halffield training to help their patients to analyze and utilize more efficiently visual-spatial information. Their techniques, however, are geared primarily for adults and are focused on the visual modality.
Tomatis's Audio-Psycho-Phonology Habilitation and "Dysmetric Dyslexia" In the auditory domain, Tomatis (1978) advocated audio-psycho-phonology training as a form of habilitation for dyslexic children. The general principle is that the voice could produce only what it hears and that many dyslexics have difficulties listening, particularly to high-frequency sounds. Within this psychoacoustic framework, Tomatis suggested that dyslexia can occur in children whose hearing acuity is intact but whose ability to listen and to communicate is impaired. He conceived of the inner ear as a kind of ''charging organ" that energizes sensory stimulation into neural energy to keep the brain alert. This is analogous to a dynamo charging and recharging a central battery. Furthermore, the inner ear in concert with the vestibular system coordinates and integrates various sensory and motor functions to achieve equilibrium. As well, the inner ear is a major .. vector" or force in establishing laterality (a leading right ear and .. dominance" of the left hemisphere). If the ''charge'' of the brain is inefficient, the central equilibrium is disturbed, and laterality is not established. Consequently, speech and language problems may result. To keep this "charge" in a dynamic state, to maintain the sensory functions in balance,
337
and to develop right ear dominance, some •'structuring dynamics'' of listening and speech skills must be established. This is done through the Tomatis listening training program (LTP), which incorporates an "Electronic Ear" as the hardware or the main programming component. There are several stages in the habilitation, all emphasizing auditory stimulation and auditory vocal exercises. First, the frequency of the acoustic signals are filtered out through a gating mechanism in the Electronic or Phonic Ear. The augmentation of high frequencies and attenuation of low frequencies and the exaggeration of low frequencies and reduction of high frequencies are meant to produce a positive therapeutic effect. Then listening to prefiltered sounds or to one's own filtered voice via almost instantaneous feedback through the Phonic Ear is said to train auditory discrimination. These two stages and the next one of the gradual attenuation of signal intensity to the left ear all aim at restoring normal listening development. It should be noted that the Tomatis audiopsycho-phonology is a supplementary form of habilitation. It is intended as a complement to a strong academic program and is not meant to be used on its own. This LTP seems to have attracted some following and there have been claims made about its efficacy in helping dyslexics. Although some shadowy evidence of short-term gains in achievement have been reported for the dyslexics so treated, many, if not most, of the studies are neither constructively critical of the theoretical framework nor overly sophisticated in their research designs. The one exception is the careful, 2-year study of the efficacy of the Tomatis LTP by Kershner, Cummings, and Clarke (1986). These researchers evaluated the pretest, posttest, and follow-up performance of three groups of a total of 42 children aged 8 to 12 with .. developmeiJtal learning disabilities" from a school recognized for its remedial work. The 16 learning-disabled children in the experimental group were given 100 hours of school-based Tomatis LTP, over the school year. The 16 children in the placebo group were given a non-Tomatis audiovisual feedback tutorial program to control for placebo and Hawthorne effects; and 10 learning-disabled children formed the "no treatment" group. The pre- and posttest design was used with blind testing on a number of achievement tests, cognitive, neuropsychological, psycholinguistic measures, social-emotional scores, and dichotic listening. Multivariate analyses of variance (MANOVA) with age-adjusted scores were computed. The experimental and placebo groups were further subtyped according to both a neuro-
338
CHAPTER 18
psychological dual processing model and an achievement pattern model in order to minimize the effects of heterogeneity and to study the possible interactions of selective treatment effects for subtypes of dyslexics. In general, the results suggest that the Tomatis LTP did not have an effect on children's audiophono development as tested on the key Tomatis Listening Test. In particular, the claim of treatment effect on hemispheric information processing was not substantiated in dichotic listening; nor were there treatment effects on specific subtypes of learning disorders. Furthermore, any apparent gains made by the experimental students were found to be the "combined result of individual care, motivation and effective remedial instruction, and not the LTP" and that ''time taken from school hours for such activities [LTP] is without empirical justification'' (Kershner et al., 1986, p. 43). This very careful empirical study serves to alert us as to the validity of the theoretical conception of the LTP and the questionable value of the resource-withdrawal Electronic Ear training for these children. Just as Tomatis attributed much of normal listening and language development to the inner ear in connection with the vestibular system, a hypothesized relation between dyslexia or what is termed "dysmetric dyslexia" (DD) and cerebellar-vestibular (c-v) functions has been proposed (Levinson, 1980). Levinson's thesis is that dyslexia results from a c-v-related nystagmus or oculomotor dyscoordination. This dysfunction disturbs the temporal-spatial sequence of visual symbols, leads to abnormal electronystagmograms (ENOs), and contributes to the slower single-target blurring speed as found in one sample of 300 dysmetric dyslexics and 25 control subjects. As a "solution to the riddle dyslexia," Levinson prescribes "c-v harmonizing agents" including such anti-motion sickness drugs as methylphenidate (Ritalin). He also suggests oculomotor training including the use of a three-dimensional reader to present written materials in temporal sequence at a fixed locus so as to improve reading functioning. Although the author and others with "solutions" to dyslexia have no doubt treated a number of dyslexics, their basic tenets need to be substantiated and their claims must be empirically verified with rigorous experiments involving doubleblind controls and refined experimental designs (see Masland & Usprich, 1981, for review).
Bakker's "Balance Model" Subsumed under the category of direct ''training of the brain'' but quite different from the forego-
ing audio-psycho-phonology program is Bakker's (1973, 1979, 1984; Bakker, Smink, & Reitsma, 1973; Rourke, Bakker, Fisk, & Strang, 1983) "balance model" to explain reading disabilities in children. Bakker hypothesized that the hemispherereading relationship is dependent on the phase of the learning-to-read process. Proficient early reading, which demands more of the perceptual processes, may rely more on right cerebral laterality. Laterstage fluent reading, which requires more of linguistic processes, may depend more on left hemisphere laterality. Following the above hypothesis, Bakker postulated two main types of dyslexias. His basic tenet is that some children learn to read using mainly left hemisphere strategies (as shown in right ear advantage (REA) of verbal dichotic listening tasks) at the wrong time or that they tend to overlook the spatialperceptual features (mainly right hemisphere activities) of the text. These children, who rely unduly on linguistic-semantic aspects of reading probably because of a functional overdevelopment of the left hemisphere, are termed L-type (for linguistic) dyslexics. Other children may rely overly on right hemisphere strategies as reflected in left ear advantage in verbal dichotic stimulation. These children, who may have a functionally overdeveloped right hemisphere, are termed P-type (for perceptual) dyslexics. The L-type dyslexics tend to make more substantive reading errors such as omissions and additions; the Ptype dyslexics tend to make more time-consuming errors such as repetitions and fragmentations. The reading error patterns by ear dominance in relation to the L-type and P-type dyslexias are shown in Figure 1, which partially summarizes a collaborative study between the present author and Bakker with 38 Dutch severely disabled readers (Bakker, 1979). The rightear-dominant dyslexics (L-type) make more omission and substitution errors (class 2 and class 8); the left-ear-dominant dyslexics (P-type) tend to make more time-consuming errors (class 12). In addition to their different patterns of reading errors, the L-type and P-type dyslexics also show hemisphere-specific electrophysiological parameters (Bakker, Licht, Kok, & Bouma, 1980; Bakker & Licht, 1986). Analyses of event-related potentials (ERPs) to words and figures reveal that a significant proportion of the variance can be accounted for by the right temporal components in reading acquisition by young children. There are also interactions with age of slow positive wave activities at the temporal sites and there are decreases at the parietal sites. These interactions suggest that left hemisphere strategies for some reading components are established after I year and before 2 years of reading instruction. Those
REMEDIATION 24 23 22
o----o LED
i.
9I
I
---RED
'o /r-: ,a_
21
,' t
20
I
19
"Qz 18 ..."" 16 17
... ...... 1111:
)(
15
~ 14
"'01111:
13
1111:
12
...... 1101
... ......0"" ...z0 ... ...v 1111: A.
11
10
9 8
F.
7
'•'•
I I
I
6
I I
I I I I I I
5 4 3
I
2
I
I
I
I I I I
1 I 1 I I I II
\I
! 2
3
4
5 6 7 8 9 10 CLASSES OF ERRORS
11
12
FIGURE I. Type of reading error by ear dominance (LED, left ear dominance; RED, right ear dominance). (After Bakker, 1979, Figure 4.)
children who are inefficient in using, or changing to, the appropriate reading strategies in concert with the demand of the reading tasks are likely to experience reading difficulties. The electrophysiological measures suggest that L- and P-type dyslexics can be distinguished as showing different cerebral activities during reading. The ERP results also point to the possibility of selective visual and auditory stimulation of the right hemisphere in L-type dyslexics, and of the left hemisphere in P-type dyslexics to improve their reading performance. Such hemisphere-specific stimulation studies were carried out with the two types of dyslexics delineated according to the ear advantage of verbal dichotic listening tasks and reading errors and the ef-
339
fects were carefully evaluated (Bakker, Moerland, & Goekoop-Hoefkens, 1981; Bakker & Vinke, 1985). In general, the L-type dyslexics were given training with words flashed to the left visual hemifield; the P-type dyslexics received words flashed to the right visual hemifield; and the controls for both types received central field training or no training at all. Significant interactions were found as to the amplitude and latencies of some peaks in the parietal and temporal areas and the asymmetric effects were most noticeable for the parietal area (Bakker & Licht, 1986; Bakker & Vinke, 1985). These results indicate the "rightening" of activities following right hemisphere stimulation with L-dyslexics and ''leftening'' of activities following left hemisphere stimulation in P-dyslexics. There were some between hemispheric changes of activities, which tended to correlate with improvements in pre- and posttest reading performance. Nontreated L-dyslexics seemed to persist in left hemisphere strategies as suggested by the leftening of early parietal positivity over time; nontreated P-dyslexics apparently continued to rely on right hemisphere strategies as inferred from the rightening of parietal positivity over time. These neurophysiological results seem to provide some evidence for the differentiation of L- and P-dyslexics. The findings also suggest that direct right hemisphere stimulation of L-dyslexics would lead to greater reading accuracy and reading efficiency. However, direct left hemisphere stimulation in Pdyslexics yielded some equivocal results in that reading accuracy and efficiency measures were relatively unaffected (Bakker & Licht, 1986; Bakker & Vinke, 1985). Bakker and his colleagues further point out that whereas their more recent studies confrrm their results of an earlier pilot study (Bakker et al., 1981), there are some variations owing to different training procedures and stimulus materials. The pilot study showed the temporal sites to be affected by hemisphere-specific stimulation, whereas the more current studies revealed a predominantly parietal effect. The Bakker concept linking cerebral processing and early reading strategies is intriguing. It emphasizes the different and conjoint contributions of the two hemispheres, especially at the learning-to-read stage. His neuropsychological and psychophysiological findings may be explained according to the novelty model of Goldberg and Costa ( 1981). These researchers suggest that the left hemisphere is more capable of unimod&l processing, storage of compact codes of information; whereas the right hemisphere is more suited for intermodal integration and for novel stimuli. Thus, right hemisphere functions are more critical at the acquisition of new descriptive systems whereas left hemisphere functions are better at util-
340
CHAPTER 18
izing routine codes. The Bakker postulates could be accommodated within this framework in that children learning to read likely view the reading tasks as both novel and perceptually complex and it is only later that they shift to a linguistically coded system. The Goldberg-Costa theoretical position goes beyond relating functional cerebral asymmetry and information processing. It explains what information the subjects process and how they process it. The Bakker balance model, which is supported by his ERP studies, was tested by Donders and van der Vlugt ( 1984) in their analysis of eye movement (EM) patterns of two age groups ("younger" and "older") of good and poor readers during slide-presented reading and arithmetic activities. The EM patterns (and by inference, the reading strategies) of young poor readers were found to be different from the EM patterns of all other children. The authors suggested that the balance model needs further validation as it was probably only the group of older good readers that might be said to employ a left hemisphere strategy in reading. The use of EM patterns to infer laterality patterns is of some interest. There is evidence that dyslexics show more frequent fixations and regressions than their controls (Pavlidis, 1985, 1986). Pavlidis found significant differences in almost all EM pat., terns in dyslexics and their nondyslexic "retarded" controls matched for reading and chronological ages, but not between the "retarded" and normal readers. Furthermore, there was little overlap between the dyslexics and all other readers in the number of regressions. There is, however, variant evidence that when older disabled readers and younger normal subjects were matched in word recognition and their EMs were measured in the same textual material, there was no significant difference between the groups in the EM patterns (Olson, Kliegl, & Davidson, 1983). The erratic EM patterns found in dyslexics are more likely the result, rather than the cause, of poor reading. Pavlidis has advanced the possibility that both EMs and reading performance may be subserved by some neurological substrates. This claim will need to be further tested. What is clear is that EM variables such as frequency of fixations, fixation duration, frequency of regressions, average saccade length all reflect orthographic and phonological processes during reading and that disabled readers vary in their use of small or large units of orthographic or phonological codes.
Callosal Connection Parenthetically, it may be pointed out that Orton ( 1925) observed that some dyslexic children could
read better words shown upside down in a mirror fashion. He interpreted the disorderly EMs or EMs not in the preferred direction in relation to the lack of cerebral dominance. A more plausible hypothesis was advanced by Geschwind and Galaburda (1985) to explain these observations. They suggested that in presenting words in a mirror fashion, the next words to be read would lie to the left of the fixation point and hence would be projected to the right hemisphere. If the left hemisphere of dyslexic children functions less efficiently than the right, then the visual processing of the language would be more adequate on the right side of the brain. This line of interpretation seems to bolster Bakker's balance model of early reading. There is another intriguing possibility. Geschwind and Galaburda ( 1985) further hypothesized that when words are presented in the traditional left-toright fashion for these dyslexic children, the words either fail to reach, or are slow in doing so, the more efficient right hemisphere because the necessary callosal connections are often poorly formed in many dyslexics. The abnormal corpus callosum leading to inadequate or inefficient interhemispheric connections is also alluded to by Hiscock and Kinsbourne (1987) as a possible "speculation" in connection with anomalous hemispheric specialization. A more affirmative position of the potential role of interhemispheric collaboration in reading acquisition was suggested by Gladstone and Best (1985). These authors analyzed the ''contemporary mainstream views'' of incomplete cerebral dominance derivable from Orton (1925, 1937) and of left hemisphere dysfunction discussed by a number of authors in Knights and Bakker (1976) in relation to early reading and reading disabilities. Gladstone and Best found these mainstream views inadequate to account for different sets of data in reading disabilities. Drawing on research on callosal contributions to attention, bimanual coordination, handedness, and gender differences, they presented an alternative explanation that deficient callosal functions could be a cause for some types of dyslexia. Their corollary is that reading acquisition depends on the interhemispheric communication with decreasing degrees of collaboration needed as the skill develops. These notions of the more important role of the right hemisphere in younger children learning to read and in reading failures seem to be consonant with Bakker's hypothesis of a shift from right to left hemisphere strategies. They also seem to relate to the Goldberg and Costa ( 1981) concept of the right hemisphere being more
suited for new descriptive systems and the left hemisphere as more efficient for categorical, routine codes.
REMEDIATION
As a summary of these sections on the efficacy of "training the brain" in remediating dyslexic children, Bakker's (1984) view that psychological stimulation may have some effects on the physiology of the human brain has some theoretical and empirical bases. He marshalls evidence from a number of neuroanatomical, neurophysiological, neurochemical, and neuropsychological studies to show the modifiability of the animal brain through environmental manipulation, sensory stimulation, and systematic training. Knowledge gained in animal studies contributes to our understanding of habilitation in humans. There are neuropsychological programs based on the adaptational capacities of brain-impaired children to .fit their therapeutic needs (see Rourke et al .. 1983). With these pathological cases, we need to know the location and age of onset of their brain damage, the nature of the lesion, its severity and rate of development. All these parameters are important determinants of responses to treatment. With braindifferent children, the volume Education and the Brain (Chall & Mirsky, 1978) published by the National Society for the Study of Education attests to the interest in linking the neurosciences, psychology, and education. Various chapters in that book by neuroscientists indicate the role of cerebral specialization in cognition and the importance of timely and appropriate environmental situations in the growth and development of the brain. This is a reminder of Bakker's (1984) concept of the brain as a "dependent variable.''
The 11Working Brain" The indirect and direct "training the brain" approaches all seek to elucidate the left-right axis responsible for higher cortical functions. These discussions of the neuropsychological basis of learning disorders must go beyond the brain-behavior relationship to tease out the more complex levels of analysis. Kinsbourne and Hiscock (1983) state that each higher mental function "comprises a complex set of component skills and even the component skills may engage various brain structures. Nevertheless, if all critical components of a function are represented within the same hemisphere, it may be said that the function is lateralized" (p. 166). Similar statements are found in Luria and Artem'eva (1978): "Neuropsychology is concerned with the analysis of cerebral mechanisms of mental processes, and its primary subject matter is the case in which circumscribed local brain lesions cause specific changes in mental processes" (p. 283). Luria emphasizes mental activities as a complex functional system reflecting the
341
conjoint activities of a whole group of cerebral zone and the importance of the principle of double dissociation between cerebral loci and cognitive processes. This principle is well documented in clinical studies by Luria (l966a,b, 1970, 1973, 1977). For example, dysfunction in the parieto-occipital region of the left hemisphere disturbs the spatial organization of perception and movement shown in such activities as map reading, directional sense, and grammatical relationship of the kind: "father's brother" as distinct from ''brother's father''; but such a pathological locus produces no disturbance of processes involving sequential activities as required in speech fluency, or the playing of musical melodies. Although advocating "syndrome analysis" of mental disorders, Luria also draws attention to the need for rigorous techniques such as factor analysis and Bayesian analysis to ensu~ consistency of neuropsychological studies. The more functional analytical perspective of Luria provides the theoretical framework for the present author's study of cognitive processing in dyslexic and less skilled readers. The position taken is that specific reading disabilities are best understood as the interaction of neuropsychology, cognition, language, and education (Leong, 1987). The recent articles in Developmental Review by Crowder (1984), Mann (1984), Morrison (1984), and Wolford and Fowler (1984) have rekindled the debate on the nature of the disabilities in dyslexics. Are these deficiencies specific to reading or are they more general? The consensus from that "debate" is that language plays a large role in reading and its difficulties and that poor readers also show deficiencies in some cognitive functions other than reading (Crowder, 1984). The interplay between neuropsychology, cognitive psychology, language, and education is also emphasized as important in unraveling reading disabilities by Doehring, Trites, Patel, and Fiedorowicz (1981). In this section on cognitive processing of disabled readers based on Luria's (1966a,b, 1970, 1973, 1977) "working brain" model, the basic neuropsychological concept is sketched below. This is followed by a summary of the application of the simultaneous-successive syntheses and planning model.
Luria's Three Basic "Blocks" In Luria's writing the "higher cortical functions'' of language, reading, and spelling are the results of different components of a complex system working in concert. A function or more exactly a functional system is explained as: The product of complex reflex activity comprising: uniting excited and inhibited areas of the nervous sys-
342
CHAPTER 18 tern into a working mosaic, analysing and integrating stimuli reaching the organism, forming a system of temporary connections, and thereby ensuring the equilibrium of the organism with its environment. (Luria, 1966a, p. 23)
Futhennore, a function is stated as "a complex and plastic system perfonning a particular adaptive task and composed of a highly differentiated group of interchangeable elements" (Luria, 1966a, p. 26). Thus, according to Luria, psychological functions cannot be adequately explained by some "morphological schemes'' of cerebral localization, but are themselves complex, organized activities of a whole system. The dynamic grouping of the connections may vary even though the task itself remains unchanged. The apparatus for the whole system is the upper associative layers of the cerebral cortex, the cortical connections arising in the secondary associative nuclei of the thalamus, and the overlapping zones uniting different boundaries of cortical analyzers. Clinical observations of patients with gunshot wounds and brain tumors show that a disturbance of a particular complex function does not arise in association with a narrowly circumscribed lesion of one part of the cortex. A lesion of the same area of the brain at different stages of ontogenesis may lead to quite different consequences. Moreover, the cortical intercentral relationship does not remain the same at different stages of development of a function. In discussing complex functional systems, Luria (l966a,b, 1970, 1973) distinguishes "three principal functional units" of the brain. The first basic block regulates the energy tone of the cortex. This block includes the upper and lower parts of the brain stem, the reticular fonnations, and the hippocampus. Damage to the first block, namely, the loss of the selectivity of cortical actions and of nonnal discrimination of stimuli, will bring about marked changes in behavior such as disturbances in wakefulness, instability of memory traces. The second basic block plays an important part in the analysis, coding, and storage of information. Located in the posterior part of the brain, the second block consists of a hierarchical organization of these cortical areas: a primary zone that sorts and records secondary infonnation, a secondary zone that organizes the infonnation further and codes it, and a tertiary zone where data from different sources overlap and are combined to lay the groundwork of behavior. A lesion in a primary zone results in a sensory defect; a lesion in the secondary zone interferes with the analysis of the sensory stimuli; a lesion in the tertiary zone can cause complex disturbances as visual disorientation in space. The third basic block comprising the frontal
lobe is involved in the fonnation of intentions and programs for behavior. This block controls and regulates human behavior. In Luria's words, these blocks can be approximated to ''a unit for regulating tone or waking, a unit for obtaining, processing and storing information arriving from the outside world and a unit for program-
ming, regulating and verifying mental activity''
(Luria, 1973, p. 43, author's italics). It should, however, be emphasized that any consCious activity is always a complex functional system and takes place through the combined, concerted working of all three brain units. It should also be borne in mind that the working zones underpinning complex cognitive functions are modified by the individual's experience in his or her acquisition of language, reading, and writing. The progressive emergence of these higher cortical functions may also be accompanied by neural organization and reorganization and carried out by different constellations of cortical zones. Although the ''material basis'' for the higher cortical functions is the brain as a whole, ''the brain is a highly differ-
entiated system whose parts are responsible for different aspects of the unified whole'' (Luria, 1966a,
p. 35, author's italics).
Simultaneous-Successive Syntheses and Planning Luria's view of the working brain provides the neuropsychological model of cognitive processing in both nonnal and atypical children. Of particular interest are his two basic fonns of integrative activity: simultaneous (primarily spatial, group) and successive (primarily temporally organized series) syntheses at the perceptual, memory, and intellectual levels (Luria, 1966a,b, 1973). For example, simultaneous synthesis at the perceptual level may be shown in: copying of geometric figures, drawing of a map, performance on Koh's Block; and at the memory level in: arithmetic difficulties and ''grammatical structure involving arrangement of elements into one simultaneous scheme." In successive synthesis, examples are counting sequences of tapping, digit span, and serial learning such as drawing ''0 + + - '' while keeping the correct order. In Luria's terms, simultaneous-successive syntheses can be identified with the functions of specific parts of the cerebral cortex, although conscious activity as a complex functional system is the result of concerted working of different brain units. The occipital-parietal areas have evolved to be mainly responsible for simultaneous synthesis. The anterior regions, particularly the frontal-temporal area, are more important for
REMEDIATION
successive synthesis. Both of these regions are concerned with coding and storage of information. The Luria model also assumes that the two modes of information processing are available to the individual, depending on his or her habitual mode of activities and the demand of the task.
Coding Processes. The Luria model provides the neuropsychological basis for the process-based Kaufman Assessment Battery for Children (K-ABC) (Kaufman & Kaufman, 1983) with its simultaneous processing, sequential processing, and achievement scales. As the rationale of the K-ABC, its diagnostic utility and remedial applications have been analyzed and critiqued elsewhere (Miller & Reynolds, 1984) and are further elaborated on in this volume, the work is not discussed here. Instead, attention is turned to the Luria model as first operationalized by Das, Kirby, and Jarman (1975, 1979) in their research program with different groups of children: retarded and nonretarded, high and low achievers, children from different SES and ethnic backgrounds. Das et al. emphasized the relative independence of simultaneous and successive modes of cognitive activities and the importance of processes, rather than products of learning. Leong (Das, Leong, & Williams, 1978; Leong, 1976, 1980, 1982, 1984; Leong & Sheh, 1982) predicated his studies of patterns of impairment of dyslexics and nondyslexics, below-average readers and their controls, and unselected samples of readers on the Luria-Das paradigm. He modified, refined the basic Das battery, added some tasks, and used more rigorous methods of analysis such as different factor analysis models (typically principal component analysis, alpha factor and promax oblique factor analyses) and factor matching to achieve "method-independent" and more reliable results. The findings are remarkably consistent in that the basic and relatively independent (orthogonal) constructs or factors of simultaneous and successive syntheses emerged with the same basic battery of tasks and the resultant factor scores were found to be discriminating between the different reading groups. The general results are in line with the Das et al. (1975, 1979) studies. In Leong's research program, the typical marker tasks that loaded on the simultaneous dimension were: Raven's Coloured Progressive Matrices (1947), Figure Copying (Ilg & Ames, 1964), Memory-for-Designs (Graham & Kendall, 1946, 1960). For the successive dimension, the marker tasks were: Auditory Serial Recall (recall of four-word series of phonetically similar words), ITPA Auditory Memory or digit span (Kirk, McCarthy, & Kirk, 1968), and
343
sentence repetition devised by Leong. A cross-modal task or Auditory-Visual Coding was found to straddle both the simultaneous and successive dimensions. The results from the different factor analyses and factor matching show that the simultaneous-successive dimensions are differentiated. The same relative orthogonal factor structure is shown in the cognitive patterns of dyslexics and nondyslexics (Leong, 1976) and with skilled and less skilled readers (Leong, 1980). From the basic differential patterns of components or factors, it is logical to derive factor scores as more parsimonious constructs and to use these scores in multiple regressions or other multivariate analyses for prediction and explanation of reading proficiency. The Luria-Das paradigm and the Leong refinements for studying the cognitive patterns of disabled readers have some attraction. For one thing, the paradigm attempts to steer clear of the traditional memory-reasoning distinction with its implied lower-to-higher hierarchy. For another, the neuropsychological framework focuses attention on the workings of cognitive processes and the possibility of training some precursors to reading. There are, however, limitations to the two-dimensional approach to cognitive processing. Tasks requiring higher mental activities are too complex to be amenable to some dichotomous analysis even though the paradigm is a viable one. Furthermore, the simultaneous-successive model would need to be further tested with more varied simultaneous-successive tasks of different levels of complexities, with larger sample sizes and more sophisticated multivariate analyses. Within the stricture of the existing conceptual and methodological frameworks, the simultaneoussuccessive syntheses approach seems to explain the available data well. However, certain aspects in interpreting both the postulate and the findings of the Luria-Das paradigm would need to be further explored. First, even though the simultaneous and successive factors are relatively independent, this psychometric finding must also be psychologically meaningful. Rather than conceiving of two distinct constructs, we would do well to think of the "simultaneous-successive matrix" or "simultaneous-successive syntheses" to emphasize the integrative nature of the processes. Careful reading or rereading of Luria shows his very cautious approach to the conventional concept of simultaneous and successive syntheses which ''are not sufficiently accurate'' and his implicit assumption of a complex matrix embodying these syntheses. By simultaneous synthesis is meant the "synthesis of successive (arriving one after the other) elements into simultaneous spatial
344
CHAPTER 18
schemes" and by successive synthesis is meant "the and locations. Both the Euclidean and projective syssynthesis of separate elements into successive se- tems provide the framework for ''multiplicative clasries" (Luria, 1966b, p. 74). The former process is sification." It is also likely that the poorer perforcharacterized by "surveyability," the latter by mance of the "retarded" and "backward" readers in "order" and "kinetic melody." These charac- Leong's studies (1976, 1980) suggests their use of a teristics are close to the "logical multiplications" in "Gestalt algorithm" rather than an "analytic althe Piagetian concept of logical development (Piaget gorithm" in solving matrices (see Hunt, 1974). This & lnhelder, 1956). More important for our purpose, qualitative interpretation of individual performance Luria wrote of "integrative activity" and "analyt- will enhance the inference of processes from factor ico-synthetical'' activities of the cortex to underline loadings and factor patterns of groups. It will also the interrelatedness of simultaneous-successive syn- highlight the different procedures a child can draw on theses (Luria, 1966b, p. 83). The following quota- in solving a cognitive task and the importance of tion explains the holistic and localized "complex analyzing learning strategies, whether simultaneous functional systems'' of the brain working in an inte- or successive. grative manner: Planning Component. In their relatively recent Syntheses of elements into simultaneous groups and Das and his associates (Das, 1980, formulation, successive series may fonn a part of any analyticoHeemsbergen, 1983) have extended & Das 1984; synthetic activity and are two fonns of working of the their work in coding processes (simultaneous and same brain sbUcture, two necessary aspects of each successive syntheses) to investigate Luria's third neuro-dynamic process. The nervous processes constituting the worlc of any analyzer always take place in block functions. Planning or planful behavior genertime and are always dependent on certain spatially orally refers to judgment, decision-making, and evalganized sbUctures .... The correlation of dynamics uation of activities. The characteristics of planning with SbUcture . . . WOUld be impossible Without taking include the encoding and recoding of information, into account both these aspects of nervous activity. the selection of appropriate ''programs,'' the evaluaOLuria, 1966b,p. 79) tion and execution of action. These and other tasks These complementary-different functions are in ac- are suggested by Das as efficient for assessing planning: visual search, trail making, planned composicord with current views of cerebral mechanisms. Over and above the quantitative aspect of per- tion, syllogistic reasoning, and the game of strategies formance, the Luria postulate carries with it the im- Mastermind. Of particular interest among these tasks plicit emphasis of the qualitative aspect. Attention is are visual search, which requires the subject to search directed to how a function suffers rather than what visually different geometric shapes, letters, and functions are deficient or inefficient. An example of number sets on an overhead transparency and to the qualitative performance from the Luria dictum is make rapid decisions matching the stimulus to the the solution of analogy items in Raven's Coloured target; and trail making, which involves the ability to Progressive Matrices, one of the marker tasks for see relationships and to shift from one stimulus to the simultaneous synthesis. Within the context of the other. Leong, Cheng, and Das (1985) recently tested early growth oflogic in the child, Inhelder and Piaget (1964, pp. 151-154) speak of the Raven's series of the validity of planful behavior as a construct within tests as a good example of ''multiplicative classifica- the Luria-Das paradigm with elementary school tion,'' which is mastered at about 8 years of age. By children. Factor analyses according to different multiplicative classification is meant ''classing each factor models showed that the simultaneous-sucelement simultaneously in terms of [the] two additive cessive syntheses and planning dimensions are well orders.'' Thus, from the kernel 2 X 2 matrix involv- differentiated (see Figure 2) and that the simuling two sets of elements such as red and blue squares taneous-successive-planning dimensions contribute and circles, the combination can be extended to in- to reading proficiency as shown in multiple regresclude shape, color, line, position, and the like. Suc- sion analyses using the factor scores derived from the cessful performance on the Raven's requires both domains. There is further statistical evidence that the "reasoning" and "spatial ability" as suggested by Luria-Das model fits the data well (Leong, Cheng, Kirby and Das (1978). A child's poor performance on the Raven's Lundberg, Olofsson, & Mulcahy, submitted). In two might be attributed to his/her failure to establish Eu- studies, one involving 129 eleven-year-old and the clidean relationships and to coordinate projective other involving 164 ten-year-old unselected readers, viewpoints as perspectives, sections, projections, Leong et al. tested the goodness of fit of the simul-
REMEDIATION
SIMULTANEOUS COMPONENT
.9 j
.8
.7
VI
.6
ct
-
<< ... o ....
.I
-.1 -.2 -.3
2
3
4
5
.9
.8
.7 .6 ct
<< ... o ....
345
taneous-successive syntheses, planning, and speed . Some of the results are summarized in Table 1. x2 is a goodness (or badness) of fit measure in that large x2 values indicate a bad fit and small values a good fit. The coefficient of determination is a generalized measure of the reliability of the whole measurement model. A high value indicates a good fit between the conjoint measurable variables (x variables) and the conjoint unobserved or unobservable variables (the simultaneous, successive, planning, and speed domains). There is thus further psychometric evidence for the Luria-Das construct of simultaneous-successive syntheses and planning and the variant construct to include the speed domain .
.2 .1
-.1
-.2 -.3
Instructional Implications 2
34
5
6789 SUCC ESSIVE COMPONENT
11'1
ct
o z ti c
<< ... g 2
3
4
5
6
7
VARIABLES Cognitive Processing of Combined Group (N :: 129)
FIGURE 2. Simultaneous-successive syntheses and planning components. Variables: I , Raven's Coloured Progressive Matrices; 2, Figure copying; 3, Memory-for-Designs (negative signs reversed); 4, Auditory serial recall; 5, ITPA auditory memory; 6, Sentence repetition; 7, Visual search time; 8, Trail-making time (A); 9, Trail-making time (B).
taneous-successive and planning model and the elaboration of that model to include the speed component. They used a maximum likelihood approach with the LISREL (linear structural relationship) model (Joreskog & Sorbom, 1984) to test their hypotheses. In essence, the maximum likelihood structural estimation using LISREL is both exploratory · and confirmatory with specification of parameters and measurement errors. The approach emphasizes both theory and measurement and allows for hypothesis testing of how well alternative models fit the data. The Leong et al. study found that the threedomain simultaneous-successive syntheses and planning fit the data well, but an even better fit is provided by the four-domain model with simul-
On the instructional aspect using the simultaneous-successive paradigm there is evidence of its relationship to reading and language disabilities (Cummins & Das, 1977, 1978). There is further evidence that these processes can be trained (Das et al., 1979). Kirby and Das ( 1977) showed that both simultaneous and successive processing are necessary, but neither by itself is sufficient, for high performance in reading and intelligence tests. For the simultaneous dimension, Lawson and Kirby (1981) found some success in training a "Gestalt strategy" and an "analytic strategy" in processing Raven's tasks and the different strategies in turn related to the performance of their subjects. Leasak, Hunt, and Randhawa (1982) carried out an intervention program specifically designed to improve simultaneous processing and found significant improvement in both reading and arithmetic in their Grade 4 experimental children. For the successive dimension, Krywaniuk and Das ( 1976) found that remedial programs designed to augment successive processing were successful and had an effect on decoding subskills. More recently, Kirby and Robinson (1987) investigated the effects of simultaneous-successive syntheses on a number of reading tasks in 105 elevenyear-old children. Their results further uphold the viability of the simultaneous-successive syntheses as a framework for remediation with simultaneous processing being implicated in lexical access and semantic explication and successive synthesis more responsible for decoding and syntactic analysis. In particular, the results suggest that reading-disabled children are likely to use an inappropriate strategy at a particular stage of their reading. They probably employ simultaneous processing in the early reading stage for both word recognition and syntactic analy-
346
CHAYfER 18
TABLE 1. LISREL VI Analysis of 10 Variables for Simultaneous Synthesis, Successive Synthesis, Planning, and Speed (N = 164) Maximum likelihood (ML) of components (standard errors in parentheses) Manifest variables
Simultaneous
Raven's Coloured Matrices (x 1)
0.582 (0.092) 0. 711 (0.092) -0.555 (0.090)
Figure copying (x 2) Memory-for-designs• (x3)
Auditory serial recall (x4 ) ITPA auditory memory (x 5 ) Sentence repetition (x6)
Trail making time A (x7 ) Trail making time B (x 8 ) Visual search time I (x9 ) Visual search time 2 (x 10)
Successive
Planning
Speed
0.427 (0.089) 0. 784 (0.096) 0.644 (0.091) 0.644 (0.119) 0.850 (0.141) 0.865 (0.229) 0.355 (0.118)
•Error scores for memory-for-designs (x 3 ):
= 32.69, p = 0.290 Goodness of fit index (GFI) = 0.963 Adjusted goodness of fit index (AGFI) = 0.929 Total coefficient of determination p = 0. 991 Chi square, X229
sis tasks for which successive processing could be more appropriate. It is this overuse or inappropriate use of simultaneous processing and the inadequate or inefficient use of successive processing that may explain some of their difficulties in reading.
Frontal Lobe Involvement in Planning Of the several components in the Luria-Das formulation, the planning domain merits particular attention. This is an area that presents challenges in conceptualization and methodology. Several tasks hold promise in identifying planful behavior consonant with Luria's (l966a,b) neuropsychological concept of programming and regulating different activities. The Block Design of the WISC-R (Wechsler, 1974) is a possibility, except that more spatial processing resources seem to be called for by the task. The Wisconsin Card Sorting Test (WCST) as a test of perseveration and concept shift from welllearned responses to new stimuli is another possibility. Milner (1963) found that patients with dorsolateral lesions were most impaired on the WCST and suggested that it is not a defect of abstract thinking that leads to the impairment but a generalized inability to shift behavior when needed to do so. Robinson, Heaton, Lehman, and Stilson (1980) showed that the WCST discriminates significantly between the neurologically normal and patients with localized frontal lobe brain lesions. An even more promising task for planning, which combines the in-
formation-processing approach with the neurological underpinning, is the Tower of Hanoi task (Simon, 1976), the processing of which requires the attainment of the goal by decomposition into subgoals with minimum individual moves. A modified form of the Tower of Hanoi task has been shown to relate to Luria's views on frontal lobe functions (Shallice, 1982). These are just further suggestions to refine the tasks with which to infer Luria's third block functions of executive intentions, maintaining and directing activities. In addition to the need to refine behavioral tasks, there are neuropsychological implications for the role of the functional capacity of frontal systems in dyslexics. Denckla and her colleagues (Denckla, 1983; Duffy, Denckla, Bartels, & Sandini, 1980; Duffy, Denckla, Bartels, Sandini & Kiessling, 1980) investigated this issue in 13 ten-year-old ''pure dyslexic'' boys compared with ll controls and speculated on frontal lobe contributions to children's growing capacity to learn language. The dyslexic and control children all underwent sophisticated EEG recording including brain electrical activity mapping (BEAM) (see also Duffy & McAnulty, 1985). BEAM is an elegant topographic imaging technique to record continuously changing spectral and spatialtemporal information with computerized analyses and display of latency, amplitude, and frequency band information accompanying each image. The location and degree of abnormality for each individual are analyzed with a technique known as significance probability mapping (SPM). The goal of SPM is to
REMEDIATION
delineate regional topographic differences of brain activity using various statistical analyses of data. The ''pure dyslexic'' boys showed the alpha-type EEG rhythmicity in the medial frontal regions but no alpha blocking in the posterior regions, while they were engaged in language-related cognitive activities including reading. Alpha rhythmicity is characteristic of the resting or idling state of the brain, whereas alpha blocking suggests active engagement in cognitive tasks. In contrast to the dyslexics, the normal controls displayed alpha blocking or marked EEG rhythmic changes while performing similar cognitive activities. What is particularly significant is that for the dyslexics, much of the resting alpha rhythm organization as shown in BEAM readings was found in the frontal convexity of the brain. The other cerebral regions distinctive in the dyslexics were the left frontal area (near Broca's area), the left midtemporal area (the auditory associative area), all from EEG data and the left posterolateral quadrant (Wernicke's area) from the evoked potential data. These between-group electrophysiological differences including the involvement of the bilateral medial frontal lobes found by Denckla, Duffy, and their colleagues with the BEAM technique are corroborated by the cortical blood flow studies of Lassen, Ingvar, and Skinhfllj ( 1978) in normal subjects during speech and reading. The BEAM and related findings led Denckla (1983) to make the "bold suggestion" that "a pure dyslexic, with no deficit in spoken language, appears to have some subset of general verbal learning deficits associated with left convexity frontal lobe" (p. 41, author's italics). There is the further suggestion that ''pure dyslexia'' may represent the dysfunction of the entire complex bihemispheric system normally involved in language-related activities. This hypothesis does not seem to be incompatible with the abnormalities found primarily in the left posterior hemisphere with dyslexics in that these abnormalities may interfere with functions over large cerebral areas. Duffy and his colleagues (Duffy, Denckla, Bartels, & Sandini, 1980; Duffy, Denckla, Bartels, Sandini, & Kiessling, 1980; Duffy & McAnulty, 1985) emphasize that the BEAM findings with the small number of "pure dyslexics" may not be universally applicable to all dyslexics. Nevertheless, their studies further alert us to the need to formulate research hypotheses or further explore the role of Luria's third block and to devise more refined paper-and-pencil tasks as clinical instruments. Parenthetically, Luria considered the left frontal lobe to be more involved in verbal processes and the right frontal lobe more concerned
347
with emotional behavior, even though he did not attach great importance to the left-right distinction in the frontal lobe areas. The condition of developmental dyslexia may represent a disability involving a complex and widely distributed cerebral system with breakdown in accessing the writing system. The research challenge should be the integration of information processing and clinical neuropsychology to understand better language and reading disorder (Ellis, Miller, & Sin, 1983). Marin, Glenn, and Walker (1982) stressed "the relationship between the molecular nature of the brain and the processes or operations which transform the neural activity into the integrated abilities that characterize human life'' (p. 253). Rourke (1982) used the metaphor "dynamic neuropsychology'' to emphasize the development of the brain and the development of the individual as well as the interactions of both aspects to explain learning and learning difficulties. Rourke's developmental neuropsychological model proposed these axes of importance in human neuropsychology: (I) progression from "lower" to "higher" cerebral centers, (2) progression from posterior to anterior regions of the cerebrum, and (3) the left-right hemisphere reciprocal relationship. The significance of his proposal is the interaction of these developmental dimensions as a framework or a neurodevelopmental model of central processing in children, especially those with learning disorders.
Interaction of Cognitive Processing, Language, and Reading Thus far, we have offered some evidence from the neuropsychological, statistical, and instructional perspectives for the viability of the simultaneoussuccessive syntheses and planning model and its variant paradigm including speed as correlates for reading proficiency. Both the three-domain and the four-domain cognitive processing models are not incompatible with the Luria-Das paradigm. Whichever model that is used for research and habilitation purposes must go beyond the statistical findings and must take into account the integrative nature of the mechanisms involved in information processing. In particular, theories should be developed to delineate the nature of the interaction among components of reading, language, and neuropsychological variables and tasks should be designed to assess these interrelated subskills. The aim is to provide a profile of reading abilities or disabilities in such a multifaceted context (Doehring, 1984; Rourke, 1982).
348
CHAPTER 18
A good example of this interactive approach is the detailed investigation carried out by Doehring et al. ( 1981) in 88 clinic-referred disabled readers ranging from 8 to 14 years of age. From the Q-factor analyses of the 39 reading-related and language-related tasks, Doehring et al. found different patterns or subgroups of dyslexic readers. The largest group (Type 0 for oral reading disability) tended to be much poorer in oral reading of words, syllables, and letters than in their visual and auditory-visual matching performance. Type A (associative reading 9isability) tended to be very poor on auditory-visual matching tasks. The third type, Type S (sequencing reading disability), was much poorer in reading syllables and words than in reading single letters. Thus, the careful study of Doehring et al. further elucidates Vellutino' s ( 1979) verbal deficit conceptualization of dyslexia. Another powerful study, the Florida longitudinal project by Satz and his colleagues (see Fletcher, Satz, & Morris, 1984; Satz & Morris, 1981, for representative works), has again identified as dominant the different clusters of ''global language difficulties," "selective language difficulties,'' but with fairly ''normal'' performance on nonlanguage perceptual and neuropsychological tasks. Thus, from the clinical and cognitive neuropsychological literature, there is the ongoing quest for a better understanding of both the nosology and pathogenesis of specific reading disabilities. The nosology or etiology delineates the ''disease'' entities, whereas pathogenesis considers the particular symptoms and symptom complexes that lead to the "disease" entities. From the habilitation point of view, pathogenetic rather than etiological considerations are more likely to provide some direct principles and useable approaches. Conceptually, we need to develop a theory-based, empirically verifiable framework involving actual reading and not just reading-related tasks and to assess the effects of systematic remediation. These proposals are made forcefully by Doehring et al. (1981). Ever their own best critic, they acknowledge the need for such a theoretical framework and emphasize the importance of "a working theory that makes possible a unified assessment and explanation of the entire pattern of deficit'' (p. 241). Methodologically, these issues should be addressed: wider sampling of reading subskills and neuropsychological tasks, broader coverage of age ranges and ethnic variables of the children studied, strengthening of criterion measures, judicious choice of statistical methods, and further validation of results (Fletcher et al., 1984; Satz & Morris, 1981). It is only through the careful working and reworking of hypotheses and the twin processes of confirmation
and refutation of hypotheses that we gain a better understanding of reading disabilities.
The Role of Language Access Although simultaneous-successive syntheses and planning together with the speed variant have been shown to be antecedent factors of reading proficiency in grade school children, the further research question is the role of language. Specifically, the monitoring, control, and repair of, and general reflection on language over and above language usage may provide the intermediate link between cognitive processing and school achievement including reading. This metacognitive knowledge variously referred to as "reflective abilities," "general development of consciousness," or "metalinguistic abilities" is termed "language awareness" by Downing and Leong (1982, Chapter 6) and more recently "language access" by Leong (1987). In essence, this concept is implicit in Vygotsky (193411962, 1978) and Luria (1979, 1981). In their seminal writings, Vygotsky and Luria discussed the notions of the internalization of higher psychological functions; the regulatory functions of speech in human cognition; the mediational role of language in thinking; and consciousness as affective and cognitive phenomena. These key Vygotsky and Luria concepts of the roles of speech and language in higher cognitive functions have been applied to pedagogy, especially to the teaching and learning of reading (Elkonin, 1963, 1973). Much of the current work (Downing & Valtin, 1984; Sinclair, Jarvella, & Levelt, 1978; Tunmer, Pratt, & Herriman, 1984) on bringing to the awareness level of the learner the various linguistic activities, and relating these activities to reading as a deliberately acquired, language-based skill, owes to the integrative works of Vygotsky and Luria. Related to the internalization of language for higher cortical functions is the need for the reader to access his or her tacit knowledge of grammar (see Leong, 1987). This grammatical knowledge is intuitive knowledge and hen.ce accessible whereas strategies such as speech perception and parsing mechanisms are inaccessible empirically, if not linguistically. Language access is thus a mental activity that interacts with other cognitive activities, modifies them, and is in turn modified by them. Studies of language access generally refer to situations where children perform actions, the results of which are apparent to them even though their awareness may lag behind their success in their ac-
REMEDIATION
tions. Much of the Soviet literature is in this direction. Luria (1946, p. 61) in his glass window theory pointed out that even though a child actively uses grammar, "he is still not able to make the word and verbal relations an object of his consciousness." He goes on to say: "In this period [of the child's development] a word may be used but not noticed by the child, and it frequently seems like a glass window through which the child looks at the surrounding world without making the word itself an object of this consciousness and without suspecting that it has its own existence, its own structural features.'' It should be noted that the Vygotsky and Luria concept of consciousness is paradigm-specific to the Marxist philosophy and is focused on how the individual reflects on, or is influenced by, the socio-cultural environment in which he or she functions (Luria, 1981). Nevertheless, the basic .concepts as embodied in Elkonin's (1963, 1973) experimental work have a direct bearing on fostering children's awareness of language. Elkonin has developed practical approaches to help early readers and disabled readers in understanding the concepts of language and in reasoning about the writing system. One of Elkonin's key notions is his emphasis on the need to distinguish between "perceived phonemes" and their embodiment of the natural flow of speech and the further need for phonematic analysis, which enables children to gain an understanding of the language system. To attain this phonematic analysis, Elkonin suggests a series of stages. They range from establishing the concept of the task, to mastering the concept with objects, with overt oral speech, then transferring the operation to the mental level and finally internalizing of these series of activities. Even allowing for the difference in the transparency-opaqueness continuum between the Cyrillic and other alphabetic writing systems, the Soviet approach to language access as represented by Luria and the role of phonematic analysis in early reading as emphasized by Elkonin are directly applicable to English. These basic concepts are explicated in the language-oriented approach to reading and reading disabilities by Liberman and her colleagues at the Haskins Laboratories (Liberman, 1983; Liberman, Shankweiler, Liberman, Fowler, & Fischer, 1977; Shankweiler, Liberman, Mark, Fowler, & Fischer, 1979). Working within the language access framework and drawing on the research programs of the Haskins group, the present writer has attempted to tease out the contributions of cognitive processing, language access, and reading. An early study in relation to Luria's phonematic analysis and the Haskins group's phonological awareness involved the perfor-
349
mance of Grade 1, 2, and 3 children in segmenting syllables into phonemes, words into syllables, phonological representation and understanding of sentence structure (Leong & Haines, 1978). The Leong and Haines results as shown in analyses of variance and canonical correlations confirmed the Liberman et al. ( 1977) findings of the role of phonemic segmentation in early reading. These results are taken to mean that, for young children, their awareness of words and sentences is at the subsidiary level. Their acquisition of verbal skills is facilitated if their understanding is brought to the focal level. The reflection on, and manipulation of, words and sentences can be taught in the form of word games and play activities, and will go some way toward helping the child in reading acquisition (Bradley & Bryant, 1985; Leong, 1987). Although there is a considerable volume ofliterature on the importance of phonological awareness in early reading, questions are often raised as to the precise nature of this contribution (Wagner & Torgesen, 1987). Are the different facets of phonological awareness prerequisites to, or facilitators, consequences, or correlates of reading? In a critical review of the development of individual differences in reading ability, Stanovich (1986) hypothesized that "if there is a specific cause of reading disability at all, it resides in the area of phonological awareness. Slow development in this area delays early code-breaking progress and initiates the cascade of interacting achievement failures and motivational problems" (p. 393). In a series of elegant studies, Lundberg and his reading research group in Umea (see Lundberg, 1985, for representative work) have provided strong evidence for the causal effect (in the descriptive sense) of phonological awareness and early reading. The Umea researchers show that phonemic awareness (rhyme recognition, segmental analysis, phoneme elision and addition, phoneme synthesis) can be developed in prereaders outside the context of formal reading instruction (Olofsson & Lundberg, 1983). Furthermore, phonemic training contributes to the long-term development of accurate concepts of reading in preschool children and young readers (Olofsson & Lundberg, 1985). The Umea group further investigated the interactions of language (both comprehension and production), cognitive development as independent latent variables and "metaphonological'' abilities and reading/ spelling as dependent latent variables in a follow-up of 46 dyslexics compared with 44 controls all drawn from a longitudinal study involving over 700 children (Torneus, 1984). Subserving the latent phonological
350
CHAPTER 18
awareness domain were the measurable tasks of segmental analysis, sound blending, position analysis, and segment deletion. Using the maximum likelihood approach with different LISREL analyses performed separately for reading and spelling, Torneus found strong support that reading/spelling is directly dependent on metaphonological abilities and only indirectly on other cognitive and language skills. The present writer's two recent studies (Leong, 1984; Leong et al., submitted) are in the direction of teasing out the reciprocal and interactive effects of cognitive processing, language awareness, and reading proficiency in older children (Grades 4 to 6). It is postulated that for older children, language awareness at the syntactic-semantic level involving such tasks as the disambiguation of ambiguities and the solution of incongruities and anomalies should contribute more of the variance to reading than cognitive processing (Leong, 1986; Leong & Carrier, 1986). The multiple regression analyses of the 1984 Leong study provide some answers to the relative contributions to the variance in reading by the language awareness aggregate, and simultaneous-successive syntheses residualized for language awareness. These analyses show clearly the predominant contribution to reading by language awareness, whether considered as separate entities or as an aggregate. In a path analysis, the direct effect of simultaneous processing on reading as given by the path coefficient is 0.050, whereas the total of indirect effect as mediated by language awareness is 0.254.
The direct effect of successive processing on reading as given by the path coefficient is 0.115, whereas the total of the indirect effect is 0.216. In contrast, the language awareness tasks have a much greater direct effect on reading as shown by the path coefficient of 0.680. Thus, within the framework of the postulated "causal" model, there is some evidence from the path analysis to support the assertion that the effect of simultaneous-successive syntheses on reading is mediated by language awareness and the latter in turn has a greater direct effect on the deliberately taught and school-based task of reading. In addition to the findings of the greater direct effects of language awareness than cognitive processing on reading, Leong and Carrier (1986) have further shown in two experiments that less skilled readers and dyslexics experienced difficulties with the processing of sentential ambiguities as compared with the dyslexics' chronological-age and readingage controls. Encouraged by these findings, Leong et a/. (submitted) further tested the extension of the model linking cognitive processing, language access, and academic achievement with a maximum likelihood approach. The generalized path model involving simultaneous-successive syntheses and planning, language awareness, and achievement including reading is shown in Figure 3. As the four-domain model involving speed as an additional domain provides a better fit to the data, these results rather than the three-domain ones are summarized below. The exogenous simultaneous-
Simultaneous Processing
Awareness Planning
School
1 - - - . - J (Hierarchy of
monitoring, control, repair, reflection processes)
Successive Processing
FIGURE 3. Postulated path diagram showing direct and indirect effects of cognitive processing and language awareness on achievement
including reading.
REMEDIATION
successive syntheses, planning, and speed and the endogenous language access domain have both direct and indirect effects on reading, but with different forces. The direct effect of simultaneous processing on academic performance as given by the path coefficient is 0.532, the indirect effect via language access is 0.196, and the total effect is 0.728. The direct effect of successive synthesis on academic performance is 0.199 as compared with the indirect effect via language access of 0. 707 for a total effect of 0.906. These higher direct and indirect effects contrast with those provided by the planning and speed domains. Planning related to academic performance has a direct effect of 0.192, an indirect effect via language access of -0.066 and a total effect of 0.127. Speed (latency scores) related to academic performance has a direct effect of -0.071, an indirect effect of -0.051, and a total effect of -0.123. Furthermore, the interaction between the endogenous domain is more from language access to academic performance as shown by the beta coefficient of 0.478 and not the other way around (beta coefficient of 0.369). These results both support and elucidate Leong's (1984) path analysis showing the greater direct effect of language awareness than simultaneous or successive synthesis on reading. The maximum likelihood results suggest the primacy of the coding processes over planning and speed, probably because of the need to refine the planning tasks. In structural terms, simultaneous processing has a greater direct effect on reading and related academic skills than successive processing, the effect of which is mostly mediated via language access. The Leong et al. (submitted) findings also suggest the need to encourage readers to reflect on language at the syntactic-semantic level. Teachers for their part should teach language and reading not so much as a rigid, formalized system, but as an activity to be enjoyed, to be manipulated. As an example, the anomalous sentence "*Stones read books" can be explained in terms of selection restriction rules in that the activity of reading should take an animate subject. The same sentence is acceptable with the insertion of the negative such as "Stones do not read books," or in the broader context such as ''My three-year-old brother tells me that stones read books," given the world knowledge that young children are inclined to fantasies. It is with this kind of "disembedded modes of thinking" that teachers can help children make human sense of incongruities and come to accept them (Donaldson, 1978, p. 82). This approach in helping children to use their knowledge more ex-
351
plicitly and systematically, to manage a set of options when unknowns are encountered, to develop reflective language skills, promotes literacy development and is effective in habilitation with dyslexics.
Summary This chapter focuses on some neuropsychological models that may contribute to the remediation of children with developmental dyslexia. Indirect and direct approaches to ''training the brain'' are discussed. Luria's "working brain" model of simultaneous-successive syntheses and planning seems to offer promise for analyzing cognitive processes. Such a model together with the inclusion of the speed domain fits well the data from unselected and poor readers as tested with different models offactor analysis and the maximum likelihood approach using LISREL. Mediating cognitive processing and reading proficiency is the latent domain of language access involving the disambiguation of ambiguities, the resolution of anomalies and related reflective tasks on the syntactic-semantic aspects of language. It is suggested that the training of coding processes and planning strategies and the development of reflective language skills should help children with reading disabilities. AcKNOWLEDGMENT
I thank Dirk J. Bakker of the Free University of Amsterdam for his insightful comments. Any shortcomings are necessarily my own.
References Affolter, F., & Stricker, E. (Eds.). (1980). Perceptual processes as prerequisites for complex human behaviour: A theoretical model and its application to therapy. Bern: Huber. Ayres, A. J. (1972). Sensory integration and learning disorders. Los Angeles: Western Psychological Services. Bakker, D. J. (1973). Hemispheric specialization and stages in the learning to read process. Bulletin of the Orton Society, 23, 15-27. Bakker, D. J. (1979). Hemispheric differences and reading strategies: Two dyslexias? Bulletin of the Onon Society, 29, 84100. Bakker, D. J. (1984). The brain as a dependent variable. Journal of Clinical Neuropsychology, 6, 1-16. Bakker, D. J., & Licht, R. (1986). Learning to read: Changing horses in mid-stream. In G. T. Pavlidis&D. E. Fisher(Eds.),
352
CHAPTER 18
Dyslexia: Its neuropsychology and treatment (pp. 87-95). New York: Wiley. Bakker, D. I., Licht, R., Kok, A., & Bouma, A. (1980). Cortical responses to word reading by right- and left-eared normal and reading-disturbed children. Journal of Clinical Neuropsychology, 2, 1-12. Bakker, D. I., Moerland, R., & Goekoop-Hoefkens, M. (1981). Effects of hemisphere-specific stimulation on the reading performance of dyslexic boys: A pilot study. Journal ofClinical Neuropsychology, 3, 155-159. Bakker, D. I., Smink, T., & Reitsma, P. (1973). Ear dominance and reading ability. Cortex, 9, 301-312. Bakker, D. I., & Vinke, I. (1985). Effects of hemisphere-specific stimulation on brain activity and reading in dyslexics. Journal of Clinical and Experimental Neuropsychology, 7, 503525. Baron, J., Treiman, R., Wilf, J. F., & Kellman, P. (1980). Spelling and reading by rules. In U. Frith (Ed.), Cognitive processes in spelling (pp. 159-194). New York: Academic Press. Beauvois, M.-F., & Derouesne, J. (1979). Phonological alexia: Three dissociations. Journal of Neurology, Neurosurgery and Psychiatry, 42, 1115-1124. Beauvois, M. -F., & Derouesne, I. (1981 ). Lexical or orthographic dysgraphia. Brain, 104, 21-50. Benton, A. L., & Pearl, D. (Eds.). (1978). Dyslexia: An appraisal of current knowledge. New York: Oxford University Press. Bradley, L., & Bryant, P. (1985). Rhyme and reason in reading and spelling. Ann Arbor: University of Michigan Press. Chall, J., & Mirsky, A. F. (Eds.). (1978). Education and the brain. Chicago: University of Chicago Press. Crowder, R. G. (1984 ). Is it just reading? Comments on the papers by Mann, Morrison and Wolford and Fowler. Developmental Review, 4, 48-61. Cruickshank, W. M. (1975). The learning environment. In W. M. Cruickshank & D. P. Hallahan (Eds. ), Perceptual and learning disabilities in children (Vol. I, pp. 227-277). Syracuse, NY: Syracuse University Press. Cummins, J., & Das, I. P. (1977). Cognitive processing and reading difficulties: A framework for research. Alberta Journal of Educational Research, 23, 245-256. Cummins, J., & Das, J.P. (1978). Simultaneous and successive linguistic processes. International Journal of Psychology, 13, 129-138. Das, J. P. (1980). Planning: Theoretical considerations and empirical evidence. Psychological Research, 4 I, 141-151. Das, J.P. (1984). Simultaneous and successive processes and KABC. Journal of Special Education, 18, 229-238. Das, J.P., & Heemsbergen, D. B. (1983). Planning as a factor in the assessment of cognitive processes. Journal of Psychoeducational Assessment, 1, I-IS. Das, J. P., Kirby, J., & Jarman, R. (1975). Simultaneous and successive synthesis: An alternative model for cognitive abilities. Psychological Bulletin, 82, 87-103. Das, J.P., Kirby, J. R., & Jarman, R. (1979). Simultaneous and successive cognitive processes. New York: Academic Press. Das, J. P., Leong, C. K., & William~, .N. H. (1978). The: rc:lationship between learning disability and simultaneous-sue-
cessive processing. Journal of Learning Disabilities, II, 618-625. Denckla, M. B. (1983). Learning for language and language for learning. In U. Kirk (Ed.), Neuropsychology of language, reading, and spelling (pp. 33-43). New York: Academic Press. Doehring, D. G. (1984). Subtypingofreadingdisorders: Implications for remediation. Annals of Dyslexia, 34, 205-216. Doehring, D. G., Trites, R. L., Patel, P. G., & Fiedorowicz, C. A. M. (1981). Reading disabilities: The interaction of reading, language, and neuropsychological deficits. New York: Academic Press. Donaldson, M. (1978). Children's minds. Glasgow: Fontana/Collins. Donders, J., & van der Vlugt, H. (1984). Eye-movement patterns in disabled readers at two age levels: A test of Bakker's balance model. Journal of Clinical Neuropsychology, 6, 241256. Downing, I., & Leong, C. K. (1982). Psychology ofreading. New York: Macmillan. Downing, J., & Valtin, R. (Eds.). (1984). Language awareness and learning to read. Berlin: Springer-Verlag. Duane, D. D., & Leong, C. K. (Eds.). (1985). Understanding learning disabilities: International and multidisciplinary views. New York: Plenum Pres~. Duffy, F. H., Denckla, M. B., Bartels, P. H., & Sandini, G. (1980). Dyslexia: Regional differences in brain electrical activity by topographic mapping. Annals ofNeurology, 7, 4 I 2420. Duffy, F. H., Denckla, M. B., Bartels, P. H., Sandini, G., & Kiessling, L. S. (1980). Dyslexia: Automated diagnosis by computerized classification of brain electrical activity. Annals of Neurology, 7, 421-428. Duffy, F. H., & McAnulty, G. B. (1985). Brain electrical activity mapping (BEAM): The search for a physiological signature of dyslexia. In F. H. Duffy & N. Geschwind (Eds.), Dyslexia: A neuroscientific approach to clinical evaluation (pp. 105-122). Boston: Little, Brown. Elkonin, D. B. (l %3). The psychology of mastering the elements of reading. In B. Simon & I. Simon (Eds.), Educational psychology in the U.S.S.R. (pp. 165-179). London: Routledge & Kegan Paul. Elkonin, D. B. (1973). U.S.S.R. In J. Downing (Ed.), Comparative reading (pp. 551-579). New York: Macmillan. Ellis, A. W., Miller, D., & Sin, G. (1983). Wernicke's aphasia and normal language processing: A case study in cognitive neuropsychology. Cognition, 15, lll-144. Fletcher, J. M., Satz, P., & Morris, R. (1984). The Florida longitudinal project: A review. InS. A. Mednick, M.A. Harway, & K. M. Finello (Eds.), Handbook of longitudinal research (pp. 280-304). New York: Praeger. Frith, U. (1978). From printto meaning and from print to sound or how to read without knowing how to spell. Visible Language, 12, 43-54. Frith, U. (1979). Reading by eye and writing by ear. In P. A. Kolers, M. E. Wrolstad, & H. Bouma (Eds.), Processing of visible language (Vol. I, pp. 379-390). New York: Plenum Press.
REMEDIATION Frith, U. ( 1980). Unexpected spelling problems,ln U. Frith (Ed.), Cognitive processes in spelling (pp. 495-515). New York: Academic Press. Geschwind, N., & Galaburda, A.M. (1985). Cerebrallateralization. Biological mechanisms, associations, and pathology: II. A hypothesis and a program for research. Archives of Neurology. 42, 521-552. Gladstone, M., & Best, C. T. (1985). Developmental dyslexia: The potential role of interhemispheric collaboration in reading acquisition. In C. T. Best (Ed.), Hemispheric function and collaboration in the child (pp. 87-118). New York: Academic Press. Goldberg, E., & Costa, L. D. (1981). Hemisphere differences in the acquisition and use of descriptive systems. Brain and Language, 14, 144-173. Graham, F. K., & Kendall, B.S. (1946). Performance. of braindamaged cases on a memory-for-designs test. Journal of Abnormal Social Psychology, 41, 301-314. Graham, F. K., & Kendall, B. S. (1960). Memory-for-Designs test: Revised general manual. Perceptual and Motor Skills, 11, 147-188 (Monograph Suppl. No. 2-VII). Hiscock, M., & Kinsboume, M. (1987). Specialization of the cerebral hemispheres: Implications for learning. Journal of Learning Disabilities, 20, 130-143. Hunt, E. (1974). Quote the Raven? Never more! In L. Gregg (Ed.), Knowledge and cognition (pp. 129-157). Hillsdale, NJ: Erlbaum. Ilg, F. L., & Ames, L. B. (1964). School readiness: Behavior tests used at the Gesell Institute. New York: Harper & Row. lnhelder, B., & Piaget, J. (1964). The early growth of logic in the child: Classification and seriation (translated by A. E. Lunzer & D. Papert). London: Routledge & Kegan Paul. J0reskog, K. G., & SOrbom, D. (1984). USREL VI: Analysis of linear structural rellltionships by the methad of maximum likelihood. Chicago: Scientific Software. Kaufman, A., & Kaufman, N. (1983). The Kaufman Assessment Battery for Children: Interpretive manual. Circle Pines, MN: American Guidance Service. Kershner, J., Cummings, R. L., & Clarke, K. A. (1986). Two
year evaluation of the Tomatis Listening Training Program with learning disabled children. Toronto: The Ontario Institute for Studies in Education, University of Toronto. Kinsboume, M., & Hiscock, M. (1983). The normal and deviant development of functionallateralization of the brain. In M. M. Haith & J. J. Campos (Eds. ), Handbook ofchild psychology (4th ed., Vol. 2, pp. 157-280). New York: Wiley. Kirby, J. R., & Das, J.P. (1977). Reading achievement, IQ, and simultaneous-successive processing. Journal ofEducational Psychology, 69, 564-570. Kirby, J. R., & Das, J. P. (1978). Skills underlying Coloured Progressive Matrices. Alberta Journal of Educational Research, 24, 94-99. Kirby, J. R., & Robinson, G. L. W. (1987). Simultaneous and successive processing in reading disabled children. Journal of Learning Disabilities, 20, 243-252. Kirk, S. A., McCarthy, J. J., & Kirk, W. (1968). Examiner's manual: Illinois Test of Psycholinguistic Abilities (rev. ed.). Urbana: University of Illinois Press.
353
Knights, R. M., & Bakker, D. J. (Eds.). (1976). The neuropsychology of learning disorders: Theoretical approaches. Baltimore: University Park Press. Krywaniuk, L. W., & Das, J. P. (1976). Cognitive strategies in native children: Analysis and intervention. Alberta Journal of Educational Research, 22, 271-280. Kuhn, T. S. (1970). The structure ofscientific revolutions (2nd ed. enlarged). Chicago: University of Chicago Press. Lassen, N. A.,lngvar, D. H., & Skinh.j, E. (1978). Brain function and blood flow. Scientific American, 239, 62-71. Lawson, M. J., & Kirby, J. R. (1981). Training in information processing algorithms. British Journal of Educational Psychology, 51, 321-335. Leasak, J., Hunt, D., & Randhawa, B.S. (1982). Cognitive processing, intervention and achievement. Alberta Journal of Educational Research, 28, 257-266. Leong, C. K. (1976). Lateralizationinsevere1ydisab1edreadersin relation to functional cerebral development and synthesis of information. In R. M. Knights & D. J. Bakker (Eds.), The neuropsychology of learning disorders (pp. 221-231). Baltimore: University Park Press. Leong, C. K. (1980). Cognitive patterns of 'retarded' and belowaverage readers. Contemporary Educational Psychology, 5, 101-117. Leong, C. K. (1982). Promising areas of research into learning disabilities with emphasis on reading disabilities. In J. P. Das, R. F. Mulcahy, & A. E. Wall (Eds.), Theory andresearch in learning disabilities (pp. 3-26). New York: Plenum Press. Leong, C. K. (1984). Cognitive processing, language awareness, and reading in grade 2 and grade 4 children. Contemporary Educational Psychology, 9, 369-383. Leong, C. K. (1986). The role of language awareness in reading proficiency. In G. Pavlidis & D. F. Fisher (Eds.), Dyslexia: Its neuropsychology and treatment (pp. 131-148). New York: Wiley. Leong, C. K. (1987). Children with specifiC reading disabilities. Lisse, The Netherlands: Swets & Zeitlinger. Leong, C. K., &Carrier, I. C. (1986). Readers' comprehension of ambiguous sentences. International Journal ofExperimental Research in Education, 23, 218-233. Leong, C. K., Cheng, S.C., & Das, J.P. (1985). Simultaneoussuccessive syntheses and planning in Chinese readers. International Journal of Psychology, 20, 19-31. Leong, C. K., Cheng, S. C., Lundberg, 1., Olofsson, A., & Mulcahy, R. (submitted). The effects ofcognitive processing,
hlnguage access on academic perfomumce-Linear structural equation modelling. Leong, C. K., & Haines, C. F. (1978). Beginning readers' awareness of words and sentences. Journal of Reading Belulvior, 10, 393-407.
Leong, C. K., & Sheh, S. (1982). Knowing about languageSome evidence from readers. Annals of Dyslexia, 32, 149161. Levinson, H. N. (1980). A solution to the riddle dyslexia. Berlin: Springer-Verlag. Liberman, I. Y. (1983). A language-oriented view of reading and its disabilities. In H. R. Myklebust (Ed.), Progress in learn-
354
CHAPTER 18
ing disabilities (Vol. V, pp. 81-101). New York: Grone & Stratton. Liberman, I. Y., Shankweiler, D., Liberman, A.M., Fowler, C., & Fischer, F. W. (1977). Phonetic segmentation and recoding in the beginning reader. In A. S. Reber & D. L. Scarborough (Eds.), Toward a psychology of reading (pp. 207225). Hillsdale, NJ: Erlbaum. Lundberg, I. (1985). Longitudinal studies of reading and reading difficulties in Sweden. In G. E. Mackinnon & T. G. Waller (Eds.), Reading research: Advances in theory and practice (Vol. 4, pp. 65-105). New York: Academic Press. Luria, A. R. (1946). On the pathology of grammatical operations [in Russian]. lzvestiya APN RSFSR, No. 17. Luria, A. R. (1966a). Higher cortical functions in man. New York: Basic Books. Luria, A. R. (1966b). Human brain and psychological processes. New York: Harper & Row. Luria, A. R. (1970). Traumatic aphasia: Its syndromes, psychology and treatment. The Hague: Mouton. Luria, A. R. (1973). The working brain. London: Penguin. Luria, A. R. (1977). NeuropsYchological studies in aphasia. Lisse, The Netherlands: Swets & Zeitlinger. Luria, A. R. (1979). The making of mind: A personal account of Soviet psychology (M. Cole & S. Cole, Eds.). Cambridge, MA: MIT Press. Luria, A. R. (1981). Language and cognition (1. V. Wertsch, Ed.). New York: Wiley. Luria, A. R., & Artem'eva, E. Y. (1978). Two approaches to an evaluation of the reliability of psychological investigations (Reliability of a fact and syndrome analysis). In M. Cole (Ed.), The selected writings of A. R. Luria (pp. 282-293). White Plains, NY: Sharpe. Mann, V. A. (1984). Reading skill and language skill. Developmental Review, 4, 1-15. Marin, 0. S.M., Glenn, C., & Walker, J. A. (1982). Changing perspectives in cognitive neuropsychology. International JournalofPsychology, 17,231-257. Masland, R., & Usprich, C. (1981). [Review of A solution to the riddle dyslexia]. Bulletin of the Orton Society, 31. 256-26 I. Mattis, S. (1978). Dyslexia syndromes: A working hypothesis that works. In A. L. Benton & D. Pearl (Eds.), Dyslexia: An appraisal of current knowledge (pp. 43-58). New York: Oxford University Press. Miller, T. L., &Reynolds, C. R. (Eds.). (1984). Kaufman Assessment Battery for Children [Special issue]. Journal of Special Education, 18(3). Milner, B. ( 1963). Effects of different brain lesions on card sorting. Archives of Neurology, 9, 90-100. Morrison, F. J. (1984). Reading disability: A problem in rule learning and word decoding. Developmental Review, 4, 3647. Olofsson, A., & Lundberg, I. (1983). Can phonemic awareness be trained in kindergarten? Scandinavian Journal of Psychology, 24, 35-44. Olofsson, A., & Lundberg, I. (1985). Evaluation of long term effects of phonemic awareness training in kinderganen: Illustrations of some methodological problems in evaluation research. Scandinavian Journal of PsYchology, 26. 21-34. Olson, R. K., Kliegl, R., & Davidson, B. J. (1983). Dyslexic and
nonnal readers' eye movements. Journal of Experimental Psychology: Human Perception and Performance, 9, 816825. Orton, S. T. (1925). "Word blindness" in school children. Archives of Neurology and Psychiatry. 14, 581-615. Onon, S. T. (1937). Reading, writing, and speech problems in children. New York: Norton. Pavlidis, G. T. (1985). Eye movements in dyslexia: Their diagnostic significance. Journal ofLearning Disabilities, /8, 4250. Pavlidis, G. T. (1986). The role of eye movements in the diagnosis of dyslexia. In G. T. Pavlidis & D. F. Fisher (Eds.), Dyslexia: Its neuropsychology and treatment (pp. 97-1 10). New York: Wiley. Piaget, J., & Inhelder, B. (1956). The child's conception of space (F. J. Langdom & J. L. Lunzer, Trans.). London: Routledge & Kegan Paul. Raven, J. C. (1947). Progressive Matrices, Sets A, Ab and B. London: Lewis. Robinson, A. L., Heaton, R. K., Lehman, R. A. W., & Stilson, D. W. (1980). TheutilityoftheWisconsinCardSortingTest in detecting and localizing frontal lobe lesions. Journal of Consulting and Clinical PsYchology, 48, 605-614. Rourke, B. P. (1982). Central processing deficiencies in children: Toward a developmental neuropsychological model. Journal of Clinical NeuropsYchology, 4, 1-18. Rourke, B. P., Bakker, D. J., Fisk, J. L., &Strang,J. D. (1983). Child neuropsychology: An intraduction to theory. research, and clinical practice. New York: Guilford Press. Rutter, M. (1978). Prevalence and types of dyslexia. In A. L. Benton & D. Pearl (Eds.), Dyslexia: An appraisal of current knowledge (pp. 3-28). New York: Oxford University Press. Satz, P., & Morris, R. (1981). Learning disability subtypes: A review. In F. J. Pirozwlo & M. C. Wittrock (Eds.), NeuroPsYChological and cognitive processes in reading (pp. 109141). New York: Academic Press. Seymour, P. H. K., & Porpodas, C. D. (1980). Lexical and nonlexical processing of spelling in dyslexia. In U. Frith (Ed.), Cognitive processes in spelling (pp. 443-473). New York: Academic Press. Shallice, T. (1982). Specific impairment of planning. Philosophical Transactions ofthe Royal Society ofLondon, B298, 199-
209. Sbankweiler, D., Liberman, I. Y., Mark, L. S., Fowler, C. A., & Fischer, F. W. (1979). The speech code and learning to read. Journal of Experimental Psychology: Human Learning and Memory, 5, 531-545. Simon, H. A. (1976). Identifying basic abilities underlying intelligence performance of complex tasks. In L. B. Resnick (Ed.), The nature of intelligence (pp. 65-98). Hillsdale, NJ: Erlbaum. Sinclair, A., Jarvella, R. J., & Levelt, W. J. M. (Eds.). (1978). The child's conception oflanguage. Berlin: Springer-Verlag. Spreen, 0. (1976). Neuropsychology of learning disorders: Postconference review. In R. M. Knights & D. J. Bakker (Eds. ), The neuropsychology of learning disorders: Theoretical approaches (pp. 445-467). Baltimore: University Park Press. Stanovich, K. E. (1986). Matthew effects in reading: Some conse-
REMEDIATION quences of individual differences in the acquisition of literacy. Reading Research Quanerly, 21, 360-406. Tomatis, A. (1978). Education and dyslexia. France-Quebec: Les Editions. Tomeus, M. (1984). Phonological awareness and reading: A chicken and egg problem? Journal of Educational Psychology, 76, 1346-1358. Tunrner, W. E., Pratt, C., & Herriman, M. L. (Eds.). (1984). Metalinguistic awareness in children: Theory, research, and implications. New York: Springer-Verlag. Vellutino, F. R. (1979). Dyslexia: Theory and research. Cambridge, MA: MIT Press. Vygotsky, L. S. (1962). Thought and language (E. Hanfmann & G. Vakar, Eds. & Trans.). Cambridge, MA: MIT Press. (Original work published 1934). Vygotsky, L. S. (1978). Mind in society: The development of
355
higher psychological processes (M. Cole, V. John-Steiner, S. Scribner, & E. Souberman, Eds.). Cambridge, MA: Harvard University Press. Wagner, R. K., & Torgesen, J. K. (1987). The nature of phonological processing and its causal role in the acquisition of reading skills. Psychological Bulleti!J, /OJ, 192-212. Wechsler, D. (1974). Manual for the Wechsler Intelligence Scale for Children-Revised. New York: Psychological Corporation. Weinberg, J., Piasetski, E., Diller, L., & Gordon, W. (1982). Treating perceptual organization deficits in non-neglecting RBD stroke patients. Journal of Clinical Neuropsychology, 4. 59-75. Wolford, G., & Fowler, C. A. (1984). Differential use of partial information by good and poor readers. Developmental Review, 4, 16-35.
19 Neuropsychological Approaches to the Remediation of Educational Deficits PHYLLIS ANNE TEETER
The primary purpose of this chapter is to introduce neuropsychological principles and approaches related to the remediation of brain-related educational deficits. General theories of how the brain functions following damage or dysfunction will be discussed in terms of prediction and outcome expectations, including: neurodevelopmental factors that affect recovery of functions; how the severity of deficits affects remediation; and the impact of associated medical and psychological deficits on remediation. A review of several neuropsychological remediation programs will be presented, including Reitan's REHABIT Program and Kaufman's Sequential or Simultaneous Program. Finally, specific remedial techniques for treating brain-related deficits in reading, math, spelling, and reasoning will be discussed. Although this chapter will provide a theoretical basis for neuropsychological remediation, readers are cautioned that there are still many questions to be answered because so few remedial approaches for specific brain-related deficits have been empirically tested.
Applying Neuropsychological Theory to the Remediation of Educational Deficits The field of child neuropsychology has made enormous strides in recent years with regard to how the brain functions in children and young adolescents. General knowledge concerning how brain damage or dysfunction affects a child's educational PHYLLIS ANNE TEETER • Department of Educational Psychology, University of Wisconsin, Milwaukee, Wisconsin 53201
and psychological development has grown considerably with the advent of increased systematic research. The neuropsychological basis of specific childhood disorders has been widely studied, including learning disabilities (Gaddes, 1980; Rourke, 1985), dyslexia (Hynd & Cohen, 1983), math deficits (Strang & Rourke, 1985), spelling disorders (Sweeney & Rourke, 1985), hyperactivity (Rutter, 1983), and attentional deficits (Douglas, 1983). Despite these advances in our understanding of the brain-behavior relationship in specific learning disorders, the area of remediation is an evolving study and specific neurocognitive treatment approaches are still in their infancy. Rourke, Bakker, Fisk, and Strang (1983, p. 153) indicated that there are "more questions than answers with respect to the remediation of brain-related deficiencies." The application of neuropsychological theory to treatment is predicted on the valid and comprehensive assessment of a child's individual strengths and weaknesses across a variety of language, reasoning, motor, perceptual, and memory tasks. Valid neuropsychological assessment is particularly complex because the clinician must determine whether the child has a brain-related deficit, a developmental delay, or a neuropsychiatric disturbance (Hooper & Boyd, 1986). There are a number of factors that further affect treatment outcomes, including: the age of the child when the damage occurred; the nature and severity of the neuropsychological deficits; the presence or absence of specific strengths; the premorbid status of the child; and other environmental and motivational factors (Rourke et al., 1983). For these reasons, remedial programs should be individually devised and monitored (Hartlage & Reynolds, 1981). The emphasis on an individual approach makes systematic research on the effectiveness of treatment problematic. Furthermore, it is difficult to make gen357
358
CHAPTER 19
eralizations from one child to another because the neurocognitive deficits, the ability structures and environmental factors vary across individuals. This may also explain why so many studies investigating treatment effectiveness have produced equivocal results. For example, many studies involve the pre- and posttesting of groups of brain-damaged children who have very heterogeneous abilities (Rourke et al., 1983). In order to determine a child's remedial needs and to design a remedial program, it is important to identify the etiology of the problem; that is, are the deficits a result of acquired (e.g., traumatic injury) or neurodevelopmental (e.g., learning disability) anomalies? The deficits and the course of recovery for each injury may differ significantly. See Berg (1986) for a discussion on recovery of function following closed-head injuries in children, and Hynd ( 1986) for a discussion of educational interventions for children with developmental learning disorders. The purpose of this chapter is to provide the reader with principles that can be applied to the treatment of children with brain-related educational deficits. Several important issues that impact on remediation will be presented.
Factors Affecting Outcome In the following section a number of factors that interact with and affect treatment outcomes will be discussed. Three major areas will be included: neurodeve1opmental issues; the nature and severity of deficits; and associated medical and psychological deficits.
Neurodevelopmental Issues Age of Onset of Injury. Neurodevelopmental factors increase the complexity of both the neuropsychological assessment and the treatment of childhood brain-related disorders. The age of onset of injury has been the focus of many studies investigating whether early or late damage is most debilitating. This issue is far from being fully resolved, but several studies provide a basis for predicting probable outcomes. Reitan (1974) argued that early damage may be more significant than is later damage. In a longitudinal, crosssectional study, Reitan ( 1981) compared the effects of injury on children 4, 8, 12, and 16 years of age. Distinctly different learning curves were plotted for each age group. Although each brain-damaged group showed learning deficits when compared to ''normal" peers, the older group came closer to approx-
imating the learning capacity of their nonimpaired counterparts. The effects of early damage reduced the overall developmental potential of the young children, and they never achieved the performance levels of their normal peers. These results led Reitan ( 1981) to conclude that the longer the brain was intact, the greater was its potential to develop higher-level language and cognitive abilities. Plasticity theories challenge Reitan's conclusions, and are based on the observation that young children show a greater capacity for recovery of function than do adults following left hemisphere lesions. For example, Alajouanine and Lhermitte (1965) found that children (aged 6 through 15 years) showed remarkable recovery oflanguage abilities 1 year after insult to the left hemisphere. Basser (1962) and Hecaen (1976) also reported more severe language deficits in adults following left hemisphere damage, with milder disorders present in children. Although both hemispheres appear to be anatomically different at birth (Witelson & Pallie, 1973) and functionally asymmetrical, both appear to be able to assume activities normally monitored by the opposite hemisphere following injury (Kolb & Whishaw, 1980). Golden and Wilkening ( 1986) reported that transfer of language to the right hemisphere is optimal prior to the age of 2 years. The transfer is less complete for older children, and the deficits mimic those found in adults. Kolb and Whishaw (1980) provided a developmental matrix concerning cerebral asymmetry and transfer of functions across the hemispheres following injury. In this model, based on hemidecortication studies, both hemispheres have the capacity to assume lower-level functions of the opposite hemisphere. However, neither is able to fully assume the functions of the other side. For example, if the left hemisphere is removed, the right hemisphere can perform simple language tasks without compromising simple or complex visuoconstructional capacities; and simple visuospatial tasks can be performed by the left hemisphere without interfering with complex language abilities. The transfer of more complex activities of either hemisphere seems to be in jeopardy, however. Kolb and Whishaw (1980) suggested that at birth there is considerable overlap in the functional capacity of the two hemispheres because cognitive functions at this age are primarily lower-level. This overlap is significantly decreased by the age of 5 years because the developing "new functions" are more specialized. This, however, does not mean that the hemispheres are "bc~oming" more spe~ialized with age, but that higher-level abilities are more specialized. These results support the notion of plasticity
REMEDIATION OF EDUCATIONAL DEFICITS
but not equipotentiality because of the apparent loss of higher-level skills (Kolb & Whishaw, 1980). Other studies also have shown that general intelligence is lowered when one hemisphere is seriously compromised (Milner, 1975). These theories have addressed transfer of functions only in children with severe brain pathology. The course and nature of the transfer of functions across or within hemispheres for other milder dysfunctions (i.e., developmental dyslexia) is not adequately understood. However, Luria's (1963) theory of functional systems provides us with a plausible hypothesis for how transfer might occur. Generally, the basic assumption is that any specific behavior can be performed by more than one functional system. When injury occurs to one system, alternative functional systems may develop either spontaneously or through specific remedial approaches (by changing the input or output demands of the task). This process of developing alternative functional systems will be discussed in more detail later in this chapter. Boll and Barth ( 1981) pointed out another important difference between children who have sustained lateralized damage that results in the removal of one hemisphere, and children who have sustained generalized damage where surgery is not feasible. By removing a damaged brain area, the abnormal influence of this dysfunctional system on cognitive activities may be less significant than the effects of continued influence from a pathological system. Boll and Barth ( 1981) cited a number of studies reporting that intact systems (or hemispheres) may receive interference or competition from damaged areas when healthy structures attempt to assume a specific function for the impaired cortical region. Consequently, milder brain impairment may be more detrimental to the overall functioning of the brain than the absence of localized or lateralized cortical regions (Boll & Barth, 1981). . Det~rmining the effects of early versus late injury IS obviOusly a complex issue, which is dependent upon the specific task being measured and the normal developmental sequence of acquisition of that task (Wilkening & Golden, 1982). Wilkening and Golden ( 1982) further theorized that different brain systems may be involved in the execution of specific behaviors at different ages. That is, whereas the behavior may be essentially the same, the brain substrate and the psychological processes may be entirely different. Teuber and Rudel (1962) showed that certain behaviors may not differentiate brain-damaged from normal individuals at earlier ages, but with time the same tasks can be highly discriminatory. Thus, the effects of some impairment may not be apparent until
359
.later developmental stages. Furthermore, it has been hypothesized that children may actually ••grow into a deficit when the functions subserved by the destroyed tissue become increasingly more crucial for the behavioral repertoire during development'' (Rourke et al., 1983, p. 92).
Neurodevelopmental Stages. Golden and Wilkening ( 1986) advanced a developmental paradigm based on Luria's concept of the functional units of the brain. They outlined five stages: ( 1) the development of the reticular activating system (from birth to 12 months); (2) the development of primary sensory and motor cortices (age range similar to Stage 1); (3) the development of secondary sensory and motor cortices (birth through age 5 years); (4) the development of tertiary, association areas in the parietal lobe (5 through 8 years); and (5) the development of association regions of the frontal cortex (1 0-12 years through adolescence). Golden and Wilkening (1986) suggested that it is almost impossible to predict whether a 4-year-old child has damage to the tertiary parietal lobe because the effects of damage may not be apparent until about the ages of 8 to 12 years. Obviously when designing remedial strategies, clinicians must be able to formulate reasonable predictions about the developmental patterns that specific disabilities may take. Predictions are most useful when they are based on an interaction between neuropsychological test results and their associated correlates of brain dysfunction, and are interpreted in a developmental framework (Rourke et al., 1983). Boll and Barth (1981) indicated that the type, size, extent, and location of damage must also be considered ~~fore one can determine the nature of specific cogmttve losses. Other factors impact upon our ability to predict the outcome of brain damage, including: the socioeconomic status of the family; the results of computerized axial tomography; loss of consciousness following· injury; and the treatment following injury, such as surgery or cranial irradiation (Wilkening & Golden, 1982). Rourke et al. (1983) suggested that predictions should be considered in terms of short-term and long-term consequences, and therapy approaches should include both components. Nature and Severity of Neurocognitive Deficits The first step in designing remedial programs is to determine the nature and severity of the child's dysfun~tion. A comprehensive clinical neuropsychologtcal assessment, including intelligence and achievement testing, is the starting point for deter-
360
CHAPTER 19
mining the child's remedial needs. Three child batteries are currently in use: the Halstead Neuropsychological Test Battery for Children, ages 9 through 14; the Reitan-Indiana Neuropsychological Test Battery, ages 5 through 8; and Lucia-Nebraska Neuropsychological Battery-Children's Revision. Individual tests for brain damage are also available, including: measures of visual-spatial abilities (e.g., Benton's Visual Retention Test), measures of memory abilities (e.g., Benton's Sentence Memory Test), and measures oflateral dominance (e.g., Dean's Test of Lateral Dominance). Although individual tests may be employed, a comprehensive battery approach is necessary to determine the full range of neuropsychological strengths and weaknesses. A broad range of behaviors should be assessed to determine the functional integrity of neural structures and cortical systems that mediate cognitive and sensory-motor abilities. A thorough evaluation of the child's language, intellectual, achievement, memory, sensoryperceptual (auditory, visual, and tactile), motor, and reasoning abilities is necessary. See Salvia and Ysseldyke ( 1985) for a review of psychometrically sound achievement tests. Selz and Reitan ( 1979) supported a multidimensional interpretative model, involving four levels of inference: (1) analyzing the child's level of performance; (2) analyzing patterns of performance; (3) analyzing pathognomonic signs; and (4) analyzing right-left performance differences. The level of performance is determined by using a normative approach. Because normal or abnormal levels of performance cannot unequivocally indicate normal or abnormal brain functioning, the presence of pathognomonic signs and the pattern of performance are necessary to fully evaluate the integrity of brain functions. Also the level of performance may be similar for children with distinctively different types of dysfunction. Patterns of performance typically include comparisons between: Verbal and Performance IQ scores; Wechsler subtest patterns; scores on the Trail Making Test, Part A and B (Halstead-Reitan Battery); and scores on the Speech Sounds Perception Test (Halstead Battery). Patterns of performance can be useful for lateralizing and localizing specific deficits. For example, significantly lower Verbal IQ (compared to Performance IQ), lower Trails B score (compared to Trails A), and poor performance on the Speech Sounds Perception Test are often associated with left hemisphere dysfunction. Pathognomonic signs are also useful because these specific deficits occur almost exclusively in brain-impaired children. The Aphasia Screening Test (Halstead-Reitan) provides powerful discrimi-
nators, including signs for: dyscalculia, central dysarthria, dysnomia, and dysgraphia, indicative of left hemisphere dysfunction; and constructional dyspraxia, indicative of right hemisphere dysfunction. The presence of these signs must be interpreted in terms of the development of the child. That is, it is imperative to determine whether the skill in question has been developed prior to injury before one can infer brain pathology. Lateralized sensory and motor deficits are among the most valid signs of cerebral damage or dysfunction (Reitan, 1981), where unilateral errors implicate the contralateral hemisphere. Rourke et al. (1983) indicated that the "general adaptational capacity" of the brain is a critical variable in judging the child's ability to respond to treatment. They suggested that the four levels of inference can be helpful in determining the adaptational capacity of the brain. Specific scores on the Category Test,· the Seashore Rhythm Test, and the Tactual Performance Test (Halstead-Reitan) are good measures of the child's ability to profit from feedback and to attend (Rourke etal., 1983). Scores on these tests are assessed from a qualitative as well as a quantitative approach. Rourke et al. ( 1983) concluded that children with higher general adaptational skills are more likely to profit from remediation, and may profit from more varied remedial approaches. This is consistent with Luria's (1965) contention that the more systems that are functional, the greater is the potential for identifying alternative functional systems.
Associated Medical and Psychological Deficits There a number of medical and psychological factors that should be considered before planning a remedial program for a child with brain-related educational deficits. Among the most important are situations that may require drug intervention (e.g., seizures and hyperactivity) or intensive psychotherapy (e.g., severe depression). These factors will be briefly discussed in this section.
Seizures and Behavioral Sequelae. Prevalence figures for epilepsy are difficult if not impossible to gather; however, estimates range from I to 2% in the general population.(Bolter, 1986). Bolter (1986) cited several studies indicating that the majority of recurring seizures first appear during early childhood (25% before 5 years; 25% between the ages of 5 and 14 years) and adolescence (10% between 15 and 19 years). The etiology of seizures can also vary considerably according to Bolter (1986) and occur for numerous reasons including: head traumas, diseases, drugs, toxins, congenital abnormalities, and meta-
REMEDIATION OF EDUCATIONAL DEFIOTS
bolic disorders. Bolter (1986) further indicated that many clinicians view epilepsy as the most common childhood neurological problem. There are conflicting data concerning the relationship between seizures and behavioral disorders. Corbett and Trimble (1983) surveyed eight studies where the frequency of behavior disorders in children with epilepsy ranged from 12% to 67%; whereas others suggest that the presence of EEG abnormalities does not predict whether a child will exhibit psychological or adjustment problems (Boll & Barth, 1981 ). Stores, Hart, and Piran (1978) found that the most consistent deficit of individuals with epilepsy was an attention disorder. Stores et al. also found that some children with epilepsy actually performed better in a high-stimulus environment that was otherwise distracting to normals. Other studies have suggested that EEG abnormalities do not necessarily indicate poor treatment prognosis (Hughes, 1968; Tymchuk, Knights, & Hinton, 1970). When seizures occur in the first year of life (for reasons other than high temperatures), Chevrie and Aicardi (1978) found the prognosis to be more guarded, with as many as 50% of 313 children showing signs of severe mental retardation. Boll and Barth (1981) indicated that prognosis is best when: (1) the etiology of the seizure is unknown; (2) the child is older; (3) there is a history of seizures in other family members; and (4) seizures are generalized and of short duration. Although cognitive deficits do not accompany all forms of seizures, some children do evidence progressive intellectual deterioration (Corbett & Trimble, 1983); and in one study, one in five children with complicated seizures displayed severe reading retardation. Seizure type appears to be associated with varying degrees of neuropsychological deficits. Tonic-clonic (grand mal) seizures have the most negative impact on generalized functioning compared to other seizures; whereas petit mal seizures typically do not significantly impair performance on measures of neuropsychological and emotional functioning (Berg, 1986). The psychosocial adjustment of children with seizures can be an important variable in treatment, because they often experience social isolation and rejection from peers (Corbet & Trimble, 1983). Parental attitudes also appear to be a critical factor to the child's adjustment to seizures. Some parents have lowered expectations for their child's academic potential (Long & Moore, 1979); and Kleck (1968) found that children learn feelings of shame about their condition from their parents. The clinician working with a child with seizures will want to consider the need for family therapy to avoid these prob-
361
lems. The child with seizures also needs careful monitoring by a physician to ensure proper drug dosages when medication is necessary. The psychologist will want to monitor closely the child's reaction to medication, and should act as a conduit sharing this information with the pediatrician and the parents.
Side Effects of Anticonvulsants. The side effects of common anticonvulsants should be considered when planning treatment programs. Corbett and Trimble ( 1983) reviewed a number of studies investigating the side effects of several anticonvulsants (i.e., phenytoin, valporate sodium, carbamazepine, phenobarbitone), and provided useful information concerning the behavioral and cognitive impairments associated with medication. Classical cerebellar dysfunction has been found in children who demonstrate signs of toxicity following long-term treatment with phenytoin; carbamazepine has been associated with deterioration of behavioral and cognitive abilities in some childem (Corbett & Trimble, 1983). Several studies have reported beneficial side effects with valproate sodium, including increased visuomotor coordination, alertness, and general school adjustment. Many anticonvulsants have an adverse effect on cerebral metabolism, resulting in low levels of serum folic acid. In one study of children with complex epilepsy requiring multiple drugs, serum folic acid levels were related to signs of neurosis, depression, and intellectual deterioration (Corbett & Trimble, 1983). Although psychiatric disorders have been found in individuals with folic acid deficiencies, the administration of folic acid has not been shown to reverse behavioral disorders (Corbett & Trimble, 1983). Although this review suggests that there is a need for further research on side effects, it does point out that psychologists and educators should work closely with the physician to get as much information as possible prior to formulating an educational treatment plan for children with seizures. Attention Deficit Disorder and Hyperactivity.
Interest in the diagnosis and treatment of children with attention deficit disorders (ADD) with hyperactivity has increased considerably in the past ten years. Barkley ( 1982) and others (Douglas, 1983) provided numerous classification criteria and outlined associated features of this complex disorder. Although cognitive deficits can be present (Lambert & Sandoval, 1980), many researchers stress the importance of differentiating children with an ADD from those with primary learning disabilities (LD) (Barkley, 1982; Douglas & Peters, 1979). For the purposes of this chapter, discussion of ADD (with hyperactivity) will be limited to "pure" cases without evidence of pri-
362
CHAPTER 19
mary language processing or visual-perceptual disorders, which are often found in LD children. Generally, behavioral characteristics for ADD include: ''impaired attention and effort, poor inhibitory control, arousal modulation problems, and inclination to seek immediate reinforcement" (Douglas, 1983, p. 323). These deficits negatively affect cognitive development, because these mechanisms interfere with the child's ability to formulate mental representations or schemata (Douglas, 1983). Inadequate motivational and ineffective coping skills may also interrupt problem-solving abilities. Douglas (1983) suggested that differences in cognitive abilities between ADD and normal children increase with age due to the nature of more complex learning. That is, "later learning often requires more deliberate, active, and conscious effort than early learning,'' and ''later learning is dependent upon early learning'' (Douglas, 1983, p. 283). Poor planning, organizational, and problem-solving strategies are seen on complex tasks due to the child's inability to regulate responses, to inhibit impulsivity, and to exhibit sufficient effort. Douglas ( 1983) believes these factors are most critical, and challenges research suggesting that poor selective attention and distractibility result in a child's failure to select the relevant dimensions or elements necessary to complete a task.
observed after long-term drug usage so that adult growth does not appear to be affected (Rapoport, 1983). Despite these findings, most investigators indicate that more research is necessary to understand fully how stimulants work on the central nervous system.
Psychological Factors. The incidence of emotional disturbance in children with brain damage is significantly high, and should not be overlooked in any remedial plan. Rutter, Graham, and Yule (1970) and Rutter, Tizard, and Whitmore (1970) found the incidence of emotional disturbance for brain-impaired children to be approximately 6 times higher than for the normal population. Boll and Barth ( 1981) suggested that there are a number of factors that are not usually associated with emotional adjustment that may influence this occurrence. These factors primarily involve: the child's self-image concerning their impairment or loss of functions; the difficulties they may have in learning; and the way they interact with others (Boll & Barth, 1981). Attempts to identify specific types of psychiatric disorders in brain-injured children have not been successful. Rutter, Chadwick, and Shaffer (1983) indicated that the pattern of psychiatric disturbance is relatively the same for normal and head-injured children. When differences were found, the psychologiSide Effects of Stimulants. When children with cal disorder manifested as disinhibited social behavhyperactivity require drug treatment, stimulants such ior (Rutter et al., 1983). Specifically, only children as methylphenidate are administered (Barkley, 1977; with very severe head injury displayed signs of a Taylor, 1983). In a review of several studies, Taylor "frontal-lobe syndrome," where social behaviors (1983) initially found that children show decreased were significantly impaired and were characterized restlessness, impulsivity, and inattention following by: outspokenness that was socially inappropriate; medication. Despite these behavioral improvements, delving into personal information of self and others; learning and academic achievement do not neces- and undressing in socially inappropriate situations sarily improve with drug treatment alone (Rapoport, (Rutter et a/., 1983). 1983). Studies combining drug treatment with beThe prevalence of frontal lobe damage in chilhavior modification typically show greater cognitive dren sustaining closed-head injury appears to be high improvements than drug treatment alone (Gittelman, because many accidents involve an impact to the Abikoff, Klein, & Mattes, 1979). Side effects of back of the head (Berg, 1986). In accidents of this stimulants have been reported in a number of studies nature the brain is thrust forward, resulting in tissue investigating how social-emotional variables are in- damage opposite the original impact site or in the fluenced. In two studies, stimulants (i.e., meth- frontal lobes. Berg (1986, p. 125) indicated that damylphenidate) reduced the social interaction of boys age to the frontal lobes can result in behaviors that with peers and parents (Barkley & Cunning- impair social behaviors and judgments, and ''may ham, 1979; Whalen, Henker, Collins, McAuliffe, & likely impede or completely arrest development of Vaux, 1979). Although children on medication functions that are critical to adequate functioning as showed more compliance, they initiated fewer in- an adult." Although early studies with brain-injured chilteractions and appeared to be more unhappy or depressed than were their hyperactive counterparts on a dren suggested that they were likely to show signs of placebo. Other side effects have been reported, in- hyperactivity and noncompliance, this has not been cluding decreases in weight and height gains. How- confirmed (Rutter et al., 1983). Research evidence ever, these effects seem to be short-term, and are not indicates that there are no unitary or stereotypic emo-
REMEDIATION OF EDUCATIONAL DEFICITS
tional patterns that result from childhood brain dysfunction. However, children with brain damage appear to be at risk for displaying emotional problems, and they must develop coping mechanisms to deal with their particular deficits. Treatment plans must address the child's social and emotional needs to avoid significant disturbances that may be more debilitating than the brain dysfunction itself. The clinician should look for signs of depression, withdrawal, and poor self-esteem prior to and during treatment. Psychotherapy may be warranted and a psychiatric evaluation may be needed.
General Approaches to Educational Treatment After a comprehensive neuropsychological evaluation, the child's functional status can be determined and steps can be taken to plan an appropriate remedial program. Remedial programs usually combine a number of different techniques and strategies that are individually designed according to the ~hild' s neuropsychological, educational, and psychological needs. Once a child's needs are determined, there are a number of general principles that affect the way particular remedial strategies are employed. Depending upon the theoretical orientation of the professionals working with the child, a program may be designed to: (1) attack the child's weaknesses; (2) teach to the child's strengths; or (3) address both strengths and weaknesses (Rourke et al., 1983). The adoption of one of these theoretical approaches will affect both the techniques employed and the course of the remediation program. In this section, a review of the three orientations will be briefly presented.
Attacking Weaknesses
363
proaches is that remediation focuses on training-specific processing deficits (e.g., form perception, perceptual-motor match, auditory discrimination). Whereas there may be some improvement in the process itself, there is no positive transfer of learning to reading or other academic areas. Similar conclusions have been reached concerning techniques directed at "training the brain," such as the Dolman-Delacato Neurological Approach (Hynd & Cohen, 1983; Hartlage & Reynolds, 1981; Zarske, 1982). Rourke et al. (1983) argued that there are instances when "attacking weaknesses" is not only warranted but may be preferred. First, for young children the remediation of neuropsychological deficits may help in the reorganization of damaged cortical areas. Second, older children may avoid using a deficit area completely, which can impede the remedial process. For example, Rourke et al. (1983) suggested that older children with auditory-perceptual deficits may ignore verbal imput and rely almost exclusively on visual input. Third, direct remediation of deficits should begin as early as possible for children with head injuries. Rourke et al. indicated that this reduces the possibility of deficits becoming entrenched or characteristic of the child's learning style. Fourth, deficits that are prevasive in nature should be remediated directly, such as attentional and motor deficiencies. Fifth, children with mild impairment may show marked improvement. Finally, direct remediation may be the only approach possible for children with pervasive brain dysfunction, such as mental retardation. Rourke et al. indicated that direct remediation of deficits should be abandoned in cases where the child is not responding well to treatment, or is developing emotional problems. Reynolds ( 1981) was more critical of the '' deficit" approach to remediation for several reasons. First, he asserted that this method is likely to be ineffective because the focus of remediation is on brain areas that are damaged or dysfunctional. Second, by teaching the child through methods that require activation of cortical areas that are not intact, we increase the child's potential for failure and this may be harmful to the individual (Hartlage & Reynolds, 1981). Finally, Reynolds (1981) pointed to research indicating that these remedial practices have been found to be ineffective.
When attacking weaknesses, direct remediation of the child's neuropsychological deficits is undertaken. Deficit training has a long-standing precedence in educational circles (Ayres, 1974; Frostig & Home, 1964; Kephart, 1971; Kirk & Kirk, 1971). Hammill and Larsen (1974) conducted an intensive review of the literature of several of these approaches, most notably the psycholinguistic training program proposed by Kirk and Kirk (1971). Goodman and Hammill (1973) also reviewed the perceptual-motor training program introduced by Frostig Teaching to Neuropsychological Strengths and Home ( 1964). Almost without exception, effectiveness studies of these programs do not support Strength approaches support the practice of their utility (Hynd & Cohen, 1983). One of the pri- identifying the child's most intact abilities, then planmary reasons for the negative results of these ap- ning and implementing a remedial program that
364
CHAPTER 19
focuses on these strengths. This approach is dependent upon an ipsative interpretative model, where a child's pattern and level of performance are analyzed in terms of relative strengths and weaknesses based on the individual ability structures or profiles (Reynolds, 1981). Although normative interpretation should be part of the analysis of test scores, sufficient information concerning individual strengths and weaknesses cannot be gathered when this is the only method of analysis. Selz and Reitan ( 1979) developed a set of rules for classification of children, using the Halstead Neuropsychological Battery, incorporating both normative and ipsative interpretations of test scores for classifying children as brain damaged, learning disabled, or normal. Raw scores are converted to a fourpoint scale (0 for normal performance and 3 for abnormal performance), which allows for a more direct comparison across different tests. Scaled scores can be summed and cutoff scores can be used to determine the severity of deficits. However, more importantly, "the scaled score conversion provides a means of evaluating the relative quality of a child's performance in different areas: one can determine quickly whether a child's 3s are cropping up in the areas of motor performance, reasoning, or language skills" (Selz, 1981). There are a number of reasons for adopting a "strength" model of remediation. First, this method may be helpful for children who are resistant to direct remediation of weaknesses (Rourke et al., 1983). When self-confidence is low or when a failure syndrome emerges as a result of frustration, a strength approach should be adopted. Second, by teaching to the child's strengths we may reduce the possibility of the child falling farther and farther below peers in academic areas. Finally, Luria suggested that recovery of function following cortical damage can be achieved ". . . by the replacement of the lost cerebral link by another which is still intact" (Luria, 1963, p. 55). For example, a child with an impaired auditory system could be taught to differentiate simple sounds using visual or nonverbal images.
A Combined Remediation Approach A combined approach would incorporate both strengths and weaknesses in the remedial plan. In this model, the children could be taught new information through their strengths while deficiency areas would receive direct remediation. Rourke et al. (1983) suggested that a ''mixed'' intervention strategy could be used for a child with weaknesses in verbal commu-
nication and organizational abilities, and strengths in visual-spatial abilities. The child would receive speech therapy to improve expressive abilities. Concurrently, the child would also be taught to use visual imagery to help monitor verbal expression. Thus, deficits would be directly attacked and the child would be taught strategies that capitalize on individual strengths. Obviously the distinction between these approaches is critical when designing intervention programs. Decisions about whether to attack a child's weaknesses should be made in the context of the severity of the damage, and the emotional and adaptational capacity of the child. When a direct approach is taken and the child is not progressing, a strength approach is recommended. Once a child starts to experience success with learning again and his or her self-concept is strong, a mixed approach could then be implemented. However, the clinician should introduce remedial techniques for weaknesses in small time units every day. Sessions should be designed so that the child can end each lesson with an accomplishment. Techniques for remediating weaknesses should be introduced early in the teaching session, and the child should finish by practicing highly improved or consistently strong skills. The following specific neuropsychological remedial programs utilize various aspects of the strength/weakness paradigm.
Specific Neuropsychological and Neurocognitive Programs of Remediation Presently, a number of neuropsychological remedial programs are available for children with brain-related cognitive deficits. Two of these programs will be briefly reviewed: (l) REHABIT, the Reitan Evaluation of Hemispheric Abilities and Brain Improvement Training (Reitan, 1980), and (2) the Kaufman-Sequential or Simultaneous program (Kaufman, Kaufman, & Goldsmith, 1984). Although these intervention programs are based on neuropsychological and cognitive theories, each is still in the developmental stage and in the process of being empirically validated.
Reitan's REHABIT The beginning of a formal study of child clinical
neuropsychological assessment can be traced to Ralph Reitan's early work with Ward Halstead in the
REMEDIATION OF EDUCATIONAL DEFICITS
1950s. The Halstead Neuropsychological Test Battery for Children, for ages 9 through 14 years, was standardized in 1954 and was followed by the development of the Reitan-Indiana Neuropsychological Test Battery for Children, for ages 5 through 9 years (Reitan & Davison, 1974). In 1965, the first report of an extensive data base was published after 12 years of research (Boll, 1974). The development of these batteries provided a standard battery for assessing childhood brain dysfunction, and offered child clinical neuropsychologists a methodology for obtaining valid descriptive information about the behavioral consequences of brain pathology. More recently, Reitan ( 1980) developed a rehabilitation program for children with brain-related disorders, and again pioneered a critical phase in child clinical neuropsychology. REHABIT is the most comprehensive neuropsychological remediation program presently available for children with a variety of deficits. TheREHABIT materials incorporate the basic principles of the brain-behavior relationship as identified in the Halstead-Reitan Neuropsychological Test Batteries (Reitan, 1980). Diagnosis of the child's brain-related deficits is the first step of this remedial program, and results from the Halstead-Reitan are used to determine which training modules should be used. Reitan (1980) did not make a distinction between training principles for adults and children. Reitan suggested that when children have injury to specific brain areas, the deficits are similar to those observed in adults. The major difference in remediation for specific deficits is that children generally require less complex levels than do adults.
Three Phases of REHABIT REHAB IT is organized in three phases (Reitan, 1980): (1) the evaluation of brain-related deficiencies using the Halstead-Reitan batteries; (2) the training of specific deficits using the test items from the Halstead-Reitan batteries; and (3) the training of deficit areas using special REHABIT materials. These phases were meant to proceed one to the other. Assessment then is the starting point. According to Reitan (1980, p. 1), "inadequate evaluation of the brain-behavior relationships for the individual subject has been the greatest impediment in brain-training efforts.'' Reitan further found that items from the neuropsychological test batteries can be used to train basic brain functions in individuals with impairment. For example, Reitan found that individuals with reasoning and abstraction deficiencies have responded
365
most favorably and have shown increased abilities when repeatedly trained on alternate forms of the Category test. Finally, Reitan identified a number of other teaching materials that can be used to train children with brain impairment. The materials used in the third phase of REHABIT have been gathered from a variety of training procedures and vary from simple to complex tasks. Training procedures have been developed for three fundamental areas: (1) language and verbal abilities; (2) abstract thinking, reasoning, and problem-solving skills; and (3) visual-spatial, motor, and sequential abilities (Reitan, 1980). Reitan (1980) emphasized the importance of training general skills that require integrated brain functioning, especially reasoning, abstraction, and logical thinking. These general skills may be more critical for overall brain functioning than are highly specialized skills.
Five Tracts for Training Specific Abilities These materials are presented in five tracts for training general abstraction abilities. Reitan (1980) described the tracts as: (1) Tract A, materials for expressive-receptive language and verbal skills; (2) Tract B, materials for abstraction, reasoning, organization, and logical analysis in the verbal domain; (3) Tract C, materials for general abstraction, reasoning, and organizational skills; (4) Tract D, materials for abstraction emphasizing visual-spatial, manipulation, and sequential processing; and (5) Tract E, materials for basic visual-spatial and manipulation skills. The materials are arranged in order from simple to complex. Based on the results of the neuropsychological evaluation, materials are selected from each tract depending on the type and severity of the deficits identified. Some individuals may require training in all five tracts, whereas others may require training in one specified tract. Reitan (1980) suggested that training should begin at a level where the individual can be successful, and should proceed to more complex materials. In doing this, the individual experiences success and develops a positive attitude about the training program. Reitan ( 1980) provided a comprehensive list of materials, described the ability functions necessary for completing each task, and indicated the primary brain areas involved in each task. For example, a child with spatial-sequential and abstraction deficits could be trained with the "What Follows" item, which progresses from simple to moderate levels of difficulty. The functions involved in the completion
366
CHAPTER 19
of this task are organization and picture arrangement skills, which are primarily monitored by the right hemisphere but also require intact general brain functioning. (A list of the publishers and distributors of these materials is provided in Reitan, 1980.)
Validation of REHABIT Presently, no published studies have empirically tested the REHAB IT procedures. However, Reitan (1980) reported that this program was used with success for a number of years at the Reitan Neuropsychological Laboratory at the University of Arizona. The procedures and materials available through REHABIT provide a preliminary step in the much needed link between assessment and remediation of brain-related disorders. This program also has a strong theoretical basis and has been clinically tested. However, controlled research studies are needed to validate these procedures, and to identify which kinds of deficits respond best to treatment.
Kaufman-Sequential or Simultaneous (K-SOS) Remedial Program Cognitive processing theories have been used to investigate the neuropsychological basis and remediation of severe achievement deficits. Das, Kirby, and Jarman (1979) discussed a comprehensive theory, originally reported in Luria's (1966) theory of functional units of the brain, relating simultaneous and sequential processing abilities to brain functioning and to academic performance. Simultaneous processing involves a synthesis of stimuli into integrated units that typically have visual-spatial overtones. Sequential processing involves the sequential ordering of information, with heavy language and time-sequence overtones. These two processes have been associated with right and left hemisphere functions, respectively (Cohen, 1973), or with posterior and anterior cortical regions (Das et al., 1979). Kaufman and Kaufman (1983) incorporated this neurocognitive processing theory into a comprehensive test battery, the Kaufman Assessment Battery for Children (K-ABC) and, more recently, used this theory to develop a procedure for remediating cognitive deficits in school-aged children (Kaufman et al., 1984). The K-SOS approach includes methods for interpreting scores on the K-ABC from a practical perspective, and for developing remedial strategies for children with learning deficits (Kaufman et al., 1984). The K-SOS approach outlines two characteristic learning styles, suggests how these styles may
affect classroom behavior, provides examples of remedial techniques, and shows how K-ABC scores can be translated into a remedial plan (Kaufman et al., 1984). In this remedial approach, children are taught through their mental processing strengths based on scores from the K-ABC. Although K-SOS represents a ''strength'' method of remediation, it is not analogous to the "visual" versus "auditory" methods of remediation. Das et al. (1979) stated that it is not the content of the task (i.e., verbal versus nonverbal) that influences the mode of processing the child uses, but it is an interaction between the demands of the task and the experience of the learner that influences the utilization of either processing strategy. Thus, most cognitive tasks can be approached and solved using either simultaneous or sequential strategies. However, Caplan and Kinsboume (1981) indicated that some tasks can be adequately processed by either cognitive strategy, whereas others may require a specific approach for the most efficient or optimal performance. Kaufman et al. (1984) began with the premise that children who have similar simultaneous and sequential skills will learn when lessons are presented through either processing mode. However, children with poorly developed skills in one of the processing strategies may tend to solve a problem using their intact strategy. Kaufman et al. recommended that clinicians identify the processing demands of a task, and then modify or adapt these tasks based on the child's learning characteristics. Kaufman et al. provided a thorough description of the characteristics of the sequential and the simultaneous learner, and outlined possible techniques and methods for teaching specific subject areas.
Characteristics of the Sequential Leamer According to Kaufman et al. (1984, p. 6), "the sequential learner solves problems best by mentally arranging small amounts of information in consecutive, linear, step-by-step order.'' Sequential processing is evident in the following activities (Kaufman et al., 1984); (1) understanding, remembering, and using arithmetic facts; (2) learning to spell words; (3) matching sounds to letters; (4) understanding grammatical rules; (5) following the procedures of a science project; (6) remembering events in history; (7) problem-solving when a series of discrete steps are involved; and (8) following directions or rules. Children with strengths in sequential processing may approach tasks by breaking them down into smaller, meaningful units. For example, when learn-
REMEDIATION OF EDUCATIONAL DEFICITS
ing a new vocabulary word, sequential learners might associate the new word with one they already know; e.g., "houseboat" is made up of the words "house" and ··boat,'' and it must mean that a '·boat'' can be a "house" or a "house" can be a "boat." If a child has strengths in sequential processing with marked weaknesses in the simultaneous mode, Kaufman et al. (1984) suggested they might have trouble when: identifying sight vocabulary; comprehending paragraphs and sentences, especially when inferences are required; using arithmetic facts; and benefitting from visual aids, like graphs and charts.
Characteristics of the Simultaneous Learner The simultaneous learner evidences different processing strengths, and "solves problems best by integrating and synthesizing many parallel pieces of information at the same time" (Kaufman et al., 1984, p. 6). Simultaneous strategies are evident when: (1) identifying and discriminating shapes of letters and geometric forms; (2) recognizing the global meaning of a picture or chart; (3) comprehending the essence of a paragraph or story; (4) understanding math concepts; and (5) using visual images or aids to solve a problem. Using the example of the word "houseboat," the simultaneous learner might develop a mental image of a •'house'' moving across a lake or a house shaped like a boat. The child who has strengths in simultaneous processing, with marked weaknesses in sequential, may have difficulty: learning a phonic approach to reading; analyzing the sequence of steps in a math problem; recalling essential details of a picture or story; and following complex verbal directions (Kaufman et al., 1984). By analyzing the goals and objectives of an instructional exercise, the clinician can match methods of presentation with the learner's processing strengths. The child can then be taught strategies for approaching tasks or solving problems that capitalize on their abilities.
Guidelines and Activities for Classroom Instruction Kaufman et al. (1984) provided a number of useful guidelines for implementing remedial programs for children. Ideas were discussed for presenting materials and modifying instructions that best match the learning style of the sequential and the simultaneous learner. General instructional principles provide the clinician with a theoretical basis
367
upon which to design classroom activities that capitalize on the child's processing strengths. Specifically, Kaufman et al. described the following activities and principles. For the sequential learner, materials should be presented in segments, where the global skill or lesson objective is introduced in a step-by-step fashion. Questions should be interspersed throughout the segments to encourage and reinforce comprehension and understanding of the construct being taught. Teach the child to verbalize directions and to employ verbal memory strategies. New skills should be introduced and rehearsed in a series of steps. Once the new skill has been mastered, the steps can be used as an outline or logical structure for problem-solving. Simultaneous learners will benefit from instruction when presented in the following ways (Kaufman et al., 1984). Introduce the lesson objective immediately. Give the children an overview of the concept or skill prior to having them begin the task. Encourage the children to use visual imagery when they read and write. Visual aids, pictures, graphs, models, and concrete objects should be used whenever possible. Although the simultaneous learner can grasp the ''whole'' picture, details of the task or lesson need to be pointed out and discussed. For a list and description of numerous classroom activities for the sequential and the simultaneous learner, refer to the K-SOS manual (Kaufman et al., 1984).
Validation of the K-505 Approach Preliminary research with this cognitive remedial approach has been promising. Gunnison, Kaufman, and Kaufman (1982) reported that studies utilizing the serial and simultaneous processing theory have found almost perfect performance when instruction is matched to the learner's processing strengths. In two separate studies, Gunnison et al. (1982) matched remedial reading lessons to the children's simultaneous or sequential processing strengths as measured by the K-ABC. In both instances, results were positive as children showed substantial gains over traditional teaching methods and over methods that did not match the children's learning style. Although the theory has been criticized for dichotomizing complex mental processes, and the KABC has been questioned on psychometric grounds (Bracken, 1985; Goetz & Hall, 1984), the K-SOS remedial techniques should be more thoroughly investigated before they are dismissed. Rourke et al.
368
CHAPTER 19
(1983) highlighted some of the major impediments to conducting research with remedial programs, including: individual differences in the nature and severity of neuropsychological deficits make it difficult to assign groups of children to the same treatment program; evaluating the effectiveness of programs is difficult because many assessment instruments may be related to the techniques used in the treatment; and the treatment needs of children vary over time, which might mean modifying the original program. These same problems will influence research efforts to determine the effectiveness of intervention programs that follow any theory or principle of cognitive or neuropsychological functioning. Thus, the most significant contribution of the K-SOS may be that it provides the clinician with a theoretical basis upon which to design and test the effectiveness of remedial programs for children with cognitive deficits.
Remedial Techniques for Testing Brain-Related Deficits in Reading, Spelling, Mathematics, and Reasoning
been identified (Horton, 1981). Reinforcement, contracting, charting, and self-monitoring can be built into the child's treatment program to increase skills in specific areas. By integrating results from the neuropsychological assessment into a behavioral plan, methods can be devised that capitalize on the child's strengths while implementing operant procedures to increase motivation and learning (Reynolds, 1981). It is obviously beyond the scope of this chapter to provide a detailed presentation of strategies for each, or even one, of the achievement areas mentioned above. Therefore, the following examples of activities or programs will serve to illustrate how neuropsychological principles can be integrated into curricular approaches for the remediation of cognitive deficits. Many of the techniques that will be presented rely on the diagnosis of subtypes of learning disorders. Subtype classifications evolved out of research showing that learning-disabled children were not a homogeneous group. Hooper and Boyd (1986) cited 27 separate studies that investigated subtypes of reading disabilities alone. Although the investigation of group trends can be useful, the clinician should always do a careful study of the individual child prior to adopting general remedial techniques or methods. In the following sections, a number of general remedial programs will be discussed. The characteristics of subtypes will also be presented, with ideas for implementing specific techniques based on an analysis of the child's strengths and weaknesses in the different academic areas.
In the following sections a number of techniques for treating general deficits in reading, spelling, mathematics, and reasoning in children with brainrelated cognitive dysfunction will be presented. These techniques should be used only after a comprehensive assessment of the child's neuropsychological strengths and weaknesses. There is a normal developmental sequence of skills in each of these academic areas, so the clinician must do a careful task analysis of achievement levels to determine specific deficits for the individual child. The clinician should also Remedial Programs for Teaching Reading work closely with the child's teacher when deciding Skills which techniques and activities should be impleLerner (1981) described a number of reading mented. It is necessary to have a thorough underprograms for children with learning disorders, and standing of what has been tried in the classroom, and four of these programs will be briefly discussed; (I) how the child has responded to specific approaches and tasks. The clinical neuropsychologist and the the Language Experience Approach; (2) the Linteacher must work together and combine their areas guistic Approach; (3) the Fernald Method; and (4) the of expertise to plan specific classroom activities. The Orton-Gillingham Method. These techniques focus neuropsychologist can provide a theoretical frame- primarily on teaching decoding skills, and other techwork for understanding the child's deficits, and the niques should be employed for teaching reading comteacher can provide suggestions for which curricular prehension. See Lerner ( 1981) and Vallett ( 1974) for materials and methods would be best for instruction. a more in-depth treatment of the reading process. This interchange should not be limited to communication through reports, but should be done in plan- Language Experience Approach ning sessions so that ideas can be presented and acThe Language Experience method can be used tively challenged. Behavioral principles can also be applied in the for developing reading and other language arts abilitreatment program, to increase academic proficiency ties. Materials for this method are based on the once the child's neuropsychological deficits have child's own language and experiences (Lerner, 198 I;
REMEDIATION OF EDUCATIONAL DEFICITS
Hynd & Cohen, 1983). The child starts by dictating a story to the teacher. Reading and spelling lessons are developed directly from these stories. The basic assumptions for the child are: "What I can think about, I can talk about. What I can say, I can write (or someone can write for me). What I can write, I can read. I can read what others write for me to read" (Lerner, 1981, p. 299). This method has a high interest factor because the instructional materials are based on the child's own experience and creativity. Lerner (1981) indicated that this approach can be used successfully with young children just learning to read, and older children with a history of reading problems.
Linguistic Reading Approach The linguistic approach teaches children to read through a controlled introduction of words with consistent letter-sound relationships, and regular spelling rules (Lerner, 1981). This method relies heavily on the decoding process, where the phoneme-grapheme relationship is stressed. Reading materials are carefully selected, so that consonant-vowel-consonant pateros are presented in a uniform manner. For example, the child is taught that the short "a" sound by reading the words, "can, Nan, van, fan, Dan, pan, man, tan, ran" (Lerner, 1981, p. 302). The child is then taught to write sentences and paragraphs using these same words. This approach differs from other phonetic approaches in that the letter sounds are blended immediately into words with regular spelling patterns and not in isolation (Lerner, 1981). Initially the child learns to differentiate only the beginning sounds, because the middle and ending sounds are consistent.
Fernald's Visual-Auditory-Kinesthetic-Tactile Approach
(VAKT)
Fernald's VAKT method employs four sensory systems simultaneously, to teach children to read (Hynd & Cohen, 1983). Hynd and Cohen (1983) outlined the following steps to this reading program, where the child begins by writing a story and asking for unknown words. In the first stage, new words are printed in large letters on a piece of paper, and the child traces over the letters while saying the word aloud. This step is repeated until the child can write the word two times without looking at the paper. During this stage, clay or sand can also be used for tracing. In the second stage, the tactile (tracing) component is dropped, and the child employs only the
369
"look-say-write" steps. In stage three, the kinesthetic (writing) component is discontinued and the child simply looks and says the word. In the last stage, the child learns to recognize the similarity between word parts in new words and known words. Hynd and Cohen ( 1983) indicated that in the VAKT method, the child is never read to, and the child never sounds out new words phonetically.
The Orton-Gillingham Method The Orton-Gillingham approach also employs tactile and kinesthetic exercises for teaching reading to children who fail to learn through other methods that emphasize visual-perceptual techniques (Hynd & Cohen, 1983). This program differs from the Fernald method in that is stresses the auditory components of the reading process. Hynd and Cohen ( 1983) outlined the following steps in this method. First, the child is taught individual letters and sounds, and these are combined into letter groups using flash card presentations. Different vowel and consonant sounds are represented in different colors, and once blending is fluid, words are introduced. The child uses tracing techniques in the early steps of letter-sound identification (Lerner, 1981). At the word analysis stage, the teacher says the word and the child identifies the letter sounds. In the next phase, the teacher says the word, the child repeats the word, identifies the individual letters, and writes the word while repeating the word aloud. Phonetically consistent words are taught first, and once simple words (three letters) are mastered, the child uses these in sentences and short paragraphs. Hynd and Cohen (1983) indicated that children receive written material that has been carefully screened to ensure that words can be attacked phonetically. Lerner (1981) also reported that independent reading is curtailed until basic phonics are learned.
Specific Reading Approaches for Dyslexic Subgroups After a careful review of traditional remedial reading programs, Hynd and Cohen (1983) concluded that no one approach is best for all dyslexic children. Based on research findings with subtypes of dyslexia, Hynd and Cohen provided recommendations for which programs can be expected to be most effective for different groups. They analyzed the learning characteristics of dylexics using Boder' sand Mattis's classification systems. Boder's diagnostic categories include three major groups of dyslexia: ( 1)
370
CHAPTER 19
the dysphonetic; (2) the dyseidetic; and (3) the alexic (Boder, 1971). The dysphonetic dyslexic reader has difficulty with the phonetic analysis of words due to deficits in integrating sounds with their letter counterparts. These children rely on visual-perceptual cues when reading, and analyze words primarily based on the structure or configuration of the letters. Conversely, the dyseidetic dyslexic has intact phonetic analysis skills, with deficits primarily in the visual perception of words. When reading, the dyseidetic dyslexic can sound out words phonetically, and spelling errors are phonetically accurate. The alexic dyslexic child has deficits in both the auditory and visual analysis of words. These children have the most severe reading deficits because they do not appear to have an area of strength in the reading and decoding process. Mattis (1978) also identified three distinct subgroups of dyslexic children: (I) a language disorder group; (2) an articulatory and graphomotor disorder group; and (3) a visual-perceptual disorder group.
Remediation for Dysphonetic and Dyseidetic
Readers
Hynd and Cohen (1983) recommended several reading programs that could be used for each subgroup. The Orton-Gillingham and other phonetic reading programs (i.e., Distar and SRA) were recommended for the dyseidetic, the language disorder group (once letter recognition skills have been developed), and the visual-perceptual disorder group. The Fernald program was recommended for the alexic group, and for some dysphonetic readers. The LookSay and Language Experience methods were suggested for the dysphonetic, the language disorder, and the graphomotor disorder groups. Finally, the Linguistic Structural Analysis approach could be implemented in later stages for the dysphonetic and graphomotor disorder groups. Hynd and Cohen ( 1983) indicated that three remedial programs were not appropriate for any of the diagnostic groups: (1) the Doman-Delacato program; (2) the ITPA Psycholinguistic program; and (3) the Visual Perceptual programs of Kephart and Frostig.
Remediation Based on Kaufman's Theory Kaufman et al. ( 1984) suggested that sequential and simultaneous learners can be taught phonics through different exercises. For example, the following activities would be appropriate for the sequential learner. The teacher selects a sound and finds a char-
acter or name with that sound. The teacher then writes the letter on the board and says the sound while the child points to the letter on the board. Then several other letters are printed on the board and the child identifies the target sound. Then the teacher writes different letters, and while the teacher says each sound, the child claps when he or she hears the target sound. This same procedure is repeated with the teacher's lips covered so that the child cannot rely on lip-reading to identify the target sound. This procedure emphasizes recall and auditory discriminations. Kaufman et al. modified the phonic lesson for the simultaneous learner in the following ways. The teacher writes a letter on a piece of paper and the child traces the letter. The child is taught to visualize the shape of the letter, and cues about the letter shape are provided. For example, the letter "m" looks like a ''roller coaster, mountains, waves, inchworm, bunny ears" (Kaufman eta/., 1984, p. 5). Each time the child hears the sound, he or she traces the letter. Individual letters are then combined, and the child is taught to find them in words. For example, the "at" sound-letter relationship is taught and the child finds "at" in the word "cat." Then other words like "fat, cat, sat, and mat" are introduced, and the child tells how these words are the same and how they are different (Kaufman et al., 1984, p. 5). The major emphasis for the simultaneous learner is on teaching the shapes of letters, and associating the sounds with the physical characteristics of the letter.
Remediation Techniques for Spelling
Deficits
Generally, spelling deficits are associated with reading disabilities, but this relationship is not a perfect one. Occasionally, children with normal decoding and comprehension skills have trouble spelling, but children with decoding deficits almost always are poor spellers (Lerner, 1981). Recently, Sweeney and Rourke ( 1985) investigated the differences in psycholinguistic abilities in children who are poor spellers. They compared children with spelling deficits who are phonetically accurate to others who are phonetically inaccurate in spelling. Sweeney and Rourke did not fmd significant differences in the Wechsler Performance IQ, spelling levels on the Wide Range Achievement Test (WRAT), scores on a test of auditory discrimination, scores on the Wechsler Digit Span subtest, or scores on a visual closure test. Difficulties appeared to be most significant for the phonetically inaccurate spellers on tasks involving memory for sentences, sound blending, and
REMEDIATION OF EDUCATIONAL DEFICITS
memory for digits reversed (Sweeney & Rourke, 1985). This subgroup displayed significant deficiencies on complex receptive language tasks, similar to patients with cerebral dysfunction in the secondary zones of the temporal lobe (Sweeney & Rourke, 1985). These deficiencies were found to be more pervasive than were deficits found in the phonetically accurate group. The phonetically accurate spelling group showed difficulty when analyzing visual-spatial information, and showed patterns of deficits similar to those in individuals with cerebral dysfunction to the tertiary zone where the parietal-occipitaltemporal lobes converge (Sweeney & Rourke, 1985). Both groups did not perform as well as normal spellers on the Information, Comprehension, and Vocabulary subtests of the Wechsler scale. The research of Sweeney and Rourke (1985) provides evidence that spelling deficits are not a unitary disorder, and that subgroups exist with different deficiencies. Although these subgroups may appear similar in many academic areas, the treatment approaches to remediating spelling deficits would differ (Sweeney & Rourke, 1985). The following activities are recommended for remediating spelling deficits for these two subgroups.
Activities for Phonetically Accurate Spellers
371
involving the introduction of individual letters and sounds, then progressing to segmenting the sound composition of words. A sight-word approach to reading is recommended (Sweeney & Rourke, 1985), and it seems likely that this approach could be extended to spelling as well. Lerner (I 981) described a spelling approach that involves teaching children from "frequency-of-use" word lists. AU of the studies analyzing frequency-of-use words suggest that 60% of our writing is composed of only 100 words (Lerner, 1981). See Lerner ( 1981) for this frequencyof-use word list. This approach could be tried if the phonetically inaccurate speller fails to learn through other techniques. Another method for teaching spelling to older children is the test-teach-test approach. This method differs from traditional approaches, which adopt a teach-test-teach paradigm. Spelling words can be generated from traditional spelling workbooks or graded frequency lists. The child is initially tested on a list of words, and then studies only those words he or she is unable to spell. This reduces the potential of having the children study words they already can spell. Lerner (1981) also recommended the use of crossword puzzles and games like Scrabble to increase spelling. It should be noted that children with significant visual-perceptual disorders may have trouble with crosswords.
Sweeney and Rourke ( 1985) suggested that phonetically accurate spellers would benefit from strat- Remediation of Mathematics Deficits egies that emphasize ''right hemisphere strategies,'' Math computation is a complex academic skill such as: ( l) the use of sight-word reading approaches; (2) the identification of visual-spatial features of that involves numerous operations to solve even a words; and (3) the use of imagery techniques to men- single basic problem. Strang and Rourke (1985) sugtally picture the shapes of the words. Other tech- gested that there are as many as 33 steps involved in niques might include tactile-kinesthetic components, calculating a two-digit multiplication problem (i.e., where the spelling words are repeatedly traced, pro- 62 x 96 = ). Also, a simple error can be costly in nounced, spelled aloud, and written. For older chil- calculation problems, where similar errors in reading dren, new spelling words could be introduced using (i.e., misreading a word in a sentence) will not necesthe child's own stories, as explained in the language sarily interfere with the overall comprehension of a sentence (Strang & Rourke, 1985). However, cogexperience approach for reading. nitive deficits have been associated with math deficits, including reading disabilities and visual-memory disorders (Strang & Rourke, 1985). Other Activities for Phonetically Inaccurate Spellers noncognitive factors can also interfere with math perSweeney and Rourke ( 1985) were more guarded formance, such as attentional, motivational, and in their prognosis for improvement of phonetically emotional disturbances (Strang & Rourke, 1985). inaccurate spellers because these individuals tend to Strang and Rourke (1985) identified subtypes show more generalized linguistic deficits. For this of children with arithmetic disabilities, including: reason, they suggested that this group will probably (I) children with a pattern of deficits suggestive of only learn information that is "extremely elemen- left hemisphere dysfunction; and (2) children with a tary, overlearned'' and involves ''redundant oral dis- pattern of deficits suggestive of right hemisphere course" (Sweeney & Rourke, 1985, p. 163). They dysfunction. Of these two groups, the ''right hemirecommended that remediation be highly repetitive sphere'' group showed the most severe pattern of
372
CHAPTER 19
neuropsychological deficits, including: visual-perceptual, visual-organization, tactile-perceptual, psychomotor, and conceptual disorders (Strang & Rourke, 1985). The "left hemisphere" group showed intact skills on the right hemisphere tasks, but demonstrated deficits on verbal and auditory-perceptual tasks. Error patterns for the "right hemisphere" group were pronounced in the following categories: organizing spatial information; attending to visual detail; following procedures; shifting mental sets; writing numbers; remembering numerical facts; and reasoning (Strang & Rourke, 1985). Although the two groups did not differ in terms of math achievement scores, they demonstrated considerably different underlying neuropsychological deficiencies, which impact significantly on the type of intervention strategies that should be utilized.
Strategies for the "Right Hemisphere" Group In a comparative sense, these children have more intact verbal abilities that can be utilized in the math area. Strang and Rourke (1985) suggested the following strategies for this group. (1) Start math remediation by selecting a simple calculation problem where the child has difficulty. (2) Provide a detailed verbal explanation of the steps involved in the calculation process .. Each step should be described, then placed in context of the entire or ''whole'' operation. The child should be taught to verbalize the necessary procedures. (3) The teacher then presents the steps verbally. For example, in the first step you must name the mathematical sign in the problem; in the second step you must look at and move your pencil to the right side of the problem, and so on. (4) Have the child describe the steps involved in the operation. (5) Have the child write the "rules" or steps, and use these when solving other problems. (6) Once the child can progress through the first five steps, the visual component can be added. Have the child verbally direct the teacher through the problem. (7) Concrete aids can be introduced at this stage. Strang and Rourke ( 1985) indicated that this might be one of the most difficult stages of the remediation. (8) Trial problems can be given at this stage. The child may use graph paper to reduce errors due to number and column placement. (9) Graph paper may need to be color coded to highlight the right and left sides of the paper. ( 10) Have the child read the problem aloud before calculating the answer. Once this is mastered, have the child read the question and the answer before starting the next problem. ( 11) Teach the child to use a hand calculator to check for accuracy. The
problem should be recalculated (by hand) if the answer is not correct. (12) The teacher should keep a record of the errors, task analyze these errors, and modify the remedial plan if necessary. (13) Relate math concepts to something relevant or practical to the child. That is, computation skills can be related to shopping. See Strang and Rourke (1985) for a more thorough discussion of these steps. There are other strategies that may prove to be helpful for the "right hemisphere" group. The child can be taught to make verbal associations relating math concepts. For example, the child can be shown the relationship between multiplication and division by the following elaboration: Multiplication 3X3=9
is related to isthesameas
Addition: 3+3+3=9
Multiplication 3X3=9
is related to is the same as
Division: 9/3 = 3
Other mathematical computations can be explained by using verbal analogies or verbal associations. For example, in a multiplication problem involving two double-digit numbers: 25 X 20
This problem can be solved in several ways. 1. 25 X 20 = 500 2. 25 added 20 times
00 +50 500
25 + 25 + 25 + 25 + 25 25+25+25+25+25 25 + 25 + 25 + 25 + 25 25 + 25 + 25 + 25 + 25 = 500
Shaded areas could be used to demarcate the columns in the above example. Also this last example should be utilized only after the child has successfully progressed through the 13 steps outlined by Strang and Rourke ( 1985), and has achieved some initial success in basic computational skills in each of the number operations involved in the verbal analogy.
Strategies for "Left Hemisphere" Group The "left hemisphere" group has relative strengths in visual-perception and visual-organization, so these abilities can be emphasized in the following ways. Provide the child with visual or physical aids whenever possible. Lerner ( 1981) suggested using concrete objects, such as: movable disks that the child can manipulate; a "rectangle array" con-
REMEDIATION OF EDUCATIONAL DEFICITS
373
taining rows and columns of ovals or squares; Michenbaum ( 1977) developed a cognitive-benumber lines so the child can see the relationship havioral training program to teach children appropribetween numbers when adding, subtracting, multi- ate verbal mediators to increase planning skills. The plying, and dividing; flash cards; and addition and program has several steps wherein the child is taught multiplication charts or wheels. Vallett (1974) pro- to: ( 1) define the problem, ''What is it I have to do?''; vided an example of an activity using both the "ar- (2) focus attention to the task; (3) provide self-reinray" and the number line for addition and forcement, "Good, I'm doing fine"; and (4) selfsubtraction. evaluate and self-correct, "That's okay .... Even if I make an error I can go on slowly" (Michenbaum & ()() ()() ()()() Burland, 1979, p. 427). Other problem-solving techniques can be used 0 1 2 3 4 5 6 to improve effectiveness and flexibility in planning and decision-making strategies. D'Zurilla and Gold()() ()()() 3+2= fried ( 1971) developed a method for systematically ~ teaching problem-solving skills. ''Means-end'' or consequential thinking can be taught in structured 4 + = 6 activities by discussing a situation with the child that has been problematic. Have the child generate as 0 1 2 3 4 5 6 many solutions to the problem as he or she can think 6 of, regardless of the effectiveness of the solution. + 2 Then have the child try the solution to determine if it 6-3= is effective. Once a set of effective strategies have 6 2 been developed, then discuss what skills (or mate2 4 rials) are necessary for implementing a particular solution. If the child needs skill training, such as selfLerner ( 1981) provided a general timetable for assertiveness, then this should be the next step. Finalwhen basic math computational skills are introduced ly, the child should practice these skills in classroom in the elementary grades. The clinician should be situations. familiar with these sequences when planning remedial programs for math deficits. See Lerner ( 1981) and Vallett (1974) for further details on other math Concluding Remarks skills. The primary purpose of this chapter was to discuss neuropsychological principles and approaches to remediating brain-related cognitive deficits in children. Factors that affect outcome and prognosis were Children with neuropsychological and cognitive also presented, including neurodevelopmental isdeficits often have associated deficiencies in reason- sues, medical problems, and psychological disoring and problem-solving skills. These deficits are ders. These topics were related to the severity of often overlooked in remedial programs because rea- deficits and the neuropsychological implications of soning skills are seldom taught as a separate skill. dysfunction to determine the course and nature of the When children cannot organize their work, cannot treatment program. Several neuropsychological and decide which is the best solution to a problem, or neurocognitive remedial programs were also recannot change mental sets to meet different situa- viewed. Finally, specific procedures and techniques tions, they generally fall behind in assignments and were outlined for treating cognitive deficits in chilthey can develop secondary behavioral problems. If dren with brain dysfunction. The successful treatthese reasoning deficits persist, these children may ment of cognitive deficits is predicated on a valid and have trouble learning basic concepts. Reitan (1980) comprehensive assessment of the child's neuropsyviewed the reasoning process as one of the most chological and academic functioning, and must be important prerequisites for other cognitive learning. the first step of any remedial program. The REHABlT program provides a number of acAll of the specific programs and techniques intivities designed to remediate weaknesses in this volved the identification of individual strengths and area. weaknesses, and many were based on the diagnosis
Remedial Strategies for Reasoning and Planning Deficits
374
CHAPTER 19
of subtypes of cogmt1ve deficits. Procedures for Boll, T. J., & Barth, J. T. (1981). Neuropsychology of brain damage in children. InS. Filskov & T. J. Boll (Eds.), Handidentifying and treating subclassifications of readbook of clinical neuropsychology. New York: Wileying, spelling, and math disabilities have increased Interscience. considerably in the past 5 years. However, issues of J. F. (1986). Epilepsy in children: Neuropsychological Bolter, subtyping are far from being resolved in the litereffects. In J. E. Obrzut & G. Hynd (Eds.), Child neuropsyature, and problems of assessment and differentiation chology (Vol. 2). New York: Academic Press. are still unresolved. Although further investigation is Br11cken, B. (1985). A critical review of the Kaufman Assessment needed to substantiate and validate these procedures · Battery for Children (K-ABC). School Psychology Review, and techniques, there is a substantial theoretical basis 14, 21-36. upon which to guide research endeavors. There is a Caplan, B., & Kinsboume, M. (1981). Cerebral lateralization, general need for empirical studies that test specific preferred cognitive mode, and reading ability in normal children. Brain and Language, 14, 349-370. remedial techniques with homogeneous brain-injured groups. Despite the difficulty in controlling for Chevrie, J. J., & Aicardi, J. (1978). Convulsive disorders in the frrst year of life: Neurological and mental outcome and mornumerous variables that influence treatment outtality. Epilepsia, 19, 67-74. comes, the field of child neuropsychology is in dire Cohen, G. (1973). Hemispheric differences in serial versus paralneed of such a data base.
ACKNOWLEDGMENT
The preparation of this chapter was supported by U.S. Department of Education Grant G-00-8302986.
References Alajouanine, T., & Lhennitte, F. (1965). Acquired aphasia in children. Brain, 88, 653-662. Ayres, A. J. (1974). Sensory integration and learning disorders. Los Angeles: Western Psychological Services. Barkley, R. A. (1977). The effects of methylphenidate on various types of activity level and attention in hyperkinetic children. Journal of Abnormal Child Psychology, 5, 351-369. Barkley, R. A. (1982). Specific guidelines for defining hyperactivity in children (attention deficit disorder). In B. Lahey & A. Kazdin (Eds.), Advances in clinical child psychology. New York: Plenum Press. Barkley, R. A., & Cunningham, C. E. (1979). The parent-child interactions of hyperactive children and their modification by stim~lant drugs. In R. Knights & D. Bakker (Eds.), Treatment of hyperactive and learning disordered children. Baltimore: University Park Press. Basser, L. (1962). Hemiplegia of early onset and the faculty of speech with reference to the effects of hemispherectomy. Brain, 85, 427-460. Berg, R. A. (1986). Neuropsychological effects of closed-head injury in children. In J. E. Obrzut & G. Hynd (Eds.), Child neuropsychology (Vol. 2). New York: Academic Press. Boder, E. (1971). Developmental dyslexia: Prevailing diagnostic concepts and a new diagnostic approach. In H. Myklebust (Ed.), Progress in learning disabilities. New York: Grone & Stratton. Boll, T. J. (1974). Behavioral correlates of cerebral damage in children aged 9-14. In R. M. Reitan & L. Davison (Eds.), Clinical neuropsychology: Cu"ent status and applications. Washington, DC: Winston.
lel processing. Journal of Experimental Psychology, 97, 349-356. Corbett, J. A., & Trimble, M. R. (1983). Epilepsy and anticonvulsant medication. In M. Rutter (Ed.), Developmental neuropsychiatry. New York: Guilford Press. Das, J.P., Kirby, J. R., & Jarman, R. F. (1979). Simultaneous and successive processes. New York: Academic Press. Douglas, V. I. (1983). Attentional and cognitive deficits. In M. Rutter (Ed.), Developmental neuropsychiatry. New York: Guilford Press. Douglas, V. 1., & Peters, K. G. (1979). Toward a clearer definition of the attentional deficit of hyperactive children. In G. A. Hale & M. Lewis (Eds.), Attention and the development of cognitive skills. New York: Plenum Press. D'Zurilla, T. J., & Goldfried, M. R. (1971). Problem solving and behavior modification. Journal ofAbnormal Psychology, 78, 107-126. Frostig, M., & Home, D. (1964). The Frostig program for the development of visual perception. Chicago: Follett. Gaddes, W. H. ( 1980). Learning disabilities and brain function: A neuropsychological approach. New York: Springer-Verlag. Gittelman, R., Abikoff, H., Klein, D. F., & Mattes, J. A. (1979). A controlled trial of behavior modification and methylphenidate in hyperactive children. Paper presented at the annual meeting of the American Academy of P~ychiatry, Atlanta. Goetz, E. T., & Hall, R. J. (1984). Evaluation of the Kaufman Assessment Battery for Children from an information-processing perspective. Journal of Special Education. 48(3), 281-294. Golden, C. J., & Wilkening, G. N. (1986). Neuropsychological basis of exceptionality. In R. Brown & C. Reynolds (Eds.), Psychological perspectives on childhood exceptionality. New York: Wiley-Interscience. Goodman, L., & Hammill, D. (1973). The effectiveness of the Kephart-Getman activities in developing conceptual-motor and cognitive skills. Focus on Exceptional Children, 4, 1-9. Gunnison, J. A., Kaufman, N. L., & Kaufman, A. S. (1982). Sequential and simultaneous processing applied to remediation. Academic Therapy, 17, 297-307. Hammill, D., & Larsen, S.C. (1974). The effectiveness of psycholinguistic training. Exceptional Children, 4/,5-15. Hartlage, L. C., & Reynolds, C. R. (1981). Neuropsychological
REMEDIATION OF EDUCATIONAL DEFIOTS assessment and the individualization of instruction. In G. W. Hynd & J. E. Obrzut (Eds.), Neuropsyclwlogical assessment and the school-age child: Issues and procedures. New York: Grune & Stratton. Hecaen, H. (1976). Acquired aphasia in children and the ontogenesis of hemispheric specialization. Brain and Language, 3, ll4--134. Hooper, S. R., & Boyd, T. A. (1986). Neurodevelopmentallearningdisorders. In J. E. Obrzut & G. Hynd (Eds.), Child neuropsychology (Vol. 2). New York: Academic Press. Horton, A. M. (1981). Behavioral neuropsychology in the schools. School Psychology Review, 10(3), 367-372. Hughes, J. (1%8). Electroencephalography and learning. In H. Myklebust (Ed.), Progress in learning disabilities. New York: Grune & Stratton. Hynd, C. R. (1986). Educational interventions in children with developmental learning disorders. In J. E. & G. W. Hynd (Eds.), Child neuropsychology. (Vol. 2). New York: Academic Press. Hynd, G. W., & Cohen, M. (1983). Dyslexia: Neuropsychological theory, research, and clinical differentiation. New York: Grune & Stratton. Kaufman, A. S., &Kaufman, N. L. (1983). Kai¢nanAssessment Battery for Children: Interpretive manual. Circle Pines, MN: American Guidance Service. Kaufman, A. S., Kaufman, N. L., & Goldsmith, B. Z. (1984). KSOS: Kaufman sequential or simultaneous. Circle Pines, MN: American Guidance Service. Kephart, N. C. (1971). The slow learner in the classroom (2nd ed.). Columbus, OH: Merrill. Kirk, S. A., & Kirk, W. D. (1971). Psycholinguistic learning disabilities: Diagnosis and remediation. Urbana: University of Illinois Press. Kleck, R. (1968). Self-disclosure patterns amang epileptics. Hanover, NH: Dartmouth University Press. Kolb, B., & Whishaw, I. Q. (1980). Fundamentals of human neuropsychology. San Francisco: Freeman. Lambert, N. M., & Sandoval, J. (1980). The prevalence oflearning disabilities in a sample of children considered hyperactive. Journal of Abnormal Child Psychology, 48, 566-574. Lerner, J. (1981). Learning disabilities: Theories, diagnosis, and teaching strategies (3rd ed.). Boston: Houghton Mifflin. Long, C. G., & Moore, J. R. (1979). Parental expectations of their epileptic children. Journal of Child Psychology and Psychiatry, 20, 299-312. Luria, A. R. (1963). Restoration of function after brain injury. New York: Macmillan Co. Luria, A. R. ( 1966). Higher conical functions in man. New York: Basic Books. Mattis, S. ( 1978). Dyslexia syndromes: A working hypothesis that works. In A. L. Benton & D. Pearl (Eds.), Dyslexia: An appraisal of current knowledge. New York: Oxford University Press. Michenbaum, D. ( 1977). Cognitive-behavior modification: An integrative approach. New York: Plenum Press. Michenbaum, D., & Burland, S. (1979). Cognitive behavior modification with children. School Psychology Digest, 8(4 ), 426433. Milner, B. (1975). Psychological aspects offocal epilepsy and its
375
neurological management. Advances in Neurology, 8, 299321. Rapoport, J. L. (1983). The use of drugs: Trends in research. In M. Rutter (Ed.), Developmental neuropsychiatry. New York: Guilford Press. Reitan, R. M. (1974). Psychological effects of cerebral lesions in children of early school age. In R. M. Reitan & L. Davison (Eds.), Clinical neuropsychology: Current status and applications. Washington, DC: Winston. Reitan, R. M. (1980). REHABIT-Reitan evaluation of hemispheric abilities and brain improvement training. Tucson: Reitan Neuropsychology Laboratory and University of Arizona. Reitan, R. M. (1981). Effectsofage-of-onsetofbraindamageon later development. Presented at the Reitan Neuropsychological Workshop, Chicago. Reitan, R. M., & Davison, L. (Eds.). (1974). Clinical neuropsychology: Current status and applications. Washington, DC: Winston. Reynolds, C. R. (1981). Neuropsychological assessment and the habilitation of learning: Considerations in the search for the aptitude x treatment interaction. School Psychology Review, 10(3), 343-349. Rourke, B. (1985). Neuropsychology of learning disabilities: Essentials of subtype analysis. New York: Guilford Press. Rourke, B., Bakker, D. J., Fisk, J. L., & Strang, J. D. (1983). Child neuropsychology: An introduction to theory, research, and clinical practice. New York: Guilford Press. Rutter, M. (1983). Developmental neuropsychiatry. New York: Guilford Press. Rutter, M., Graham, P., & Yule, W. (1970). Clinics and developmental medicine. A Neuropsychiatric Study in Childhood. London: Spastics Medical. Rutter, M., Tizard, J., & Whitmore, M. (Eds.). (1970). Education, health and behavior. London: Longman. Rutter, M., Chadwick, 0., & Shaffer, D. (1983). Head injury. In M. Rutter (Ed.), Developmental neuropsychiatry (pp. 83111). New York: Guilford Press. Salvia, J., & Ysseldyke, J. (1985). Assessment in special and remedial education (3rd ed.). Boston: Houghton Mifflin. Selz, M. (1981). Halstead-Reitan neuropsychological test batteries for children. In G. W. Hynd & J. E. Obrzut (Eds.), Neuropsychological assessment and the school-age child: Issues and procedures. New York: Grune & Stratton. Selz. M., & Reitan, R. M. (1979). Rules for neuropsychological diagnosis: Classification of brain function in older children. Journal of Consulting and Clinical Psychology, 47, 258-
264.
Stores, G., Hart, J., & Piran, N. (1978). On attentiveness in school children with epilepsy. Epilepsia, 19, 169-175. Strang, J.D. & Rourke, B. P. (1985). Arithmetic disability subtypes: The neuropsychological significance of specific arithmetical impairment in childhood. In B. P. Rourke (Ed.), Neuropsychology of learning disabilities: Essentials of subtype analysis. New York: Guilford Press. Sweeney, J. E., & Rourke, B. P. (1985). Spelling disability subtypes. In B. P. Rourke (Ed.), Neuropsychology of learning disabilities: Essentials of subtype analysis. New York: Guilford Press.
376
CHAPTER 19
Taylor, E. (1983). Drug response and diagnostic validation. In M. Rutter (Ed.), Developmental neuropsychiatry. New York: Guilford Press. Teuber, H. L., & Rudel, R. G. (1962). Behavior after cerebral lesions in children and adults. Developmental Medicine and Child Neurology, 4, 3-20. Tymchuk, A. J., Knights, R. M., & Hinton, G. C. (1970). Neuropsychological test results of children with brain lesions, abnormal EEG's and normal EEG's. Canadian Journal of Behavioral Science, 2, 322-329. Vallett, R. E. (1974). The remediation of learning disabilities: A handbook of psychoeducational resource programs (2nd ed.). Belmost, CA: Fearon Publications. Whalen, C. K., Henker, B., Collins, B. E., McAuliffe, S., &
Vaux, A. (1979). Peer interaction in a structured communication task: Comparisons of normal and hyperactive boys and of methylphenidate (Ritalin) and placebo effects. Child Development, 50, 388-401. Wilkening, G. N., & Golden, C. J. (1982). Pediatric neuropsychology: Status, research and theory. In P. Karoly, J. Steffen, & J. O'Grady (Eds.), Child health psychology: Concepts and issues. Elmsford, NY: Pergamon Press. Witelson, S. F., & Pallie, W. (1973). Left hemisphere specialization for language in the newborn: Neuroanatomical evidence of asymmetry. Brain, 96, 646-671. Zarske, J. (1982). Neuropsychological intervention approaches for handicapped children. Journal ofResearch and Development in Education, 15, 66-75.
20 The Biofeedback Treatment of Neurologica l and Neuropsych ological Disorders of Childhood and Adolescence ROBERT L. HODES
Biofeedback is one of several bemavioral treatments designed to increase an individual's self-regulation of physiology. It employs instrumentation to provide patients with both immediate and precise information about otherwise occult physiological processes. In clinical settings, biofeedback is typically used in conjunction with other behavioral or medical interventions to reduce the frequency of distressing symptoms or to minimize physical impairments. The mechanisms underlying biofeedback's clinical efficacy are unclear. Different theorists have advanced varying explanations including operant conditioning of discrete physiological responses (Miller, 1969), the learning of a generalized relaxation response (Silver & Blanchard, 1978), and the production of cognitive changes promoting an increased sense of selfcontrol and self-efficacy (Turk, Meichenbaum, & Berman, 1979; Holroyd et al.. 1984). Despite these different viewpoints, agreement exists that motivated individuals are able to use biofeedback signals to learn voluntary control over a variety of physiological parameters. The popular press has made "biofeedback" almost a household word and the technique has been offered as a potential cure for a myriad of ills of the mind and body. Although many of the early claims were unfounded, biofeedback has earned a well-deserved reputation as an effective treatment for several chronic pain syndromes, psychophysiological comROBERT L. HODES • Depanment of Neurology, University of Wisconsin, Madison, Wisconsin 53792.
plaints, and neuromuscular disorders. In addition, many contemporary clinical applications of biofeedback assume that the self-regulated modification of physiology leads not only to a reduction in physical symptoms, but also to useful behavioral and cognitive changes. For example, the use of EMG biofeedback for children with attention deficit disorders assumes that biofeedback prompts not only a reduction in muscle tension, but also reduces disruptive classroom behavior, improves academic performance, and increases the child's sense of self-control and self-esteem (Braud, 1978; Omizo, 1980a).
Biofeedback with Disorders of Childhood and Adolescence This chapter reviews the published work on biofeedback training in neurological and neuropsychological disorders of childhood and adolescence. More general reviews of the application of biofeedback in pediatric medicine are provided by Finley (1983) and Andrasik and Attanasio (1985). In the broadest sense, all biofeedback learning studies deal with the modification of neurally controlled phenomena. This chapter is limited to a review of the application of biofeedback technology to the symptoms of neurological disease or neuropsychological dysfunction. In some of these studies, the evidence of a neurological lesion is unequivocal; in others, such evidence is presumptive (Cleeland, 1981). Specific disorders considered include cerebral palsy. epilep377
378
CHAPTER 20
sy, neurogenic fecal incontinence, attention deficit disorder with hyperactivity, learning disabilities, and migraine headaches. When available, controlled group treatment outcome studies are given emphasis over both controlled and uncontrolled case studies. For each disorder, the literature review emphasizes four basic points relevant to the clinical use of biofeedback. These issues are: 1. Are patients able to learn the desired response? 2. Does physiological self-regulation lead to other clinically relevant changes in either physiology, behavior, or cognition? 3. How effective is biofeedback relative to either control or alternative treatment procedures? 4. Are therapeutic gains maintained over time? The vast majority of applications of biofeedback have been with adults. Many of the techniques and insights generated by that literature are directly applicable when working with children. For example, the treatment of migraine headaches in children is closely modeled after treatment protocols used with adults (Labbe & Williamson, 1984). However, biofeedback training with children does require attention to several issues, such as varying developmental levels, not typically addressed when working with adult clients. The relevancy of these issues to the literature reviewed in this chapter will also be discussed. Finally, this chapter concludes with comments summarizing the current state of knowledge and offering suggestions for future inquiries.
Cerebral Palsy Cerebral palsy refers to a group of nonprogressive motor disorders resulting from damage to the central nervous system during prenatal or perinatal periods. Several types of movement impairments may be seen including spasticity, involuntary movements, and incoordination. Although not invariably present, associated symptoms may include sensory abnormalities and mental deficiency. Cerebral palsy results from multiple etiologies and occurs with an estimated incidence of between 0.1 and 0.2% in children (Berkow, 1977). Of the various abnormalities of muscle tone and movement found in cerebral palsy, spasticity and rigidity have received the most attention from biofeedback clinicians. The essential feature of spasticity is an increased reactivity to an exciting stim-
ulus, most notably rapid stretch. Clinically, this is observed by asking the patient to relax and then to passively extend a limb. In normal muscles, no muscular resistance is observed. Spastic muscles, in contrast, reflexively contract to the attempted extension and resist the clinician's efforts (DeBacher, 1983). The degree of resistance is often determined by the velocity of the passive movement and the position of the limb. In rigidity, muscles are almost continuously hypertonic. In contrast to spastic muscles, rigid muscles resist movement throughout their entire range of motion and at slow movement velocities (Adams & Victor, 1977).
Biofeedback Treatment of Movement Disorders in Cerebral Palsy As defined above, spasticity and rigidity both involve hypertonicity of skeletal muscles. Increase in the severity of involuntary movements as a result of environmental stimulation and emotional distress is also well known. Based on earlier work with progressive muscle relaxation training in patients with cerebral palsy, Finley, Niman, Standley, and Wansley (1977) employed frontal region, EMG biofeedback to produce a generalized state of muscle relaxation and associated emotional calm. The effect of this intervention on motor and speech behavior of four children with cerebral palsy was analyzed employing an ABAB design. All children were able to successfully reduce frontal EMG after an initial 12 training sessions. EMG activity increased during the first reversal phase but only one of the four children was able to reinstitute control during the second treatment period. Similar results were obtained for forearm flexor EMG, where a reduction was observed during the initial training phase, but increases in flexor tension were found during the second training phase. The results were more encouraging for speech and motor skills. All children manifested statistically significant improvement in speech and/ or motor measures during initial training. These gains deteriorated during the reversal phase and were reinstituted during the second training phase. Although statistical significance was demonstrated, judgments concerning the clinical significance of performance gains were not provided by the authors. In a follow-up study, Finley, Etherton, Dickmen, Karimian, and Simpson (1981) studied 15 cerebral palsied children who primarily manifested spasticity. In this study, the biofeedback was contingent upon the combined EMG activity from frontal region and forearm flexor sites. These children also received
BIOFEEDBACK
tangible rewards (e.g., candy, toys) for meeting performance standards involving reductions in both tonic and phasic EMG. For some children, rewards were immediate, whereas for other children, rewards were given at the end of the training session. The results indicated that children in both groups learned to reduce their combined EMG, but larger magnitude reductions were found for children receiving immediate reinforcement for performance changes. Data on functional changes in motor behavior were not reported. Cataldo, Bird, and Cunningham (1978) employed EMG biofeedback to increase neuromuscular control in two children with choreoathetoid cerebral palsy. The first child was given feedback from the right biceps with the effects of biofeedback evaluated with an ABA design. The results suggested that the subject was able to produce muscle relaxation with contingent feedback, and that control deteriorated during the no feedback phase, but was reinitiated during the final feedback sessions. Although functional motor control was not formally assessed, the authors noted that hospital staff indicated no noticeable changes in adaptive functioning. For the second child, feedback was given from multiple muscle groups using a multiple baseline across-behaviors design. The data presented describe appropriate increases or decreases in muscle control that follow the intent of the study design. In addition, muscle control generalized to no feedback conditions. Finally, the authors reported that anecdotal observations indicated functional improvements in some muscle groups. In summary, only preliminary data exist on the utility of EMG biofeedback in increasing neuromuscular control in children with cerebral palsy. These initial data suggest that patients are able to learn to reduce undesirable levels of muscle tension. However, no study has demonstrated continued selfregulation of muscle activity during long-term follow-up. In this regard, Seeger and Caudrey (1983) noted that therapeutic gains in gait training made with sensory feedback from a load-sensitive insole were not maintained after a follow-up period of 18-24 months. The available data on the impact of enhanced muscular control on functional activity have been mixed. Greater use of reliable observation methods (e.g., Bird & Cataldo, 1978) or automated recording devices of functional activity (e.g., Seeger, Caudrey, & Scholes, 1981) is clearly indicated. Finally, it is revealing to compare this early literature on the biofeedback treatment of cerebral palsy with the more extensive literature on the treatment of adult patients with spasticity (e.g., spasticity fol-
379
lowing strokes). The training of neuromuscular control in cerebral palsied children has typically taken place while the affected extremity or joint is at rest. This is somewhat surprising given the definition of spasticity as hypertonicity produced by stretch. In greater accordance with this physiological fact, clinicians attempting to reduce spasticity in adult patients (e.g., DeBacher, 1983) typically train muscle control not only while the muscle is at rest, but also during passive and active movement of the muscle and during activation of antagonistic muscle groups. The application of these more sophisticated training procedures with cerebral palsied children is indicated.
Epilepsy Epilepsy refers to a group of related syndromes characterized by repeated paroxysmal disturbances in brain function. The resulting seizures may have one or more clinical manifestations including impairments of consciousness, sensory or motor abnormalities, or disturbances in cognition or emotional functioning (Dodrill, 1981). Seizure disorders can have a variety of etiologies that are too numerous to detail in this review. Although epilepsy may appear at any age, it most commonly emerges relatively early in life. It has been estimated that 64% of epileptics first develop seizures by the end of elementary school and 77% by the end of adolescence. The incidence of epilepsy has been estimated as 1-2% of the population in the United States (Epilepsy Foundation of America, 1975). Treatment of epilepsy is primarily pharmacological with anticonwlsant medications providing varying degrees of seizure control for 70 to 80% of children (Johnston & Freeman, 1981). For a small percentage of patients, surgical excision of abnormally active brain tissue provides a second treatment alternative. Therefore, despite the many advances in medical management, approximately 20% of individuals with epilepsy have inadequate control of seizures.
EEG Biofeedback Treatment of Seizures Feedback training for various parameters of EEG activity has been used as a method of reducing seizures poorly controlled by medications. Several feedback protocols have been attempted with the patient often required to meet multiple contingencies, such as suppressing slow-wave activity while simultaneously generating increased activity in a faster
380
CHAPTER 20
frequency band. Uniformly, the training period required has been long, often several months. The accumulated evidence from such arduous training suggests that certain epileptics can learn "something" that will allow for reduction of seizure frequency (Cleeland, 1981). The reader is cautioned, though, that the research discussed below has generally employed patient populations that are a mixture of both pediatric and adult epileptics. Differences in the utility of EEG biofeedback across age ranges have not been systematically evaluated. The biofeedback treatment of seizures began in the early 1970s. It is rooted in the work by Sterman and others on the sensorimotor rhythm (SMR). This rhythm refers to 12- to 14-Hz activity maximally recorded over the sensorimotor cortex. It was originally identified in the waking EEG of cats, where SMR' s most obvious behavioral correlate is immobility and muscular inhibition (Donhoffer & Lissak, 1962; Roth, Sterman, & Clemente, 1967; Howe & Sterman, 1972). Clinical applications of SMR training were initiated after it was observed that cats trained to increase SMR power demonstrated increased threshold for seizures when challenged with a convulsion-producing dose of monomethylhydrazine (Sterman, LoPresti, and Fairchild, 1969). In this way, EEG biofeedback mimicked the therapeutic mechanism of anticonvulsant medications. Early studies of EEG training attempted to demonstrate its utility for controlling a variety of seizure disorders. Two strategies for producing EEG normalization emerged. One involved the enhancement of intermediate EEG frequencies, such as the SMR, which are thought to inhibit epileptogenic brain activity(Sterman&Friar, 1972;Seifert&Lubar, 1975; Kuhlman, 1978). A second set of strategies involved training patients to suppress the excessive slow-wave activity and/or EEG spikes characteristic of seizure states. This latter training was provided either by itself (Cott, Pavloski, & Black, 1979) or in combination with biofeedback training to increase either SMR (Sterman, Macdonald, & Stone, 1974; Lubar & Bahler, 1976; Finley, 1976) or beta-range EEG activity (Sterman & Macdonald, 1978; Wyler, Robbins, & Dodrill, 1979). When EEG activity was recorded over the sensorimotor region, these various training procedures produced significant seizure reduction in 60 to 80% of patients. The most consistent criticism of these early studies has been the lack of adequate experimental controls. Recent work from the laboratories of Lubar and Sterman has been responsive to this need. Lubar (1982) provided patients with one of three types of EEG normalization training followed by an altered
feedback phase where patients were trained to increase epileptiform activity. Training ended with a final EEG normalization period. Dependent measures included self-report of seizure activity and neuropsychological testing. In general, seizure activity paralleled the feedback contingency; five of the eight patients showed clinical improvement during the final treatment phase with an average seizure reduction of 39. 7%, whereas four patients relapsed when epileptiform EEG was reinforced. As EEG training was provided using double-blind procedures, Lubar's data provide the strongest evidence to date supporting the specific efficacy of biofeedback training in regulating seizure activity. Neuropsychological testing, including the Halstead-Reitan Battery and either the WAIS or WISC, showed little or no change following EEG training, replicating earlier findings of Wyler et al. (1979). Sterman (1982) reported data relevant both to the maintenance of therapeutic gains over time and to the generalization of EEG changes to nonfeedback periods. Fifteen patients with poorly controlled seizures received EEG training to reduce abnormally low (1-5Hz) and high (20-25Hz) frequencies while simultaneously increasing intermediate (10-15 Hz) activity. Feedback was also contingent on the absence of high-voltage spiking. For five patients, EEG training was preceded by 6 weeks of symptom selfmonitoring; for a second group of five patients, 6 weeks of noncontingent (yoked) feedback was provided. The results strongly supported the role of contingent feedback in controlling seizure rates. Seizure reduction only occurred when the EEG normalization feedback contingency was imposed, with 13 of 15 patients reducing seizures by a mean frequency of about 60%. Furthermore, seizure rates were still reduced by approximately 42% at follow-up. Finally, Sterman and Shouse (1980) performed spectral analyses of sleep recordings taken both before and after EEG conditioning. During periods of maximal clinical improvement, EEG changes showed increased power for intermediate EEG frequencies but decreased power for both low (0-3 Hz) and high (20-33 Hz) frequency bands. These data suggest that EEG training generalized from the daytime laboratory, when feedback training occurred, to the sleep EEG. In summary, several investigators have obtained data supporting the utility of biofeedback for the control of a variety of seizure disorders. Even mentally retarded children (e.g., Lubar, 1982) have been able to profit from complicated feedback contingencies to alter their EEG frequency distributions and reduce seizure rate. An evolution in training has
BIOFEEDBACK
occurred with earlier studies of the effects of specific frequency band training (e.g., SMR training) giving way to efforts to produce EEG normalization across multiple frequency bands. Comparisons of different training procedures are just beginning and it is not clear if one approach holds any superiority over the others in producing changes in either seizure rate or neuropsychological functioning. Studies that have compared alternative frequency training procedures have employed small sample sizes, limiting the power of the design to offer reliable recommendations (Lubar, 1982). Similarly, although some data suggest that treatment gains may be maintained over time (Sterman, 1982), the reported sample sizes are too small to allow for a reliable conclusion. As noted by Lubar, ''It seems appropriate that wide-scale clinical trials be initiated to determine whether EEG feedback conditioning can become a valuable adjunct in the treatment of epilepsy" (Lubar, 1982).
381
absent contractions of the external anal sphincter when the rectum is distended (White et al., 1972). However, inadequate sensation in the rectum to cue external anal sphincter contractions has also been implicated (Wald, 1983). Medical management of fecal incontinence in myelomeningocele typically involves placing the child on a regular schedule of enemas or suppositories and timed defecations following meals. This regimen keeps the rectum relatively empty. As a consequence, reflexive internal anal sphincter relaxations are minimized, allowing internal sphincter tone to oppose successfully the movement of fecal matter out of the rectum. For children who become constipated on this regimen, a variety of stool softeners may also be employed. This regimen was effective in 58.6% of the children studied by White et al. (1972).
Biofeedback Treatment of Fecal Incontinence
Fecal Incontinence in Myelomeningocele Myelomeningocele or spinal bifida is a major cause of neurogenic fecal incontinence in children and adolescents. This disorder involves a congenital neural tube defect with resulting lesions of some of the nerve tracts in the lower spinal cord. These lesions may interfere with the afferent and efferent nerves that innervate the rectum and anal sphincters, compromising the child's ability to voluntarily control bowel habits. Although not medically serious, fecal incontinence often leads to significant social problems and embarrassments. An understanding of the treatment of fecal incontinence requires some familiarity with the normal physiology of bowel movements. The movement of stool out of the rectum is normally opposed by the tonic constriction of the internal anal sphincter. However, when sufficient amounts of stool or gas enter the rectum, the internal sphincter will reflexively relax. This requires a voluntary contraction of the external sphincter in order to avoid soiling (Marzuk, 1985). From this brief description, it can be seen that normal toilet training in children requires three factors: voluntary control of external anal sphincter contractions during the period of internal anal sphincter relaxation, adequate sensations from the rectum to allow for accurate timing of the external sphincter contraction, and appropriate self-toileting behaviors (Whitehead et al., 1986). Fecal incontinence in myelomeningocele is generally attributed to weak or
Engel, Nikoomanesh, and Schuster (1974) described the first use of biofeedback to treat fecal incontinence in myelomeningocele. As described by these authors and others, biofeedback treatment of fecal incontinence involves the use of three balloon pressure transducers inserted in the patient's anus and rectum. One balloon is placed in the proximal rectum, a second in the internal anal sphincter, and the third in the external anal sphincter. Inflation of the balloon in the rectum simulates the presence of stool. Pressure measurements from the other two balloons measure muscle tone in the internal and external anal sphincters in response to this rectal distension. By observing the manometric or pressure changes in these three balloons, patients are given feedback concerning these three physiological activities. Patients use this information to learn to produce voluntarily external anal sphincter contractions that ( 1) occur in response to rectal distension and (2) outlast the phase of internal sphincter relaxation. As outlined by Whitehead, Parker, Masek, Cataldo, and Freeman (1981), biofeedback training in myelomeningocele patients usually proceeds in four phases. An initial assessment phase tests the rectosphincteric reflex of the internal anal sphincter, the strength of the voluntary contraction of the external sphincter, and the patient's subjective sensory threshold for rectal distension. In the second phase, patients are trained to make skillful contractions of the external anal sphincter in the absence of rectal distension. If possible, patients are trained to voluntarily contract the external anal sphincter. If this proves too difficult, voluntary contractions of the
382
CHAPTER 20
nearby gluteal muscles may be adequate (Wald, 1981). During this phase, patients are typically given visual feedback in the form of observation of manometric tracings from the external anal sphincter. Verbal and tangible reinforcers are also often employed, especially with younger children. In phase three, patients are trained to make voluntary external sphincter contractions in response to rectal distension. Training often begins with distensions at the level of subjective appreciation and gradually progresses to increasingly larger amplitude distensions. During this phase, patients observe manometric tracings from all three balloons to aid them in producing sphincter contractions of adequate amplitude, duration, and timing. In the final phase, patients are required to continue to demonstrate skillful control of the external sphincter but without visual feedback and without any cues concerning the onset or duration of rectal distension and internal anal sphincter relaxation. Uncontrolled case studies (Cerulli, Nikoomanesh, & Schuster, 1979; Whitehead et al., 1981; Ward, 1981, 1983; Shepherd, Hickstein, & Shepherd, 1983) have all found manometric biofeedback to produce clinically significant reductions in fecal soiling in the majority of children treated. Success rates have varied from 46 to 96%. It should be noted that these improvement rates are for children who have invariably failed to achieve continence via more standard medical management. In addition to demonstrating biofeedback's utility, these preliminary data have suggested tentative guidelines for selecting patients who are most likely to benefit from manometric biofeedback training. Wald (1983) observed that patients who benefitted from biofeedback had significantly lower thresholds of rectal sensation than did patients who failed to improve. He reported that of the 43% of his sample of children with myelomeningocele who had impaired appreciation of rectal distension, all failed to respond to biofeedback therapy. Shepherd et al. ( 1983) also found a high incidence of impaired rectal sensation, with 18 of 22 children assessed as having either subnormal or absent sensation. However, they observed that following behavioral training for continence, previously absent rectal sensation developed in four patients. In those cases, impaired rectal sensation was associated with fecal retention and megarectum. The authors suggested that biofeedback training may be effective for these children following appropriate medical management of their megarectum. Recently, Whitehead and colleagues reported a study comparing biofeedback and behavior modification approaches (Whitehead et al., 1986). After a 2-week baseline period, all subjects received behav-
ior modification training to teach self-initiation of bowel movements. Briefly, this involved regular toileting after meals with reinforcers for successful bowel movements or for accident-free days. In addition, the program used enemas and/or stool softeners under specified conditions to prevent fecal impaction. For one group, biofeedback training was initiated following I month of behavior modification training. For a second group, biofeedback training was provided after a 3-month delay. The results were complex and were interpreted by the authors as suggesting that clinically significant reductions in incontinence occurred primarily during the behavior modification phase of training. The authors also used Markov chain modeling to analyze the separate contributions of behavior modification and biofeedback. This analysis suggested that biofeedback and behavior modification "had differential effects on different types of incontinent states; namely, behavior modification was the only treatment effective in causing patients to move from infrequent incontinence (staining or one accident/day) to continence, whereas biofeedback was more effective than behavior modification for highfrequency (more than once a day) incontinence." Replicating Wald (1983), children with impaired rectal sensation showed a poor response to biofeedback training. However, these same children were able to improve bowel control, with four of these six patients achieving a 75% or greater reduction in episodes of incontinence. This improvement was presumably the result of the addition of behavior modification to these patients' training regimen. Six-month or longer follow-up data were available on 66% of the patients in the Whitehead et al. ( 1986) study. These data indicated that both the frequency of incontinence and the number of enemas per week were still significantly reduced relative to baseline levels. Sphincter strength in patients who previously met a satisfactory training criterion was also well maintained. However, the number of selfinitiated bowel movements regressed to pretraining levels. The results of their experiments led both Shepherd et al. (1983) and Whitehead et al. (1986) to similar conclusions. Both argued for a partitioning of children with myelomeningocele into two groups based on physiological criteria. In one group, rectal sensation is minimal or absent and augmented reflex activity of the external anal sphincter is present. In this group, chronic constipation and/or megarectum occurs. Whitehead et al. ( 1986) also characterized these patients as being more likely to have spinal cord lesions at L-2 or above. These patients seem to benefit the most from behavior modification combined
BIOFEEDBACK
with medical management to produce regular, complete evacuations of stool from the rectum. The second group is characterized by normal or near-normal rectal sensation, inadequate reflex activity and muscle tone in the external sphincter, and multiple daily bowel movements. These patients benefit the most from biofeedback training to augment the strength of voluntary contractions in the external anal sphincter. Preliminary data suggest that therapeutic changes are maintained at follow-up.
Attention Deficit Disorder with Hyperactivity The essential features of attention deficit disorder (ADD) are signs of developmentally inappropriate inattention and impulsivity (APA, 1980). DSMIII recognizes two subtypes of this disorder, ADD with and without hyperactivity. They both share signs and symptoms relating to inattention and impulsivity, but problems with excessive motor activity are only found in the former category. Associated features vary but most often include negativism or noncompliant behavior, mood lability, low frustration tolerance, and low self-esteem. Specific developmental disorders (i.e., learning disabilities) are also relatively common. In the past, this disorder has received a variety of names, most typically minimal brain dysfunction and hyperactivity. DSM-III opted for the term attention deficit disorder because attentional difficulties are prominent and ubiquitous and definite evidence for cerebral brain dysfunction is not found in most children. In addition, the hyperactivity often observed in these children frequently diminishes during adolescence, while difficulties in attention persist. The majority of the literature on biofeedback and ADD selected children with excessive motor activity. As most of these studies labeled these children as hyperactive, this review will adopt that more common albeit outdated label. The onset of ADD with hyperactivity is typically by the age of 3 although the diagnosis is usually made during elementary school. This disorder is estimated to be I0 times more common in boys than in girls (APA, 1980). Lambert, Sandoval, and Sassone (1978) reported the prevalence of this disorder among elementary school children as ranging from I to 6%. The traditional treatment of hyperactivity involves one of two approaches (Walden & Thompson, 1981 ) . The most popular treatment approach has been medication management using CNS stimulants. It was initially assumed that these drugs had a para-
383
doxical effect in hyperactive children, producing therapeutic sedation rather than stimulation. However, in the early 1970s, Satterfield and associates observed that hyperactive children displayed both autonomic and electrocortical hypoarousal (Satterfield & Dawson, 1971). This suggested that stimulants work not by producing a cortical sedation but by stimulating cortical inhibitory mechanisms that are immature or inadequate in hyperactive children. Studies evaluating the use of stimulants in children with ADD have consistently found a reduction in hyperactive behavior (Conners, 1972). However, expected gains in academic achievement have not been found (see Denkowski et al., 1983). Side effects have also proven to be a problem, limiting the use of these medications. Finally, behavioral gains are not always maintained into early adulthood when medications are discontinued (Ackerman, Dykman, & Peters, 1977; Weiss, Kruger, Danielson, & Elman, 1975). Behavior modification provides the most common alternative to medications. This approach involves the use of social learning theory to weaken maladaptive behavior and to strengthen systematically behaviors incompatible with inattention, impulsivity, and hyperactivity (O'Leary. Pelham, Rosenbaum, & Price, 1976). Behavior modification techniques are often employed in conjunction with classroom management strategies (e.g., removing classroom distractors, providing a structured classroom routine) that structure the learning environment to accommodate the behavioral and attentional style of the hyperactive child. Behavior modification is clearly effective in changing specific behaviors, such as in-seat behavior or time on task. Unfortunately, generalization to other syndrome behaviors such as academic performance, or generalization across time when contingencies are removed has been unsatisfactory. In response to these deficiencies, clinicians proposed in the early 1970s that biofeedback may provide a useful primary or adjunctive intervention in treating ADD with hyperactivity. Most investigations of the utility of biofeedback have employed either EMG biofeedback to reduce generalized muscle tension or SMR biofeedback to promote a brain state associated with inhibited motor activity (Sterman, 1982).
EMG Biofeedback Braud, Lupin, and Braud (1975) are generally credited with first proposing the use of frontal region EMG biofeedback with hyperactive children. Braud's (1978) rationale for the utility of EMG bio-
384
CHAPTER 20
feedback was that ''hyperactive children are not only overactive but also overly tense. The authors feel that these tension levels could aggravate the symptomatology." Since this initial hypothesis, eight group outcome studies have been published evaluating frontal region biofeedback in hyperactive children and adolescents. Five studies (Braud, 1978; Bhatara, Arnold, Lorance, & Gupta, 1979; Denkowski. Denkowski, & Omizo, 1983; Dunn & Howell, 1982; Omizo, 1980b) reported EMG data for patients receiving from 3 to 12 biofeedback training sessions. All studies reported reductions in frontal region EMG following biofeedback. Further, three of these studies employed appropriate control procedures allowing the conclusion that the observed reduction in EMG is not the result of habituation across repeated testing occasions. Omizo (1980b) obtained similar results; however, in his study biofeedback was confounded with taped relaxation instruction. Three qualifying points need to be made. First, the decision to train a reduction in frontal region EMG makes the implicit assumption that muscle tension is elevated in hyperactive children. Borkovec and Sides (1979), among others, argued persuasively that the choice of a relaxation strategy should be matched with the response domains showing excessive levels of arousal. Surprisingly, this assumption has only been confirmed in one study (Braud, 1978) where levels of frontal region EMG during rest were significantly higher in hyperactive than in control children. No other laboratories have replicated this important assumption. Second, it is often assumed that a reduction of frontal region EMG leads to a generalized state of relaxation at multiple body sites. This state of relaxation ultimately translates into a decrease in overactivity. Unfortunately, none of the studies measured EMG from additional muscle locations. Burish (1983) reviewed the research on frontal region EMG biofeedback in adults and concluded that this type of biofeedback does not reliably produce a generalized reduction in muscle tension or autonomic activity. Although the situation may be different with children, the existing research does not encourage the frequently made assumption that frontal region biofeedback in hyperactivity leads to a generalized relaxation response. Finally, Gargiulo and Kuna ( 1979) questioned whether it makes theoretical sense to train reductions in physiological arousal in hyperactive children. Based on the Satterfield and Dawson ( 1971) finding
of hypoarousal in hyperactivity, these authors suggested that relaxation might be an inappropriate reg-
ulatory strategy. In support of this, Whitmer ( 1977) trained one group of hyperactive children to decrease frontal region EMG, but trained a second group to increase EMG. Consistent with the underarousal theory, reductions in hyperactive behavior were only found for subjects trained to increase EMG. All eight studies have employed multiple outcome measures in evaluating the utility of biofeedback therapy. Parent and/ or teacher ratings of hyperactive and disruptive behaviors were made in four studies. A striking disparity exists in the ratings made by parents versus teachers. Parent rating scales consistently indicated decreased behavior problems following biofeedback training (Bhatara et al., 1979; Braud, 1978; Dunn & Howell, 1982). In contrast to these positive findings, both studies assessing teacher ratings (Bhatara et al., 1979; Denkowski & Denkowski, 1984) failed to find an improvement in classroom behavior. None of the researchers in this area have commented on this apparent difference between classroom and home behavior change. Possible explanations include procedural differences between the studies (however, note that Bhatara et al. measured both parent and teacher ratings), differences between classroom and home settings in provoking hyperactive behavior, differences in the amount of contact between parents and researchers versus the contact between teachers and researchers, differences in the level of objectivity between parents and teachers, and different levels of expectancy of change between parents and teachers. Neuropsychological testing has involved either the Bender Gestalt, the Digit Span and Coding subtests from the WISC-R, or other tests sensitive to attention and impulsivity. In general, the results suggest improved performance on these measures. Braud (1978) found that biofeedback training led to fewer errors on the Bender test and improved scores on the Illinois Test of Psycholinguistic Abilities. No difference between biofeedback subjects and controls was found on Digit Span and Coding. Dunn and Howell ( 1982) found that hyperactive subjects improved on the Bender Gestalt and WISC-R subtest following active biofeedback training but not following placebo treatment. Finally, Omizo and Michael ( 1982) measured both errors and response latency on the Matching Familiar Figures Test. They found decreases in errors and an increase in latency following biofeedback training and concluded that these changes were consistent with decreased inattention and impulsivity in their subjects. In other response domains, the data are more ambiguous. Academic skills were measured in only two studies (Denkowski et al., 1983; Denkowski &
BIOFEEDBACK
Denkowski, 1984). Although the former study found improvements in reading vocabulary and comprehension following biofeedback, similar improvements were not obtained by the same investigators in their second study. The authors attributed this discrepancy to differences in the tests used to measure academic abilities; however, numerous other procedural differences between the studies (such as the number of training sessions and the age of hyperactive children) may also be important. Finally, in five studies, Omizo, the Denkowskis, and associates assessed a variety of biofeedback-induced changes in either locus of control or self-esteem. These data are important to the hypothesis that biofeedback ''may demonstrate to hyperactive children that they have control over their physical behavior and this perception of self-control then generalizes to enable more selective socioeducational behavior" (Denkowski et al., 1983). Four studies assessed locus of control. In two out of four studies, subjects demonstrated increased internality following biofeedback but not following placebo treatment (Denkowski et al., 1983; Omizo, 1980a). Similarly, of the three studies measuring either selfconcept or self-esteem, two (Omizo, 1980a,b) found favorable changes, whereas one (Denkowski et al., 1983) obtained no advantage for subjects receiving biofeedback. Hence, the literature is almost evenly split between studies supporting the conclusion that biofeedback produces favorable changes in self-concept and locus of control and negative findings in these areas. Clearly, more research is needed before a confident conclusion can be drawn. Control procedures varied across the eight studies reviewed. Braud (1978) employed a no-treatment control group. Bhatara et al. (1979) connected control subjects to an inoperative biofeedback unit but provided no other placebo treatment. Omizo (l980a,b) and Denkowski and Denkowski (1984) attached subjects to inoperative equipment and also played for control subjects either neutral tapes or tapes of children's stories. This contrasted to the biofeedback subjects, who listened to relaxation tapes. Denkowski et a/. (1983) employed a similar procedure but substituted neutral conversations with the children for the neutral tapes. It seems unlikely that any of these control procedures were as credible as biofeedback for either parents or children. The only study that used a potentially credible control procedure was Dunn and Howell (1982). In this study, all children received 10 sessions of placebo treatment followed by 10 sessions of biofeedback treatment. Relationship and play therapy served as placebo treatments (these indeed had no significant effect on
385
either physiology or hyperactive behaviors). Unfortunately, Dunn and Howell made no attempt to assess the placebo's credibility, so that one does not really know if the placebo treatment prompted the same degree of expectancy for change as did the biofeedback treatment (Borkovec & Nau, 1972). These methodological weaknesses make it impossible to determine the role that the contingent presentation of the biofeedback signal plays in producing the clinical improvement obtained in these research studies. Three studies compared EMG biofeedback to taped relaxation training. Braud (1978) and Dunn and Howell ( 1982) obtained comparable improvements in muscle tension, hyperactive behavior, and psychological test data between biofeedback and relaxation. In the Denkowski and Denkowski (1984) study, the results were largely negative for both groups, with one analysis supporting a shift in locus of control toward internality only for subjects receiving relaxation training. Hence, the data do not support the employment of the more expensive biofeedback training over the more cost-effective taped relaxation training. Because these studies were initial attempts to demonstrate the efficacy of biofeedback, it is not surprising that scant attention was paid to demonstrating the maintenance of change over time. Indeed, this crucial issue was assessed in only one study (Bhatara et al., 1979) and in that case, reductions in hyperactive behaviors, as rated by parents, regressed to pretreatment levels after 12 weeks posttreatment. Clearly, the inclusion of an adequate follow-up period is a prerequisite for all future treatment outcome research in this area.
EEG Biofeedback Nail (1973) provided the only controlled group outcome study of EEG biofeedback as a treatment for hyperactivity. In her study, 48 children with hyperkinesis associated with learning disabilities were assigned to either a no-treatment group, contingent alpha biofeedback, and false or noncontingent feedback. Subjects in the latter two groups also received brief relaxation exercises. Her results suggested little or no difference between the veridical and placebo biofeedback groups. Alpha amplitude increased in 9 of the 16 veridical training patients and in 7 of the 16 placebo patients. Measures of hyperkinetic and maladaptive behavior and overall academic achievement did not significantly differ between the biofeedback and control groups.
386
CHAPTER 20
More recently, Lubar and his colleagues reported a series of case studies employing an EEG biofeedback contingency designed to train an increase in SMR in the absence of excessive theta wave activity (Lubar & Shouse, 1976; Shouse & Lubar, 1979; Lubar & Lubar, 1984). Lobar's case studies provide strong preliminary support for the efficacy of this form of biofeedback. Five patients were evaluated with reversal designs and 6 were evaluated as uncontrolled case studies. Of the 11 subjects reported, 10 learned to increase SMR and/or beta wave activity while facial EMG and gross motor activity decreased. Observational measures of classroom behavior improved in 8/ 13 categories for 4 of 5 children for whom these data were available. Electrocortical and to a lesser extent behavioral improvement reversed when the biofeedback contingencies were withdrawn. For 6 of these children, information was presented on· academic test scores and/or school grades. In each case, unspecified "considerable improvements'' were reported. These data suggest that SMR biofeedback warrants an evaluation with a controlled group outcome study or with a larger series of controlled case studies. As Lubar and his colleagues treated patients with multiple therapies (EEG biofeedback was combined in various studies with medications, academic remediation, and feedback on muscle activity) it will also be necessary to determine the relative contribution of EEG biofeedback. Finally, SMR biofeedback requires a relatively large investment of both human and technological resources and its cost effectiveness relative to other interventions needs to be addressed.
specific developmental disorders when referring to learning disabilities. Learning disabilities are generally thought to be about twice as common among males as among females (APA, 1980). Hyperactivity and other behavioral disorders are often associated with learning disabilities, and delinquency may be a complication of learning disabilities among adolescents (Sattler, 1982). The treatment of learning disabilities traditionally involves some form of remedial education training, with the specific training methods tailored to the cognitive and academic strengths and weaknesses of the child. Behavior modification principles may also be employed to alter behavior patterns, such as overactivity, that impair the child's ability to profit from educational experiences. Although remediation is often effective, it is not uncommon for LD children to demonstrate a pattern of academic difficulties throughout childhood and adolescence with many individuals showing cognitive problems in adult life (APA, 1980). As noted above, it is not uncommon for children with hyperactivity to display difficulties in learning, either as a primary neuropsychological problem or secondary to the behavior disruption inherent in hyperactivity. It is not surprising, then, that the biofeedback treatment of learning disabilities closely parallels the treatment of hyperactivity. Both EMG-based relaxation techniques and EEG biofeedback have been employed.
EMG Biofeedback in Learning Disabilities
Children with Learning Disabilities Learning disabilities refer to deficiencies in academic performance that are not attributable to lowered intelligence, emotional disturbance, sensorimotor problems, or significant socioeconomic disadvantage (Rourke, 1981). The etiology of a learning disability is usually unclear. The definition of learning disabilities implies that they arise from neuropsychological inefficiencies in the perception, organization, and/ or integration of information. The specific type of brain-related disability has been found to vary among learning-disabled (LD) children. An excellent discussion of this issue is found in Rourke (1981, 1985). Learning disabilities can affect any academic skill, but most typically involve language-related problems. DSM-III employs the term
Carter and Russell ( 1980) reasoned that stress and muscle tension have a debilitating effect upon academic attainment in LD children. They speculated that relaxation training would ''bring about a cognitive reorganization or integration which allows the recipient to use his abilities more efficiently.'' To test this, they gave four LD males, aged 8-13, ten sessions ofEMG biofeedback from the forearm flexors combined with handwriting training. They were able to demonstrate both a 62% decrease in forearm muscle tension and a mean increase of 0.65, 0.68, and 0.53 grade level equivalences in reading, spelling, and arithmetic abilities, respectively. The authors followed these uncontrolled case studies with two controlled group outcome studies (Carter & Russell, 1985). In the first study, 32 male elementary students with learning disabilities were randomized either to a combined biofeedback/taped
BIOFEEDBACK
relaxation/handwriting training condition or to a notreatment control group. Each student also completed a test battery measuring intellectual functioning, academic achievement, auditory memory, perceptual motor skills, and handwriting ability. The results demonstrated significantly greater improvement for the experimental group on all measures except the arithmetic subtest of the Wide Range Achievement Test. EMG was only measured for experimental subjects, with those subjects demonstrating a significant reduction in forearm muscle tension following biofeedback/relaxation training. In a replication study, 30 elementary school LD males were randomized either to 18 sessions of a similar biofeedback/relaxation/handwriting training procedure or to a no-treatment control group. The results replicated the main findings from study I, with significantly larger gains made by experimental subjects on all dependent measures except general intelligence and auditory memory. Finally, with an independent group of 20 LD males, the authors obtained significant improvements on the Tennessee Self-Concept Scale and on the Child Behavior Rating Scale. Taken as a whole, the authors suggest that (I) LD children are able to learn to reduce relevant muscle tension levels, and (2) these changes are associated with improvement in cognitive abilities, academic achievement, behavioral adjustment, and self-concept. Omizo and Williams (1982) argued that "relaxation training is effective because the learning-disabled child gains control over his muscle levels, thus improving his ability to perform on visual/perceptual tasks that involve focusing on relevant cues and ignoring irrelevant stimuli." In a controlled outcome study, the authors randomized 32 children, ages 811 , either to three sessions of combined frontal region EMG biofeedback, taped relaxation training or to a placebo treatment involving listening to neutral tapes. The study found significant decreases in frontal region EMG only for subjects in the experimental group. In addition, compared to the control subjects, subjects receiving biofeedback/relaxation training made significantly fewer errors on the Matching Familiar Figures Test and significantly increased their responding latency. These findings were interpreted as reflecting increased attention to task and decreased impulsivity in the relaxation-trained LD pupils. Finally, control and experimental subjects failed to differ on a measure of locus of control, with both groups showing little change from pre- to posttesting. Unlike the studies previously described, Hunter, Russell, Russell, and Zimmerman (1976) trained digital skin temperature increases to LD and normal
387
children. The authors reasoned that temperature feedback would improve academic skills by teaching LD children a "fully attentive, attuned, and stable internal state in which background noise is reduced to a minimum.'' Hunter et al. randomized LD and normal children, ages 7-9 years, either to five sessions of "consistent" temperature feedback or to a false feedback "mixed reinforcement" procedure (the success feedback signal followed increases, decreases, and stable temperature patterns). The results demonstrated modest (OSF) but statistically reliable increases in skin temperature for both LD and normal children with veridical reinforcement. However, veridical feedback and false feedback LD groups did not vary in improvement on scores on a brief neuropsychological battery. In summary, EMG and temperature biofeedback training do seem to be associated with increased voluntary control over trained physiology. EMG biofeedback was also associated with desirable changes in behavior, academic achievement, and less consistently self-report measures of self-concept and locus of control. Unfortunately, the interpretation of these findings is problematic. First, all of the studies combined biofeedback with other training procedures, making it impossible to identify the unique properties of biofeedback training. Second, although all of the authors assumed that biofeedback leads to a sense of relaxation, no attempt has been made to demonstrate a relaxation response involving multiple response systems. Third, all of the studies employed control procedures that may have been inadequate to control for nonspecific treatment effects, such as expectancy for change. As before, treatment credibility was not assessed. Fourth, none of these studies included a follow-up period to determine the stability of behavior change over time. Finally, the rationale behind the choice of relaxation as a specific intervention for learning disabilities has varied across studies and is poorly articulated at best. Greater conceptual clarity would aid clinicians in deciding which children might benefit the most from a relaxation-based intervention.
EEG Biofeedback in Learning Disabilities As noted above, many theories of learning disability assume some underlying neurological dysfunction. Comparison of the EEG of normal and LD children became one method of searching for this. Roberts ( 1966) demonstrated excessive amounts of slow-wave EEG activity, particularly in the 3- to 4-
388
CHAPTER 20
Hz frequency band, in LD children. More recently, Lubar et al. (1985) performed power spectral fast Fourier analyses of EEG in 69 children with learning disabilities and 34 controls. Among other findings, LD children differed from controls by having significantly more power in theta and low alpha frequency bands, i.e., more slow-wave EEG activity. Based on these and other observations, it has been suggested that biofeedback techniques might prove useful in training LD children to alter dysfunctional patterns of EEG activity, and that this modification would produce salutary changes in learning and behavior. Gracenin and Cook (1977) gave eight LD children 10 sessions of alpha biofeedback. During the last 6 sessions, subjects were asked to increase alpha while reading. Although statistical analyses were not reported, the authors suggested that four of eight subjects learned to increase alpha amplitude and duration. However, alpha-trained subjects did not differ from no-treatment controls on improvements in oral reading and reading comprehension. Cunningham and Murphy (1981) hypothesized from earlier research (Murphy, Darwin, & Murphy, 1977) that different types of cognitive tasks (e.g., visual spatial versus verbal) are associated with different patterns of electrocortical arousal. They suggested that LD children may lack the ability to produce these patterns of cortical arousal in a fashion necessary to perform different types of cognitive tasks. Based on this assumption, 24 LD adolescents, ranging in age from 13.1 to 17.9, were assigned to one of three groups: ( 1) EEG biofeedback to produce a pattern of increased right hemisphere and decreased left hemisphere frequencies, (2) training to decrease EEG frequencies in both hemispheres, and (3) a notreatment control group. The results were complex and will only be briefly summarized here. First, EEG power data found lower left than right hemisphere cortical arousal across both verbal and nonverbal tasks. These data support theories of left hemisphere hypoarousal in LD children (e.g., Satz, Rardin, & Ross, 1971). In addition, both biofeedback groups displayed decreasing left hemisphere baseline frequencies across sessions. In terms of achievement test data, subjects trained to increase right hemisphere and decrease left hemisphere arousal showed a significant improvement on the Wide Range Achievement Test (WRAT) Arithmetic subtest. Other subjects showed no change in arithmetic scores. None of the subjects displayed significant improvement on measures of spelling, reading, or spatial abilities. Finally, a strong correlation (r = 0.9) was found between subjects' ability to increase
right hemisphere frequencies and improvement on WRAT Arithmetic. Carter and Russell (1981) reported a single group outcome study on four LD children chosen on the basis of having a WISC-R Verbal IQ score at least 15 points below their Performance IQ (the mean Performance-Verbal discrepancy was 25.34 IQ points). These elementary school aged boys received 16 sessions of EEG biofeedback training to voluntarily produce either alpha or beta activity coupled with taped relaxation training. Due to equipment problems, the EEG data were not reported. However, posttest data suggested a marked decrease of 14.67 points in the discrepancy between Verbal and Performance IQ with the mean Verbal IQ increasing from 76.3 to 88.7. Tansey (1984) also reported single group outcome data on six elementary school aged boys with a history of learning disabilities. Unlike the subjects in the previous study, these children varied in the relative strength of their Verbal and Performance IQs. Tansey's subjects received a variable number of biofeedback training sessions to increase bilateral SMR amplitude. Tansey reasoned that many forms of remedial therapy for LD children work by providing external, task-relevant stimulation to the child's sensorimotor cortex. He argued that SMR biofeedback provides for internal cerebral stimulation. Further, by providing bilateral EEG biofeedback, remediation should also occur for learning deficits caused by deficiencies in normal interhemispheric interactions. The results strongly supported the efficacy of SMR biofeedback. First, as a group, SMR amplitude showed a mean increase of 138% above baseline levels. Second, large increases, ranging from 7 to 25 points, were found for all six children in WISC-R Verbal, Performance, and Full-Scale IQ scores. These IQ increments were felt to be much larger than the change that would be expected by maturation alone. Finally, SMR training attenuated the discrepancy between Verbal and Performance IQ scores for the four children with significant pretraining asymmetries in ability levels. It is premature to offer a reliable opinion on the utility of EEG biofeedback as a therapy for LD children. In part, this is caused by the lack of procedural overlap between the four published articles in this area. However, several observations can be made. First, LD children and adolescents demonstrated reliable changes in physiology from pre- to posttesting. Unfortunately, the lack of appropriate control measures makes it impossible to determine if these changes are due to the contingent presentation of the
BIOFEEDBACK
feedback signal, or are caused by other factors, such as confounding treatments, nonspecific treatment effects, or mere instructions to attempt to control EEG. Second, psychoeducational data were limited. The two studies reporting achievement testing found, at best, only minimal changes in academic skills. Also the absence of appropriate controls makes it difficult to separate the effects of biofeedback training from the influence of both maturation effects and practice effects caused by repeated exposure to tests. Third, none of the studies obtained follow-up data to determine if reported benefits of treatment were maintained. Finally, with the exception of the study by Carter and Russell, no attempt was made to sample a group of LD children or adolescents with relatively homogeneous neuropsychological characteristics. Rather, an invariant intervention is assumed to be equally effective for all LD individuals [see Doehring, Hoshko, & Bryans (1979) and Leslie, Davidson, & Batey (1985) for criticisms of this strategy in LD research in general].
Childhood Migraines Despite the large literature on the biofeedback treatment of adult headaches, little is known about its utility in pediatric headaches. Of the studies that have been completed, all have investigated biofeedback treatment of migraine headache.
Migraines in Children and Adolescents Migra4le headaches are characterized by paroxysmal alterations of cerebral blood flow. The pain of migraine is felt to be caused by excessive vasodilation of the cranial arteries (Fenichel, 1985), with some arguing that substances such as bradykinin and histamine play a role by sensitizing primary pain afferents during migraine attacks (Saper, 1983). The source of this vascular instability is uncertain, but neural and humoral factors may both be involved. The prevalence of migraine in children and adolescents is estimated to be between 3 and 7% (Bille, 1962; Hoelscher & Lichstein, 1984). In early childhood, migraine has been reported to be at least as frequent in boys as in girls, but this gender ratio changes in late adolescence and early adulthood, when females are more commonly affected (Rigg, 1975). Although simple analgesics are often effective for childhood headaches, they typically provide little
389
relief for migraine sufferers who present at headache clinics. Unfortunately, little systematic research has been done on medication protocols for childhood migraines. For adult patients with infrequent migraines, abortive agents such as ergotamine or Midrin are employed. These medications are vasoconstrictors and presumably work by minimizing the vasodilation phase of the migraine. For patients with more frequent migraines, prophylactic agents are recommended. A variety of medications have been recommended, the most consistent efficacy being claimed for beta adrenergic blocking agents, such as propanolol (Fenichel, 1985). In addition, physicians often counsel patients to avoid things that trigger migraine attacks. Common triggers are strong emotions, fatigue, dietary factors including a variety of foods and alcohol, and oral contraceptives (Saper, 1983). The decision to supplement simple analgesics with more powerful medications is one that many physicians are hesitant to make. Medications such as ergotamine and propanolol have a number of undesirable side effects. In addition, many physicians are reluctant to prescribe maintenance medications chiefly due to concerns about eventual drug dependency or to concerns about the unknown effects of long-term use of these medications on children (Hoelscher & Lichstein, 1984). This reluctance has motivated a search for effective, nonpharmacological treatments of migraines in both children and adults.
Biofeedback Treatment of Migraines The research on the biofeedback treatment of migraines in adults provides a model for the interventions used with children and adolescents. In a series of controlled group outcome studies, both digital skin temperature biofeedback and cephalic vasomotor biofeedback have proven more effective than no treatment in reducing migrainous symptoms (Blanchard & Andrasik, 1982). There is some dispute over the mechanism(s) underlying the effectiveness of biofeedback training. Some (e.g., Elmore & Tursky, 1981) emphasize the role of biofeedback in training patients to voluntarily produce cerebral artery vasoconstriction during the headache phase of the migraine. For others, the efficacy of digital skin temperature biofeedback rests on the observation that the vasodilation or headache phase of the migraine is felt to be a reaction to a preceding phase of excessive vasoconstriction. These proponents argue that train-
390
CHAPTER 20
ing subjects in peripheral vasodilation (increased digital skin temperature) produces a generalized decrease in sympathetic tone (smooth muscle relaxation) that minimizes the prodromal vasoconstrictive phase and aborts the headache before it starts (Sargent, Green, & Walters, 1973). Despite this uncertainty, there is agreement that both procedures are effective, with 50 to 70% of adults being significantly improved following biofeedback treatment (Labbe & Williamson, 1984). The treatment of childhood migraine with behavioral interventions has a relatively short history, with only isolated reports appearing as early as the mid-1970s. During this period, the literature was limited to controlled and uncontrolled case studies and single group outcome studies. These reports explored the utility of a variety of behavioral self-regulation methods, the most common intervention being skin temperature biofeedback with autogenic training. Autogenic training is a relaxation method developed around the tum of the century in Germany by Schultz and more recently reintroduced by Luthe (Schultz & Luthe, 1969). The method involves passive concentration on a systematic set of phrases and images suggesting control of physiological responding. The combination of biofeedback and autogenics was popularized by clinicians at the Menninger Foundation who labeled this integrated approach autogenic biofeedback training (Green, Green, Walters, Sargent, & Meyer, 1975). In a review of this literature, Hoelscher and Lichstein (1984) concluded that, "it appears that skin temperature biofeedback with autogenic training is associated with significant reductions in migraine headache activity.'' However, they cautioned that design limitations made it impossible to conclude that behavioral interventions were reliably superior to no treatment or placebo treatment. The reader is referred to the Hoelscher and Lichstein paper for a more detailed discussion of these early studies. Since this review, one single group and two controlled group outcome studies have been reported (see Andrasik, Blake, & McCarran, 1985, for an excellent review of the current published and unpublished literature on the treatment of pediatric headaches). Werder and Sargent (1984) treated 21 children, ages 7-17, with approximately 7 hours of biofeedback (skin temperature and EMG) and relaxation (autogenics, progressive relaxation, guided imagery) training. Nineteen of the children had migraine headaches; the other two were diagnosed as mixed vascular-muscle contraction headaches. Treatment resulted in a 71% reduction in mean weekly headache hours with a concomitant 87% reduction
in medication usage. Improvement was maintained over a follow-up period of 1-3 years. Labbe and Williamson ( 1984) compared autogenic feedback and a no-treatment control group with regard to migraines in 28 children ages 7-16. Autogenic feedback led to significant improvement in headache intensity, frequency, and duration measured posttreatment and at l month follow-up. In addition, 93% (13/14) of the patients receiving autogenic feedback reported at least a 50% improvement in their average headache rating. Six-month followup data were also available on 8 of the 14 treated patients: 5 were found to be improved or symptomfree. In contrast, there were no mean group differences in pre and post ratings of headaches for the control subjects, with only 7 and 14% rating at least a 50% improvement at posttreatment and l month follow-up. Finally, across all training sessions, patients were able to produce a statistically significant increase in digital skin temperature while the feedback display was available. However, no differences were found between the treatment and control group during the pre- and posttreatment self-control assessment; for example, patients were not able to increase skin temperature in the absence of feedback. Masek, Russo, and Varni (1984) reported on their research project evaluating a comprehensive pain management program for childhood migraines. Eighteen children ages 8-12 were randomized to one of three groups: (1) EMG biofeedback, meditative exercises [similar to Benson's (1975) secular TM], and pain behavior management (operant control of pain behaviors), (2) progressive muscle relaxation, meditative relaxation, and pain behavior management, or (3) a waiting list control group. The data supported clinically significant reductions in headache activity in both treatment groups averaging 60 and 82% across all outcome measures for the biofeedback and relaxation groups, respectively. In contrast, patients in the control group reported a 19% increase in symptoms. These changes were maintained at 1 year follow-up. The authors concluded that the behavioral management of migraines is very effective, but that EMG biofeedback does not appear to be an essential component. These preliminary studies encourage the use of behavioral techniques in the treatment of vascular headaches in children. Impressive reductions in headache activity have been found in all group outcome studies with benefits maintained over followup. What seems less clear is the specific role that biofeedback plays in contributing to these positive outcomes. In the three recent group outcome studies, biofeedback was invariably combined with other
BIOFEEDBACK
physiological, cognitive, or behavioral pain management interventions. In addition, only one of these studies (Labbe & Williamson, 1984) measured physiological responding and their trained patients were unable to increase their skin temperature in the absence of a biofeedback display. Finally, the report by Masek eta/. (I 984) found no advantage in the use of EMG biofeedback over progressive muscle relaxation. Unfortunately, this study included only six subjects in each intervention, limiting the power of the design when comparing treatment alternatives. It appears that much more research needs to be done in these areas. Relevant unanswered questions are: I. Is biofeedback either alone or in combination with other interventions more effective than suitable placebo interventions in reducing migraine headache activity in children? Although it is of some academic interest to determine if biofeedback works in isolation, it is very unusual for biofeedback to be clinically applied in pain management without combining it with other behavioral treatments (Roberts, 1985). 2. If biofeedback is uniquely effective, which modality is most appropriate for which patients? For example, skin temperature biofeedback may work best as a prophylactic treatment for migraine headaches, as it theoretically minimizes the sympathetic overactivity during the initial or prodromal phase of the migraine. In contrast, temporal artery pulse volume biofeedback might have more benefit as an abortive treatment for modifying the vasodilation during the headache phase of the migraine. Finally, all forms of biofeedback may be unsuitable for patients with strong operant contributions to their pain complaints (Haber, Kuczmierczyk, & Adams, 1985). 3. Much work still needs to be done explicating the physiological mechanisms underlying biofeedback's effectiveness. Labbe and Williamson (1984) did not find evidence for impressive posttraining selfregulation of skin temperature. Despite this, patients reported decreased headache activity both immediately posttraining and at follow-up. This discrepancy between skin temperature and headache changes requires further investigation. Possible explanations of this discrepancy include: (1) it may be more relevant to measure cephalic temperature than digital skin temperature; (2) even though patients cannot increase skin temperature, biofeedback training may result in a decrease in the vasomotor variability or instability characteristic of migraines, (3) biofeedback and related techniques may be effective by modifying other physiologically relevant parameters implicated in the onset of migraines (e.g., blood pressure), or (4) biofeedback's efficacy may be a result of cognitive changes associated with successful control of the
391
feedback signal and not primarily regulation of vasomotor activity (Holroyd eta/., 1984).
General Discussion Several impressions emerge from this literature review. The existing data indicate that children as young as 5 years old are able to learn to control targeted physiology following biofeedback training. It also seems clear that many patients show clinically significant improvements in symptom control following biofeedback interventions. These changes occur despite the fact that biofeedback is often attempted only when patients have not responded to previous medical interventions. Although encouraging, the significance of many of these findings is compromised by numerous methodological shortcomings (see Cleeland, I 98 I ; Cobb & Evans, 1981; Kewman & Roberts, 1983). With the exception of the exemplary programmatic research on the treatment of fecal incontinence and epilepsy, the. following problems are common in the studies reviewed in this chapter. 1. Control procedures are typically inadequate to rule out the contributions of nonspecific changes in attitude and motivation. Adequate designs, such as the use of credible noncontingent feedback (Lubar, 1982), the comparison of biofeedback with other active treatments (Whitehead eta/., 1986), and the use of appropriate single case experimental designs (Bird & Cataldo, 1978) are rarely employed. 2. Interpretation is often clouded by the mixing of biofeedback with other behavioral interventions including relaxation training, behavioral skills training, and behavior modification. This is a thorny issue in biofeedback studies as additional treatment approaches, such as relaxation training, are often used as home practice aids designed to help maintain symptom control in the home or school situation. Although there may be much clinical wisdom in applying a treatment package rather than an isolated biofeedback intervention, these confounds obviously make it impossible to determine which treatment should be credited for the obtained therapeutic gains. 3. Information concerning potentially important patient characteristics is usually not reported. For example, hyperactive children vary widely in their levels of cortical and autonomic arousal (Finley, 1982). Despite this, not one study considered measures of physiological arousal when selecting patients for biofeedback interventions, and only one investigation (Denkowski, Denkowski, & Omizo,
392
CHAPTER 20
1984) correlated baseline measures of muscle tension with therapeutic gains. The implications of individual differences for the choice of a biofeedback intervention have been inadequately explored. 4. Follow-up data are rarely collected. Erosions in therapeutic gains are the norm and this possibility must be assessed. When appropriate, treatments should be altered to include elements promoting the maintenance of symptom control (e.g., the gradual fading of biofeedback contingencies, providing feedback both during resting conditions and during situational challenges). 5. There is a tradition in behavioral medicine for the use of relaxation as an intervention with broadspectrum healing powers. This has resulted in the prescription of relaxation even when no compelling rationale for its usage has been empirically documented. For example, the use of frontal region, EMG biofeedback in the treatment of learning disabilities has as much justification as the use of increased exercise, adequate nutrition, or other formulas for improving general well-being. In addition, Burish has argued that frontal region, EMG biofeedback provides only minimal benefit as a generalized relaxation procedure. Indeed, he and his colleagues have employed this form of biofeedback as a control intervention when evaluating the impact of other more potent relaxation procedures (Shirley, Burish, & Rowe, 1982). Taken as a whole, the data do not support the relatively widespread use of this intervention for the pediatric problems reviewed in this chapter.
Patient-Treatment Interactions and the Art of Biofeedback The successful application of appropriate biofeedback interventions with children and adolescents often requires modifications in the training procedures developed for adults. Experienced clinicians have long recognized this point and helpful guidelines are provided by both Linkenhoker (1983) and Attanasio et al. (1985). For example, whereas adults typically tolerate the procedures involved in electrode applications, younger children may require greater reassurance about the safety of instrumentation. To date, most of the available advice is based on unsystematic observations of patients. However, some useful data are beginning to emerge. Clinical researchers have commented upon the challenges presented by the sometimes fragile motivation level and attention span of children referred for biofeedback training. Denkowski and Denkowski
(1984) observed that across several research studies of EMG biofeedback with hyperactive children, a tendency emerged for physiological learning to plateau by the fourth session. They tentatively attributed any lack of further learning to boredom with the feedback task. One possible strategy for preventing boredom is to reduce the length of biofeedback sessions or to include a larger number of rest breaks (Attanasio et al., 1985). Unfortunately, little is known about the impact of these procedural changes. Others have attempted to maintain attention either by employing tangible reinforcers or by altering the nature of the feedback display. Finley et al. ( 1981) developed an automated reward system involving a universal feeder that presented children with reinforcers, such as candy or small toys, contingent upon appropriate changes in physiology. The use of this system led to greater and more rapid reductions in EMG in a group of cerebral palsied children than did the use of an audio feedback signal alone. Other recommended changes in reinforcement have included the use of music rather than a feedback tone (Walmsley, Crichton, & Droog, 1981) and the use of video game-like feedback displays. It will be important, though, to determine if the enhanced stimulation provided by these changes leads to counterproductive increases in physiological arousal. Many of the studies reviewed in this chapter combined biofeedback with medical interventions, mostly commonly medication. Despite this, interactions between behavioral and pharmacological therapies have received little attention. Linkenhoker (1983) alerted clinicians to potential dangers caused by failures to alter the dosage of medications whose physiological effects parallel the physiological consequences of biofeedback. For example, Seeburg and DeBoer ( 1980) reported a case study where EMG biofeedback training in a diabetic reduced that patient's need for Insulin, leading to harmful side effects until her medication regimen was adjusted. In a more positive vein, Surwit, Allen, Gilgor, and Duvic (1982) investigated the concurrent effects of autogenic training and sympathetic blocking agents on skin temperature changes in patients with Raynaud's disease. Statistically significant changes in skin temperature were only obtained when both interventions were applied simultaneously. On a related issue, Cleeland ( 1981) suggested that the addition of behavioral training methods to medical regimens might allow for a reduction of medications, such as anticonvulsants, which often suppress symptoms only at levels that are either toxic or produce unpleasant side effects. Finally, researchers are beginning to identify
BIOFEEDBACK
patient variables that predict success or failure with biofeedback interventions. Denkowski et al. (1984) employed multiple regression analyses to relate patient age, pretreatment EMG level, degree of hyperactivity, and locus of control to decreases in EMG levels following biofeedback training. The subjects were 59 hyperactive males, aged 8 through 15. Only locus of control predicted posttreatment changes in EMG, with internal locus of control associated with better outcomes. Whitehead et al. (1986) also found that age did not predict response to biofeedback training in their patients with neurogenic incontinence. Children as young as 5 were able to successfully profit from biofeedback training as long as tangible reinforcers were used to maintain the attention of their younger patients. The authors did find, however, that physiological parameters were important in determining which patients benefit from biofeedback versus behavior modification approaches. Quite obviously, we are only beginning to understand the impact of demographic, cognitive, behavioral, and physiological characteristics of patients on the biofeedback learning experience. Much worthwhile research is waiting to be done.
References Ackennan, D. T., Dykman, R. A., & Peters, J. E. (1977). Teenage status of hyperactive and non-hyperactive learning disabled boys. American Journal of Orthopsychiatry. 47, 577-
5%. Adams, R. D., & Victor, M. (1977). Principles of neurology. New York: McGraw-Hill. American Psychiatric Association. (1980). Diagnostic and statistical manual of mental disorders (3rd ed.). Washington, DC: Author. Andrasik, F., & Attanasio, V. (1985). Biofeedback in pediatrics: Current status and appraisal. In M. C. Wolraith & D. K. Routh (Eds. ), Advances in developmental and behavioral pediatrics. Greenwich, CT: JAI Press. Andrasik, F., Blake, D., & McCarran, M. (1985). A biobehavioral analysis of pediatric headache. In N. A. Krasnego, J. D. Arastek, & M. S. Cataldo (Eds.), Child health behavior: A behavioral pediatrics perspective (pp. 394-434). New York: Wiley. Attanasio, V., Andrasik, F., Burke, E. J., Blake, D. D., Kabela, E., & McCarran, M. S. (1985). Clinical issues in utilizing biofeedback with children. Clinical Biofeedback and Health, 8, 134-141. Benson, H. (1975). The relaxation response. New York: Morrow. Berkow, R. (1977). The Merck manual of diagnosis and therapy. (13th ed.). Rahway, NJ: Merck. Bhatara, V., Arnold, L. E., Lorance, T., & Gupta, D. (1979). Muscle relaxation therapy in hyperkinesis: Is it effective? Journal of Learning Disabilities, 12, 49-53.
393
Bille, B. (1962). Migraine in school children. Acta Paediatrica (Suppl.), 136, 1-151. Bird, B., & Cataldo, M. (1978). Experimental analysis of EMG feedback in treating dystonia. Annals of Neurology, 3, 310315. Blanchard, E. B., & Andrasik, F. (1982). Psychological assessment and treatment of headache: Recent developments and emerging issues. Journal ofConsulting and Clinical Psychology, 50, 859-879. Borkovec, T. D., & Nau, S. D. (1972). Credibility of analogue therapy rationales. Journal of Behavior Therapy and Experimental Psychiatry, 3, 257-260. Borkovec, T. D., & Sides, J. K. (1979). Critical procedural variables related to the physiological effects of progressive relaxation: A review. Behavior Research and Therapy, 17, 119126. Braud, L. W. (1978). The effects of frontal EMG biofeedback and progressive relaxation upon hyperactivity and its behavioral concomitants. Biofeedback and Self-Regulation, 3, 69-89. Braud, L. W., Lupin, M. N., & Braud, W. G. (1975). The use of electromyographic biofeedback in the control of hyperactivity. Journal of Learning Disability, 7, 420-425. Burish, T. G. (1983). EMG biofeedback in the treatment of stressrelated disorders. In C. Prokop & L. Bradley (Eds.), Medical psychology: Contributions to behavioral medicine (pp. 395421). New York: Academic Press. Carter, J. L., & Russell, H. L. (1980). Biofeedback and academic attainment ofLD children. Academic Therapy, 15, 483-486. Carter, I. L., & Russell, H. L. (1981). Changes in verbal-performance IQ discrepancy scores after left hemisphere EEG frequency control training-A pilot report. American Journal of Clinical Biofeedback, 4, 66--67. Carter, J. L., & Russell, H. L. (1985). Use of EMG biofeedback procedures with learning disabled children in a clinical and an educational setting. Journal ofLearning Disability, 18, 214-216. Cataldo, M. F., Bird, B. L., & Cunningham, C. E. (1978). Experimental analysis of EMG feedback in treating cerebral palsy. Journal of Behavioral Medicine, 1, 311-322. Cerulli, M. A., Nikoomanesh, P., & Schuster, M. M. (1979). Progress in biofeedback conditioning for fecal incontinence. Gastroemerology, 76, 742-746. Cleeland, C. S. (1981). Biofeedback as a clinical tool: Its use with the neurologically impaired patient. InS. B. Filskov & T. I. Boll (Eds.), Handbook ofclinical neuropsychology (pp. 734753). New York: Wiley. Cobb, D. E., & Evans, J. R. (1981). The use of biofeedback techniques with school aged children exhibiting behavioral and/or learning problems. Journal of Abnormal Child Psychology, 9, 251-281. Connors, C. K. (1972). Pharmocotherapy of psychopathology in children. In H. Quay & I. S. Werry (Eds.), Psychopathological disorders of childhood (pp. 316-347). New York: Wiley. Cott, A., Pavloski, R. P., & Black, A. H. (1979). Reducing epileptic seizures through operant condition of central nervous system activity: Procedural variables. Science, 203, 7375. Cunningham, M. D., & Murphy, P. J. (1981). The effects of
394
CHAPTER 20
bilateral EEG biofeedback on verbal, visual-spatial, and creative skills in learning disabled male adolescents. Journal of Learning Disabilities, 14, 204-208. DeBacher, G. (1983). Biofeedback in spasticity control. In J. V. Basmajian (Ed.), Biofeedback: Principles and practice for clinicians (2nd ed., pp. 111-129). Baltimore: Williams & Wilkins. Denkowski, K. M., & Denkowski, G. C. (1984). Is group progressive relaxation training as effective with hyperactive children as individuai"EMG biofeedback treatment? Biofeedback and Self-Regulation, 9, 353-364. Denkowski, K. M., Denkowski, G. C., & Omizo, M. H. ( 1983). The effects of EMG-assisted relaxation training on the academic performance, locus of control, and self-esteem of hyperactive boys. Biofeedback and Self-Regulation, 8, 363-375. Denkowski, K. M., Denkowski, G. C., &Omizo, M. H. (1984). Predictors of success in the EMG biofeedback training of hyperactive male children. Biofeedback and Self-Regulation, 9, 253-264. Dodrill, C. B. (1981). Neuropsychology of epilepsy. In S. B. Filskov & T. J. Boll (Eds.), Handbook of clinical neuropsychology (pp. 366-398). New York: Wiley. Doehring, D., Hoshako, 1., & Bryans, B. (1979). Statistical classification of children with reading problems. Journal of Clinical Neuropsychology, 1, 5-16. Donhoffer, H., & Lissak, K. ( 1962). EEG changes associated with the elaboration of conditioned reflexes. Acta Physiologica Academy of Science, Hungary, 21, 249-255. Dunn, F. M., & Howell, R. J. (1982). Relaxation training and its relationship to hyperactivity in boys. Journal ofClinical Psychology, 38, 92-100. Elmore, A.M., & Tursky, B. (1981). A comparison of two psychophysiological approaches to the treatment of migraine. Headache, 21, 93-101. Engel, B. T., Nikoomanesh, P., & Schuster, M. M. (1974). Operant conditioning of rectosphincteric responses in the treatment of fecal incontinence. New England Journal of Medicine, 290, 646-649. Epilepsy Foundation of America. (1975). Basic statistics on the epilepsies. Philadelphia: Davis. Fenichel, G. M. (1985). Migraine in children. Neurology Clinics, 3, 77-94. Finley, W. W. (1976). Effects of sham feedback following successful SMR training in an epileptic: A follow-up study. Biofeedback and Self-Regulation, 1, 227-235. Finley, W. W. (1983). Biofeedback with children. In C. E. Walker & M. C. Roberts (Eds.), Handbook of clinical child psychology (pp. 1050-1068). New York: Wiley: Finley, W. W., Etherton, M.D., Dickman, D., Karimian, D., & Simpson, R. W. (1981). A simple EMG-reward system for biofeedback training of children. Biofeedback and Self-Regulation, 6, 169-180. Finley, W. W., Niman, C. A., Standley, J., & Wansley, R. A. (1977). Electrophysiologic behavior modification of frontal EMG in cerebral palsied children. Biofeedback and Self-Regulation, 2, 59-79. Gargiulo, R. M., & Kuna, D. J. (1979). Arousal level and hyperkinesis: Implications for biofeedback. Journal of Learning Disabilities, 12, 4-5.
Gracenin, C. T., & Cook, J. E. (1977). Alpha biofeedback and LD children. Academic Therapy. 12, 275-279. Green, E. E., Green, A. M., Walters, E. D., Sargent, J. D., & Meyer, R. G. (1975). Autogenic feedback training. P-sychotherapy and Psychosomatics, 25, 88-98. Haber, J. D., Kuczmierczyk, A. R., & Adams, H. E. (1985). Tension headaches: Muscle overactivity or psychogenic pain? Headache, 25, 23-29. Hoelscher, T. J., & Lichstein, K. L. (1984). Behavioral assessment and treatment of child migraine: Implications for clinical research and practice. Headache, 24, 94-103. Holroyd, K. A., Penzien, D. B., Hursey, K. G., Tobin, D. L., Rogers, L., Holm, J. E., Marcille, P. J., Hall, J. R., &Chila, A. G. (1984). Change mechanisms in EMG biofeedback training: Cognitive changes underlying improvements in tension headache. Journal of Consulting and Clinical Psychology, 52, 1039-1053. Howe, R. C., & Sterman, M. B. (1972). Cortical-subcortical EEG correlates of suppressed motor behavior during sleep and waking in the cat. Electroencephalography and Clinical Neurophysiology, 32, 681-695. Hunter, S. H., Russell, H. I., Russell, E. D., &Zimmerman, R.I. (1976). Control of fingertip temperature increases via biofeedback in learning disabled and normal children. Perceptual and Motor Skills, 43, 743-755. Johnston, M. V., & Freeman, J. M. (1981). Pharmacologic advances in seizure control. Pediatric Clinics ofNorthAmerica, 28, 179-194. Kewman, D. C., &Roberts, A. H. (1983). An alternative perspective on biofeedback efficacy studies: A reply to Steiner and Dince. Biofeedback and Self-Regulation, 8, 487-497. Kuhlman, W. N. (1978). EEG feedback training of epileptic patients: Clinical and electroencephalographic analysis. Electroencephalography and Clinical Neurophysiology, 45, 699710. Labbe, E. L., & Williamson, D. A. (1984). Treatment of childhood migraine using autogenic feedback training. Journal of Consulting and Clinical Psychology, 52, 968-976. Lambert, N. M., Sandoval, J., & Sassone, D. (1978). Prevalence of hyperactivity in elementary school children as a function of social system definers. American Journal of Orthopsychiatry, 48, 446-463. Leslie, S. C., Davidson, R. J., & Batey, 0. B. (1985). Purdue pegboard performance of disabled and normal readers: Unirnanual versus bimanual differences. Brain and Language, 24, 359-369. Linkenhoker, D. (1983). Tools of behavioral medicine: Applications of biofeedback treatment for children and adolescents. Developmental and Behavioral Pediatrics. 4, 16-20. Lobar, J. F. (1982). EEG operant conditioning in severe epileptics: Controlled multidimensional studies. In L. White & B. Tursky (Eds.), Clinical biofeedback: Efficacy and mechanisms (pp. 288-310). New York: Guilford Press. Lubar,J. F., &Bahler, W. W. (1976). Behavioral management of epileptic seizures following EEG biofeedback training of the sensorimotor rhythm. Biofeedback and Self-Regulation, I. 77-104. Lobar, J. F., Bianchini, K. ]., Calhoun, W. H., Lambert, E. W., Brody, Z. H., & Shabsin, H. S. (1985). Spectral analysis of
BIOFEEDBACK
EEG differences between children with and without learning disabilities. Journal of Learning Disabilities, 7, 403-408. Lobar, J. F., & Shouse, M. N. (1976). EEG and behavioral changes in a hyperkinetic child concurrent with training of the sensorimotor rhythm (SMR): A preliminary report. Biofeedback and Self-Regulation, I, 293-306. Lobar, J. 0., & Lubar, J. F. (1984). Electroencephalographic biofeedback of SMR and beta for treatment of attention deficit disorders in a clinical setting. Biofeedback and Self-Regulation, 9, 1-23. Marzuk, P.M. (1985). Biofeedback for gastrointestinal disorders: A review of the literature. Annals ofInternal Medicine,103, 240-244. Masek, B. J., Russo, D. C., & Varni, J. W. (1984). Behavioral approaches to the management of chronic pain in children. Pediatric Clinics of North America. 3/, 1113-1131. Masterson,]. P.,Jr., & Turley, W. B. (1980). Biofeedback training with children. American Journal ofClinical Biofeedback, 3. 137-143. Miller, N. E. (1969). Learning of visceral and glandular responses. Science, 163, 434-445. Murphy, P. J., Darwin, I., & Murphy, D. (1977). EEG feedback training for cerebral dysfunction: A research program with learning disabled adolescents. Biofeedback and Self-Regulation, 2, 288. Nail, A. (1973). Alpha training and the hyperkinetic child: Is it effective? Academic Therapy, 9, 5-19. O'Leary, K. D., Pelham, W. E., Rosenbaum, A., &Price, G. H. (1976). Behavior treatment of kinetic children: An experimental evaluation of its usefulness. Clinical Pediatrics. 15, 510-515. Omizo, M. M. (1980a). The effects of biofeedback-induced relaxation training in hyperactive adolescent boys. The Journal of Psychology, /05, 131-138. Omizo, M. M. (1980b). Theeffectsofrelaxationandbiofeedback training on dimensions of self-concept (DOSC) among hyperactive male children. Educational Research Quarterly, 5, 22-30. Omizo, M. M., & Michael, W. B. (1982). Biofeedback-induced relaxation training and impulsivity, attention to task, and locus of control among hyperactive boys. Journal of Learning Disabilities, /5, 414-416. Omiro, M. M., & Williams, R. E. (1982). Biofeedback-induced relaxation training as an alternative for the elementary school learning disabled child. Biofeedback and Self-Regulation, 7, 139-148. Rigg, C. A. (1975). Migraine in children and adolescents. Acta Paediatrica Scandinavica (Suppl.), 256, 19-24. Roberts, A. C. (1966). The aphasic child. Springfield, IL: Thomas. Roberts, A. H. (1985). Biofeedback: Research, training and clinical roles. American Psychologist, 40, 938-941. Roth, S. R., Sterman, M. B., & Clemente, C. D. (1967). Comparison of EEG correlates of reinforcement, internal inhibition and sleep. Electroencephalography and Clinical Neurophysiology, 23, 509-520. Rourke, B. T. (1981). Neuropsychological assessment of children with learning disabilities. In S. Filskov & T. Boll (Eds.), Handbook of clinical neurop.rychology (pp. 453-480). New York: Wiley.
395
Rourke, B. T. (1985). Neuropsychology of learning disabilities: Essentials of subtype analysis. New York: Guilford Press. Saper, J. R. (1983). Headache disorders: Current concepts and treatment strategies. Boston: Wright. Sargent, J.D., Green, E. E., & Walters, E. D. (1973). The use of autogenic feedback techniques in the treatment of migraine headaches. Psychosomatic Medicine, 35, 129-135. Satterfield, J., & Dawson, M. (1971). Electrodermal correlates of hyperactivity in children. Psychophysiology, 8, 191197. Sattler, J. M. (1982), Assessment of children's intelligence and special abilities (2nd ed.). Rockleigh, NJ: Allyn & Bacon. Satz, P., Rardin, D .• & Ross, J. (1971). An evaluation of a theory of specific developmental dyslexia. Child Development, 42, 2009-2021. Schultz, J. H., & Luthe, W. (1969). Autogenic therapy (Vol. 1). New York: Grune & Stratton. Seeburg, K. N., & DeBoer, K. F. (1980). Effects of EMG biofeedback in diabetes. Biofeedback and Self-Regulation, 5, 289-294. Seeger, B. R., & Caudrey, D.J. (1983). Biofeedback therapy to achieve symmetrical gait in children with hemiplegic cerebral palsy: Long-term efficacy. Archives of Physical Medicine and Rehabilitation, 64, 160-162. Seeger, B. R., Caudrey, D.J., & Scholes, I. R. (1981). Biofeedback therapy to achieve symmetrical gait in hemiplegic cerebral palsied children. Archives ofPhysical Medicine andRehabilitation, 62, 364-368. Seifert, A. R., & Lobar, J. F. (1975). Reduction of epileptic seizures through EEG biofeedback training. Biological Psychology, 3, 81-109. Shepherd, K., Hickstein, R., & Shepherd, R. (1983). Neurogenic faecal incontinence in children with spina bifida: Rectosphincteric responses and evaluation of a physiological rationale for management, including biofeedback conditioning. Australian Pediatric Journal, 19, 97-99. Shirley, M., Burish, T., & Rowe, C. (1982). Effectiveness of multiple-site EMG biofeedback in the reduction of arousal. Biofeedback and Self-Regulation, 7, 167-184. Shouse, M. N., & Lobar, J. F. (1979). Operant conditioning of EEG rhythms and Ritalin in the treatment of hyperkinesis. Biofeedback and Self-Regulation, 4, 299-312. Silver, B. V., & Blanchard, E. B. (1978). Biofeedback and relaxation training in the treatment of psychophysiological disorders: Or are the machines really necessary? Journal ofBehavioral Medicine, 1, 217-239. Sterman, M. B. (1982). EEG biofeedback in the treatment of epilepsy: An overview circa 1980. In L. White & B. Tursky (Eds.), Clinical biofeedback: Efficacy and mechanisms (pp. 311-330). New York: Guilford Press. Sterman, M. B., & Friar, L. (1972). Suppression of seizures in an epileptic following sensorimotor EEG feedback training.
Electroencephalography and Clinical Neurophysiology, 33, 89-95. Sterman, M. B., & Macdonald, L. R. (1978). Effects of central cortical EEG feedback training on the incidence of poorly controlled seizures. Epilepsia, 19, 207-222. Sterman, M. B., &Shouse, M. N. (1980). Quantitative analysis of training, sleep EEG and clinical response to EEG operant
396
CHAPTER 20
conditioning in epileptics. Electroencephalography and Clinical Neurophysiology, 49, 558-576. Stennan, M. B., LoPresti, R. W., & Fairchild, M. D. (1969). Electroencephalographic and behavioral studies of monomethylhydrazine toxicity in the cat. (Technical report, AMRL-TR-69-3). Air Systems Command, Wright-Patterson Air Force Base, Ohio. Stennan, M. B., Macdonald, L. R., & Stone, R. V. (1974). Biofeedback training of the sensorimotor rhythm in man: Effects on epilepsy. Epilepsia, 15, 395-416. Surwit, R., Allen, L., Gilgor, R., & Duvic, M. (1982). The combined effect of prazosin and autogenic training on cold reactivity in Raynaud's phenomenon. Biofeedback and Self-Regulation, 7, 537-544. Tansey, M. A. (1984). EEG sensorimotor rhythm biofeedback training: Some effects on the neurologic precursors of learning disabilities. International Journal of Psychophysiology, 1, 163-177. Turk, D. C., Meichenbaum, D. H., & Bennan, W. H. (1979). Application of biofeedback for the regulation of pain: A critical review. Psychological Bulletin; 86, 1322-1338. Wald, A. (1981 ). Use of biofeedback in treatment offecal incontinence in patients with meningomyelocele. Pediatrics, 68, 45-49. Wald, A. (1983). Biofeedback for neurogenic fecal incontinence: Rectal sensation is a detenninant of outcome. Journal of Pediatric Gastroenterology and Nutrition, 2. 302-306. Walden, E. L., & Thompson, S. A. (1981). A review of some alternative approaches to drug management of hyperactivity in children. Journal of Learning Disability. 14, 213-217. Walmsley, R. P., Crichton, L., & Droog, D. (1981). Music as a
feedback mechanism for teaching head control to severely handicapped children: A pilot study. Developmental Medicine and Child Neurology, 23, 739-746. Weiss, G., Kruger, E., Danielson, V., & Elman, M. (1975). Effects of long-tenn treatment with methylphenidate. Canadian Medical Association Journal, 112, 159-165. Werder, D. S., & Sargent, J. D. (1984). A study of childhood headache using biofeedback as a treatment alternative. Headache. 24, 122-126. White, J. J., Suzuki, H., El Shafie, M., Kumar, M., Haller, J. A., & Schnaufer, L. (1972). A physiologic ratiOArue for the management of neurologic rectal incontinence in children. Pediatrics, 49, 888-893. Whitehead, W. E., Parker, L. H., Bosmajian, L., Morrill-Corbin, E. D., Middaugh, S., Garwood, M., Cataldo, M. F., & Freeman, J. (1986). Treatment of fecal incontinence in children with spina bifida: Comparison of biofeedback and behavior modification. Archives of Physical Medicine andRehabilitation, 67, 218-224. Whitehead, W. E., Parker, L. H., Masek, B. J., Cataldo, M. F., & Freeman, J. M. (1981). Biofeedback treatment of fecal incontinence in patients with myelomeningocele. Developmental Medicine and Child Neurology, 23, 313-322. Whitmer, P. 0. (1977). EMG biofeedback manipulation of arousal as a test of the overarousal and underarousal theories of childhood hyperactivity. Dissertation Abstracts international, 38, 3423B. (University Microfilms No. 77-28941, 95). Wyler, A. R., Robbins, C. A., & Dodrill, C. B. (1979). EEG operant conditioning for control of epilepsy. Epilepsia, 20, 279-286.
21 Approaches to the Cognitive Rehabilitation of Children with Neuropsychological Impairment JEFFREY W. GRAY AND RAYMOND S. DEAN
Historically, clinical neuropsychology in North America has focused on the prediction and localization of cortical dysfunction (e.g., Boll, 1974; Reitan, 1974; Dean, 1985a). Although this emphasis on diagnosis continues, the growing sophistication of radiological techniques may deemphasize this diagnostic role. A number of authors have argued that clinical neuropsychology will focus on the understanding of the patient's neurological deficits and the planning of rehabilitation approaches (Dean, 1982, 1985a; Diller & Gordon, 1981; Piasetsky, 1981). Cognitive rehabilitation involves the retraining of mental skills or abilities impaired as the result of neurological disorders. Along these lines, Diller and Gordon (1981) have stressed the utility of neuropsychology, with its foundation in brain-behavior relationships, in cognitive rehabilitation. Indeed, arecent survey of practicing neuropsychologists indicated that nearly 50% of the respondents reported cognitive rehabilitation to be part of their clinical practice (Seretny, Dean, Gray, & Hartlage, 1986). Thus, it appears that although the major focus in neuropsychology may continue to be on diagnosis, an increasing number of clinicians seem to be expanding their practice to include some form of cognitive therapy. In light of recent data suggesting the utility of neuropsychological information in the treatment of cognitive deficits in adults (see Diller & Gordon, JEFFREY W. GRAY • Neuropsychology Laboratory, Ball RAYMOND S. State University, Muncie, Indiana 47306. DEAN • Neuropsychology Laboratory, Ball State University, Muncie, Indiana 47306; and Indiana University School of Medicine, Indianapolis, Indiana 46282.
1981) this chapter examines the present status of cognitive rehabilitation treatment with neuropsychologically impaired children. Basic theoretical approaches to rehabilitation are examined, with emphasis on implications for cognitive rehabilitation. In addition, a number of specific cognitive rehabilitation programs are reviewed with a focus on their potential clinical utility with children.
Theoretical Models An array of theoretical approaches to the rehabilitation of the neurologically impaired have been offered (e.g., Barth & Boll, 1981; Bolger, 1981; Diller & Gordon, 1981). Although related, each views neuropsychological impairment and treatment of such deficits rather uniquely. One basic approach to rehabilitation centers around the patient's performance on specific measures of cognitive and neuropsychological functioning. Within this model, dysfunction is conceptualized in terms of impaired performance on psychometric tests. As such, remedial strategies focus upon ameliorating individual test-specific deficits. Although therapy emphasizes specific deficits, generalized cognitive/neuropsyc_hological improvement is hypothesized to occur (e.g., Gudeman, Golden, & Craine, 1978). Using what could be viewed as a mastery learning approach, the ''psychometric'' model (Diller & Gordon, 1981) of cognitive rehabilitation relies on the analysis of task requirements and the systematic shaping and cuing of patient responses (see Block, 1971 , for a review of mastery learning). Early in the rehabilitation process, patients are provided with nu397
398
CHAPTER 21
merous cues and as rehabilitation progresses cues are month), patients receiving the scanning exercises faded out until the patient is able to perform the spe- were compared to groups of brain-damaged patients cific task unaided. With the use of discriminative on a standard occupational therapy regimen. With stimuli and small increments in the difficulty level, minor exceptions, those patients receiving the exthe probability of failure is virtually eliminated for perimental treatment showed significant improvethe patient at any given level of the program. Ben- ment in attention to the complete visual field, comYishay and his colleagues (e.g., Ben-Yishay et al., pared to the control group (Diller & Weinberg, 1979; Diller et al., 1974) reported success in the 1977). Of interest, the patient's ability to scan selecrehabilitation of cognitive deficits using such a "sat- tively all pertinent visual information was reported to urated cuing" approach with brain-damaged adults. generalize to other more complex tasks (e.g., readSpecifically, this treatment strategy has been re- ing, arithmetic, copying). ported useful in remediating deficits on rather comA third model of rehabilitation can best be charplex cognitive tasks as represented by the Block De- acterized as behavioral engineering (Diller & Gorsign and Similarities subtests from the Wechsler don, 1981). From this view of rehabilitation, the paAdult Intelligence Scale (Ben-Yishay et al., 1970a; tient's impairment is defined in terms of operationalDiller, 1976) as well as psychomotor difficulties sec- ired behavioral deficits. Inherent here is the assumpondary to neurological impairment (Ben-Yishay & tion that behavioral deficits are maintained by enSchoon, 1975). vironmental conditions. Therefore, the major objecA second basic approach to cognitive rehabilita- tive in treatment is to identify and systematically tion focuses upon groups or patterns of behaviors modify the environmental antecedents that are as(e.g., Barth & Boll, 1981). Unlike the "psycho- sumed to underlie the problem behavior. To this end, metric" model of rehabilitation, deficits are treated rehabilitation therapists utilize a number of common as parts of a whole rather than as individual dysfunc- behavioral analytic techniques (see Wilson & tions. Whereas the "psychometric" approach tore- O'Leary, 1980, for a review). In keeping with a behavioral orientation, the habilitation focuses predominantly on the patient's response to a given stimulus, the more "biolog- initial stages of treatment are marked by the identifiically" oriented model emphasizes elements of the cation of target behaviors amenable to therapy. Once stimulus itself (Diller & Gordon, 1981). Indeed, one behavioral deficits have been operationalized and of the major tasks of the therapist in this approach is their antecedents identified, the emphasis shifts to the to identify the specific components of a stimulus that establishment of a reliable baseline of behaviors from contribute to the patient's deficit. From this perspec- which to evaluate progress. Finally, the target behavtive, one facet of treatment is to alter the stimulus in ior(s) is made contingent on a systematic program of such a way as to offset the patient's behavioral defi- reinforcement. As was true with the ''psychometric'' cits. Moreover, proponents of this modal of cognitive or task analytic model, the target behavior is systemrehabilitation argue that these stimulus alterations atically shaped in such small increments that failure may bring to the attention of the patient both the is minimized. The difference between the psychonature and extent of his or her disabilities as well as metric and behavioral engineering modes related to potential compensatory strategies. Thus, a major the reliance upon a standardized neuropsychological goal of this approach is to facilitate the patient's un- assessment in the former. Horton (1979a,b, 1981) argued in favor of such derstanding of the impairment. It has been argued that such insight is important if rehabilitation is to a behavioral approach with neuropsychologically imgeneralize to other problem behaviors (Diller & Gor- paired adults and children, citing a relatively large don, 1981). However, it is not clear in this mode of literature. Indeed, in support of this stance, a number therapy how to plan for or facilitate generalization. of investigators have reported behavioral techniques Perhaps the best example of this path to re- to be effective in treating hyperactivity, impulsivity, habilitation can be found in reported attempts to re- and perseveration in brain-damaged children (e.g., mediate spatial neglect in brain-damaged patients Hall & Broden, 1967; Krop, 1971) and short-term (e.g., Diller, 1976; Diller et al., 1974; Diller & . memory disorders in adult head-injured patients Weinberg, 1977; Weinberg & Diller, 1968; Wein- (e.g., Cooke, 1973). However, it is not clear how berg et al., 1977). In accord with the "biological" such an atomistic therapy generalizes to system defimodel of rehabilitation, patients are initially trained cits experienced by the neurologically impaired to perform visual scanning tasks designed to promote patient. Although relatively little theoretical attention awareness of the visual field deficit. Following a predetermined amount of time (usually about l has been directed upon the cognitive rehabilitation of
COGNmVE REHABILITATION
neuropsychological impairment in children (e.g., learning-disabled and head-injured children), anumber of authors have argued that the most appropriate approach with these children may be one that focuses upon their neurological strengths (e.g., Hartlage & Lucas, 1973a,b; Hartlage & Reynolds, 1981; Hartlage & Telzrow, 1983, 1984; Reynolds, 1981; Telzrow, 1985). Specifically, remedial strategies are planned in a task analytic fashion that are thought to complement the child's observed mode of processing information. In an attempt to empirically test this strength model, Hartlage and Lucas (1973a) compared the word recognition performance of a group of normally functioning first grade students receiving reading instruction based on their processing strengths with a group of children in traditional reading instruction. The two groups were comparable in reading readiness at the beginning of the school year. At end of one school year, the group receiving the neuropsychologically based instruction significantly outperformed the traditionally instructed control group. Interestingly, teachers' ratings of the child's reading skills paralleled the findings with the Reading subtest of the Wide Range Achievement Test. Thus, it appears that at least with this sample the matching of remedial strategies to information-processing strengths was successful.
Cognitive Rehabilitation Programs One of the most common sequelae of brain damage in children is a generalized impairment in cognitive functions. This most often involves deficits in memory, attention, and perception (Levin, Benton, & Grossman, 1982; Levin & Eisenberg, l979a,b). With such neuropsychological impairment, it is not surprising that a significant number of these children subsequently experience difficulties in school that require special education services (e.g., Klonoff, Low, & Clark, 1977). Although cognitive deficits have been clearly documented with brain-damaged children, few rehabilitation/remedial programs have been developed specifically to ameliorate such deficits in children. In contrast, a number of programs have been offered for brain-damaged adults. With this in mind, several of the more well-known attempts at cognitive rehabilitation with adults will be reviewed and implications d~wn for children. It is important to note that although intuitively these programs appear to hold clinical utility for brain-damaged children, there are little data to support this argument.
399
Cognitive rehabilitation has been broadly defined as a systematic effort to teach patients to overcome intellectual deficits arising from brain dysfunction (Task Force on Head Injury, 1984). This effort is seen to involve the reinforcement and strengthening of previously learned patterns of cognitive behavior, as well as the establishment of new patterns of cognitive activity that compensate for neurological systems too impaired for functional return to occur (Task Force on Head Injury, 1984). Consistent with this working definition, Ben-Yishay (1981) argued that the objective of cognitive rehabilitation must be to .. overcome, i.e., to correct (and if that is not feasible, then at the very least, to significantly ameloiorate), the effects of generic cognitive deficits in such a way as to enable the individual patient to find alternative and adequate means of achieving specific functional goals" (p. 20). As part of a comprehensive rehabilitation program, Ben-Yishay and his colleagues (e.g., BenYishay & Diller, 1981) have developed a systematic or hierarchical program of cognitive rehabilitation. Divided into five separate, yet interrelated modules, treatment emphasizes. the remediation of both lower level (e.g., attention) and more complex (e.g., verbal-abstract reasoning) cognitive deficits. Modules are organized such that patients initially receive tasks requiring more rudimentary cognitive skills (e.g., psychomotor manipulations), followed by more complicated exercises involving abstract reasoning and mental manipulation. The patient is presented with each module in such a fashion that failure is minimized. To this end, the remedial tasks are initially broken down into their smallest logical component parts. The patient is then guided through the tasks with the aid of prompts. The cues are made less explicit with each trial until the patient is able to perform tasks within the module without error or prompting. In this way the patient procedes in a sequential fashion until the final module is reached. Depending on the nature and degree of the neuropsychological impairment, this program is often highly ly redundant, which is seen to promote generalization (Silver et al., 1983). Ben-Yishay (1981) prioritized domains in cognitive rehabilitation as they related to (1) self-help and daily living; (2) psychomotor, perceptual, and cognitive skills that underlie successful vocational or school functioning; and (3) socioemotional/interpersonal skills. Telzrow (1985) stressed independent eating, dressing, and toileting; adequate perceptualmotor skills; proficiency in reading, writing, and arithmetic; and interpersonal relationships at home and at school as rehabilitation priorities with chil-
400
CHAPTER 21
dren. Clearly, the goal of any rehabilitation program must be the reintegration of the patients in their premorbid environment. For the pediatric patient, school performance must be the focus. As a prelude to implementing cognitive rehabilitation, a complete assessment of the patient's overall neuropsychological functioning is necessary. Specifically, the patient's ability to formulate, plan, and implement goal-directed behaviors, selectively attend to a stimulus, process and retain various forms of information, grasp the essential nature of problem situations, and verbal interaction must be evaluated. Only with such information can the neuropsychologist plan a program of rehabilitation that fits the patient's unique needs while taking advantage of specific strengths. This information is necessary in estimating the patient's readiness to benefit from different remedial strategies (Ben-Yishay, 1983). This patient-specific regimen of rehabilitation is appropriate with both adults and pediatric patients. Of particular interest to the present discussion, Ben-Yishay (1981) offered essentially two conditions to the successful implementation of cognitive remediation. Perhaps most important, the patient must be aware of the potential benefits of the specific exercise and must be motivated to participate in the process. If one makes the assumption that repeated failure on a task may result in motivational deficits (e.g., see Dean & Rattan, 1986; Fowler & Peterson, 1981; Seligman, 1975), it follows that a cognitive rehabilitation program must be constructed to emphasize success and reduce error. As mentioned previously, Ben-Yishay ( 1981) espoused a remedial approach in which tasks are "orchestrated" in such a fashion that success is "guaranteed." Of course, errorless learning is not unique to Ben-Yishay' s ( 1981) program. In fact, small increments of new learning that reduces errors are the root of programmed instruction (see Anderson & Faust, 1967). However, it is clear that programming for success serves to elevate patient motivation and continued participation. Linked to motivation in a complex fashion, a successful cognitive remedial program also depends on the patient's ability to focus attention on the task at hand while ignoring irrelevant stimuli (Ben-Yishay, 1981 ). Underlying the importance of this condition, a number of investigators have showed that attentional deficits may be a concomitant of childhood learning disorders (e.g., Hallahan, 1975; Hallahan, Gajar, Cohen, & Tarver, 1978; Hallahan, Tarver, Kaufman, & Grabeal, 1978; Ross, 1976). Consistent with Ben-Yishay's (1981) methods with braindamaged adults, a number of behavioral techniques have been successfully employed with learning-dis-
abled children manifesting attentional deficits (e.g., Brown & Alford, 1984). Ben-Yishay and his colleagues (e.g., BenYishay, Diller, Gerstman, & Gordon, 1970; BenYishay, Gerstman, Diller, & Haas, 1970; BenYishay, 1983) have examined the clinical utility of a "modular" approach to the remediation of cognitive deficits resulting from cerebral insult. Small sample sizes, restricted patient selection criteria, and assessment procedures that overlap with training tasks notwithstanding, results from these studies suggest that the program may be effective in ameliorating a number of cognitive, perceptual, and psychomotor deficits. Indeed, brain-damaged patients undergoing the treatment regimen have been shown to improve significantly more than those patients receiving a "standard'' rehabilitation program. However, in light of the fact that the results of these preliminary investigations may have been confounded by variables other than treatment, the findings remain equivocal. Indeed, with the selection bias in these investigations, the generalizability of this treatment program to braindamaged patients in general would seem premature. Moreover, one is given to question as to the extent to which the improvement reported in many studies is a function of the remedial program or simply the result of spontaneous recovery (e.g., Diller & Gordon, 1981). Although future research in this area is needed prior to the use of such an approach on a large-scale clinical basis, it seems clear that highly behaviorally oriented programs incorporating principles of task analysis, reinforcement, "saturated" cuing, and shaping may be effective with both neuropsychologically impaired adults and children. A number of investigations using stringent controls have found these behavioral techniques to be useful in treating cognitive deficits found in pediatric and adult patients with neurological deficits (e.g., Brown & Alford, 1984; Cooke, 1973; Hall & Broden, 1967; Krop, 1971). A heuristic value of a behavioral focus may well be related to our primitive understanding of brain-behavior relationships.
A Developmental Approach to Cognitive Rehabilitation Based on a sequential approach to remediation, Bolger ( 1981) developed a comprehensive program of cognitive retraining. Underlying the specific methods is the hypothesis that efficient processing of information is dependent upon the ability to execute cognitive operations in an automatic fashion. Con-
COGNITIVE REHABILITATION
401
ceptualized within a limited capacity framework sug- and exerting adequate cognitive controls over their gesting that only a given amount of attentional capac- behaviors. The actual tasks involve cognitive-beity exists at any point in time (e.g., Case, 1972; Case, havioral techniques. For instance, Bolger and colKurland, & Goldberg, 1982; Hasher & Zacks, 1979), leagues have used cognitive-behavior modification it is hypothesized that many cognitive tasks place an to improve the patient's ability to maintain emotional excessive cognitive demand upon brain-damaged pa- self-control. At this point in the training the patient is tients. Moreover, it has been suggested that these involved in canned •'video games'' aimed at impatients may experience a significant reduction in proving basic math and reading skills. One such available mental capacity evidenced by deficits in game that Bolger (1981) suggested is "Hangman" both storage and retrieval of information (e.g., (marketed by Atari), which is seen to remediate spellBolger, 1981). With this theory as a basis, remedial ing and dysnomic difficulties. Underlying Bolger's (1981) approach to cogefforts are designed to improve the patient's execution of cognitive operations and provide processing nitive rehabilitation is the notion that rudimentary strategies to reduce the cognitive demands. Bolger skills (e.g., attention, word recognition) behaviors (1981) argued that the goal of cognitive remediation must be made ••automatic'' before more higher level •• . . . is to increase the mental capacity of the cognitive skills can be addressed. Based on this reaindividual to process larger amounts of stimuli with soning, each task is practiced repeatedly even after more accuracy and with greater attention to sub- the initial mastery has been reached. This "overtleties" (p. 67). Such an increase in the patient's learning'' is seen as facilitating the patient's ability to ability to process information is seen as a necessary perform a number of important cognitive functions component to the performance of complex cognitive spontaneously. Indeed, such "automatic" processtasks. For Bolger (1981), remedial tasks focusing on ing is a prerequisite for successful educational or rudimentary (e.g., perceptual and attentional) pro- vocational generalization. Also inherent within this cesses as well as higher cortical functions are present- program is an emphasis on remedial tasks that are ed to the patient continuously throughout the re- both entertaining and reinforcing to aid the patient's habilitation program. Of particular importance in this motivation and concentration (Bolger, 1981). Although the cognitive therapy described by program is the emphasis on the patient's ability to Bolger (1981) has clinical appeal, there are a lack of integrate these high-level cognitive functions. Similar to the modular approach of Ben-Yishay data showing its benefits over spontaneous recovery. (e.g., 1981), a three-step paradigm was offered by Future research with groups that differ in specific Bolger (1981). The primary emphasis of the first neurological deficits would allow the examination of stage is prolonging on-task behavior and stresses the the utility of the approach. patient's ability to focus and sustain attention. Consistent with the assumption that the patient must enjoy tasks for the remediation to be successful (Bolger, 1981), this program has utilized a number of Computer Programs in Cognitive commercial microcomputer games. For example, the Rehabilitation popular electronic game "Simon" has been used in With the availability of microcomputers has an attempt to improve the patient's attentional and come a growing reliance on cognitive rehabilitation on-task behaviors. Upon completion of the first phase, the patient software. Indeed, the use of video games and cogadvances to the next stage, which focuses on the nitive rehabilitation routines is not uncommon in discrimination of visual and auditory information. therapy (Bracy, Lynch, Sbordone, & Berrol, 1985). Early in training the emphasis is on the remediation A number of features make microcomputers attracof eye-hand coordination, reaction time, and visual tive in rehabilitation. One of the clearest involves the scanning deficits. As the patient progresses, a fact that computer programs present remedial tasks in number of video games designed to improve visual a novel fashion (Lynch, 1981). Other than the fact discrimination are introduced. Finally, a set of rather that these programs offer the portability and adaptinnovative tasks have been designed to aid the patient ability necessary for most clinical settings, they also allow "individualization" of treatment in a costin developing spatial skills. After meeting the minimum requirements for effective manner. In addition, the cost of the base stage 2, the patient progresses to a third and final equipment continues to fall within the means of pamodule. This training is designed to improve skills in tients, thus enabling the practice of skills in the making data-based decisions, thinking abstractly, home.
402
CHAPTER 21
A number of computer-based cogmuve remediation programs have been developed for use with brain-damaged adults and children (e.g., Bracy, 1983). Clinically based, these programs focus on improving the patient's functioning in selective and sustained attention, verbal and nonverbal auditory and visual discrimination, and stimulus differentiation and generalization. Although a number of computerbased rehabilitation programs have been argued to be effective in remediating cognitive deficits (e.g., Bracy, 1983; Bracy et al., 1985), little empirical $Upport exists for such claims. In the main, these claims are supported by case studies showing cognitive improvement in both children and adults who have been involved with the programs (Bracy, 1983). Although of interest, the crucial test of any program comes with a comparison with untreated groups. As opposed to the programs described above, which require psychological supervision, computer programs with ·unsubstantiated validity can be misinterpreted as a panacea. Generally speaking, computer programs are organized such that patients can work at home and typically involve approximately 2 to 5 hours per day. Similar to the therapy programs previously discussed, this approach initially focuses on more rudimentary cognitive functions such as attention and stimulus discrimination and generalization. Upon mastery of these basic cognitive skills, the patient is administered computer-based programming that provides training in higher level areas such as memory and problem-solving. However, unlike the clinically based programs described above, little professional supervision or evaluation is involved. Generally speaking, the utilization of video games in rehabilitation settings has centered around reading and math skills, memory, visual-perceptual functioning, and attention and concentration (see Lynch, 1979, 1981, for a review). For example, a number of Atari games require patients to label and spell words of varying degrees of difficulty, as well as to add, subtract, multiply, and divide simple and complex number strings. Video games are available that are said to emphasize short-term and long-term memory of both visual and verbal stimuli (Lynch, 1979, 1981). Other games focus more on motor coordination, attention, planning, and visual scanning (Lynch, 1979, 1981). Aside from the systematic repetition of rehabilitation routines, microcomputers hold a clear potential for automating and individualizing cognitive rehabilitation in an economical fashion. Indeed, the use of computers, programmed to automatically record data, would free the neuropsychologist
from record keeping, allowing a focus on structuring cognitive therapy that meets the patient's assessed needs. Moreover, unlike most labor-intensive approaches to cognitive rehabilitation, the psychologist could monitor several patients' performances simultaneously. Of particular interest to the present discussion, it seems that computer-based cognitive rehabilitation may lend itself to use with children as well as adults. Although little evidence supports this notion, Lynch (1979) suggested that such a computer-based program has been successful in improving the phonetic reading ability of children with learning disorders. Parenthetically, the microcomputer has been argued to be a valuable tool in diagnosis as well as remediation of cognitive deficits (Lynch, 1981). Indeed, the microcomputer holds a very clear potential in neuropsychological assessment. However, with the proliferation of computer-based assessment procedures, one should examine reliability and validity data with the same rigor applied to any standardized assessment procedure. Although a number of authors have extolled the virtues of a computer-based rehabilitation program, few controlled studies accounting for spontaneous recovery exist. To be sure, the majority of supporting evidence has been based on case studies. Consequently, the generalizability of these results is highly questionable. Investigations using larger numbers of patients, adequate control groups, and appropriate criterion measures would allow a test of the potential for many unsubstantiated claims.
Neurobehavioral Approach to Cognitive Rehabilitation For many, the assessment of neuropsychological functions is tantamount to understanding the problem. It is noteworthy that few cognitive rehabilitation strategies consider the potential interaction between neuropsychological impairment and emotional characteristics. Indeed, attempts that have focused on various underlying cognitive processes are often made to the exclusion of the patient's behavior history and learned methods of coping with failure (e.g., Dean, 1978). This is a rather curious state of affairs when one considers the frequency with which patients having neurological disorders also present with maladaptive emotional patterns (see Boll, 1981; Dean, 1985b). These emotional factors would seem especially prominent in children with cognitive deficits because reintegration into their premorbid environment involves return to school.
COGNITIVE REHABILITATION
This emotional-cognitive dysfunction is portrayed quite clearly in children with long-standing learning disorders. Indeed, negative reactions to specific cognitive tasks and school in general exist in a large number of school-age children but may be masked by seemingly unrelated behaviors (e.g., withdrawal, lack of compliance). So too, recent research indicates that neurologically impaired children may develop maladaptive methods of coping with the stress of cognitive failure (Bender, 1985; Dean & Rattan, 1986). Because of the length of time thy have attempted to cope with the disorder, this cognitive impairment-emotional reaction may be more clearly portrayed with congenital learning disorders. However, such problems are evident in a study by Klonoff and Low (1974) of children with closed head injuries. In fact, children may develop what would be likened to a phobic reaction in an attempt to cope with perceived cognitive dysfunction. Aversive reactions are seen to go beyond the immediate therapy session to the creation of an emotional reaction to those cognitive skills impaired. Unlike the adult who may attempt to avoid situations in which cognitive deficits are highlighted, the child who is expected to return to school has few choices in participating. Thus, what begins as a neuropsychological dysfunction may lead to a response set of failure-aversion-failure, as the child attempts to cope with the stress of cognitive demands. Systematic desensitization is a behavioral therapy aimed at modifying phobic responses. This procedure includes the identification of an individual's hierarchy of aversive reactions to stimuli and then proceeds to pair positive (reinforcing) events with those that have produced negative reactions (see Lang, 1964; Wolpe, 1969). This approach has been shown to be successful in desensitizing children's aversions and irrational emotional responses (see Wolpe, 1969). With such a perspective in mind, it would seem that this treatment would be of utility in treating children's acquired aversive reactions to cognitive tasks. Specifically, this procedure may be applicable during the actual process of cognitive therapy. Suggested some years ago by Severson (1970) with learning-disabled children, this format would allow the cognitive therapist to offer both systematic desensitization of emotional aspects and cognitive rehabilitation simultaneously. In general, ''canned programs'' that attempt to focus on children's processing deficits have not been shown to remediate cognitive impairment (e.g., Ayres, 1972; Kirk & Becker, 1963). However, the utility of neuropsychological procedures, which al-
403
low individualization, seems promising in planning treatment that capitalizes on individual strengths (Golden, 1978; Hartlage, 1975; Hynd & Obrzut, 1981; Luria, 1963; Reynolds, 1981; Rourke, 1976). Support for this diagnostic position comes from a number of studies using a neuropsychological orientation (Hartlage, 1975; Hartlage & Lucas, 1973a; Hynd & Obrzut, 1981). These findings have led some to argue in favor of therapy based on the assessment of neuropsychological processing styles (Gunnison, Kaufman, & Kaufman, 1982; Kaufman & Kaufman, 1979; Reynolds, 1981). However, few investigators have integrated this approach with more emotional-learning-based interventions. It has been hypothesized that children with neurological deficits cannot be treated simplistically from either a cognitive or a mental health point of view (Dean, 1982). It would seem that children with cognitive impairment would benefit from an approach that offers academic remediation while attempting to modify negative emotional responses. In a recent study in our laboratory, both the goals of cognitive therapy and desensitization of negative emotional reactions were considered simultaneously. Each of these objectives was applied in a systematic fashion during therapy sessions. Sessions, while concentrating on cognitive skills, were structured so as to desensitize the child's emotional reactions and reinforce appropriate coping behaviors. Following a complete neuropsychological assessment, a hierarchy of remedial tasks were constructed for each child along an approach-avoidance continuum. Much like Ben-Yishay's (1981) program, sessions were structured using a task analytic approach. In effect, this procedure is akin to the philosophy of systems analysis, where aspects of a task are systematically explored. Moreover, task analysis structures cognitive tasks in a step-by-step fashion, with the patients experiencing success, regardless of their presumed level of cognitive functioning. In this mode of cognitive rehabilitation, many of the desirable qualities of neuropsychological tests were used. A neuropsychological evaluation provided an understanding of how children best process information. This evaluation also allowed establishment of a benchmark to follow the success of therapy. In addition, however, the observation of children as they learn was retained and considered valuable in structuring individual programs. Rather than seeking to make differential diagnosis between various underlying neurological disorders, this program approached the cognitive deficit through the use of a system that reinforced attempts,
404
CHAPTER 21
as well as successes, on the approach-avoidance hierarchy. Patients choose rewards that correspond to the task level they select from their individually prescribed cognitive skills hierarchy. Near the top of this task-based hierarchy have been placed important cognitive skills that are basic to remediation but may also have generated avoidance behavior due to a neurological disorder. Levels of the child's hierarchy ranged from the most obviously cognitively related tasks to simply talking with the therapist. Of interest also, the patient is able to choose tasks anywhere on the approach-avoidance continuum and receives differential levels of reinforcement depending on the perceived difficulty of the task. In an attempt to study the utility of this approach, a group of children with long-standing cognitive deficits were seen once a week for 8 months. The results showed significant gains in rehabilitating reading deficits and aiding patients in coping with the stress offailure. Specifically, when compared with a control group, children in the treatment group made significant gains in academic skills, rated classroom behaviors, and the ability to respond concomitant with their measured skill level even after obvious failure. Interestingly, although rehabilitation was gained in specific cognitive areas, these impaired children showed little improvement in specific neuropsychological functions. Thus, the therapy would appear to offer the child methods of compensation for neuropsychological impairment.
as the academic productivity of an 11-year-old male with attentional deficits. Importantly, the behavioral effects of this remedial strategy were maintained over a follow-up examination at 2 Y2 months. Although the single subject design of this investigation obscures the generalizability of the findings, it appears that a self-monitoring procedure may be effective in the remediation of some attentional deficits. The validity of the approach was supported in a related study by Brown and Alford (1984), who used a self-instructional technique to remediate both attentional deficits and academic difficulties in 20 learning-disordered children. Indeed, children receiving the cognitive training showed significant improvement on tests of learning aptitude, reading recognition, and the Matching Familiar Figures Test compared to the control group. In a somewhat different approach to selective attention deficits, Ben-Yishay and his colleagues (e.g., Ben-Yishay, Diller, & Rattok, 1978; BenYishay et al., 1980) developed a set of tasks that focus on systematic remediation. Organized in a hierarchy, these tasks require the patient to actively respond to stimulus lights, estimate time, consciously scan and identify various stimulus signals, and freely discriminate auditory and visual stimuli. The lack of a control group notwithstanding, Rattok et al. ( 1982) reported that these techniques significantly improved the selective attention of head-injured patients. A number of rather innovative techniques have been proposed for use with brain-damaged patients who experience perceptual deficits. For example, Diller and his associates (e.g., Diller et al., 1974; Strategies of Cognitive Rehabilitation Diller & Weinberg, 1977) reported the effective use of visual cancellation exercises in the remediation of Although relatively few empirically based pro- visual scanning deficits. Initially the patient is taught grams of cognitive rehabilitation presently exist, a to utilize a left/righ~ field anchor to compensate for number of specific techniques have been studied in his or her visual neglect. After this anchoring proefforts to treat attentional, perceptual, and memory cedure is learned, visual stimuli are presented in the deficits often associated with neurological impair- neglected field until the patient is able to follow the ment. The frequency of attentional deficits in chil- stimuli to the outermost edges of that visual field. dren with brain damage and neurologically based Weinberg et al. (1977) reported data consistent with learning disabilities is well documented (Brown & anchoring skill generalization to paper and pencil Alford, 1984). Rehabilitation programs for the most exercises as well as reading tasks. It has been well established that brain-damaged part have included some form of cognitive self-control or, if you will, self-monitoring techniques. In the individuals often present with deficits in encoding, main, research in this area has reported success in storage, and/or retrieval of information (e.g., Levin increasing attention (Brown & Alford, 1984; Hal- et al., 1982). Remediating memory difficulties has lahan & Sapona, 1983). The procedure involves long occupied neuropsychological research and teaching the patient to cognitively monitor on-task clinical efforts. These disorders may relate to both behavior via an array oflearned self-messages (e.g., attention problems as well as actual functional disor"Was I paying attention?"). Using such a self- ders of regulation, storage, or retrieval of informamonitoring technique, Hallahan and Sapona (1983) tion. Consistent with the notion that children with reported improvement of the on-task behavior as well neuropsychologically based learning disorders may
COGNmVE REHABILITATION
fail to spontaneously utilize mnemonic strategies such as rehearsal (e.g., Bauer, 1977; Hallahan & Sapona, 1983; Tarver, Hallahan, Kaufman, & Ball, 1976), remedial efforts often focus on the systematic instruction of the child in the effective use of mnemonic strategies (Hallahan & Sapona, 1983). Along these same lines, various mnemonic techniques have been used with brain-damaged patients (e.g., Gianutsos & Gianutsos, 1979; Jones, 1974; Leftoff, 1981; Lewinshon, Danaher, &Kikel, 1977). For instance, Leftoff (1981) reported that similar to normals when verbal information (i.e., high-frequency nouns) was presented in a consistent order, patients with left hemispheric dysfunction recalled significantly more words than when the words were presented in a random fashion. Based on these data, Leftoff concluded that ordered information may serve as a salient cuing device. In other words, patients were better able to retrieve verbal information if it was presented in an organized fashion. In a related investigation, Jones (1974) showed that visual imagery was effective in improving the pairedassociate learning of patients with documented left temporal-lobe lesions. In this study, Jones initially presented patients with word pairs accompanied by drawings depicting the two words interacting with one another (e.g., for the word pair elephant-bouquet, the picture depicted an elephant holding a bouquet of flowers). After a number of practice trials, the patient was required to develop his or her own unique images for each pair of words. Importantly, a followup of a number of the original patients indicated that they continued to use this technique on a daily basis (Jones, 1974). Although cognitive rehabilitation with neuropsychologically impaired children is in its infancy, it appears that a number of strategies have been effectively utilized to remediate cognitive deficits. Indeed, the early relevant data seem quite promising.
best (Adamovich, Henderson, & Auerbach, 1985). Whereas Ben-Yishay's (e.g., 1983) and our program have received some support, it is clear that with some 50% of neuropsychologists reporting involvement in cognitive rehabilitation, the research base of this practice is wanting. Although many of these programs have intuitive appeal for use with neuropsychologically impaired children, little data support the use of these programs. It seems clear that neuropsychology has the potential to continue contributing to rehabilitation in general and cognitive retraining specifically. Indeed, the inertia of the field is moving toward a treatment emphasis (Dean, 1982, 1985a). Neuropsychology with its emphasis on brain-behavior relationships may be the most relevant approach to remediating neurologically related cognitive deficits. Although such a trend in neuropsychology is clear (see Serebly eta/., 1986), we may be entering the field without benefit of a firm empirical foundation.
References Adamovich, B. B., Henderson, J., & Auerbach, S. (1985). Cognitive rehabilitation of closed head injured patients. San Diego: College-Hill Press. Anderson, R. L., & Faust, G. F. (1967). The effects of strong formal prompts in programmed instruction. American Educational Research Joumal, 4, 345-352.
Ayres, A. J. (1972). Sensory integration and learning disorders. Los Angeles: Western Psychological Services. Barth, J. T., & Boll, T. J. (1981). Rehabilitation and treatment of central nervous system dysfunction: A behavioral medicine perspective. In Medical psychology: Contributions to behavioral medicine (pp. 241-266). New York: Academic Press. Bauer, R. H. ( 1977). Memory processes in children with learning disabilities: Evidence for deficient rehearsal. Journal of Experimental Child Psychology, 24, 415-430.
Bendet, W. N. (1985). Differences between learning disabled and non-learning disabled children· in temperament and behavior. Learning Disability Quarterly, 8, 11-18.
Conclusion A number of cognitive rehabilitation programs presently exist. However, there is a general paucity of research on the efficacy of these programs. Where data do exist, there are inherent difficulties in making clear conclusions concerning the cognitive rehabilitation approaches useful with neuropsychologically impaired patients. Problems relating to the heterogeneity of samples, lack of control groups, and failure to control for spontaneous recovery combine to make the results of such investigations equivocal at
405
Ben-Yishay, Y. (1981). Cognitive remediation after TBD: Toward a definition of its objectives, tasks, and conditions. In Working approaches to remediation of cognitive thficits in brain damaged persons (Rehabilitation Monograph No. 62). New York: New York University Medical Center, Institute of Rehabilitation Medicine. Ben-Yishay, Y. (1983). Cognitive remediation viewed from the perspective of a systematic clinical research program in rehabilitation. Cognitive Rehabilitation, 1, 4-6. Ben-Yishay, Y., Ben-Nachum, Z., Cohen, A., Gerstman, L., Gordon, W.,Gross, Y.,Hofien,D.,Piasetsky,E.,&Rattok, J. (1978, June). Working approaches to remediation of cognitive deficits in brain damaged persons. Supplement to the sixth annual worlcshop for rehabilitation professionals, New York.
406
CHAPTER 21
Ben-Yishay, Y., & Diller, L (1981). Rehabilitation of cognitive Newsletter for Research in Mental Health and Behavioral and perceptual defects in people with traumatic brain Sciences, I5, 43-46. damage_ International Journal ofRehabilitation Research, 4, Dean, R. S. (1978). The use of the WISC-R in distinguishing 208-210. learning disabled and emotionally disturbed children. JourBen-Yishay, Y., Diller, L, Gerstman, L., & Gordon, W. (1970). nal of Consulting and Clinical Psychology, 46, 381-382. Relationships between initial competence and ability to profit Dean, R. S. (1982). Neuropsychological assessment. In T. R. from cues in brain damaged individuals. Journal ofAbnormal Kratochwill (Ed.), Advances in school psychology (VoL 2, Psychology, 75, 248-259. PP- 171-201). Hillsdale, NJ: Erlbaum. Ben-Vishay, Y., Diller, L, & Rattok, J_ (1978). A modular apDean, R. S. (1985a). Perspectives on the future of neuropsychological assessment. In B. S. Plake & J_ C. Witt (Eds.), proach to optimizing orientation, psychomotor alertness, and Buras-Nebraska series on measurement and testing: Future purposive behavior in severe head trauma patients. In Working approaches to cognitive deficits in brain damage (Reof testing and measurement (pp. 203-244). Hillsdale, NJ: habilitation Monograph No. 59). New York: New York UniErlbaum. versity Medical Center, Institute of Rehabilitation Medicine. Dean, R. S. (1985b). Neuropsychological assessment. In J. D. Ben-Yishay, Y., Diller, L, Rattok, J., Ross, B., Schaier, A., & Cavenar, R. Michels, H. K. Brodie, A. M. Cooper, S. B. Scherger, P. (1979, May). Working approaches to remediaGuze, L. L Judd, G. L. Klerman, & A. J. Solnit (Eds.), tion of cognitive deficits in brain damaged persons. SupplePsychiatry (pp. 1-175). Philadelphia: Lippincott. ment to the seventh annual workshop for rehabilitation pro- Dean, R. S., & Rattan, A. L (1986). Measuring the effects of fessionals, New York. failure with learning disabled children. Presented at the AnBen-Yishay, Y., Gerstman, L, Diller, L., & Haas, A. (1970). nual Convention of the National Academy of NeuropsyPrediction of rehabilitation outcomes from psychometric pachologists. rameters in left hemiplegics. Journal of Consulting and Diller, L (1976). A model for cognitive retraining in rehabilitaClinical Psychology, 34, 436-441. tion. The Clinical Psychologist, 29, 13-15. Ben-Yishay, Y., Rattok, J., Ross, B., Lakin, P., Cohen, J., & Diller, L, Ben-Yishay, Y., Gerstman, L J., Gordon, W., WeinDiller, L. (1980). A remedial module for the systematic ameberg, J., Mandleberg, L, Schulman, P., & Shah, N. (1974). lioration of basic attentional disturbances in head trauma paStudies in cognition and rehabilitation in hemiplegia (Retients. In Working approaches to cognitive deficits in brain habilitation Monograph No. 50). New York: New York Unidamaged persons (Rehabilitation Monograph No. 61). New versity Medical Center, Institute of Rehabilitation Medicine. York: New York University Medical Center, Institute of ReDiller, L, & Gordon, W. A. (1981). Rehabilitation and clinical habilitation Medicine. neuropsychology. InS. B. Filskov & T. J. Boll (Eds.),HandBen-Yishay, Y., & Schoon, C. (1975, February). The developbook of clinical neurology (pp. 702-733). New York: Wileyment of a tool for retraining psychomotor skills (Purdue PegDiller, L V., & Weinberg, J. (1977). Hemiinattention in reboard). Paper presented at the International Neuropsycholhabilitation: The evolution of a rational remediation program. ogy Society meeting, Boston. Advances in Neurology, I8, 63-82. Block, J. H. (Ed_)_ (1971). Mastery learning: Theory and pracFowler, J_ w:, & Peterson, P. L (1981). Increasing reading pertice. New York: Holt, Rinehart, & Winston. sistence and altering attributional style of learned helpless Bolger, J. P. (1981). Cognitive retraining: A developmental apchildren. Journal of Educational Psychology, 73, 251-260. proach. Clinical Neuropsychology, 4, 66-70. Gianutsos, R., & Gianutsos, J_ (1979). Rehabilitating the verbal Boll, T. J. (1974). Behavioral correlates of cerebral damage in recall of brain-injured patients by mnemonic training: An experimental demonstration using single-case methodology. children nine through fourteen. In R. M. Reitan & L A. Journal of Clinical Neuropsychology, I, 117-135. Davison (Eds.), Clinical neuropsychology: Current status Golden, C. J_ (1978). Diagnosis and rehabilitation in clinical and applications (pp. 171-191). New York: Wiley. neuropsychology. Springfield, IL: Thomas. Boll, T. J _(1981 ). The Halstead-Reitan neuropsychology battery. InS. B. Filskov & T. J. Boll (Eds.), Handbook of clinical Gudeman, H., Golden, C., & Craine, J. (1978). The role of neuropsychological evaluation in the rehabilitation of the brainneuropsychology (pp. 42-65). New York: Wiley. injured patient: A program in neurotraining. JSAS Catalog of Bracy, 0. L (1983). Computer based cognitive rehabilitation. Selected Documents in Psychology, 8, 44. (MS No. 1693) Cognitive Rehabilitation, I, 7-8, 18. Gunnison, J. A., Kaufman, N. L, & Kaufman, A. S. (1982). Bracy, 0., Lynch, W., Sbordone, R., & Berro!, S. (1985). CogSequential and simultaneous processing applied to remedianitive Rehabilitation, 3, 10-27. tion. Academic Therapy, I7, 297-307. Brown, R. T., & Alford, N. K. (1984). Ameliorating attentional Hall, R., & Broden, M. (1967). Behavior changes in brain-injured deficits and concomitant academic deficiencies in learning children through social reinforcement. Journal ofExperimendisabled children through cognitive training. Journal of tal Child Psychology, 5, 463-479. Learning Disabilities, I7, 20-26. Hallahan, D. P. (1975). Distractibility in the learning disabled Case, R. (1972). Validation of a neo-Piagetian capacity construct. child. In W. M. Cruickshank & D.P. Hallahan (Eds.), PerJournal of Experimental Child Psychology, I4, 45-49. ceptual and learning disabilities in children, research and Case, R., Kurland, D. M., & Goldberg, J. (1982). Operational efficiency and the growth of short-tenn memory span. Jour-
theory (VoL 2, PP- 210-240). Syracuse: Syracuse University
nal of Experimental Child Psychology, 33, 386-403. Cooke, N. (1973). Neuropsychology: From theory into practice.
Press. Hallahan, D. P., Gajar, A. H., Cohen, S. B., & Tarver, S. G.
COGNITIVE REHABILITATION (1978). Selective attention and locus of control in learning disabled and normal children. Journal of Learning Disabilities, 11, 231-236. Hallahan, D.P., & Sapona, R. (1983). Self-monitoring of attention with learning disabled children: Past research and current issues. Journal of Learning Disabilities, 16, 616-620. Hallahan, D.P., Tarver,S. G., Kaufman,J. M.,&Grabeal, N. L. (1978). A comparison of the effects of reinforcement and response cost on the selective attention of learning disabled children. Journal of Learning Disabilities, 11, 430-438. Hartlage, L. C. (1975). Neuropsychological approaches to predicting outcomes of remedial education strategies for learning disabled children. Pediatric Psychology, 3, 23-28. Hartlage, L. C., & Lucas, D. G. (1973a). Group screening for reading disability in flfSt grade children. Journal of Learning Disabilities, 6, 48-52. Hartlage, L. C., & Lucas, D. G. (1973b). Pre-reading expectancy screening scales. Jacksonville, IL: Psychologists and Educators. Hartlage, L. C., & Reynolds, C. R. (1981). Neuropsychological assessment and the individualization of instruction. In G. W. Hynd & J. E. Obrzut (Eds.), Neuropsychological assessment and the school-age child: Issues and procedures (pp. 355378). New York: Grune & Stratton. Hartlage, L. C., & Telzrow, C. F. (1983). The neuropsychological basis of educational intervention. Journal of Learning Disabilities, /6, 521-528. Hartlage, L. C., & Telzrow, C. F. (1984). Rehabilitation of persons with learning disabilities. Journal ofRehabilitation, 50, 31-34. Hasher, L., & Zacks, R. T. (1979). Automatic and effortful processes in memory. Journal of Experimental Psychology: General, 108, 356-388. Horton, A. M., Jr. (1979a). Behavioral neuropsychology: A clinical case study. Clinical Neuropsychology, 1(3), 44-47. Horton, A. M., Jr. (l979b). Behavioral neuropsychology: Rationale and research. Clinica/Neuropsyclwlogy, 1(2), 20-23. Horton, A. M., Jr. (1981). Behavioral neuropsychology in the schools. School Psychology Review, /0(3), 367-372. Hynd, G. W., & Obrzut, J. E. (1981). School neuropsychology. Journal of School Psychology, 19, 45-50. Jones, M. K. (1974). Imagery asamnemonicaidaftcrlefttemporal lobectomy: Contrasts between material-specific and generalizedmemorydisorders.Neuropsychologia, 12,21-30. Kaufman, D., & Kaufman, P. (1979). Strategy training andremedial techniques. Journal of Learning Disabilities, 12, 416-419. Kirk, S. A., & Becker, W. (Eds.). (1963). Conference on children with minimal brain impairment. Urbana: University of Illinois. Klonoff, H., & Low, M. (1974). Disordered brain function in young children and early adolescents: Neuropsychological and electroencephalographic correlates. In R. M. Reitan & L. A. Davison (Eds.), Clinical neuropsyclwlogy: Current status and applications (pp. 121-165). New York: Wiley. Klonoff, H., Low, M. D.• & Clark, L. (1977). Head injuries in children: A prospective five year follow-up. Journal ofNeurology, Neurosurgery, and Psychiatry, 40, 1211-1219. Krop, H. (1971). Modification of hyperactive behaviorofa brain-
407
damaged, emotionally disturbed child. Training School Bulletin (Vineland), 68, 49-54. Lang, P. (1964). Experimental studies of desensitization psychotherapy. In J. Wolpe, A. Salter, & L. J. Peyna (Eds.), The conditioning therapies (pp. 260-291). New York: Holt, Rinehart, & Winston. Leftoff, S. ( 1981). Learning functions for unilaterally brain damaged patients for serially and randomly ordered stimulus material: Analysis of retrieval strategies and their relationship to rehabilitation. Journal of Clinical Neuropsychology, 3, 301-313. Levin, H. S., Benton, A. L., & Grossman, R. G. (1982). Neurobehavioral consequences ofclosed head injury. New York: Oxford University Press. Levin, H. S., & Eisenberg, H. M. (1979a). Neuropsychological impairment after closed head injury in children and adolescents. Journal of Pediatric Psychology, 4, 389-402. Levin, H. S., & Eisenberg, H. M. (J979b). Neuropsychological outcome of closed head injury in children and adolescents. Childs Brain, 5, 281-292. Lewinsohn, P.M., Danaher, B. G., & Kikel, S. (1977). Visual imagery as a mnemonic aid for brain injured persons. Journal of Consulting and Clinical Psychology, 45, 717-723. Luria, A. R. (1963). Restoration of function after brain injury. New York: Macmillan Co. Lynch, W. J. (1979). A guhle to Atari Videocomputer programs for rehabilitation settings. Palo Alto, CA: Veterans Administration Medical Center. Lynch, W. J. (1981). The use of electronic games in cognitive rehabilitation. In L. E. Trexler (Ed.), Cognitive rehabilitation: Conceptualization and intervention (pp. 263-274). New York: Plenum Press. Piasetsky, E. B. (1981). The relevance of brain-behavior relationships for rehabilitation. In L. E. Trexler (Ed.), Cognitive rehabilitation: Conceptualization and intervention (pp. 115130). New York: Plenum Press. Rattok, J., Ben-Yishay, Y., Ross, B., Lakin, P., Silver, S., Thomas, L., & Diller, L. (1982). A diagnostic-remedial system for basic attentional disorders in head trauma patients undergoing rehabilitation: A preliminary report. In Working approaches to remediation of cognitive deficits in brain damaged persons (Rehabilitation Monograph No. 64). New York: New York University Medical Center, Institute of Rehabilitation Medicine. Reitan, R. M. (1974). Methodological problems in clinical neuropsychology. In R. M. Reitan & L. A. Davison (Eds.), Clinical neuropsychology: Current status and applications (pp. 140-191). New York: Wiley. Reynolds, C. R. (1981). Neuropsychological assessment and the habilitation of learning: Considerations in the search for the aptitude x treatment interaction. School Psychology Review, 10, 343-349. Ross, A. 0. (1976). Psyclwlogical aspects of learning disabUities and disorders. New York: McGraw-Hill. Rourke, B. P. ( 1976). Issues in the neuropsychological assessment of children with learning disabilities. Canadian Psyclwlogical Review, 17, 89-102. Seligman, M. E. P. (1975). Helplessness: On depression, development, and death. San Francisco: Freeman.
408
CHAPTER 21
Seretny, M. L., Dean, R. S., Gray, J. W., & Hartlage, L. C. (1986). The practice of clinical neuropsychology in the United States. Archives ofClinical Neuropsychology, 1. 90-
94. Severson, R. A. (1970). Behavior therapy with severe learning disabilities. Unpublished manuscript, University of Wisconsin. Silver, S., Ben-Yishay, Y., Rattock, J., Ross, B., Lakin, P., Piasetsky, E., Ezrachi, 0., & Diller, L. (1983). Occupational outcomes in severe TBD's following intensive cognitive remediation: An interim report. In Working Approaches toRemediation of Cognitive Deficits in Brain Damaged Persons (Rehabilitation Monograph No. 66). New York: New York University Medical Center, Institute of Rehabilitation Medicine. Tarver, S. G., Hallahan, D.P., Kaufman, J. M., & Ball, D. W. (1976). Verbal rehearsal and selective attention in children with learning disabilities: A developmental lag. Journal of Experimental Child Psychology, 22, 375-385. Task Force on Head Injury. (1984). Standards for cognitive re-
habilitation. Erie, PA: American Congress of Rehabilitation Medicine. Telzrow, C. F. (1985). The science and speculation of rehabilitation in developmental neuropsychological disorders. In L. C. Hartlage & C. F. Telzrow (Eds.), The neuropsychology of individual differences (pp. 271-304). Elsford, NY: Pergamon Press. Weinberg, J., & Diller, L. (1968). On reading newspapers by hemiplegic denial of visual disability. Paper presented at the annual meeting of the American Psychological Association, New York. Weinberg, J., Diller, L., Gordon, W. A., Gerstman, L. J., Lieberman, A., Lakin, P., Hodges, G., & Ezrachi, 0. (1977). Visual scanning training effect on reading-related tasks in acquired right brain damage. Archives of Physical Medicine and Rehabilitation, 58, 479-486. Wilson, G. T., & O'Leary, K. D. (1980). Principles of behavior therapy. Englewood Cliffs, NJ: Prentice-Hall. Wolpe, J. (1969). The practice of behavior therapy. Elmsford, NY: Pergamon Press.
22 Neuropsychological Aspects of Epilepsy Introduction and Overview LAWRENCE C. HARTLAGE AND PATRICIA L. HARTLAGE
Although estimates of the prevalance of childhood epilepsy in the United States vary, review of a number of epidemiological studies suggests that this condition may affect up to one million children (Hartlage & Telzrow, 1984). As for normal children, children with epilepsy differ from one another on a variety of dimensions. Unlike normal children, however, children with epilepsy are subject to three conditions that contribute to increased variability. Two of these conditions are primarily neuropsychological in nature. One involves the fact that epilepsy is symptomatic of some type of brain dysfunction, and such factors as locus, type, extent, age of onset, and seizure manifestations may each or in combination have implications for how the child's adaptive behavior may be affected. The other neuropsychological condition involves the effects of anticonvulsant drug therapy on the child's development or manifestation of adaptive behavior. The third condition represents an interaction between such social factors as the reactions of the child's parents, peers, and teachers; the neuropsychological substrates of appropriate adaptation to the requirements of given ages; and the effects of anticonvulsant medication on both peer response and the child's potential for utilization of underlying neuropsychological assets. Although for a given child with epilepsy these three conditions are likely to be interactive, it may be helpful to overview each condition as a separate entity before attempting to address the more complex issues involved in the interactions of these conditions. LAWRENCE C. HARTLAGE
o
Department of Psychol-
ogy, University of Arkansas, Fayetteville, Arkansas 72701.
PATRICIA L. HARTLAGE
o
Department of Pediatrics and
Neurology, Medical College of Georgia, Augusta, Georgia 30902.
Neuropsychological Substrates of Childhood Epilepsy Seizures can be classified in a number of ways. Perhaps the most widely used current classification involves three major types: generalized tonic-clonic, partial complex, and absence seizures. Although different on a number of dimensions, this classification is similar to the earlier classification of grand mal, psychomotor, and petit mal types of seizures. In children, there may be less correspondence between a specific neurological abnormality and discrete behavioral symptoms. Further, the incidence of types of seizures differs between children and adults, to the extent that in a number of respects childhood epilepsy represents a somewhat different phenomenon than adult epilepsy. Although not all children with epilepsy have intellectual deficits, for many years it was reported than an unselected population of children with epilepsy were likely to have mean IQ ranging up to a standard deviation lower than average (Henderson, 1953; Somerfild-Ziskind & Ziskind, 1940; Sullivan & Gahagan, 1935; Whitehouse, 1971). More recent research has identified factors that contribute to our understanding of correlations of depressed IQ in children with epilepsy. Frequency of seizures is associated with mental deficit, greater frequency being correlated with more severe intellectual deficit (Farwell, Dodrill, & Batzel, 1985; Keith, Evert, Green, & Gage, 1955). Type of seizure appears to be an important correlate of intellectual ability in that intellectual levels within a given seizure classification are much more homogeneous than those found among the spectrum of childhood epilepsy. In general, absence types of 409
410
CHAPTER 22
seizures represent the class least likely to show intellective deficit (Collins & Lennox, 1946; O'Leary, Seidenberg, Berent, & Boll, 1981; Zimmennan, Burgemeister, & Putnam, 1948), generalized tonicclonic seizures most likely to show greatest intellectual deficit, and partial complex seizures resulting in intennediate levels of intellectual impainnent. Although these findings are likely fairly compatible with impressions of clinicians who work with unselected populations of children with epilepsy, a carefully screened sample of tertiary center childhood epilepsy patients shows Full Scale IQ levels of 70 for minor motor; 74 for atypical absence; 96 for partial plus generalized; 98 for partial; 99 for generalized tonic-clonic; 97 for classic absence plus generalized tonic-clonic; and 106 for classic absence (Farwell et al., 1985). In this sample, the lowest Full Scale IQ scores were found to differ (p < 0.05) between the minor motor and atypical absence children and all other types. The finding of highest IQ scores in the children with only classic absence type of seizures is similar to the average IQ levels of 106 and 113 reported, respectively, by Zimmennan et al. ( 1948) and Collins and Lennox ( 1946) for samples of children with classic absence seizures. Another important correlate of cognitive or intellectual deficit in childhood epilepsy relates to age of onset. There is fairly consistent agreement that earlier onset is associated with greatest intellectual impainnent (O'Leary et al., 1981). Many years ago, research indicated that IQ scores were lower among children with epilepsy whose seizures began before age 5 (Sullivan & Gahagan, 1935), and more recent research has shown that seizure onset during the first year of life is related to greatest impainnent (Scarpa & Carassini, 1982). An important component of age of onset of seizures is that of duration of seizures, in that earlier onset typically is related to duration. Recent computation shows that overlap (r) between intelligence and age of onset accounts for 9% of variance, with duration accounting for 16% and number of years with seizures accounting for most (38%) variance (Farwell et al., 1985). It is likely that studies of intellectual findings among children with epilepsy will not show a great deal of unifonnity across settings because the types of children with epilepsy who are evaluated at different types of referral centers may differ on variables related to intellectual abilities. In tertiary centers, such as major medical center epilepsy programs, there is a likelihood that the more difficult cases will be studied because the relatively uncomplicated cases are less likely to be referred there. In a large series of epileptic children, Gregoriades (1972)
found that 11% had classic absence seizures, whereas a recent report from a major tertiary care epilepsy program found only 6% of children with this problem (Farwell et al., 1985). Similarly, Gregoriades found that 50% of children in his sample had only generalized tonic-clonic seizures, whereas Farwell and colleagues' sample had only 25% of this simple seizure type. In light of findings concerning intellectual correlates of seizure types, it is obvious that conclusions concerning intellectual levels among children with epilepsy may be expected to vary as a function of the types of children seen in a given setting. Although there are no specific data relating to this phenomenon, it seems likely that the extent of intellectual deficit in childhood epilepsy may be somewhat exaggerated by the fact that careful psychometric studies of children with epilepsy are somewhat more likely to be reported from large epilepsy programs, to which the more complicated and difficult-to-manage seizure cases may be referred. Fonnal neuropsychological studies involving assessment of children with epilepsy typically have not found a specific profile or pattern of impainnent. In tenns of seizure types, one recent study of children tested with the age-appropriate Reitan Neuropsychological Battery (Reitan & Davison, 1974) found no significant impainnent among children with only classic absence seizures; and only very mild or mild impairments in children with minor motor, atypical absence, and classical absence plus generalized tonic-clonic seizures (Farwell et al., 1985). These researchers also found differences among seizure types on a screen for aphasia, with the aphasia screening scale differentiating between children with minor motor and atypical absence seizures compared with other seizure types, significance being more pronounced (p < 0.01) for younger children. Somewhat similar findings were reported by Matthews and Klove ( 1967) who found that children with various types of epilepsy perfonned poorer than controls on neuropsychological measures. Somewhat more specific findings were reported by Epir, Renda, and Baser (1984) in a study of Turkish children, wherein receptive language and drawing ability were more impaired in children with epilepsy. The children in the Epir et al. study were somewhat atypical from those usually studied in that 80% were not receiving seizure medication. With respect to reliability, one of the few such studies was reported by Dodrill and Troupin (1975), who found that the majority of neuropsychological measures used for evaluation of patients with epilepsy did not demonstrate significant practice effects. However, they concluded that perfonnance on Wechsler scales may be more affected
NEUROPSYCHOLOGICAL ASPECTS OF EPILEPSY
by anticonvulsant medication than many other neuropsychological measures, thus raising the possibility that specific tests used for assessment in epilepsy may have reliability differentially affected by anticonvulsant drug therapy.
Medication-Performance Interactions The possible influence of anticonvulsant medication on the mental performance of children with epilepsy is of obvious relevance to neuropsychological study, in that the vast majority of children with epilepsy who receive neuropsychological evaluation will likely be taking one or more anticonvulsant medications at the time of testing. This factor has, however, tended to be ignored in much neuropsychological research in childhood epilepsy. Perhaps the first long-range approach to studying anticonvulsant medication effects on mental test performance was reported by Lennox (1942), who studied more than 1000 patients of varying ages who had received anticonvulsant medications for seizure control. He concluded that there was no relationship between anticonvulsant medications and psychological function. Similar conclusions were reached by other early investigators, who studied such anticonvulsant medications as phenobarbital (SomerfeldZiskind & Ziskind, 1940), diphenylhydantoin (Loveland, Smith, & Forster, 1957), and primidone (Loveland et al., 1957; Roye & Martin, 1959; Chaudhry & Pond, 1961). Thus, for approximately three decades there was both belief, and research evidence in support of this belief, that any behavioral impairment in children with epilepsy was not related to their anticonvulsant medication. A problem in the investigation of possible relationships between anticonvulsant medication and mental function in children involved the individual variability in drug metabolism and the considerable variability in size found in children at different ages. At a given dosage level of anticonvulsant medication, for example, a child weighing 20 pounds might show a different response than a child weighing 80 pounds. Such fourfold variability in body weight is relatively uncommon in adults, so that the technical considerations involved in studying drug-behavior interactions in children posed a considerably greater challenge. Increased variability of metabolism in children, as opposed to adults, may be related to the fact that adults typically have achieved a stable body habitus, whereas children at different growth stages may have considerably different growth and metabolism rates.
411
With the advent of gas-liquid chromatography (Woodbury, Penry, & Schmidt, 1972), it became possible to assess the amount of anticonvulsant medication in the child's serum at the time of mental testing. Approximately corresponding to the increased availability of gas-liquid chromatography to investigators in epilepsy, was the recognition of relationships between the medications used for seizure control and mental test performance in both children and adults (ldestrom, Schalling, Carlquist, & Sjoquist, 1972; Reynolds & Travers, 1974; Trimble & Reynolds, 1976). Specific relationships between given anticonvulsant medications and performance on given psychological tests were studied in a series of investigations wherein serum anticonvulsant levels were determined on serial visits with psychological testing done at the time serum levels were collected (Hartlage, 1984; Hartlage & Linz, 1984; Hartlage, Noonan, & Prim, 1983; Hartlage, Nance, Noonan, & Shaw, 1983; Hartlage, McCuiston, & Noonan, 1982; Hartlage, Hynd, & Telzrow, 1981). For the major types of anticonvulsant medications studied (phenobarbital, primidone, carbamazepine, diphenylhydantoin, and valproic acid), most striking relationships were found between phenobarbital and primidone and neuropsychological measures. The neuropsychological measures most sensitive to serum anticonvulsant levels were coding, digit symbol, and symbol digit tests, all of which correlated (p < 0.01) with phenobarbital. Two of these measures (symbol digit and digit symbol) correlated (p < 0.05) with primidone serum levels. Neuropsychological measures that correlated with phenobarbital serum levels (p < 0.05) were finger oscillation (bilateral), digit span forward and total, and a subscale involving a rapid symbol marking from the General Aptitude Test Battery. Primidone correlated (p < 0.05) with nondominant hand finger oscillation and Minnesota Rate of Manipulation Test, bilateral Minnesota Rate of Manipulation Test, digit spao forward, and the General Aptitude Test Battery subscale. With respect to actual levels of performance, high serum barbiturate levels were associated with raw scores of 15.2, 15.5, and 11.6, compared with low serum barbiturate level scores of37.2, 27.9, and 31.1, respectively, for performance on digit symbol, coding, and symbol digit measures (Hartlage, 1981). Relationships between neuropsychological measures and other types of anticonvulsant medications (carbamazepine, diphenylhydantoin, and valproic acid) did not exceed chance levels, and there were no significant relationships between either phenobarbital or primidone and measures of word knowledge, concept formation, or dominant hand performance on the
412
CHAPTER 22
Minnesota Rate of Manipulation Test. Dodrill (1975) reported that diphenylhydantoin depressed performance on a variety of motor tasks, but not with tasks emphasizing higher motor function, and MacLeod, Dekaban, and Hunt (1978) found relationships between phenobarbital concentration and short-term but not long-term memory. Skilbeck (1984) found memory impairment on reaction time for memory scanning affected by diphenylhydantoin, with slower reaction times related to higher dosage levels. These and other findings suggest that the relationship between anticonvulsant medications and neuropsychological test performance may be fairly specific, both with respect to anticonvulsant medication and the abilities affected (MacLeod et al., 1978). In any case, however, there appear to be relationships between some medications used for seizure control and performance on special mental ability measures, to an extent that merits consideration in the evaluation of the performance of a given child with epilepsy.
Social and Emotional Correlates of Childhood Epilepsy Although there has been fairly consistent agreement that children with epilepsy differ from their nonepileptic peers on a number of social and emotional variables (Bagley, 1971; Livingston, 1972), there is no good agreement concerning either the nature or the possible causes of these differences. Early work implicated psychopathology related to temporal lobe dysfunction (Gibbs, Gibbs, & Forster, 1948) as etiologic, although more recent work has indicated that as a group, individuals with epilepsy do not manifest behavior disturbances at a level different from those found in groups with chronic illness or nonepileptic neurological disorder (Whitman, Hermann, & Gordon, 1980). However, these same investigators found that when psychopathology was present in individuals in each of the three groups, the possibility of a more serious psychopathology was greatest in persons with epilepsy. This raises the possibility that, although epilepsy may not necessarily be etiologic in social-emotional problems, it may be a condition that exacerbates such problems. Such a possibility does not necessarily preclude the potential for specific types or causes of epilepsy to be associated with increased likelihood of social-emotional problems, and there is speculation that temporal lobe epilepsy muy huve different sociul-emotionul correlates than other types. The nature of these correlates is not a matter of common agreement, in that Ounsted
( 1969) found that children who experienced only temporal lobe epilepsy were not likely to have social-emotional pathology, whereas others have reported contradictory findings (Bear, 1979; Blumer, 1975; Glass & Mattson, 1973; Waxman & Geschwind, 1975). Yet other investigators have found that combined seizure types are more likely to be associated with emotional pathology (Hermann, Dikmen, & Wilensky, 1982), whereas no relationships between seizure types and emotional pathology have been found by other researchers (Lacher, Lewis, & Kupke, 1979; Hermann & Stevens, 1980; Matthews & Klove, 1967; Standage & Fenton, 1975). Although there is no common agreement concerning relationships between seizure types and behavioral pathology, there is agreement that emotional adjustment among individuals with epilepsy may be impaired. In a recent study of epilepsy in four countries (Canada, Finland, German Democratic Republic, and the United States), Dodrill et al. (1984) found a number of common problems, with emotional adjustment always representing the major area of concern. Although these findings were limited to adults, they are compatible with those of Bagley (1971) with children. A feature of social-emotional problems in childhood epilepsy, when compared with healthy nonepileptic and chronically ill diabetic children, has been found to involve the attribution of control over their lives to external sources (Matthews, Barabas, & Ferrari, 1982), with a secondary feature involving lower self-concept. This feeling of external control over their lives may be related to a finding of increased dependency in epileptic children who were matched with (otherwise healthy) tonsillectomy patients (Hartlage, Green, & Offutt, 1972). Other investigators have hypothesized that dependency in epileptic children may be due to parental attitudes (Bayley & Schaefer, 1960; Heathers, 1953), and this hypothesized relationship has received some support from Hartlage and Green (1972), who also found parental attitudes to be related to academic and social achievement in children with epilepsy. When compared with matched controls, children with epilepsy tended to perform less well academically than their ability levels would suggest, as well as tending to have poorer language usage than what would be suggested by their measured communication abilities (Green & Hartlage, 1971). Although no causal relationships between epilepsy and academic underachievement have been identified, there are a number of factors that may be involved. One involves the possibility of some depressing effect of anticonvulsant medication on academic performance. Another factor may relate to de-
NEUROPSYCHOLOGICAL ASPECfS OF EPILEPSY
pendency, of whatever origin it may have. Yet another possible factor suggested by Dodrill ( 1980) involves some interrelationship between neuropsychological substrates of learning. Although presumably impaired neuropsychological functioning might also be expected to depress intellectual level, Matthews, Barabas, and Ferrari ( 1983) reported the incidence of learning disorders to range from 15 to 30% among children with epilepsy, and Breger (1975) also identified learning impairments in children with epilepsy. Thus, there is agreement that problems in so~ cial-emotional spheres are not uncommon among children with epilepsy, although the causes of these problems remain unclear. Although researchers have identified a number of possible causes or at least correlates of such problems, the interpretation of findings and implementation of intervention ap~ proaches for a given child still must depend heavily on the skills and insights of the clinician working with the child.
Diagnostic and Classificatory Issues Although mention of major classes of childhood seizures has been made, it is important to emphasize that there are many manifestations of seizures in children. Noted epileptologists have said, "Epile~ sy . . . can masquerade in so many forms that any busy doctor will be treating it knowingly or unknowingly" (Bashes & Gibbs, 1972, p. 3). These protean manifestations of epilepsy may be one reason for the lack of agreement among researchers studying specific aspects of epilepsy such as the relationship between epilepsy and social-emotional adjustment in children. It may be helpful to keep in mind that epilepsy represents a symptom rather than a specific disease entity, so that in addition to multiple manifestations there may be multiple causes. In general, epilepsy represents evidence of an irritative reaction to some type of brain injury. If the nature of the brain injury is known, such as might result in the case of an identified neoplasm, the resultant epilepsy may be referred to as symptomatic. In the more common cases where no specific cause is identified, epilepsy may be referred to as idiopathic. Obviously, the effects of a specific (e.g. , tumor) cause may differ from those of unknown or nonspecific causes, on a number of cognitive and behavioral dimensions. Another determinant of how epilepsy may affect a ·given child relates to how much of the brain is involved, either as an underlying cause or in the sei-
413
zure itself. Although a seizure may have a focal discharge, it may spread and become generalized. It is with these types of seizures that surgical disconnection of the cerebral hemisphere may be attempted to prevent the spread of discharge to the nonaffected hemisphere. Although some types of seizures have certain common EEG profiles, the absence of such a profile does not necessarily imply that this type of seizure is not present: approximately 25% of patients with generalized tonic-clonic seizures have completely normal EEG findings, in both waking and sleep tracings (Boshes & Gibbs, 1972). For this reason, careful history and sustained observation may be necessary to determine the type of seizure involved. Although this may suggest that the EEG is of relatively little use in seizure diagnosis, there are a number of conditions in which the EEG shows a strong relationship with neuropsychological status. In infants and young children with a history of frequent brief spasms or quivering spells, an abnormal EEG pattern referred to as hypsarrhythmia may predict significantly impaired mental function or retardation. Much less likely to predict mental impairment or even subsequent seizures are febrile convulsions, which may appear in approximately 3% of children below 5 years of age. In general, these types of seizures tend to resolve with age, and not be followed by any other type of seizure disorder. Although there is a fairly strong familial incidence of this type of epile~ sy, what is inherited appears to be a low convulsive threshold for increased body temperature, which is a limited defect associated with spontaneous recovery (Garvin, 1970; Livingston, Bridge, & Kajdi, 1974). Some seizure types are relatively age specific, with absence seizures rare in children below 2 years of age and most common between 5 and 19 years of age: these seizures tend to disappear with increasing age and are relatively rare in adults (Bashes & Gibbs, 1972). Another example of age-related seizures involves what John Hughlings Jackson ( 1931) referred to as epileptic equivalents, and which have subsequently been called psychomotor seizures, temporal lobe epilepsy, and partial complex seizures. For many years it has been recognized that this type of disorder typically has a spike focus in the anterior temporal areas (Gibbs et al., 1948), and is most common among adults, whereas children with this type of seizure are more likely to have a spike focus in the midtemporal area (Boshes & Gibbs, 1972). Further, a given child may demonstrate a series of seizure manifestations at different stages of life (Dreifuss, 1975).
414
CHAPTER 22
A unique feature of epilepsy, when compared with most other chronic diseases, is that most of the time the symptoms (seizures) are not present. Further, it is relatively uncommon for the child with convulsive seizures to demonstrate them on examination, so that the diagnosis is not as straightforward as it might be with many other childhood disorders. The primary laboratory diagnostic tool, the EEG, may well not be definitive because estimates of normal (i.e., nonepileptic) individuals with abnormal EEGs are approximately 25% (Kaufman, 1981), and nearly that percentage of individuals with recurrent seizures do not have abnormal EEGs (Harris, 1976). Although many texts in child development, educational psychology, and general psychology have used the classic terminology for seizure classification (e.g., grand mal, petit mal, psychomotor epilepsy), since 1970 the International League Against Epilepsy has used a classification system based on description of seizure types. This classification system has been widely accepted by the medical neurology community, and classifies seizures into four major groups. Group I (partial seizures) begin locally, and may have no impairment of consciousness, although there may be motor, sensory, or autonomic symptoms. A subtype within this class, generally accompanied by impairment of consciousness, may have cognitive, affective, and psychomotor symptoms. Group II (generalized) seizures include absence and generalized tonic-clonic seizures. Group III seizures are unilateral, and Group IV contains unclassified seizures (Hartlage & Telzrow, 1984). Although issues involving classification are important for epileptologists, and for researchers who wish to define patient populations, the complexity of criteria for classification of seizures tend to preclude understanding by professionals from other specialties. As a result, the symptoms (epilepsy) of a variety of manifestations of central nervous system disorders may be treated as if this represented a meaningful unitary classification in neuropsychological research or clinical practice. Given the heterogeneity of epilepsy, it is often difficult to generalize across studies, even involving children who have seizures within a given class, in such a way as to generate hypotheses with implications for neuropsychological assessment or intervention for a given child with epilepsy.
Treatment and Rehabilitation Although surgical treatment of epilepsy has been in use for many years (Penfield & Flanigin,
1950), and continues to attract interest (Green & Pootrakul, 1982; Flanigin, King, & Gallagher, 1985; King eta/., 1986; Whittle, Ellis, & Simpson, 1981), the principal treatment involves medical (anticonvulsant drug) therapy. Since early in the 20th century, phenobarbital and diphenylhydantoin have been widely used for seizure control. The Epilepsy Branch of the National Institute for Neurological and Communicative Diseases and Stroke (NINCDS) has documented 14 other drugs currently in use in the United States (NINCDS, 1980). Although these anticonvulsant medications differ from one another on a variety of pharmacological properties and behavioral effects, there is no universal agreement concerning which anticonvulsant medications may be specifically indicated for treatment of seizures of a given type for an individual child. The reported spectrum of psychotropic action of anticonvulsant medications in general is expected to produce psychomotor improvement (diminished retardation); cognitive improvement (especially concentration and attention); and affective improvement (diminished irritability, anxiety, and depression) (Parnas, Gram, & Flachs, 1980). It is interesting to note that, as reported earlier in this chapter, apparently the opposite effect of psychomotor improvement appears to represent the rule rather than the exception when barbiturates are used for seizure control. Valproic acid is a drug that, although in common European use for some time, in the United States has been in use for approximately a decade. Its chemical structure differs completely from other antiepileptic drugs, and it appears to have very few and relatively harmless side effects (Parnas et al., 1980). Thus, the spectrum of psychotropic effects of anticonvulsant medications is a broad one, and it is probably premature to generalize to behavioral concomitants or effects of antiepileptic drugs as a generic group. Further compounding the problems in this respect, it is not uncommon for children with chronic seizure disorders to be treated concurrently with more than one anticonvulsant medication, so that longitudinal, large-sample studies of children receiving only one anticonvulsant drug are both difficult and rare. One novel approach to studying effects of (anticonvulsant drug) therapy on adaptive behavior, with some control for environmental influences, involves the study of operant responding rate, using a sample of mentally retarded residents of an institution. In this study, valproic acid therapy did not affect operant responding rate, whereas patients receiving barbiturate therapy tended to show greatest performance decline, although there were interactions between sei-
NEUROPSYCHOLOGICAL ASPECTS OF EPILEPSY
zure types and drug effects on performance (Gay, 1984). Although such focused studies can contribute to understanding of the interaction between treatment and rehabilitation, the context of rehabilitation generally is much broader. Typically, such factors as level of mental function, psychosocial and interpersonal adjustment, and seizure type and frequency are variables with important implications for rehabilitative prognosis. These variables, in turn, may interact with treatment variables; for example, seizure type and frequency may influence the regimen of drug (or surgical) therapy, which in turn may influence level of mental function and psychosocial adjustment and how teachers, parents, and peers react to the child; how well the child may perform academically with respect to his or her level of mental abilities; and so on. Prognosis for rehabilitative outcome of the child with epilepsy thus represents a multifaceted issue, with so many potentially pertinent variables as to suggest that specific prognostic statements for a given child may need to be done on an individualized, case-by-case basis. The adjustment requirements for a child with epilepsy, as for any other child, change with age. The child with epilepsy who has been overprotected by solicitous parents may have special difficulty in separating from this protective environment to enter school and make adjustments to teachers and peers who may be much less protective or supportive. In middle school years the child with epilepsy may encounter problems with self-esteem and peer acceptance due to being "different," especially if seizures occurring during school attendance are remarkable and violent. Further adjustment difficulties during this age span may relate to learning problems, whether due to neuropsychological impairment or medication effects. Participation in some playground activities or contact sports may be prohibited, and represent yet another indication of being different. Even out-of-school activities, such as swimming or bicycle riding, can be limited, providing additional limitations to normal peer interactions or skill development. Adolescence represents a time of special adjustment difficulty for the child with epilepsy. Such issues as driving a car, an important part of most adolescents' lives, may be precluded by solicitous parents or concerned physicians. Rejecting peers may provide other important sources of psychosocial frustration. The emerging need for independence may be clouded by accumulative academic deficiencies, limited vocational or educational options, and
415
rejecting social attitudes (Hartlage & Roland, I 971 ; Hartlage, Roland, & Taraba, 1971; Hartlage & Taraha, 1971; Hartlage, 1974). Further complicating the problems of the adolescent with epilepsy is the possibility that the endocrine changes of adolescence may precipitate different or increased seizures, or may make stable levels of seizure control on given medications less stable, thus precipitating more seizures or the need to establish a new regimen of effective anticonvulsant therapy.
Summary Consideration of the many dimensions on which childhood epilepsy may vary, and the imposing combinations and permutations of interaction among these dimensions that may affect children at differing ages who suffer from different degrees and types of epilepsy, enhances appreciation of the complexity of issues facing the child neuropsychologist who works in this field. As with many specialty areas in child clinical neuropsychology, more has become known about the field in recent years than was known for the preceding hundreds of years. It is hoped that, in the brief span allocated to the topic relative to its complexity, it has been possible to capture and expose some of the challenges and accomplishments related to the topic. An expanded hope is that the reader may find something of value and relevance to his or her professional work, with the result that increased research and clinical service in this field may be of help in solving the many remaining questions, and thereby help to ameliorate the burden of epilepsy on afflicted children and their families.
References Bagley, C. ( 1971). The social psychology of the epileptic child. Coral Gables, FL: University of Miami Press. Bayley, N., & Schaefer, E. (1960). Maternal behavior and personality development data from the Berkeley growth study. Psychiatric Research Reports of the American Psychiatric Association, 13, 155-173. Bear, D. M. (1979). Temporal lobe epilepsy: A syndrome of sensory-limbic hyperconnection. Cortex, 15, 357-384. Blumer, D. (1975). Temporal lobe epilepsy and its psychiatric significance. In D. F. Benson & D. Blumer (Eds.), Psychiatric aspects of neurologic disease (pp. 171-198). New York: Grune & Stratton. Boshes, L. D., & Gibbs, F. A. (1972). Epilepsy handbook (2nd ed.). Springfield, IL: Thomas. Breger, E. (1975). Psychiatric consultation to the epileptic child
416
CHAPTER 22
and his family: A study of 60 cases-Part 3: Follow-up study. Maryland State Medical Journal, 47-50. Chaudhry, M. R., & Pond, D. A. (1961). Mental deterioration in epileptic children. Journal of Neurology, Neurosurgery & Psychiatry, 24, 213-219. Collins, L. A., & Lennox, W. G. (1946). Intelligence of 300 private epileptic patients. Research Publication of the Association for Nervous and Mental Disorder, 26, 586-597. Dodrill, C. B. (1975). Diphenylhydantoin serum levels, toxicity, and neuropsychological performance in patients with epilepsy. Epilepsia, 16, 593-600. Dodrill, C. B. (1980). Interrelationships between neuropsychological data and social problems in epilepsy. In R. Canger, F. Angeleri, & J. Penry (Eds.), Advances in epileptology (pp. 191-197). New York: Raven Press. Dodrill, C. B., Beir, R., Kasparik, M., Tacke, 1., Tacke, U., & Tan, S. (1984). Psychosocial problems in adults with epilepsy: Comparison of findings from four countries. Epilepsia, 25, 176-183. Dodrill, C. B., & Troupin, A. S. (1975). Effects of repeated administration of a comprehensive neuropsychological battery among chronic epileptics. Journal ofNervous and Mental Disease, 161, 185-190. Epir, S., Renda, R., & Baser, N. (1984). Cognitive and behavioural characteristics of children with idiopathic epilepsy in a low-income area of Ankara, Turkey. Developmental Medicine and Child Neurology, 28, 200-207. Farwell, J. R., Dodrill, C. B., & Batzel, L. W. (1985). Neuropsychological abilities of children with epilepsy. Epilepsia, 26, 395-400. Aanigin,H. F., King, D. W., &Gallagher, B. B. (1985). Surgical treatment of epilepsy. InT. Pedley & B. Meldrum (Eds.), Recent advances in epilepsy (pp. 297-339). Edinburgh: Churchill Livingstone. Garvin, J. S. (1970). Significance of convulsions with fever. Clinical Electroencephalography, 1, 41-44. Gay, P. (1984). Effects of antiepileptic drugs and seizure type on operant responding in mentally retarded persons. Epilepsia, 25, 377-386. Gibbs, F. A., Gibbs, E. F., & Furster, B. (1948). Psychomotor epilepsy. Archives of Neurology, 60, 331-340. Glass, D. W., & Mattson, R. (1973). Psychopathology and emotional precipitation of seizures in temporal lobe and nontemporal lobe epileptics. Proceedings, American Psychological Association Convention, 427. Green, J. B., & Hartlage, L. C. (1971). Comparative performance of epileptic and non-epileptic children and adolescents. Diseases of the Nervous System, 32, 418-421. Green, J. R., & Pootrakul, A. (1982). Surgical aspects of the treatment of epilepsy during childhood and adolescence. Arizona Medicine, 39, 35-38. Gregoriades, A. D. (1972). A medical and social survey of 231 children with seizures. Epilepsia, 13, 13-20. Harris, H. (1976). Clinical and biochemical aspects of epilepsy. In H. F. Bradford & C. D. Marsden (Eds.), Biochemistry and neurology (pp. 213-240). New York: Academic Press. Hartlage, L. C. (1974). Ten year changes in attitudes toward different types of handicaps. lnteramerican Journal ofPsychology, 8, 1-2.
Hartlage, L. C. (1981). Neuropsychological assessment of anticonvulsant drug toxicity. Clinical Neuropsychology, 3. 20-
22. Hartlage, L. C. (1984, August). False positive lateralizing signs related to anticonvulsant medications. Paper presented at the American Psychological Association, Toronto. Hartlage, L. C., & Green, J. B. (1972). The relation of parental attitudes to academic and social achievement in epileptic children. Epilepsia, 13, 21-26. Hartlage, L. C., Green, J. B., & Offutt, L. (1972). Dependency in epileptic children. Epilepsia, 13, 27-30. Hartlage, L. C., Hynd, G., & Telzrow, C. (1981, October). Canonical variate analysis ofanticonvulsant medication effects in neurological measures. Paper presented at the National Academy of Neuropsychologists, Orlando. Hartlage, L. C., & Linz, T. (1984, October). Differential stability ofneuropsychological tests with epileptic patients. Paper presented at the National Academy of Neuropsychologists, San Diego. Hartlage, L. C., McCuiston, M., & Noonan, M. (1982, October). Stability of correlations between serum phenobarbital levels and neuropsychological test performance. Paper presented at the National Academy of Neuropsychologists, Atlanta. Hartlage, L. C., Nance, L., Noonan, M., & Shaw, J. (1983, February). A neuropsychological screening battery for drug toxicity. Paper presented at the International Neuropsychological Society, Mexico City. Hartlage, L. C., Noonan, M., & Prim, J. (1983, October). Anticonvulsant serum level vs. dosage correlated with neuropsychological test performance. Paper presented at the National Academy of Neuropsychologists, Houston. Hartlage, L. C., & Roland, P. E. (1971). Attitudes toward different types of handicapped workers. Journal of Applied Rehabilitation Counseling, 2. 155-161. Hartlage, L. C., Roland, P. E., & Taraba, D. (1971). Perceptions of various types of disabilities by employers. Psychological Aspects of Disability, 18, 122-124. Hartlage, L. C., & Taraba, D. (1971 ). Implications of differential employer acceptance of individuals with physical, mental and social handicaps. Rehabilitation Research and Practice Review, 2, 45-48. Hartlage, L. C., & Telzrow, C. F. (1984). Neuropsychological aspects of childhood epilepsy. In R. Tarter & G. Goldstein (Eds.), Advances in clinical neuropsychology (pp. 159179). New York: Plenum Press. Heathers, G. (1953). Acquiring dependence and independence: A theoretical orientation. Journal of Genetic Psychology, 87,
277-291. Henderson, P. (1953). Epilepsy in school children. British Journal of the Preventive Society of Medicine, 7, 9-13. Hermann, B. P., Dikmen, S., & Wilensky, A. (1982). Increased psychopathology associated with multiple seizure types: Fact or artifact? Epilepsia, 23, 587-595. Hermann, B. P., & Stevens, J. R. (1980). Interictal behavioural correlates of the epilepsies. In B. P. Hermann (Ed.), A multidisciplinary handbook of epilepsy (pp. 272-307). Springfield, IL: Thomas. ldestrom, C. M., Schalling, D., Carlquist, U., & Sjtiquist, F. (1972). Behavioral and psychophysiological studies: Acute
NEUROPSYCHOLOGICAL ASPECTS OF EPILEPSY effects of diphenylhydantoin in relation to plasma level. Psychological Medicine, 2, 111-120. Jackson. J. H. (1931). Epilepsy and epileptiform convulsions. In 1. Taylor (Ed.), Selected writings (Vol. I, pp. 193-247). London: Hodder & Staughton. Kaufman, D. H. (1981). Clinical neurologyfor psychiatrists. New York: Grone & Stratton. Keith, H. M., Evert, J. C., Green, M. W., & Gage, R. P. (1955). Mental status of children with convulsive disorders. Neurology, 5. 419-425. King, D. W., Flanigin, H. F., Gallagher, B. B., So, E. L., Murvin, A. J., Smith, D. B., Oommen, K. F., Feldman, D. S., & Power, J. (1986). Temporal lobectomy for partial complex seizures: Evaluating results, and I year follow-up. Neurology, 36, 334-339. Lacher,D., Lewis, R., & Kupke, T. (1979). MMPI in differentiation of temporal lobe and nontemporal lobe epilepsy: Investigation of three levels of test performance. Journal ofClinical and Consultant Psychology, 47, 186-188. Lennox, W. G. (1942). Brain injury, drugs and environment as causes of mental decay in epilepsy. American Journal of Psychiatry, 99, 174-180. Livingston, S. (1972). Comprehensive management of epilepsy in infancy, childhood, and adolescence. Springfield, IL: Charles C. Thomas. Livingston, S., Bridge, E., & Kajdi, L. (1974). Febrile convulsions: A clinical study with special reference to heredity and prognosis. Journal of Pediatrics, 31, 509-512. Loveland, N., Smith, B., & Forster, F. (1957). Mental and emotional changes in epileptic patients on continuous anticonvulsant medication. Neurology, 7, 856-865. MacLeod, C. M., Dekaban, A., & Hunt, L. (1978). Memory impairment in epileptic patients: Selective effects of phenobarbital concentration. Science, 202, 1102-1104. Matthews, C. G., & Klove, H. (1%7). Differential psychological performances in major motor, psychomotor, and mixed seizure classifications of known and unknown etiology. Epilepsia, 8, 117-128. Matthews, W., Barabas, G., & Ferrari, M. (1982). Emotional concomitants of childhood epilepsy. Epilepsia, 23, 671-684. Matthews, W., Barabas, G., & Ferrari, M. (1983). Achievement and school behavior among children with epilepsy. Psychology in the Schools, 26, 10-13. NINCDS. (1980). Epilepsy, the N/NCDS research program. Bethesda, MD: National Institutes of Health. O'Leary, D. S., Seidenberg, M., Berent, S., & Boll, T. J. (1981). Effects of age of onset of tonic-clonic seizures on neuropsychological performance in children. Epilepsia, 22, 197-204. Ounsted, C. (1969). Aggression and epilepsy: Rage in children with temporal lobe epilepsy. Journal of Psychosomatic Research, 13, 237-242. Pamas, J., Gram, L., & Flachs, H. (1980). Psychophar-
417
macological aspects of antiepileptic treatment. Progress in Neurobiology, /5, 119-138. Penfield, W., & Flanigin, H. (1950). Surgical therapy of temporal lobe seizures. Archives of Neurology and Psychiatry, 64, 491-500. Reitan, R. M., & Davison, L.A. (Eds.). (1974). Clinical neuropsychology: Current status and applications. Washington, DC: Winston. Reynolds, E. M., & Travers, D. (1974). Serum anticonvulsant concentrations in epileptic patients with mental symptoms. British Journal of Psychiatry, 124, 440-445. Roye, D., & Martin, F. (1959). Standardized psychometrical test applied to the analysis of effect of anticonvulsant medication on the intellectual proficiency of young epileptics. Epilepsia, I, 189-207. Scarpa, P., & Carassini, B. (1982). Partial epilepsy in childhood: Clinical and EEG study of 261 cases. Epilepsia, 23, 333341. Skilbeck, C. (1984). Computer assistance in the management of memory and cognitive impairment. In B. Wilson & N. Moffat (Eds.), Clinical manasement of memory problems (pp. 112-131). New York: Aspen. Somerfield-Ziskind, E., &Ziskind, E. (1940). Effect of phenobarbital on the mentality of epileptic patients. Archives of Neurology and Psychiatry, 125, 678-685. Standage, K. F., & Fenton, G. W. (1975). Psychiatric symptom profiles of patients with epilepsy: A controlled investigation. Psychological Medicine, 5, 152-160. Sullivan, E. B., & Gahagan, L. (1935). On intelligence of epileptic children. Genetic Psychology Monograph, 17, 309-375. Trimble, M. R., &Reynolds, E. M. (1976). Anticonvulsant drugs and mental symptoms: A review. Psychological Medicine, 6, 169-178. Waxman, S. G., & Geschwind, N. (1975). The interictal behavior syndrome of temporal lobe epilepsy. Archives of General Psychiatry, 32, 1580-1588. Whitehouse, D. ( 1971 ). Psychological and neurological correlates of seizure disorders. Johns Hopkins Medical Journal, 129, 36-42. Whitman, S., Hermann, B. P., & Gordon, A. (1980). Psychopathology in epilepsy: How great is the risk? Presented at the annual meeting of the American Epilepsy Society, San Diego. Whittle,l. R., Ellis, H. 1., & Simpson, D. A. (1981). The surgical treatment of intractable childhood and adolescent epilepsy. Australian and New Zealand Journal of Surgery, 51, 190-
196. Woodbury, D. M., Penry, J. K., & Schmidt, R. P. (Eds.). (1972). Antiepileptic drugs. New York: Raven Press. Zimmerman, F. T., Burgemeister, B., & Putnam, T. (1948). The ceiling effect of glutamic acid upon intelligence in children and adolescents. American Journal of Psychiatry, /04, 593.
23 The Neuropsychology of Epilepsy Psychological and Social Impact THOMAS L. BENNETT AND LINDA K. KREIN
Epilepsy is a nervous system disturbance that abruptly interferes with ongoing behavior, perception, movement, consciousness, or other brain functions. Individual attacks are called seizures, and when the problem is persistent it is called either a seizure disorder or epilepsy. Seizures are relatively common among infants, children, and adolescents. Probably 8 of every 1000 children experience some sort of seizure activity, even if it is only a single occurrence of a febrile seizure (Lechtenberg, 1984). Occasionally, a seizure disorder will disappear as a child matures, but in a majority of cases, childhood epilepsy persists into adulthood, and in about 80% of adults with epilepsy this condition developed when they were children. At least 5 of every 1000 adults in the United States have epilepsy (Lechtenberg, 1984). The various types of epilepsy were described in Chapter 21 of this volume. The purpose of this chapter is to describe the neuropsychology of epilepsy, and the emphasis will be to discuss the emotional/behavioral and cognitive concomitants of epilepsy. We use the term concomitants to underscore the fact that epilepsy is a complex phenomenon, and the behavioral and cognitive events associated with it are the product of a complex interaction among neurological, medication, and psychosocial variables (Hermann & Whitman, 1986).
THOMAS L. BENNETT
AND
LINDA K. KREIN •
Department of Psychology, Colorado State University, Fort Collins, Colorado 80523.
Variables Producing Behavioral and Cognitive Correlates of Epilepsy Neurological Variables Cognitive and behavioral changes associated with epilepsy are most commonly attributed to neurophysiological dysfunction associated with ictal (seizure) events or long-term alterations in the central nervous system associated with repeated discharge. An example of the latter would be the kindling phenomenon (e.g., see Post, 1983) or Bear's (1979) hypothesis that personality changes associated with complex partial seizures reflect a hyperconnectivity or a hyperexcitability of the limbic system. A number of variables, as reviewed by Hermann and Whitman (1986), have been posited as determining the magnitude of behavioral changes. In general, more significant effects are thought to result if the seizure disorder starts at an early age, in patients with poor seizure control, in individuals who have had the disorder for a relatively long period of time, and if the person exhibits multiple seizure types. Complex partial seizures, particularly if of temporal lobe origin, are typ~ ically believed to produce more obvious cognitive and behavioral changes than most seizure types. Antiepilepsy drugs or anticonvulsant medications have recently been implicated in producing negative cognitive and emotional effects in patients with seizure disorders. As a general rule, toxic blood serum levels of antiepilepsy drugs adversely affect
419
420
CHAPTER 23
behavior and cognition (Reynolds, 1983), but adverse biobehavioral effects are sometimes associated with serum levels of antiepilepsy drugs that are within the therapeutic range (Thompson, Huppert, & Trimble, 1981). Polypharmacy increases the risk of epilepsy patients developing cognitive and emotional disorders (Thompson & Trimble, 1982). Generally, phenobarbital and primidone (20% of which is metabolized into phenobarbital) are thought to produce the most significant effects, but phenytoin has also been implicated. Sodium valproate and carbamazepine have been found to produce less significant effects. Generally, antiepilepsy drugs can magnify behavioral and cognitive changes produced by the seizure disorder itself, but in contrast, carbamazepine has been argued to produce positive psychotropic effects (Dalby, 1975).
Psychosocial Variables Recently, many theorists have highlighted ways in which unique social and interpersonal stresses associated with having a seizure disorder may contribute to behavioral and emotional phenomena attributed to epilepsy (Whitman & Hermann, 1986). Significant stressors include fear of seizures and related fears of injury and death, perceived stigma of being "epileptic," perceived discrimination (particularly in employment) against the person with epilepsy, loss of control over one's life due to the unpredictability of seizures, social support (or lack of it), and treatment of the patient by family in the home environment.
Contributions of Neuropsychology to the Assessment and Treatment of Epilepsy Regardless of the dynamic interactions involving neuropsychological, psychosocial, and medication variables that contribute to the constellation of symptoms that constitute epilepsy, behavioral and cognitive correlates of this disorder are observed, and the neuropsychologist can contribute in many ways to its assessment and treatment. Because patients with epilepsy frequently present with a variety of cognitive and psychomotor deficits, as will be described in this chapter, neuropsychological assessment can be a valuable aid in establishing the severity of these impairments and monitoring the effects of treatments on these deficits. Specifically, neuropsychological evaluation before and after treatment
is begun, and before and after major medication changes, could be used to determine if such changes produced positive effects for the patient; the evaluations could help determine an antiepilepsy medica.tion regimen that would strike an optimal balance between seizure control and adverse cognitive effects (Trimble & Thompson, 1986). Finally, knowledge of the nature and extent of neuropsychological deficits may be of help in devising remedial programs or compensatory strategies to alleviate these deficits. The neuropsychologist can provide a direct role in treatment through traditional group and individual psychotherapy. The neuropsychologist can be an educational source for the child and his or her family about the nature of epilepsy, can help the child deal with psychosocial stressors associated with this disorder, and can help the family or teachers of the child devise methods to manage objectionable behavior. As indicated, remedial or compensatory strategies can be devised to alleviate cognitive deficits. The neuropsychologist may also be involved in stressmanagement and/ or biofeedback training whose goal it is to reduce seizure frequency. Behavioral approaches to the treatment of epilepsy will be described in the last major section of this chapter. Thus, the contributions of the neuropsychologist to the assessment and treatment of epilepsy can be significant, and indeed, it is not uncommon for a neurologist to request that a neuropsychologist do a behavioral/neuropsychological assessment during the initial diagnostic evaluation of a child thought to have epilepsy. A common consultation question is, "What are the relative contributions of epilepsy and emotional factors to a child's behavior?" To accomplish this, the neuropsychologist must have an understanding of the neurophysiology, cognitive deficits, and emotional consequences of epilepsy. The goal of this chapter is to provide an overview of cognitive deficits and emotional consequences of this disorder and describe behavior treatment approaches available to the neuropsychologist.
Epilepsy and Emotional and Behavioral Disorders Introduction Most children who have epilepsy without any other nervous system pathology have no obvious emotional or behavioral disorders, but of course, there are exceptions. Problems with behavior are more likely to occur if other neurological signs, such
THE NEUROPSYCHOLOGY OF EPILEPSY
as limb weakness or dyscoordination, accompany the epilepsy. Overall, most authorities agree that emotional and behavioral disorders are more common among children with epilepsy compared to nonepileptic, healthy children or children with other chronic, nonneurological disorders, but estimates of the scope of this problem vary widely (Corbett & Trimble, 1983). For example, some studies have used a restricted sample by investigating the incidence of behavior disorders in a school populated only by epileptic children or in a clinic for children with epilepsy. Such approaches inflate the incidence of behavioral problems in children with epilepsy by only investigating children with complicated seizure disorders. Thus, Bridge's (1949) study in the United States, which concluded that 46% of 742 children attending an epilepsy clinic had personality disorders, is of interest only from a historical perspective. An epidemiological approach is required to obtain accurate population statistics. Such investigations were conducted in the 1970s by Rutter and his colleagues (Rutter, Graham, & Yule, 1970) on the Isle of Wright and by Mellor (1977) in Scotland. These studies used standardized rating scales to assess behavior, and found an incidence of 27-29% psychiatric disorders in children with epilepsy compared with 12-15% in nmtched controls. These children did not experience additional neurological disorders concurrently with epilepsy. Estimates of epilepsy rose as high as 67% when children with complicated epilepsy who attended a special school for children with epilepsy were evaluated (Corbett & Trimble, 1967). (In cases of complicated epilepsy, the epilepsy is not idiopathic, but rather is associated with identified neuropathology or structural lesion of the brain.) A variety of behavioral and emotional disorders have been noted in children with epilepsy including . increased irritability, temper outbursts, violence, aggression, hyperactivity, difficulty socializing with other children, low self-esteem, lowered self-expectations, impaired participation in family activities, and difficulties meeting the common demands that any child must face (Keating, 1961; Lechtenberg, 1984). Although these problems are seen at a higher frequency in children with epilepsy than in matched controls, the basis for these behavioral disorders is not immediately apparent. Several investigators have attempted to identify variables that might be associated with increased incidence of emotional and behavioral problems in children. For example, Holdsworth and Whitmore (1974), investigating a sample of 85 children with epilepsy who attended ordinary schools, reported
421
that 21% exhibited prominent deviant behavior. They concluded that the best predictors of behavior disorders were frequency of seizures during the past 12 months, educational retardation, and the absence of phenobarbital as an antiepilepsy medication. Stores ( 1978, 1980) concluded that boys with epilepsy had more psychiatric problems than girls, particularly those with a left temporal lobe epileptiform focus. Ounsted (1969) reported that rage outbursts in children with complex partial seizures of temporal lobe origin were associated with neurological and psychosocial factors including neurological insult by head trauma, age of onset of seizures, and the presence of a disordered home. Pritchard, Lombroso, and Mcintyre ( 1980) also reported that early age at onset of epilepsy was associated with an increase in behavioral and emotional problems, and Rutter et al. ( 1970) found that a lower socioeconomic level and the psychiatric status of the child's mother increased the risk of psychopathology in children with epilepsy. Finally, Hermann ( 1982) found that children with good neuropsychological functioning who had epilepsy were more aggressive, showed more psychopathology, and were less socially competent than a matched group of children with epilepsy whose · neuropsychological functioning was poor. Taken together, these studies demonstrate that a variety of factors contribute to the increased incidence of behavioral dysfunction observed in children with epilepsy. They suggest that behavioral and emotional disorders reflect a complex interaction among neurological, medication, and psychosocial variables. Examples of these variables will be discussed next.
Neurological Mechanisms of Emotional and Behavioral Problems The view that neurological variables predispose individuals with epilepsy to psychopathology is best supported in studies that investigated adult patients with complex partial seizures of temporal lobe origin. These patients exhibited more psychopathology than patients with focal seizures originating from regions of the cortex other than the temporal lobes (Gunn, 1977; Hermann, Black, & Chhabria, 1981; Sherwin, Peron-Magnan, Bacaud, Bonis, & Tolairach, 1982; Stevens, 1966). But there have been exceptions (e.g., Glass & Mattson, 1973). In addition, patients with complex partial seizures of temporal lobe origin have been reported to show a constellation of personality changes that allegedly represent an organic personality syndrome (Bear, 1979; Bear
422
CHAPTER 23
& Fedio, 1977), but whether this syndrome is indeed specific to complex partial seizures has been disputed (e.g., Mungas, 1982; Stevens, 1982). The central personality attributes seen in patients with complex partial seizures include deepened emotions, increased aggressiveness, alterations in social interactions (interpersonal viscosity circumstantiality), hyperreligiosity, hyposexuality, and hypergraphia (Bennett, 1987). The neurological basis for these personality changes is posited to be a hyperconnectivity or hyperexcitability of the limbic system (Bear, 1979), the system of the brain that processes and mediates emotional feelings and responses (Bennett, 1982). The neuropathological process by which hyperconnectivity develops is thought to be analogous to the experimental phenomenon of kindling, which has been studied extensively in recent years (e.g., Post, 1983). According to the kindling model, repeated discharge from an epileptic focus would result in a lowering of threshold to excitation and seizure in adjacent structures to which neurons from the focus project. The altered excitability of the latter structures, in this case the limbic system, would produce or underlie personality changes seen in many patients with complex partial seizures. A compelling argument for the view that personality changes seen in the individual with complex partial seizures are neurologically based and reflect hyperexcitability or hyperconnectivity in limbic system neurons is its antithesis, the Kluver-Bucy syndrome (Kluver & Bucy, 1939). This syndrome refers to several behavioral changes, first observed in monkeys, that result from bilateral temporal lobectomy. In contrast to the personality attributes characteristic of the complex partial seizure patient, the temporallobectomized animal (and human; see Marlowe, Mancall, & Thomas, 1975) shows a decrease rather than an increase in aggression. Sexual behavior and drive are increased rather than decreased. Social cohesion is diminished rather than heightened, and exploratory behavior rather than excessive attention to detail is prominent. These findings support the view that neurological factors contribute to behavior and emotional problems seen in children with epilepsy. The majority of the research that has evaluated the relationship between personality and epilepsy has been conducted using adults. Nevertheless, it is logical to assume that similar factors, such as limbic system hyperconnectivity, contribute to the increased tendency toward emotional and behavioral problems in children with epilepsy.
Medication Variables in Emotional and Behavioral Problems Adverse personality reactions from some antiepilepsy drugs have been well documented, and as will be reviewed later in this chapter, some antiepilepsy drugs produce cognitive deficits as well. Phenobarbital, used in treating generalized seizures, will at times produce hyperactivity in children, although it is generally sedating in adults. Children taking phenobarbital will, as a result, exhibit inattentiveness, impulsiveness, and aggressiveness. Hyperemotionality and irritability may appear. Primidone, a drug popular in the management of complex partial seizures, can produce many of the same side effects as phenobarbital, but typically they are not as pronounced. This results from the fact that approximately 20% of primidone is metabolized by the brain into phenobarbital. In both cases, the hyperactivity may be so extreme that family, social, and school activities are severely disrupted. Carbamazepine, the drug of choice for complex partial seizures, can, in contrast, produce positive psychiatric effects (Dalby, 1975). Generalized absence (petit mal) seizures are typically treated with ethosuximide. This medication can also produce irritability, hyperactivity, aggressiveness, and inattentiveness. Children being maintained on ethosuximide appear to have an increased susceptibility to disturbances of sleep, especially night terrors. Patients with myoclonic, absence, and other types of epilepsy who do not respond sufficiently to treatment with conventional antiepilepsy medications are often administered clonazepam. Behavioral and emotional side effects include withdrawn behavior, depression, mood swings, insomnia, and auditory hallucinations. Thus, a variety of antiepilepsy medications may produce behavioral and emotional side effects. As a result, a compromise must often be struck between optimal seizure control and negative psychiatric effects of antiepilepsy drugs.
Psychosocial Variables in Emotional and Behavioral Problems As indicated, psychosocial factors perform a major role in the establishment, maintenance, and severity of emotional and behavioral problems seen in children with epilepsy. These variables include fear of seizures, social stigma, loss of control, and treatment by family members and teachers.
THE NEUROPSYCHOLOGY OF EPILEPSY
Fear of Seizures Mittan and his associates (e.g., Mittan, 1986) have shown that fear of seizures may lead to significant psychosocial impairment in people with epilepsy, and their work suggests further that a high level of fear about seizures is associated with significant psychopathology. The most common fear (66%) in their patients was fear of death due to seizures. Patients typically believed that they would die as a result of suffocation from swallowing their tongue. Many other fatal consequences were feared as well. The second most common fear was that they would just suddenly die due to a seizure or that they would have a fatal accident during a seizure. Many thought that if they turned blue during a seizure (cyanosis) they would die. Some thought that if they had epilepsy they must also have a brain tumor. Thus, a majority of patients are concerned about a variety of causes of death from epilepsy, and this is a significant source of psychological distress. Fear of brain damage was almost as common as was fear of death. Interestingly, fewer than I of 20 patients ever discuss these possibilities with their neurologist, and few neurologists ever bring up the topic. Indeed, there appears to be a tacit understanding not to discuss this topic. The fears about death and brain damage can have a pervasive effect on the person's psychological adjustment in that they can lead to decreased coping ability, social withdrawal, poorer emotional adjustment, and disruption of family life. Mittan's conclusions are based on data collected with adults, but it is reasonable to assume that similar fears exert a pervasive influence over children with epilepsy. Research is needed to verify this possibility.
Perceived Stigma It has long been believed that the stigma of having epilepsy could predispose the individual to various emotional and behavioral problems. A recent survey of adults by Arntson and his colleagues appears to support this hypothesis (Arntson, Droge, Norton, & Murray, 1986). Patients in their survey complained of depression and anxiety, and they endorsed statements about how epilepsy had stigmatized their lives and left them with feelings of helplessness. In reviewing the relationship between stigma and the development of psychopathology, West ( 1986) concluded that ''felt stigma'' and a deep sense of shame about "being epileptic" are more charac-
423
teristic of the situation experienced by individuals with epilepsy than is actual "enacted stigma" by others. The perception of being different or stigmatized is likely related to emotional and behavioral problems shown by children. This phenomenon has its roots in parental attitudes toward and treatment of the child and is expressed in such parental behavior as concealment of the child's epilepsy, attempts to keep the child "normal" and to deny the existence of epilepsy, and a pattern of parenting characterized as being overly restrictive and overprotective. Concealment, and these other strategies, are eventually uncovered, and in the end, they perpetuate feelings of shame and stigmatization by the child.
Loss of Control Because of the unpredictability of seizures, children may feel a loss of control over their lives. Their belief is ''no matter what I do, I have a seizure.'' As a result, the child may be prone to develop a general expectation that "my life is under the control of others, and nothing I do matters.'' That is, they are likely to develop a belief in an external locus of control (e.g., Matthews, Barabas, & Ferarri, 1982). In general, patients with epilepsy have a higher likelihood of believing in an external locus of control than either healthy people (DeYellis, DeYellis, Walston, & Walston, 1980) or diabetic patients (Matthews et al., 1982). Because individuals with a belief in an external locus of control exhibit a higher incidence of psychopathology than do individuals with a belief in an internal locus of control (Lefcourt, 1976), it is reasonable to assume that this factor has a role in the development of psychopathology in children with epilepsy. Matthews and Barabas (1986) not only concluded that this factor is involved in emotional and behavioral problems, but also discussed some of the mechanisms by which an external locus of control develops in children with epilepsy. One factor, described earlier, is that children with epilepsy are likely to perceive that the occurrence of a seizure is unavoidable. Another is that many of these children are overprotected by their families, teachers, and others who, as a result, do not give the children the opportunity to fully participate in decisions that directly affect them. Such an environment would confirm the child's belief that "I have no power in determining my future." Matthews and Barabas ( 1986) asserted that development of a belief in an external locus of control in
424
CHAPTER 23
children with epilepsy can impede their motivation to learn, to socially interact and form friendships with others, to act appropriately with others, and to pursue happiness. Chronic underachievement can result from the child's inability to recognize his or her own responsibility in academic success and failure. As with the general population, this belief system can produce feelings of helplessness and hopelessness, anxiety, and depression.
Parental Treatment Two aspects of parental treatment that might contribute to behavioral and emotional problems were discussed earlier. It was indicated that (I) parental treatment of children is one factor that can result in children developing an external locus of control and (2) by their attitudes, either overt or covert, parents can also contribute to a child's belief that he or she is stigmatized by epilepsy. Unfortunately, parents of children with epilepsy may have unrealistically limited views of their children's potential. Basically, if the parents believe that the child will not achieve anything major in life, the child will develop this view as well. This expectation may be directly or indirectly communicated to the child, but as a result, the child may underachieve because of a limited self-concept. The limited expectations may lead to negative feelings of self-worth as well, and this, in tum, can lead to emotional and behavioral problems. Parental fears of their child having a seizure in public may also be adopted by the child. This will accentuate any fears the child has about having seizures, and it will decrease the child's confidence that he or she can manage outside the family. As a consequence, the child becomes overly dependent on and clings to the family. This phenomenon is another hindrance to establishing normal peer relationships and another reason why children with epilepsy often have difficulty developing and maintaining relationships with other children.
The Adolescent with Epilepsy Normal adolescent behavior can be a complication for individuals with epilepsy (Lechtenberg, 1984). The crucial developmental issues during this time of life are independence, identity, and conformity. Part of independence is getting a driver's license, and the adolescent may deny that she or he has epilepsy if it will interfere with obtaining this goal.
Identity may be negatively affected if peers know that the individual has to take antiepilepsy medication or suffers from epilepsy and, as a result, treat him or her as being different. Noncompliance with medication is a risk for the adolescent. First, the teenager may resent the fact that epilepsy and the medication is controlling his or her life. Second, they may stop taking antiepilepsy drugs to counteract the perceived stigma of feeling different. Finally, they may quit taking medication to assert their independence from their parents who, for as long as can be remembered, have been reminding them to take their antiepilepsy drugs. Reckless behavior may likewise emerge to oppose their families' excessive cautiousness. Alcohol and other recreational drugs used by teenagers can alter the metabolism of antiepilepsy drugs and, as a side effect, can increase the incidence of seizure activity. Excessive alcohol makes an individual more vulnerable to seizure activity. The seizure occurs when the alcohol level of the blood falls.
Effects of Seizure Disorders on Intellectual and Cognitive Ability Introduction Controversy has long surrounded the extent to which epileptic seizures themselves account for intellectual and cognitive deficits. However, there is general agreement that observable impairments of memory, language, attention and concentration, psychomotor speed, and planning ability are present as neuropsychological consequences of seizure disorders in a large percentage of the epileptic population (Bennett, 1986; Dodrill, 1981; Folsom, 1953; Reitan, 1974; Rodin, Katz, & Lennox, 1976). Whether these deficits would be alleviated through cognitive rehabilitation, as are deficits produced by other neurological factors such as head injury and stroke (Bracy, 1983), has yet to be formally investigated. Historically, cognitive deficits and intellectual impairment have been recognized as an element of the symptomatology of epilepsy for well over 100 years (Blumer, 1984). Attempts to measure these observed dysfunctions were, however, only as sophisticated as the psychometric tools available. As a result, the earliest studies reflect a degree of confusion and misunderstanding surrounding what was actually being measured. The concept of "epileptic dementia'' had gained widespread popularity by the tum of the century, and in the absence of more rele-
THE NEUROPSYCHOLOGY OF EPILEPSY
425
vant psychometric instruments, cognitive function in Seizure Type and Frequency patients with epilepsy was evaluated primarily with In addition to etiological factors, type and freIQ measures (Stores, 1971) and projective personality tests such as the Rorschach (Reitan, 1974). For- quency of seizures constitute important variables in tunately, more recent work reflects a movement the determination of the nature and extent of intellecaway from the evaluation of intellectual abilities tual and cognitive dysfunction. A number of studies within a global intelligence perspective and subjec- have shown generalized major motor seizures to be tive personality assessment toward the measurement associated with greater intellectual and cognitive impairment than are other types of seizures. An early of specific cognitive abilities and impairments. The literature concerning cognitive deficits as- study by Zimmerman, Burgemeister, and Putnam sociated with epilepsy is varied with respect to meth- (1951) investigated intellectual ability in children odology. Results are frequently contradictory, and and adults using the Stanford-Binet, Wechslerrelatively few studies have been done with children. Bellevue, and Merrill-Palmer Performance Tests. Although results of adult studies should not be uni- Mean IQ measured in children and adults with formly generalized to children (Fletcher & Taylor, idiopathic petit mal seizures ranged from 10 to 14 1984; Rourke, Bakker, Fisk, & Strang, 1983) when points higher than in patients whose seizures were interpreted within a framework of developmental and described as grand mal (generalized major motor). maturational considerations, findings in adult studies Using theWAIS and items from the Halstead-Reitan may be useful in understanding the neuropsychologi- battery, Matthews and Klove (1967) found that adult cal consequences of seizure disorders in children, patients who experienced generalized tonic-clonic and in guiding further research efforts in this area. or major motor seizures demonstrated greater overall Results of well-designed adult studies appropriate to intellectual-cognitive impairment than patients with the discussion of the following aspects of epilepsy other types of seizures regardless of whether etiology will therefore be presented in addition to currently was known or unknown. Wilkus and Dodrill (1976) also observed poorer performance by adults with published studies with children. EEG evidence of generalized discharge, compared to a focal seizure group. In addition, they noted that · more frequent seizures were associated with greater Etiology deficits. This negative correlation between cognitive Of the intellectual correlates associated with ability and frequency of seizures was also reported by epilepsy and the variables that alter them, perhaps the Dikmen and Matthews (1977) in a study of 72 adults most predictable is that of etiology and its rela- with major motor seizures of known and unknown tionship to IQ. In his review of research concerned etiology. with intellectual and adaptive functioning in epilepA similar relationship between seizure frequensy, Tarter (1972) summarized studies in which cy and intellectual-cognitive impairment has been etiological factors were considered. IQ scores of in- reported in children. In an early study, Keith, Ewert, dividuals whose seizures were idiopathic ranged Green, and Gage ( 1955) reviewed medical records of from 4 to 11 points higher than scores attained by 296 children and found a regular decreasing progrespatients whose seizures were secondary to known sion in percentage of retarded children as frequency pathology. These differences were observed in both of seizures decreased. This relationship was coninstitutionalized and noninstitutionalized children sistent across all seizure types considered. Also, and adults. cases in which etiology was known showed a greater The same relationship between severity of IQ incidence of retardation (73%) than those in which impairment in known versus unknown etiology has the cause of seizures could not be attributed to been found to occur on measures of neuropsychologi- organic abnormality (22.2%). cal functions as well. In one of a series of investigaMore recently, Farwell, Dodrill, and Batzel tions, Klove and Matthews (1966) found that al- (1985) evaluated a large group of children whose though adult patients with seizures of unknown ages ranged from 6 to 15 years. Within each seizure etiology (idiopathic seizures) demonstrated neuro- type studied, lower seizure frequency was associated psychological impairment as measured by the with greater intelligence test scores (WISC-R). In Halstead battery, when compared to nonepileptic addition, seizure type was found to be a discriminatcontrols, the greatest degree of impairment was noted ing factor when both IQ and neuropsychological in patients whose seizures were a result of known functions were evaluated. The minor motor and brain pathology. atypical absence groups showed statistically signifi-
426
CHAPTER 23
cant lower IQ scores than all other groups. However, children with partial or generalized tonic-clonic seizures demonstrated proportions of Full-Scale IQ scores comparable to those observed in the control group. When considered together, children with epilepsy showed significantly greater neuropsychological impairment than controls as measured by the ageappropriate Halstead-Reitan battery. Overall neuropsychological impairment was found to differentiate between seizure types with greater sensitivity than Full-Scale IQ (WISC-R). Children with minor motor or atypical absence seizures showed mild impairment. The majority of children who experienced only classical absence seizures demonstrated no detectable neuropsychological impairment, but when seizure types were mixed (classic absence plus generalized tonic-clonic) impairment was again evident. Seizure type has been found to affect selected cognitive functions differentially: Quadfasel and Pruyser ( 1955) compared cognitive abilities in adult male patients with generalized seizures versus complex partial seizures and found that memory was impaired only in the partial seizure group. Fedio and Mirsky (1969) assessed the performance of outpatient groups of children (6 to 14 years old) who had left temporal lobe epileptic focus, right temporal lobe focus, or centrencephalic epilepsy (generalized seizures). Children were evaluated using measures of attention, verbal and nonverbal learning and memory, and IQ. The performance of children with epilepsy regardless of seizure type was below that of the epileptic control group. Of greater interest, however, was the pattern of deficits observed between seizure types and within the temporal lobe seizure groups. Children with left temporal lobe seizure focus showed learning and memory deficits on measures that required delayed recall of verbal material whereas children with right temporal lobe focus had greater difficulty with recall tasks involving visuospatial abilities. Significant differences between performance on measures of short-term memory were not evident between groups. Further, children whose seizures were centrencephalic in nature performed at a significantly lower level on tasks of sustained attention than did the temporal lobe groups, but did not demonstrate either short-term or long-term memory impairment. Patterns of intellectual performance on the Wechsler Intelligence Scales (WAIS or WISC-R) that varied with seizure type were observed by Giordani et at. (1985). Adults and children with partial
More than a century ago, Gowers recognized the relationship between early onset of seizure disorder and poor prognosis for mental functioning (Brown & Reynolds, 1981). In general, current research supports this observation. Studies of intellectual and neuropsychological functions in children with epilepsy, regardless of seizure type, indicate that onset of seizures early in life and a consequently
bol (or Coding), Block Design, and Object Assembly than did patients with either generalized or partial
noted that in studies of children, long duration of
seizures performed better on Digit Span, Digit Sym-
secondarily generalized seizures although significant differences between groups on Full-Scale IQ scores were not present. In contrast, some studies have not shown a clear relationship between seizure type and cognitive impairment (Arieff & Yacorzynski, 1942; Scott, Moffett, Matthews, & Ettlinger; 1967) or frequency of seizures and greater intellectual impairment (Delaney, Rosen, Mattson, & Novelly, 1980; Loiseau et al., 1980; Scott et al., 1967). O'Leary et al. (1983) found only one variable that showed a significant difference in performance between various groups of children with seizure disorders. Children with partial seizures performed significantly better on the Tactile Performance Test (TPT-total time) than children with generalized seizures. The partial seizure group in this study, however, was composed of simple partial, complex partial, and partial secondarily generalized seizure types, and this wide variation of seizure types within one group may have accounted for the limited differences seen when groups were compared. Because seizure classifications and their inclusion criteria have not been consistent, particularly in the earlier studies, and populations tested have not been uniform across investigations (institutionalized versus noninstitutionalized), direct comparisons between studies are not always appropriate. The study of seizure type and frequency and its effect on intellectual and cognitive function is further complicated by the severity of seizures and the antiepileptic medications necessary to achieve adequate seizure control. It is also possible that in some cases, the association between observed cognitive deficits and frequency of seizures is due to the extent of cerebral damage which is responsible for both. When considered as a whole, however, current studies suggest that the extent of intellectual and cognitive dysfunction in epilepsy varies with type of seizure, and increases with greater seizure frequency.
Age at Onset and Duration of Disorder
long duration of seizure disorder places children at higher risk for cognitive dysfunction. It should be
THE NEUROPSYCHOLOGY OF EPILEPSY
seizure disorder is necessarily associated with early onset. Many studies of the effect of age at onset in the past have considered only major motor (generalized) seizures. Dikman, Matthews, and Harley (1977) found that adult patients with early onset of major motor seizures (0-5 years of age) obtained significantly lower Verbal, Performance, and Full-Scale IQ scores (W AIS) than a group of patients with later onset of seizures ( 10-15 years of age). Both seizure groups showed impaired neuropsychological functions (Halstead-Reitan) compared to a nonepileptic control group. However, differences in performance between the early and late-onset epileptic groups were not significant. On the other hand, Matthews and Klove ( 1967) found that early onset of major motor seizures resulted in greater impairment of both intellectual and neuropsychological abilities. This difference was observed in both idiopathic seizures and seizures secondary to known pathology. More recently, O'Leary eta/. (1983) studied the effects of early onset of epilepsy in children 9 to 15 years of age with partial versus generalized seizures. Results indicated that both groups of children with early seizure onset performed more poorly on measures of neuropsychological abilities than children whose seizures began at a later age. These findings are consistent with observations of the effect of seizure onset by Farwell et al. (1985), who studied a variety of seizure types, and Scarpa and Carassini (1982) in their study of children with partial seizures. As in studies of seizure frequency, investigation of the effects of age at onset is complicated by anticonvulsant medications. Anticonvulsants have been found to affect cognitive performance in both children and adults (Brown & Reynolds, 1981; Trimble, 1981). In cases of early seizure onset, the effects of anticonvulsants on the developing brain become an important consideration as well as the subsequent long-term drug therapy that must follow. These issues, and current studies that address them will be discussed in further detail later in this chapter.
Epilepsy and General Intelligence Disagreement over the extent of the relationship between epilepsy and intelligence in both children and adults is evident throughout the history of modem research in epilepsy. Early studies of intelligence in children with epilepsy (Collins, 1941; Dawson & Conn, 1927; Fox, 1924; Patterson & Fonner, 1928; Reed, 1951) were limited to cases so severe as to require institutionalization and thus were not representative of the epileptic population as a whole. Not
427
surprisingly, these studies showed the IQ of children with epilepsy to be well below that of children without seizures. More recent research has shown that there is not a simple one-to-one relationship between epilepsy and intelligence and has further acknowledged that both intellectual and cognitive impairment in epilepsy appear to vary as a function of several interrelated factors: location of lesion or epileptogenic focus, etiology, seizure type and frequency, age at onset and duration of seizure disorder, and anticonvulsant drug management. In addition, it has been known for some time that test-retest measures of IQ in children and adults with epilepsy fluctuate more widely and unpredictably in either direction than in the nonepileptic population (Patterson & Fonner, 1928; Sullivan & Gahagan, 1935). This is perhaps due to the frequent but inconsistent altered electrical activity of the brain in epilepsy and the serum anticonvulsant levei at the time of testing. General intelligence in epilepsy was investigated in a retrospective study by Lennox ( 1942) in which the mental status of 1905 patients with various seizure types and etiologies was reviewed. Two-thirds of these patients were described by Lennox as ''mentally normal." However, only 2% were considered above normal and 36% were in the low normal to below normal range, resulting in a distribution in which lower IQ scores were overrepresented. Realizing the fallibility of a subjective evaluation of this kind, Lennox conducted a study in 1951 in association with Collins (Lennox & Lennox, 1960). Using the Wechsler-Bellevue Scale, they found that mean IQ was significantly higher in patients seen in private practice compared to general or veterans hospital practice. Further, patients with complex partial (known as psychomotor seizures at that time) seizures as their only seizure type demonstrated relatively better intellectual ability than patients with other types of seizures or mixed seizure types. In this study as in others, early onset of seizures was associated with a poorer prognosis for mental ability (Brown & Reynolds, 1981; Dikman, Matthews, & Harley, 1977; Klove & Matthews, 1969; Rodin, 1968). Numerous studies have reported the mean IQ of children with epilepsy to be below average (Carlberg & Kavale, 1980; Dennerll, Broeder, & Sokolov, 1964; Halstead, 1957; Keith et al., 1955; O'Leary et al., 1983). Recently, Ellenberg, Hirtz, and Nelson ( 1986) conducted a large-scale longitudinal investigation that confirmed previous observations that, as a group, children with seizure disorders tend to attain lower scores on tests of intellectual function than children without seizures. Children in this study were
428
CHAPTER 23
tested with the Bayley Mental and Motor scales at 8 months of age, the Stanford-Binet at 4 years of age, and the Wechsler Intelligence Scale for Children (WISC-R) at 7 years. Not all researchers have found this relationship between general intelligence and epilepsy. Rutter et al. (1970) reported a distribution of intelligence in children with epilepsy that closely approximated that of the normal population as did Angers and Dennerll (1962) who concluded that IQs of noninstitutionalized adults with epilepsy were comparable to those of the nonepileptic population. Current studies of both children and adults have given more consideration to methodological control and representativeness of the sample studied, seizure type and frequency, age at seizure onset, etiology, duration of seizure disorder, medical intervention, and psychosocial factors that interact to affect intellectual and cognitive performance (e.g., Farwell et al., 1985; Giordani et al., 1985; O'Leary et al., 1983). Results of the majority of studies suggest that IQ scores vary as a function of all of these factors. However, there still remain unanswered questions concerning the relative contribution of each factor to the impairment of general intelligence.
Epilepsy and Intellectual Deterioration The stability of intellectual ability in children and adults with epilepsy has been the subject of several studies, some of which have reported deterioration and others have not (Tarter, 1972). Recently, Ellenberg et al. (1986) compared Full-Scale IQs of7year-olds who had experienced one or more nonfebrile seizures to IQs of their seizure-free siblings. Mean IQs of the children with epilepsy did not differ significantly from previous assessment at age 4. In a substudy, children who had developed seizures in the interval between the first (at age 4) and second. test (at age 7) were evaluated. Analysis of this subset of children did not show a significant difference in IQ. These data suggest that the occurrence of seizures does not result in mental deterioration; however, the maximum duration of seizure disorder possible in the substudy was 3 years. It has been proposed that selected intellectual and cognitive deficits caused by some seizure disorders may require up to 10 years to become evident (Mirsky, Primae, Ajmone-Marsan, Rosvold, & Stevens, 1960). It should also be noted that two different psychometric measures, the Stanford-Binet (at age 4) and the WISC-R (at age 7), were used in testing. Although these tests are comparable and much information can be gained from them, neither test was developed with the intent of
detecting the neuropsychological deficits that may contribute significantly to overall intellectual deterioration. The question of whether seizures are responsible for a gradual decline in intellectual ability is not easily answered, and is complicated by several factors. First, as indicated, a wider variation in testretest scores (upward or downward) is observed in patients with epilepsy than in the nonepileptic population (Brown & Reynolds, 1981) and this factor makes evaluation of the extent of decline in scores difficult. As an example, Fox (1924) in his study of students in a special school for children with epilepsy found that scores of37% of the children retested after 1 year declined, 22% improved, and 41% remained unchanged. Second, as in studies of intellectual and cognitive impairment, investigations of intellectual deterioration are compounded by the factors that complicate assessment of performance: seizure type and frequency, age at onset and duration of seizure disorder, and the presence of an underlying progressive pathology that is responsible for the seizures. Nevertheless, animal studies have indicated that neuronal degeneration occurs as a result of recurrent seizures that inhibit brain protein synthesis and growth (Harris, 1972). A third factor complicating the study of intellectual decline over time is the effect of anticonvulsants on intellectual and cognitive abilities. As indicated, recent research has shown that certain anticonvulsants cause impairment of selected cognitive functions (Corbett, Trimble, & Nichol, 1985). In addition, more adverse neuropsychological deficits might be expected in patients whose seizures are frequent and began early in life, due to the higher doses of antiepileptics or combinations of medication necessary to achieve maximum seizure control and the longer periods of time these drugs must be taken (Dikmen & Matthews, 1977; Dodrill, 1981). Bourgeois, Prensky, Palkes, Talent, and Busch (1983) conducted a prospective study of IQ stability in children aged 18 months to 16 years who bad various types of seizure disorders. Children were evaluated within 2 weeks of initial diagnosis, and yearly thereafter for an average of 4 years. In 45 of the 72 children tested, a nonepileptic sibling was evaluated in parallel. The mean IQ for all of the children with epilepsy considered together initially did not differ significantly from the siblings, nor did the scores change appreciably over time. Eight of the children did, however, show a decrease in IQ of 10 points or greater. The authors explained these findings in terms of drug toxicity, seizure type and frequency, and early onset. Further analysis revealed that of
THE NEUROPSYCHOLOGY OF EPILEPSY
these factors, the two best predictors of intellectual deterioration were age at onset and number of drugs to which the child developed toxicity. This study draws attention to one of the many problems that arise in the treatment of children with epilepsy, the trade-off between maximum seizure control and possible adverse drug side effects. Finally, psychosocial factors may play a role in the assessment of intellectual deterioration. For example, Carlberg and Kavale ( 1980) suggested that the placement of children with epilepsy in special schools may be partially responsible for depressed performance in measures of intelligence. Also, as in school performance, parents and teachers may have a less optimistic view of an epileptic child's ability and potential. Deterioration in performance may then be a result of lowered expectation as weJI as neurological and medication variables (Lechtenberg, 1984). Some of the inconclusive and negative results in past studies of intellectual deterioration may be explained by the history of the development of anticonvulsant drugs. Sullivan and Gahagan (1935) found that premorbidly brighter patients showed greater deterioration in intellectual ability over ·time, but improved seizure control tended to slow the rate of deterioration. Barbiturate anticonvulsants and bromide were, however, the only available chemical means of seizure control until the introduction of phenytoin (Dilantin) in 1938, and were not effective in controlling all types of seizures. The development of newer anticonvulsant drugs and therapeutic interventions such as biofeedback have gradually resulted in better seizure control than in the past. Because of the numerous factors that affect test performance, intellectual deterioration is perhaps more effectively considered individually and relative to premorbid ability rather than in terms of preservation of "average intelligence." For example, a person of superior intellectual ability who acquires a seizure disorder may subsequently score within the average range of intelligence; however, this individual will have experienced marked deterioration in mental ability, which may interfere with his/her previous work and leisure activities, and this loss may be extremely difficult for the individual to accept.
Educational Attainment in Children with Epilepsy Surveys of schools in England (Pond & Bidwen, 1959), Italy (Pazzaglia & Frank-Pazzaglia, 1976), and the United States (Green & Hartlage, 1971) have indicated that children with epilepsy are
429
overrepresented among students who have difficulty in school. These results suggest that the school problems experienced by the child with epilepsy cannot be explained as a function of curriculum or classroom environment across cultures. The most frequent term teachers use to describe problem classroom behavior in epileptic children is "inattentiveness" (Stores, 1978), which is descriptive of both cognitive and behavioral neuropsychological correlates of epilepsy. Bennett-Levy and Stores (1984) described results of a study in which they used an evaluation of classroom performance rather than psychometric test scores to assess the cognitive and behavioral dysfunction in children with epilepsy. After discussions with teachers and observation of classroom behavior, a questionnaire relevant to learning difficulties experienced by epileptic children was constructed. Four factors that were indicative of learning difficulties and therefore predictive of poor educational attainment were identified: concentration, mental processing of information, alertness, and self-confidence. Teacher ratings of children with epilepsy in this study did not differ significantly from ratings of their nonepileptic classmates on concentration, mental processing, or self-confidence; however, significant deficits in alertness were noted in the children with epilepsy, and this deficit was evident in both boys and girls. It was further suggested that the lowered alertness in these children was not associated with type of seizure or drug effects, because the children with epilepsy who were no longer taking anti epileptic medication were also significantly less alert. Until recently, educational attainment has been examined within the context of general intellectual ability (Stores, 1971), but the focus of current research has gradually turned to the specific underlying reasons for poor academic performance in children with epilepsy. The previously discussed cognitive impairments in memory, language, attention concentration, psychomotor speed, and planning ability that have been observed in children and adults with epilepsy represent deficits in mental abilities that are essential to learning and consequently to successful academic performance. Children with epilepsy are in special education programs three times as often as other children (Harrison & Taylor, 1976). Furthermore, academic achievement is below expectancy even when epileptic children have been placed in age-appropriate grade levels in normal schools, and is below that which might be predicted based on intellectual differences alone (Green & Hartlage, 1971; Rutter et a/., 1970). Farwell eta/. ( 1985) investigated school experience of children with epilepsy and found that
430
CHAPTER 23
seizure type was an important determinant in a child's academic experience. Twenty-seven percent of the children with epilepsy (all seizure types considered together) were in special education or had failed a grade, versus only 20% of nonepileptic controls. The poorest academic experience occurred in children with atypical absence or minor motor seizures. Seventeen of the 20 children (85%) with these seizure types had been held back a grade or were in special education classes. Farwell et al. (1985) further compared academic achievement using the Wide Range Achievement Test (WRAT), as a function of seizure type and found that children with classic absence seizures averaged 5 months ahead of grade placement, whereas children with other seizure types averaged from 4 months to 18 months behind grade level. A longitudinal study of children in England, Scotland, and Wales (Ross, Peckham, West, & Butler, 1980) found that by age 11, children with epilepsy who attended normal schools accomplished nonacademic tasks such as copying designs without difficulty, but their scores in reading comprehension and mathematics were below those of their nonepileptic cohorts. In addition, prolonged absence from school was common among the children with epilepsy. A number of other authors have noted that for children with epilepsy, school achievement, particularly in reading and arithmetic, averages behind grade level. Green and Hartlage (1971) observed that WRAT scores in a group of children with epilepsy averaged approximately 1 year behind their grade level placement in reading and 1 year 8 months behind in arithmetic. Still greater deficits in arithmetic skills were noted by Bagley ( 1971) who estimated that children with epilepsy averaged 23.1 months behind their nonepileptic classmates. The learning and behavioral educational problems that are characteristic of many children with epilepsy have been attributed by various researchers to neuropsychological impairment, medical treatment (anticonvulsants), and psychosocial factors (e.g., lower parent and teacher expectations and social stigma surrounding the child with epilepsy). It has been suggested (Serafetinides, 1970) that certain behavioral disturbances, for example, aggressive behavior associated with temporal lobe epilepsy, result from learning deficits-in effect, the child fails to "learn to behave." On the other hand, the belief that social factors generate cognitive deficits that in turn result in poor academic performance is not without support (Lechtenberg, 1984; Matthews & Barabas, 1986).
Epilepsy and Memory A relatively large number of past studies of individuals with epilepsy have focused on verbal memory impairment, which has been observed by others as well as reported by epileptic patients themselves. As early as 1876, Samt, a Berlin psychiatrist, suggested that the presence of memory dysfunction may be useful in the diagnosis of epilepsy (Blumer, 1984). In 1881, Gowers also described memory deficits in terms of impaired recent acquisiton of information in patients with epilepsy (Hill, 1981). Quadfasel and Pruyser (1955) found that a group of adult males with temporal lobe seizures showed significantly poorer verbal memory (Wechsler Memory Scale) than patients with generalized seizures. Glowinski (1973) compared short-term memory in relation to a secondary distraction task in 30 patients with chronic generalized seizures and 30 with chronic unilateral temporal lobe focus. The temporal lobe group showed significantly greater deficits on the Wechsler Memory Scale (WMS) than the group with generalized seizures. In contrast, Mirsky et al. ( 1960) failed to find a differential memory deficit when they compared WMS scores of adults with temporal lobe foci to scores of a matched group with generalized seizures. However, patients with seizures other than those originating in the temporal lobes may have inadvertently been included in the focal seizure group. Several subsequent studies also found no difference between memory function when seizure types were compared (Mignone, Donnelly, & Sadowsky, 1970; Scott et al., 1967). Past studies have often investigated memory in patients with varying methods of clinical substantiation of seizure type, and specific memory functions have not always been consistent or clearly described. As a result, findings have been mixed. Inconsistencies noted across studies may also reflect differential sensitivity of the measures that have frequently been used to evaluate cognitive abilities and misinterpretation of what abilities the tests actually measure. Laterality effects have been evident in memory studies with both children and adults. As previously described in part, Fedio and Mirsky (1969) observed greater verbal memory deficits in children with left temporal lobe foci. Conversely, children whose seizures originated in the right temporal lobe showed greater nonverbal or visuospatial memory deficits. Ladavas, Umilta, and Provinciali ( 1979) investigated both short-term and long-term memory in adult patients with specific left or right temporal lobe focus.
THE NEUROPSYCHOLOGY OF EPILEPSY
In the absence of evidence of structural lesions or other brain pathology, patients with left temporal lobe foci were shown to have greater deficits in longterm verbal memory, whereas presence of right temporallobe foci was consistent with nonverbal memory dysfunction. No differences in short-term memory were observed. Delaney et al. ( 1980) studied immediate and delayed recall of verbal and nonverbal material in adult patients with left or right temporal lobe foci and a group of patients with frontal lobe foci who were matched on age at seizure onset, duration of disorder, and seizure frequency. Both left and right temporal lobe groups showed impairment of immediate verbal recall when compared to the frontal focus and control groups. When the two temporal lobe groups were compared, the group with left temporal lobe foci demonstrated poorer performance on verbal tasks than the group with right temporal lobe foci. Nonverbal tasks resulted in poorer performance by the right temporal lobe group. Other researchers have recently reported similar evidence of Iateralized verbal and nonverbal deficits in focal epilepsy (Masui et al., 1984; Mungas, Ehlers, Walton, & McCutchen, 1985). These observations of lateralized cognitive processes in nonsurgical patients are consistent with results of studies that have considered the effects of temporal lobectomy in severe seizure disorders that are refractory to other forms of medical intervention (e.g., Milner, 1958; Novelly et al., 1984). Lateralization of verbal and nonverbal memory impairment in epilepsy has not been demonstrated in all studies. As indicated, Quadfasel and Pruyser ( 1955) found verbal memory deficits in patients with temporal lobe seizures, but similar deficits in nonverhal memory were not noted and lateralization was not apparent. However, some of the patients tested in this study exhibited bilateral temporal lobe involvement, making detection of left/right differences improbable. Although Glowinski ( 1973) noted differences in verbal and nonverbal memory with respect to left and right temporal lobe foci, the magnitude of these differences was not statistically significant. Some researchers have found that attention deficits related to seizure type occur in epilepsy and may in fact contribute to poor performance on learning and memory measures. Children and adults whose seizures are generalized have shown greater attention and concentration deficits than patients with partial seizures (Fedio & Mirsky, 1969; Glowinski, 1973; Kimura, 1964; Lansdell & Mirsky, 1964; Mirsky & Van Buren, 1965). Loiseau, Signoret, and Strube (1984) tested patients ranging in age from 15 to 36 years on learning, memory, and attention. Although
431
the sample tested was small, significant differences due mainly to the generalized seizure group were noted on all tests. When epileptic groups were considered independently, the performance of patients with partial seizures was similar to that of the nonepileptic group on learning and memory tasks. Loiseau et al. concluded that epileptic patients suffer from both learning and attention disabilities. Memory impairment in patients with epilepsy has also been explained in terms of language deficits by Mayeux, Brandt, Rosen, and Benson (1980) who investigated both interictal memory and language dysfunction in volunteers with idiopathic seizure disorders. Participants were matched for age, seizure frequency, duration of seizure disorder, and level of education. Patients with EEG confirmation of right temporal lobe foci, left temporal lobe foci, or generalized seizures were evaluated on measures of intelligence, auditory_ and visual memory, and language. No significant differences were observed between groups on the WAIS (verbal or performance subscales) or memory measures including WMS (overall or subtests), Rey-Ostereith Complex Figure Test, or Benton Visual Retention Tasks. However, markedly significant deficits in the left temporal lobe group were noted on the Boston Naming Test and ·Controlled Word Association Test. The authors suggested that anomia demonstrated in temporal lobe epilepsy patients may be incorrectly interpreted by patients and their relatives as poor memory. Mayeux et al. further suggested that the verbosity and circumstantiality observed in some patients with temporal lobe epilepsy (Bear & Fedio, 1977) may be the expression of a compensatory mechanism for anomia. Few formal studies have directly investigated memory disturbances in noninstitutionalized children with epilepsy whose intelligence is normal or above. Research more often has been concerned with school success or failure and has considered memory deficits as part of learning disorders in general. A recent study by Camfield et al. ( 1984) compared cognitive ability, personality profiles, and school success in children with epilepsy. Groups tested were children with clinically well-documented unilateral left versus right temporal lobe origin of seizures. The children were tested using the WISC-R, the Halstead-Reitan battery, and the Wide Range Achievement Test (WRAT). Parents completed the Personality Inventory for Children (PIC). In contrast to Fedio and Mirsky's (1969) study with children, significant left versus right differences were not found on measures of cognitive ability, although the combined left and right temporal lobe groups as a whole
432
CHAPTER 23
scored significantly lower than the nonepileptic control group on the WISC-R, Halstead-Reitan, and WRAT. Failure to find left-right differences in cognitive abilities may have been due to the wide range ofchildren tested (6-17 years), the wide range of duration of seizure disorder (1-12 years), and the relatively low seizure frequency. Results of past studies indicate that duration of seizure disorder and seizure frequency are positively correlated with cognitive dysfunction in epileptic children (Fedio & Mirsky, 1969; O'Leary eta/., 1983). In addition, the test battery used in this study may not have been adequately sensitive to the selective cognitive impairment seen in temporal lobe epilepsy. Although the studies reviewed represent varied results with respect to the association between degree of impairment and seizure type, frequency, and duration, considered together, they suggest that in children and adults, cognitive deficits in memory exist as a neurological consequence of epilepsy, that these deficits are present to differing extents in generalized seizure disorders and partial seizure disorders, and that verbal and nonverbal memory functions appear to be Iateralized. Discrepancies in results can, in part, be explained by differences in test measures used across experiments, and the extent to which intervening variables such as seizure type and frequency, age of onset, duration of disorder, and anticonvulsant medication were controlled.
Intellectual and Cognitive Effects of Antiepileptic Drugs Intellectual and cognitive impairments, especially memory deficits, in the epileptic population were observed and noted in the literature long before the earliest anticonvulsant drugs were known (Trimble & Thompson, 1981). A recent study (Epir, Renda, & Baser, 1984) in which untreated children with epilepsy performed more poorly on the Peabody Picture Vocabulary Test and a drawing task than their nonepileptic siblings and a control group lends support to the earlier observations. Thus, cognitive deficits can result from seizures themselves. The disentangling of medication effects from seizure effects represents a formidable challenge to researchers. However, there is increasing evidence that many anticonvulsants affect cognitive abilities and thereby add to the impairment of function caused by seizures (American Academy of Pediatrics, 1985; Corbett et al., 1985; Reynolds & Trimble, 1985). Toxic doses of virtually all anticonvulsants can affect mental functioning. It is, however, the pos-
sibility of cognitive impairment resulting from serum concentrations of antiepileptic drugs in the therapeutic range that is of most concern in the long-term treatment of epilepsy. As indicated, detrimental effects on neuropsychological abilities have been reported for most anticonvulsants. The nature and extent of these deficits vary with the drug or combination of drugs administered, as well as the serum concentration of the drug. Polytherapy (administration of more than one anticonvulsant) has been found to result in greater deficits (MacLeod, Dekaban, & Hunt, 1978; Shorvon & Reynolds, 1979; Thompson & Trimble, 1982). The reduction of polytherapy in chronic epileptic patients, when possible, has contributed to a better understanding of the subtle effects anticonvulsants have on cognitive function (Fischbacher, 1982; Shorvon & Reynolds, 1979; Thompson & Trimble, 1983). A study of 312 children with epilepsy (Trimble, Corbett, & Donaldson, 1980) indicated that children whose IQs deteriorated 10 to 40 points in an interval of at least I year had significantly higher blood levels of phenytoin and primidone. A similar trend was found for phenobarbitone even though all serum drug concentrations were within the optimum therapeutic range. The depletion of folate levels by anticonvulsants, in particular phenytoin and phenobarbitone, has been suggested as the underlying mechanism of cognitive deterioration in children and adults with epilepsy (Reynolds, 1970). This study supported this hypothesis. The group with low folate levels also contained significantly more children who were being treated with phenytoin. The antiepileptic drugs commonly prescribed for children, their indication(s) for use, usual maintenance dosages, and reported cognitive and behavioral side effects are summarized in Table I. It should be noted that this information is representative of the average, and individual cases may vary.
Phenobarbital. Results of early studies (Grinker, 1929; Lennox, 1942; Somerfeld-Ziskind and Ziskind, 1940) led to the conclusion that despite its sedative properties, phenobarbital had no effect on cognitive ability. However, with the development of more sensitive neuropsychological tests and the medical technology necessary to monitor accurately blood levels of anticonvulsants, further investigations began to implicate antiepileptic drugs in the impairment of intellectual and cognitive dysfunction. Hutt, Jackson, Belshum, and Higgins (1968) tested the effect of phenobarbitone on nonepileptic volunteers and found that it impaired sustained attention and psychomotor performance.
Barbiturate anaIogue
Hydantoin
Succinimide
Carboxylic acid derivative Benzodiazepine
Iminostllbene derivative structurally similar to the tricyclic antidepressants
Primidone (Mysoline)
Phenytoin (Dilantin)
Ethosuximide (Zarontin)
Valproic acid (Depakene) Clonazepam (Kionopinb)
Carbamazepine (Tegretol)
"American Hospital Formulary Service. bformerly Clonopin.
Long-acting barbiturate
Chemical classification
Phenobarbital (Luminal)
Generic name (trade name)
Major motor Psychomotor
Absence Minor motor Absence Minor motor Myoclonic
Absence
Major motor
Major motor Psychomotor
Major motor
Primary indications
8
Varies with age: younger than 6 years= 10-20 mglkg per day; 6-12 years = not to exceed I g/day
Not to exceed 0.2
10-60
20
4-8
in children 8 or older
250 mg three or four times daily
10-25 in children younger than
3-5
Usual pediatric maintenance dose (mg/kg per day)a
Reported behavioral side effects
Impaired performance on motor tasks
Impaired attention and concentration, problem-solving and visuomotor performance Impaired attention
Insomnia, aggression, irritability, hyperactivity, antisocial acts, disobedience, social withdrawal, auditory hallucinations, depression, mood swings Agitation, insomnia, irritability, hyperemotionality
Mood changes, drowsiness, irritability, insomnia, hyperactivity, disturbed sleep (especially night terrors), inattentiveness Drowsiness
Fatigue, emotionality, involuntary movements
Hyperactivity, fussiness, disturbed sleep, irritability, disobedience, depressive symptoms, lethargy, hyperemotionality, inattentiveness See phenobarbital (side effects not as pronounced)
Impaired short-term memory and concentration
Reported cognitive side effects
TABLE 1. Commonly Prescribed Anticonvulsants for Children
e
~
~
~
I
tll
6!
434
CHAPTER 23
Primidone. Primidone is a barbiturate analogue that is metabolized to phenobarbital and phenylethylmalonamide (PEMA) whose actions may be synergistic. There are little data available with respect to the effect of primidone on cognitive ability in children, although its sedative, and consequently cognitive, effects may be expected to parallel those of phenobarbital. Phenytoin. When phenytoin (Dilantin) was initially introduced as an antiepileptic drug, it was thought to improve alertness (Trimble, 1981). More recent research has indicated that phenytoin has adverse effects on psychomotor performance (Idestrom, Schalling, Carlqvist, & Sjoqvist, 1972; Thompson & Trimble, 1982), concentration (Andrewes, Tomlinson, Elwes, & Reynolds, 1984; Dodrill & Troupin, 1977), memory (Andrewes eta/., 1984; Thompson & Trimble, 1982), and problem solving (Dodrill & Troupin, 1977). Of additional concern is the possibility of a phenytoin-induced encephalopathy, which has been observed without other clear neurological evidence of toxicity (Reynolds, 1970). In patients with preexisting mental retardation or brain damage, further deterioration in intellectual functioning in the absence of classical signs of toxicity may be intetpreted as an effect of the underlying pathology rather than due to medication-induced encephalopathy. Ethosuximide. Browne et a/. (1975) studied the effects of ethosuximide on psychometric performance and noted an improvement in 17 of 37 children. Blood levels of ethosuximide were monitored and remained within therapeutic ranges over the 8week duration of the study. These findings were inconsistent with earlier reports (Guey, Charles, Coquery, Roger, & Soulayrol, 1967). However, 15 of 25 children in the study by Guey et at. were mentally retarded and taking other drugs in addition to ethosuximide. Nevertheless, there have been additional reports of psychosis or encephalopathy resulting from ethosuximide administration (Roger, Grangeon, Guey, & Lob, 1968). Valproic Acid (Sodium Valproate). Minimal adverse cognitive effects on psychological test performance were noted with the use of valproic acid (Trimble & Thompson, 1984). In addition, it had previously been suggested that administration of sodium valproate improved alertness and school performance (Barnes & Bower, 1975). On the negative side, there have been some reports of sodium valproate-induced encephalopathy similar to that seen with phenytoin (Davidson, 1983).
Carbamazepine. Schain, Ward, and Guthrie ( 1977) evaluated the cognitive consequences of replacing phenobarbital and primidone with carbamazepine treatment in children with major motor or psychomotor seizure disorders. A battery of measures intended to assess general intelligence, problem-solving ability, and inattentiveness was administered. Substantial improvement in problem-solving measures were noted with carbamazepine use. In addition, the children appeared to be more alert and attentive than when they were treated with phenobarbital or primidone and seizures were still adequately controlled. Thompson and Trimble (1982) also found carbamazepine to have a less detrimental effect on cognitive functioning in adults. Considerable progress has been made in the anticonvulsant control of seizures since the accidental discovery of bromide at the tum of the century; however, many questions remain concerning the effects of these drugs on cognitive functioning.
Behavioral Treatment of Seizure Disorders Although a variety of antiepilepsy drugs are available to treat seizure disorders, 20% of epileptic patients report unsatisfactory control of their seizure activity (Masland, 1974). Higher levels of antiepilepsy medication or polydrug therapy increase the likelihood of cognitive and emotional side effects of these drugs. As a result, a compromise must often be achieved between minimal level of medication side effects and optimal level of seizure reduction. The child may have to tolerate a low level of continuing intermittent seizure activity because of the negative effects produced by the medication. As an alternative, behavioral treatment of epilepsy may provide additional relief from the seizure disorder. Biofeedback has recently been shown to produce a significant reduction in frequency of seizures, even in patients with very intractable seizure disorders. Additional benefit is probably realized if the biofeedback is combined with stress management training. Both individual and group psychotherapy also yield beneficial results.
Biofeedback Biofeedback has been used successfully over the past two decades to treat a variety of psychophysiological disorders. Biofeedback, and specifi-
THE NEUROPSYCHOLOGY OF EPILEPSY
cally EEG biofeedback, as a modality for treating poorly controlled seizures has been shown to be of significant therapeutic value in a number of reports published since Sterman and Friar ( 1972) published their initial case study. Sterman and Friar treated a 23-year-old female with a long-term seizure disorder. The seizures were of a nocturnal generalized tonic-clonic type and occurred at a frequency of one or two per month. She had a childhood history of febrile convulsions but there was no family history of epilepsy. There was no evidence of a localized brain lesion producing the seizure activity. She was treated with a combination of phenytoin and mephobarbital. She was trained to produce 12- to 15-Hz activity monitored with scalp electrodes placed over the sensorimotor strip. During training, she showed an increase in sensorimotor activity with depression of the alpha rhythm. She exhibited a rapid decrease in seizures and became seizure-free during several months of training. No long-term follow-up data are available, but positive personality changes were reported: she became more outgoing, more confident, and more interested in her appearance. She also reported that her sleep became more restful. Subsequent relevant case studies and research on this procedure have been reviewed by Lubar (1984) and Sterman (1984). To date, more than 50 published studies have demonstrated that EEG biofeedback can play a significant role in the management of poorly controlled seizures. Unfortunately, due to the cost and technological requirements of establishing an EEG biofeedback laboratory, this procedure is used only by a small number of clinicians and researchers in selected medical and university settings. Lubar ( 1984) estimated that Jess than 250 patients nationwide have had EEG biofeedback training for seizure disorders. The most effective EEG biofeedback approach for epilepsy appears to be a combination of training in at least two waveforms. The first is the 12- to 15-Hz sensorimotor rhythm (SMR), which the individual learns to increase. The second is the 4- to 7-Hz theta rhythm, which the client is trained to decrease. Patients are typically taught to reduce the incidence of epileptiform spiking or sharp waves as well. SMR is recorded via scalp electrodes placed over the sensorimotor cortex (Rolandic cortex). SMR is typically difficult to observe in unfiltered EEG records, but it can, at times, be fairly prominent in the records of highly trained individuals. This EEG rhythm is generated by pools of neurons located in the ventrobasal nuclear complex, most likely in the ventral posteriomedial nucleus. This nucleus is in-
435
volved in transmission of somesthetic information to the sensorimotor cortex. SMR, like alpha, theta, and other sinusoidal EEG rhythms, reflects an "idling" phenomenon in the brain, and this may be associated with an inhibitory process (Andersen & Andersson, 1968). However, these sinusoidal rhythms do not reflect the same process, can occur independently of one another, and have different neural generators (e.g., Shabsin, Bahler, & Lubar, 1979; Sterman & Friar, 1972). Thus, these rhythms reflect idling phenomena in independent, but at times related, neurobehavioral systems. In the case of SMR, the idling or inhibitory process appears to be for movement or motor reflexes, a view supported by some of the early research by Sterman and his colleagues (Sterman & Wywricka, 1967; Sterman, Wywricka, & Roth, 1969). SMR biofeedback training-induced EEG changes that accompany seizure reduction are assumed by Sterman ( 1984) to result from a decrease in abnormal thalamocortical excitability. Wyler (1984) offered a different interpretation. He believes that the beneficial effects of SMR training reflect a resulting prevention and disruption of a synchronization of single neurons into an epileptogenic neuronal aggregate. The fact that SMR training appears to be as effective in reducing the frequency of seizures without motor components as it is in reducing seizures with motor components is support for Wyler's view (Bennett, 1977). The procedure used for EEG biofeedback in patients with epilepsy at most treatment facilities can be summarized as follows (Bennett, 1977). The unfiltered EEG, recorded from the sensorimotor cortex, is passed into two filters. An SMR filter selectively transmits 12- to 15-Hz activity, and the SMR is further processed by an amplitude discriminator. Only EEG waves that are above a predetermined amplitude requirement result in positive feedback to the patient. The threshold (minimum amplitude required for positive feedback) is gradually increased as training progresses. In our program, for example, a green light comes on if the threshold is surpassed; a series of small LEOs indicates the relative waxing and waning of amplitude within an above-threshold train of SMR activity. Patients are trained to keep the large green light on as constantly as possible and to illuminate as many of the LEDs as they can. The unfiltered EEG is also passed into a theta filter that selectively transmits 4- to 7-Hz activity to an amplitude discriminator. If the amplitude from this filter surpasses a predetermined threshold, negative feedback occurs (orange light is illuminated).
436
CHAPTER 23
The patients attempt to keep this orange light off as much as possible. Similar negative feedback results following the occurrence of spike waves, sharp waves, or excessive movements. As training progresses, the patient becomes more adept at increasing the incidence of SMR and decreasing epileptiform activity. Correlated with these EEG changes is a reduction in seizure frequency. The observed changes in EEG patterns during training are not restricted to the formal biofeedback sessions; they are observed in the sleep records of trained patients as well (Sterman, 1984). On the average, seizure frequency decreases by 60% (Lubar, 1984; Sterman, 1984), an impressive result given the fact that the patients who are referred for biofeedback are typically only those with poorly controlled seizures. Furthermore, both single- and double-blind investigations have demonstrated the effectiveness of this procedure (Sterman & McDonald, 1978; Lubar et al., 1981). Therefore, the beneficial effects that result are not simply the product of placebo or nonspecific treatment effects. Since late 1980, one of us (T.L.B.) has treated approximately 30 patients with SMR biofeedback training (see Bennett, 1987), including children as young as 8. Results comparable to those reported by Lubar and Sterman have been obtained. It is important to note that we do not view biofeedback training as a "cure" for epilepsy (see also Lubar & Deering, 1981). The procedure does produce positive effects, but the individual is still at risk for seizures even though seizures have been reasonably well-controlled or entirely suppressed for a long period of time. The "epileptogenic focus or seizure generator process" has not been removed. For this reason, those who use these procedures should not encourage their patients to discontinue all antiepilepsy medications. On the other hand, it may be possible with better seizure control following biofeedback training to lower medication levels while remaining in the therapeutic range or to reduce the number of drugs taken for those individuals on polydrug therapy. These procedures would decrease the potential of negative cognitive and emotional side effects of the drugs, and it is certainly true that biofeedback has none of the common side effects that antiepilepsy drugs have.
Psychotherapy Although biofeedback appears to produce significant improvement for individuals with seizure
disorders, as it does for a variety of psychophysiological disorders, the greatest benefits of this procedure accrue when it is used as part of a total treatment program. For example, for children whose seizures are exacerbated by stress, stress management training can be very beneficial. Many patients report that the connection between stress and their seizures is obvious. Many sources of stress are noted by patients, and one approach is to help the child identify significant stressors in his or her life and the reactions they precipitate. Some of these are obvious, such as stresses related to the home, family, friends, and school; others may be more difficult to express, such as fears of injury or death from seizures. Once the child learns to identify the stressors and reactions to them, behavioral intervention into the stress-seizure chain is possible, and the child can master more adaptive and consequently less catastrophic responses to the stressors. Cognitive psychotherapeutic approaches can be helpful in reconditioning or substituting adaptive responses ~nd thought patterns for maladaptive ones. Part of stress management training can include conditioned relaxation and other techniques used to help the child maintain an overall lower level of reactivity. In the first author's epilepsy treatment program, a series of tape recordings are used for home practice including progressive muscle relaxation, autogenics, and visualization. A visualization of epilepsy tape is also sometimes used in which the child or adult is guided in visualizing his or her seizure focus and controlling its abnormal pattern of electrical activity. This latter exercise generally is positively reviewed by the patients, partly because of its educational value. Education is a critical component for anyone with epilepsy. As Mittan (1986) pointed out, "people are not born knowing how to cope with epilepsy, and understanding does not come with the first seizure" (p. 116). As a result, the child needs to learn about epilepsy; common fears that he or she might have must be brought out and frankly discussed. The child must realize that there is nothing wrong in fearing seizures or their consequences. By learning about them, the strength of the fears can be diminished. Family members are typically no more knowledgeable than is the child with seizures, and they must be educated about and allowed to frankly discuss epilepsy as well. The child's friends or classmates may also want to discuss epilepsy and how to deal with seizures. The first author once talked to 80 fifth graders about a classmate with complex partial
THE NEUROPSYCHOLOGY OF EPILEPSY
seizures. It alleviated their fears (especially about the risks associated with seizures), and it helped them treat their classmate more as "one of the gang." Both children and adults with complex partial seizures have wondered whether the symptoms of their epilepsy are a sign of psychiatric disorder. For this reason, the child needs to be given a simple explanation about neural mechanisms of emotional behavior and how a seizure involving the limbic system can generate emotional feelings such as anger or fear in the absence of emotion-evoking environmental events. Thus, the child needs to separate appropriate or typical feelings from those that are simply the product of a seizure discharge. The approaches discussed thus far all involve individual psychotherapy. The first author has had success in using group psychotherapy with adults suffering from epilepsy. It should work equally well with children. Sharing with individuals experiencing similar seizure events, fears, and stresses can provide consensual validation of what might otherwise seem to be bizarre experiences, inappropriate emotions, and feelings of being stigmatized. It can help individuals develop practical, effective strategies for dealing with their epilepsy, and can strengthen their self-esteem and renew sagging feelings of selfworth.
Conclusions It is hoped that the reader has learned several things from this chapter. First, elilepsy is a complex phenomenon that includes seizures themselves, changes in behavior and emotional responses, and alterations in cognitive and intellectual processes. Second, these components are the product of complicated interactions among neurophysiological, medication, and psychosocial variables. These phenomena were discussed in this chapter. · Third, the neuropsychologist can perform an important role in both the diagnosis and the treatment of this disorder in children. A ~ontribution to diagnosis can be made through clinical interview and neuropsychological assessment. Neuropsychological assessment can also serve a function in planning educational and cognitive treatments to compensate for cognitive deficits. The neuropsychologist can also perform a role in treatment through application of behavioral approaches-education, behavior management, biofeedback, stress management, and psychotherapy-to the control of seizure disorders.
437
Because of the complexity of this disorder, many children do not achieve adequate seizure control, and these behavioral approaches offer a promising alternative to traditional therapy with anticonvulsant medications.
References American Academy of Pediatrics. (1985). Behavioral and cognitive effects of anticonvulsant therapy. Pediatrics, 76(4), 644-647. Andersen, P., & Andersson, S. A. (1968). Physiological basis of the alpha rhythm. New York: Appleton-Century-Crofts. Andrewes, D. G., Tomlinson, L., Elwes, R. D., & Reynolds, E. H. ( 1984). The influence of carbamazepine and phenytoin on memory and other aspects of cognitive function in new referrals with epilepsy. Acta Neurologica Scandi1UJvica, Suppl. 99, 23-30. Angers, W. P., & Dennerll, R. D. (1962). Intelligence quotients of institutionalized and non-institutionalized epileptics. Journsl of Psychological Studies, 13, 152. Arieff, A. J., & Yacorzynski, G. K. (1942). Deterioration of patients with organic epilepsy. Jour1UJl of Nervous and Mental Disease, 96, 49-55. Arntson, P., Droge, D., Norton, R., & Murray, E. (1986). The perceived psychosocial consequences of having epilepsy. In S. Whitman & B. P. Herman (Eds.), Psychopathology in epilepsy: Social dimensions (pp. 143-161). New York: Oxford University Press. Bagley, C. (1971). The social psychology of the child with epilepsy. London: Routledge & Kegan Paul. Barnes, S. E., & Bower, B. D. (1975). Sodium valproate in the treatment of intractable childhood epilepsy. Developmental Medicine and Child Neurology, 17, 175-181. Bear, D. M. (1979). Temporal lobe epilepsy: A syndrome of sensory-limbic hyperconnection. Cortex, 15, 357-384. Bear, D., & Fedio, P. (1977). Quantitative analysis of interictal behavior in temporal lobe epilepsy. Archives of Neurology, 34, 454-467. Bennett, T. L. (1982). Introduction to physiological psychology (pp. 243-269). Monterey, CA: Brooks/Cole. Bennett, T. L. (1987). Neuropsychological aspects of complex partial seizures: Diagnostic and treatment issues. /nter1UJtio1UJI Journal of Clinical Neuropsychology, 9, 37-45. Bennett-Levy, J., & Stores, G. (1984). The nature of cognitive dysfunction in school children with epilepsy. Acta Neurologica Scandinavica, Suppl. 99, 79-82. Blumer, D. ( 1984). Psychiatric aspects of epilepsy. Washington, DC: American Psychiatric Press. Bourgeois, B. F., Prensky, A. L., Palkes, H., Talent, B. K., & Busch, S. G. (1983). Intelligence in epilepsy: A prospective study in children. An1UJls of Neurology, 14(4), 438-444. Bracy, 0. D. (1983). Computer based cognitive rehabilitation. Cognitive Rehabilitation, 1, 7-8. Bridge, E. M. (1949). Epilepsy and convulsive disorders in children. New York: McGraw-Hill.
438
CHAPTER 23
Brown, S. W., & Reynolds, E. H. (1981). Cognitive impairment in epileptic patients. In E. H. Reynolds & M. R. Trimble (Eds.), Epilepsy and psychiatry. Edinburgh: Churchill Livingstone. Browne, T. R., Dreifuss, F. E., Dyken, P. R., Goode, D. J., Penry, J. K., Porter, R. J., White, B. J., & White, P. T. ( 1975). Ethosuccimide in the treatment of absence (petit mal) seizures. Neurology, 25, 515-525. Camfield, P.R., Gates, R., Rosen, G., Camfield, C., Ferguson, A., & MacDonald, G. W. (1984). Comparison of cognitive ability, personality profile, and school success in epileptic children with pure right versus left temporal lobe EEG foci. Annals of Neurology, 15(2), 122-126. Carlberg, C., & Kavale, K. (1980). The efficacy of special versus regular class placement for exceptional children: A metaanalysis. Journal of Special Education, 14, 295-309. Collins, A. L. (1941). Psychometric records of institutionalized epileptics, Journal of Psychology, 11, 359-370. Corbett, J. A., & Trimble, M. (1967). Neuropsychiatric aspects of anticonvulsant treatment in children with epilepsy. Paper presented at the IVth World Congress of Psychiatry, Honolulu. Corbett, J. A., & Trimble, M. R. (1983). Epilepsy and anticonvulsant medication. In M. Rutter (Ed.), Developmental neuropsychiatry (pp. 112-129). New York: Guilford Press. Corbett, J. A., Trimble, M. R., & Nichol, T. C. (1985). Behavioral and cognitn
phenytoin. Neurology, 27, 1023-1028. Drug Information. (1985). Anticonvulsants. In G. K. McEvoy
(Ed.), American Hospital Formulary Service. American Society of Hospital Pharmacists, 28(12), 811-840. Ellenberg, J. H., Hirtz, D. G., & Nelson, K. B. (1986). Do seizures in children cause intellectual deterioration? The New England Journal of Medicine, 314(11), 1085-1088. Epir, S., Renda, Y., & Baser, N. (1984). Cognitive and behavioral characteristics of children with idiopathic epilepsy in a low-income area of Ankara, Turkey. Developmental and Medical Child Neurology, 26, 200-207. Farwell, J. R., Dodrill, C. B., & Batzel, L. W. (1985). Neuropsychological abilities of children with epilepsy. Epilepsia, 26(5), 395-400. Fedio, P., & Mirsky, A. F. (1969). Selective intellectual deficits in children with temporal lobe or centrencephalic epilepsy. Neuropsychologia, 7, 287-300. Fischbacher, E. (1982). Effect of reduction of anticonvulsants on well being. British Medical Journal, 285, 425-427. Fletcher, J. M., & Taylor, H. G. (1984). Neuropsychological approaches to children: Towards a developmental neuropsychology. Journal of Clinical Neuropsychology, 6, 39. Folsom, A. (1953). Psychological testing in epilepsy. I. Cognitive function, Epilepsia, 2. 15-22. Fox, J. T. (1924). Response of epileptic children to mental and educational tests. British Journal ofMedical Psychology, 4. 235-248. Giordani, B., Berent, S., Sackellares, J. C., Rourke, D., Seidenberg, M., O'Leary, D. S., Dreifuss, F. E., & Boll, T. J. (1985). Intelligence test performance of patients with partial and generalized seizures. Epilepsia, 26(1), 37-42. Glass, D. H., & Mattson, R. J. (1973). Psychopathology and emotional precipitation of seizures in temporal lobe and nontemporal lobe epileptics. Proceedings of the 81 st Annual Convention of the American Psychological Association, 8, 425-426. Glowinski, H. (1973). Cognitive deficits in temporal lobe epilepsy. Journal ofNervous and Mental Disease, 157. 129137. Harris, A. B. (1972). Degeneration in experimental epileptic foci. Archives of Neurology, 26, 434-449. Green, J. B., & Hartlage, L. C. ( 1971). Comparative performance of epileptic and non-epileptic children and adolescents (on tests of academic, communicative, and social skills). Diseases of the Nervous System, 32(6), 418-421. Grinker, R. R. (1929). The proper use of phenobarbital in the treatment of the epilepsies. Journal of the American Medical Association, 93, 1218-1219. Guey, J., Charles, C., Coquery, C., Roger, J., & Soulayrol, R. (1967). Study of psychological effects of ethosuximide (Zarontin) on 25 children suffering from petit mal epilepsy. Epilepsia, 8, 129-141. Gunn, J. ( 1977). Epileptics in prison. New York: Academic Press. Halstead, H. (1957). Abilities and behavior of epileptic children. Journal of Mental Science, 103, 28-47. Harrison, R. M., & Taylor, D. C. (1976). Childhood seizures: A 25-year follow-up. Lancet, I. 948-951. Hermann, B. P. ( 1982). Neuropsychological functioning and psychopathology in children with epilepsy. Epilepsia. 23. 545554. Hermann, B. P., Black, R. B., & Chhabria, S. (1981 ). Behavioral
THE NEUROPSYCHOLOGY OF EPILEPSY problems and social competence in children with epilepsy. Epilepsia, 22, 703-710. Hermann, B. P., & Whitman, S. (1986). Psychopathology in epilepsy: A multi-etiologic model. InS. Whitman & B. P. Hermann (Eds.), Psychopathology in epilepsy: Socwl dimensions (pp. 5-37). New York: Oxford University Press. Hill, D. (1981). Historical review. In E. H. Reynolds & M. R. Trimble, Epilepsy and psychiatry (pp. 1-11). Edinburgh: Churchill Livingstone. Holdsworth, L., & Whitmore, K. ( 1974). A study of children with epilepsy attending ordinary schools. I. Their seizure patterns, progress, and behavior in school. Developmental Medicine and Child Neurology, 16, 746-758. Hutt, S. J., Jackson, P. M., Belsham, A. B., & Higgins, G. (1968). Perceptual-motor behavior in relation to blood phenobarbitone level. Developmental Medicine and Child Neurology, 10, 626-632. ldestrom, C. M., Schalling, D., Carlqvist, V., & Sjoqvist, F. (1972). Behavioral and psychophysiological studies: Acute effects of diphenylhydantoin in relation to plasma levels. Psychological Medicine, 2, 111-120. Keating, L. E. (1961 ). Epilepsy behavior disorders in school children. Journal of Mental Science, 107, 161-180. Keith, H. M., Ewert, J. C., Green, M. W., & Gage, R. P. (1955). Mental status of children with convulsive disorders. Neurology, 5, 419-425. Kimura, D. (1964). Cognitive deficit related to seizure patterns in centrencephalic epilepsy. Journal of Neurology, Neurosurgery, and Psychiatry, 27, 291-295. Klove, H., & Matthews, C. G. (1966). Psychometric and adaptive abilities in epilepsy with differential etiology. Epilepsia, 7, 330-338. Klove, H., & Matthews, C. G. (1969). The neuropsychological! evaluation of the epileptic patient. Wisconsin Medical Association Journal, 68, 296-301. Kluver, H., & Bucy, P. C. (1939). Preliminary analysis of the functions of the temporal lobe in monkeys. Archives ofNeurology and Psychiatry, 42, 919-1000. Ladavas, E., Umilta, C., & Provinciali, L. (1979). Hemisphere dependent cognitive performances in epileptic patients. Epi/epsia, 20, 493-502. Lansdell, H., &Mirsky, A. F. (1964). Attentioninfocalandcentren-
cephalic epilepsy. Experimental Neurology, 9, 463-469. Lechtenberg, R. (1984). Epilepsy Qlld the family. Cambridge, MA: Harvard University Press. Lefcourt, H. M. (1976). Locus ofcontrol: Current trends in theory and research. Hillsdale, NJ: Erlbaum. Lennox, W. G. (1942). Brain injury, drugs, and environment as causes of mental decay in epileptics. American Journal of Psychiatry. 99, 174-180. Lennox, W. G., & Lennox, M.A. (1960). Epilepsy and reklted disorders. Boston: Little, Brown. Loiseau, P., Signoret, J. L., & Strube, E. ( 1984). Attention problems in adult epileptic patients. Acta Neurologica Scandinavica, Suppl. 99, 31-34. Loiseau, P., Strube, E., Broustet, D., Battellochi. S., Gomeni, C., & Morselli, P. L. (1980). Evaluationofmemoryfunction in a population of epileptic patients and matched controls. Acta Neurologica Sandinavica. Suppl. 80. 58-61.
439
Lubar, J. F. (1984). Applications of operant conditioning of the EEG for the management of epileptic seizures. InT. Elbert, B. Rockstroh, W. Lutzenberger, & N. Birbaumer (Eds.), Self-regulation of the brain (pp. 107-125). Berlin: SpringerVerlag. Lubar, J. F., & Deering, W. M. (1981). Behavioral approaches to neurology. New York: Academic Press. Lubar, J. F., Shabsin, H. S., Natelson, S. E., Holder, G. S., Whitsett, S. F., Pamplin, W. E., & Krolikowski, D. E. (1981). EEG operant conditioning in intractible epileptics. Archives of Neurology, 38, 700-704. MacLeod, C. M., Dekaban, A., & Hunt, E. (1978). Memory impairment in epileptic patients: Selective effects of phenobarbital concentration. Science, 202, 1102-1104 .. Marlowe, W. 8., Mancall, E. L., & Thomas, J. J. (1975). Complete Kluver-Bucy syndrome in man. Cortex, II, 53-59. Masland, R. L. (1976). Epidemiology and basic statistics on the epilepsies: Where are we? Paper presented at the Fifth National COIIIference on the Epilepsies, Washington, DC. Masui, K., Niwa, S., Anzai, N., Kameyama, T., Saitob, 0., & Rymar, K. (1984). Verbal memory disturbances in left temporal lobe epileptics. Cortex, 20, 361-368. Matthews, C. G., & Klove, H. (1967). Differential psychological performances in major motor, psychomotor, and mixed seizure classifications of known and unknown etiology. Epilepsia, 8, 117-128. Matthews, W. S., & Barabas, G. (1986). Perception of control among children with epilepsy. InS. Whitman & B. P. Hermann (Eds.), Psychopathology in epilepsy: Social dimensions (pp. 162-182). New York: Oxford University Press. Matthews, W. S., Barabas, G., & Ferrari, M. (1982). Emotional concomitants of childhood epilepsy. Epilepsio, 23, 671-681. Mayeux, R., Brandt, J., Rosen, J., & Benson, D. F. (1980). Interictal memory and language impairment in temporal lobe epilepsy. Neurology, 30, 120-125. Mellor, D. ( 1977). A study of epilepsy with particuklr reference to behaviour disorder. Unpublished M.D. thesis, University of Leeds. Mignone, R. J., Donnelly, E. F., & Sadowsky, D. (1970). Psychological and neurological comparisons of psychomotor and non-psychomotor epileptic patients. Epilepsia, I I, 345-359. Milner, B. (1958). Psychological dificits produced by temporal lobe excision. Research Publications, Association for Research in Nervous and Mental Disease, 36, 244-257. Mirsky, A. F., Primae, D. W., Ajmone-Marsan, C., Rosvold, H. E., & Stevens, J. R. (1960). A comparison of the psychological test performance of patients with focal and non-focal epilepsy. Experimental Neurology, 2, 75-89. Mirsky, A. F., & Van Buren, J. M. (1965). On the nature of the absence in centrencephalic epilepsy: A study of some behavioral, electroencephalographic and autonomic factors. Electroencephalography and Clinical Neurophysiology, 18, 334338. Mittan, R. J. (1986). Fear of seizures. In S. Whitman & B. P. Hermann (Eds.), Psychopathology in epilepsy: Socwl dimensions (pp. 90-121). New York: Oxford University Press. Mungas, D. (1982). Interictal behavior abnormality in temporal lobe epilepsy: A specific syndrome or nonspecific psychopathology? Archives of General Psychiatry, 39, 108-111.
440
CHAPTER 23
Mungas, D., Ehlers, C., Walton, N., & McCutchen, C. B. (1985). Verbal learning differences in epileptic patients with left and right temporal lobe foci. Epilepsia, 26, 340-345. Novelly, R. A., Augustine, E. A., Mattson, R. H., Glaser, G. H., Williamson, P. D., Spencer, D. D., & Spencer, S. S. (1984). Selective memory improvement and impairment in temporal lobectomy for epilepsy. Annals of Neurology, 15, 64-67. O'Leary, D. S., Lovell, M. R., Sackellares, J. C., Berent, S., Giordani, B., Seidenberg, M., & Boll, T. J. (1983). Effects of age of onset of partial and generalized seizures on neuropsychological performance in children. Journal of Nervous and Mental Disease, 171(10), 624-629. Ounsted, C. (1969). Aggression and epilepsy: Rage in children with temporal lobe epilepsy. Journal of Psychosomatic Research, 13, 237-242. Patterson, H., & Fonner, A. (1928). Some observations on the intelligence quotient in epileptics. Psychiatric Quarterly. 2, 542-548. Pazzaglia, P., & Frank-Pazzaglia, L. (1976). Record in grade school of pupils with epilepsy: An epidemiological study. Epilepsia, 17, 361-366. Pond, D. A., & Bidwell, B. H. (1959). A survey of epilepsy in fourteen general practices. II. Social and psychological aspects. Epilepsia, 1, 285-295. Post, R. M. (1983). Behavioral effects of kindling. In M. Parsonage (Ed.), Advances in epileptology: XIVth International Symposium (pp. 173-180). New York: Raven Press. Pritchard, P. B., Lombroso, C. T., & Mcintyre, M. (1980). Psychological complications of temporal lobe epilepsy. Neurology, 30, 227-232. Quadfasel, A. F., & Pruyser, P. W. (1955). Cognitive deficits in patients with psychomotor epilepsy. Epilepsia, 4, 80-90. Reed, H. B. (1951). The intelligence of epileptics. Journal of Genetic Psychology, 78, 145-152. Reitan, R. M. ( 1974). Psychological testing of epileptic patients. In 0. Magnus & L. de Haas (Eds.), The epilepsies (pp. 559575). Amsterdam: Elsevier. Reynolds, E. H. (1970). Neurological aspects of folate and B 12 metabolism. Clinics in Haematology, 5, 661-694. Reynolds, E. H. (1983). Mental effects of antiepileptic medication: A review. Epilepsia, 24(Suppl. 2), 85-%. Reynolds, E. H., & Trimble, M. R. (1985). Adverse neuropsychiatric effects of anticonvulsant drugs. Drugs, 29, 570581. Rodin, E. A. (1968). The prognosis of patients with epilepsy. Springfield, IL: Thomas. Rodin, E. A., Katz, M., & Lennox, K. (1976). Differences between patients with temporal lobe seizures and those with other forms of epileptic attacks. Epilepsia, 17, 313-320. Roger, J., Grangeon, H., Guey, J., & Lob, H. (1968). Psychiatric and psychological effects of ethosuccimide treatment in epileptics. Encephale, 57, 407-438. Ross, E. M., Peckham, C. S., West, P. B., & Butler, N. R. (1980). Epilepsy in childhood: Findings from the national child development study. British Medical Journal, 280, 207210. Rourke, B. P., Bakker. D. J., Fisk, J. L.. & Strang, J.D. (1983). Child neuropsychology: An introduction to theory, research, and clinical practice. New York: Guilford Press.
Rutter, M., Graham, P., & Yule, W. (1970). A neuropsychiatric study in childhood. Philadelphia: Lippincott. Scarpa, P., & Carassini, B. (1982). Partial epilepsy in childhood: Clinical and EEG study of 261 cases. Epilepsia, 23, 333341. Schain, R. J., Ward, J. W., &Guthrie, D. (1977). Carbamazepine as an anticonvulsant in children. Neurology, 27, 476-480. Scott, D., Moffett, A., Matthews, A., & Ettlinger, G. (1967). Effects of epileptic discharges on learning and memory in patients. Epilepsia, 8, 188-194. Serafetinides, E. A. ( 1970). Psychiatric aspects of temporal lobe epilepsy. In E. Niedemeyer (Ed.), Epilepsy: Recent views on theory, diagnosis and therapy of epilepsy (p. 155). Basel: Karger. Shabsin, H. S., Bahler, W. W., & Lubar, J. F. (1979). A comparison of 12-15Hz Rolandic activity during eyes-opened and eyes-closed conditions and its occurrence with occipital alpha. Paper presented at the annual meeting of the Southwestern Psychological Association. Sherwin, 1., Peron-Magnan, P., Bancaud, J., Bonis, A., & Talairach, J. (1982). Prevalence of psychosis in epilepsy as a function of the laterality of the epileptogenic lesion. Archives of Neurology, 39, 621-625. Shorvon, S. D., & Reynolds, E. H. (1979). Reduction of polypharmacy for epilepsy. British Medical Journal, 2. 10231025. Somerfeld-Ziskind, E., & Ziskind, E. ( 1940). Effect of phenobarbital on the mentality of epileptic patients. Archives of Neurology and Psychiatry, 43, 70-79. Sterman, M. B. (1984). The role of sensorimotor rhythmic EEG activity in the etiology and treatment of generalized motor seizures. InT. Elbert, B. Rockstroh, W., Lutzenberger, & N. Birbaumer (Eds. ), Self-regulation of the brain (pp. 95-106 ). Berlin: Springer-Verlag. Sterman, M. B., & Friar, L. (1972). Suppression of seizures in an epileptic following sensorimotor EEG biofeedback training. Electroencephalography and Clinical Neurophysia/ogy, 33, 89-95. Sterman, M. B., & MacDonald, L. R. (1978). Effects of central cortical EEG feedback training on incidence of poorly controlled seizures. Epilepsia, 19, 207-222. Sterman, M. B., & Wywricka, W. (1967). EEG correlates of sleep: Evidence for separate forebrain substrates. Brain Research, 6, 143-163. Sterman, M. B., Wywricka, W., & Ruth, S. R. (1969). Electrophysiological correlates and neural substrates of alimentary behavior in the cat. Annals of the New York Academy of Sciences, 157, 723-739. Stevens, J. R. (1966). Psychiatric implications of psychomotor epilepsy. Archives ofGeneral Psychiatry, 14, 461-471. Stevens, J. R. (1982). Does epilepsy cause psychopathology and personality disorders? In H. Akimoto, H. Kazamatsuri, M. Seino, & A. Ward (Eds.), Advances in epileptology: Xlllth Epilepsy International Symposium (pp. 199-204). New York: Raven Press. Stores, G. (1971). Cognitive function in children with epilepsy. Developmental Medicine and Child Neurology, 13(3), 390393. Stores, G. (1978). School-children with epilepsy, at risk for learn-
THE NEUROPSYCHOLOGY OF EPILEPSY
441
ing and behavior problems. Developmental and Medical sant drugs, and seizures. Acta Neurologica Scandinavica Child Neurology. 20(4), 502-508. Suppl .. 89, 31-41. Stores, G. (1980). Children with epilepsy: Psychosocial aspects. Trimble, M. R., & Thompson, P. J. (1984). Sodium valproateand In B. P. Hermann (Ed.). A multidisciplinary handbook of cognitive function. Epilepsia. 25(Suppl. 1}, 560-564. epilepsy (pp. 224-242). Springfield, IL: Thomas. Trimble, M. R., & Thompson, P. J. (1986). Neuropsychological Sullivan, E. B., & Gahagan, L. ( 1935). On intelligence of epilepaspects of epilepsy. In I. Grant & K. M. Adams (Eds.), tic children. Genetic Psychology Monographs. /7(5), 309Neuropsychological assessment of neuropsychiatric disor375. ders (pp. 321-346). New York: Oxford University Press. Tarter, R. E. (1972). Intellectual and adaptive functioning in epi- West, P. (1986). The social meaning of epilepsy: Stigma as a lepsy: A review of 50 years of research. Diseases of the potential explanation for psychopathology in children. InS. Nervous System, 33(12), 763-770. Whitman & B. P. Hermann (Eds.), Psychopathology in epiThompson, P. J., Huppert, F. A., & Trimble, M. R. (1981). lepsy: Social dimensions (pp. 245-265). New York: Oxford Phenytoin and cognitive function: Effects on normal volunUniversity Press. teers and implications for epilepsy. British Journal of Whitman, S., & Hermann, B. P. (Eds.). (1986). Psychopathology Clinical Psychology. 20. 155-162. in epilepsy: Social dimensions. New York: Oxford University Thompson, P. J., & Trimble, M. R. (1982). Anticonvulsant drugs Press. and cognitive functions. Epi/epsia, 23, 531-544. Wilkus, R. J., & Dodrill, C. B. (1976). Neuropsychological correThompson, P. J., & Trimble, M. R. (1983). The effect of anticonlates of the EEG in epileptics. I. Topographic distribution and vulsant drugs on cognitive function: Relation to serum levels. average rate of epileptiform activity. Epilepsia, 17, 89Journal of Neurology. Neurosurgery, and Psychiatry, 46, 100. 227-233. Wyler, A. F. (1984). Operant conditioning of single neurons in Trimble, M. (1981). Anticonvulsant drugs, behavior, and cogmonkeys and its theoretical application to EEG operant connitive abilities. Current Developments in Psychopharditioning in human epilepsy. InT. Elbert, B. Rockstroh, W. macology, 6, 65-91. Lutzenberger, & N. Birbaumer (Eds. ), Self-regulation of the Trimble, M. R., Corbett, J., & Donaldson, J. (1980). Folic acid brain (pp. 85-94). Berlin: Springer-Verlag. and mental symptoms in children with epilepsy. Journal of Zimmerman, F. T., Burgemeister, B. B., & Putnam, T. J. (1951). Neurology, Neurosurgery, and Psychiatry, 43, 1030-1034. Intellectual and emotional makeup of the epileptic. Archives Trimble, M. R., & Thompson, P. J. (1981). Memory, anticonvulof Neurology and Psychiatry, 65, 545-556.
24 Neuropsychological Effects of Stimulant Medication on Children's Learning and Behavior RONALD T. BROWN AND KATHI A. BORDEN
Historical Overview Bradley's (1937) seminal publication of the ameliorative effects of Benzedrine on behavior-disordered children had significant impact on pediatric psychopharmacology and to a large extent altered child psychiatric and neurological services in this country. According to Bradley and his colleagues, the main clinical effects of stimulant drugs were reduced hyperactivity, distractibility, and impulsivity in children who had not responded to traditional psychotherapy. Ironically, Bradley's earlier clinical observations, which were based on insufficient investigation and inadequate methodology, are generally consistent with contemporary research findings, which generally support the efficacy of stimulants in improving performance on a wide array of cognitive tasks and decreasing disruptive behaviors (Gittelman, 1983). Building upon Bradley's work and attempting to diagnostically classify behavior-disordered children into more homogeneous subgroups, Laufer and his associates (Laufer, Denhoff, & Solomons, 1957) determined that amphetamines acted most positively in the behaviorally disordered children who exhibited the ''hyperkinetic impulse'' disorder. Methylphenidate was introduced in the 1950s to avoid the untoward side effects and potential abuses of amphetamines. In 1956 the National Institute of Mental Health
RONALD T. BROWN • Division of Child and Adolescent Psychiatry, Emory University School of Medicine, Atlanta, Georgia 30322. KATHI A. BORDEN • Department of Psychology, Pepperdine University, Los Angeles, California 90034.
(NIMH) established the Psychopharmacology Branch Service Center to serve as a primary source of federal support for psychopharmacological research, and greater attention began to be paid to the disparity between empirical validation and the clinical use of stimulants. The first NIMH grant in pediatric psychopharmacology was awarded to Leon Eisenberg, MD, of Johns Hopkins University, who investigated the effects of stimulants on children's behavioral and cognitive functioning. In collaboration with Conners (Conners & Eisenberg, 1963), he replicated the clinical findings presented by Bradley and his group (Bradley, 1937), thus empirically establishing the short-term efficacy of stimulant medication through well-controlled clinical trials. Despite the widespread clinical use of stimulant medication in the 1950s, it was not until the 1960s that extensive efforts were made to study systematically the effects of stimulants in children. From those efforts grew a reliable and standardized rating system by which to assess stimulant medications (Conners, 1969). Also during this time, research on the effects of stimulant drugs became more methodologically rigorous, and much greater emphasis was placed on assessing the target response in carefully controlled clinical trials. More important, careful attention was paid to the systematic documentation of side effects, which had been clinically described as early as the 1930s (Bradley, 1937) but had never been carefully documented. Because most research on stimulant use with children (including our own work) has been focused on attention deficit disorders (ADD), this chapter is largely devoted to the ADD population. Some have considered the 1970s a "professional and lay backlash" (Dulcan, 1986) to the en443
444
CHAPTER 24
thusiastic use of stimulants; teachers and physicians were accused of "drugging children, especially poor ones, into submission" (Dulcan, 1985, p. 383). As a result, research efforts in pediatric psychopharmacology increased further, and even more methodologically rigorous studies proliferated. Some of the more fruitful studies provided valuable research data on the characteristics of children who would particularly benefit from the use of stimulants (Satterfield, Cantwell, & Satterfield, 1974). In the virtually unexplored areas of pediatric psychopharmacology, studies were concentrated on the dosages of stimulant medication (Brown & Sleator, 1979; Sprague & Sleator, I 977) (pediatric dosages had simply been extrapolated from adult dosage levels). Further, the side effects of stimulants in children were studied much more systematically, for example, the possibility of growth suppression (Safer, Allen, & Barr, 1972), the potential for deleterious drug interactions (Fisher & Wilson. 1971), and the longterm effects of stimulants (Weiss, 1983). Well-controlled clinical trials were conducted with pemoline (Cylert), which was introduced for pediatric use during this period (Conners, 1972). The late 1970s and the early 1980s have been a very exciting tertiary stage in the refinement of drug trials and systematic study of stimulants for children's learning and behavioral problems. Recent publications (Gittelman, 1983; Rapoport, 1983) have typically underscored the generally accepted notion that stimulants are the drugs of choice for decreasing restlessness and impulsive behaviors and for improving attention span in hyperactive and behavior-disordered children. Research is under way to more carefully systematize target response to cognitive, social, and academic variables (Pelham, Bender, Caddell, Booth, & Moorer, 1985), and well-designed research (Douglas, Barr, O'Neill, & Britton, 1986; Pelham et al., 1985; Stephens, Pelham, & Skinner, 1984) has begun to address the general ineffectiveness of stimulants on academic achievement. In addition, the effects of stimulants on growth have been studied more systematically. Recent research efforts in the pharmacokinetics (absorption, metabolism, distribution, and excretion) of stimulants are providing greater knowledge of the physiological mechanisms that will, we hope, translate into clinical application (Rapoport, 1983). Although a prodigious amount of empirical research has generally attested to the superiority of stimulants over other nonsomatic therapies (Gittelman, 1983; Rapoport, 1983), the combination of stimulant medications and other psychotherapeutic approaches is also being studied (Gadow, 1985b). Although new areas await investigation, consid-
erable progress has been made in the past 30 years to examine more carefully the effects of stimulant medication on children's learning and behavior. This progress has been largely due to the interdisciplinary efforts of the neuropsychologist, the pediatric neurologist, the child psychiatrist, the pediatrician, and the pharmacologist. Distinct but complementary knowledge from various disciplines has resulted in a more complete understanding of the actions and mechanisms of stimulant medication in the research laboratory and in the clinical setting.
Frequently Used Stimulant Medications Psychostimulants may be defined as the class of drugs that produces excitation of the central nervous system. The most commonly prescribed are the dextroamphetamines, methylphenidate, and pemoline, each of which is widely used in treating behavior and learning problems, which are typically associated with externalizing disorders in children.• It has been estimated that I to 2% of all elementary school children are prescribed some type of stimulant medication (Bosco & Robin, 1980). Although rates of diagnosis and stimulant drug treatment have been found to vary widely among school districts, stimulant drug treatment has not been found to correlate to socioeconomic status (Bosco & Robin, 1980), thus contradicting the unsupported claim in the 1970s that these agents were being used to "drug" poor children into submission (Dulcan, 1985). According to recent research, the rate of prescribing stimulant medication has increased significantly in the past several years, particularly for older children and adolescents (Safer & Krager, 1985), so the need for empirical knowledge of the long- and short-term effects · of psychostimulants has become even more pressing. Agreement has been fairly widespread that the main indications for stimulant drug therapy are hyperactivity 2 and short attention span (Rapoport, 1Deanol
is a stimulant that has yielded some positive clinical findings (Donnelly & Rapoport, 1985); however, it will not be covered here, as it is not in general use and its pharmacology is not well described. Caffeine is another stimulant that will not be discussed here, as no significant improvement has been found as a function of therapeutic caffeine use (Donnelly & Rapoport, 1985). 2Jn accordance with the Diagnostic and Statistical Manunl ofMental Disorders (American Psychiatric Association, 1980), children clinically diagnosed as hyperactive are referred to as having attention deficit disorder with hyperactivity (ADD/H). Because many
NEUROPSYCHOLOGICAL EFFECTS OF STIMULANT MEDICATION
445
1983). In fact, Barkley (1977), in reviewing 110 studies of more than 4200 ADD/H children, found that 75% could be judged as improved after a brief course of stimulant drugs. Although much research has been conducted on the behavioral and cognitive effects of the stimulants, we have little data on the cause, nature, and effects of any biochemical abnormalities that may exist in the groups that have been treated effectively with stimulants (Dulcan, 1986). Moreover, we know little about the pharmacodynamics (mechanisms of action and interactions with biological receptors) of psychostimulants. The most commonly used stimulants-dextroamphetamine (Dexedrine) and methylphenidate (Ritalin), both of which are sympathomimetic amines, and pemoline (Cylert)-and their basic effects are presented in Table 1.
Despite considerable debate in the pediatric literature about the appropriate clinical dosage of dextroamphetamine, we have few empirical studies to guide clinical efforts. Wender (1971) and Solomons (1973) recommended a starting dose of 5 mg three times per day, yet they emphasized that many children can tolerate much higher doses (20 mg per day is considered a median dose). Others have advocated lower doses for children (Gross & Wilson, I974). According to some experts (Ross & Ross, 1982), one advantage of dextroamphetamine over methylphenidate is that the former, because it is more resistant to gastric juices, need not be administered on a special schedule related to meals (methylphenidate, which is destroyed by gastric activity, must be administered 30 minutes before meals).
Dextroamphetamine
Methylphenidate
The oldest of the psychostimulants, dextroamphetamine is believed to potentiate both dopamine and norepinephrine by stimulating the release of newly synthesized dopamine into the synaptic cleft, inhibiting presynaptic uptake and inhibiting monoamine oxidase (Dulcan, 1986). Following the oral ingestion of dextroamphetamine, peak plasma levels are believed to occur 3 to 4 hours after administration; the elimination half-life of the drug ranges from 6 to 7 hours in children and is even more rapid in adults (Dulcan, I986). The behavioral effects typically occur I to 4 hours after oral administration and often correspond to the absorption phase. Little relationship has been found between behavioral response ratings across individuals and plasma amphetamine levels (Rapoport, I983). Dextroamphetamine may be administered in tablet form or as a sustained-release (S-R) capsule, which acts for 6 to I8 hours. No significant differences have been found in elimination half-life or behavioral response resulting from the two dosage forms (Ross & Ross, I982). Although tablets offer the advantage of systematic dosage monitoring, the S-R form requires only one daily administration, thus eliminating the need for school personnel to administer medication. Some very recent research has suggested adverse effects if the child chews the capsule rather than swallows it (Dulcan, 1985).
Methylphenidate, which is structurally related to dextroamphetamine, is a piperidine derivative that acts by releasing stored dopamine from the reserpinesensitive presynaptic vasicular pool, decreasing dopamine reuptake, and inhibiting monoamine oxidase (Dulcan, I986). Similar to the dextroamphetamines, methylphenidate is poorly bound to plasma proteins, but it is rapidly metabolized to ritalinic acid. Peak serum concentration typically occurs I to 3 hours after oral administration, although peak plasma concentrations vary from person to person (perhaps because of individual idiosyncratic rates of absorption) (Chan et al., 1983). The elimination halflife of methylphenidate is rather short (2 to 4 hours), so accumulation is virtually nonexistent. Methylphenidate is given to approximately two thirds of all children who have ADD and who are given stimulants. It is typically administered in two divided doses, although some experts consider a single morning dose (20 mg) sufficient to improve school performance (Huey, 1985). Single doses of SR forms are also available. Whitehouse and his associates (Whitehouse, Shah, & Palmer, 1980) compared S-R methylphenidate and standard doses and found no differences in clinical and side effects. The dependent measures they used were quite global and thus did not permit sensitive comparison of the two dosage forms. Moreover, because the clinical effects have been found to dissipate 4 hours after administration (Ross & Ross, 1982), many practitioners consider twice daily doses necessary to sustain clinical effects. Research using a more sensitive measurement system of the efficacy of the S-R form needs to be undertaken.
of the studies reviewed in this chapter were published prior to 1980 and used previous diagnostic criteria, the tenn ADD/H here will signify any subjects referred to in research reports as either hyperactive, attention deficit disordered, or attention deficit disordered with hyperactivity.
Diagnosis
Methylphenidate (Ritalin)
Pemoline (Cylert)
2.5 mg 2 to 3 times/day I mg/kg per day 3 to 4 hours 6 to 7 hours I to 4 hours
I hour I hour
Onset of behavioral effect Duration of behavioral effect
I hour 3 to 4 hours
0.6-1.7 10-60 5 mg 2 to 3 times/day 0.3 mg to 2 mg/kg per day I to 3 hours 2 to 3 hours I to 3 hours
0.3-1.25
5-40
5, 10, 20, SR-20 0.3-0.7
5, 10 0.15-0.5
Pharmacology
0.5-3.0 18.75-37.5 18.75 mg each morning 75 mg/day in single daily dose I to 3 hours 4 to 15 hours If prescribed as indicated, several weeks after treatment begins, therapeutic effect is generally sustained over several hours Variable 6 to 8 hours
18.75; 37.5, 75.0 0.5-2.5
ADD/H ADDIH ADDIH Residual ADD/H Residual ADD/H Residual ADD/H ADD without hyperactivity; conduct disorder; behavioral impulsivity; inattentiveness; specific developmental (learning) disorders; emotional lability Physical and neurologic examination (height and weight, blood pressure, pulse, and dyskinetic movements), routine laboratory tests (CBC with differential, blood chemistry profile, urinalysis)
How supplied (mg) Single-dose range Daily dose range mg/kg per day mg/day Initial dosage Therapeutic dosage Peak plasma level Plasma half-life Peak clinical effect
Workup
Indications Definite Possible Conjectural
Dextroamphetamine (Dexedrine)
TABLE 1. Indications, Adverse Effects, and Medical Management of Stimulants in Children
~
1:1:1
~
~
Follow-up
Drug interactions
Relative contraindications Toxicity/overdose
Serious but unusual adverse reactions and precautions
Common adverse reaction Less frequent adverse reactions
Marked anxiety, agitation or psychosis, glaucoma, verbal-motor tic (Tourette's syndrome) Irritability, restlessness, agitation, nausea, diarrhea, High fever, sweating, high pallor, flushing, fever, sweating, pallor, flushing, arrhythmias, tachyarrhythmias, tachycardia, significant bycardia, significant hypertension, delirium, tremor, pertension, delirium, tremor, convulsions, convulsion, coma coma Increase blood level of tricyclic antidepressants, increase metabolism of phenytoin, oppose effect of antihypertensives, acetazolamide increases renal absorption of amphetamines Record height and weight every 3 to 6 weeks, follow Record height and weight every 3 to 6 blood pressure, pulse, and dyskinetic movements, perweeks, follow blood pressure, pulse, and form yearly physical examination and routine laboradyskinetic movements, perform yearly tory tests physical examination and routine laboratory tests (liver function tests are particularly important)
Special considerations
Difficulty in falling asleep, mild elevation of pulse and blood pressure Decreased appetite (temporary), crying and dysphoria, growth retardation (height and weight, mild), drowsiness, anxiety, and irritability Psychotic thoughts, lowered seizure threshold, worsenPsychotic thoughts, lowered seizure threshold, ing of tic disorder or dyskinesia, potential for medicaworsening of tic disorder or dyskinesia, potion abuse, hypertension tential for medication abuse, hypertension, hepatocellular injury, elevated serum glutamic pyruvic transaminase
Side effects
~
e
I ~ I I ~
;
448
CHAPTER 24
According to some clinical literature (Gilman, Goodman, & Gilman, 1980), methylphenidate is vulnerable to destruction by gastric acidity, thus requiring that the drug be administered 30 minutes before meals. Swanson and his colleagues (Chan eta/., 1983; Swanson, Sandman, Deutsch, & Baren, 1983) questioned this standard administration practice (Gilman et a/., 1980) and specifically investigated the effect of food on the absorption of an oral dose of methylphenidate. Taking the drugs 30 minutes before or with breakfast produced no significant differences in the children's cognitive, behavioral, or physiological performance. In fact, the plasma concentration peaked sooner when the drug was ingested after a meal rather than before, with no significant differences in plasma concentration. As is true of dextroamphetamine, little relationship has been found between plasma level and oral doses of methylphenidate. To complicate this matter even further, dose-response curves vary in accordance with the effects measured (Sprague & Sleator, 1977). Thus, the appropriate dosage of methylphenidate has been a source of debate throughout the pediatric literature (Gittelman, 1983). For example, to better understand the effects of stimulants on learning, several investigators have studied the relationship between dosage of methylphenidate and target response (Brown & Sleator, 1979; Brown, Slimmer, & Wynne, 1984; Sprague & Sleator, 1977). Some of that research has provided information on the methylphenidate dosage levels that seem to enhance the learning performance of ADD/H children. The data from all three studies indicate that learning performance in children is most improved when methylphenidate is restricted to a low dosage and that peak cognitive results are obtained by administering 0.3 mg/kg body weight. Most important, high doses (1.0 mg/kg) were detrimental to learning performance on a short-term memory task. Lower doses (Brown & Sleator, 1979) enhanced careful performance on the Matching Familiar Figures Test (Kagan, 1965), a clinical index of impulsivity on which ADD/H children generally respond rapidly or make many errors. In contrast, when the classroom behavior of each child was examined, the teacher reported continued improvement in the child's social behavior as the dose increased from 0. 3 mg/kg to 1.0 mg/kg. Although these studies have been supported by subsequent empirical studies (Brown et al., 1984; Douglas et at., 1986), not all studies have found impairment in cognitive performance and learning at higher doses (Charles, Schain, & Zelniker, 1981; Stephens et al., 1984; Whalen & Henker, 1984). Many investigators have interpreted
their data as supporting a linear relationship between doses of methylphenidate on cognitive and behavioral indices (Stephens et al., 1984; Rapport, Stoner, DuPaul, Birmingham, & Tucker, 1985; Rapport, DuPaul, Stoner, Birmingham, & Masse, 1985b). The disparity in findings is particularly troublesome to practitioners seeking optimal doses for their patients. The high degree of variability among the children for whom methylphenidate may be considered appropriate may in fact decrease generalizability between studies unless the conditions and subject samples are precisely replicated. Further, the side effects have been shown to increase with higher doses (Sprague & Sleator, 1977). The careful monitoring (cognitively and behaviorally) of individual patients and the systematic documentation of side effects is indeed necessary to achieve the desired clinical response. The comparative efficacy of methylphenidate and dextroamphetamine has been investigated in well-controlled clinical trials (Arnold, Christopher, & Huestis, 1978; Weiss, Minde, Douglas, Werry, & Sykes, 1971). Conners (1971) reported that the drugs seemed equally efficacious on most outcome measures except that methylphenidate was superior on several subtests of the WISC and produced fewer side effects. Weiss et al. (1971) found methylphenidate slightly more efficacious and the side effects comparable to those of dextroamphetamine. Although some investigators have found that behavioral ratings improve more with dextroamphetamine than with methylphenidate (Arnold et at., 1978), others have found methylphenidate more effective in improving learning (Fish, 1971). Some researchers have suggested that a poor response to one of these medications warrants a trial of the other.
Pemoline A third stimulant medication-magnesium pemoline-is structurally dissimilar to dextroamphetamine and does not have similar sympathomimetic activity, but it has compared favorably with dextroamphetamine and methylphenidate in short-term trials (Dulcan, 1986). The particular advantage of pemoline over other stimulants is that only a once-daily administration is necessary because its half-life averages 12 hours (Dulcan, 1986). Although the mechanisms of action and the pharmacokinetics are far less understood than for the other stimulant medications, one of the clear drawbacks of pemoline is its slower onset of action. Compared with methylphenidate and dextroamphetamine, in which
NEUROPSYCHOLOGICAL EFFECTS OF STIMULANT MEDICATION
changes may be discerned a few hours following ingestion, the beneficial clinical effects of pemoline may not be noted for several weeks (Dulcan, 1986). Kinsboume and Swanson (1980) reported, however, that an initial larger dosage of pemoline (recommended dose is 18.75 mg to 37.5 mg) followed by two daily doses of the recommended amount produced an immediate effect. The clinical effects of pemoline have been sustained for as long as 2 weeks following its discontinuation (Conners & Taylor, 1980), and the half-life has increased markedly with continued administration (Sallee, Stiller, Perel, & Bates, 1985). Although the short-term efficacy of pemoline has compared favorably with the other stimulants (Conners & Taylor, 1980), there has been some question as to whether pemoline is as effective long-term. Some researchers have suggested that it is not (Swanson, Kinsboume, Roberts, & Zucker, 1978). Severe dysphoria has been observed following the cessation of pemoline therapy (Brown, Borden, Spunt, & Medenis, 1985), and abnormal liver function tests, particularly during prolonged therapy (Gillman et al., 1980), have been noted. Moreover, its safety and efficacy for children under the age of 6 have not been established (Gilman et al., 1980). The advantages of pemoline over the other stimulants are the singledose schedule and less significant cardiovascular effects (Page, Bernstein, Janicki, & Michaelli, 1974). Nonetheless the few studies that have been published have suggested that more studies must be conducted before the long-term clinical efficacy of pemoline can be endorsed.
Clinical Uses ADD
Treatment Prevalence The clinical use of psychostimulant medication remains the most common treatment for ADD/H (Gadow, 1981). Gadow estimated that 300,000 to 600,000 children in the United States were receiving stimulant medication to control hyperactivity. Statements have been made as to higher rates to stimulant drug usage; in fact, the press once reported that 20 to 25% of the children in Omaha schools were receiving these drugs (Weiss & Hechtman, 1979). Although those percentages were later disclaimed, they underscore the enthusiasm for treating ADD/H children with stimulant drugs.
449
Despite the widespread usage of stimulant medication for this population, some experts have concluded that these figures may represent the overuse and abuse of stimulants by many practicing professionals (Weiss & Hechtman, 1979). From our clinical and research experience, we are inclined to agree. Nonetheless, the debate about the potential overuse and abuses of stimulants has spawned many empirical studies on the efficacy of stimulant medication. Having gathered several hundred reprints and preprints as well as many literature reviews while preparing this chapter, we can certainly affirm that psychostimulant medication is the most meticulously studied and best researched topic in pediatric psychopharmacology. Some practitioners, in fact, have endorsed the use of stimulants for ADD/H children because it is the best-documented therapy in child psychiatry (Gittelman, 1983).
Cognitive Effects Many careful studies have clearly demonstrated that stimulant drugs consistently improve performance on basic laboratory tasks of cognitive performance (Barkley, 1977). The efficacy of stimulants has been thoroughly documented in the areas of sustained attention and vigilance (Solanto, 1984). Specifically, methylphenidate treatment has significantly increased the number of ADD/H children's correct responses on the Continuous Performance Task (Mirsky & Rosvold, 1963) and has improved their simple reaction time on this task (Solanto, 1984); it has also improved performance on the Matching Familiar Figures Test (Kagan, 1965) and on the Porteus Mazes, two frequently used laboratory measures of impulsivity, both of which require the inhibition of an incorrect response (Brown & Sleator, 1979; Rapport et al., 1985b). On other cognitive tasks, including rote learning, memory, and concept formation, stimulants have been found to enhance performance. For example, on the paired-associate learning test, in which the subject is presented with word pairs and subsequently asked to respond with the stimulus (initial) word or picture alone, there have been several reports of fewer errors after methylphenidate administration (Solanto, 1984). Weingartner et al. (1980), who found that amphetamines improved the cued recall of acoustically processed words in ADD/H subjects, interpreted their findings to suggest that dextroamphetamine treatment merely stimulates memory function rather than correcting or altering a basic deficit in cognitive processing. (The clinical implica-
450
CHAPTER 24
tions of this interpretation are discussed later in this section.) Gan and Cantwell (1982) found that although methylphenidate improved acquisition on a basic learning task, it had no effect on retention 24 hours after learning took place. Dykman and his colleagues (Dykman, Ackerman, & Oglesby, 1979), examining the effect of methylphenidate on a visual search task, found a decreased frequency of aftersearch lapses and extraneous responses in ADD/H children. The neuropsychological measures that have demonstrated greater variability in stimulant drug effects are the Bender Visual Motor Gestalt Test, the Frostig Developmental Test of Visual Perception, and particular scales on the Wechsler Intelligence Scale for Children, such as the picture completion subtest (Kavale, 1982). It has been suggested that because ADD/H children typically use more immature strategies for approaching cognitive tasks than do normally developing children, stimulant drug improvement may tend to occur more frequently on rote or simple tasks than on measures that require higherorder information processing (Douglas & Peters, 1979). This hypothesis has been corroborated by Adams (1982), who found that methylphenidate had limited effects on a task requiring higher-order schemata.
Behavioral and Motor Effects A prodigious number of well-controlled studies have consistently demonstrated that psychostimulants dramatically improve the behaviors of ADD/H children, sometimes to the point that the children are indistinguishable from their normal peers (Abikoff & Gittelman, 1985; Whalen et al., 1978). Ullmann and Sleator (1985), using a multidimensional teacher rating scale, demonstrated that methylphenidate dramatically improves attention and helps somewhat in decreasing activity level but that it often has minor effects on deficient social skills and aggressive-oppositional behavior. Improvement on parent and teacher rating scales of hyperactivity has been shown in many studies (Rapoport, 1983). Positive changes have also been reported with stimulant drug treatment in mother-child interactions (Barkley, Karlsson, Pollard, & Murphy, 1985) and in teacher-child interactions (Whalen et al., 1978; Whalen, Henker, &Dotemoto, 1981). Barkley (1981) found that as children's compliance to maternal commands increased (as a function of stimulants), mothers provided more positive attention. Similarly, teachers used significantly fewer control-
ling, guiding, or disciplinary actions with children who were receiving methylphenidate as compared with children on placebo doses (Whalen, Henker, & Dotemoto, 1981). Dextroamphetamine and methylphenidate have been found equally effective in ameliorating behavioral difficulties in ADD/H children. Concerning motoric effects, the psychostimulants have decreased activity levels in structured settings, but the effects have been more variable in less restricted situations (Donnelly & Rapoport, 1985). One study, which took place in a naturalistic setting and used 24-hour monitors, demonstrated that dextroamphetamine significantly decreased activity in a structured classroom and significantly increased activity during physical recreation (Porrino, Rapoport, Behar, Ismond, & Bunney, 1983). Similarly, Whalen and her associates (Whalen et al., 1978) found that methylphenidate reduced ADD/H boys' gross motor movements, vocal noise, and disruption in a classroom to a level indistinguishable from the boys' normal peers. Methylphenidate has also been found to improve handwriting (Whalen, Henker, & Finch, 1981) in ADD/H children.
Paradoxical Effects The effects of stimulants on ADD/H children have been regarded as paradoxical, and a favorable drug response bas often been viewed as confirmation of a diagnosis of ADD/H. Contrary to this popular clinical lore, "normal" children and adults have cognitive and behavioral effects that are similar to the effects of psychostimulants on ADD/H children (Rapoport, 1983). Moreover, normal and ADD/H children have similar physiological responses (e.g., heart rate slows more during the foreperiod of a reaction time task). This finding suggests that stimulant actions are not specific to ADD/H children and that exploring the efficacy of stimulants with other clinical populations (e.g., children who have learning disabilities and conduct disorders) may indeed be warranted.
Learning Disabilities Most of the studies on the effects of stimulant drugs and learning performance have been conducted with ADD/H children, some of whom were also learning disabled (Gadow, Torgesen, Greenstein, & Schell, 1986). Few researchers have examined the efficacy of stimulants on non-ADD/H children who
NEUROPSYCHOLOGICAL EFFECTS OF STIMULANT MEDICATION
have learning disabilities. Although we found no figures on the number of learning-disabled children who receive medication for learning problems alone, clinical experience tells us that the number is probably quite small (Gadow et al., 1986). Learning disabilities is clearly a heterogeneous diagnostic category. Children who have learning disabilities are typically categorized by a deficit in a specific academic area, such as reading, arithmetic, or spelling. The diagnostic category is further complicated because the terms learning disability, hyperactivity, minimal brain dysfunction, and attention deficit disorder are used interchangeably. In fact, these terms are often lumped under the diagnostic rubric learning disability. To complicate matters even further, much research has shown that many children with learning disabilities have attentional deficits (as do their ADD counterparts) and are more susceptible to extraneous distractors (Douglas & Peters, 1979). Because learning-disabled and ADD children share similar symptomatology (Douglas & Peters, 1979), it is logical and enticing to expect learning-disabled groups to respond to stimulants similarly as do ADD!H children. It is of course quite dubious that stimulants enhance knowledge of complex academic tasks. Thus, if psychostimulants were to be effective for children with specific learning disabilities, it would only be reasonable to expect these children to learn better because their improved attentional capacity would increase their capacity for instruction (Gittelman, 1983). A review of the empirical literature has revealed very few stimulant drug studies of non-ADD/H, learning-disabled children. Of those few, most have been conducted with reading-disabled children. Because reading disabilities make up the largest category of learning disabilities (Bryan & Bryan, 1978), we mainly review tbe studies of children with reading disabilities (we do mention stimulants and arithmetical disabilities). Huddleston, Staiger, Frye, Musgrave, and Stritch ( 1961) conducted the first placebo-controlled study of stimulant drug effects on reading performance. Deanol, which is a relatively weak and less potent stimulant than methylphenidate, dextroamphetamine, or pemoline, was compared with placebo doses in 60 retarded readers for 8 weeks. Scores on the Gates Reading Test showed no drug effects on reading performance. In a methodologically rigorous study of 6 I children who had been referred by teachers because of poor academic performance, Gittelman-Klein and Klein (1976) investigated the efficacy of methylphenidate versus placebo in matched controls for
451
12 weeks. The subjects, none of whom demonstrated signs of hyperactivity or behavior disorders, were 2 years behind in reading achievement. The findings indicated significantly improved performance on laboratory measures of attention, cognitive performance, and arithmetic achievement for the methylphenidate group. The methylphenidate and placebo groups showed no differences in reading achievement. These findings may suggest that stimulants significantly affect attentional deployment but have little effect on complex academic tasks such as reading. In an attempt to isolate a subgroup of readingdisabled children who might be positive methylphenidate responders, Gittelman (1980) examined stimulant drug responses in reading-disabled children who showed evidence of neurological soft signs (minimal brain dysfunction) and in their reading-disabled peers who showed no such signs. The data failed to indicate that the reading-disabled children with minimal brain dysfunction are more likely to improve in reading because of methylphenidate therapy. This study is of prime clinical importance because it does not support the notion that learningdisabled children who have neurological deficits coupled with a specific learning disability (reading disorder) are necessarily good candidates for stimulant drug therapy. As stimulants have been found particularly efficacious in ameliorating the attentional deficits of ADD/H children, a reasonable hypothesis is that stimulant medication may enhance attentional capacity in learning-disabled children, thus making them more receptive to reading instruction. To learn whether stimulant medication may facilitate a special education or remedial approach to learning disabilities, Gittelman and her associates (Gittelman, Klein & Feingold, 1983) compared two groups of reading-disabled children in a double-blind study. Both groups received reading remediation; one group was given a placebo, and the other was given methylphenidate. Although the latter group showed enhanced attentional deployment on a series of cognitive tasks, the clinical advantage induced by methylphenidate was not dramatic (Gittelman, 1983). Although support for the efficacy of stimulants in the academic domain (Gadow et al., 1986) has been limited, some of the disappointing findings have been attributed to high doses of stimulants, which in comparison with relatively lower doses, have decreased cognitive effects (Brown & Sleator, 1979; Brown et al., 1984; Sprague & Sleator, 1977). Aman and Werry ( 1982) examined the effects of low doses of methylphenidate (0.35 mg/kg body weight)
452
CHAPTER 24
in a crossover design utilizing three drug conditions: methylphenidate, diazepam (Valium), and a placebo. Subjects were 15 children who lagged 2 years behind their mental age in reading achievement. No differences were found, probably because of the short time (6 days) that the children received stimulant medication. The length of drug treatment is crucial in studying children with learning disabilities because the capacity for improved learning over time, not improved performance per se, is the variable of interest (Gittelman, 1983). We found only one methodologically adequate study that included an arithmetic achievement test as a dependent measure (Gittelman-Klein & Klein, 1976). Clearly, additional research is needed before any definitive statement can be made about learning disabilities and stimulants. In comparison to the extensive research on the effects of stimulants in ADD/H children, studies of the effects of stimulants on learning disabilities have been decidedly disappointing. Aman (l982a) reviewed six fairly rigorous studies of stimulant drug effects in children who had academic difficulties. His conclusion agrees with Gittelman's (1983) advice that clinicians should use caution in prescribing stimulants for learning-disabled children without hyperactivity until research shows more promise in using stimulants with this population.
Conduct Disorders We found no studies on the efficacy of stimulant drugs in conduct-disordered children. As mentioned, the idea that stimulants have a paradoxical effect on ADD/H children and that the effects are reversed at puberty has not been supported empirically (Rapoport, 1983). This calls into question the diagnostic specificity of stimulant drug response and, more importantly, suggests that stimulants may be effective for clinical populations whose symptomatology is similar to that of ADD/H groups. In fact, the subjects in clinical trials of stimulants have been fairly heterogeneous in age, intellectual functioning, degree of attentional deficits, and amount of aggression (Rapoport, 1983), even when all the subjects were designated ADD/H in accordance with DSM-III criteria (American Psychiatric Association, 1980). For example, in some clinical trials with ADD/H patients, conduct disorders and symptoms of aggression were exclusionary criteria (Varley, 1983); in other studies (Pelham et al., 1985), including research from our laboratory (Brown, Wynne, & Medenis, 1985; Brown, Borden, Wynne, Schleser,
& Clingerman, 1986), subjects included children with dual diagnoses-conduct disorder and ADD/H. Whether stimulants will prove effective for conduct disorders remains to be addressed, but we are encouraged by the comprehensive reviews of stimulant drug effects (Pelham & Murphy, 1986) that have suggested that the target behaviors demonstrating clearest improvement are the disruptive and antisocial behaviors. If that suggestion proves accurate, the response to stimulant drugs in conduct-disordered groups may be similar to the effects observed in their ADD/H peers. Rapoport (1983) pointed out that many subjects in the earlier studies were unselected delinquent populations (Risenberg et al., 1963), suggesting that a "pure" conduct-disordered group would in fact benefit from stimulants. Further supporting the clinical use of stimulants for conduct-disordered children are the clinical gains in sociability and the reductions in aggression seen in ADD/H children (Rapoport, 1983). A double-blind crossover study of the response to stimulants in a group of conduct-disordered children with attentional problems and in a group without attentional problems should be conducted. The systematic study of a group with dual diagnoses of ADD/Hand conduct disorders versus a group with "pure" conduct disorders would also be fruitful.
Mental Retardation In a lucid and thorough review, Gadow ( 1985b) noted that stimulants are the psychotropic drugs most commonly prescribed for the mentally retarded, particularly the educable and the trainable mentally retarded youngsters who are hyperactive. This seems counter to clinical intuition in that neuroleptics are widely used in institutions serving the mentally retarded in this country (Gadow, 1985a) and the efficacy of stimulants for behavior disorders in this population has been questioned (Aman, 1982b). Surprisingly, according to a very careful and thorough, albeit limited, review of studies in which stimulants were given to retarded youngsters, the drugs produced therapeutic benefits comparable to those observed in youngsters who were not mentally retarded. It was suggested that the degree of improvement from stimulants relates directly to the severity of retardation, the more severely mentally retarded children benefiting the least (Gadow, 1985a). In the very early clinical trials, mentally retarded children were commonly included as subjects (Cutler, Little, & Strauss, 1940; Molitch & Sullivan, 1937). In those investigations, mentally retarded
NEUROPSYCHOLOGICAL EFFECTS OF STIMULANT MEDICATION
subjects improved in much the same way as their nonretarded peers. In only a few recent studies, however, have the subjects been retarded (Poling & Breuning, 1983; Varley & Trupin, 1982); in most studies, mentally retarded children have been excluded (Kavale, 1982). Nonetheless, findings clearly support the efficacy of stimulant drug effects on a wide array of dependent measures, including caregiver ratings and psychometric tests. Poling and Breuning (1983), in their research with trainable mentally retarded children, found a curvilinear dose-response relationship in the behavioral ratings of caregivers; that is, when methylphenidate doses were particularly high, the behavioral ratings of teachers deteriorated. This finding is in stark contrast to studies with ADD/H children (Charles et al., 1981; Rapport et al., 1985a,b; Sprague & Sleator, 1977), in which a linear relationship between dose of medication and teacher ratings has been found. Poling and Breuning's findings lend tentative support to Aman's (1982b) hypothesis that an attentional deficit underlies retardation and that it can be exacerbated by higher doses of a stimulant drug. Despite the widespread clinical use of psychostimulants for the mentally retarded, the area has been decidedly neglected in pediatric psychopharmacology research. Gadow (l985a) offered several explanations for that neglect, including the fact that most research-oriented hyperactivity clinics exclude mentally retarded subjects because often there are too few for systematic study. Although this population is readily available in the public schools, conducting pharmacological research in an educ~tional setting presents many obstacles and difficulties. Also, the clinical myth that pharmacological intervention would have little effect with retarded persons has unfortunately been promulgated throughout the scholarly and research literature (Klein, Gittelman, Quitkin, & Rifkin, 1980). Collaboration among investigators and single-subject designs will, we hope, solve some of these problems (Gadow, 1985a).
453
(Gilman et al., 1980) for use with preschoolers; dextroamphetamine is recommended only for children aged 3 and older. Nonetheless, there has been some debate concerning how early in life stimulants may be used without compromising the safety and wellbeing of the child. Gross and Wilson ( 1974) recommended 2.5 mg of amphetamines daily for children up to the age of 4; others argued for this dosage twice daily (Wender, 1971). In a double-blind placebo trial in preschool hyperactive children, Conners (1973) examined the effects of methylphenidate on cognitive, motoric, and psychological test performance according to teacher and pediatrician's behavioral ratings. The results were decidedly disappointing: The children's overt behavior was little affected by stimulant treatment. In another study, improvement occurred in only 3 of28 preschoolers receiving methylphenidate (Schleifer al., 1975). These are consistent with Conners's (1973) results, which offer little support for improved behavior or improved psychometric test scores of preschoolers in response to stimulant drugs. In addition to the limited efficacy of stimulants for preschoolers, an even greater concern has been the untoward side effects. Although Conners (1973) found minimal side effects, Schleifer et al. (1975) found that methylphenidate was associated with insomnia, anorexia, increased solitary play, and poor peer relationships. The mean daily dose for both studies, however, was more than 10 mg, a dosage far exceeding the recommended daily dosage for preschool children (Gross & Wilson, 1974; Wender, 1971). Another pessimistic note was sounded in the findings presented by Zara (1973): The positive responders to stimulants were less verbally competent in an experimental learning situation when they were receiving stimulants. In general, the clinical efficacy of stimulant medication has been more variable in preschool children than in older groups, and the incidence of side effects, including dysphoria, anorexia, and insomnia, has been particularly high in preschoolers. Such findings lead us to concur with Dulcan (1985), who Developmental Issues cautioned that stimulants be used for preschoolers Although most stimulant research on learning only ''in the most severe cases where parent training and behavior problems has been conducted with pri- and placement in a highly structured, well-staffed mary school populations, the potential efficacy of preschool program have been unsuccessful or are not stimulants in other age groups has recently been possible" (p. 395). recognized.
et
Preschoolers
Adolescents
Methylphenidate has not been specifically approved by the Food and Drug Administration
According to clinical lore, stimulant medication is to be used with preadolescent children experienc-
454
CHAPTER 24
ing attentional difficulties, and the symptomatology is simply "outgrown" at adolescence (Gross & Wilson, 1974). This notion has been one reason for the refusal to administer psychostimulant medication to these youths beyond puberty, despite the persistence of cognitive, emotional, and social problems well into adolescence (Hoy, Weiss, Minde, & Cohen, 1978) and even adulthood (Wender, Reimherr, & Wood, 1981). Clampit and Pirkle (1983) speculated on other reasons that psychostimulants are believed to work only until adolescence. At puberty, the most visible symptoms (e.g., overactivity), which are so responsive to stimulant medication, often diminish, and cognitive and social dysfunctioning assume center stage. Thus, many practitioners have falsely reasoned that stimulants are no longer needed because the visible symptoms have dissipated. A second possible reason that stimulant medication is not used with adolescents is that the school structure changes and the adolescent experiences more social freedom and less classroom structure. Although the youngster may no longer seem to need the medication, the difficulty does not result from the structure imposed by the elementary school classroom and the adolescent with attentional problems may be a "silent sufferer" (Clampit & Pirkle, 1983, p. 816). A third possible reason is the high level of variability in the behavior of normal adolescents and in particular adolescents who have externalizing disorders such as ADD/H. Differentiating the changes due to stimulant medication and the changes due to other factors is difficult. A fourth possible reason is a dated theoretical tenet rooted in the minimal brain dysfunction theory of hyperactivity: children who have attentional deficits are experiencing difficulty with an ''immature'' central nervous system (Clampit & Pirkle, 1983), a lag in neurological development, and they will "catch up" at puberty. Thus, it was erroneously reasoned that psychostimulants would not be effective at puberty and might even have behaviorally stimulating effects. The notion that stimulant medication does not act therapeutically on adolescents undoubtedly explains the dearth of stimulant drug trials with this group. Contrary to this popular misconception, researchers have recently reached consensus that adults (Wender eta/., 1981) and adolescents (Cantwell, 1979) once diagnosed as having attentional problems can benefit markedly from stimulant drug therapy. Open trials of stimulant medication (Lerer & Lerer, 1977; MacKay, Beck, & Taylor, 1973; Safer & Allen, 1975a,b) have clearly refuted the myth
(Clampit & Pirkle, 1983) that psychostimulants are ineffective for ADD adolescents. Moreover, two recent double-blind studies (Coons, Klorman, & Borgstedt, 1987; Varley, 1983) and one single-blind trial (Garfinkel et al., in press), as well as current research in our laboratory (Brown & Sexson, 1986), have yielded very encouraging data that moderate doses of methylphenidate may improve behavioral ratings and enhance cognitive functioning in adolescents. Many practitioners share the notion that psychostimulants have a paradoxical effect on children of elementary school age and that the stimulant effects are reversed at puberty. In truth, there are many similarities in the stimulant drug effects in children and in adolescents, including improved behavioral ratings and enhanced cognitive functioning (Klorman, 1986). Additional support for the use of pharmacotherapy at adolescence has come from comprehensive reviews (Solanto, 1984; Whalen & Henker, 1984) concluding that the target behaviors that demonstrate the clearest improvements with stimulants are the disruptive and antisocial behaviors, which are generally characteristic of adolescents. In addition, the observation that adolescents with ADD/H manifest cognitive impairments similar to those of their younger counterparts (Hoy et al., 1978) suggests that judiciously monitored stimulant medication would yield cognitive as well as behavioral results (Brown & Sexson, 1986; Coons etal., 1987). All three of the empirical studies that have addressed this issue (Coons et al., 1987; Garfinkel et al., in press; Varley, 1983) have shown significant behavioral improvements in adolescents treated with stimulants. See Table 2 for a summary of the studies on the efficacy of stimulants in adolescents. Despite Varley's (1985) statement that "not a single study has reported a negative stimulant drug effect outcome in adolescence" (p. 216), important questions remain. Although some researchers have suggested that a chief difference in the stimulant drug regimens of children and of adolescents is the marked variability in the effective dosages for adolescents, we have no systematic data about the effective dosage range for adolescents. Although the range of methylphenidate dosages in adolescent studies-! 0 mg to 60 mg per day-appears comparable to the absolute dosages used with pediatric populations, the effective dosages for adolescents are typically lower (mg/kg) than the dosage for prepubescent children. Specific dose responses and dose requirements for adolescents still need to be empirically established. Also, is the variability in effective dosages in fact marked for adolescents? Although clinical trials have
Psychopathic and neurotic juvenile delinquents MBD
Korey (1949)
Lerer & Lerer ( 1977)
ADD/H
ADD/H
Garfinkel et al. (in press)
(1987)
Delinquent boys
at.
Population
Eisenberg et al. (1963)
Coons et
Study
27
20
17
21
19
N
Methylphenidate
Benzedrine sulfate
Dextroamphetamine Dextroamphetamine
Methylphenidate
Stimulant
Individually titrated; mean dose not reported 5 mg/day, Week I; 20 mg/day, Week2 40 mg, total daily dose
40 mg/day
Placebo; 25 mg/ day, Week I; 40 mg/day, Weeks 2 and 3
Dose
Open
Double-blind
Double-blind
Double-blind
Double-blind
Type of trial Results
Improvement noted on Conners's teacher ratings, academic performance, and behavior control
Subjects more compliant; behavior improved
NO
NR
NR
NR
NO
Abuse
(continued)
Weekly ratings by parents on the Conners's questionnaire and . global ratings of outcome indicated superior behavior under methylphenidate; significant increases in accuracy noted on Continuous Performance Test Reduction in aggression and gains in sociability and manageability Improved cognitive response
TABLE 2. Studies Examining Efficacy of Stimulants in Adolescents a
m
z
0
~
0
~
~
!
{/)
q
~
8
0
C)
I
~
Delinquent boys
ADD/H (some patients also had Axis U diagnosis of conduct disorder) ADD/H; all patients previously determined to be responsive to stimulants
Maletzk:y ( 1974)
Safer & Allen (1975b)
22
24
28
10
N
Methylphenidate
Methylphenidate and dextroamphetamine
Dextroamphetamine
Methylphenidate
Stimulant Dose
Placebo; 0.15 mg/kg plus 0.3 mg/kg b.i.d.
mg total daily dose Not specified
S mg per dose/40
30 mg, total daily dose
aNo, none observed; NR, not reported; MBD, minimal brain dysfunction; b.i.d., twice daily.
Varley (1983)
MBD
Population
MacKay et al. (1973)
Study
TABLE 2. (Continued)
Double-blind
Open
Open
Type of trial
Reductions in pretreatment ratings of restlessness, inattention, and aggression, by ratings on an adapted version of Conners's Teacher Questionnaire Significantly lowered hyperactivity ratings, by teacher and parent Conners's ratings; clinical effects only slightly greater with the higher dose
Improvement noted in academic functioning, Raven matrices, and EEG abnormalities Improvement in aggression and sociability
Results
NR
NO
NO
NO
Abuse
~
I
~
NEUROPSYCHOLOGICAL EFFECfS OF STIMULANT MEDICATION
suggested the short-term efficacy of methylphenidate for adolescents, the major side effects have not been systematically reported-an important issue, particularly given the sensitivity of adolescents to medication (Cantwell, 1985). For example, some researchers have mentioned dysphoria and irritability following methylphenidate administration (Donnelly & Rapoport, 1985). Whether these effects are increased in adolescents who are receiving stimulant medication will be a fruitful area for study. The side effects influencing the cardiovascular system, blood pressure, and pulse rate may be even more important in adolescents than in younger children (Cantwell, 1985). The need for additional well-controlled clinical trials with ADD adolescents is underscored by the research efforts of Safer and Krager (1985), who found a prominent increase in the rate at which stimulant medications were prescribed from 1975 to 1983 for secondary school pupils.
Adults In a series of impressive follow-up studies, the outcomes for adults previously diagnosed as hyperactive have been examined (Thorley, 1984; Weiss, 1983). The problems of the adults seemed to mirror the difficulties of their younger counterparts, including attentional difficulties, impulsivity, temper outbursts, poor organization, an inability to complete tasks, low self-esteem, and problems in social interaction (Woods, 1986). A logical extension of that research would be an examination of the efficacy of stimulants in adult populations previously diagnosed as ADD/H. According to the limited literature in this area, adults may benefit significantly from psychostimulants; Woods (1986) estimated, in fact, that 60 to 80% of these adults would respond therapeutically to psychostimulant medication. As do their younger counterparts, adults tend to experience the direct effects of stimulant medication, including similar side effects (Ross & Ross, 1982). In reviewing positive clinical response in adults who had received stimulant medication for more than 5 years, Woods recommended that drug therapy be continued in adults as long as they show undesirable symptoms, which typically abate during later adulthood. Despite the promise of psychostimulants for adult populations, much variability has been reported in the types of stimulants and the effective dosages for adults (Ross & Ross, 1982; Woods, 1986). Woods noted that most of the patients he studied required daily doses of 20 to 40 mg, but he pointed out that specifying dosages is diffitult be-
457
cause of individual variables. For example, response to methylphenidate has been established with as little as 1 to 2 mg, but some patients have required significantly larger doses (up to 120 mg). Despite the promise of stimulants for adult populations, Ross and Ross (1982) cautioned that adults are often noncompliant and actively resist pharmacotherapy despite the fact that they typically undergo treatment voluntarily and experience marked symptomatic relief when given stimulant drug therapy. Although stimulant drug therapy increases attentional capacity in adults, adults often become depressed as awareness of their life situations develops (Woods, 1986). Ross and Ross suggested that such dysphoria may result when the adult discontinues drug therapy prematurely. Unlike adults who have been diagnosed as depressed and who seek treatment because of their pain, ADD/H adults may have adapted to their pathology and consider themselves normal, making compliance even more difficult (Woods, 1986). Research on stimulant drug treatment for adults (e.g., specific dosages for specific symptoms) is an area ripe for further investigation. However, because of the unique problems experienced by adult populations, such research will undoubtedly prove challenging.
Assessing Therapeutic Responses Psychophysiological Correlates of Stimulant Drug Response Extensive research has been conducted on the behavioral effects of stimulant medication. Most of this research has been chiefly concerned with ameliorating overt clinical symptoms and has thus used a variety of behavioral measures to evaluate clinical response (Ross & Ross, 1982). Recently, however, efforts have been increasingly devoted to investigating physiological and biochemical processes in an effort to understand the outcome of stimulant drugs for various disorders (Solanto, 1984; Yellin, 1986) and ultimately to predict more efficiently the response to stimulant drug treatment. As Yellin pointed out, clinical improvement in target symptoms is simply not sufficient for the evaluation of drug efficacy because some psychotropic medications, although particularly effective in improving specific target behaviors, have deleterious effects on other psychophysiological processes.
458
CHAPTER 24
Autonomic Nervous System Variables Particular interest has focused on autonomic nervous system variables in ADD/H children in an attempt to assess whether cognitive processes such as attention are mediated by impairments in arousal and orienting responses. Satterfield and Dawson (1971) proposed that children with ADD/H are underaroused in comparison with normally developing children and that psychostimulants act by normalizing the arousal level. Although that hypothesis has not received widespread empirical support (Yellin, 1986), agreement is fairly consistent that stimulants do affect arousal level (Rosenthal & Allen, 1978; Solanto, 1984; Yellin, 1986). In general, stimulants produce increases on measures of arousal, including skin conductance levels, frequencies of spontaneous skin conductance responses, heart rate, cardiac deceleration to an orienting stimulus, and decreased heart rate variability (Yellin, 1986). These measures have also been linearly related to dosage (Rapoport, 1983) and positively correlated with decreased reaction time and improved performance on measures of sustained attention. Moreover, the cardiovascular effects of stimulants have been remarkably similar for all age groups, although dextroamphetamine generally produces more severe cardiovascular changes than does methylphenidate or pemoline (Huessey, Cohen, Blair, & Wood, 1979). Zahn, Rapoport, and Thompson ( 1980) compared the psychostimulant effects on autonomic nervous system responses in normal and ADD/H prepubescent boys. Both groups demonstrated similar behavioral (reduced motor activity), cognitive (decreased impulsivity), and autonomic responses (increased heart rate, increased cardiac deceleration during orienting, and decreased skin temperature). The investigators concluded that stimulants have similar behavioral and autonomic effects in ADD/H and normal children, although the magnitude of the effects may differ. They also noted that the beneficial effects on behavior were not necessarily due to the activating properties of stimulants, as the increased autonomic responses of the placebo group were not accompanied by reduced motor activity, reduced impulsivity, and enhanced attentional deployment.
Central Nervous System Variables The interest in measuring CNS effects in clinical populations and the effects of psychostimulants on these variables has evolved because various stages of information processing, such as selective attention
and decision making (Donchin, Ritter, & McCallum, 1978), have been related to CNS functioning. It is hoped that the study of CNS variables will help us to better understand how stimulant drugs affect information processing. A number of studies have suggested a greater overall percentage of EEG alpha waves, a smaller percentage of beta waves, reduced alpha waves, and increased EEG variability in ADD/H children (Yellin, 1986). These data have been interpreted as indicating lower cortical arousal in these children. According to Surwillo's (1980) data, the EEGs of children with ADD/Hare immature in relationship to their chronological age and the stimulants have a maturing effect on the EEGs. Stimulants have also been found to have a normalizing effect on beta frequencies. Event-related potentials (ERPs), also known as evoked potentials, are electrical changes in the brain in response to specific stimuli. They may occur in response to internal events (i.e., the K-complex during Stage 2 sleep, which may be a response to cardiovascular events) or to external events, such as visual or auditory stimuli (Yellin, 1986). In many studies the amplitude of several of the late negative and positive components of ERPs has been smaller in ADD/H subjects than in controls, particularly on tasks requiring active attention (Rosenthal & Allen, 1978; Yellin, 1986). Interestingly, the positive late component (P300) has occurred 300 msec after the initial stimulus, which has been sensitive to taskrelevant aspects of stimuli. Specifically, the P300 components in ADD/H children (Michael, Klorman, & Salzman, 1981) and adolescents (Loiselle, Stamm, Maitinsky, & Whipple, 1980) were of smaller amplitude than in normal controls. Most importantly, the smaller P300 components of the ADDIH subjects were correlated with poorer performance on the attentional measures. Unfortunately, this finding is not specific to ADD/H groups but has also been found in learning-disabled and schizophrenic populations. Methylphenidate has been associated with improved performance on the attention tasks (reduced reaction time and fewer errors) and a concomitant increase in the amplitude of P300 (Michael et al., 1981). However, methylphenidate has been shown to increase P300 amplitude in normal children and · adults, though less dramatically and during experimental tasks requiring more effort (Coons, Peloquin, & Klorman, 1981). In short, these studies suggest that improved attention subsequent to the administration of psychostimulants is correlated with a normalizing of several nervous system variables. However, the specific ac-
NEUROPSYCHOLOGICAL EFFECTS OF STIMULANT MEDICATION
459
tions of stimulants on central and autonomic nervous al., 1985) have led investigators to examine other system variables are not well understood. According hypotheses. to one somewhat encouraging study (Yellin, 1986), A second possibility is that the duration of dextroamphetamine may act on the CNS by activat- pharmacotherapy has been too short to allow the deing the inhibitory norepinephrine receptors. Al- tection of improvement on standardized academic though promising, the data also suggested that a sin- tests (Sprague & Berger, 1980). Specifically, the gle mechanism does not sufficiently explain the brief intervention used in the studies, in combination pharmacological effect of ERPs because different with academic measures of limited reliability and cortical regions are differentially affected by stim- validity (Gadow & Swanson, 1985), which have ulants. been designed to detect only gross changes in Although research on the psychophysiological achievement level, imply that to show significant response to stimulants is in its infancy, exciting hy- improvement on standardized tests, a child would potheses about the dynamics of psychophysiological have to gain academic skills at an extremely high processes are being evaluated. With increasingly so- rate, one much higher than expected for the normal phisticated measurement of these processes and child. greater subject homogeneity, the widespread clinical One widely accepted explanation for this disuse of psychophysiological procedures in the assess- crepancy has been the failure to use assessment dement of stimulant drug responses will likely be com- vices sensitive to the level of change expected during monplace in the near future. the relatively brief treatment periods of most clinical trials (Sprague & Berger, 1980). This hypothesis has been tested in recent studies in which academic meaLearning sures have replaced the traditional standardized achievement tests. The new measures, which contain several items to tap each specific skill, do appear Academic Achievement more sensitive. Pelham and his colleagues (Pelham et Psychostimulants have been found to affect at- al., 1985; Stephens et al., 1984) have developed tention span and concentration favorably and to im- several tests to assess stimulant drug effects, for exprove classroom behavior (Gittelman, 1983). More- ample, a spelling task based upon the requirements over, many parents and teachers have reported that for spelling tasks in school. In the first study (Stechildren show marked improvement in their school- phens et al., 1984), children were presented with work in response to stimulants (Rapoport, Quinn, word lists and led through practice trials before being Bradbard, Riddle, & Brooks, 1974; Sleator, von tested. In the second study (Pelham et al., 1985), Neumann, & Sprague, 1974). These reports have arithmetic work sheets and multiple-choice reading been contradicted, however, by the repeated failure comprehension problems supplemented the spelling of investigators to detect these improvements on task, and the children were given a week of structured standardized academic tests (Brown, Wynne, & practice before the test. Numerous problems adapted Medenis, 1985; Brown et al., 1986; Gadow, 1983, to grade levels were presented on each task. Another 1985b). Because the basic cognitive skills needed for important innovation was the use of average daily academic tasks have been so positively affected by performance scores on reading and arithmetic tasks, stimulants (Swanson & Kinsbourne, 1979), re- thus reducing as a source of variance the day-to-day searchers have been puzzled by the lack of improve- variability commonly reported for ADDIH children. ment on standardized tests and have proposed several In a similar study, arithmetic and spelling tasks hypotheses to explain the discrepancy. were designed with increased sensitivity to the shortDrug dosage levels that are high enough to cause term gains observed with stimulants (Douglas et al. , learning to deteriorate have been proposed as one 1986). These tasks were scored for the number of explanation for the rather disappointing findings in items attempted as well as the number correct. the academic domain (Sprague & Berger, 1980). As Again, average daily scores were used to reduce the mentioned earlier, some evidence bas suggested that effects of day-to-day variability. dose response differs across domains of functioning The positive findings with these newer academ(Sprague & Sleator, 1977). Dosages determined by ic measures are very encouraging. They support the parent and teacher ratings of behavior are often set hypothesis that the measurement procedure used in higher than the level at which learning and cognition earlier studies (Sprague & Berger, 1980), not the are optimally influenced (Sprague & Berger, 1980). inefficacy of stimulants, resulted in a failure to detect Questions about dose-response curves (Rapport et academic gains. Whether stimulants affect learning
460
CHAPTER 24
by increasing attention and concentration or in some ylphenidate and pemoline on spelling word retention. other way is unknown, but the use of sensitive aca- Because of this finding, they cautioned against using demic measures should help elucidate stimulant drug both medications concurrently or switching the two drugs frequently for a given child. Although more effects in the academic domain. Long-term stimulant drug outcome studies for recent studies lack real support for state-dependent ADD/H children have typically used special class learning between placebo and active drug conditions, placement, grade retentions (holding back), and practitioners and researchers need to be aware of the grades to evaluate stimulant treatment (Weiss, possibility and should evaluate learning outcome Hechtman, Hopkins, Perlman, & Wenar, 1979). both on and off medication (Brown et al. , 1986). Such measures are useful because they have "realworld" implications for evaluating drug efficacy (Gadow & Swanson, 1985) and may be used to assess Behavior outcome over longer periods. Academic productivity may be measured on Direct Observations specific academic tasks (Douglas et al. , 1986) or by Direct observations of behaviors have provided teacher ratings of structured activities (Edelbrock & Rancurello, 1985). The use of such ratings has been important data concerning the efficacy of stimulant efficient and cost-effective (Gadow & Swanson, drugs. These observations are typically conducted in a classroom setting by multiple raters who have been 1985). extensively trained in the observational system. All behavioral categories are precisely defined and a selected period for observation is broken into subunits State-Dependent Learning ranging from 5 or more seconds to I minute (Ross & Ross, 1982). For a representative sample of the State-dependent learning has been implicated in child's behavior, observations are typically made for psychostimulant effects on learning (Swanson & several days at different times during the day. ObserKinsbourne, 1976, 1979). State-dependent learning is characterized by decrement in transfer between vational systems have a threefold advantage: (I) the medicated and undrugged states (i.e., information behavior to be observed can be defined by clear operlearned while the child is medicated is not easily ational criteria, (2) interrater agreement is readily retrieved if the child is not medicated when tested). established, and (3) the technique requires little More important, a decrement in transfer to the non- equipment and is quite simple (Ross & Ross, 1982). drug state occurs when learning takes place in the The disadvantages of direct observations are that medicated state. Swanson and Kinsbourne (1976) evaluating multiple categories can be quite demandcautioned that the effects of stimulants on ADD/H ing of the rater and the procedures are typically quite children are state dependent. They found that nearly costly. The classroom observational schedules that 30% more errors were made in response to items have been used successfully in stimulant drug studies learned while the children were receiving methare the Revised Stony Brook Observation Code ylphenidate than were made while the children were (Abikoff, Gittelman-Klein, & Klein, 1977) and the receiving placebos. This finding has evoked concern among the professional community because it has Classroom Observation System (Whalen et al. , serious implications for the use of stimulants as the 1978). Several other reliable observational systems sole treatment for ADD/H children. If the effects of are sensitive to stimulant drug effects (Blunden, stimulant medication are in fact state dependent, Spring, & Greenberg, 1974; O'Leary, Romanczyk, ADD/H children will demonstrate poor retention of Kass, Dietz, & Santagrossi, 1971). See the publicapreviously learned material when treatment ends, or tions of Ross and Ross (1982) and Sandoval (1977) once drug therapy has begun, the children will have for more information on observational systems. to be treated with stimulants indefinitely (which most physicians are unwilling to do) (Douglas, 1980). State-dependent learning effects with stimulants in ADD/H children have not been found in other studies (Aman & Sprague, 1974; Gan & Cantwell, 1982; Gittelman, 1982; Stephens et al., 1984), although Stephens et al. did find a significant statedependent learning interaction between meth-
Checklists and Ratings Behavioral checklists and rating scales have played a central role in evaluating stimulant drug trials. Their advantages are many, including simplicity, cost-effectiveness, and reduced subjectivity in parents' and teachers' judgments of improvement
NEUROPSYCHOLOGICAL EFFECTS OF STIMULANT MEDICATION
in response to stimulants. Rating scales are typically valuable when behavior must be synthesized and integrated to evaluate drug treatment response. Of the host of rating scales, most have proven effective and sensitive to psychostimulant effects (Edelbrock & Rancurello, 1985) and are available to teachers and parents as well as mental health professionals. (See Edelbrock and Rancurello, 1985, for a thorough and lucid review of the rating scales that may be used in drug trials.) Typically, rating scales, whether completed by parents or teachers, tap similar domains of child behavior (Edelbrock & Rancurello, 1985), thus suggesting that multiple sources of behavior may prove useful for evaluating stimulant drug effectiveness. The major standardized rating scales have consistently demonstrated drug sensitivity (Kavale, 1982), and global ratings by teachers and parents have not differed significantly. Interestingly, teachers and parents are less inclined than clinicians to rate drug-treated subjects as globally improved (Kavale, 1982); Despite the sensitivity of behavioral ratings to stimulant drug effects, rating scales have been less powerful predictors of stimulant drug response than have laboratory measures of attention and concentration (Kavale, 1982; Edelbrock & Rancurello, 1985). High behavioral ratings of activity are weakly associated with positive drug response (Edelbrock & Rancurello, 1985); ratings by teachers are somewhat better predictors than those of parents. High ratings of anxiety by teachers or by parents predict a poor response to stimulant medication (Edelbrock & Rancurello, 1985).
Measures of Personality and Social Functioning Other relevant measures are personality and temperament scales as well as sociometric ratings. Because of concerns about dysphoria, learned helplessness, and anxiety resulting from initiation and cessation of stimulant drug therapy (Brown, Borden, Spunt, & Medenis, 1985; Cantwell & Carlson, 1978; Whalen & Henker, 1980, 1984), the systematic quantification of these effects through the use of rating scales, structured interviews, and objective personality tests is certainly necessary. Although such tests and rating scales are readily available to clinicians and researchers, few of them have been used in drug trials. The greater use of such instruments is necessary for future research. Sociometric ratings and self-rating scales have been used in some stimulant trials (Brown, Wynne, & Medenis, 1985), and
461
although stimulants seem to improve self-ratings (Brown, Wynne, & Medenis, 1985), few studies have systematically examined sociometric ratings in response to stimulant drug therapy. Although personality change has not necessarily been an ultimate goal of stimulant use, the evaluation of personality change, particularly over long periods, would be especially interesting to the pharmacological researcher.
Ecological Measures According to the pediatric psychopharmacology literature, too few real-world measures have been used in evaluating drug outcomes. Weiss et a/. (1979), for example, used measures of truancy and law violations in the Montreal follow-up studies of ADD/H children. Such measures are particularly appropriate for long-term (several years) follow-up studies.
Psychological Testing
Clinical Evaluation The best approach to a clinical assessment of stimulant drug effects may be to examine the clinical trials that proposed to assess the individual symptoms or deficits for which the medication was initially intended. Because the effects of stimulants tend to be so widespread, crossing a number of behavioral and cognitive domains, Loney (1986) recommended that a comprehensive assessment be cross-situational, multidimensional, and multidisciplinary. The diagnostic approaches range from traditional paper-andpencil tests to computers (Kiee, 1986). Intelligence tests have also been used to assess the efficacy of stimulant drug therapy (Ross & Ross, 1982). A detailed examination of individual subtests from the Wechsler Intelligence Scale for ChildrenRevised (WISC-R) can help determine whether a child has demonstrated clinical response to stimulants. For example, the Freedom from Distractibility Factor (Kaufman, 1979), which consists of digit span, arithmetic, and coding subtests, has been responsive to stimulant drug effects (Brown et a/., 1984; Brown, Wynne, & Medenis, 1985). WISC-R subtests that reflect academic history and social reasoning would not be expected to be as responsive to stimulant drug effects. Kagan's Matching Familiar Figures Test (Kagan, 196.5), which was originally designated a test of reflection-impulsivity, has demonstrated drug sen-
462
CHAPTER 24
sitivity and sensitivity across doses (Brown & Sleator, 1979; Brown eta/., 1984). Ease of administration and scoring makes it valuable in clinical and in laboratory settings. Its limited normative data permit only pre-post comparisons rather than comparisons with a child's peer group. . . Several simple paper-and-penctl tests, mcluding the Bender Gestalt, Trail Making Test, and Stroop Color Word Test (see Klee, 1986, for areview), are available for assessing drug responses, although their responsiveness to medication has varied somewhat across studies (Kavale, 1982). Stimulants are also commonly used for children who have continually underachieved academically. Although standardized achievement tests in short-term and long-term studies have demonstrated a fairly poor response to stimulant drugs, criterion-related tests may demonstrate greater sensitivity. (Gado:-v & Swanson, 1985) to stimulants. Readily available measures such as school grades and records of promotion may also help in quantifying stimulant drug response. . During the past 5 years, many computenzed tasks have been developed, some of which have claimed sensitivity to stimulant drug effects (Garfinkel, 1986). Whether computerized assessment, a new means of assessing drug response in children, will withstand the empirical test of time will have to be determined through future research.
Neuropsychological Evaluation No standard batteries of neuropsychological assessment instruments are available for measuring specific responses to stimulant drugs. In fact, data concerning psychostimulant drug effects on neuropsychological measures are scant. If we a~sume that the neuropsychologist's task is to determme the _aspect of the functional system that is affected by stimulants, neuropsychological assessments would certainly be of theoretical and clinical importance because they would help determine greater diagnostic specificity of stimulants. [See Lezak's (1983) excellent compendium on neuropsychological techniques.] Just as neuropsychological techniques and instruments await careful trials with psychostimulants, standardized tests of neurological soft signs also deserve systematic investigation. The Physical and Neurological Examination for Soft Signs (PA~~~S) (Close, 1973) measures performance on activities such as eye tracking, synergy, balance, and ~raph esthesia. ADD/H children who were responsive to dextroamphetamine scored significantly higher than
nonresponders on this test (Shekim, Dekirmenjlan, & Chapel, 1979). Whether this test will help in predicting the responses to methylphenidate and pemoline in diagnostic subgroups should be a fruitful investigative topic.
Predicting Responses The ability to predict the response to stimulants would be of great clinical use. Time and worry would be reduced, as would unnecessary adverse side effects. Prediction of response may be dichotomized as the prediction of positive outcome and of deleterious side effects. Barkley (1976) found that measures of attention and concentration were the best predictors of response to stimulant drug treatment; the severity of attentional impairment seems related to the efficacy of stimulant drug response. Barcai's (1971) Finger Twitch Test has been calle~ the single be~t predictor of drug response, although Its psychometric properties have not been established (Loney, 1986). Halliday, Rosenthal, Naylor, and Callaway (1976) found that recordings of neurochemical activity were associated with stimulant drug response. Loney found no single variable clinically useful in predicting response to medication; even the combination_of age at referral, birth complications, and the seventy of overactivity and attentional difficulties accounted for only 25% of the variance in ADDJH children's responses to stimulants. Considering a multitude ?f predictors is clearly important, and more systematic research must develop precision in predicting drug response. Until better prediction is possible, a clinical trial remains the practitioner's best tool. The predictors of side effects have not been well documented. Clearly, the effects are related to the specific stimulants prescribed, th_e ~osa~e level, the use of drug holidays, and the adminiStration schedule (Loney, 1986). Certain family fact~rs ~ay also p~e dict side effects; for example, a family history of t1cs has been associated with the occurrence ofTourette 's syndrome following stimulant administration (Huey, 1985).
Limitations Stimulants have been demonstrated to improve the attention, concentration, cognitive impulse c~n trol, memory, and adult-rated and observed behaviOr of children with learning and behavioral problems. However, a variety of limitations and adverse effects have caused concern.
NEUROPSYCHOLOGICAL EFFECTS OF STIMULANT MEDICATION
Questions have been raised about the efficacy of stimulants in the social and academic spheres. Stimulant therapy is usually initiated during elementary school, and a child who is given stimulant medication usually has a history of problems in school and at home. Parents, teachers, school administrators, and peers have experienced these children's uncontrolled, often annoying behaviors for months or even years. That people in these children's social milieu already perceive them as problematic is supported by parent, teacher, and classmate evaluations showing that ADD/H children have more difficulty with their peers than do other children (Ross & Ross, 1982). Perhaps even more important, these children rate themselves as less accepted by peers (Campbell, Endman, & Bernfeld, 1977); Pelham (1980) found that rejection occurred as quickly as 2 hours after an ADD/H child met another child! A history of behavior problems that is known to others leads to the labeling of these children and expectations about their future behavior. Although medication may produce change (e.g., the observable behavior of ADD/H children), the negative expectations of others may delay or even prevent the formation of new, more accurate opinions of them. Thus, overattentiveness from teachers and teasing by peers may continue. ADD/H children's self-expectations may be quite low, and they may become bewildered at the continued lack of acceptance by their peers. Even worse, peers may use the medication as additional evidence that ADD/H children are "different" or "crazy" and may tease them about the pills. Some ADD/H children may be teased less as their disruptive behavior decreases. If positive attention does not increase dramatically, the children may increase their disruptive behavior to gain attention. This problem is most difficult in older ADD/H children who have not acquired the social skills and practice necessary to gain positive social attention. In conclusion, the negative expectations of the child and others, and the child's lack of knowledge about how to gain positive peer attention may strongly attenuate the potential influence of medication in the social domain. The short-term efficacy of psychostimulants in decreasing the symptoms of attentional deficits and behavioral difficulties has been well documented (Ross & Ross, 1982), but we know that stimulants directly affect behavior only while in the patient's system in sufficient quantity (Brown eta/., 1986). Until recently, it was commonly believed that ADD/H children outgrew their difficulties at adolescence (Weiss, 1983). Because of recent evidence that ADD/H patients experience difficulties into adoles-
463
cence and adulthood (Weiss, 1983), the duration of medication effects has become a greater concern. Unfortunately, because of ethical considerations, random assignment to long-term stimulant treatment is not possible. Limited studies have compared two types of long-term outcome research studies: studies of adolescents and adults who had taken medication briefly in childhood and studies of adolescents and adults who had taken medication almost continuously since childhood. Because of the alarming implication that patients might need to be maintained on stimulants for many more years than originally thought necessary, the search for longer-acting alternatives has intensified (Sprague, 1983). The combination of stimulants with various psychotherapies offers hope. Many clinicians and researchers have assumed that educational programs and psychotherapy have greater effects when children are receiving medication than when they are unmedicated (in an overactive, inattentive state). Further, changes attained through psychotherapy have been expected to last longer, resulting in the maintenance of therapy gains long after medication is discontinued. Unfortunately, this expectation has not been supported by most empirical studies (Brown et al.. 1986), although some researchers have reported long-term results (Hinshaw, Henker, & Whalen, 1984). (For a thorough overview, see the excellent comprehensive reviews by Gadow, 1985b, and Pelham & Murphy, 1986). In summary, stimulants are effective only while in a child's system (gains are not maintained beyond treatment), and psychotherapy and educational interventions combined with medication do not appear to improve the maintenance of treatment gains. One possible solution lies in the efforts to develop more effective nonpharmacological therapies. Because many people are opposed to drug therapy, these efforts will certainly continue. Although a second option is to continue children on stimulant medication indefinitely, the concern about serious long-term side effects increases with the duration of treatment.
Iatrogenic Effects
Physical The most common physical side effects of stimulants-anorexia and insomnia (Gittelman-Klein, Klein, Katz, Saraf, & Pollack, 1976)-are usually transient and may be less severe with certain medications (e.g., methylphenidate). Stimulants are frequently withheld late in the day to prevent insomnia,
464
CHAPTER 24
but Kinsbourne ( 1973) cautioned against this practice because of the overactivity "rebound" that occurs when the stimulant medication wears off. He argued that withholding evening medication results in maximum overactivity at bedtime. According to this line of reasoning, the use of stimulants late in the day should decrease rather than exacerbate insomnia. An alternative strategy has been to decrease the dosage when insomnia or anorexia becomes a problem (these side effects seem more severe with higher doses; Cantwell & Carlson, 1978). Relatively low dosages may be effective for many children in various domains of functioning (Sprague & Sleator, 1977) and should result in less insomnia or anorexia while promoting desired therapeutic effects. Lower dosages often decrease other side effects such as headaches, abdominal pain, and gastrointestinal distress. Concern has also been expressed about the cardiovascular effects of stimulant drugs. Stimulants are known to cause increased heart rate, respiration, and blood pressure (Gilman et al., 1980). Fortunately, these effects are transitory, decreasing as the medication is metabolized and eliminated from the body and as tolerance to the medication develops (Cantwell, 1979; Safer & Allen, 1975a). In addition, the changes in cardiovascular functioning are minimal (Brown et al., 1984), and no abnormalities have been found on children's electrocardiograms, even following long-term stimulant therapy (Rapoport et al .. 1974). Nonetheless, the clinical implications of mild but long-term changes in the cardiovascular system are not known (Brown eta/., 1984). Mild changes in blood pressure and heart rate have been observed after medication has been discontinued (Boileau, Ballard, Sprague, Sleator, & Massey, 1976). The effects of these changes must be studied, and until they have been thoroughly elaborated, caution must prevail in prescribing these drugs. In addition to these common side effects, several rare, but serious, physical problems can be caused by stimulants. For example, blood dyscrasias may result from long-term stimulant administration. Preliminary evidence from one study has linked Hodgkin's disease to stimulant use: More patients with Hodgkin's disease reported a history of amphetamine use than did nonpatient controls (Newell & Henderson, 1973). Differences in the characteristics and experiences of the two groups, not the history of amphetamine use, may have accounted for the finding. The onset of tics, similar to those characteristic of Tourette's syndrome, has been reported in children treated with methylphenidate and pemoline
(Sleator, 1980). This finding is not sufficient to infer cause and effect, but the onset of tics following stim~ ulant administration should be carefully monitored. One researcher reported grand mal seizures as a side effect of methylphenidate administration (Chamberlin, 1974). Although this may have been an idiosyncratic finding, practitioners and researchers should remain alert to this possible effect. Growth suppression and weight loss have been reported with dextroamphetamine and moderate to high doses of methylphenidate (Safer & Allen, 1973; Safer et al., 1972). The investigators later reported growth spurts when the children were taken off medication during school vacations (Safer, Allen, & Barr, 1975). Several researchers have used methodological and measurement issues to refute the finding of growth suppression (Sprague, 1977). In a study of prepubescent children, no height suppression occurred with moderate doses of methylphenidate (Kalachnik, Sprague, Sleator, Cohen, & Ullmann, 1981). However, more recent studies have associated stimulant therapy in adolescence with shorter adult stature (Loney, Whaley-Klahn, Ponto, & Adney, 1981). Although the height-suppressing potential of stimulants continues to be debated, the evidence of weight loss in children on stimulants has been more consistent (Roche, Lipman, Overall, & Hung, 1979). The effects of stimulants on height and weight may be due to the anorectic effect of the medication, but other theories have cited drug action on pituitary secretions or direct action on bone and cartilage development (Ross & Ross, 1982). Until the issues of growth suppression and weight loss are settled, Eisenberg's (1972) advice still seems wise: Children on stimulant medication should be monitored carefully for physical side effects.
Psychosocial The most often reported psychosocial side effects of stimulants are emotional changes. Many parents and children have reported increased sensitivity and emotionality in children taking stimulants (Gittelman-Klein et al., 1976), for example, an increase in observations of depression, fearfulness, anger outbursts, or dysphoria. These symptoms are highly distressing to families and have been reported by parents in our laboratory as reasons to discontinue the medication. Other parents have reported that the children become unemotional, passive, too quiet, or socially withdrawn (Gittelman-Klein et al., 1976), changes
NEUROPSYCHOLOGICAL EFFECfS OF STIMULANT MEDICATION
that are also alarming to families. The far less frequently reported symptoms of psychosis, such as hallucinations (Lucas & Weiss, 1971 ), usually abate when the medication is discontinued but are extremely upsetting to family members and persons at the child's school who observe this reaction. The professional labeling of a child has powerful effects. Children who are referred for stimulant treatment have been labeled as deviant by parents, teachers, and peers long before the children came to the attention of professionals, but the confirmation of a label by a mental health or medical professional who prescribes medication alters the label, changing it from "deviant" or "uncooperative" to "sick." Medication may lead to additional labeling and teasing by peers, and medication administration procedures may be the first information a new teacher receives about the child, biasing the teacher's first impressions. Parents and teachers may view children who are given medication as more seriously disturbed (Henker & Whalen, 1980) and less capable of taking an active part in solving their behavior problems (Borden, 1986) than are children who are not given medication. According to research with adults, altering labels and expectations may be quite difficult, despite observable behavior changes (Rosenhan, 1973). Perhaps more disconcerting is the influence that medication may have on the medicated child's beliefs. Researchers have hypothesized that placing a child on psychoactive medication may negatively influence self-efficacy (Henker & Whalen, 1980), and the children's external attributions for problem solutions may increase (Bugenthal, Whalen, & Henker, 1977). Increased external attributions may be related to the lack of maintenance of treatment gains. Internally attributed changes are believed to be better maintained than those attributed to external factors (Bandura, Jeffrey, & Gajdos, 1975), although the evidence for this claim has been disputed (Grimm, 1980). Concern has been voiced that children who receive stimulants to treat neuropsychological disorders may be at increased risk for addiction to prescribed drugs. Children who learn to rely on drugs to solve problems might continue to use drugs to cope with stress. Taking prescribed stimulant medication is often reinforced by adults as well as by the child's own positive behavior change, perhaps strengthening the use of medication to solve problems. Further, children who have learning and behavior problems and are rejected by peers may begin to abuse drugs or sell their medication to gain peer acceptance. Arguments against these concerns include the
465
children's documented dislike of the medication and their avoidance of taking it (Sleator, Ullmann, & von Neumann, 1982). Also, children have rarely reported craving or euphoric effects (Huey, 1985). Finally, the social stigma associated with psychoactive medication is quite aversive to these children. Recent studies of drug abuse in children who had previously taken stimulants have been plagued with methodological problems. However, the evidence from studies of ADD/H adolescents and adults has disconfirmed the fear that medicated children are more at risk of drug and alcohol abuse than unmedicated children (Henker, Whalen, Bugenthal, & Barker, 1981) or their normal peers (Gadow & Sprague, 1980). In fact, according to some empirical data, a positive clinical response to stimulants may be associated with a lower probability of drug and alcohol abuse than found in the general population (Blouin, Bomstein, & Trites, 1978; Loney, Kramer, & Milich, 1981). Despite the limited evidence that stimulant therapy increases the risk for later drug abuse, several issues qualify that risk. First, many aggressive children, including those diagnosed as conduct disordered, are at risk for juvenile delinquency and drug abuse. Drug abuse may in fact be a problem for aggressive children for whom medication is prescribed. However, prescribing stimulant medication is not likely to cause drug abuse. Second, as the use of stimulants for adolescents and adults increases, more occurrences of drug abuse may be observed. Tolerance to some of the physiological effects of stimulants has been observed in as few as 2 to 5 months following the commencement of pharmacotherapy (Allen & Safer, 1979; Weiss, Kruger, Danielson, & Elman, 1975), and withdrawal of the medication has caused side effects, including severe depression (Brown, Borden, Spunt, & Medenis, 1985). Thus, addiction to prescribed stimulants may indeed be occurring, and psychological dependence on the drug may develop in some patients. For example, Rosen, O'Leary, and Conway (1985) reported the case of a 12-year-old boy whose behavior had improved drastically when he was given methylphenidate. His behavior deteriorated rapidly when medication was discontinued but again improved when a placebo was introduced. Although psychological dependence differs from the physical drug dependence defined in DSM-III (American Psychiatric Association, 1980), this child's psychological dependence on stimulants was maladaptive, and he required a program of attribution retraining before the medication could be discontinued. Psychological as well as physical addiction to medication should be
466
CHAPTER 24
carefully monitored, and we need interventions to curtail these complications.
Societal A final concern about the use of stimulants relates to the use of medication as a form of social control. Stimulants prescribed incorrectly by misdiagnosis may mask the symptoms of the true problems, thus preventing proper treatment (Volkmar, Hoder, & Cohen, 1985). Even when prescribed appropriately, unusually high doses, which subdue social behavior, often have been used. This suppression of behavior, referred to as a "chemical straightjacket" (Ross & Ross, 1982), has become a greater concern as the reports of widely ranging doses increase (American Academy of Pediatrics, 1975). Controlled, apathetic behavior in children is intolerable to some adults and to some researchers (Bosco & Robin, 1980). Others argue that this "control" is really increased freedom for a child who has been incapable of prolonged impulse control and attention to stimuli (Gittelman-Klein & Klein, 1975). Consistent with legal and ethical principles, the least restrictive alternative should be used in treatment. In the case of medication, this implies titrating dosages to the lowest effective level for each child. In addition, it implies that a course of medication should be used only when other treatments have failed and for the shortest time possible.
Compliance Although the encouraging findings on the shortterm efficacy of stimulants are not matched by data on the long-term outcome of stimulants (Weiss, 1983), many of the poor long-term outcomes may be due to a failure of patients and families to adhere to the treatment procedures. One type of nonadherence-attrition-is the premature discontinuation of treatment. A second type is the alteration of drugadministration procedures (e.g., increasing or decreasing the dose level). Both types are common in pediatric medicine (Sleator, 1985) and have been documented in ADD/H children treated with stimulants (Brown, Borden, & Clingerman, 1985; Brown, Borden, Wynne, Clingerman, & Spunt, 1987; Firestone, 1982). To maintain treatment gains, a long-term course of pharmacotherapy is often indicated. However, in Firestone's (1982) pioneering study of adherence, 20% of his patients had discontinued treatment by the 4th month. By the end of the lOth month, this figure
had increased to nearly 50%. Most important, fewer than I 0% of these families consulted a project staff member before they stopped giving the medication. In a study in which compliance was monitored through pill counts, the 34 subjects who completed a 3-month treatment program did not take an average of at least 25% of the prescribed dosages (Brown et al., 1987). Because of nonadherence, attempts to determine the efficacy of these medications in clinical settings and research trials may be yielding spurious results. Obviously, if a child is not taking medication appropriately, it is difficult to evaluate the efficacy of the chemical itself. The long-term outlook for children on stimulants might well be improved if adherence were improved (Firestone, 1982; Sleator, 1985). Investigators have identified patients at risk for high levels of nonadherence (Brown et al., 1987; Firestone, 1982). Identification of high-risk patients combined with an examination of the causes of nonadherence might help in the design of interventions to improve adherence rates. The many hypotheses about the causes of nonadherence include the children's dislike of the medication (Sleator et al., 1982). When we have asked children in our laboratory why they dislike the medication, some have mentioned side effects, such as stomachaches, or said that "it makes my head feel funny." Others have complained about difficulty swallowing the pill or remembering to take it. Still others have complained about the social ramifications, particularly when they have to take the medication at school. In addition to direct teasing by peers, these children probably recognize more subtle changes in how others view or interact with them (Barkley et al., 1985). Low adherence rates are also related to parents' concerns about having their children take medication for long periods (Whalen, Henker, & Hinshaw, 1985); they worry about negative side effects, including addiction and health problems. The parents' feelings about the medication, however, are probably not independent of their children's dislike of it. A third reason for poor adherence may be related to the behavior of the child's physician, neuropsychologist, or psychotherapist. The child and the family may not feel close enough to the professional staff to express their concerns about treatment. In addition, some professionals fail to monitor adherence (Solomons, 1973). Another factor in nonadherence is the intrusive routine of administering stimulants. Sometimes, when medication is prescribed, the length of the therapy is left open. Having no end in sight may further
NEUROPSYCHOLOGICAL EFFECTS OF STIMULANT MEDICATION
discourage the children and their parents. Finally, medication may communicate to family members that a child's symptoms are out of the child's or the family's control (Borden, 1986). The belief in an inability to effect improvements may be considered a form of "learned helplessness," which can compound a family's demoralization. The incomplete results of treatment, the inability to envision meeting hoped-for goals within a reasonable time, and the lack of a sense of control over symptoms may act synergistically, leaving families too discouraged to follow through on treatment plans. Adherence must be assessed before unambiguous conclusions about medication efficacy may be drawn. Attending to parents as well as children in developing therapeutic rapport and providing support to parents and children will help them maintain adherence. In addition, families must be given adequate information about the treatment and provided with strategies for overcoming the social and pragmatic obstacles to taking the medication as prescribed. Finally, the practitioner must monitor adherence carefully before evaluating the efficacy of the therapy. In general, however, the more valid the monitoring approach, the more intrusive and costly it will be (Sleator, 1985).
Summary and the Direction of Future Research The past decade has witnessed the marked expansion of well-controlled empirical research in the use of stimulants for children. The clinical community's enthusiasm for the increased use of stimulants has unfortunately been accompanied by some instances of overuse and misuse of stimulants for pediatric populations. The stimulants of choice are dextroamphetamine, methylphenidate, and pemoline. The short-term efficacy of these drugs has been thoroughly documented, although the long-term benefits of stimulants remain uncertain. Research is sorely needed on the relationship of dosage, plasma level, and behavioral response. The specificity of various symptom responses across different dosages also begs for investigation. The treatment of ADDIH with psychostimulants is probably the best documented and researched treatment in all of child psychiatry. According to most of the studies reviewed, stimulants most influence the laboratory measures of attention and impulsivity, as well as behavioral ratings and observations in ADD/H children. Few studies, however,
467
have demonstrated their effectiveness on tasks of higher-order information processing. Clearly, more research is needed with ADD/H children to isolate the specific cognitive processes that are affected by psychostimulants. Although the studies with learning-disabled children have been few and have not proven to be particularly promising, this area is particularly ripe for future investigation. We know little about the efficacy of stimulants for children with specific learning disabilities, such as arithmetic, writing, and spelling disabilities, disorders for which stimulants have proven effective in ADD/H samples. A careful examination of studies in which ADD/H children have participated indicates that subjects have been quite heterogeneous in degree of aggression and level of intellectual functioning (Gadow, 1985b; Rapoport, 1983). Because many children with aggressive symptomatology and children of differing levels of intelligence have proven to be positive responders to stimulants, it is reasonable to hypothesize that stimulants may prove to be a viable therapeutic modality for mentally retarded and conduct-disordered children. Nonetheless, research on the use of stimulants with these two populations is scant. This notion of diagnostic specificity also has implications for examining stimulant responses in subtypes of ADDIH children. For example, future research efforts might focus on comparing the responses in aggressive and nonaggressive ADD/H children; other studies might evaluate drug response in subjects of differing degrees of intellectual functioning. The clinical use of stimulants with adolescents, a new avenue of interest in the past several years, has demonstrated promise. The use of stimulants for adolescents has definitely increased (Safer & Krager, 198~). despite the very few research studies with this population. More systematic research in this area remains a high priority for pediatric psychopharmacology. Most of the well-controlled studies with stimulants have incorporated behavioral and cognitive variables as dependent measures. Little is known, however, about psychophysiological responses to stimulants. In fact, the specific actions of stimulants on central and autonomic nervous system variables remain unclear. With the development of new psychophysiological instruments, this area is destined for marked expansion in the late 1980s. Although it was once believed that stimulants exert little influence in the academic domain, recent studies have been more encouraging. Much more systematic research is needed, particularly well-controlled, long-term studies of stimulant drug effects on
468
CHAPTER 24
academic tasks. Although we have finally learned to evaluate appropriately the response to stimulants, undue optimism will not be warranted until we have evaluated long-term efficacy in the academic as well as the behavioral domain. Important treatment decisions such as when to initiate or discontinue medication have been the topic of much clinical and research literature, but no studies have yielded definitive data. Treatment decisions continue to be chiefly clinical decisions, and a clinical trial is still the practitioner's best tool. Unfortunately, the practitioner has little qualifiable information about terminating stimulant treatment for a child. In fact, only one rating scale allows the practitioner to document side effects systematically, thus permitting the quantification of such effects (Barkley, 1981). Although we have used this instrument with considerable success in our laboratory, additional scales with adequate reliability and validity will be necessary for the careful monitoring of possible adverse drug effects in children. Clearly, we need systematic study of individual variables in adverse effects and in the prediction of responses. The limited generalizability of stimulant drugs, along with the realization that children may need to be maintained on stimulants indefinitely to sustain desired clinical effects, has been particularly disappointing to researchers and practitioners alike. In fact, these drawbacks have caused some practitioners to be unduly pessimistic about the use of stimulants. As Gittelman ( 1983) pointed out, though, no medical therapy ameliorates a disease or completely reverses symptoms to the extent that the symptoms are no longer visible when pharmacotherapy is discontinued. Thus, before indicting simulants as ineffective, the practitioner must consider the alleviation of symptoms the only reasonable expectation. Although we expected combinations of stimulants and nonsomatic therapies to prove most efficacious and thereby lead to greater generalization, this has not been borne out (Brown, Wynne, & Medenis, 1985; Brown et al., 1986). In comparison with behavioral and cognitive therapies, stimulants are in fact superior. Thus, as Gittelman (1983) observed, until we find actual cures for attentional deficits and behavioral difficulties, the use of stimulants for symptom alleviation is probably "not a bad bargain" (p. 446). Despite the particularly encouraging effects of stimulants in the short-term management of children with behavioral and learning disorders, the possible long-term adverse effects on growth and learning, as well as on self-concept, are of some concern. This area is particularly interesting and merits intensive investigation. An inherent problem in the systematic
study of the long-term effects of stimulants is the obvious ethical limitations in placing children in nomedication or placebo-control groups for long periods. Collaboration among investigators in research laboratories throughout the country may be one answer to this dilemma and one that will help us determine any possible long-term adverse effects in the physiological or the psychosocial domain. Pending further, more definitive studies, clinicians and parents must continue to struggle with many unanswered questions about stimulant medication. Until we have more definitive answers about long-term effects, research remains a high priority, particularly joint research by the psychologist and the physician. ACKNOWLEDGMENTS
This chapter was supported in part by BRSG S07 RR05364 from the Biomedical Research Supported Grant Program, Division of Research Resources, National Institutes of Health, awarded to R.T.B. The authors are indebted to Marie. Morgan for her careful editing of the manuscript and to Martha Hagan for her skill in typing the manuscript. The authors are also grateful to Kenneth Gadow for making available to them so many reprints of his work. Our appreciation also goes to the Emory University School of Medicine, Behavioral and Developmental pediatric residents and child psychiatry fellows as well as Sandra Sexson for their thoughts and clinical insights, which in part shaped some of our thinking and views in preparing the chapter. The opinions expressed herein, however, are those of the authors, and they are not necessarily shared by others working in the field.
References Abikoff, H., & Gittelman, R. (1985). Hyperactive children treated with stimulants: Is cognitive training a useful adjunct? Archives of General Psychiatry, 42, 953-961. Abikoff, H., Gittelman-Klein, R., & Klein, D. F. (1977). Validation of a classroom observation code for hyperactive children.
Journal of Consulting and Clinical Psychology, 45, 772783. Adams, W. ( 1982). Effect of methylphenidate on thought processing time in children. Developmental and Behavioral Pediatrics, 3, 133-135. Allen, R. P., & Safer, D. (1979). Long-term effects of stimulant therapy for HA children: Risk benefits analysis. In M. J. Cohen (Ed.), Drugs and the special child (pp. 189-202). New York: Gardner.
NEUROPSYCHOLOGICAL EFFECTS OF STIMULANT MEDICATION Aman, M. G. (1982a). Psychotropic drugs in the treatment of reading disorders. In R.N. Malatesha & P. G. Aaron (Eds.), Reading disorders: Varieties and treatments (pp. 453-471 ). New York: Academic Press. Aman, M. G. (1982b). Stimulant drug effects in developmental disorders and hyperactivity-Toward a resolution of disparate findings. Journal of Austism and Developmental Disorders, 12, 385-398. Aman, M. G., & Sprague, R. L. (1974). The state-dependent effects of methylphenidate and dextroamphetamine. Journal of Nervous and Mental Disease, 158, 268-279. Aman, M.G., & Werry, J. S. (1982). Methylphenidate and diazepam in severe reading retardation. Journal of the American Academy of Child Psychiatry, 1, 31-37. American Academy of Pediatrics, Council on Child Health. ( 1975). Medication for hyperkinetic children. Pediatrics, 55, 560-561. American Psychiatric Association. ( 1980). Diagnostic and statistical manual of mental disorders (3rd ed.). Washington, DC: Author. Arnold, L. E., Christopher, J., & Huestis, R. (1978). Methylphenidate vs. dextroamphetamine vs. caffeine in minimal brain dysfunction. Archives of General Psychiatry, 35, 463-473. Bandura, A., Jeffrey, R. W., & Gajdos, E. (1975). Generalizing change through participant modeling with self-directed mastery. Behavior Research and Therapy, 13, 141-152. Barcai, A. ( 1971). Predicting the response of children with learning disabilities and behavior problems to dextroamphetamine sulfate. Pediatrics, 47, 73-80. Barkley, R. A. (1976). Predicting the response of hyperkinetic children to stimulant drugs: A review. Journal of Abnormal Child Psychology, 4, 327-348. Barkley, R. A. (1977). The effects of methylphenidate on various measures of activity level and attention in hyperkinetic children. Journal of Abnormal Child Psychology, 5, 351-369. Barkley, R. A. (1981). Hyperactive children: A handbook for diagnosis and treatment. New York: Guilford Press. Barkley, R. A., Karlsson, J., Pollard, S., &Murphy, I. V. (1985). Developmental changes in the mother-child interactions of hyperactive boys: Effects of two dose levels of Ritalin. Journal of Child Psychology and Psychiatry, 24, 705-715. Blouin, A. G. A., Bomstein, R. A., & Trites, R. L. (1978). Teenage alcohol use among hyperactive children: A five-year follow-up study. Journal of Pediatric Psychology, 3, 188194. Blunden, D., Spring, C., & Greenberg, L. (1974). Validation of the classroom behavior inventory. Journal ofConsulting and Clinical Psychology, 42, 84-88. Boileau, R. A., Ballard, J. E., Sprague, R. L., Sleator, E. K., & Massey, B. H. (1976). Effect of methylphenidate on cardiorespiratory responses in hyperactive children. Research Quarterly, 47, 590-596. Borden, K. A. (1986). Attributional outcomes: The subtle messages of treatments for attention deficit disorder. Unpublished doctoral dissertation, University of Illinois, Urbana-Champaign. Bosco, J. J., & Robin, S. S. (1980). Hyperkinesis: Prevalence and treatment. In C. K. Whalen & B. Henker (Eds. ), Hyperactive children (pp. 173-187). New York: Academic Press.
469
Bradley, C. (1937). The behavior of children receiving Benzedrine. American Journal of Psychiatry, 94, 577-585. Brown, R. T., Borden, K. A., & Clingerman, S. R. (1985). Adherence to methylphenidate therapy in a pediatric population: A preliminary investigation. Psychopharmacology Bulletin, 21, 28-36. Brown, R. T., Borden, K. A., Spunt, A. L., & Medenis, R. (1985). Depression following pemoline withdrawal in a hyperactive child [letter]. Clinical Pediatrics, 24, 174-175. Brown, R. T., Borden, K. A., Wynne, M. E., Clingerman, S. R., & Spunt, A. (1987). An examination of compliance with pharmacological and cognitive treatments for attention deficit disorder. American Academy ofChild and Adolescent Psychiatry, 26, 521-526. Brown, R. T., Borden, K. A., Wynne, M.A., Schlesser, R., & Clingerman, S. R. (1986). Methylphenidate and cognitive therapy with ADD children: A methodological reconsideration. Journal of Abnormal Child Psychology, 4, 481-497. Brown, R. T., & Sexson, S. B. (1986, August). Methylphenidate in the treatment of ADD adolescents. Paper presented at the annual meeting of the American Psychological Association. Washington, DC. Brown, R. T., & Sleator, E. K. (1979). Methylphenidate in hyperkinetic children: Differences in dose effects on impulsive behavior. Pediatrics, 64, 408-411. Brown, R. T., Slimmer, L., & Wynne, M. E. (1984). How much stimulant medication is appropriate for hyperactive school children? Journal of School Health, 54, 128-130. Brown, R. T., Wynne, M. E., & Medenis, R. (1985). Methylphenidate and cognitive therapy: A comparison of treatment approaches with hyperactive boys. Journal ofAbnormal Child Psychology, 13, 69-87. Bryan, T. H., & Bryan, J. H. (1978). Understanding learning disabilities. Palo Alto, CA: Alfred. Bugenthal, D. B., Whalen, C. K., & Henker, B. (1977). Causal attributions of hyperactive children and motivational assumptions of two behavior change approaches: Evidence for an interactionist position. Child Development, 48, 874-884. Campbell, S. B., Endman, M. W., & Bemfeld, G. (1977). A three-year follow-up of hyperactive preschoolers into elementary school. Journal ofChild Psychology and Psychiatry, 18. 239-249. Cantwell, D. P. (1979). Use of stimulant medication with psychiatrically disordered adolescents. Adolescent Psychiatry, 7, 375-388. Cantwell, D. P. (1985). Pharmacotherapy of ADD in adolescence: What do we know, where should we go, how should we do it? Psychopharmacology Bulletin, 21, 251-257. Cantwell, D.P., & Carlson, G. A. (1978). Stimulants. In J. S. Werry (Ed.), Pediatric psychopharmacology: The use of behavior-modifying drugs in children (pp. 171-207). New York: Brunner/Mazel. Chamberlin, R. W. (1974). Convulsions and Ritalin? [Letter]. Pediatrics, 54, 658-659. Chan, Y.-P. M., Swanson, J. M., Soldin, S. S., Thiessen, J. J., MacLeod, S. M., & Logan, W. (1983). Methylphenidate hydrochloride given with or before breakfast. II. Effects on plasma concentration of methylphenidate and ritalinic acid. Pediatrics, 72, 56-59.
470
CHAPTER 24
tion ofthe attentional deficit of hyperactive children. In G. A. Charles, L., Schain, R., & Zelniker, T. (1981). Optimal dosages Hale & M. Lewis (Eds.), Attention and the development of of methylphenidate for improving the learning and behavior cognitive skills (pp. 173-247). New York: Plenum Press. of hyperactive children. Developmental and Behavioral PeDulcan, M. K. (1985). Attention deficit disorder: Evaluation and diatrics, 2, 78-81. treatment. Pediatric Annals, 14, 383-398. Clampit, M. K .• & Pirkle, J. B. (1983). Stimulant medication and Dulcan, M. K. (1986). Comprehensive treatment of children and the hyperactive adolescent: Myths and facts. Adolescence, adolescents with attention deficit disorders: The state of the 18, 8ll-822. art. Clinical Psychology Review, 6, 539-569. Close, J. (1973). Scored neurological examination. PsychopharDykman, R., Ackerman, P. T., & Oglesby, C. (1979). Selective macology Bulletin, 9, 142-148. and sustained attention in hyperactive, learning-disabled, and Conners, C. K. (1969). A teacher rating scale for use in drug normal boys. Journal of Nervous and Mental Disease, 167, studies with children. American Journal of Psychiatry, 126, 288-297. 152-156. Edelbrock, C., & Rancurello, M. (1985). Assessment of childConners, C. K. (1971). Recent drug studies with hyperkinetic hood hyperactivity: Rating scales and their applications. children. Journal of Learning Disabilities, 4, 476-483. Clinical Psychology Review, 5, 429-445. Conners, C. K. (1972). Psychological effects of stimulant drugs in Eisenberg, L. ( 1972). The hyperkinetic child and stimulant drugs. children with minimal brain dysfunction. Pediatrics, 49, New England Journal of Medicine, 287, 249-250. 702-708. Conners, C. K. (1973, September). A Clinical comparison of Eisenberg, L., Lachman, R., Moiling, P. A., Lockner, A., Mizelle, J. D., & Conners, C. K. (1963). A psychopharmethylphenidate and placebo in the treatment of children macological experiment in a training school for delinquent with minimal brain dysfunction. Paper presented at the annual meeting of the American Psychological Association, boys: Methods, problems, findings. American Journal of Orthopsychiatry, 33, 431-447. Montreal, Canada. Conners, C. K., & Eisenberg, L. (1963). The effects of meth- Firestone, P. (1982). Factors associated with children's adherence to stimulant medication. American Journal of Orthopsychiaylphenidate on symptomatology and learning in disturbed try, 52, 447-457. children. American Journal of Psychiatry, 120, 458-464. Conners, C. K., & Taylor, E. (1980). Pemoline, methylphenidate, Fish, B. (1971). The "one child, one drug" myth of stimulants in hyperkinesis. Archives of General Psychiatry. 25, 193-203. and placebo in children with minimal brain dysfunction. ArFisher, K. C., & Wilson, W. P. (1971). Methylphenidate and the chives of General Psychiatry, 37, 922-930. hyperkinetic. Diseases of the Nervous System, 32, 695-698. Coons, H. W., Klorman, R., & Borgstedt, A. D. (1987). Enhancing effects of methylphenidate on sustained attention and Gadow, K. D. (I 981). Drug therapy for hyperactivity: Treatment procedures in natural settings. InK. D. Gadow & J. Loney event-related potentials of adolescent patients with attention (Eds. ), Psychosocial aspects of drug treatment for hyperacdeficit disorders. Psychophysiology, 24, 572-573. tivity (pp. 325-378). Boulder, CO: Westview. Coons, H. W., Peloquin, L., & Klorman, R. (1981). Effect of methylphenidate on young adults' vigilance and event related Gadow, K. D. (1983). Pharmacotherapy for behavior disorders: Typical treatment practices. Clinical Pediatrics, 22, 48-53. potentials. Electroencephalography and Clinical NeuGadow, K. D. (1985a). Prevalence and efficacy of stimulant drug rophysiology, 51, 373-387. use with mentally retarded children and youth. PsychopharCutler, M., Little, J. W., & Strauss, A. A. (1940). The effect of macology Bulletin, 21, 291-303. Benzedrine on mentally deficient children. American Journal Gadow, K. D. (1985b). Relative efficacy of pharmacological, of Mental Deficiency, 45, 49-65. behavioral, and combination treatments for enhancing acaDonchin, E., Ritter, W., & McCallum, W. C. (1978). Cognitive demic performance. Clinical Psychology Review, 5, 513psychophysiology: The endogenous components of the ERP. 533. In E. Callaway, P. Tueting, & S. H. Koslow (Eds.), EventGadow, K., & Sprague, R. L. (1980, September). An anterospecrelated potentials in man (pp. 349-411). New York: Acative follow-up of hyperactive children into adolescence: Licit demic Press. and illicit drug use. Paper presented at the annual meeting of Donnelly, M., & Rapoport, J. L. (1985). Attention deficit disorthe American Psychological Association, Montreal, Canada. ders. In J. M. Wiener (Ed.), Diagnosis and psychopharmacology of childhood and adolescent disorders (pp. 179- Gadow, K. D., & Swanson, H. L. (1985). Assessing drug effects on academic performance. Psychopharmacology Bulletin, 197). New York: Wiley. 21, 877-886. Douglas, V.I. (1980). Treatment and training approaches to hyGadow, K. D., Torgesen, J., Greenstein, J., & Schell, R. (1986). peractivity: Establishing internal or external control. In C. K. Learning disabilities. In M. Hersen & S. E. Breuning (Eds.), Whalen & B. Henker (Eds.), Hyperactive children: The soPharmacological and behavioral treatment: An integrated cial ecology of identification and treatment (pp. 283-317). approach (pp. 149-196). New York: Wiley. New York: Academic Press. Gan, J., & Cantwell, D. (1982). Dosage effects of methylpheniDouglas, V. I., Barr, R. G., O'Neill, M. E., & Britton, B. G. date on paired associate learning: Positive/negative placebo (1986). Short-term effects of methylphenidate on the cogresponders. Journal of the American Academy of Child P.rynitive, learning and academic performance of children with attention deficit disorder in the laboratory and classroom. chiatry. 21. 237-242. Garfinkel, B. D. (1986). RecFnt developments in attention deficit Journal of Child Psychology and Psychiatry, 27. 191-21 I. Douglas, V. 1., & Peters, K. G. (1979). Toward a clearer definidisorder. Psychiatric Annals, 16, 11-15.
NEUROPSYCHOLOGICAL EFFECTS OF STIMULANT MEDICATION Garfinkel, B. D., Brown, W. A., Klee, S. H., Braden, W., Beauchesne, H., & Shapiro, S. K. (in press). Neuroendocrine and cognitive responses to amphetamine in adolescents with a history of ADD. Journal of Clinical Phormacology. Gilman, A. G., Goodman, L. S., &Gilman, A. (1980). Goodman and Gilman's the pharmacologic basis of therapeutics (6th ed.). New York: Macmillan Co. Giuelman, R. (1980). Indications for the use of stimulant treatment in learning disorders. Journal of the American Academy of Child Psychiatry, 19, 623-636. Giuelman, R. (1982). Controlled study of methylphenidate in combination with academic instruction. Psychophormacology Bulletin, 18, 112-113. Gittelman, R. (1983). Hyperkinetic syndrome: Treatment issues and principles. In M. Rutter (Ed.), Developmental neuropsychiatry (pp. 437-449). New York: Guilford Press. Gittelman, R., Klein, D. F., & Feingold, I. (1983). Children with reading disorders. II. Effects of methylphenidate in combination with reading remediation. Journal of Child Psychology and Psychiatry, 24, 193-212. Gittelman-Kiein, R., & Klein, D. F. (1975). Are behavioral and psychometric changes related in methylphenidate-treated, hyperactive children? International Journal of Mental Health, 4. 182-198. Gittelman-Klein, R., & Klein, D. F. (1976). Methylphenidate effects in learning disabilities: Psychometric changes. Archives of General Psychiatry, 33, 655-664. Gittelman-Klein, R., Klein, D., Katz, S., Saraf, K., & Pollack, E. (1976). Comparative effects of methylphenidate and thioridazine in hyperkinetic children. Archives of General Psychiatry, 33, 1217-1231. Grimm, L. G. (1980). The maintenance of self- and drug-attributed behavior change: A critique. Journal of Abnormal Psychology, 89, 282-285. Gross, M. D., & Wilson, N. C. (1974). Minimal brain dysfunction. New York: Brunner/Mazel. Halliday, R. A., Rosenthal, J. H., Naylor, H., & Callaway, E. (1976). Average evoked potential predictors of clinical improvement in hyperactive children treated with methylphenidate: An initial study and replication. Psychophysiology, 13. 429-440. Henker, B., & Whalen, C. K. (1980). The changing faces of hyperactivity: Retrospect and prospect. In C. K. Whalen & B. Henker (Eds. ), Hyperactive children (pp. 321-363). New York: Academic Press. Henker, B., Whalen, C. K., Bugenthal, D. B., & Barker, C. ( 1981 ). Licit and illicit drug patterns in stimulant treated children and their peers. In K. D. Gadow & J. Loney (Eds.), Psychosocial aspects of drug treatment for hyperactivity (pp. 443-462). Boulder, CO: Westview. Hinshaw, S. P., Henker, B., & Whalen, C. K. (1984). Selfcontrol in hyperactive boys in anger-inducing situations: Effects of cognitive-behavioral training and methylphenidate. Journal of Abnormal Child Psychology. 52, 739-749. Hoy, E., Weiss, G., Minde, K., &Cohen, N. (1978). The hyperactive child at adolescence: Cognitive, emotional, and social functioning. Journal ofAbnormal Child Psychology, 6, 311324. Huddleston, W., Staiger, R. C., Frye. R., Musgrave, R. S., &
471
Stritch, T. (1961). Deanol as aid in overcoming reading retardation. Clinical Medicine, 8, 1340-1342. Huessey, H. R., Cohen, S.M., Blair, C. L., & Wood, P. (1979). Clinical explorations in adult minimal brain dysfunction. In L. Bellak (Ed.), Psychiatric aspects of minimal brain dysfunction (pp. 19-36). New York: Grune & Stratton. Huey, L. Y. (1985). Attention deficit disorders. In R. Michaels, J. 0. Cavenar, K. H. Brodie, A.M. Cooper, S. B. Guze, L. L. Judd, G. L. Klennan, & A. J. Solnit (Eds.), Psychiatry (Vol. 3, pp. 1-31). Philadelphia: Lippincott. Kagan, J. (1965). Reflection-impulsivity and reading ability in primary grade children. Child Development, 36, 609-628. Kalachnik,J. E.,Sprague, R. L.,Sleator, E. K.,Cohen,M. N.,& Ullmann, R. K. (1981). Use of the RWT stature prediction formula in monitoring growth suppression effects in methylphenidate. Unpublished manuscript, University of Illinois. Kaufman, A. S. (1979). Intelligence testing with the WISC-R. New York: Wiley. Kavale, K. (1982). The efficacy of stimulant drug treatment for hyperactivity: A meta-analysis. Journal of Learning Disabilities, 15, 280-289. Kinsboume, M. (1973). Minimal brain dysfunction as a neurodevelopmentallag. Annals of the New York Academy of Sciences, 205, 263-273. Kinsboume, M., & Swanson, J. M. (1980). Evaluation of symptomatic treatment of hyperactive behavior by stimulant drugs. In R. Knights & D. Bakker (Eds.), Treatment ofhyperactive and learning disordered children (pp. 207-217). Baltimore: University Park Press. Klee, S. H. ( 1986). The clinical psychological evaluation of attention deficit disorder. Psychiatric Annals, 16, 43-46. Klein, D. F., Gittelman, R., Quitkin, F., & Rifkin, A. (1980). Diagnosis and drug treatment of psychiatric disorders: Adults and children. Baltimore: Williams & Wilkins. Klonnan, R. (1986). Attention-deficit disorder in adolescents. In R. A. Feldman & A. R. Stiffman (Eds.), Advances in adolescent mental health. Greenwich, Cf: JAI Press. Korey, S. R. (1949). The effects of Benzedrine sulfate on the behavior of psychopathic and neurotic juvenile delinquents. Psychiatric Quarterly, 18, 127-137. Laufer, M. W., Denhoff, E., & Solomons, G. (1957). Hyperkinetic impulse disorder in children's behavior problems. Psychosomatic Medicine, 19, 38-49. Lerer, R. J., & Lerer, P. (1977). Response of adolescents with minimal brain dysfunction to methylphenidate. Journal of Learning Disabilities, 10, 35-40. Lezak, M. D. (1983). Neuropsychological assessment. New York: Oxford University Press. Loiselle, D., Stamm, J. S., Maitinsky, S., & Whipple, S. C. (1980). Evoked potential and behavioral signs of attention dysfunctions in hyperactive boys. Psychophysiology, 17, 193-201. Loney, J. (1986). Predicting stimulant drug response among hyperactive children. Psychiatric Annals, 16, 16-19. Loney, J., Kramer, J., & Milich, R. S. (1981). The hyperactive child grows up: Predictors of symptoms, delinquency and achievement at follow-up. In K. D. Gadow & J. Loney (Eds.), Psychosocial aspects of drug treatment/or hyperactivity (pp. 381-415). Boulder, CO: Westview.
472
CHAPTER 24
Loney, J., Whaley-Klahn, M. A., Ponto, L. B., & Adney, K. (1981). Predictors of adolescent height and weight in hyperkinetic boys treated with methylphenidate. Psychopharmacology Bulletin, 17, 132-134. Lucas, A. R., & Weiss, M. (1971). Methylphenidate hallucinosis. Journal of the American Medical Association, 217, 10791081. MacKay, M. C., Beck, L., & Taylor, R. (1973). Methylphenidate for adolescents with minimal brain dysfunction. New York State Journal of Medicine, 73, 550-554. Maletzky, B. M. (1974). o-Amphetamine and delinquency: Hyperkinesis persisting? Diseases of the Nervous System, 35, 543-547. Michael, R. L., Klorman, R., & Salzman, L. F. (1981). Normalizing effects of methylphenidate on hyperactive children's vigilance performance and evoked potentials. Psychophysiology, 17, 193-201. Mirsky, A. F., & Rosvold, H. E. (1963). Behavioral and physiological studies in impaired attention. In Z. Votava, M. Horvath, & 0. Vinar (Eds.), Psychopharmacological methods: Proceedings of a symposium on the effects of psychotropic drugs on higher nervous activity (pp. 302-315). Elmsford, NY: Pergamon Press. Molitch, M. J., & Sullivan, J.P. (1937). The effect of Benzedrine sulfate on children taking the new Stanford Achievement Test. American Journal of Onhopsychiatry, 7, 519-522. Newell, G. R., & Henderson, B. E. (1973). Case-control study of Hodgkin's disease. I. Results of the interview questionnaire. Journal of the National Cancer Institute, 51, 1437-1441. O'Leary, K. D., Romanczyk, R. G., Kass, R. E., Dietz, A., & Santagrossi, D. (1971). Procedures for classroom observation of teachers and children. Unpublished manuscript, State University of New York, Stony Brook. Page, J. G., Bernstein, J. E., Janicki, R. S., & Michelli, P. A. (1974). A multi-clinic trial of pemoline in childhood hyperkinesis. In C. K. Conners (Ed.), Clinical uses of stimulant drugs in children (pp. 98-129). Amsterdam: Excerpta Medica. Pelham, W. E. (1980, September). Peer relationships in hyperactive children: Description and treatment effects. In R. Milich (Chair), Peer relationships among hyperactive children. Symposium conducted at the annual meeting of the American Psychological Association, Montreal, Canada. Pelham, W. E., Bender, M. E., Caddell, J., Booth, S., & Moorer, S. H. (1985). Methylphenidate and children with attention deficit disorder: Dose effects on classroom academic and social behavior. Archives of General Psychiatry, 42, 948-952. Pelham, W. E., & Murphy, H. A. (1986). Behavioral and pharmacological treatment of attention deficit and conduct disorders. In M. Hersen & S. E. Brenning (Eds.), Pharmacological and behavioral treatment: An integrative approach. New York: Wiley. Poling, A., & Brenning, S. E. (1983). Effects of methylphenidate on the fixed-ration performance of mentally retarded children. Pharmacology, Biochemistry, and Behavior, 18, 531544. Porrino, L. J., Rapoport, J. L., Behar, D., Ismond, D., & Bunney, W. E. (1983). A naturalistic assessment of the motor activity of hyperactive boys. II. Stimulant drug effects. Archives of General Psychiatry, 40, 688-693.
Rapoport,J. L. (1983). The use of drugs: Trends in research. In M. Rutter (Ed.), Developmental neuropsychiatry (pp. 385403). New York: Guilford Press. Rapoport, J., Quinn, P., Bradbard, G., Riddle, D., & Brooks, E. (1974). Imipramine and methylphenidate treatments of hyperactive boys. Archives of General Psychiatry, 30, 789793. Rapport, M.D., DuPaul, G. J., Stoner, G., Birmingham, B. K., & Masse, G. (1985b). Attention deficit disorder with hyperactivity: Differential effects of methylphenidate on impulsivity. Pediatrics, 76, 938-943. Rapport, M.D., Stoner, G., DuPaul, G. J., Birmingham, B. K., & Tucker, S. (1985). Methylphenidate in hyperactive children: Differential effects of dose on academic, learning, and social behavior. Journal of Abnormal Child Psychology, 13, 227-244. Roche, A. F., Lipman, R. S., Overall, J. E., & Hung, W. (1979). The effects of stimulant medication on the growth of hyperactive children. Pediatrics, 63, 847-850. Rosen, L.A., O'Leary, S. G., & Conway, G. (1985). The withdrawal of stimulant medication for hyperactivity: Overcoming detrimental attributions. Behavior Therapy, /6, 538544. Rosenhan, D. L. (1973). On being sane in insane places. Science, 179, 250-258. Rosenthal, R. H., & Allen, T. W. (1978). An examination of attention, arousal, and learning dysfunctions of hyperkinetic children. Psychological Bulletin, 85, 689-715. Ross, D. M., & Ross, S. A. (1982). Hvperactivity. New York: Wiley. Safer, D. J., & Allen, R. P. (1973). Factors influencing the suppressant effects of two stimulant drugs on the growth of hyperactive children. Pediatrics, 51, 660-667. Safer, D. J., & Allen, R. P. (1975a). Side effects from long-term use of stimulants in children. In R. Gittelman-Klein (Ed.), Recent advances in child psychopharmacology (pp. 109122). New York: International Arts and Science Press. Safer, D. J., &Allen, R. P. (1975b). Stimulanttreatmentofhyperactive adolescents. Diseases of the Nervous System, 36, 454457. Safer, D., Allen, R., & Barr, E. (1972). Depression of growth in hyperactive children on stimulant drugs. New England Journal of Medicine, 287, 217-220. Safer, D. J., Allen, R. P., & Barr, E. (1975). Growth rebound after terminati
NEUROPSYCHOLOGICAL EFFECTS OF STIMULANT MEDICATION
Schleifer, M., Weiss, G., Cohen, N., Elman, M., Cvejic, H., & Kruger, E. (1975). Hyperactivity in preschoolers and the effect of methylphenidate. American Journal of Orthopsychiatry, 45, 38-50. Shekim, W. 0., Dekirmenjlan, H., & Chapel, J. L. (1979). Urinary MHPH excretion in minimal brain dysfunction in older children. American Journal of Psychiatry. 136, 667-671. Sleator, E. K. ( 1980). Deleterious effects of drugs used for hyperactivity on patients with Gilles de Ia Tourette's syndrome. Clinical Pediatrics, 19, 453-454. Sleator, E. K. (1985). Measurement of compliance. Psychopharmacology Bulletin, 21. 1089-1093. Sleator, E. K., Ullmann, R. K., & von Neumann, A. (1982). How do hyperactive children feel about taking stimulants and will they tell the doctor? Clinical Pediatrics. 21. 474-479. Sleator, E. K., von Neumann, A., & Sprague, R. L. (1974). Hyperactive children: A continuous long-term placebo-controlled follow-up. Journal of the American Medical Association. 227. 316-327. Solanto, M. V. (1984). Neuropharmacological basis of stimulant drug action in attention deficit disorder with hyperactivity: A review and synthesis. Psychological Bulletin, 95, 387-409. Solomons, G. (1973). Drug therapy: Initiation and follow-up. Annals of the New York Academy of Sciences. 205, 335-344. Sprague, R. L. (1977). Psychopharmacotherapy in children. In M. F. MacMillan & S. Henao (Eds.), Child psychiatry: Treatment and research. New York: Brunner/Mazel. Sprague, R. L. (1983}. Behavior modification and educational techniques. In M. Rutter (Ed.), Developmental neuropsychiatry (pp. 404-421). New York: Guilford Press. Sprague, R. L., & Berger, B. D. (1980), Drug effects on learning performance: Relevance of animal research to pediatric psychopharmacology. In R. M. Knights & D. J. Bakker (Eds.), Rehabilitation, treatment and management of learning disorders (pp. 167-183). Baltimore: University Park Press. Sprague, R. L., & Sleator, E. K. (1977). Methylphenidate in hyperkinetic children: Differences in dose effects on learning and social behavior. Science, 198, 1274-1276. Stephens, R. S., Pelham, W. E., & Skinner, R. (1984). Statedependent and main effects of methylphenidate and pemoline on paired-associate learning and spelling in hyperactive children. Journal of Consulting and Clinical Psychology, 51,
104-113. Surwillo, W. W. (1980). Changes in the electroencephalogram accompanying the use of stimulant drugs (methylphenidate and dextroamphetamine) in hyperactive children. Biological Psychiatry, 12, 787-799. Swanson, J. M., & Kinsbourne, M. (1976). Stimulant-related state-dependent learning in hyperactive children. Science. 192, 1354-1357. Swanson, 1. M., & Kinsbourne, M. ( 1979). The cognitive effects of stimulant drugs on hyperactive (inattentive) children. In G. Hales & M. Lewis (Eds.), Attention and the development of cognitive skills (pp. 249-274). New York: Plenum Press. Swanson,J., Kinsbourne, M., Roberts, W., &Zucker, K. (1978). Dose response analysis of the effect of stimulant medication on the learning ability of children referred for hyperactivity. Pediatrics, 61, 21-29. Swanson, J., Sandman, C., Deutsch, C., & Baren, M. (1983). Methylphenidate hydrochloride given with or before break-
473
fast. I. Behavioral, cognitive and electrophysiologic effects. Pediatrics, 72, 49-55. Thorley, G. (1984). Review offollow-up and follow-back studies of childhood hyperactivity. Psychological Bulletin, 96, 116-
132. Ullmann, R. K., & Sleator, E. K. (1985). Attention deficit disordered children with or without hyperactivity: Which behaviors are helped by stimulants? Clinical Pediatrics, 24, 547-
552. Varley, C. K. (1983). Effects of methylphenidate in adolescents with attention deficit disorder. Journal of the American Academy of Child Psychiatry, 22, 351-354. Varley, C. K. (1985). A review of studies of drug treatment efficacy for attention deficit disorder with hyperactivity in adolescents. Psychopharmacology Bulletin. 21, 216-221. Varley, C. K., & Trupin, E. W. (1982). Double-blind administration of methylphenidate to mentally retarded children with attention deficit disorder: A preliminary study. American Journal of Mental Deficiency. 86, 560-566. Volkmar, F. R., Hoder, E. L., &Cohen, D. J. (1985). Inappropriate uses of stimulant medication. Clinical Pediatrics. 24,
127-130. Weingartner, H., Ebert, M. H., Mikkelsen, E. J., Rapoport, J. L., Buchsbaum, M.S., Bunney, W. E., & Caine, E. D. (1980). Cognitive processes in normal and hyperactive children and their response to amphetamine treatment. Journal of Abnormal Psychology, 89, 25-37. Weiss, G. (1983). Long-term outcome: Findings, concepts, and practical implications. In M. Rutter (Ed.), Developmental neuropsychiatry (pp. 422-449). New York: Guilford Press. Weiss, G., & Hechtman, L. T. (1979). The hyperactive child syndrome. Science, 205, 1348-1354. Weiss, G., Hechtman, L., Hopkins, J., Perlman, T., & Wenar, A. ( 1979). Hyperactives as young adults: A controlled-perspective ten-year follow-up of 75 children. Archives of General Psychiatry, 36, 675-681. Weiss, G., Kruger, E., Danielson, V., & Elman, J. (1975). Effects of long-term treatment of hyperactive children with methylphenidate. Canadian Medical Association Journal,
112. 159-165. Weiss, G., Minde, K., Douglas, V., Werry, J., & Sykes, D. (1971). Comparison of the effect of chloropromazine, dextroamphetamine, and methylphenidate on the behavior and intellectual functioning of hyperactive children. Canadian Medical Association Journal, 104, 20-25. Wender, P. H. (1971). Minimal brain dysfunction in children. New York: Wiley. Wender, P. H., Reimherr, F. W., & Wood, M.D. (1981). Diagnosis and drug treatment of attention deficit disorder (minimal brain dysfunction) in adults: A replication. Archives of General Psychiatry, 38, 449-456. Whalen, C. K., Collins, B. E., Henker, B., Alkus, S. R., Adams, D., & Stapp, J. (1978). Behavior observations of hyperactive children and methylphe11idate (Ritalin) effects in systematically structured classroom environments: Now you see them; now you don't. Journal ofPediatric Psychology, 3, 177-187. Whalen, C. K., Henker, B., & Dotemoto, S. (1981). Teacher response to the methylphenidate (Ritalin) versus placebo status of hyperactive boys in the classroom. Child Development, 52, 1005-1014.
474
CHAPTER 24
Whalen, C. K., Henker. B., & Finch, D. (1981). Medication effects in the classroom: Three naturalistic indicators. Journal of Abnormal Child Psychology, 9, 419-433. Whalen, C. K., Henker, B., & Hinshaw, S. P. (1985). Cognitivebehavioral therapies for hyperactive children: Premises, problems, and prospects. Journal of Abnormal Child Psychology, 13, 391-410. Whalen, C. K., & Henker, B. (1980). The social ecology ofpsychostimulant treatment: A model for conceptual and empirical analysis. InC. K. Whalen& B. Henker(Eds.),Hyperactive children: The social ecology of identification and treatment (pp. 3-51). New York: Academic Press. Whalen, C. K., & Henker, B. (1984). Hyperactivity and the attention deficit disorders: Expanding frontiers. Pediatric Clinics of North America, 31, 399-427.
Whitehouse, D .• Shah, U., & Palmer. F. B. (1980). Comparison of sustained-release and standard methylphenidate in the treatment of minimal brain dysfunction. Journal of Clinical Psychiatry, 41, 282-285. Woods, D. (1986). The diagnosis and treatment of attention deficit disorder-residual type. Psychiatric Annals, 16, 23-28. Yellin, A.M. (1986). Psychophysiological correlates of attention deficit disorder. Psychiatric Annals, 16, 29-34. Zahn, T. P., Rapoport, J. L., & Thompson, C. L. (1980). Autonomic and behavioral effects of dextroamphetamine and placebo in normal and hyperactive prepubertal boys. Journal of Abnormal Child Psychology, 8, 145-160. Zara, M. M. (1973). Effects of medication on learning in hyperactive four-year-old children. Dissertation Abstracts International, 34, 2407A.
25 Nonstimulant Psychotropic Medication Side Effects on Children's Cognition and Behavior MANUEL L. CEPEDA
Nonstimulant psychotropic medications are used to attenuate a limited number of target symptoms associated with childhood psychopathology. For some disorders, as the underlying disease process improves, cognitive performance also improves. The cognitive improvement is probably not a direct effect of the medication, however. In addition to the primary effect on behaviors targeted for improvement, all medications have side effects. These may affect behavior and cognition. This chapter summarizes the primary or desired effects of the major classes of nonstimulant psychotropic medications, and reviews the behavioral and cognitive side effects that might influence psychological testing.
Psychotropic Medication and Psychological Testing Those engaged in psychological testing occasionally obtain and interpret data from children using psychotropic medications. Most frequently, these are the stimulant medications covered in the preceding chapter by Brown and Borden. However, the major tranquilizers, anxiolytics, and antidepressants will also be encountered. There is always a concern that the behavioral and cognitive side effects of these medications may adversely influence the test results. This concern will be addressed for the more global psychological instruments administered to children,
MANUEL L. CEPEDA • Depanment of Psychiatry, University of South Alabama College of Medicine, Mobile, Alabama 36617.
such as Wechsler-type scales (e.g., WISC-R, WPPSI, KABC), Bender Gestalt, and achievement tests. It is unfair to the child who is helped by a medication to request arbitrarily of the physician or parent that the child be free of medication when tested. While on medication, the child may actually be giving a more useful picture of test performance. It is also misleading or a disservice to state in every psychological testing report based on data taken while the child was on medication that the report may be invalid, or not represent a true picture of the child's performance because of medications. Not every behavior that "interferes" during testing is due to a medication. Most medications, in routine doses, do not impair test performance. Some medications do have specific behavioral side effects that are easily recognizable. If the behavior of concern is not one of these specific side effects, the medication probably is not causing the problem. Behavioral observations made during testing may reflect that the child "did not seem to hear what was said," or "seemed in a fog," or "appeared sleepy and could not concentrate." Seldom are behaviors such as this a side effect of nonstimulant psychotropic medications. Dissociative states, situational stress, or oppositional behavior patterns far more often account for these common responses than does medication. Sometimes, it takes hours of diagnostic play therapy to discover the child's entire repertoire of mental mechanisms and behavioral responses. The psychometrician seeing the child only for testing cannot put the behaviors seen in a true context. Many educational and counseling personnel working with children are unfamiliar with medications. Some are philosophically opposed to drug use. All too quickly, 475
476
CHAPTER 25
behaviors or test results may be erroneously attributed to the influence of medication. This chapter is written to help place these issues into perspective and help clarify the specific concerns that should be placed in psychological test reports from a child-psychiatric perspective. It will also serve as a guide for discussing concerns about possible medication side effects with the prescribing physician. Hospitalized children have ptore severe symptoms and may be prescribed higher doses more quickly than would be given to outpatients. Those testing children on a hospital inpatient unit frequently talk with the referring physician when a clarification needs to be made about medication effects such as sedation. In the school or outpatient clinic setting, similar consultation should be available although the need may not be as frequent. For such a consultation, a parent should first give permission.
Psychopharmacology with Children
were being used to treat symptoms. No diagnostic nomenclature of childhood psychopathology adequately identified the children who might be responsive to medication. No treatment-responsive symptom is unique to a specific diagnosis. In contrast to the psychopharmacological treatment of adult-onset major mental disorders such as schizophrenia or the major affective disorders, childhood psychopharmacology was at this time in its infancy and quite confusing to mental health clinicians who saw children being treated for symptoms that may have responded to medication by physicians who still argued about the diagnosis. In 1973, a special issue of the Psychopharmacology Bulletin (National Institute of Mental Health, 1973) summarized the current needs in the field of childhood psychopharmacology and presented a standardized package of rating scales and data sheets that had been found valuable in clinical research. This brought much needed uniformity to the field of childhood psychopharmacology. This information has recently been updated (National Institute of Mental Health, 1985). Although improvements will continue, there is now a considerable literature on the reliability and validity of the psychopharmacology assessment rating scales. A vast improvement in the diagnostic nomenclature for children came with the publication of the Diagnostic and Statistical Manual of Mental Disorders, 3rd edition (DSM-lli; American Psychiatric Association, 1980). The DSM-lli has brought research-grade diagnostic criteria that can be applied uniformly by researchers and clinicians. It is now in widespread use. Refinements in the diagnostic categories, based on both clinical use and statistical studies, are now being made (American Psychiatric Association, 1985). The latest areas of research are looking at the relationships between drug dosage and plasma levels. Hopefully, these pharmacokinetic studies can be used to establish therapeutic blood ranges necessary for clinical response. Several texts summarize both the history of childhood psychopharmacology and its current use (Weiner, 1977, 1985; Werry, 1978).
The use of psychotropic medications in the treatment of childhood psychopathology began in the late 1930s with the stimulant medication Benzedrine. It was almost 15 years later, following the introduction of the antipsychotic& such as chlorpromazine (Thorazine) and thioridazine (Mellaril), that other drugs were administered for the control of childhood behavior disorders. Most of the early target behaviors involved psychomotor excitement, restless behavior, anxiety, and hyperactivity. Many of the initial drug trials involved children with a diagnosis of delinquent behaviors, brain damage, cerebral palsy, or mental retardation. Quickly, antianxiety medications such as hydroxyzine (Atarax) and later, the tricyclic antidepressants such as imipramine (Tofranil) were added. Most recently, the antimania drug l~thium and anticonvulsants are being tried in conjunction with childhood psychopathology. The field of childhood psychopharmacology has always been one of symptom treatment. Although many behavioral concomitants of specific diagnoses are responsive to medication, the underlying disease process has yet to be treated directly. This point distinguishes childhood psychopharmacology from the psychopharmacology Nonstimulant Psychotropic Drugs of many adult mental disorders. Five general groups of neuroleptic medications During the 1960s, research methodology improved considerably. Single case studies and clinical are given to children: the antipsychotic, antidepresobservations were replaced with controlled studies sant, antianxiety, antimania, and stimulant medicausing rating scales, double-blind research designs, tions. The first four are discussed in this chapter. For placebo, and statistical analysis. Research has al- each, a general outline of the clinical characteristics ways been hampered by the fact that medications of the diagnostic categories associated with the use of
NONSTIMULANT PSYCHOTROPIC MEDICATION
the medication will be given. Then, the specific target symptoms that are responsive to medication will be covered. The discussion of side effects follows. Side effects are physiological, psychological, and behavioral responses to medication that are not pertinent to the main purpose of the medication. All drugs have side effects. Side effects are not necessarily undesirable. Often, they are only a nuisance. Frequently the patient does not realize that a side effect is occurring because no real problem is caused. Effect and side effect can change as the purpose for which the medication is prescribed changes. Some medications impair motor performance or may cause bizarre motor responses. This may impede testing. Sedative effects common to the initial administration of some drugs may likewise cause a problem. One of the earlier reviews (Baker, 1968) of the effects of psychotropic drugs on psychological testing concluded that there were few or no changes in psychological test performance as a result of drug treatment. Common clinical tests such as the Bender Gestalt, Draw-A-Person, Wechsler Adult Intelligence Scale, Rorschach, Thematic Apperception Test, and Minnesota Multiphasic Personality Inventory were reviewed. The author hypothesized that the lack of drug effects demonstrable may be due to the relative insensitivity of the clinical tests. A similar conclusion was drawn concerning the HalsteadReitan Neuropsychological Battery (Howard, Hogan, & Wright, 1975). This study controlled for age differences and considered drugs individually and in combination. The subjects were patients using antipsychotic, antianxiety, antidepressant, and sedative drugs. The conclusion held true for a wide dosage and range. Specialized research instruments do demonstrate the pharmacological effects of psychotropic medication upon cognitive and behavioral function (Nicholson & Ward, 1984). Variables studied include body sway, driving performance, vigilance, pupillary function, ambulatory motor activity, saccadic eye movements, and various pencil-and-paper tests. Most studies involve the single dose administration of a drug to a healthy volunteer. This is not analogous to the clinical setting where patients take medication over a period of time and adapt to the initial side effects. Although not of relevance to the clinical testing of children, the reports from the journal edited by Nicholson and Ward (1984) do represent the type of research that is necessary to demonstrate psychotropic medication effects on physiological and cognitive functioning through psychological tests.
477
Antipsychotics The antipsychotics are used to treat psychotic symptoms in adults or adolescents with major mental disorders. These include schizophrenia, major depression with psychotic features, bipolar disorders with manic symptoms, and delirium or dementia with associated psychotic symptoms. The symptoms most responsive to drugs are those of acute or sudden onset and include the disorganized thinking of the schizophrenic disorder, delusional thinking or hallucinations of any psychotic process, and the confusion associated with delirium or dementia. Symptoms that are slow, gradual, and insidious in onset are poorly responsive. Psychotic symptoms accompanied by anxiety, agitation, and motoric hyperactivity are usually amenable to medication treatment. Medicationresponsive psychotic symptoms seen in adult disorders are not common to childhood psychopathology. No childhood disorders respond as well to antipsychotic medication. That childhood disorders are distinct from adultonset disorders was made clear by the change in diagnostic nomenclature for the DSM-lli. The psychotic reaction of childhood and childhood schizophrenia diagnoses have been updated. The current nomenclature is pervasive developmental disorder. This reflects the extent of the possible impairments due to disordered growth and development in multiple systems. There may be arrests or distortions in the development of language, perception, social skills, reality testing, and motor movement. Although childhood autism (onset before 30 months of age) is distinguished from the childhood-onset pervasive developmental disorders (onset after 30 months of age and before 12 years), this may be an unnecessary distinction. Probably there are several common etiological factors that could serve as insult to account for development of the problem. The age of onset may determine the specific clinical presentation. A lack of responsiveness to other people is the hallmark of infantile autism. The deficits in language development include echolalia, pronominal reversal, bizarre intonations, and bizarre motor patterns. There is a resistance to change in the environment. The child may respond to people as he or she does to inanimate objects without giving cognizance to the human qualities of the relationship. If the onset comes after the child has developed speech and a relationship with a caretaker has been established, the clinical presentation is different. Although there may be a regression to a more autistic level of relating to others, many of the behaviors may be near normal.
478
CHAPTER 25
There may be episodes of severe or excessive anxiety. Affect may be inappropriate to the situation. Bizarre motoric patterns such as walking on tiptoes or choreoathetoid-like hand or finger movements may develop. Speech abnormalities are seen. Self-mutilation may occur. Excessive fantasy play may occur in settings where other children the same age are working and talking with a reality-oriented or task-directed manner. Most current research suggests that a biological insult during the first trimester of pregnancy accounts for at least some of the earlier-onset and more autistic disorders. The data are less clear for the later-onset disorders. It is only with onset of illness toward the end of latency that the clinical picture may be the same as that seen with the adult-onset schizophrenic disorders. If hallucinations or delusions are present, the problem is not one of pervasive developmental disorder. Nomenclature designed for adult psychopathology would then apply even though a child is involved. One other diagnostic category in childhood is treated with antipsychotic medication. These are children with tic disorders. At present, few therapists feel a tic disorder represents a specific psychodynamic conflict. Historically, it was thought to be so. Some 20% of children develop a transient tic at some time. The tic disorder of concern is Tourette's disorder. This involves multiple tics, including vocal tics. The onset may be as early as age 2 and symptoms may last a lifetime. They also remit and exacerbate with a variable time course. Those who use psychoptherapy to treat the tics point out that symptoms exacerbate during stress. Intervention is directed toward stress reduction. Fewer psychotherapists at present try to analyze the tic as a symptom representing psychological conflict. Both behavioral and medical management programs have the same problem: differentiating therapeutic response from the natural course of the disorder. When medications are used, usually haloperidol (Haldol) is prescribed. There is no good reason why this should be chosen over the other antipsychotic medications except that it has been studied more in the treatment of Tourette's syndrome. The target symptoms for treatment of childhood disorders with antipsychotic medications (with the exception of the treatment of multiple tics, which came later) remain the same as when the medications were first used in the 1950s. Drugs are used to treat the psychomotor agitation/hyperactivity, excessive and incapacitating anxiety, and aggressive behaviors toward self and others common to a number of diagnoses. As a consequence, the mentally retarded who
have behavior problems of this nature are given antipsychotic medications. The brain-damaged child with aggressive and uncontrollable behavior may be medicated. Antipsychotics are often tried in the treatment of any child with a pervasive developmental disorder even though now it is considered to be a developmental disorder involving multiple systems rather than a psychotic process. Sometimes children who have a conduct disorder with an aggressive component may be medicated. The common behavioral side effects of the antipsychotic drugs involve CNS-mediated behaviors. One possible side effect is the acute dystonic reaction. This involves rather sudden onset spasms of several common muscle groups. Although a dystonic reaction may look frightening to an observer, most cause little if any physical discomfort. The jaw may pull over to one side and look as if it is dislocated. The head and neck may twist to one side in a torticollis position. The child may complain of his tongue pulling into the back of his mouth, or that he feels like he is swallowing his tongue. Both eyes may roll up (oculogyric crisis) leaving only the bottom part of each pupil and iris showing. Rarely, the truncal muscles may be involved. The child may twist into a opisthotonos position with the head pulled backwards and back arched with arms drawn up and legs extended back. With the exception of the tongue complaint, the acute dystonic reactions should be easily recognizable. Dyskinesias should also be easily recognizable. These include facial tics, lip-smacking, tongue movements, blinking, and muscle spasms. Most of these symptoms can occur in children who are not receiving antipsychotic medication. Thus, it is difficult to assign an etiology to these observations. They may not be due to medications. Akathisia is easily recognized in adults. It consists of an apparent inability to sit still. The legs may look as if they want to pace as they continue to move in a "walking" pattern even while the individual is sitting. In children, the "fidgetiness" of the hyperactive child may be mistaken for akathisia. A parkinsonian picture may be seen. There may be muscular rigidity, a pill-rolling hand tremor, drooling, and masklike faces. The reduced movement may be confused with catatonia or with the appearance of a retarded depression. A troublesome side effect is tardive dyskinesia. This involves abnormal mouth movements that may look like persistent chewing, lip-smacking, or repetitive tongue protruding. Another name for this is buccolingual-masticatory (BLM) movements. They may go away if medication is lowered or discon-
NONSTIMULANT PSYCHOTROPIC MEDICATION
tinued and may be hidden or masked if medication is increased. Sometimes BLM movements increase in intensity and do become disfiguring. The picture is confusing because withdrawal dyskinesias are initially indistinguishable from tardive dyskinesia. The withdrawal dyskinesias may appear transiently when the relative dose of antipsychotic medication is decreased. The literature that cites incidences of tardive dyskinesia probably includes withdrawal dyskinesia symptoms also. Those at greatest risk are elderly women. Although reported in children, the incidence probably is not as great. The dystonic reactions, most dyskinesias, akathisias, and parkinsonian symptoms should clear following treatment with an antiparkinsonian drug such as benztropine mesylate (Cogentin) or with diphenhydramine (Benadryl). Some of the antipsychotic medications may cause sedation when first given or anytime that very high doses are prescribed. This effect should abate over several days or weeks as the child adapts to the medication or as the physician decreases the initial treatment dose to lower maintenance levels. If antipsychotic medication is withdrawn, the clinical effects persist for about 24 to 36 hours. Essentially all of the drug will be metabolized and cleared from the body within 3 or 4 days following cessation. At this point, there is no longer any clinical effect. With prolonged administration, metabolites may persist in the urine for several months. These latter findings are not of clinical importance because there is no pharmacological effect. Most research that considers the effects of antipsychotic medications upon cognition has studied children with mental retardation or with autism. Some 70% of this latter group have an IQ in the range of the mentally retarded. When research subjects have medication-responsive target behaviors, it is difficult to design any research using common psychological tests that specifically assess the effects of antipsychotic medications upon cognition alone. Most studies look at the effects of medications both upon behavior and upon learning ability (Anderson et al., 1984; Campbell, Anderson, Small, Perry, Green, & Caplan, 1982; Campbell et al., 1982; Helper, Wilcott, & Garfield, 1963; Sprague, Barnes, & Werry, 1970; Weise, O'Reilly, & Hesbacher, 1972; Werry, Weiss, Douglas, & Martin, 1966; Wong & Cock, 1971). Several make the point that as the symptoms of aggression, restlessness, and poor concentration improve, tasks including learning and school achievement improve. Most of the improvement seems to come following the first I to 2 months of medication treatment. The study by Wong and Cock (1971) suggested that the antipsychotic medi-
479
cation not only did not impair performance on the visuomotor tasks of the WISC, but accounted for an improved IQ. Results such as this most often involved studies of children with an autistic process. When antipsychotic medications were given to children with a variety of diagnoses that included nonpsychotic processes (such as adjustment reaction, psychoneurotic reaction, and personality disturbance), the results more often were inconclusive or actually showed some deterioration in test performance. It could be argued that in these latter groups, the symptoms of psychopathology do not directly impede the process of testing in the same sense that the symptoms of autism impede testing. Thus, in the autistic child, the use of medications may have alleviated target symptoms and allowed testing to take place. For the other diagnostic categories, medication either had no effect because there were no symptoms that impeded the process of testing or a sedative effect actually impaired the test performance. Repeatedly, the comment was made that any untoward effects on testing or learning were probably related to either the sedation seen at high doses, or the sedation seen upon initial administration. This was more true when the drugs were used for treatment of the aggressive and explosive components of the conduct disorders, and the hyperactive component of the attention deficit disorder. Similar conclusions have been drawn in studies with adults (Braff & Saccuzzo, 1982; Spiegel & Keith-Spiegel, 1967; Weiss, Robinson, & Dasberg, 1973). In an analogous manner to the studies that showed an improvement in cognitive performance on selected clinical psychological tests with autistic children, those symptomatic with schizophrenia also improved on similar tests following the administration of antipsychotics. Again, the improvement came with a reduction in the severity of symptoms of illness. It was postulated that those with severe hyperactivity had excess dopaminergic activity and that haloperidol, a dopamine antagonist, reduced the hyperactivity and fidgetiness. This allowed an improvement in test performance. The following represents clinical guidelines. Antipsychotic medication treatment of children with a pervasive developmental disorder, in reasonable clinical doses and without obvious sedation, should not adversely affect the test results on clinical instruments such as the Wechsler Scales or Bender Gestalt. If anything, the medication will improve performance by reducing many of the target symptoms that impair test performance. These symptoms include hyperactivity, fidgetiness, aggressivity, and concentration problems. It is possible that the same med-
480
CHAPTER 25
ications, used to treat similar symptoms in children with attention deficit disorders or conduct disorders, may impair performance on those items that are timed or require motor performance. This is difficult to demonstrate in the common psychological tests used clinically and may only be of concern if the dose of medication is actually causing sedation. Adverse effects of medication upon test results will be more frequent just after the medication is started. At reasonable clinical doses, they will be of less or negligible concern after the child has been on medication for at least 4 to 8 weeks. If testing is to be used for educational placement, it should come 4 to 8 weeks after the child is on a stable maintenance dose. If the child must be tested free of antipsychotic medication, a 3- or 4-day delay following cessation of drug should be sufficient.
Antidepressants The antidepressants were developed to treat symptoms of melancholia in adolescents and adults experiencing depression as part of a major depression or the depressive component of a bipolar disorder or schizoaffective disorder. Symptoms most clearly responsive to these drugs include crying spells, decreased appetite and weight loss, diurnal mood variation (less energy upon arising than later in the day), difficulty falling asleep, middle of the night awakening, early morning awakening, decreased libido, and impairment of concentration. Less responsive to medications are subjective feelings of depression and anhedonia. Drugs are of benefit in the treatment of a dysthymic disorder (depressive neurosis) only when the previously listed biological concomitants of depression are present. Antidepressants do not alleviate subjective feelings of depression or life circumstance dysphoria unaccompanied by these biological changes. Although clinicians do not question the concept of depression with melancholia in adolescents, the concept of childhood depression has caused difficulty. Young age and limited cognitive development are probably protective of the development of the nihilistic depressive delusional beliefs that are seen in adults. It is uncommon for the target symptoms (biological concomitants of depression) that so clearly respond to antidepressants in adults to persist in children. When they occur, it is usually for only a brief period of time. For adults, approximately 3 weeks of antidepressant treatment is needed for clinical symptoms of depression to respond. For children, the delay in response may be even longer. For this reason,
medication is usually prescribed only after symptoms have persisted for at least 30 days. Adolescents with the same melancholic symptoms as adults seem to respond more poorly to antidepressants than do adults. To account for the paucity of melancholic symptoms in children, the psychodynamic concepts of masked depression, somatic equivalent of depression, and acting out of defense against underlying depression were formulated. Currently, psychosocial deficits and impairments in relationships are being discussed as diagnostic of depression in the schoolage child. At best, the children meeting psychodynamic formulation criteria for depression may also meet the diagnostic criteria for a dysthymic disorder without symptoms of melancholia. In the early 1970s, the tricyclic antidepressant imipramine (Tofranil) was used to treat hyperactive children. Most of the target clinical improvements included improved conduct in general, improved attention span, and reduced motor activity. Most studies concluded that the antidepressant was effective and also that the degree of improvement was eventually the same as with stimulant medication (Quinn & Rapoport, 1975; Rapoport, Quinn, Bradbard, Riddle, & Brooks, 1974; Werry, Aman, & Diamond, 1979). These medications never gained popularity, however. Today the stimulant medications are used almost exclusively. An increased incidence of side effects with the antidepressants may have led physicians to choose the stimulant medications. Enuresis is still being treated with imipramine. Although this drug is the prototype antidepressant for use with this problem, any tricyclic antidepressant should do equally as well. The most optimistic reports come in the treatment of secondary nocturnal enuresis. It is called secondary because the child does have periods free of enuresis. Primary enuresis (never has been dry) is less responsive. About half of those with secondary enuresis become asymptomatic with medication. Probably only half of this group sustain the improvement. Imipramine treatment is superior to placebo treatment, but has not been demonstrated to be superior to the many other alternative treatments (bell and pad, token economy, brief family or individual psychotherapy). There is a literature on this topic (Behrle, Elkin, & Laybourne, 1956; Kardash, Hillman, & Werry, 1968). Some children with severe separation anxiety disorders are treated with high doses of imipramine. The child with this disorder may experience unrealistic worry that harm will befall a caretaker if they are out of sight. Associated behaviors include refusal to attend school so that the child may stay home with
NONSTIMULANT PSYCHOTROPIC MEDICATION
the caretaker and insistence on sleeping with the caretaker. Physical complaints such as headaches or stomachaches on school days only are reported. Anxiety may be seen. Social withdrawal, apathy, sadness, or difficulty concentrating may be so severe as to be judged melancholic components of depression. Most medication treatment for this disorder comes only after psychotherapeutic and psychosocial/educational therapies have failed. It is always used in conjunction with other therapies. This is not a common treatment and the current literature is still sparse (Gittelman-Klein & Klein, 1971, 1973). There are no dramatic behavioral side effects of antidepressants as there were with the antipsychotics. Children are more comfortable if the dose is increased by increments every 2 or 3 days so that the therapeutic dose is reached after a week or so. If the dose is increased too rapidly, the child may complain of sleepiness, tiredness, or drowsiness. These effects are transient and should abate over a week or two. Mild fine hand tremors may be seen or there may be complaints of a dry mouth or blurred vision for reading. These latter symptoms may be anticholinergicmediated side effects that are more of a problem with some tricyclics than with others. These should abate also. This may take a few months if the child is unduly sensitive or the dose must remain high. Rarely, a child will complain of difficulty paying attention even on a reasonable dose. A different tricyclic may need to be prescribed. Because this is such a common symptom of preexisting psychopathology, a decision to attribute these complaints to medication should be made with care. When tricyclic antidepressants are stopped following prolonged administration, all biological activity or clinical medication effect should dissipate over a 7- to 10-day period. This may be even shorter with young children than with adults. The cognitive effects of single dose administration of antidepressants have been studied (DiMascio, Heninger, & Klerman, 1964; Ross, Smallberg, & Weingartner, 1984; Thompson & Trimble, 1982). It is clear that some antidepressants are more sedating than others. Subjective reports of sleepiness peak between 2 and 3 hours and are gone by 7 hours after the drug is ingested. The same complaints do not apply during long-term administration. Care should be taken not to extrapolate from effects following single dose administration to effects following prolonged administration. The latter group has an opportunity to adapt to the medication effects common to single dose administration. The more sedating tricyclics did impair performance on serial addition and digit sub~titution tasks when the drug was first ad-
481
ministered. The effects upon concentration for these tasks may have again become apparent during withdrawal following prolonged administration. The memory and attention changes during antidepressant treatment of adults with depression have received more attention (Glass, Uhlenhut, & Weinreb, 1978; Glass, Uhlenhuth, Hartel, Matuzas, & Fischman, 1981; Henry, Weingartner, & Murphy, 1973; Lamping, Spring, & Gelenberg, 1984; Legg & Stiff, 1976; Sternberg & Jarvik, 1976; Keeler, Prange, & Reifler, 1966). The effects of antidepressants upon cognitive function are unclear. Probably gross clinical measures such as the Wechsler Adult Intelligence Scale are not sensitive enough to reflect medication effect. When patients are tested for IQ changes following treatment with antidepressants, scores do improve. But medication effects on test scores cannot be separated from practice effect and from improvement due to alleviation of depression. Depressed patients usually show an impairment in short-term memory tasks without long-term memory impairment. The greater the clinical improvement with medications, the more short-term memory improved. Imipramine probably does facilitate psychomotor tasks such as tapping or reaction time. It is possible that those tricyclic antidepressants with higher anticholinergic effects may potentiate memory disturbances in depressed patients while they are clinically depressed and when drug treatment is first started. Again, this was not demonstrated on common clinical measures, but could be seen on specialized memory tasks. The short-term memory (improved accuracy without decreased speed) improvement may come just prior to improvement in the clinical level of depression. Cognitive effects of antidepressants have been studied in children being treated for enuresis, hyperactivity, childhood depression, and aggression (Brumback & Staton, 1980; Campbell, Small, et al., 1982; Kupietz & Balka, 1976; Rapoport, 1965; Staton, Wilson, & Brumback, 1981; Werry, Dowrick, Lampen, & Vamos, 1975). The children with depression who received medication did have melancholic symptoms. Psychological testing after successful drug treatment usually showed an improvement in the pretreatment scores on such measures as the Wechsler Intelligence Scale for Children. It may take 2 or 3 months of drug therapy before these gains can be shown. It was hypothesized and suggested that the symptoms of depression precluded accurate testing and that the educational and intellectual performance of a'depressed child should be assessed after the depression is in remission. Although not a clinical measure, the Continuous Performance Test was used
482
CHAPTER 25
in one study to show that amitriptyline (Elavil) facilitated vigilance performance. Several of the more clinical studies commented that children receiving an antidepressant concentrated better. In those studies that compared pre- and posttreatment IQs for children treated for conduct problems or enuresis, no significant differences were found. The following clinical guidelines apply for routine psychological and psychoeducational testing. If the child is receiving tricyclic antidepressants to treat a depression with a melancholic component, testing should be delayed until clinical improvement is seen if it is to be used for long-term educational placement. The delay may need to be as long as 2 to 3 months after medication is started. Testing earlier than this probably reflects current function only and may be a poor predictor of future performance. For the child receiving antidepressants for problems other than depression, the initial sedative and/or anticholinergic effect that may be seen in the first week or two of administration may influence the test results. It would be best to delay testing until the child has been on medication a couple of weeks. If the child must be tested free of antidepressant medication, a 7to 10-day delay will be necessary. There is no research evidence to show that long-term administration of tricyclic antidepressants impairs performance on any routine psychological or psychoeducational batteries.
sufficient to convince physicians to incorporate lithium carbonate into the psychopharmacological armamentarium for use in childhood behavior disorders, research on the use of lithium carbonate to control aggression in childhood has afforded an opportunity to look at the cognitive effects. Lithium has been used as an antiaggression drug for school-age children with undersocialized and aggressive conduct disorders (Platt, Campbell, Green, & Grega, 1984). This is a disorder in which the child displays a persistent and repetitive pattern of aggressive conduct. There may be physical violence against persons or property, including vandalism, fire setting, mugging, and assault. Fighting is the predominant symptom for the younger child. In usual therapeutic doses, monitored via serum blood levels, lithium does not produce CNS-mediated behavioral side effects. One study (Judd, Hubbard, Janowsky, Huey, & Takahashi, 1977) reported that in normal volunteers, therapeutic levels of lithium did impair performance on the Digit Symbol subtest of the W AIS and the Trail Making A Test. Another study concluded that lithium caused a decrement in W AIS IQ (Aminoff, Marshall, Smith, & Wyke, 1974) in patients being treated for Huntington's chorea. This is a progressive neurological disorder. Dementia is a part of the expected deterioration and the concern was whether the lithium potentiated the process. Other studies have looked at memory tasks (Bonnel, Etevenon, Benyacoub, & Slowen, 1981 ; Christodoulou, Kokkevi, Lykouras, Antimanic Drugs Stefanis, & Papadimitriou, 1981; Huey, Janowsky, Lithium carbonate is used for the treatment of Judd, Abrams, Parker, & Clopton, 1981; Kusumo & the manic phase of bipolar disorders in adults. It is of Vaughan, 1977; Marusarz, Wolpert, & Koh, 1981; unclear benefit in the treatment of the depressive Squire, Judd, Janowski, & Huey, 1980). All of these phase and for many patients does not offer pro- studies used either specialized scales or selected parts phylaxis against recurring depression. Because of more common tests. clinical improvement of manic symptoms takes 7 to No consistent conclusion can be drawn. It might 10 days with lithium alone, an antipsychotic medica- be that those with a concomitant dementing process tion that can reduce symptomatology over 3 or 4 days or those older are at more risk for memory dysfuncis often used first. If the manic phase of a bipolar tion when lithium is used. It may be that those who disorder does occur in children, it must be rare. It is are the most ill show more impairment. There is some evidence that lithium does not affect the more global seen in the adolescent. The behavioral symptoms of a manic episode measures such as theWAIS or the Wechsler Memory include an increase in activity or physical rest- Scale in nondemented patients. In contrast to the lessness, a pressure to keep talking, racing thoughts, other drugs covered, studies with lithium did not grandiosity, decreased sleep, and distractibility. The show an improvement in cognitive function with individual in a manic episode often becomes irritable clinical improvements. There are insufficient data when confronted with structure and control or if gran- from which to generalize concerning children and diose schemes are thwarted. Physical and verbal ag- testing. gression may result. Property may be destroyed. Hospital personnel attempting to controi the patient Antianxiety Drugs may be hurt if not trained to handle the physical Antianxiety medications are in widespread use in outbursts. Although the research evidence has not been the symptomatic relief of anxiety in adults. Anxiety is
NONSTIMULANT PSYCHOTROPIC MEDICATION
a universal phenomenon and is not limited to psychopathological states. Rating scales show that the depressed patient frequently complains more vehemently of anxiety than does the patient with a generalized anxiety disorder. Anxiety can be very painful to those experiencing a psychotic episode. Children experience anxiety also. This may be part of a specific diagnosis such as an overanxious disorder, a separation anxiety disorder, or an obsessive-compulsive disorder. Although antianxiety medications lead the list of medications prescribed for adults, it is relatively uncommon for the same drugs to be prescribed for children. Some physicians do prescribe anxiolytics for children as a part of symptomatic treatment of anxiety. This is usually for only very brief periods of time. There is no clinical literature advocating the use of anxiolytics to treat the common childhood diagnoses associated with anxiety. There is a corresponding paucity of literature on the cognitive and behavioral side effects of anxiolytics in children. One study that evaluated anxiolytic drug effects on the cognitive function of children concluded that at therapeutic doses, there is no adverse effect on cognition (Ferguson & Simeon, 1984). Most studies of adults conclude that at common therapeutic doses, little if any difference between placebo and drug groups can be demonstrated (Healey, Pickens, Meisch, & McKenna, 1983; Pishkin, Fishkin, Shurley, Lawrence, & Lovallo, 1978; Zimmermann-Tansella, 1984). Single dose administration may reduce the speed of performance on some items. At high doses, the sedative effect will impair cognitive performance. As with the antimanic drugs, there are insufficient research data with the use of anxiolytic drugs in children to comment on the cognitive effects in relationship to routine psychological testing. The clinical effect for single dose administration of most anxiolytics is about 4 to 6 hours. The half-life for most is much longer and drugs may be detected through blood and urine assays for I to 2 days following single dose administration. If the child must be free of anxiolytic medication effect at the time of testing, a delay of I day following the last dose should be sufficient.
483
are a vast improvement over the earlier research. Yet, for the physician prescribing medication, treatment is still directed toward a limited number of target symptoms such as motoric hyperactivity, aggressivity, psychomotor excitement, and enuresis. Much prescribing is on an empirical basis and not for treatment of specific psychopathological disturbances. Nonstimulant psychotropic medications never constitute the sole treatment of any childhood psychopathological disorder. They are an adjunct at best. Programs involving special education placements and psychological interventions (psychotherapeutic and behavioral management) are the mainstay of treatment. All medications have side effects. A side effect is any pharmacological action that occurs in addition to the amelioration of target symptoms. That a side effect occurs is neither desirable nor undesirable unless it causes discomfort or dysfunction. Some behavioral and cognitive side effects may influence or interfere with psychological testing. Most behavioral and cognitive side effects of medications on test performance are easily recognized. Sedation is associated with both the initial administration and high doses of some drugs. Children adapt to common doses quickly. The recommended delay in testing following the start of drug management may be only a week or two. For some, the recommended delay is also dependent on the diagnosis. For the latter situation, if the examiner must wait until clinical response occurs, the delay may be as long as 2 months. The CNS-mediated side effects for the antipsychotics are very specific. Probably, dissociative responses and oppositional behaviors, neither a side effect of drugs, account most frequently for the concerns that medications are adversely affecting psychological testing results. There is little research evidence to show that nonstimulant psychotropic medications in common doses, once the child is on a regular administration regime, affect the routine psychological tests used with children. The tests are just not sensitive enough to be influenced by most nonstimulant psychotropics. There is some evidence that with the use of medications to treat major depression or symptoms associated with the pervasive developmental disorders, there actually is improvement in test performance. Summary This chapter has presented clinical guidelines for use when testing children who are receiving antiChildhood psychopharmacology started in the psychotic, antidepressant, and anxiolytic drugs at the late 1930s. The search for behavior-modifying drugs time of testing. These guidelines must be used in for use with children continues today. Current re- conjunction with a knowledge of the class of drug search methods with rating scales sensitive to behav- prescribed, the child's diagnosis, whether the dose is ioral changes and pharmacokinetic studies that may common or high, and the length of time the child has be able to link blood levels with medication response been receiving the medication.
484
CHAPTER 25
References American Psychiatric Association. (1980). Diagnostic and statistical manual of mental disorders (3rd ed.). Washington, DC: Author. American Psychiatric Association. (1985). Draft: DSM-lll-R in development. Washington, DC: Author. Aminoff, M. 1., Marshall, 1., Smith, E., & Wyke, M. (1974). Cognitive function in patients on lithium therapy. British Journal of Psychiatry, 125, 109-112. Anderson, L. T., Campbell, M., Grega, D. M., Perry, R., Small, A. M., & Green, W. H. (1984). Haloperidol in the treatment of infantile autism: Effects upon learning and behavioral symptoms. American Journal of Psychiatry, 141, 11951202. Baker, R. R. ( 1968). The effects of psychotropic drugs on psychological testing. Psychological Bulletin, 69, 377-387. Behrle, F. C., Elkin, M. T., & Layboume, P. C. (1956). Evaluation of a conditioning device in the treatment of nocturnal enuresis. Pediatrics, 17, 849-856. Bonne!, H. L., Etevenon, P., Benyacoub, J., & Slowen, P. (1981). Intellectual efficiency in manic-depressive patients treated with lithium. Acta Psychiatrica Scandinavica, 64, 423-430. Braff, D. L., & Saccuzzo, D.P. (1982). Effect of antipsychotic medication on speed of information processing in schizophrenic patients. American Journal of Psychiatry, 139, 1127-1130. Brumback, R. A., & Staton, R. D. (1980). Neuropsychological study of children during and after remission of endogenous depressive episodes. Perceptual and Motor Skills, 50, 11631167. Campbell, M., Anderson, L. T., Small, A.M., Perry, R., Green, W. H., & Caplan, R. (1982). The effects of haloperidol on learning and behavior in autistic children. Journal of Autism and Developmental Disorders, 12, 167-175. Campbell, M., Small, A.M., Green, W. H., Jennings, M.A., Peny, R., Bennett, W. G., Pedron-Gayol, M., & Anderson, L. ( 1982). lithium and haloperidol in hospitalized aggressive children. Psychopharmacology Bulletin, 18, 126-130. Christodoulou, G. N., Kokkevi, A., Lykouras, E. P., Stefanis, C. N., & Papadirnitriou, G. N. (1981). Effects of lithium on memory. American Journal of Psychiatry, 138, 847-848. DiMascio, A., Heninger, G., & Kleeman, G. L. (1964). Psychopharmacology of imipramine and desipramine: A comparative study of their effects in normal males. Psychopharmacologia, 5, 361-371. Ferguson, H. B., & Simeon, I. G. (1984). Evaluating drug effects on children's cognitive functioning. Progress in Neuropsychopharmacology and Biological Psychiatry, 8, 683-686. Gittelman-Kiein, R., & Klein, D. F. (1971). Controlled imipramine treatment of school phobia. Archives of General Psychiatry, 25, 204-207. Gittelman-Klein, R., & Klein, D. F. (1973). School phobia: Diagnostic con~iderations in the light of imipramine effect. Journal of Nervous and Mental Disease. 156, 199-215. Glass, R. M., Uhlenhuth, E. H., Hartel, F. W., Matuzas, W., & Fischman, M. W. (1981). Cognitive dysfunction and im-
ipramine in outpatient depressives. Archives ofGeneral Psychiatry, 38, 1048-1051. Glass, R. M., Uhlenhuth, E. H., & Weinreb, H. (1978). Imipramine-reversible cognitive deficit in outpatient depressives. Psychopharmacology Bulletin, 14, 10-11. Healey, M., Pickens, R., Meisch, R., & McKenna, T. (1983). Effects of clorazepate, diazepam, lorazepam, and placebo on human memory. Journal of Clinical Psychiatry, 44, 436439. Helper, M. M., Wilcott, R. C., & Garfield, S. L. (1963). Effects of chlorpromazine on learning and related processes in emotionally disturbed children. Journal of Consulting Psychology, 27, 1-9. Henry, G. M., Weingartner, H., & Murphy, D. L. (1973). Influence of affective states and psychoactive drugs on verbal learning and memory. American Journal ofPsychiatry, 130, 966-971. Howard, M. L., Hogan, T. P., & Wright, M. W. (1975). The effects of drugs on psychiatric patients' performance on the Halstead-Reitan Neuropsychological Test Battery. The Journal of Nervous and Mental Disease, 161, 166-171. Huey, L. Y.,Janowsky, D. S.,Judd, L. L., Abrams, A., Parker, D., & Clopton, P. (1981). Effects of lithium carbonate on methylphenidate-induced mood, behavior, and cognitive process. Psychopharmacology, 73, 161-164. Judd, L. L., Hubbard, B., Janowsky, D. S., Huey, L. Y., & Takahashi, K. I. (1977). The effect of lithium carbonate on the cognitive functions of normal subjects. Archives of General Psychiatry, 34, 355-357. Kardash, S., Hillman, E. S., & Werry, I. (1968). Efficacy of imipramine in childhood enuresis: A double-blind control study with placebo. Journal of the Canadian Medical Association, 99, 263-266. Keeler, M. H., Prange, A. I., & Reitler, C. B. (1966). Effects of imipramine and tbioridazine on set and action. Diseases ofthe Nervous System, 27, 798-802. Kupietz, S. S., & Balka, E. B. (1976). Alterations in the vigilance performance of children receiving amitriptyline and methylphenidate pharmacotherapy. Psychopharmacology, 50, 29-33. Kusumo, K. S., & Vaughan, M. (1977). Effects of lithium salts on memory. British Journal of Psychiatry, 131, 453-457. Lamping, D. L., Spring, B., &Gelenberg, A.J. (1984). Effects of two antidepressants on memory performance in depressed outpatients: A double blind study. Psychopharmacology. 84, 254-261. Legg, J. F., & Stiff, M. P. (1976). Drug-related test patterns of depressed patients. Psychopharmacology, 50, 205-210. Marusarz, T. Z., Wolpert, E. A., & Koh, S.D. (1981). Memory processing with lithium carbonate. Journal of Clinical Psychiatry, 42, 190-192. National Institute of Mental Health. (1973). Pharmacotherapy of children. Psychopharmacology Bulletin, Special Issue, 9, 1195. National Institute of Mental Health. (1985). Rating scales and assessment instnJments for use in pediatric psychopharmacology research. Psychophamwcology Bulletin, 21, 7131125. Nicholson, A. N., & Ward, J. (1984). Psychotropic drugs and
NONSTIMULANT PSYCHOTROPIC MEDICATION
485
performance. British Journal of Clinical Pharmacology: ment of childhood major depressive illness. Perceptual and Supplement I to Volume /8. Motor Skills, 53, 219-234. Pishkin, V., Fishkin, S.M., Shurley, J. T., Lawrence, B. E., & Sternberg, D. E., & Jarvik, M. E. (1976). Memory functions in Lovallo, W. R. (1978). Cognitive and psychophysiologic depression. Archives of General Psychiatry, 33, 219-224. response to doxepin and chlordiazepoxide. Comprehensive Thompson, P. J., & Trimble, M. R. (1982). Non-MAOI antiPsychiatry, /9, 171-178. depressant drugs and cognitive functions: A review. PsychoPlatt, I. E., Campbell, M., Green, W. H., &Grega, D. M. (1984). logical Medicine, 12, 539-548. Cognitive effects of lithium carbonate and haloperidol in Weiner, J. M. (1977). Psychapharmacology in childhood and treatment-resistant aggressive children. Archives of General adolescence. New York: Basic Books. Psychiatry, 41, 657-662. Weiner, J. M. (1985). Diagnosis and psychopharmacology of Quinn, P. 0., & Rapoport, J. L. (1975). One year follow-up of childhood and adolescent disorders. New York: Wiley. hyperactive boys treated with imipramine or methylpheni- Weise, C. C., O'Reilly, P. P., & Hesbacher, P. (1972). Perdate. American Journal of Psychiatry, 132. 241-245. phenazine-amitriptyline in neurotic underachieving stuRapoport, J. (1965). Childhood behavior and learning problems dents: A controlled study. Diseases of the Nervous System, treated with imipramine.JnternationalJournal ofPsychiatry, 33, 318-325. 1. 635-642. Weiss,A.A.,Robinson,S.,&Dasberg,H. (l973).Changesinmental Rapoport, J. L., Quinn, P. 0., Bradbard, G., Riddle, K. D., & functioning accompanying antipsychotic drug therapy. Annals Brooks, E. (1974). Imipramine and methylphenidate treatofPsychiatry andReiDted Disciplines, 11, 134-140. ments of hyperactive boys. Archives of General Psychiatry, Werry, J. S. (1978). Pediatric psychopharmacology: The use of 30, 789-793. behavior modifying drugs in children. New York: Ross, R. J., Smallberg, S., & Weingartner, H. (1984). The effects Brunner/Mazel. of desmethylimipramine on cognitive function in healthy sub- Werry, I. S., Aman, M.G., & Diamond, E. (1979). Imipramine jects. Psychiatry Research, 12, 89-97. and methylphenidate in hyperactive children. Journal of Spiegel, D. E., & Keith-Spiegel, P. (1967). The effects of carChild Psychology and Psychiatry, 21, 27-35. phenazine, trifluoperazine, and chlorpromazine on ward be- Werry, J. S., Dowrick, P. W., Lampen, E. L., & Vamos, M. J. havior, physiological functioning, and psychological test (1975). Imipramine in enuresis-Psychological and physioscores in chronic schizophrenic patients. The Journal ofNerlogical effects. Journal of Child Psychology and Psychiatry, vous and Mental Disease, 144, 111-116. 16, 289-299. Sprague, R. L., Barnes, K. R., & Werry, J. S. (1970). Meth- Werry, J. S., Weiss, G., Douglas, V., & Martin, M.A. (1966). ylphenidate and thioridazine: Learning, reaction time, acStudies on the hyperactive child: The effect of chlortivity, and classroom behavior in disturbed children. Ameripromazine upon behavior and learning ability. Journal of the can Journal of Onhopsychiatry, 40, 615-628. American Academy of Child Psychiatry, 5, 292-312. Squire, L. R., Judd, L. L., Janowsky, D. S., & Huey, L. Y. Wong, G. H., & Cock, R. J. (1971). Long term effects of (1980). Effects of lithium carbonate on memory and other haloperidol on severely emotionally disturbed children. Auscognitive functions. American Journal of Psychiatry, 137, tralia and New Zealand Journal of Psychiatry, 5, 296-300. 1042-1046. Zimmermann-Tansella, C. (1984). Psychological performance of Staton, R. D., Wilson, H., & Brumback, R. A. (1981). Cognitive normal subjects on tasks commonly used in evaluation of improvement associated with tricyclic antidepressant treatanxiolytic drugs. Perceptual and Motor Skills. 58, 803-810.
IV New Aspects of Neuropsych ology
26 Establishing Neuropsychology in a School Setting Organizatio n, Problems, and Benefits RUTH ADLOF HAAK
At this time, it is not standard practice to operate a neuropsychological testing component in public schools. This chapter traces the problems of introducing such a component into the public school; the model of operation that is presently proposed for such a component; the emerging, though still unclear, benefits of pursuing this task; and the general conceptual model of neuropsychological underpinning that seems necessary for an adequate public school, based on our experiences of neuropsychologically assessing public school students for a decade. The operation to be described is that of the Balcones Special Services Cooperative, a special education service unit shared across school district lines in Travis County, Texas. The special education unit is relatively small, serving a student population of about 1200 special education students. Excluding the teaching staff, the central staff of this unit numbers approximately 20 persons, primarily assessment personnel but also including some occupational therapists, physical therapists, counselors, and special itinerant teachers. The unit is directed by a combination licensed psychologist-special education director whose formal training is in counseling psychology and whose subsequent training in neuropsychology was self-acquired postgraduation. The situation is thus fairly representative of many special units across the United States that may wish to contemplate introducing a neuropsychological testing component into their public school assessment capabilities; i.e., staff members are well trained in assessment and in their
RUTH ADLOF HAAK • Balcones Special Services Cooperative, Austin, Texas 78746.
separate functions but more interested-at least in the beginning-than skilled in neuropsychology. Neuropsychological assessment capabilities were introduced into the above special education situation at a time when grant monies were more available for innovative educational practices (10 years ago). This was only one of several innovations introduced into the system at that time, the purpose of all of them being to increase expertise in the assessment and subsequent programming of special education students. That remains the purpose of the neuropsychological testing component today, which has operated without interruption since its introduction.
Operational Model and Operational Issues Balcones Special Services Cooperative (BSSC) serves three (formerly five) school districts that view this service unit as their primary available resource for dealing with children who are experiencing serious learning problems or emotional/behavioral problems in school. All schools served by BSSC do have regular counseling components as well as special state and federal programs to serve their students. Nevertheless, any student who is experiencing serious difficulties in these schools will probably be referred for assessment and possible services to special education. We believe this attitude is proper because so many children are discovered to have bases for their difficulties when they are properly assessed. "Proper" assessment certainly can include neuropsychological assessment in many cases. Because special education was already the com489
490
CHAPTER 26
prehensive assessment unit in our schools, and for utility, the neuropsychological testing component was placed within the province of special education when it was introduced into our system. If children were indeed to be identified as having neuropsychological problems, it would be hard to imagine how they would not be qualified for special education. Placing neuropsychological testing within special education was done to effect a more logical flow of services. Utilitywise, it also effects a much higher degree of time saving and efficient testing, because the testing required by special education already covers much of the territory of the standard neuropsychological exam, specifically: the administration of the Wechsler scales, the gathering of health history, the gathering of developmental milestones information, the gathering of information from observers with whom the students interact (family and teachers), and consideration of emotional and sociological factors.
The Operational Model There are two possible routes of referral for neuropsychological assessment in the special education operational model to be discussed here (see Figure I). One course is that neuropsychological testing be completed as part of a new referral for special services. The second course, much more common in our experience, occurs when someone within the special education delivery system refers an alreadyplaced special student for neuropsychological assessment. The latter occurs for a number of reasons: it takes a while for teachers and staff to become acquainted with a student's total behavior and to notice certain signs of possible neurological/neuropsychological dysfunction; a student receiving services may begin to experience new difficulties and a neuropsychological exam is desired to rule out certain possibilities; a student may be referred for an emotional evaluation; or the need for neuropsychological assessment may emerge in a review of a student's case. From whatever source, once someone within the system decides that a neuropsychological exam is in order, such a request is made through the managing diagnostician of the child's case. Most often, it is the managing diagnostician who makes this referral. The diagnostician completes a short "Neuropsychological Referral" form and submits it to the director. The form is primarily designed for the referring diagnostician to pose direct questions that he or she wants (or hopes to have) answered by the neuropsycholo-
gical testing. The director reviews these questions, may suggest others or better ways of looking at the questions, initials the form, and returns it to the diagnostician, who then schedules the neuropsychological exam with the neuropsychological testing technician. When the technician has completed the testing, the neuropsychological test results are given to the managing diagnostician. At this point the diagnostician has a choice: (I) interpret these data, write a short neuropsychological report, incorporate the findings of this report into the student's comprehensive assessment, and present these data to the next committee meeting regarding the child: or (2) request the director or staff neuropsychological consultant (a diagnostician who has completed special study in neuropsychological assessment) to interpret the neuropsychological data, after which the managing diagnostician then continues the process as above. If the neuropsychological (and other) assessment data indicate that a referral to a pediatric neurologist is in order, the managing diagnostician can do so, including writing the referral letter with attached information, or may, as before, request the director or neuropsychological consultant to complete this step. If a referral is made to a neurologist, the finalization ofneuropsychological information in the comprehensive assessment report awaits, of course, the outcome of the medical tests. Parents are requested to follow through on medical referrals; however, if parents are unable to do this, BSSC pays for these neurological exams. It is very important to include the monies to pay for medical exams in one's budget; otherwise, it will often be impossible to carry a child's assessment to proper completion, and the time spent giving the neuropsychological exam will not have been productive. It is frustrating and professionally inadequate practice to have to abort a line of inquiry that appears headed toward helping solve a child's problems. The Comprehensive Assessment Report, into which all neuropsychological and neurological findings are incorporated, lays out suggested methods of intervention with the student, both instructionally and behaviorally. These methods are discussed in committee meetings that either determine whether the child will receive services or review the services the child is receiving. Any drug or medical treatments are also outlined in these meetings, including specification in the child's Individual Educational Plan (IEP) of parents' and school nurses' responsibilities regarding medication. If the neurologist has requested feedback from the school, the person to
ESTABLISHING NEUROPSYCHOLOGY IN A SCHOOL SETTING
491
Referral sources: Outside professionals
School personnel
The staff psychologist(s)
Special education committee reviews
Parents
-~~~ Neuropsychological consultant Interpretation of test data ~------------~
I
~----------------~
Managing diagnostician Director of the of the child's case 1-----+1 neuropsychological unit (or assessment unit) ~ Posing of test questions Interpretation of test data
~
~~.----------,
Neuropsychological testing technician Actual testing
~----------------~
Pediatric neurologist (?)
/ Short neuropsychological report Incorporation of neuropsychological data into Comprehensive Assessment Recommendations for remediation or programming
l Special education committee Final recommendations for remediation or programming FIGURE 1. Operational model of a neuropsychological testing component within a special education department.
supply the feedback and its form are also specified in the IEP. All special students' IEPs must be routinely reviewed by committees. In these reviews, neuropsychological testing is often first requested. It is also in these reviews that methods of intervening with the child's neuropsychological deficits are discussed and changed, if necessary. It is also not unheard of for such a child to pass through a particularly difficult developmental period and to regain adequate functioning, whereupon he or she may be dismissed from special services. In short, the special education committee deals with the implications of neuropsychological testing, as it must deal with all assessment information. There are other sources for neuropsychological referral besides those given above, for example, the managing diagnostician, the special education committee, or a teacher within the system. Many referrals come from the person completing an emotional as-
sessment upon finding indications of possible neuropsychological dysfunctions; the occupational or physical therapist may request a neuropsychological examination; outside professionals in private practice-particularly psychiatrists and psychologistswill make referrals to the school for neuropsychological testing; and even parents occasionally make such requests, especially if they are neighbors of a child who was so assessed and whose assessment led to improvement in behavior and school attainment.
Funding and Personnel One hears objections, voiced in professional gatherings, to the utilization of neuropsychological testing based on its high use of time and money. This is not a real issue within special education. The only person we have found it necessary to employ in order to operationalize the neuropsychological testing
492
CHAPTER 26
component is a half-time neuropsychological testing Problems in Instituting technician. The BSSC model retains the use of such a Neuropsychological Testing in the technician, as is customary in neuropsychological asPublic School Setting sessment. The alternative would be to attempt to teach all existing assessment personnel the skills of The major problems of instituting a neuropsyneuropsychological assessment. This would indeed chological testing component in the public school's be time intensive; in addition, it might involve trying to teach these skills to persons who do not wish to assessment capabilities are much less obvious than money, personnel, or settling upon a working model learn them. It is more efficient in practice to have one of operation. Furthermore, the major problems will person conducting all neuropsychological exams not be immediately apparent. Some arise only after a than to take large blocks of time from existing perrelative degree of familiarity and success with the sonnel's schedules. new system. Many will be realized in retrospect. There are other reasons for utilizing the services Many are not going to be resolved in the near future. of only one person to conduct the neuropsychological To institute such a component into the public tests. When instituting a new testing system, probschool's assessment system is somewhat to do the lems of reliability of test results are greatly miniproverbial "rushing in where angels fear to tread." mized by using a lone tester. A whole group of assessment personnel working with a new set of tests may obtain dubious results for quite some time. Also, Developing Staff Expertise if there is a subsequent attempt to subject the findings to research, then the use of one testing technician will Obviously, the first major problem to be faced have been a great advantage. by a school instituting a neuropsychological testing We have had only two technicians in our decade capacity into its system is the training of on-board of practice. Both were locally trained and then re- personnel. Some schools will be able to hire a profesviewed by professional neuropsychologists outside sional neuropsychologist to oversee the institution of our system. Many neuropsychologists report, and we such a system and to manage the training of existing have also found, that the personal attributes of the personnel. This would certainly be the logical way to neuropsychological testing technician are of more begin. importance than credentials. One of our technicians At the time we instituted our system, in a medihas a college degree; one did not attend college. Both um-sized American city with one of the largest unihave children and are very adept at relating to chil- versities in the nation, there was no professional dren and obtaining the best possible performances neuropsychologist in practice in the city; nor was from children. Both are acutely observational and there a course in neuropsychology at the university. often make notes on test protocols that prove of more Although this is no longer true, many other school worth than the test results. Both have insisted on systems may still find themselves in that position. So working only half-time, and perhaps this has contrib- the training of existing personnel is a large task. uted to their low level of "burnout." Funding for the This problem was ameliorated for us by the provitechnician is provided by our participating school sion of sufficient grant monies to attend many frrstdistricts, just as it is for any other member of the class, out-of-state and in-state professional training special staff. seminars. We were also able to obtain early in our The initial expense of instituting the neuropsy- development an on-site visit and review of our system chological testing component involves, of course, by a nationally known neuropsychologist (Lawrence purchasing the necessary equipment. Though this C. Hartlage). As professional neuropsychologists equipment is more expensive than other types of test- have moved into the city, we have been able to build ing equipment, the increase in expense is only rela- relationships with some of them. And we have, over tive. We purchased our equipment at the end of one time, established quite productive working relafiscal year and the beginning of the next, spreading tionships with some pediatric neurologists, which is very helpful. All these opportunities have been invaluthe expense over two years. Though certain items of neuropsychological able sources of support to our developing neuropsytesting equipment are reputed to be fragile, particu- chological assessment capabilities. larly the Halstead-Reitan Category machine, we If it is not possible to hire a professional neurohave not had that experience. OUr equipment has psychologist to institute the public school's neurofrequently been moved about for 10 years, with only psychological testing component, and if there are no minor repair problems. local universities teaching formal neuropsychology
ESTABLISHING NEUROPSYCHOLOGY IN A SCHOOL SETTING
courses, then there is just no substitute for sending staff members to professional training workshopsand a lot of them. Also, at least some outside review of the developing system may well be required. Neuropsychology is one of the more complex fields in psychology, as is well known. A sufficient knowledge base of this area to support the introduction of testing capacities is the first problem that must be faced. If this cannot be managed, the system probably should not be instituted.
Selecting Staff to Participate Because the issue of training is so obvious and even oppressive in nature, existing staff in a school assessment setting must be given a real choice about whether to participate in the introduction of a new neuropsychological testing component. The major reason for this is that even with the best of institutional support with regard to attending professional training opportunities, a person entering this area will have to devote an exorbitant amount of personal time to individual reading and study in order to begin to be competent in neuropsychological test interpretation. It is far better to have an honest refusal from a staff member to make such a commitment than to have that person either abandon the pursuit (after spending valuable staff funds for training) or to be satisfied with a cursory and superficial knowledge of the subject. This is no ordinary "in-service" undertaking! It is probably wise, therefore, for the director of the assessment unit to select a limited cadre of onstaff persons to begin instituting the new testing component. Other staff members can (and should) be kept abreast of the developments in this area through general in-service staff training sessions. It should be understood by this new group that a great deal of personal commitment is being called for; and it should also be understood by the larger group that a high percentage of the total stafrs resources for attending professional training opportunities will go to these people for awhile. Because the beginning neuropsychological "stafr' should be kept small, it appears that the smaller school system is at no disadvantage in introducing neuropsychological testing into its assessment system.
493
Inevitably, people leave and enter systems. New members of the staff will need to be added. Also, current staff members who at first considered themselves uninterested in neuropsychology will become interested. The original group in the new neuropsychological team will hardly be in midstream in their pursuit of knowledge when others will want or need to join. There are at least two methods of handling this problem. One is to have the original staff, along with their leader, conduct in-house workshops in the rudimentary knowledge they are acquiring. This is helpful to the general staff and provides an entry point for new and more involved staff. A second method is to maintain an ongoing study group in neuropsychology within the assessment staff. This method is mostly set up for the original staff and for the acquisition of new knowledge, including the study of current cases being assessed. The two methods serve somewhat different purposes, and we have employed both. New members added to the core group involved in neuropsychological assessment need to be able to attend professional training opportunities. It is therefore judicious not to expand this original group too quickly' or funds will not be sufficient to provide the necessary breadth and depth of workshop opportunities. What the above infers is that much of the concentrated study required for a staff to take on the new area of neuropsychological interpretation must be provided on "company time." It is unrealistic for any organization to expect its employees to expend the amount of energy that will be required to learn a system as complicated as neuropsychological test interpretation without an equal amount of dedication in time and funds from the organization itself.
Sustaining the Team's Motivation
Remarkably, those who choose to involve themselves with the pursuit of neuropsychological knowledge seem to have more intrinsic motivation to continue this pursuit than is observed in many areas. Nevertheless, this is a difficult area to enter: The pursuit seems unattainable. The more one learns, the less one knows. Cases one considered understood tum out to be enormously more complicated than thought at first glance. There is remarkably little supBroadening the Range of Involved Staff port for the public school psychologist or diagnostiThough problems of quality control can be kept cian performing neuropsychological testing and test at a minimum when instituting the new neuropsy- interpretation. Initial successes are very important chological testing component by keeping the number (as, for example, our discovery of three subclinical of involved staff low, this cannot continue forever. seizure cases in one family); but this initial self-confi-
494
CHAPTER 26
dence can erode with less spectacular results. Some methods we have used to attack midstream malaise are the following:
tern. Though some of the following methods may strike one as indirect, they seem to work best in our experience:
I. Put on general workshops for the teaching staff. In preparing for the workshops, one is forced to review concepts that have slipped the mind and one is forced to organize what one knows. The general school staff has some degree of sophistication and interest these days in neuropsychological functioning. If such workshops are solidly done, they can be very helpful to all concerned. Especially helpful are general workshops on basic brain functioning, attention deficit disorder, and physical problems (including seizures) that may present as behavioral problems. 2. Send one or two core staff members to a professional workshop. People always return from such a venture with new information and new enthusiasm. 3. Visit other professional neuropsychologists, those in private practice or in nearby universities. Invite them to visit your establishment. These visits create networks that eventually will prove helpful and beneficial. 4. Arrange a consultation with a neurologist with whom you are working on a child's case. This will help both parties to understand and to manage the child better. It will also lead to much more open communication and trust regarding future cases.
1. First and foremost, do the best possible job of conducting neuropsychological assessments. Avoid fanfare about the new system. Just do the job of conducting these assessments in the most careful, professional manner possible. When neuropsychological test results are fed back to staff members or the special education committee (which includes administrators and parents), use careful, nonneurological language and concentrate on how the assessment leads to ideas for intervention. There is no other method for getting support for the new testing system that will even come close to being as important as this method-do a careful, competent job on each assessment. 2. Always report neuropsychological test information to the staff members who can benefit from this information. Testing that is not reported is useless. 3. Always make practical suggestions based on the neuropsychological information. This does not mean that remediation suggestions always need to be made: one may find that remediation is not necessary. One of the major benefits of neuropsychological testing in the school is to rule out problems, as well as rule them in. But in all cases, neuropsychological assessment should lead to some kind of increasingly appropriate action-or else why do it? 4. Concentrate on the reporting of cases with administrators rather than on flashy explanations of neuropsychology. Administrators always know who their problem children are; they have multiple sources of input. Nearly all administrators' ultimate concern is the children. Solving a few difficult cases will go further than any other measure to keep support for the new assessment unit. 5. Be careful and patient with parents. No parent wants to think that something serious is wrong with his or her child. This realization, even if it is true, must usually come slowly. The staff will need to be prepared for several sessions with the parents regarding a child's neuropsychological problems. Again, avoid neurological and technical jargon. Do not try to urge parents into seeking a neurological exam with too defi-
There are long droughts in the pursuit of neuropsychological expertise. These become hard for a staff to tolerate. Inevitably the best cure for this problem is to persevere: a real breakthrough case wil1 come along, a quite difficult child will be better understood and managed, and morale will rise again for a long time.
Obtaining the Support of Others If the introduction of a neuropsychological testing component into the public school's assessment capability is to obtain credibility, the support of many nonstaff persons is needed: school administrators, board members, school staff members, and parents. These people are usually not going to attend workshops, nor do they have the patience to listen to long explanations about neuropsychology. This leaves one in the position of wondering how to get across to this larger public the benefits of the new testing sys-
ESTABLISHING NEUROPSYCHOLOGY IN A SCHOOL SETTING
495
nite ideas about what is wrong with the child to develop neuropsychologically based intervention neurologically. If a basic attitude of concern strategies (Rourke, 1985). One is operating in a new is the attitude conveyed ("This is something area-conducting neuropsychological assessment we need to check out"); if the problems are and intervention in the public school. This causes, or discussed as possible problems; if the par- at least relates to, a number of concerns. ents are convinced to take their child to a First, there is the matter of norms. Norms exist neurologist and no physical problems are in various sources for normal children, children seen found, the parents will still appreciate the in a clinical setting, hospitalized children, and chilconcern but be relieved that nothing terrible dren seen in learning-centered clinics. Satisfactory had been discovered. If a physical neu- norms for a wide range of public school children rological problem is discovered, they will be referred for a broad range of learning and behavioral even more appreciative of the careful type of problems are not easily obtained. assessment done in their school. But in either Second, and related to the first, is the matter of case, the school assessment staff will have the strongly deficient neuropsychological perforacted responsibly. mances of many public school children. These chilIf, on the other hand, the school assess- dren often score as poorly as do diagnosed, sick chilment staff has gotten heavy-handed (and be- dren; yet they are functioning in the public schools, in yond its depth) with neurological jargon and some fashion or other. Often after extensive pbysical there turns out to be no identifiable neu- examinations, they are still not found to have estabrological problem, the possibilities for polit- lished disorders. The opposite also occurs: children ical ramifications are real. Such an event with obvious neuropsychological signs from intelleccould conceivably end neuropsychological tual test results and behavior will have few if any testing as part of the school's assessment findings on neuropsychological tests. In short, the capabilities. inferences one has learned often do not hold up in this setting. Neuropsychological testing in the public Signs one has been taught to consider as pathogschools is a new venture. If it is to receive broad nomonic often are not indicative of diagnosable pasupport, it will need to proceed slowly and most cauthology. One catches oneself disregarding such tiously. The participating assessment staff, in particsigns, often to one's grief. Contradictory results, ular, needs to have this attitude strongly impressed rather than recognizable patterns, become comupon them. monplace. Obviously, ignorance and inexperience with neuropsychological data could be the culprit in The Lack of a Body of Public School many of these cases; however, even when these cases are referred out, often inconclusive findings result. Knowledge Children assessed before the age of 9 often do not Not only does the beginning neuropsycholo- appear to have the same level or pattern of problems gical assessment team in the school lack its own body as they will when reassessed later. Obviously, these of knowledge and experience, it cannot tum to a children change developmentally; however, their broad body in the literature. Obviously, this is not to patterns of brain dysfunction should not change radisay that expert knowledge in the field of child neuro- cally over a short period of time, if indeed brain psychology is not available. As examples of the work dysfunction is being adequately measured. being done in child neuropsychology, one can cite The above are only a few of the sources of conBoll (1974), Gaddes (1980), Golden and Anderson fusion encountered as one examines more and more (1979), Hartlage (1980), Hynd and Obrzut (1981), public school children using neuropsychological testKinsbourne and Caplan (1979), Knights and Bakker ing methods. Naturally, of course, sometimes stan(1976), Lezak (1976), Reitan and Davison (1974), dard teaching and standard inferences lead to producReynolds and Gutkin (1979), Rourke, Fisk, and tive diagnoses, better management, and better Strang (1983), and many others. These and other education. This happens, in fact, a great deal. But the sources within the field of neuropsychology form the times that it does not happen provide the sources of knowledge base that the school neuropsychologist concern for the person seeing large numbers of public does use. school children receiving neuropsychological assessBut there is no cumulative, broad experience ment. regarding neuropsychological assessment in the pubObviously, there is something about the public lic schools themselves. Only now has an effort begun school setting itself, its sequenced and institu-
496
CHAPTER 26
tionalized contexts and expectations, that interacts both with normal developmental stages and with normal and abnormal neuropsychological functioning. Our present conclusion is that neuropsychological testing in the public school can only be conducted and utilized as part of a total, comprehensive. and contextually related assessment system. Gaddes (1981, p. 28) arrived at this conclusion earlier. When a child is found to have neuropsychological dysfunctions, invariably there are related developmental issues, difficulties with achievement, misunderstandings between parent and child over expectations, misunderstandings between school and child over expectations, perhaps even worse disabilities in other areas, and failure on the school's part to adapt to the child. Most children assessed by neuropsychological testing display a set of nebulous findings calling for multiple adjustments in school, home, and child. When a child is assessed by neuropsychological testing, is found to display clearly recognizable signs of possible problems, is referred to the neurologist, is found to have abnormal physical findings, is placed on medication, and improves markedly in behavior and academic functioning, the nice linear process one hopes to see in good assessment has actually occurred! Thankfully, this does happen rather often, but still in only a minority of cases. The rest of the time the public school neuropsychologist is operating within the realm of general systems theory-with a 360-degree set of facts and interactions, the inferences for many of these facts and interactions as yet unknown. That will probably continue to be the state of the art in public school neuropsychological assessment for some time to come.
Benefits of Instituting Neuropsychological Testing in the Public School With such an array of problems, one may wonder why anyone would attempt to carry out such a venture. Fortunately, even though some of the benefits are only beginning to come into focus, there are even at this point certain observable benefits.
The School Is an Ideal Test Site Though the school lacks some of the advantages found in clinical test settings, it more than makes up for this with some of its own unique advantages. For one, the children are present every school day for six or seven hours. The school assessor has many lines of
constant feedback about any particular child. There is a huge ''information edge. '' The assessor also knows the various contexts in which the child may be found, for example, facts such as who is a tolerant teacher and who is a rigid disciplinarian, problems that are endemic to certain campuses, and so on. These contexts do affect the way a child behaves and can be taken into account. If a child's behavior changes for better or worse, especially within certain contexts, the school assessor can easily know this. The child can be constantly monitored and a behavioral history established. Finally, assessment on site becomes a constant rather than a one- or two-shot process. When breakdowns occur in behavior, further assessment can be done (and nearer the point of the breakdown). One does not have to give an initially excess number of tests in order to cover every possibility. Tests can always be added as needed, with relatively little difficulty in scheduling and at no cost to parents. In a true sense, assessment of a child in the public school setting is never considered to be completed. Assessment becomes an ongoing process, with relatively easy access to the child.
Student Benefits Are Extended Probably the most important reason for conducting neuropsychological testing in the public schools is that a large number of children can be assessed and benefit from this assessment-including many who would not otherwise be tested. Knowing the symptoms of neurological or neuropsychological dysfunction in children is not within the average parent's training. Taking the child on one's own to a pediatric neurologist is not within the average parent's range of actions. The very children who are most likely to exhibit symptoms of neurological or neuropsychological dysfunction are those whose families are least able to afford first-rate medical care. Therefore, the children most likely to need neuropsychological assessment are those least likely to get it-unless the public school provides the service. Finding causes for children's learning problems can be a most rewarding social venture. It is the children who profit.
Children with Neurological Dysfunctions Are Discovered When one can amplify the standard testing done in the public school setting with neuropsychological testing, one does indeed begin to discover a large
ESTABLISHING NEUROPSYCHOLOGY IN A SCHOOL SETTING
number of children with neurological problems. Especially is this true, in our experience, with subclinical seizures. Subclinical seizures, if our population is at all representative (in fact, the average socioeconomic level is higher than average), are much more common than suspected. Usually the behavior and learning of these children change radically once they are properly identified and medically treated. No longer are they viewed as mere candidates for behavior modification. The dysfunctions, disabilities, or diseases that the children are "allowed" to have are no longer limited to what the teacher or assessor knows. We have found cases in a broad range of neurological and neuropsychological disorders in our decade of performing neuropsychological testing. Only a few of these children would have eventually been assessed outside the public school setting.
School Personnel Become More Scientific The effect of having neuropsychological testing capabilities in the school and discovering many children who will be better understood and programmed through this testing is to increase the scientific perspective of the staff. Especially in special education, the staff returns to searching for causes instead of merely applying its range of known treatments. It is beyond the scope of this chapter to consider the trends in special education that have in the last few years appeared to cast special education back upon the "underachievement" concepts of the 1950s and away from the examination of brain variables. Nevertheless, when one begins to measure special students with neuropsychological testing procedures, one is cast back inevitably upon brain and cognitive concepts for explanations-because the data force it. This redirection is especially beneficial to the child, for it takes school personnel out of moralistic and motivational thinking patterns and moves them into the more objective patterns of scientific thinking about causes for the child's behavior. It is axiomatic that what one understands reasons for, one deals with much more effectively.
Assessment for Emotional Disturbance Is Oarified Emotional disturbance is a large and undelineated concept within the general area of children with problems in school. Even within the defmition of the federal law, emotional disturbance is often seen as that condition of a disturbed child when the causes of
497
his or her maladaptive behavior cannot be identified.
In this sense, neuropsychological testing is an enor-
mous aid to the proper identification of children who are emotionally disturbed from those who may be experiencing secondary emotional distress. At BSSC, we do not call a child emotionally disturbed unless the causes of the maladaptive behavioral reactions lie clearly within sociological or human interactional areas (or, of course, in those known mental diseases for which there is presently no known cause, such as schizophrenia). This means that if a child is identified as having, for example, an attention deficit disorder, the impulsive behavior, lack of reflectivity, general refusal to accept responsibility for his or her actions, and difficulties with peers do not constitute emotional disturbance. These are known and customary correlates of attention deficit disorder, even though they are somewhat emotional in nature. Nor do we call the periodic explosiveness, the circumlocution, the tendency to be overconcemed with others' behavior, or the compulsive writing of the temporal lobe epileptic emotional disturbance (or borderline personality disorder). In short, when there is a diagnosed neurological or neuropsychological condition, and the common correlates of this condition are known, these expected emotional correlates of the problem are seen as part of the problem-not as emotional disturbance. The children may still get the same treatment as they would have gotten with a·label of "emotionally disturbed": they will receive counseling, family counseling, behavior modification, special programming in an outdoor education unit, or whatever is called for-but the problems will be associated with the neurological status and not with the emotional status. We have found the above approach to greatly "clean up" both the diagnosis and subsequent treatment of emotional problems. It also greatly cleans up the attitudes and actions of those who educate or parent the child. Much of the "blame" for the child's actions is removed from the child and replaced with a more scientific and supportive approach. Again, reasons for the behavior make all the difference to the adults who interact with the child. And they make that same difference to the child, who is on the receiving end of these interactions. As this scientific attitude spreads, all levels of the staff become involved. For example, after a summer workshop last year on the subject of attention deficit disorder, a teacher called late one night to say that she had just received word from parents (whom she had herself suggested take their child to a pediatric neurologist) that the child was diagnosed as se-
498
CHAPTER 26
verely hyperactive and put on medication. She was very proud of the fact that she had known the right questions to ask the parents, for example, questions about sleep disturbance. She had a right to feel proud of herself. A neuropsychological testing component in the public school inevitably forces the entire staff to become gradually more scientific, that is, more accurate and just in their education and treatment of children. This is rewarding for all concerned. One more child had escaped being seen as merely a behavioral problem. Adult interaction with that child begins to support, not impede, development.
(/)
c:
.g ~
c:
u
E:g' Q)._ 0 ~~ ::J-!2 (I)
...
Q)
0
!
Culture
Central model Pathology
.s::.t-
Brain Ill I:
Central model
0
Pathology
I:
Our decade of experience in neuropsychological testing has led us to a number of tentative conclusions regarding the public school. One repeated findingthat it is difficult if not impossible in many cases to differentiate deviation from disability (an issue to be discussed in the next section)-has led us to look at the adequacy of the school itself. It is often the rigidity and lack of range in the educational program that apparently moves what begins as a deviation in the child into the realm of a disability. If, as Plato might have suggested. the major working material of the school is the human brain, should not the school be the brain writ large? The writer has proposed a model of an adequate neuropsychological substructure for a public school (see Figure 2). It should be emphasized that this model presents a substructure-an infrastructure. It serves as a check on the adequacy of the total program of the school. A wide range of program elements can satisfy the various demands of the model. Also, schools may differ in what they need to develop to satisfy the model. For example, one of our own school systems is presently making massive efforts to increase its range of deviant offerings. The model for a Sound Neuropsychological Substructure for a Public School acknowledges that learning takes place within a three-variable interaction: the human brain, the emotional status of that brain at any time, and the environmental stimulation that is supplied by the culture to the brain. In a scientific sense, the brain and the emotional status are the necessary conditions for learning; the culture provides the sufficient condition for learning. Within each of these three major components of the learning process-the brain, the emotional con-
Deviation
"+'
'6
Conceptual Model of a Neuropsychological Foundation for a Public School
Deviation
8
c:-a
Ill .:
I: Ill ... Q) ttl u Q) G)I: ... (I)
Emotional status
Q)~
Deviation Central model Pathology
~
In all areas there are
the goal for which is
• a centrality (central model, central tendency) which is positive, within limits • deviation which is positive, within limits • pathology which is negative to any degree
provision allowance, tolerance, provision reduction, elimination
FIGURE 2. A neuropsychological substructure for a school system.
clition, and the culture-there exists a central tendency (or model), with deviations from the model that provide for evolutionary possibilities and differences from the model that are pathological in nature, the result of insults or errors. Both the central tendency and the deviations are normal, within certain operational limits. Both can become pathological outside these limits and can constitute, along with the results of insult, accident, genetic error, and so on, the pool of pathology. The goal of education is to provide, through the culture, the development of learning in the individual. This is accomplished through proper utilization of the individual's brain in a climate of emotional stability. Education needs to be carried out in both the central tendency model and the productive deviations model, while attempting at the same time to modify any pathology and to enlighten about possible pathologies. For example: A child who is of average intelligence and of sound emotional health for his or her age is operating in the central model, both in the
ESTABLISHING NEUROPSYCHOLOGY IN A SCHOOL SETTING
brain and in the area of emotional status. If that child is exposed to the central model of education in the school (the culture), and that model is an adequate model, all should go relatively well. If that school is a deviant school-for example, a special private school with a unique philosophy of education-all will still probably go well, though the emerging child may be expected to be somewhat different in orientation from the typical public school student. If that "average" child is, however, for some reason exposed to a pathological educational system, that child may take on certain pathological behaviors. In all the above examples, the child has had the necessary conditions for learning; but in one of the examples, the sufficient condition for satisfactory learning was not there. In all cases one can think of, these sufficient conditions will vary: this variation is of no consequence unless deviation is not provided for, the central model is outside its limits (either too strong or not strong enough), or pathology is definitely present. Let us consider another example: Suppose the individual is a deviant child with a nonverbal IQ of 130 and a verbal IQ of 95. Despite neuropsychological testing and a careful medical exam, with complete history, no basis in pathology can be found for this difference. To make the case even stronger, virtually all the father's relatives seem to fit the same pattern: at least in behavior this seems to be so (all are in active occupations; they hardly speak a word at school meetings). The father is called "Gary Cooper" by the mother. The child has a deviant brain. If the family is overconcemed with standard cultural expectations, this child may not be operating in the central tendency of mental health. However, let us suppose this family thinks the child is wonderful. We then have a deviant child in the central model of emotional health. What the culture does to the child in school may then become critical. If the school tries to force the child to learn exactly in the central model, the child will have difficulty and become discontent. The discontent will amplify the difficulties. What is required is for the school to have a sufficient supply of deviant possibilities in its curricula, its methods, its materials, and its routes to graduation so that the education of this particular child can proceed "normally." A final example: Suppose that we have a pathological child, for example, an autistic child. The causes are unknown. The brain of this child is assumed to be pathological. Despite a "good" home, the emotional status of the child is pathological. The school (culture) inust now address the pathology if the child is to learn anything, central or deviant. Furthermore, when the causes of pathologies become known
499
through discovery, the school has the responsibility to enlighten all its clients about these causes so that they may be avoided in the future. Special education deals with pathological children. (Too often nowadays it also deals with deviant children and deviant children made pathological by the culture's failures.) The whole school needs to deal with instruction about pathology (as well as dealing with pathological children to the degree that their lives can be modified more toward the central model). Certainly the culture becomes pathological partly as a cumulative function of the pathologies of the individuals within it. But more than that, the culture (and the school) becomes pathological when it enforces a too-rigid adherence to its central model or when it fails to provide what is necessary for the central model in the society. The deviant person is actually better off when the central model is insufficient than when it is too rigid; but in that case, the nondeviant usually pay the price of insufficient education. As the model proposes, both central tendency and deviant brains need to be seen as ''normal''; both require emotional stability in which to learn; both should be offered an appropriate cultural menu to enhance their capabilities. In the present time frame, pathology is actually being rather well addressed by the public school. The school reforms sweeping the country are beefing up the central tendency. What is not being tended to presently in our experience and the experience of many others is proper provision for the educational needs of the deviant child. For the culture to be able to function through its schools in a neuropsychologically sound way, we do not anticipate that a major revolution is required. Walls need not be tom down. Whole programs do not require introduction or abandonment. Much of what is required is in place. The model of neuropsychological adequacy we have presented provides a basic infrastructure for the school-a guiding mechanism. It provides a way that a school can check the adequacy of its range of proceedings and offerings. Where ''holes'' exist in the content of this substructure for a particular school, there that school requires additional resources. Usually this will be found to be the area of supplying a proper range of individual opportunities for deviant types of children-without making these children feel emotionally any different. In an overall sense, the following will be required in any school in order to satisfy the demands of this neuropsychological model:
1. A comprehensive and expert (though not cumbersome) identification (assessment)
500
CHAPTER 26
system for measuring the child's abilities in a nomonic -signs, starkness of findings in children fairly wide range of areas functioning beyond prediction levels, and so on). 2. A faculty educated and operating within Obviously we need to know a great deal more about sound mental health principles (i.e., a men- public school students and their neuropsychological tal health "curriculum" for the school), conditions before we will be able to draw accurate with mental health supportive services for inferences about many findings in our test results. students (and perhaps faculty) One example may suffice: Any physician can identi3. A faculty educated and operating within fy a child with measles. But if that physician steps sound teaching methods, including teaching outside his or her office and finds a whole population methods for the deviant covered with red spots, either everybody has measles 4. An articulated, systematic curriculum (cul- or something else is going on. This dilemma faces the tural menu) that contains the following: (a) person conducting neuropsychological testing in the the customary central offerings (3R's, tradi- public school. Most of the inferences he or she has tional curriculum, in Texas-the "essential learned were gleaned from clinical cases. These inelements"); (b) a range of deviant (indi- ferences are often still highly accurate. But when vidual) offerings; (c) special education (for neuropsychological inferences drawn from certain pathologies of the brain or emotional facts do not lead to the expected conclusions-as condition) often seems the case in the public school-then the assessor faces a dilemma. This dilemma, of course, has to do with the hardly known or unknown. The public school neuropsychologist needs a lot Conceptual Issues: Neuropsychology of company: researchers, professional neuropsycholin the Public School ogists from the private sector, university faculty, and graduate students, to name a few. But this company Operating a neuropsychological assessment canecessarily needs to work continuously in this setpability within the public school is not a simple matting-to share the same experience. So far little of ter. We have addressed the operational problems this type of cooperation has been available, and when associated with that endeavor, as well as the operait is, it is short-lived. The public school needs to be tional benefits. We have also pointed out, in proposviewed as an appropriate arena for professional ing a conceptual model of neuropsychological underneuropsychologists. The research needs of this setpinning for a public school, that neuropsychological ting are extreme, and the service personnel presently functioning in the school is dependent on more than working in the schools will simply never have time to merely neuropsychological assessment. pursue research activities. At this stage in our experience the following appear to be some of the larger issues: it is still very difficult to draw neuropsychological inferences Differentiating Deviation from Pathology about a population so little studied to date; it is diffiObviously in our model, deviation is not viewed cult in many cases to differentiate brain pathology as pathology. Deviation is viewed as nature's way of from simple brain deviation in a population experi- being sure that the species survives under unforeseeaencing difficulties with the structure of the public ble circumstances. It is often difficult in neuropsyschool system; it is difficult to provide appropriate chological assessment to tell the merely deviant from educational methods for a deviant or pathological the pathological. Testing procedures of the future brain; and the issue of the ''general practice'' of need to build in controls for this necessary differneuropsychology in the schools from the expert entiation. It has been our experience that adjusting "specialized practice" of neuropsychology in the the school curriculum for a child with deviant neuroschools is only emerging. psychological test findings is the most potent way to help that child in many cases. Square pegs simply cannot be forced into round holes without emotional Neuropsychological Inferences in Public abrasion. At present, we have to rely on family histoSchool Populations ry, statements a parent makes about "how I was," The operational problems associated with mak- preferences, and behavioral data in order to try to ing inferences about the neuropsychological func- differentiate normal deviation from pathology. This tioning of public school students have been is not such a difficult problem when the child has an mentioned (proper norms, overoccurrence of pathog- 80 IQ (i.e., the child is seen as merely deviant when
ESTABLISHING NEUROPSYCHOLOGY IN A SCHOOL SETTING
501
scenario could be the school "neuropsychologist"but it need not be. The professional person with a proper sense of perspective is not going to take on a whole new field of knowledge without some proper degree of caution and respect. Unfortunately, not all persons in a profession operate that way. But as neuropsychology moves into the public schools, if this movement with so much potential to better the lives of children is to succeed, it must move in with this proper degree of caution and respect. The matter of how a psychologist should behave who takes on the new practice of Providing Remediation neuropsychology in the schools is of major concern. The new neuropsychologist or neuropsycholoIf anything outstrips Mark Twain's subject of gical assessment person in the schools can behave the weather for having a lot of talk done about it and like a general practitioner, screening for the most little else, it is the subject of individualized remediation. Despite recent efforts in this regard (e.g., obvious of problems and referring suspected clients Rourke, 1985) and the daily deluge of slick-paper to outside experts. This is an appropriate model for advertisements from education publishers purporting most entry-level persons within neuropsychology in to deliver remediation materials of this or that vari- the schools. Unfortunately, this model will only go ety, getting differentiated remediation delivered to so far. When students are referred to experts in the assessed children is still a major obstacle. Assess- private sector, often these experts do not understand ment far outstrips remediation when it comes to neu- enough about the context of the school to either (1) make proper judgments about the student or (2) have ropsychological progress. any notion of what to realistically recommend once At this point, the creative and innovative teacher the child is assessed. The school neuropsychologist is the best attack on this problem. Teachers must be encouraged to experiment with individual children, cannot avoid moving toward becoming more of an based on the best assessment guidance they can get. expert practitioner. The question of how the school psychologist is Time for sharing ideas between assessor and teacher to take on the role of the professional person in the needs to be made available. Also, when an innoschool with neuropsychological knowledge is a curvative teacher discovers a productive way of working with a child, that wisdom needs to be better preserved rent issue for debate within the profession (for one and generalized to the other teachers. This calls for a discussion, see Hynd, 1981). It will not be an easy full assessor-teacher partnership. The pressures that question to answer. The public school is not the place it used to be, many teachers presently face do not allow for this however. A few years ago it was relatively rare to creative use of time. Perhaps it is time to think about creating a special kind of teacher-an experimental find many doctoral-level faculty members in a public teacher on a teaching team, who is given a certain school. This is not at all rare anymore. So it will be amount of time from her or his schedule to work with with neuropsychology. The children who need the the assessment team. But this teacher should not be a assistance of neuropsychologists are in the public supervisor; this teacher should be the one working schools more than any other place. The movement directly with the children involved or, at the least, toward the creation of the public school expert neuropsychologist is inevitable. Let us be sure as this with the teacher teaching the child. movement is generated that the potential of this sopromising field of science is preserved by our cauThe General versus Specialized Practice of tious and discrete actions, for some day neuropsyNeuropsychology in the Schools chology should make it possible, at last, for every child in school to succeed. When a relatively new practice moves into a new area, the potential for abuse is high. Hardly anything is needed less in the public schools today References than a new breed of "expert" antagonizing the general staff with half-baked and largely nonutilitarian Boll, T. J. (1974). Behavioral correlates of cerebral damage in knowledge about an esoteric subject. This negative children aged 9 through 14. In R. M. Reitan & L.A. Davison often he or she is brain damaged). But it is a difficult problem when the nonverbal and verbal IQs differ by 30 points and neither is below average. One needs to remember this exercise is not merely academic. The child who is deviant is in trouble with the system and is being referred. Better ways to measure both the child and the system seem to be required. A full-scale attack by the profession upon the issue of deviation versus disability would seem to be a fruitful venture.
502
CHAPTER 26
(Eds.), Clinical neuropsychology: Current status and applications. Washington, DC: Winston. Gaddes, W. H. (1980). Learning disabilities and brain function: A neuropsychological approach. Berlin: Springer-Verlag. Gaddes, W. H. (1981). An examination of the validity of neuropsychological knowledge in educational diagnosis and remediation. In G. W. Hynd & J. E. Obrzut (Eds.), Neuropsychological assessment and the school-age child. New York: Grune & Stratton. Golden, C. J., & Anderson, S. (1979). Learning disabilities and brain dysfunction. Springfield, IL: Thomas. Hartlage, L. C. (1980). Neuropsychological assessment techniques for the school psychologist. In C. R. Reynolds & T. B. Gutkin (Eds. ), A handbook for the practice ofschool psychology. New York: Wiley. Hynd, G. W. (1981). Training the school psychologist in neuropsychology: Perspective, issues and models. In G. W. Hynd & J. E. Obrzut (Eds.), Neuropsychological assessment and the school-age child. New York: Grune & Stratton.
Hynd, G. W., &Obrzut, J. E. (1981). Neuropsychological assessment and the school-age child. New York: Grune & Stratton. Kinsboume, M., & Caplan, P. J. (1979). Children's learning and attention problems. Boston: Little, Brown. Knights, R. M., & Bakker, D. J. (1976). The neuropsychology of learning disorders: Theoretical approaches. Baltimore: University Park Press. Lezak. M. D. (1976). Neuropsychological assessment. New York: Oxford University Press. Reitan, R. M., & Davison, L.A. (Eds.). (1974). Clinical neuropsychology: Current status and applications. Washington, DC: Winston. Reynolds, C., & Gutkin, T. B. (1979). Predicting the premorbid intellectual status of children using demographic data. Clinical Neuropsychology, 1, 36-38. Rourke, B. (Ed.). (1985). Neuropsychology oflearning disabilities: Essentials of subtype analysis. New York: Guilford Press. Rourke, B., Fisk, J. L., & Strang, J.D. (1983). Child neuropsychology. New York: Guilford Press.
27 Current Neurops ycholog ical Diagnos is of Learning Problems: A Leap of Faith DANIEL J. RESCHLY AND FRANK M. GRESHAM
Based on our review of the literature, we have reached what may be a startling conclusion to readers of this handbook: Neuropsychological diagnoses of mild learning problems are largely irrelevant, misleading, and potentially harmful because they contribute to beliefs that probably impede rather than facilitate effective remediation. Strong statements? Yes, but entirely consistent with the available evidence, which we believe shows that neuropsychological assessment and treatment based on neuropsychological concepts have little or no treatment validity; rely on unreliable and invalid measures; and impede efforts of nonneurologically trained persons such as teachers and parents to cope with learning problems.
Scope of Chapter: Mild Handicaps The vast majority of the students classified as handicapped and served in special education programs in the United States are mildly handicapped, most often in the category of specific learning disability. Mild handicaps, LD, mild mental retardation, most behavior disorders, and speech/communication disorders can be quite serious impediments to successful school experiences and to successful adjustment in the adult years. The typical characteristics of students with mild DANIEL J. RESCHLY • Department of Psychology, Iowa FRANK M. State University, Ames, Iowa 50011-3180. GRESHAM • Department of Psychology, Louisiana State University, Baton Rouge, Louisiana 70803.
handicaps have been described by several authors in recent years (Algozzine & Korinek, 1985; Gelb & Mizokawa, 1986; Reschly, 1986, 1987a). The major characteristics are: (1) no identifiable physical basis for the problem behaviors; (2) initial identification nearly always occurs after, rather than before, school entrance; (3) problems with literacy skills, particularly reading, usually lead to teacher referral; (4) social skills problems usually accompany the achievement problems; (5) most students with mild handicaps are not officially diagnosed by educational, medical, social service, or rehabilitation agencies after they leave school, i.e., in the adult years. In contrast, the more severely handicapped (e.g., Down's syndrome or early infantile autism) nearly always have physical anomalies that lead them to be diagnosed as handicapped prior to school entrance, often soon after birth, and they are usually regarded as handicapped throughout their lives. Our discussion in this chapter is restricted to the mildly handicapped, particularly LD students. This population has an approximate prevalence of 3 to 4% according to a recent neuropsychological analysis (Hynd, Obrzut, Hayes, & Becker, 1986) as well as recent federal statistics (Algozzine & Korinek, 1985; Gerber, 1984). LD students do have serious learning problems, but they meet the criteria presented earlier for the concept of mild handicap; specifically, they do not have hard signs of neurological disorder or any other physical/biological anomaly. LD students constitute approximately 1. 8 million or 44% of the 4. 1 million students served in special education programs in the United States. It is the LD population to whom neuropsychological explanations of learning problems are 503
504
CHAPTER 27
most often applied, a population that has, we repeat, no hard signs of neurological disorder. We are not discussing school-age children or adults who do have hard signs of neurological disease or disorder. Furthermore, we wish to emphasize that our skepticism about neuropsychological diagnoses of LD and other mild handicapping conditions does not extend to basic research on brain-behavior relationships, laboratory studies of neuropsychological phenomena, or investigations of neuropsychologically based aptitude by treatment interactions. Although we are especially skeptical of the latter, we see no particular danger in pursuing research in that area. We do see danger in ''leap of faith'' assumptions about the application of abstract neuropsychological constructs to the classification and/ or treatment of students with learning problems. And this chapter is about the limited foundation for those applications.
eta/. noted that LOs can exist with other handicapping conditions, but these other conditions such as sensory impairments, mild mental retardation, or behavior disorder/emotionally disturbed cannot be the primary cause of the learning problem. Thus, these investigators recognized that LD can occur with other handicapping conditions and that there are a number of LD subtypes. The LD diagnostic construct has been severely criticized in recent years on the most fundamental bases concerning reliability and validity (Algozzine & Ysseldyke, 1983; Coles, 1978; Epps, Ysseldyke, & Algozzine, 1983; Ysseldyke, Algozzine, & Epps, 1983; Ysseldyke, Algozzine, Shinn, & McGue, 1982). Critics have asserted that no differences exist between LD students diagnosed as handicapped and low achieving students who remain in regular education programs and are regarded as "normal" learners. Furthermore, vast differences exist in state criteria used in classification of LD students and significant differences exist between districts within Uses of Neuropsychological states (Mercer, Hughes, & Mercer, 1985; Shepard, Diagnoses 1983; Shepard & Smith, 1983). Thus, current classification of LD students is, at least, unreliable. PerThe debate over neuropsychological analyses of haps even more fundamental is the criticism that curlearning problems needs a context. The appropriate rent classification of LD students is also invalid, at context is the three major ways that neuro- least in the sense of treatment validity. No unique psychological information might be used in decisions methods or strategies exist for LD students and there about students exhibiting learning problems: (I) clas- is reason to believe that the same results could be sification of students as handicapped, particularly obtained with remedial programs within regular eduLD; (2) selection of teaching strategies or instruc- cation (Heller, Holtzman, & Messick, 1982; Leintional methodology; and (3) determination of re- hardt, Bickel, & Pallay, 1982; Reynolds, Wang, & mediation objectives, such as what needs to be Walberg, 1987). taught. There is some evidence that neuropsychologDespite the negative evidence, the LD field conical diagnostic information is used for each of these tinues to flourish and the number of students classipurposes, but the most frequently discussed purposes fied as LD has grown each year since 1976. This are classification of students and/ or determination of growth cannot be understood from analysis of theory instructional strategies or methodology. Each of and research in psychology, education, special eduthese uses will be discussed in more detail below. cation, medicine, or any other discipline. The LD movement is best understood as a social and political reality, which continues to search for a solid base in Oassification of Students theory and research. Neuropsychology is an attracOne of the most important possible uses of tive possibility to those frustrated with the negative neuropsychological information is in the classifica- research on LD from other fields. Three definitions of LD have appeared over the tion of students as handicapped, particularly LD. Hynd et al. (1986) strongly endorsed the use of last 20 years and are, to varying degrees, adopted in neuropsychological concepts as the foundation for state legislation and state special education classifidefinitions of LD and in classification criteria to be cation criteria. The first of these definitions, develused in deciding whether or not a specific student is oped by the National Advisory Committee on HandiLD. H neuropsychology was the primary basis for the capped Children in 1967, suggests that LD is based LD diagnostic construct, that construct would pre- on psychological process disorders. That approach sumably involve (1) problems with school learning; was severely criticized and ultimately rejected by (2) other causes are ruled out as the primary etiology many persons in the field in the mid- to late 1970s. of the learning problem; and (3) positive evidence of Two new definitions of LDs appeared in the first half neuropsychological dysfunction is established. Hynd of the 1980s; both included phrases implicating neu-
A LEAP OF FAITH
rological dysfunction as the basis for LD. The National Joint Committee for Learning Disabilities ( 1981) used the phrase, ''these disorders are intrinsic to the individual and presumed to be due to central nervous system dysfunction.'' A similar definition adopted in 1984 by the Association for Children and Adults with Learning Disabilities referred to LD as "a chronic condition of presumed neurological origin .. .. " The explanations accompanying both of the definitions were careful to note that the neurological origin of LD was presumptive, not a criterion that had to be used in a diagnosis. Both also noted that many students with LD would not exhibit identifiable neurological dysfunctions. The newer definitions appear to be quite different in that a neurological rather than a processing basis is suggested for LD. However, as discussed later, there is more similarity in these bases for LD than might first appear (see Figure 1). Use of neuropsychological constructs in LD definitions and classification criteria is consistent
1. NACHC Definition National Advisory Committee on Handicapped Children, Con· terence sponsored by Bureau of Education for the Handi· capped, U.S. Office of Education, Washington, D.C.• Septem· ber 28, 1967:
505
with the origins of the LD movement. Cruickshank (1972, 1979), one of the pioneers, has argued vehemently that the original meaning of LD was unwisely and inappropriately broadened by educators and others in the 1960s and 1970s. Cruickshank argued that when LD was first adopted as a term in 1963, everyone knew what it meant. According to Cruickshank, it meant perceptual disorders due to neurological dysfunction. Therefore, neuropsychology might be seen as a way to return to the fundamental bases of LD, as it is closely tied to part of the traditions of the field. However, whether or not it is a useful basis for LD classification depends on the currently available reliability and validity evidence concerning neuropsychological principles. If this evidence is insufficient, then no useful purpose is served by returning to the neurological roots of LD. Reliability and validity of diagnostic constructs were discussed with great insight by Cromwell, Blashfield, and Strauss ( 1975). They established simple, straightforward criteria to judge reliability and validity of diagnostic constructs.
nificant difficulties in the acquisition and use of listen· ing, speaking, reading, writing, reasoning or mathematical abilities. These disorders are intrinsic to the individual and presumed to be due to central nervous system dysfunction. "Even though a learning disability may occur concomitantly with other handicapping conditions (e.g., sensory impairment, mental retardation, social and emotional disturbance) or environmental influences (e.g., cultural differences, insufficient/inappropriate instruction. psychogenic factors), it is not the direct result of those conditions or influences."
"'Specific learning disability' means a disorder in one or more of the basic psychological processes involved in understanding or in using language, spoken or writ· ten, which may manifest itself in an imperfect ability to listen, think, speak. read, write, spell, or to do mathematical calculations. The term includes such conditions as perceptual handicaps, brain injury, minimal brain disfunction, dyslexia, and developmental apha- 3. ACLD Definition sia. The term does not include children who have (Adopted by the Board of Directors of the Aasociation for Chil· learning problems which are primarily the result of dren and Adults with Learning Disabilities, September, 19841. visual, hearing, or motor handicaps, of mental retarReprinted from LD Fotum, Council for Learning Disabilities, dation, of emotional disturbance, or of environmen10(3), Winter 1985. pp. 12-13. tal, cultural, or economic disadvantage." "Specific Learning Disabilities is a chronic condition Note: This definition was adopted In "Proceduresfor Evaluat· of presumed neurological origin which selectively ining Specific Learning Disabilities," Federsl Register. Deterferes with the development, integration, and/or cember 29, 1977, Vol. 42, No. 250, pp. 65082-65085. (The demonstration of verbal and/or non-verbal abilities. NACHC definition appears on p. 65083.) "Specific Learning Disabilities exists as a distinct handicapping condition in the presence of average to 2. NJCLD Definition superior intelligence, adequate sensory and motor National Joint Committee For Learning Diubllitles Position systems, and adequate learning opportunities. The Paper, January 30, 1981. Reprinted in Society for LtHtrning condition varies in its manifestations and in degree of Dis11bl1ities and Remedial Education Newsletter, August, severity. 1981, Vol. 1, No.4, pp. 1-2: "Throughout life, the condition can affect self-es"Learning disabilities is a generic term that refers to a teem, education, vocation, socialization, and/or daily heterogeneous group of disorders manifested by sigliving activities."
.
FIGURE 1. Learning disability definitions.
506
CHAPTER 27
cal processing models, the most prominent theme in LD from the early 1960s to the mid-1970s, because these approaches emphasized remediation of psychological processing deficits (Hartlage, 1986; Hartlage & Reynolds, 1981; Reynolds, 1981, 1986). Their point of view, which appears to be well accepted among persons discussing neuropsychological approaches to learning problems, was expressed eloquently in Hartlage & Reynolds ( 1981): ". . . the deficit approach is doomed to failure since it teaches to dead tissue" (p. 358). Reynolds (1981) advocated a neuropsychological approach because it "matches up cognitive neuropsychological strengths with methods of presenting and acquiring information that rely most heavily on those strengths" (pp. 344-345). The ''matching up'' process described by Reynolds was justified primarily because of the inherent problems in the deficit approach: "Viewed from contemporary neuropsychological models, the deficit approach to remediation is doomed to failure since it takes damaged or dysfunctional areas of the brain and focuses training specifically on these areas" (p. 343). The fundamental assertions in the instructional methodology use of neuropsychological information rest on aptitude by treatment interaction. Aptitude, broadly, may be practically any characconceived Intervention Methods/Instructional individual that can be reliably assessed, the of teristic Strategies and then, most important, related to differential perThe second, and most prominent, use of neuro- formance under different treatment conditions (Cronpsychological information is to determine approach bach & Snow, 1977). The assumption is that persons to teaching, i.e., to specify intervention methods and will learn best if the teaching methodology or stratinstructional strategies. From a neuropsychological egy is matched specifically. to aptitude strengths. In point of view, students with learning problems are much of the neuropsychological literature the discusviewed as having quite uneven aptitudes and abili- sion focuses on finding ways around dysfunctional ties. This unevenness is assumed to arise from differ- brain areas or "dead tissue." In a later section we address the aptitude by ential efficiency or intactness of various sections of the brain. For example, students whose abilities sug- treatment interaction assumptions and the evidence gest a dysfunction in one hemisphere of the brain or on those relationships using neuropsychological in one portion of a hemisphere would, according to concepts. Using the Cromwell et al. (1975) approach, the contemporary neuropsychologists, learn more efficiently if methods were used that placed as little de- validity of an LD construct based on neuropsycholmands on that part of the brain as possible, and that, ogy will depend on the degree to which neuroat the same time, capitalized on intact neurological psychological information is related to effectiveness of interventions. If interventions or instruction can be structures or processes. The emphasis on neurological strengths is con- made more effective with application of neurosistent throughout most of the neuropsychological psychological concepts, then the LD diagnostic conliterature on learning problems. Reynolds and struct based on neuropsychological concepts is valid. Hartlage have been particularly prominent in advanc- On the other hand, if differential strategy and mething the argument that more efficient learning will odology cannot be related to these neuropsychologitake place if methodology or strategies are matched cal concepts, then these concepts are immaterial and to intact neurological structures or functions and if irrelevant to interventions, and potentially misleadneurological dysfunctions are avoided. They have ing and harmful if they undermine other intervention also been especially critical of the older psychologi- efforts. The reliability criteria have to do with the degree to which the diagnosis can be made consistently across different clinicians, different settings, and different times for the same individual. Diagnoses that are consistent and stable, and therefore reliable, have the potential for being valid. Diagnoses that are not reliable, in the sense of being stable and consistent, cannot be valid according to fundamental psychometric laws. Our contention is that an LD diagnostic construct based on neuropsychological principles is not reliable, and not surprisingly, not valid either. The validity of diagnostic constructs according to Cromwell et al. was based on the degree to which the construct was related to specification of treatments (instructional methodology, teaching strategies, or remediation techniques) and the effectiveness of those treatments. Diagnostic constructs related to unique methodology effective in remediating problems were valid; however, diagnostic constructs not related to effective interventions were invalid. We propose to apply the Cromwell et al. criteria to neuropsychological diagnoses of learning problems.
A LEAP OF FAITH
Determination of Intervention Goals A third possible use of neuropsychological concepts is in determination of intervention goals. These intervention goals would then be pursued through the most effective instructional procedures available for teaching particular kinds of cognitive operations, problem-solving strategies, or specific skills. Most of the contemporary advocates of neuropsychology, but not all, disavow this particular use. As noted in the previous section, Hartlage and Reynolds view this particular approach as doomed to failure because it rests on teaching cognitive operations that use neurologically dysfunctional areas of the brain. However, other scholars using neuropsychological constructs reach different conclusions (Das & Naglieri, in press; Naglieri & Das, in press). Das;s concepts of simultaneous and successive (Das, Kirby, & Jarman, 1979) are used widely by neuropsychological advocates (Reynolds & Kamphaus, 1986). However, in what would appear to be a break with most of the neuropsychologists in this volume, Das and Naglieri suggest specific methods for improving coding strategies such as simultaneous and successive as well as other neurologically based conceptions of cognitive processing such as arousal and planning. For Das and Naglieri, low scores on a particular neurological indicator apparently do not imply "dead tissue" but rather a cognitive process that can be improved through instructional procedures. There is another fundamental difference between Das and Naglieri's formulations, and contemporary advocates of neuropsychological explanations of LD. This has to do with assertions about actual brain structure or function based on observation of behaviors. Das and Naglieri generally avoid suggesting specific locations of neurological dysfunction, probably because of their recognition that several parts of the brain, usually including both hemispheres, are responsible even with relatively simple behavioral routines. In contrast, contemporary neuropsychologists are quite comfortable with attributing failure on simple or complex tasks, e.g., copying Bender drawings or determining conceptual likenesses on the Wechsler Similarities Subtest, to one or the other hemisphere or even to specific locations such as right occipital lobe or left temporal lobe. Because some neuropsychologists seem to have a quite literal notion of performance variations resting on specific neurological dysfunction, it makes sense for them to be very skeptical about approaches attempting to train weak processes (tissue). On the other hand, for scholars influenced by neuropsychology, particularly Luria, e.g., Das, whose model of neurological functioning is far
507
more complex, training deficit processes is entirely appropriate. The past debate over process training (Hammill & Larsen, 1974, 1978; Lund, Foster, & McCallPerez, 1978; Minskoff, 1975;-Newcomer, Larsen, & Hammill, 1975) will almost undoubtedly reappear when Das and Naglieri's assessment battery is published. However, for now, the use of neuropsychological information to determine intervention goals is not a major issue. The rest of the chapter will emphasize review of the evidence on classification and on selection of intervention strategies.
Fallacies in Neuropsychological Explanations Neuropsychological explanations of common learning problems are based on studies of highly selected and often, extraordinarily rare, individuals and then generalized to students whose developmental and neurological status are clearly very different from persons included in the basic research. These generalizations often involve inferences from persons with definite brain damage to children who have no identifiable brain injury. A number of the fallacies and the overgeneralizations made in neuropsychological explanations of mild learning problems are discussed in the following sections.
Fact versus Concept of Brain Injury The overwhelming majority of students classified as mildly handicapped, about 8% of the total school age population, as well as those classified as LD, about 3 to 4% of the school age population, have absolutely no identifiable anomaly in actual neurological structure, process, or function. This is openly acknowledged by neuropsychologists (e.g., Gaddes, 1981). The vast majority of mildly handicapped and LD students do not display any of the hard neurological signs of brain dysfunction (hard signs refer to relatively direct measures of brain structure, function, or process). Some of the assessment tools yielding ''hard'' evidence of neurological dysfunction are the electroencephalogram, pneumoencephalogram, X ray, examination of cranial nerves, and cerebral blood flow studies. These assessment devices allow descriptions of actual dysfunction or anomaly in brain structure, function, or process. A second level of neurological diagnosis is provided by the so-called "soft signs" of neurological
508
CHAPTER 27
dysfunction. Most often these soft signs are used to make inferences about neurological function and process, although it is very easy to generalize further to structure, as will be seen in interpretations quoted from model reports by neuropsychologists in later sections. These soft signs of neurological dysfunction are based on observations of behavior, most often behaviors having to do with motor coordination, perceptual functioning, and complex information processing. In many instances, contrasts are drawn between behaviors that are presumed to be based on neurological processes from different parts of the brain. These assertions about different parts of the brain are based on looking at differences in behaviors, not on hard neurological evidence. Moreover, the generalizations are from neurological studies of adults who had various kinds of neurological injuries or who were exposed to very narrowly construed experimental tasks administered under controlled conditions in laboratory studies, not children who were similar to the child being examined by the neuropsychologist. The analogue from adults to children and the assertion that a complex of symptoms exhibited by children was due to minimal brain injury rests on analogical reasoning of the 1940s, which, in tum, was based on studies of soldiers in World War I who sustained and survived serious brain injuries. Although this historical relationship is beyond the scope of this chapter, we do wish to note that much of what is asserted about minimally brain injured children today as well as the examination of children for "soft" neurological signs rests on drawing inferences from studies by Goldstein during and just after World War I and the educational interventions developed by Strauss, Werner, Lehtinen, and others at the Wayne County Training School in the 1940s (Widerholt, 1974). The critical distinction is between the fact and the concept of brain injury. The fact of brain injury must be based on hard neurological evidence. The concept of brain injury is based on a complex of symptoms first observed with adults who undeniably had brain injury. However, the brain injury etiology was generalized to children who exhibited some, but not all, of the same symptoms even though they had absolutely no hard evidence of actual brain injury. Moreover, much of the information concerning localization within hemispheres or hemispheric specialization is based on highly unusual conditions that will be discussed shortly. The fact is that when Hartlage, Reynolds, or other neuropsychologists analyze learning problems in terms of left- and right-hemisphere functioning or
anterior and posterior portions of one hemisphere or the other, or particular locations such as occipital or temporal lobe, they are applying results from studies of adults and laboratory studies using highly controlled conditions. They are not applying knowledge based on hard neurological evidence gained with children of approximately the same age as the child being studied. To put it bluntly, they deal with the concept of brain injury, a concept based on a very high level of inference, rather than the fact of actual brain dysfunction or injury. This alone should create doubts about neurological explanations of learning problems.
Overgeneralization from Atypical Subjects One of the major difficulties with neuropsychological explanations of learning problems is the basis for many of the generalizations, usually studies of highly unusual subjects or studies conducted in tightly controlled laboratory situations. The work of Sperry and his associates (Gazzaniga, 1970; Gazzaniga, Bogen, & Sperry, 1965; Sperry, 1968) on adult epileptic patients who had undergone surgical severance of the corpus callosum is one of many examples that could be cited of overgeneralization from extremely unusual subjects. Sperry and associates were studying epileptic patients whose seizure disorders were so severe and unresponsive to standard pharmacological treatment that severance of the corpus callosum, thus ensuring functional independence of left and right hemispheres, was carried out in order to control the seizure disorders. The Sperry eta/. research resulted in the conclusion that complex linguistic functions are localized in the left hemisphere and visual-spatial skills are localized in the right hemisphere. However, this generalization needs to be restricted to commissurized adult epileptic patients. It does not apply directly to children with learning problems and it does not provide a secure foundation for the frequent assertioll'S about rightand left-hemisphere localization of complex school learning. Overgeneralizations from the classic work by Sperry et a/. are frequently made by neuropsychologists describing children with learning disabilities. A WISC-R or other test tapping both verbal and spatial skills is administered. If the verbal scores are higher than the spatial scores, the diagnosis is usually some sort of right cerebral hemispheric dysfunction or weakness. An example of such an interpretation is given by Hartlage (1981): The score configurations, in addition to supporting hypotheses of fairly chronic functional superiority of left-
A LEAP OF FAITH over right-hemisphere functioning, suggest that, while both anterior and posterior portions of the right-hemisphere are depressed when compared to the left, there may be slightly Jess impairment on posterior than on anterior portions of the right-hemisphere. (p. 364)
This interpretation can be traced back to the original work of Sperry et al. ~eg~ding co~ missurized adult patients and localizatiOn of bram function. Because the data drawn from Sperry's work were based on adults whose neurological condition made severance of the corpus callosum necessary, an extraordinarily rare surgical intervention, generalizations to children who perform verbal tasks better than they put puzzles together is highly inferential, and not based on data from children. The relevance of clinical samples as models of the functioning of school-age children with no hard signs ~f neurological dysfunction remains highly questionable. The fact is that relevant studies with school-age children have not been done and it is an extreme overgeneralization to apply results of adult patients with severe brain injuries to school-age persons. A second kind of overgeneralization occurs when results from studies employing well-controlled, very specific laboratory conditions are applied to school-age children. Many of these laboratory studies do show hemispheric differences, but these differences are extremely small, a matter of a few milliseconds or slight differences in percentage correct, and are probably limited to relatively simple stimuli such as a letter, digit, word, or shapes (O'Boyle, 1986). The generalizability of these ~x perimental findings to extremely complex tasks hke reading or mathematics reasoning also tenu~us. The degree to which a barely perceptible flash ofhght measured in milliseconds generalizes to reading where exposure is longer, the information is being received by both hemispheres simultaneously, and a variety of cues are present, is highly questionable. Again, most of the studies have not been conducted with children, and even when they are, the degree to which they adequately represent complex cognitive functions remains uncertain. Overgeneralization from atypical subjects and laboratory settings is one of the most important errors in the neuropsychological explanations of learning problems.
!s
Unsophisticated, Primitive Neurology A third major problem in neuropsychological explanations is primitive, unsophisticat~d ap~recia tion of the complex, interdependent relattonsh1ps between brain and behavior. These relationships are
509
bidirectional and inseparable. Furthermore, the frequent suggestions of quite specific relationships between complex behaviors and a particul~ portion of the brain fail to appreciate the complexity of neurological functioning, and the interdependen.ce ~nd the inseparability of the hemispheres and brat~ sttes within hemispheres. To suggest that a parttc~lar complex task such as recognizing letters or assoctating sounds with particular letters or vo':"el and consonant combinations is located in a spec1fic part of the brain is a vast oversimplification and an inaccurate portrayal of brain-behavior relationships. O'Boyle (1986) pointed out that the cerebral hemispheres make interactive contributions to complex human performance on a wide variety of co~ nitive tasks. The differences between the hemispheres, typically based on studies. of atypical subjects as noted above, are usually qutte small and observable only under very well-controlled conditions. The generalizability of these small differences to understanding learning problems is far from clear, but the simultaneous functioning of several parts of the brain in any complex task is a virtual certainty. Therefore, neuropsychological interpretations of specific tasks being related to one hemisphere or ~ other or specific locations within or between hemispheres simply cannot be plausible. O'Boyle (1986) was particularly critical. to the "leap of faith" generalizations abou~ ~emtsp~er icity. According to O'Boyle, "~!early, _It 1s on~ ~mg to conclude that the right hem1sphere 1s spec1al1zed for visuo-spatial processing on the bases of superior ability to match shapes. It is quite another to ~oncl~de from the same data that we 'draw from the nght s1de of the brain'. Translation errors like this are undoubtedly responsible for many of the misguided educational applications of left-brain/right-brain differences" (p. 43). The alleged superiority of rightbrain functioning with visuospatial processing is based on well-controlled conditions and very simple tasks. Learners in classrooms, even if instructed on a one-to-one basis, are involved with a far more complex array of relevant stimuli and much more complicated tasks. Furthermore, these complicated tasks in natural classroom settings typically draw upon the processing characteristics, whatever those may be, of both hemispheres. As O'Boyle points out, the hemispheres are best understood as "equally important partners of a processing team" (p. 42). We acknowledge that hemispheric asymmetry does exist, but dispute the degree of differences in processing functions associated with relevant le~ ing tasks. Both hemispheres are undoubtedly mvolved. Even under highly controlled conditions, the
510
CHAPTER 27
differences between the hemispheres are relatively small in degree and far from universal. All of this has clear implications for neuropsychological explanations of learning problems. Those explanations nearly always apply a highly simplistic conception of neurological functioning that presupposes areas of the brain functioning independently and uniquely responsible for particular kinds of learning. There is absolutely no evidence of any kind to support this degree of independence and specialization in areas of the brain relative toLD and other learning problems. The problems described above are related to what has been referred to as "the fallacy of brain damage as the first and final cause" (Goodman, 1983). Goodman pointed out that neuropsychologists assume a "closed circuit" relationship between brain damage and behavior because they implicitly assume one-way causal relationships. That is, causality for the neuropsychologists starts with the brain and ends with behavior when, in fact, there is also good evidence that specific environmental experiences also influence brain function. Goodman pointed out that neuropsychologists typically assume that, in the presence of both the neurological symptom and a behavioral symptom, the behavioral symptom must be caused by the neurological symptom. In fact, a variety of causes could be posited in what is essentially a correlation. That correlation does not establish causality, is a dictum well understood by all psychologists, including neuropsychologists, but often ignored in asserting relationships between specific behaviors and neurological structure or function. We, along with Goodman (1983), acknowledge that all learning has a biological substrate. We even acknowledge that there are probably neurological differences between a student who can recite the A, B, C's and recognize letters and students who cannot perform those routines. To acknowledge that virtually everything is neurological does not, however, provide any validity for neuropsychological interpretations of specific behavioral events. The fact is that the underlying neurological differences among persons who have mastered or who have not mastered specific routines are simply, at this point, unknown, and accurate descriptions of the precise neurological changes associated with learning are probably decades away. But to say that everything is neurological does not move us any closer to understanding or genuine explanation of learning and human individual differences. Moreover, premature, unsophisticated neurological explanations probably do more harm than good because they deflect attention from aspects
of the learning situation that can, indeed, be utilized to improve skills (more on that later). At a minimum, though, one-way causal thinking, brain to behavior, and unsophisticated applications of laboratory findings are at best highly speculative and, at worst, perpetuate myths about the primacy of neurological status over environmental experiences. Goodman also pointed out the fallacy of brainbehavior parallelism where neuropsychologists seem to make the assumption that the effects of brain damage are highly specific. There is little evidence, however, to suggest that brain damage or dysfunction commonly takes a very specific form, such as reading problems, difficulty in processing certain kinds of information such as spatial relationships, hyperactivity, and so on. Benton (1974) noted that the best single index of the presence of brain damage in children and adults is a lower than expected total score derived from a comprehensive test battery. Thus, there are little data to support the notion of localization of brain damage leading to specific impairments reflected in behaviors observed during standard neuropsychological assessment procedures. Despite this lack of research support for brain dysfunction paralleling specific behaviors, neuropsychologists attempting to explain learning problems frequently use simple behavioral observations, say on the Bender, or test scatter, as indications of specific brain dysfunction. Discussion of problems with neuropsychological assessment procedures is given in the following section. There are numerous fallacies in neuropsychological explanations of learning problems. First, there is the confusion over the fact versus the concept of brain damage. Neuropsychologists are using the concept of brain damage, not the fact, in explanations of learning problems. These explanations are derived from studies of highly atypical subjects some of whom have extraordinary neurological characteristics, such as severed corpus callosum, as well as application of findings from extremely tightly controlled laboratory studies. These generalizations are dubious. Furthermore, the assertions of specific brain status associated with specific observed behaviors are based on "leaps of faith" with little or no evidence with students of the kind involved in the recommendations (overgeneralizations). These generalizations reflect primitive, unsophisticated applications of neurological findings as well as other fallacies such as the brain as the first and final cause of behavior and the assumption of very close relationships between neurological status and specific behaviors. These problems alone should give pause
ALEAPOFFAITH
to widespread application of neuropsychological techniques. Additional problems with neuropsychological assessment and educational implications of presumed neuropsychological states create even more formidable barriers to neuropsychological explanations of learning problems.
Fallacies in Neuropsychological Assessment One of the weakest aspects of the case for neuropsychological explanations of LDs is the nature and quality of the assessment procedures. These assessment procedures are subject to three major problems: (I) the instruments are questionable due to limitations in reliability, validity, and norms; (2) phenomena that occur at high base rates in the normal population are interpreted as indicating neurological disorders; and (3) simple behaviors are interpreted as indicating complex underlying disorders through very high levels of inference. These difficulties are discussed in the following three sections.
Questionable Instruments The instruments commonly used in neuropsychological assessment batteries are, except for the Wechsler Scales, widely regarded as having questionable technical characteristics. The instruments listed in model case studies (Hartlage, 1981, 1986; Hartlage & Reynolds, 1981; Obrzut, 1981) typically feature the Wechsler Intelligence Scale for Children-Revised (WISC-R), the Bender-Gestalt, the Illinois Test ofPsycholinguistic Abilities (ITPA), the Wide Range Achievement Test (WRAT), the Peabody Picture Vocabulary Test (PPVT), as well as various techniques for examining motor skills such as finger tapping with right and left hand, relative strength of right and left sides, and so on. Except for the Wechsler Scales, these procedures are widely regarded as having questionable technical characteristics, including inadequate reliability for making decisions about individuals, undemonstrated validity, and/or unsatisfactory norms (Salvia & Ysseldyke, 1985). Instruments with poor norms, undemonstrated validity, and inadequate reliability simply cannot yield valid information to be used in diagnosing complex characteristics such as learning problems, and certainly not the more complex, underlying neurological status of the individual. Unreliable instruments measure error. not consistent sta-
511
ble characteristics that can be used, with confidence, in prescribing remedial interventions. Calling the Bender a measure of "constructional praxis" or including the WRAT as part of a neuropsychological test battery does not confer greater technical adequacy for these weak instruments (Salvia & Ysseldyke, 1985; Witt, 1986). Instruments do not acquire essential characteristics simply because they are used to make highly speculative inferences about underlying neurological status. In fact, the same deficiencies identified when these instruments are evaluated on their own undoubtedly further complicate and substantially diminish the reliability and validity of inferences about neurological status. Even the Wechsler Scales, which have good technical adequacy for certain inferences, are used in such a way to render them technically inadequate for inferences about neurological characteristics. For example, the Wechsler Verbal and Performance IQs are frequently interpreted as indicating left- and righthemisphere specialization. Thus, a person with a higher Performance than Verbal score is regarded as having stronger left- than right-hemisphere functioning. The evidence for this assertion is dubious at best. Even more dubious is the comparison of single pairs of subtests to determine neurological integrity of different parts of the brain. Fluctuations among single pairs of subtests are known to be common (Kaufman, 1976). Furthermore, simple comparisons of subtests are complicated by the relative unreliability of difference scores. In spite of these problems, we hear the following from a well-known neuropsychologist "Even greater refinement can be obtained by comparing individual subtests, such as comparing functional integrity of left- and right-temporal lobes by the Similarities versus Picture Arrangement subtests; or, left- and right-parietal lobe integrity by the Arithmetic vs. Block Design subtests" (Hartlage, 1982, p. 300). Here there is the implicit assumption that specific Wechsler subtests can be used to determine the integrity of specific parts of the brain. There are many things wrong with this reasoning, beginning with the relative unreliability of difference scores, meaning that the difference between subtest pairs is likely to fluctuate upon retesting. This, in turn, means either that much of what is interpreted about pair differences is error or that neurological status changes fairly rapidly when testing is repeated. Although we believe in relative plasticity of neurological functioning, we have trouble believing that the ·'integrity'' of different parts of the brain is that haphazard, and, for that and other reasons, we have even greater reservations about whether individual
512
CHAPTER 27
Wechsler subtests can be used to make inferences about specific parts of the brain.
neuropsychological dysfunctions and thereby be eligible to be classified as LD based on the results of neuropsychological assessment. The high base rates make it almost certain that a very high percentage of Base Rates normal learners would meet the criteria for having a neuropsychologically based LD. This result would Base rate refers to the frequency with which some phenomenon occurs in the general population. be identical to that reported by Ysseldyke et al. Base rates are important because many of the charac- ( 1982) indicating a very high percentage (perhaps as teristics interpreted as clinicially significant by high as 90%) of normal students show one or more neuropsychologists occur frequently in the general LD symptoms. Although studies of the percentage of normal population. The high-base-rate phenomenon has two students who would show one or more significant implications. First, the patterns interpreted as signs of neurological dysfunction, using the kind of clinically significant by neuropsychologists are far criteria applied in the model case reports cited earlier, from unusual and certainly not unique characteristics have not, to our knowledge, been done, we contend of individuals. Second, a study of a representative sample of persons in the general population would that such studies do not need to be done because the yield, for virtually everyone, suggestions like hemi- results
A LEAP OF FAITH
513
Danny's
Danny's Friend's Bender FIGURE 2. Different levels of inference in interpretation of Bender reproductions. The following interpretations vary from low to high inference: poor copying skills, poor visual-motor skills, developmental lag or immaturity, neurological dysfunction, emotional symptoms, personality dynamics.
A very poor drawing might be interpreted at five different levels of inference. The lowest level of inference would be to interpret the poor Bepder drawings as indicating poor copying skills. A slightly higher level of inference might be to infer an underlying skill such as visual motor skills. A still hi~her level of inference might be to make some assertion about maturation rate or developmental level, a common interpretation of the Bender. A much higher level of inference that goes far beyond the actual behavior observed, which was-it is important to remember-merely copying a geometric figure, would be to assert some complex process like constructional praxis and to suggest an associated neurological dysfunction. A still higher level of inference would be to make assertions about personality dynamics using the Bender emotional indicators (Koppitz, 1975). There are several problems with these highly inferential interpretations, particularly interpretations having to do with either underlying physical status, as in neuropsychological interpretations of Bender drawings, or underlying personality dynamics. The high levels of inference are generally not in the student's best interests because: (I) the highly inferential interpretations are rarely supported by solid, empirical evidence; (2) the variables identified in the highly inferential interpretations are usually difficult to impossible to influence, at least directly; (3) the highly inferential interpretations are rarely related to educational interventions or psychological treatments; and (4) the highly inferential interpretations, because they focus on deep underlying problems, frequently reduce commitment toward, and a sense of efficacy about, carrying out interventions. This
latter point, although subtle, is extremely important. Problems described in terms of neurological dysfunction or deep underlying conflicts are less likely to be seen as amenable to instruction or treatment. On the other hand, straightforward and precise descriptions of the same behavior without the presumed underlying cause are much more likely to be regarded to amenable to instruction or treatment (Tombari & Bergan, 1978). One further observation about highly inferential interpretations in neuropsychology: Court testimony based on high levels of inference is often challenged effectively on cross-examination (Ziskin, 1981), because of the lack of empirical support. The problems of high base rates and high levels of inference are typically combined in neuropsychological assessment. Consider this example: ••Perhaps the most striking feature of his Wechsler profile is his uniquely depressed similarities subtest. The relative isolation of this intellectual deficit is more compatible with some sort of acquired insult to the left temporal tuea than to any generalized left hemisphere deficit, and in fact the majority of other test measures are reasonably close to normal" (Hartlage, 1982, p. 310). The similarities subtest score in the model case study just quoted was 6. The highest subtest score on the Verbal Scale was 10; on the entire Wechsler battery the highest subtest score was 11. A four-point difference between pairs of subtests on the Verbal Scale or a five-point difference between highest and lowest subtests across both Verbal and Performance Scales is a very frequent occurrence. Typical, normal children have subtest differences of that magnitude or greater. Furthermore, the suggestion that a lower Similarities score indicates "acquired insult to the left temporal area'' can only be regarded as reflecting
514
CHAPTER 27
an extremely high level of inference because the relationship between what is actually observed and the meaning attributed to that observation is, at best, extremely remote. Moreover, the observation is not based on empirical research with normal and braininjured students who also differ in learning efficiency. Rather, it is based on analogical reasoning from studies of very unusual adults. One of the major problems with neuropsychology is in the realm of assessment. The assessment instruments provide little reason for confidence in neuropsychological interpretations. These instruments clearly do not become better or somehow acquire technical adequacy because their results are interpreted as indicating neurological status. Neuropsychological assessment, in addition to being based on questionable instruments, also reflects overinterpretation of normal test score fluctuations as indicating neurological dysfunction and extremely high levels of inference that have little or no foundation from empirical research. Based on these assessment problems, it is not surprising that little or no evidence exists to support educational interventions based on neuropsychological principles.
Fallacies in Educational Recommendations Based on Neuropsychological Principles The principal recommended use of neuropsychological assessment of students with learning problems is for improved classification and/or determination of teaching methodology or strategies. The impossibility of improved LD classification was discussed in the preceding section. In this section we will review evidence on the possible use of neuropsychological principles to determine how to teach students with learning problems. The neuropsychological principles in this regard are fairly straightforward. Basically, neuropsychologists argue that intact areas, i.e., areas in the brain that are functioning properly, should be used in deciding how to teach important academic skills such as reading and mathematics (Hartlage, 1981, 1982, 1986; Reynolds, 1981; Hartlage & Reynolds, 1981). These neuropsychologists sharply reject being associated with the earlier processing view in LD, which they claim, generally accurately, involved efforts to remediate deficits rather than capitalize on strengths. For this use of neuropsychological assessment to be valid there must be evidence for aptitude by treatment interaction and there must be a set of instructional inter-
ventions that capitalize on the student's neurological strengths. The evidence for both of these requirements is negative.
Aptitude by Treatment Interaction (ATI) The most critical assumption to current neuropsychological approaches is that learning academic content such as reading or mathematic skills will be enhanced if a careful neuropsychological examination is conducted and the student's best neurological functions utilized in the methodology or strategy used to teach academic skills. The notion of ''matching up'' intervention methodology or teaching strategy to intact neurological structures or processes has tremendous intuitive appeal and almost inherent credibility. The problem is that this intuitive appeal or inherent credibility has virtually no support from empirical evidence. Reviews of ATI studies using aptitudes like processing strength or intact neurological structures/functions have been published (Arter & Jenkins, 1979; Ysseldyke & Mirken, 1982). In the absence of positive evidence concerning ATl, including the specific kinds of assertions made by Hartlage, Reynolds, Obrzut, and others cited in this chapter, it is very difficult to find support for continuation of neuropsychological interpretations and educational recommendations based on neuropsychology. This is not simply a matter of "we need more research, but should continue what we're doing now in the absence of that research," but rather, without some positive research evidence, the whole enterprise needs to be scuttled. The likelihood of positive ATl findings for current neuropsychological principles is, we believe, highly remote. We suspect the outcomes of future research concerning ATis with neuropsychological principles will be similar to findings reviewed by Arter and Jenkins (1979). A variety of reasons could be suggested for the failure of ATl studies. These reasons range from doubts about the relationship of neurological integrity to learning efficiency, the accuracy of current assessment procedures in determining neurological integrity, the relationship of neurological integrity to specific instructional procedures, particularly whether these instructional procedures do indeed tap specific neurological structures/functions; whether the instructional procedures were implemented properly, whether subjects in the study were chosen appropriately, and so on. We have no way to determine which combination of these reasons accounts for the negative findings to date. We can only say, however,
ALEAPOFFAITH
that negative ATI findings suggest we are wasting our time and the valuable time of children in conducting elaborate neuropsychological analyses in order to prescribe some kind of instructional intervention. There is no evidence to suggest this entire expensive and time-consuming enterprise is of value to anyone, except to clinicians who are more ~atisfied with a neurological explanation for behavior.
Limited Instructional Interventions One of the major weaknesses of educational applications of neuropsychological principles is the very limited range of educational treatments suggested by neuropsychological advocates. These instructional interventions are most often limited to vague generalizations about methods of teaching reading with whole word or sight word approaches recommended for students with (allegedly) better functioning· right than left hemispheres, and a more analytic or phonic method for students with (allegedly) better left than right hemispheres. Specific recommendations for students whose temporal lobes are functioning better on one than the other or whose brain anteriors are better than their brain posteriors have generally not been provided in the model reports developed by neuropsychologists (Hartlage, 1981, 1982, 1986; Hartlage & Reynolds, 1981; Obrzut, 1981). One of the weakest aspects of the neuropsychological approach, not appreciated sufficiently to date, is the relative absence of finely graduated instructional procedures that can be selected in order to capitalize on alleged neurological strengths. We doubt that these more finely differentiated methods are likely to have much effect because of all the problems discussed in other sections of this chapter. However, if all that can be done is to make a vague generalization about reading method, when in fact all reading methods undoubtedly involve complex processing ofboth hemispheres (O'Boyle, 1986), then we have to wonder about even the potential usefulness of the neuropsychological principles. Finally, it is impossible to develop teaching methods or strategies that capitalize exclusively or even primarily on a small area of the brain or on only one or the other hemisphere. For all learners, except for those extraordinarily rare adults with a severed corpus callosum, both hemispheres are working together. Finally, we agree with Obrzut ( 1981) and Gaddes ( 1981 ), both of whom expressed doubts as to whether interpretations about neurological integrity should even appear in reports used by educators in school settings. Obrzut (1981) noted, "Although
515
there are a variety of significant behavioral signs in D.K. 's performance that suggest left-hemisphere dysfunction, such a diagnosis could only be made when correlated with medical data (e.g., an EEG dysrhythmia in the left-hemisphere with none in the right). This information was not available nor was deemed necessary in planning appropriately for remedial activities" (p. 358). Gaddes (1981) expressed his reservations in the following way: "School psychologists are ill-advised to make diagnostic statements about a learning disabled child's brain unless they have had extended and competent training in clinical-neuropsychology; they do better to keep to the behavioral data on the child's sensory, cognitive, and motor strengths" (p. 40). Whatever instructional interventions are available, and those described in the neuropsychological literature seem very restricted and simple, it appears that use of neuropsychological constructs is unnecessary to "match up" the student's strengths with instructional methodology. Again, we find no basis for neuropsychological diagnosis of learning problems from the literature on instructional interventions.
Psychological Processing Revisited Although strongly disavowed by contemporary neuropsychologists, the fact is that there are numerous similarities between current neuropsychological explanations of learning problems and the older, now widely rejected, psychological process explanation of learning problems. These similarities are immediately apparent if guidelines for developing instruction to capitalize on psychological processing strengths are compared to guidelines for capitalizing on neuropsychological strengths (compare Kaufman, Goldsmith, & Kaufman, 1984, with Minskoff, Wiseman, & Minskoff, 1972). In comparing those materials, the only changes needed to establish a very high degree of similarity are to substitute auditory vocal channel for left-hemisphere or sequential processing and visual motor channel for right-hemisphere or simultaneous processing. Otherwise the content, particularly recommendations for teaching strategies, are virtually the same. Although we acknowledge the differences in the primary use of the information, the contemporary neuropsychologists claim to avoid teaching ''dead tissue," the other inadequacies of the earlier processing approach, particularly technically inadequate assessment instruments, undemonstrated ATI, and obscure relationships to instruction are relevant to current educational applications of neuropsychology.
516
CHAPTER 27
Potential Harmful Effects All that has been said thus far could be dismissed as merely an argument among persons with different theoretical orientations; a kind of paradigm war among psychologists interested in learning problems. We think this argument, however, has important consequences for children. A major problem with the neuropsychological approach to learning problems is that it is heavily steeped in the medical model tradition. In fact, we find it difficult to think of anything that has more of a ''medical model'' orientation than to make suggestions like ''damage to posterior portions of the right cerebral hemisphere involving deficits of function of both temporal and parietal lobes" (pp. 308309, Hartlage, 1982). Although neuropsychologists claim to emphasize neurological strengths, the content of their reports typically is devoted more to a description of deficits, usually in what appears to the uninitiated to be in very technical, scientific language related to specific damaged areas of the brain. Although we and most neuropsychologists understand the degree of speculation involved, most teachers will have little or no background to understand the speculative nature of the inferences or the differences seen by the neuropsychologist in intact areas of the brain as opposed to dysfunctional areas of the brain. We believe this focus on deficits couched in what are pseudoscientific behavior-brain links will lead teachers and parents to have low expectations for students for whom neuropsychological diagnoses are provided (Sandoval & Haapanen, 1981). Drawing from the learned helplessness model, parents, teachers, and the diagnosed child may come to believe that no matter what is done instructionally, it will not make any difference in terms of academic performance because the student's brain is not functioning properly. Moreover, none of these persons are likely to realize that the ''concept'' of brain injury, not the fact of brain injury, is the foundation for these neuropsychological explanations. One study that demonstrates this problem contrasted behavioral and medical model approaches to conceptualizing learning problems identified by teachers. Tombari and Bergan ( 1978) exposed teachers to either a behavioral or a medical model form of consultation. Results from a path analysis showed that teachers receiving the behavioral model were better able to define the problem behaviors and this definition of problem behaviors, in tum, led to higher expectations regarding their perceived ability to solve these problems. We suggest that further studies of this nature might be conducted using the same
basic facts for a case, but different reports developed, thus allowing contrasts in teachers' reactions to neuropsychological and behavioral explanations of learning problems. We believe that the neuropsychological reports, despite their advocates' vehement claims to create a more positive expectation through identifying neurologically functional areas of the brain, will, in fact, have the effect of producing teacher and parent beliefs that the neurologically damaged child will be very difficult, if not impossible, to change. In contrast, we believe that a behavioral assessment model with the identification of specific behaviors and description of precise objectives will lead to beliefs in greater efficacy in teaching the child (Shapiro & Lentz, 1985). Interested readers wishing to see the contrast of approaches might compare the model report in Shapiro and Lentz ( 1985) with the reports in the neuropsychological literature (Hartlage, 1981, 1986; Hartlage & Reynolds, 1981 ). It is important to note that in the contrasts suggested here the actual behavior of the student is the same. The difference is in the interpretation of that behavior. The critical questions might be framed as follows: Are children served better by an interpretation that uses a very high level of inference, internal, nonobservable states of the individual, and illusions to damaged tissue (without any actual evidence of damage), or are students better served by careful descriptions of relevant skills with thorough description of antecedent, situational, and consequent events followed by direct instruction on skills not mastered with appropriate behavior analysis principles applied in diagnosis, instruction, and evaluation of the outcomes of instruction? Our answer on that matter is obvious and we challenge the reader to weigh the alternatives and consider which is more likely to lead to better educational opportunities for students with learning problems.
Special Education Reform The current reform movement in special education may render moot much of the discussion in this chapter. The special education system of the future may very well eliminate classification of children with learning problems in order for them to receive some kind of remedial services. Future classification may place far less emphasis on diagnosis of underlying, nonobservable traits or internal states of the organism and far more emphasis on straightforward curriculum-based assessment (Grimes & Reschly, 1986; Reschly, 1987b; Shapiro & Lentz, 1985;
A LEAP OF FAITH
Shapiro, 1987). The relevance of neuropsychology to a reformed, radically different special education delivery system is, at best, remote. Pending changes in the data base for neuropsychology, particularly the clear demonstration of aptitude by treatment interactions, something not attained or even approximated to date, the reformed system described elsewhere (Graden, Zins, & Curtis, 1988) may render moot the issues debated in this chapter.
Conclusions
517
sociation, No. 3. Benton, A. L. (1974). Clinical neuropsychology of childhood: An overview. In R. Reitan & L. Davison (Eds.), Clinical neuropsychology: Current status and applications. New York: Wiley. Coles, G. S. (1978). The learning disabilities test battery: Empirical and social issues. Howard Educational Review, 48,
313-340. Cromwell, R., Blashfield, R., & Strauss, J. (1975). Criteria for classification systems: InN. Hobbs (Ed.), Issues in the classification of children. San Francisco: Jossey-Bass. Cronbach, L. J., & Snow, R. E. (1977). Aptitudes and instructional methods. New York: Wiley (Halsted Press). Cruickshank, W. M. (1972). Some issues facing the field oflearning disabilities. Journal of Learning Disabilities, 5, 380-
383. Neuropsychological approaches to leru:ning Cruickshank, W. M. (1979). Learning disabilities: Perceptual or problems, such as those exhibited by students now other? ACW Newbriejs, No. 125, March/April, 7-10. classified as LD, are irrelevant to accurate classification, immaterial to effective remediation, misleading Das, J.P., Kirby, J., & Jarman, R. F. (1979). Simultaneous and successive cognitive processing. New York: Academic Press. by suggesting brain conditions for which there is no Das, J. P., & Naglieri, J. A. (in press). Cognitive assessment solid evidence, and potentially harmful in suggesting system. New York: Psychological Corporation. deep underlying, indeed dysfunctional physical Epps, S., Ysseldyke, J. E., & Algozzine, B. (1983). Impact of structures or processes as the cause of learning probdifferent definitions of learning disabilities on the number of lems. Neuropsychology carries the danger of sugstudents identified. Journal of Psychoeducational Assessment, 1, 341-352. gesting damaged physical structures to teachers, parents, and students, with the connotation of immuta- Federal Register. (1977). Procedures for Evaluating Specific Learning Disabilities. Author, December 29, 42 (250), bility, without providing a foundation for more effec65082-65085. tive remedial or developmental educational services. Neuropsychological constructs are largely unneeded Gaddes, W. H. (1981). An examination of the validity of neuropsychological knowledge in educational diagnosis and reto make the kinds of recommendations that appear in mediation. In G. W. Hynd & I. E. Obrzut (Eds.), Neuromodel reports by neuropsychologists, and, in any psychological assessment and the school age child: Issues event, there is no evidence that the aptitude (intact and procedures. New York: Grune & Strattan. neurological structures/functions) does interact with Gazzaniga, M. S. (1970). The bisected brain. New York: Apinstructional methodology. Until such interactions pleton-Century-Crofts. can be demonstrated, we find no basis to perpetuate Gazzaniga, M.S., Bogen, J. E., & Sperry, R. W. (1965). Observations on visual perception after disconnexion of the cereneuropsychological assessment. We agree with the bral hemispheres in man. Brain, 88, 221-236. dictum, "In God we trust. All others must have data'' used as the opening for a chapter on bias in Geib, S. A., & Mizokawa, D. T. (1986). Special education and social structure: The commonality of "exceptionality." assessment by a well-known and widely respected American Educational Research Journal, 23, 543-557. psychologist (Reynolds, 1982). We believe the dicGerber, M. M. ( 1984). The Department of Education's Sixth Antum also applies to those who would explain learning nual Report to Congress on PL94- I 42: Is Congress getting the problems with neuropsychological constructs. full story? Exceptional Children, 51, 209-224.
References Algozzine, B., & Korinek, L. (1985). Where is special education for students with high prevalence handicaps going? Exceptional Children, 51, 388-394. Algozzine, B., & Ysseldyke, J. (1983). Learning disabilities as a subset of school failure: The over-sophistication of a concept. Exceptional Children, 50, 242-246. Arter, J. A., & Jenkins, J. R. (1979). Differential diagnosisprescriptive teaching: A critical appraisal. Review of Education Research, 49, 517-555. Bender, L. A. (1938). A Visual Motor Gestalt Test and its clinical use. Research Monographs of the American Psychiatric As-
Gob, D. S., Telzrow, C. I., & Fuller, G. B. (1981). The practice of psychological assessment among school psychologists. Professional Psychology, 12, 696-706. Goodman, I. F. (1983). Organicity as a construct in psychological diagnosis. InT. R. Kratochwill (Ed.), Advances in scl!ool psychology (Vol. III). Hillsdale, NJ: Erlbaum. Graden, J. L., Zins, J. E., & Curtis, M. J. (Eds.). 0988). Alternative educational delivery systems: Enhancing instructional options for all students. Washington, DC: National Association of School Psychologists. Grimes, J. P., & Reschly, D. J. (1986). Relevant Educational Assessment and Intervention Model (RE-AlM) (Project proposal funded by the United States Department of Education). Des Moines: Iowa Department of Education, Bureau of Special Education.
518
CHAPTER 27
Hammill, D., & Larsen, S. (1974). The effectiveness of psycholinguistic training. Exceptional Children, 4 I, 5-14. Hammill, D., & Larsen, S. (1978). The effectiveness of psycholinguistic training: A reaffirmation of position. Exceptional Children, 44, 402-414. Hartlage, L. C. (1981 ). Clinical application of neuropsychological test data: A case study. School Psychology Review, 10, 362366 . . Hartlage, L. C. (1982). Neuropsychological assessment techniques. In C. R. Reynolds & T. B. Gutkin (Eds.), Handbook of school psychology. New York: Wiley. Hartlage, L. C. (1986). Pediatric neuropsychology. In D. Wedding, A.M. Horton, &J. Webster(Eds.), The neuropsychology handbook: Behavioral and clinical perspectives. Berlin: Springer. Hartlage, L. C., & Reynolds, C. R. (1981). Neuropsychological assessment and the individualization of instruction. In G. W. Hynd & J. E. Obrzut (Eds.), Neuropsychological assessment and the school age child: Issues and procedures. New York: Grune & Stratton. Heller, K., Holtzman, W., & Messick, S. (Eds.). (1982). Placing children in special education: A strategy for equity. Washington, DC: National Academy Press. Hynd, G. W., Obrzut, J. E., Hayes, F., & Becker, M.G. (1986). Neuropsychology of childhood learning disabilities. In D. Wedding, A. M. Horton, & J. Webster (Eds. ), The neuropsychology handbook: Behavioral and clinical perspectives. Berlin: Springer. Kaufman, A. (1976). A new approach to interpretation of test scatter on the WISC-R. Journal of Learning Disabilities, 9, 160-168. Kaufman, A. (1979). WISC-R research: Implications for interpretation. School Psychology Digest, 8, 5-27. Kaufman, A. S., Goldsmith, B. Z., & Kaufman, N. L. (1984). KSOS: Kaufman Sequential or Simultaneous. Circle Pines, MN: American Guidance Service. Koppitz, E. (1975). The Bender Gestalt Test for young children (Vol. 2). New York: Grune & Stratton. Leinhardt, G., Bickel, W., & Pallay, A. (1982). Unlabeled but still entitled: Toward more effective remediation. Teachers College Record, 84, 391-422. Lund, K., Foster, G., & McCall-Perez, F. (1978). The effectiveness of psycholinguistic training: A reevaluation. Exceptional Children, 44, 310-321. Mercer, C. D., Hughes, C., & Mercer, A. R. (1985). Learning disabilities definitions used by state education departments. Learning Disability Quarterly, 8, 45-55. Minskoff, E. (1975). Research on psycholinguistic training: Critique and guidelines. Exceptional Children, 42, 136-144. Minskoff, E. H., Wiseman, D. E., & Minskoff, J. G. (1972). The MWM Program for Developing Language Abilities. Ridgeway, NJ: Educational Performance Associates. Naglieri, J. A., & Das, J. P. (in press). Planning, arousal, simultaneous, and successive (PASS): A model for assessment. JourTUJI of School Psychology. National Joint Committee for Learning Disabilities Position Paper. (1981, January 30). (Reprinted in the Society for Learning Disabilities and Remedial Education Newsletter. ( 1981, August). 1(4), 1-2).
Newcomer, R., Larsen, S., & Hammill, D. (1975). A response to Minskoff. Exceptional Children, 42, 144-148. O'Boyle, M. W. (1986). Hemispheric laterality as a basis for learning: What we know and don't know. In G. D. Phye& T. Andre (Eds.), Cognitive instructioTUJl psychology: Components of classroom learning. New York: Academic Press. Obrzut, A. (1981). A neuropsychological case report of a child with auditory-linguistic dyslexia. School Psychology Review, 10, 356-361. Reschly, D. J. (1986). Economic and cultural factors in childhood exceptionality. In R. T. Brown & C. R. Reynolds (Eds.), Psychological perspectives on childhood exceptionality: A handbook. New York: Wiley-Interscience. Reschly, D. J. (1987a). Learning characteristics of mildly handicapped students: Implications for classification, placement, and programming. In M. C. Wang, M. C. Reynolds, & H. J. Walberg (Eds.), The handbook of special educatiqn: Research and practice (Vols. 1-3). Elmsford, NY: Pergamon Press. Reschly, D. J. (1987b). Assessing educational handicaps. In A. Hess & I. Weiner (Eds.), Handbook offorensic psychology. New York: Wiley. Reschly, D. 1., Genshaft, J., & Binder, M.S. (1987). The 1986 NASP survey: Comparison of practitioners, NASP leadership, and university faculty on key issues. Washington, DC: National Association of School Psychologists. Reynolds, C. R. (1981). Neuropsychological assessment and the habilitation of learning: Considerations in the search for aptitude x treatment interaction. School Psychology Review, 10, 343-349. Reynolds, C. R. (1982). The problem of bias in psychological assessment. In C. R. Reynolds & T. B. Gutkins (Eds.), The handbook of school psychology. New York: Wiley. Reynolds, C. R. (1986). Transactional models of intellectual development, yes. Deficit models of process remediation, no. School Psychology Review, 15, 256-260. Reynolds, C. R., & Kamphaus, R. W. (1986). The Kaufman Assessment Battery for Children: Development, structure, and application in neuropsychology. In D. Wedding, A.M. Horton, & J. Webster (Eds.), The neuropsychology handbook: Behavioral and clinical perspectives. Berlin: Springer. Reynolds, M. C., Wang, M. C., & Walberg, H. J. (1987). The necessary restructuring of special and regular education. ExceptioTUJl Children, 53, 391-398. Salvia, J., & Ysseldyke, J. (1985). Assessment in special and remedial education (3rd ed.). Boston: Houghton Mifflin. Sandoval, J., & Haapanen, R. M. (1981). A critical commentary on neuropsychology in the schools: Are we ready? School Psychology Review, 10, 381-388. Shapiro, E. S. (1987). Behavioral school psychology. Hillsdale. NJ: Erlbaum. Shapiro, E. S., & Lentz, F. E. (1985). Assessing academic behavior: A behavioral approach. School Psychology Review, 14. 325-338. Shepard, L. A. (1983). The role of measurement in educational policy: Lessons from the identification of learning disabilities. Educational Measurement: Issues and Practice, 2, 4-8. Shepard, L. A.,_ & Smith, M. L. (1983). An evaluation of the
A LEAP OF FAITH identification of learning disabled students in Colorado. Learning Disability Quarterly, 6, 115-127. Sperry, R. W. (1968). Hemisphere disconnection and unity in conscious awareness. American Psychologist, 23, 723-733. Tombari, M., & Bergan, J. (1978). Consultant uses and teacher verbalizations, judgments, and expectancies concerning children's adjustment problems. Journal of School Psychology, 16, 312-319. Wade, T. C., & Baker, T. B. (1977). Opinions and use of psychological tests: A survey of clinical psychologists. American Psychologist, 32, 874-882. Wiederholt, J. L. (1974). Historical perspectives on the education of the learning disabled. InL. Mann&D. A. Sabatino(Eds.), The second review of special education. Philadelphia: JSE Press. Witt, J. C. (1986). Review of the Wide Range Achievement
519
Test-Revised. Journal of Psychoeducational Assessment. 4, 87-90. Ysseldyke, J. E., Algozzine, B., & Epps, S. (1983). A logical and empirical analysis of current practice in classifying students as handicapped. Exceptional Children, 50, 160-166. Ysseldyke, J. E., Algozzine, B., Shinn, M., & McGue, M. (1982). Similarities and differences between underachievers and students labeled as learning disabled. Journal of Special Education, 16, 73-85. Ysseldyke, J. E., & Mirkin, P. K. (1982). The use of assessment information to plan instructional interventions: A review of the research. In C. R. Reynolds & T. B. Gutkin (Eds.), The handbook of school psychology. New York: Wiley. Ziskin, J. (1970, lst ed.; 1975, 2nd ed.; 1981, 3rd ed.). Coping with psychiatric and psychological testimony (3rd ed.). Venice, CA: Law and Psychology Press.
28 Child Behavioral Neuropsychology ARTHUR MAcNEILL HORTON, }R.
Recognition of the overlap between neuropsychology and learning is not new. One of the first pioneers to differentiate brain and mind was the famous British neurologist John Hughlings Jackson as early as 1872. Indeed, Gaddes ( 1981) suggested that contemporary notions and conjectures regarding learning disabilities have their roots in the research of 19th century neurologists who first began to elucidate the complex interrelationship of brain structure and language. As noted by Horton and Wedding (1984), there is a long history of this work. In the last 20 years, however, interest in brainbehavior relationships has expanded greatly. Contributing to this trend were striking demonstrations of the cross-cultural validity of neuropsychological knowledge (Horton & Wedding, 1984; Golden, 1981; Luria, 1966; Reitan & Davison, 1974) as well as very successful demonstrations of the application of neuropsychological data to school-aged children (Reitan & Davison, 1974; Horton & Wedding, 1984; Hynd & Obrzut, 1981). It would appear straightforward that human neuropsychology is of great relevance to the practice of clinical child psychology. In addition to the tremendous growth of neuropsychology, there has been one other dramatic development in the practice of psychology with schoolaged children. This, of course, is the development of behavior modification/therapy programs (Rimm & Masters, 1978; Wolpe, 1973). Since the first introduction of large-scale token economics following the work of Ayllon and Azrin (Kazdin, 1978), there has been a strong and substantial adoption of behavioral methods by school-based, institution-supported, and independent providers of psychological services to ARTHUR MAcNEILL HORTON, }R. • Veterans Administration Medical Center, Baltimore, Maryland 21218; Depart-
ment of Psychiatry, University of Maryland Medical School, Baltimore, Maryland 21201; and Psych Associates, Towson, Maryland 21214.
children. The extent of the empirical data base supporting behavioral methods is only paralleled by that amassed by human neuropsychology. Given these twin developments, a natural question would be the possibility of amalgamation. As might be expected, there have been some considerable efforts devoted to the integration of assessment methods derived from human neuropsychology and treatment approaches based, at least initially, on learning theory. A new specialty that blends behavior therapy with neuropsychology has been proposed (Horton, 1979) and tentatively titled, "behavioral neuropsychology.'' Psychologists and other professionals involved in the behavioral treatment of children with learning problems that are related to neuropsychological functioning are expected to find this area of research and clinical practice valuable. The intent of this chapter is to explore the research and clinical knowledge base that underlies behavioral neuropsychology and with that objective in mind, the first step might be to delineate what is meant by behavioral neuropsychology. The first use of the term was in 1978, when a Special Interest Group (under that title) was formed under the administrative aegis of the Association for Advancement of Behavior Therapy (AABT). The first meeting of the group was held in Chicago at the annual AABT meeting and the Special Interest Group has continued to be active in AABT over the years. Shortly following the founding of the Special Interest Group, a tentative definition was advanced. This tentative definition was as follows: Behavioral Neuropsychology may be defined as the application of behavior therapy techniques to problems of organically impaired individuals while using a
Neuropsychological assessment and intervention perspective. This treatment methodology suggests that in-
clusion of data from Neuropsychological assessment strategies would be helpful in the formulation of hypotheses regarding antecedent conditions (external or
internal) for observed phenomena of psychopathology.
521
522
CHAPTER 28 That is, a Neuropsychological perspective will significantly enhance the ability of the behavior therapist to make accurate discriminations as to the etiology of patient behavior. Moreover, the formulation of a cogent plan of therapeutic intervention and its skillful implementation could, in certain cases, be facilitated by an analysis of behavior deficits implicating impairment of higher cortical functioning. (Horton, 1979, p. 20)
Although this definition is, of course, somewhat dated, it nonetheless provides a focus to further discussion. It is acknowledged that alternate interpretations of behavioral neuropsychology, as a field, might be quite valid, and that many individuals in either clinical neuropsychology or behavior therapy (or applied behavioral analysis) may have substantial reservations. The major portion of this chapter will be organized into four sections related to the specific application of behavioral neuropsychology with school-aged children. The first section will focus on theoretical issues and will attempt to elucidate how behavioral neuropsychology is compatible with both radical and contemporary variations of behaviorism. The second section is concerned with treatment planning issues and will discuss some models for constructive intervention. The third section will examine selectively some of the existing research on the application of behavioral methods with learning-disabled and structurally brain-damaged children. The fourth section will function as a concluding summary, but also will include some tentative hypotheses about possible future developments of child behavioral neuropsychology.
Theoretical Issues Behavioral Concerns The first theoretical issue to be dealt with is the relationship of radical behaviorism with neuropsychology. The second theroetical issue is how more contemporary notions of behaviorism can be reconciled with neuropsychology. These comments will be brief as more lengthy discussions are available (Horton, 1979, 1981; Horton & Puente, 1986). Simply put, the radical behaviorist model holds that the totality of human behavior can be satisfactorily accounted· for by observed stimulus-response relationships (Watson, 1913). Proponents of this view agree that the behavior of humans can be explained without the need to postulate covert or unobserved factors (Skinner, 1938). Put another way, var-
iables that are not observable stimulus-response actions are not necessary to account for human behaviors (Marr, 1984). Variables that may not be observed (i.e., inferred variables) are therefore regarded as essentially worthless by radical behaviorists. In order to understand how neuropsychological data might fit into a radical behaviorist paradigm, it is important to consider the proposition that there may be legitimate inferred variables in the functional analysis of human behavior (Mahoney, 1974). As agreed elsewhere (Horton, 1979), inferred variables might be considered to come in two distinct categories. These are intervening variables and hypothetical constructs (Craighead, Kazdin, & Mahoney, 1976). As noted by Horton (1979): . . . an intervening variable is a theoretical creation. For instance, no one has ever observed intervening variables, such as thoughts or feelings, yet they are used by cognitive behavior therapists to explain behavior. It could be said that, at least as far as we now know it, it is unlikely one would be able to directly observe an intervening variable. (Horton, 1979, p. 21)
Simply put, intervening variables exist only in theory as they are conceptual abstractions. Hypothetical constructs, on the other hand, are generally seen as more physical or empirical than intervening variables. As observed by Horton (1981): . . . a hypothetical construct is an actual physical object or process which is unobservable at the present time. For instance, hypothetical constructs in Neuropsychology tend to have physiological referents and can, if so desired, be verified. If a child evidences certain characteristics, it might be postulated that there is damage to the right parietal lobe. In this case. our hypothetical construct is based in our knowledge of brain-behavior relationships and can be verified through neurosurgical procedures. (Horton, 1981, p. 368)
In an earlier publication, Horton ( 1979) further elaborated upon the differences between intervening variables and hypothetical constructs. As observed by Horton: ... an example of this distinction would be the behavior of failing to draw a Greek cross and the explanation of this behavior by each inferred variable. An intervening variable could opt for an explaining mechanism such as an emotional state as the cause. Using a hypothetical construct, one might postulate the impairment of the right parietal lobe. While this example is grossly over-simplified, the major point should be clear. Hypothetical constructs tend to have physiological referents. The major advantage is that at some point, by some
CHILD BEHAVIORAL NEUROPSYCHOLOGY means, its existence or nonexistence can be verified. In the instance cited, neurosurgical procedures could determine the actual condition of the right parietal lobe in our subject. While in clinical practice, this is rarely done, the distinction is important. At present, methods for the direct objective verification of a thought or feeling have yet to be adequately developed. (Horton, 1979, p. 21)
Relative to neuropsychology and behavior therapy, the cardinal element of the preceding statement is that neuropsychological data are hypothetical constructs. Therefore, neuropsychological data, because they are hypothetical constructs, are in a different class of elements from intervening variables. Because of these differences, it might be argued that the inclusion of hypothetical constructs in the behavioral paradigm has markedly different implications from the inclusion of intervening variables. It is, of course, understandable that rejection of intervening variables by the S-R model of behaviorism at the time of John B. Watson (1913), and when B. F. Skinner (1938) did his early work, was a rational decision based on the inability of neuroscience to contribute to a functional analysis of human behavior. With respect to hypothetical constructs and neuroscience, however, there has been an explosive rate of development over the decades. As noted by Horton (1979): Now, however, it could be observed that the knowledge base of brain-behavior relationships has changed drastically since the days of Watson. At the time, Neuropsychology was unable to provide even rudimentary guidance for research minded behaviorists. Clearly, the most appropriate strategy has a benign neglect of the area. Today, however, is a drastically different situation. In the last 20 years, the knowledge base of brain-behavior relationships has increased geometrically (Davison, 1974). Cross-cultural research has provided such impressive validation of Neuropsychological insights that it would appear difficult to minimize their importance (Luria, 1966; Hecaen & Ajuriaguerra, 1964; Faglioni, Spinnler, & Vignola, 1969) of Neuropsychological factors, it would seem that the time for their inclusion in an enlarged behavioral paradigm is at hand. (Horton, 1979, p. 22)
Although the ultimate test of the above remarks is of necessity empirical, it should be clear that neuropsychological factors are accessible to measurement and therefore cannot be dismissed on the grounds that they are unscientific. A related but clearly independent conceptual issue might be mentioned at this point. That is, how do contemporary views of behavioral assessment and treatment blend with behavioral neuropsychology?
523
In this context, it is worth recalling that when the journal Behavioral Assessment was launched in 1979, the founding editor, Rosemary Nelson, in her editorial statement describing the role of the journal, stated that behavioral assessment emphasizes: . . . both meaningful response units and . . . their controlling variables. Behavior is defined functionally, in relation to its present controlling variables (both environmental and organismic) and to its responsiveness to intervention strategies.
Along these lines, it might be stated that contemporary behavior therapy has been characterized by an evolving clinical acumen. A portion of the increased sophistication in behavior therapy might be attributed to improved behavioral assessment techniques. Indeed, in the last decade, there has been a tendency to increasingly focus on the assessment aspects of behavior therapy. Nelson (1983), for example, stated that behavior therapy, to an increasing degree, is defined by the techniques used. For example, some authorities in the field of behavior therapy would classify "self-monitoring" as a clearly behavioral technique, while at the same time holding that the Minnesota Multiphasic Personality Inventory (MMPI) is clearly nonbehavioral (Hayes & Zettle, 1980). A difficulty with this assertion, however, is that it is quite arbitrary, not to mention its nonempirical nature (Horton & Puente, 1986). It would appear that some better method of classifying particular techniques as behavioral or nonbehavioral might be found. Interestingly, Hayes and Zettle ( 1980) discussed a particularly progressive conceptual model. Their paradigm is based on the distinctions between conceptual (how to talk about techniques) and technical (how to perform techniques) dimensions of behavioral assessment and treatment. They postulated that the most reasonable guideline is to favor conceptual as opposed to technical dimensions when attempting to distinguish between whether or not a technique can be discussed in terms of behavioral principles. Strictly speaking, who devised the technique or the topographical details of the procedure are not relevant to the decision of whether or not the technique is behavioral. The major point is that both the antecedents and consequences of an act must be viewed in order to determine what the purpose of the action was. The simple physical details matter little except to the degree they enable one to deduce the intended purpose of the action under study. In short, if it is possible to talk about a method in terms of behavioral principles, and the methods yield outcomes that can be objectively assessed, then it is behavioral.
524
CHAPTER 28
The above conceptualization of behavior therapy has some implications for behavioral neuropsychology. For example, as noted by Horton (1981): If behavior therapy is defmed in a conceptual sense, then clinical Neuropsychological assessment instruments such as the Halstead-Reitan Neuropsychological Test Battery or the Luria-Nebraska Neuropsychologkal Test Battery can be classified as "behavioral." ... If the most appropriate goal of behavior therapy is a "clinical science based upon clinical realities" (Hayes & Zettle, 1980), then it would appear that a conceptual view would be preferable. Thus, a Neuropsychological perspective could be integrated into such an enlarged and clinically realistic behavioral paradigm. Whether or not such a blend of Neuropsychology and behaviorism proves a potent addition . . . remains an empirical question, which in the best tradition of behaviorism should be objectively tested. (p. 369)
At this point, some consideration should be given to the goals of treatment planning. Many treatment programs focus on either restitution of desired behavior or amelioration of undesired behavior. An assumption of this approach is a rigid localization model of brain-behavior relationships (Horton & Puente, 1986). The result is a treatment program focused on patient deficits. The actual value of this orientation has been questioned. Reynolds (198lb) and others have suggested that a more powerful approach would be to focus on the child's strengths rather than weaknesses. Horton, Wedding, and Phay ( 1981), as well as Horton and Miller ( 1985), provide extended discussions of this issue and due to space limitation, those discussions will not be repeated here. Simply put, it is suggested that focusing on strengths is the best way to maximize treatment efficacy.
Therefore, if the necessary empirical basis can be amassed, then a neuropsychological perspective could be profitably subsumed into a sophisticated and augmented contemporary behavioral model.
Treatment Strategies
Neuropsychological Issues
Lewinsohn's Model
Over a number of years, Peter M. Lewinsohn at Important considerations relative to planning behavioral treatments for brain-impaired children in- the University of Oregon Neuropsychology Clinic clude consideration of the context of neurological has made impressive research contributions to the development and appreciation of the behavioral se- literature on the remediation of memory deficits in quelae of neuropsychological impairment. Regard- brain-damaged persons (Lewinsohn, Danaker, & ing developmental neuropsychology, Miller (1984), Kikel, 1977; Galsgow, Zeiss, Barrera, & Lewinafter reviewing the infrahuman research, observed sohn, 1977). As part of this research effort, Lewinthat in all cases, a specific recovery pattern should be sohn, with his associates and doctoral students, has expected and that the more practical the skills, the developed a valuable model for clinical work with less impaired they are, as a general rule, and that neuropsychologically impaired persons that is likely children show plasticity of neural and behavioral to work well in conceptualizing interventions with functions. Perhaps most important, Miller (1984) children. The model is divided into four steps: the concluded that intervention, particularly early inter- first two steps are concerned with assessment and the vention, appeared to facilitate recovery of function. second two steps deal with treatment: Another issue is the behavioral effects of neural I. General assessment of neuropsychological dysfunction. At least one group of authors (K.lonoff, functioning Crockett, & Clark, 1984) has concluded that there is 2. Specific assessment of neuropsychological a significant relationship between environmental facfunctioning tors and brain injury and that the sequelae of brain 3. Laboratory evaluation of intervention techinjury are related to age with younger children showniques ing emotional and personality changes and older chil4. In vivo application of intervention techdren displaying learning and memory difficulties. niques Horton and Puente ( 1986) concluded that: The following paragraphs will discuss each . . . treatment planning should take into consideration step. Regarding the first step, usually standard neuroenvironmental factors such as actual physical environpsychological batteries are used. For example, one ment as well as family structure in order to minimize might use an age-appropriate Wechsler Intelligence future occurrences of Neural impairment as well as to Scale, the Kaufman Assessment Battery for Children maximize the general ability of the office or institution(K-ABC), the age-appropriate Halstead-Reitan based treatment program.
CHILD BEHAVIORAL NEUROPSYCHOLOGY
Neuropsychological Test Battery, the Reitan-Indiana Neuropsychological Battery, or the age-appropriate Luria-Nebraska Neuropsychological Battery. The purpose of the first step is to obtain normative psychometrics. This enables a comparison of the patient with other patients on the basis of descriptive statistics. This promotes a global understanding of the patient's problem. By contrast, the second step focuses on personalized understanding of the patient's problem. Specifically, the intent is to examine in detail the precise details of the patient's problem. For example, in using the WISC-R Picture Arrangement subtest, it is possible to obtain a standard score. That is, the use of normative comparison as the standard score can be compared to other children in the normative population with great accuracy. To develop a more intuitive understanding of the child's deficits, the child might be requested to verbalize what story he or she can make out of the pictures and then progressively more help can be given the child or prompts can be used. By using this dynamic approach, a better understanding of the actual dimension of the child's deficits can be elucidated, which is often different than that reached by normative psychometrics. In the second step, the emphasis is on interindividual comparisons. The risk of the second step, of course,. is that the examiner will overinterpret the findings. On the other hand, the advantage is great flexibility to investigate the actual S-R dimension of the problem behavior and conduct a functional analysis of the behavior deficit. In the third step, the focus is on a controlled (or laboratory) setting. There, specific intervention techniques are introduced to deal with the problem behavior identified and defined in the first two steps. To extend the example of talking out loud to pictures of the Picture Arrangement subtest of the WISC-R, one technique might be to train verbal self-instruction to a variety of predesigned problem situations (i.e., puzzles, role-playing). After intervention effectiveness can be demonstrated, efforts toward generalization of the techniques might be initiated. In the fourth step, the effective intervention strategy of the laboratory is introduced to the real world. Clearly, there are major differences between a strategy working under a controlled laboratory situation and a strategy working in the real world. Additional adjustments, alternatives, and possible augmentations of treatment strength may be necessary, as well as rearrangement of environmental stimuli and contingencies. Extending the example used earlier, the patient might need to be trained to initiate his/her verbal self-instructional technique to a vari-
525
ety of situations and the technique may need to be practiced until it becomes an automatic response to emotional stimuli. Therefore, it can be seen that Lewinsohn 's paradigm provides a robust model for conceptualizing clinical behavior therapy with the brain injured. The overall framework of general and specific assessment with specific intervention in a controlling setting followed by generalization to the real world can be adapted to multiple techniques and assessment devices. Indeed, the range of quantitative and qualitative measuring devices is only limited by ingenuity. To illustrate some of these considerations, Goldfried and Davison ( 1976) proposed a framework that overlaps with steps two and three ofLewinsohn 's model. To provide a measure of conceptual depth, some attention will be devoted to explicating the Goldfried and Davison paradigm. In order to account for maladaptive behavior, Goldfried and Davison proposed four types of important variables: 1. 2. 3. 4.
Stimulus antecedents Organismic variables Response variables Consequent variables
Essentially the paradigm is a liberalization of the classical S-R model of radical behaviorism. To the traditional framework, the variables of "organismic" (perhaps it could be suggested that in addition to other physiological factors, neurological factors are included in this category) and "consequents" (the roles that reinforcing and punishing events play in determining the frequency of human actions (Skinner, 1981) are so well documented as to need no further explanation here) are added. The resulting model enables one to more adequately conceptualize the various domains that need to be considered in an enlightened behavioral assessment model. In order to futher develop this line of reasoning, it would be appropriate to devote some consideration to treatment planning issues. Due to space limitations, these remarks will of necessity be quite concise. More elaborate discussions are available elsewhere (Horton & Wedding, 1984; Golden, 1981; Luria, 1963; Miller, 1984; Horton & Sautter, 1986). Some issues selected for discussion might include the following: (1) self-efficacy, (2) personality x treatment interactions, and (3) available resources for support. Regarding self-efficacy, this notion was first proposed by Albert Bandura (1982). He suggested
526
CHAPTER 28
that how a person perceives his or her personal effectiveness is a major factor in accounting for a degree of therapeutic behavior change. To a large measure, Bandura would assert that behavior change techniques that work do so by the mechanism of inducing and increasing an individual's self-estimate of effectiveness. This personal belief system influences activities chosen as well as the amount of the persistence of effort when aversive consequences are encountered. In Bandura's initial formulation of the concept of self-efficacy, he postulated that four data sources influenced these beliefs. Stated in a somewhat oversimplified fashion, they are as follows:
University and his colleagues have pioneered a cognitive-somatic typology of anxiety-reducing activities. Some clients/patients/subjects do better with relaxation as their anxiety is somatic based; others do better with meditation as their anxiety is cognitively based. Interestingly, those with mixed cognitivesomatic symptoms do best with active sports that appear to involve both cognitive and somatic dimensions. The availability of resources is the next topic for discussion. This refers to both the environmental characteristics of the treatment setting (institutional and/ or community based) as well as personal and family qualities. In some cases, the skills of the thera1. Successful personal behavioral performance pist(s) and referral sources might qualify. It has been 2. Observing successful performances by well documented (Diller & Gordon, 1981) that famiothers ly members can serve as mentors and therapists and 3. Personal status of physiological arousal their availability in these roles can be crucial. Also, · 4. Verbal persuasion selection of therapeutic techniques is related to the These data sources are stated in the order of their ability of institutional and/ or community treatment presumed potential for successful modification of settings to provide follow-up. It is of little value to self-efficacy beliefs (i.e., personal behavioral perfor- propose a treatment modality that is impractical to mance is the most powerful, verbal persuasion is the implement. In summary, analysis and intervention utilizing weakest). Some implications of self-efficacy for treatment the aforementioned models (Lewinsohn, behavioral planning might be mentioned at this point. It should assessment and treatment planning) would appear to appear clear that if possible, in vivo behavioral per- provide a robust paradigm for a successful behavioral formance would be the preferred therapy mode. In neuropsychology program for children. dealing with brain-injured children, the use of performance-based feedback may, of course, be difficult and the necessity for special facilitative conditions Guidelines from Behavioral and assistance devices should be encouraged. In in- Neuropsychology stitutions where successful personal behavioral perAlthough the generation of behavior therapy informance is impossible to arrange, the observance of successful performances by others and so on should terventions for brain-impaired children is no easy be used. If one can propose that motivation for task, some rudimentary suggestions might be garchange is related to the reinforcement for performing nered from careful if cursory attention to basic neuan action times the individual's assessment of the roanatomical dimensions. The neuro-cortex has been likelihood of accomplishing the action successfully, described by Meier (1974) as consisting of three prithen it would appear wise to do all things possible to mary parameters: increase subjectively assessed probabilities of sucl. Left to right cess. Simply put, personal predictions of success2. Front to back ful performance are crucial factors in treatment 3. Top to bottom planning. Put another way, Horton and Wedding (1984), Similarly, personality x treatment interactions deserve careful scrutiny. Personality x treatment in- in consideration of Meier's (1974) conceptualizateractions are client/patient/subject characteristics tion, termed the left to right parameter as "laterthat influence the success of particular therapeutic ality," the front to back parameter as "caudality," methods. One example might be a client/ patient/ sub- and the top to bottom parameter as "dorsality." It ject's personal standards for self-reinforcement should be well understood that these terms are util(Goldfried & Davison, 1976). Those with very high ized in a special context in this discussion. Similarly, standards might be given relatively easy therapeutic the parameters are intended to illustrate very general goals as any failure experiences might destroy moti- and rudimentary concepts and there is an extreme vation for change. Similarly, Gary Schwartz of Yale degree of oversimplification. Knowledgeable schol-
CHILD BEHAVIORAL NEUROPSYCHOLOGY
ars could find multiple and valid exceptions by the score to this framework, but for the sake of conceptual communicative ease, the above framework will be used. In addition, the following suggestions presuppose circumscribed and localized cerebral insult. Hopefully, these suggestions will serve as hypotheses for future research that will enable the generation of more valid and generalized understanding of the delicate interplay of neuropsychology and behavior therapy.
Laterality Perhaps no other single concept has generated more research in American neuropsychology than that of laterality. The notion of laterality in the human brain has been so well accepted that it has become the stuff of popular culture. In terms of treatment planning, Horton and Wedding (1984) observed that: The two cerebral hemispheres process information in different ways. Assuming right handedness, the left hemisphere is logical and language oriented, while the right hemisphere is intuitive and concerned with spatial aspects of stimuli. (p. 216)
Similarly, Reynolds (l98la) noted that: For the vast majority of individuals, the left cerebral hemisphere appears to be specialized for linguistic, propositional, serial, and analytic tasks, and the right hemisphere for more nonverbal, spatial, appositional, synthetic, and holistic tasks .... It is most important to remember that cerebal hemispheric asymmetries of function are process-specific and not stimulis-specific. (p. 109)
527
tional interventions, emotional correlates, and prognostic predictions for three neuropsychological subtypes. In Table I, the neuropsychological profiles for children are presented. As can be seen from Table 1, Hartlage sees type I children as exhibiting 'left hemisphere dysfunctions, type II children as typifying right hemisphere dysfunction, and type III children as demonstrating dysfunction of both cerebral hemispheres. It should be well understood that Table I is a gross oversimplification of the extremely complicated and complex clinical reality. At the same time, this oversimplification still is an attempt at general rules for rational behavioral intervention procedures selection. Even with rather dramatic limitations, it demonstrates some incremental validity over a random trial-and-error procedure. As more specific understandings develop from the current research efforts in child neuropsychology, rather drastic modification of Table 1 will certainly be expected.
Caudality This dimension refers to localized cerebral impairment on the front to back or, put another way, anterior-posterior axis. Horton and Wedding (1984) noted that: There is some agreement that the frontal lobes involve the planning, execution, and verification of behavior while the posterior sections are involved with the reception, integration, and analysis of sensory information .... (p. 219)
The frontal lobes, of course, have been a topic Horton, Owens, and Hartlage (1980) observed of great interest in cognitive neuropsychology. Given that modes of communication, therapy tasks, and the wealth of material, no attempt to condense it will therapeutic management may all be influenced by be ventured here. Rather, interested readers are enhemispheric mental asymmetry. To cite but a single couraged to examine a recent paper that reviews the example, many clinicians and researchers (Broder, behavioral effects of frontal lobe lesions and the at1973; Hartlage, 1975; Mattis, French, & Rabin, tendant emotional and psychological consequences 1975; Pirozzolo, 1981) have argued eloquently and (Struss & Benson, 1984). It might be mentioned, presented data suggesting the existence of subtypes however, that it has long been recognized that frontal of children with reading disabilities. Also, they have lobe impairment is of particular clinical significance. indicated a consensus that neuropsychological as- For example, Luria (1966), in discussing his efforts sessment is important to identify reading disability at the rehabilitation of brain-injured veterans of subtypes and in guiding appropriate educational in- World War II, mentioned that those with frontal lobe tervention. Of particular interest is that the two most impairment were rarely able to leave even a sheltered common reading disability subtypes have auditory- workshop setting. It would appear that impairment of linguistic and visuospatial elements (Pirozzolo, the frontal cortex is particularly dolorous for self1981 ). Hartlage ( 1975) has been among the leaders in management skills. Indeed, Horton and Wedding this area of research and clinical practice. In his ( 1984) postulated that whether or not a lesion is in the work, he has devised a conceptualization of hemi- frontal area is more important for predicting overall spheric mental asymmetry with respect to educa- behavioral adjustment than even the overall extent of
•Adapted from Hartlage (1975).
Prognosis
Educational intervention
Neurological syndrome Emotional correlates
Neuropsychological profile Comparatively lower WISC-R Verbal than Perfonnance IQ score with consistently lowered language ability (i.e., depressed ITPA and PPVT scores) relative to perceptual motor skills (i.e., Bender-Gestalt or VMI) Left hemisphere dysfunction Reserved, tentative, and uncertain of self-efficacy Whole word or look-say reading programs and perceptually oriented instructional modes Persistent problem during academic career (after 3rd grade) but relatively good adjustment in nonacademic pursuits
Type I child
Difficulty in early school grades (K-2) but tend to do better in later elementary grades (3-6) with generally successful academic career
Right hemisphere dysfunction Impulsive and uncritical of personal performance Linguistic and aural instruction modes
Comparatively lower WISC-R Performance than Verbal IQ and consistently lowered perceptual-motor ability relative to language skills
Type II child
TABLE 1. Basic Neuropsychological Profiles for Childrena
Little ultimate academic success
Extreme structure and special placement
Generalized cerebral dysfunction Restless, irritable, and hyperactive
No consistent pattern of WISC-R strengths and weaknesses or clear superiority of either language or perceptual-motor abilities and skills
Type III child
~
~
~
CHILD BEHAVIORAL NEUROPSYCHOLOGY
brain impairment on objective measures of neuropsychological functioning. With respect to therapeutic implications with children, there appear to be a number. One example might be the use of Meichenbaum's (1977) self-instructional therapy to increase self-management skills for frontally impaired children. It might be of great interest to utilize developmentally appropriate methods like the turtle technique (Schneider & Robin, 1976). In a study with an adult with frontal lobe impairment, Horton (1984) utilized the turtle technique, a method for the self-control of impulsive behavior, to decrease temper outbursts. Because that case report was an extension of a technique originally developed for children with an adult, it would be interesting to attempt the method with a brain-injured member of the population the technique was designed for.
Dorsality This dimension is the vertical axis. Put another way, dorsality refers to the top to bottom parameter of the brain. There is some theoretical (MacLean, 1973) and experimental work (MacLean & Delgado, 1953) to suggest evolutionary distinct layers of brain tissue. Although there has been relatively little attention devoted to this dimension by clinical neuropsychologists insofar as their test batteries are concerned, (Meier, 1974; Horton & Wedding, 1984), there is considerable theoretical interest by workers concerned with the neurobiology of aggression (Bear, 1986). In this light, Paul MacLean's theory of the triune brain appears most insightful. There is, however, a dearth of well-accepted treatment implications. Horton and Puente (1986) observed: The clinical implication is that the depth of brain-impairment could have great relevance. It should be freely admitted that at present, the knowledge of brain-behavior relations, relative to dorsality, is not adequate enough to generate many meaningful treatment suggestions for impairment to developing brains.
Perhaps one of the few valuable suggestions is one advanced by Horton and Wedding (1984): When there are cortical lesions, there are often concomitant personality changes. It is also commonly observed the premorbidly controlled antisocial character traits of brain-damaged individuals are released after the onset of the brain injury. This syndrome might be
529
explained by the triune brain model of Paul MacLean. (p. 220)
Empirical Considerations In this section, research on the use of behavior modification/therapy techniques with children will be selectively reviewed. It is worth noting that the behavior modification/therapy literature, in general (i.e., adults and children), has been reviewed before. lnce ( 1976) surveyed the use of behavior modification with brain-injured persons in a review that was noted by Diller and Gordon (1981). Similarly, Horton (1979, 1982) has twice surveyed the research literature regarding the application of behavior theraPY with brain-damaged individuals. More recently, Horton and Miller (1985) again surveyed this research literature and found an increasing trend toward using behavior modification/therapy with brain-damaged individuals. Table 2 is adapted from the review of Horton and Miller ( 1985) that focuses exclusively on behavior modification/therapy studies with brain-damaged children from 1967 to 1984. A few limitations of Table 2 might be mentioned at this point. First, although the available literature was reviewed, Table 2 is, for the most part, a selective review and no claim of comprehensiveness is made. Also, studies focusing on biofeedback were not included. Although it is clear that biofeedback is often considered part of behavior therapy, it was decided to focus primarily on more specifically operant interventions as the biofeedback literature has been competently reviewed many times by others. It is felt this more focused review will serve to highlight the value of the more classical behavior therapy techniques with children. Also, for the most part, studies focused on structural brain damage rather than learning disabilities. Inspection of Table 2 reveals a number of interesting points. Of the 19 studies, only Dean ( 1984) and Denton and Citron (1983) were group design studies. Two were group case studies-Carlin and Armstrong (1968) and Salzinger, Feldman, and Portnoy (1970); and three were single-subject design studies (i.e., multiple baseline or ABAB design)Campbell and Stramel-Campbell (1982), Gajan, Schloss, Schloss, and Thompson (1984), and Muir and Milan (1982). The remainder appeared to be case studies. The types of brain-damaged patients treated included closed head injury, cerebral palsy, and Huntington's chorea, among others.
13-year-old male with minimal brain damage 10-year-old male with cerebral palsy; multiple baseline design Four boys-two with brain damage; group case study 7-year-old male with cerebral palsy; single-case design
Ninety learning-disabled (reading disorder) males, average age 10.6, range 9-13; group design study
Twenty adolescent males-seven with closed head injury; group design study
Brannigan & Young (1978)
Carlin & Armstrong (1968)
Dean (1984)
Denton & Citron (1983)
Cinciripini, Epstein, & Kotanchik (1980)
Campbell & StremelCampbell (1982)
Brain-damaged children
Population
Blyth (1969)
Author
Overcorrection for self-stimulation and differential reinforcement of attentional responses and behavior incompatible with self-stimulation Following neuropsychological assessment of strengths, a hierarchy of remedial tasks was constructed along an approach-avoidance continuum and differential levels of reinforcement were established Cognitive behavioral treatment for impulsivity
Social praise, repetition of child's responses, answers to child's questions, and a token Token rewards and fines
Social reinforcement and stimulus control and token economy Social skills training
Method
Remarks
Improved self-control
Group design with significant improvement in area of academic deficit and classroom behavior
Behavioral treatment reduced selfstimulation and increased attention
Improved social responsibility
Better social function and emotional control improved Language generalization facilitated
Improved behavior
TABLE 2. Selected Behavior Modification/Therapy Studies with Childrena.b
(11
~
~
~
~
•Adapted from Horton and Miller (1985). hSee the original for all references.
Waye (1980)
Sellick & Peck (1981)
Ribes-Inesta & Guzman (1974) Salzinger, Feldman, & Ponnoy (1970)
Murray (1978) Reidy (1979)
Muir & Milan (1982)
Krumchy & Kores (1971)
14-year-old female with deep brain lesion; case study Twenty-three families with braindamaged children (56% females, 44% males); group case study 28-rnonth-old male with cerebral palsy; case study 16-year-old female with Huntington's chorea
Three children with central nervous system dysfunction 8-year-old brain-damaged, emotionally disturbed male; case study Neurologically impaired pediatric inpatients Three children-two with cerebral palsy and one with developmental delay and seizure disorder; ABAB design _ Brain-damaged children; case study 7-year-old male with brain damage
Hall & Broden (1967)
Krop (1971)
Two head trauma youths; ABCACB design 16-year-old male with traumatic closed head injury; case study
Gajan, Schloss, Schloss, & Thompson (1984) Gerber, Major, Adams, & Spevack (1981)
Time-out contingencies and positive reinforcement
In vivo behavioral flooding/extinction
Parents instructed in behavior modification techniques
Time-out procedure Verbal commands, praise, and manual guidance Time-out, electric shock, and slaps
Reinforcement (praise standing and using binoculars) Parents reinforced for child's progress with lottery tickets
Reinforcement (praise and/or candy)
Light signal feedback versus selfmonitoring Operant reward contingencies with motorcycle pictures as rewards for performing cognitive stimulation exercises Social reinforcement
Tantrum behavior controlled Appropriate acting behavior was learned and generalized to home and school Reduced behavior of placing nonedibles in mouth Parents who carried out the program reported success in changing their child's behavior Decreased fear of physical insecurity and positive generalization Decreased temper outburst and public disrobing
Decrease in inappropriate behavior and improved appropriate behavior Children improved in language skills
Reduced hyperactive behavior
Children improved in target behaviors
Social interaction gains generalized to less structured situations Increases on Raven's progressive matrices and subscales of Pictorial Test of Intelligence
~
B
I I
6
Q
532
CHAPTER 28
In summary, the research literature on the use of behavior modification/therapy techniques with children is still in an early stage of development. Perhaps it's best to recall words written at the conclusion of an earlier review: A limitation of this literature is the generally neurologically simplistic adaptation of behavioral treatment methodology. Only recently have more sophisticated applications been attempted. . . . Still, the uniformly positive results suggest great promise in this area for future application. (Horton, 1982, p. 101)
Conclusions Recent years have witnessed impressive growth in neuropsychology. All expectations are for even more dramatic growth in neuropsychology in the next decade. This chapter has reviewed the status of a subfield of neuropsychology-behavioral neuropsychology. More to the point, attention was devoted to the ability of behavioral neuropsychology to deal with the mental, emotional, and behavioral problems of brain-damaged children. First, there is evidence that behavioral methods are effective with brain-damaged children. Second, the value of neuropsychological assessment to select behavior modification/therapy techniques is an area that will require much additional research. In truth, the majority of the research is at a case study or single-case design level, and there is clear need for well-controlled and methodologically sophisticated treatment group design research. It is clear that the final assessment of behavioral neuropsychology with children will rest on its ability to solve the mental, emotional, and behavioral problems of brain-damaged children. The desire and expectation is that this chapter will be of significant value in this effort. Although much additional research is needed, there is reason to feel that initial efforts were somewhat successful. Finally, as Horton (1982) observed: . . . It is clear that much additional work must be done in order to effectively integrate behavior therapy and clinical neuropsychology. At the same time, it should be noted that the field of therapy for the braininjured is in its infancy. Thus, it would be unrealistic to expect initial efforts on research fronts to demonstrate more than significant promise. Whether or not this promise will be fulfilled, however, is a question only the future may answer. At this point, one might reflect that Neal Miller's assertion that researchers should be bold in what they try, but cautions in what they claim is a point well taken. (p. 102)
The author's contributions to this chapter were made in his private capacity and without support or endorsement by the Veterans Administration.
References Bandura, A. (1982). Self-efficacy mechanism in human agency. American Psychologist, 37, 122-147. Bear, D. M. (1986, May). Hierarchical neurology of human aggression lecture presented to the Baltimore-Washington Neuropsychology Group meeting, Bethesda, MD. Broder, E. (1973). Developmental dyslexia: A diagnostic approach based on three atypical reading-spelling patterns. Developmental Medicine and Child Neurology, 15, 663-667. Campbell, C., & Stremel-Campbell, K. (1982). Programming "loose training" as a strategy to facilitate language generalization. Journal of Applied Behavior Analysis, 15, 295301. Carlin, A. S., & Armstrong, H. E. (1968). Rewarding social responsibility in disturbed children: A group play technique. Psychotherapy: Theory, Research and Practice, 5, 169-174. Craighead, W. E., Kazdin, A. E., & Mahoney, M. J. (1976). Behavior modification: Principles, issues and applications. Boston: Houghton Mifflin. Dean, R. S. (1984, August). Treatment of learning disorders with neuropsychological impairment: A behavioral approach. Paper presented at the American Psychological Association meeting, Toronto, Canada. Denton, A., & Citron, C. (1983, August). The development of group intervention strategies for impulsive adolescents with cognitive and language deficits. Paper presented at the annual meeting of the American Psychological Association, Anaheim, CA. Diller, L., & Gordon, W. A. (1981). Interventions for cognitive deficits in brain-injured adults. Journal of Consulting and Clinical Psychology, 49, 822-834. Faglioni, P., Spinnler, H., & Vignola, L.A. (1969). Contrasting behavior of right and left hemisphere-damaged patients on a discriminative and a semantic test of auditory recognition. Cortex, 5, 366-389. Gaddes, W. H. (1981). Neuropsychology, factor mythology, educational help or hindrance? School Psychology Review, 10(31), 322-330. Gajan, A., Schloss, P. J., Schloss, C. N., & Thompson, C. K. (1984). Effects offeedback and self-monitoring on head trauma youths' conversational skills. Journal of Applied Behavioral Analysis, 17(3), 353-358. Glasgow, R., Zeiss, A., Barrera, M., & Lewinsohn, P. (1977). Case studies on remediating brain damage deficits in brain damaged individuals. Journal of Clinical Psychology, 33. 1049-1054. Golden, C. J. (1981). Diagnosis and rehabilitation in clinical neuropsychology. Springfield, IL: Thomas. Goldfried, M. R., & Davison, G. C. (1976). Clinical behavior therapy. New York: Holt, Rinehart & Winston. Hartlage, L. C. (1975). Neuropsychological approaches to predicting outcome of remedial educational strategies for learning disabled children. Pediatric Psychology, 3, 23-28.
CHILD BEHAVIORAL NEUROPSYCHOLOGY
Hayes, S. C., & Zettle, R. D. (1980). On being "behavioral": The technical and conceptual dimensions of behavioral assessment and therapy. The Behavior Therapist, 3(3), 4-6. Hecaen, J., & Ajuriaguerra, J. (1964). Left-handedness. Manual superiority and cerebral dominance (E. Ponder, Trans.). New York: Grone & Stratton. Horton, A. M., Jr. (1979). Behavioral neuropsychology: Rationale and research. Clinical Neuropsychology, 1, 2Q-23. Horton, A. M., Jr. (1981). Behavioral neuropsychology in the schools. School Psychology Review, 10(33), 367-372. Horton, A.M., Jr. (1982). Behavioral neuropsychology: A brief rationale. The Behavior Therapist, 5(1), 100-102. Horton, A.M., Jr. (1984). Useoftheturtletechniquewithabraindamaged patient: A case study. The Behavior Therapist, 7(2), 18, 34. Horton, A. M., Jr., & Miller, W. G. (1985). Neuropsychology and behavior theory. In M. Hersen, R. Eisler, and R. Miller (Eds.), Progress in behavior modification. New York: Academic Press. Horton, A. M., Jr., Owens, R. C., & Hartlage, L. C. (1980, September). Neuropsychological assessment in private practice. Conversation hour sponsored by the American Society of Psychologists in Private Practice at the 88th Annual Meeting of the American Psychological Association, Montreal, Canada. Horton, A.M., Jr., &Puente, A. E. (1986). Behavioral neuropsychology with children. In G. W. Hynd&J. E. Obrzut(Eds.), Child neuropsychology (Vol. D). New York: Grone & Stratton. Horton, A.M., Jr., & Sautter, W. (1986). Behavioral neuropsychology. InD. Wedding, A.M. Horton, Jr., &J. S. Webster (Eds.), Handbook of clinical and behavioral neuropsychology. New York: Springer. Horton, A.M., Jr., & Wedding, D. (1984). Clinical and behavioral neuropsychology. New York: Praeger. Horton, A. M.,Jr., Wedding, D., &Phay, A. (1981). Contemporary perspectives on assessment and therapy for the braindamaged individual. In C. J. Golden, S. S. Alcaperres, F. Strider, M. A. Strider, & B. Graber (Eds.), Applied techniques in behavioral medicine. New York: Grone & Stratton. Hynd, G. W., &Obrzut, J. E. (Eds.). (1981). Neuropsychological assessment and the school-aged child: Issues and procedures. New York: Grone & Stratton. Ince, L. P. (1976). Behavior modification in rehabilitation mediums. Springfield, IL: Thomas. Kazdin, A. E. (1978). The application of operant techniques in treatment, rehabilitation, and education. InS. L. Garfield & A. E. Bergin (Eds.), Handbook ofpsychotherapy and behavior change (2nd ed.). New York: Wiley. Klonoff, A., Crockett, D. F., & Clark, G. (1984). Head trauma in children. In R. Tarter & G. Goldstein (Eds.), Advances in clinical neuropsychology. New York: Plenum Press. Lewinsohn, P. M., Danaher, B. G., & Kikel, S. (1977). Visual imagery as a mnemonic aid for brain-damaged persons. Journal of Consulting and Clinical Psychology, 45, 717-723. Luria, A. R. (1963). Restoration of function after brain injury. New York: Macmillan. Luria, A. R. (1966). Higher conical function in man (B. Haigh, Trans.). New York: Basic Books.
533
MacLean, P. D. (1973). On the evolution of three mentalities. Toronto: University of Toronto Press. MacLean, P. D., & Delgado. J. R. (1953). Electrical and chemical stimulation of frontotemporal portion of limbic system in the waking animal. Electroencephalography and Clinical Neurophysiology, 5, 91-100. Mahoney, M. J. (1974). Cognition and behavior modification. Cambridge, MA: Ballinger. Marr, M. J. (1984). Conceptual approaches and issues. Journal of the Experimental Analysis of behavior, 42, 353-362. Mattis, S., French, J. H., & Rabins, T. (1975). Dyslexia in children and adults: Three independent neuropsychological syndromes. Developmental Medicine and Child Neurology, 17, 15Q-163. Meichenbaum, D. H. (1977). Cognitive behavior modification. New York: Plenum Press. Meier, M. J. (1974). Some challenges for clinical neuropsychology .In R. M. Reitan & L.A. Davison (Eds.), Clinical neuropsychology: Current status and application. New York: Wiley. Miller, E. (1984). Recovery and management of neuropsychological impairment. New York: Wiley. Muir, K., & Milan, M. (1982). Parent reinforcement for child achievement: The use of a lottery to maximize parent training effects. Journal ofApplied Behavioral Analysis, 15(3), 455460. Nelson, R. D. (1983). Behavioral assessment: Past, present, and future. Behavioral Assessment, 5, 195-206. Pirozzolo, F. J. (1981). Language and brain: Neuropsychological aspects of developmental reading disability. School Psychology Review, 10(3), 350-355. Reitan, R. M., & Davison, L.A. (Eds.). (1974). Clinical neuropsychology: Current status and applications. New York: Wiley. Reynolds, C. R. (1981a). The neuropsychological basis of intelligence. In G. W. Hynd & J. E. Obrzut (Eds.), Neuropsychological assessment and the school-aged child. New York: Grone & Stratton. Reynolds, C. R. ( 1981 b). Neuropsychological assessment and the habilitation of learning: Considerations in the search for aptitude and treatment interaction. School Psychology Review, 10, 343-349 Rimm, D. C., & Masters, J. R. (1978). Behavior therapy. New York: Academic Press. Salzinger, L., Feldman, R. D., & Portnoy, S. (1970). Training parents of brain-injured children in the use of operant conditioning procedures. Behavior Therapy, 1, 4-32. Schneider, M., & Robin, H. (1976). The turtle technique: A method for self-control of impulsive impulsive behavior. In J.D. Krumbolts & C. E. Thorenson (Eds.), Counseling methods. New York: Rinehart & Winston. Skinner, B. F. (1938). The behavior of organi:rms. New York: Appleton-Century-Crofts. Skinner, B. F. (1981). How to discover what you have to say-A talk to students. Behavior Analyst, 4, 1-7. Struss, D. T., & Benson, D. F. (1984). Neuropsychological studies of the frontal lobes. Psychological Bulletin, 95(1 ), 33-38. Watson, J. B. (1913). Psychology from the standpoint of a behaviorist. Psychology Review, 20, 158-177. Wolpe, J. A. (1973). The practice ofbehaviortherapy. Elmsford, NY: Pergamon Press.
29 Coping and Adjustment of Children with Neurological Disorder TIMOTHY B. WHELAN AND MARIE L. WALKER
Introduction The opportunity to test the limits of one's clinical acumen is clearly apparent in the field of clinical child neuropsychology. The explosion of related theory and research, the complexity of case material, and the growing demand for applied expertise continue to challenge us daily. Indeed, the study of developmental brain-behavior relationships can be so intrinsically fascinating, so alluring, that many of us cannot imagine being satisfied in another domain of study. Yet there may be a subtle trap in all of this, the trap of becoming so enthralled with exploring brain-behavior relationships in isolation that we lose sight of tpe total experience of the child. It is the premise of this chapter that not only do children and adolescents who have sustained an insult to the brain face the likelihood of altered brain function and its attendant problems, they must also contend with the effects of neurological disorder in a social context. In other words, like children with chronic illnesses, they may face severe developmental challenges during diagnosis, hospitalization, medical intervention, rehabilitation, schooling, family development, and socialization. As a consequence, there is the risk throughout the course of the disorder for the creation of secondary deforming effects on psychosocial adjustment in addition to the primary cognitive, sensory, and motor changes commonly associated with these disorders. As with the victims of other diseases, accidents, or undesirable life events who are faced with major personal and
TIMOTHY B. WHELAN • Department of Psychiatry, Bay State Medical Center, Springfield, Massachusetts 01199. MARIE L. WALKER • Department of Educational Psychology, University of Texas, Austin, Texas 78712.
developmental crises, the presence of neurological impairment may force children and their families to question their fundamental assumptions and expectations about themselves and their world, and to react to or "cope" with multilevel effects of the disorder. It is therefore apparent that the helping professional concerned with the psychological well-being of children must remain cognizant of the nonbiological systems with which children interface. This chapter is concerned with multiple systems affecting the coping and adjustment of children with neurological disorder. Investigation of this topic often leads to psychological literature that is conceptually relevant but somewhat apart from traditional neurodiagnostically oriented neuropsychology. For instance, social psychologists investigating coping mechanisms have primarily been concerned with adults, and the pediatric psychology literature regarding coping with neurological disorders is not highly developed. However, despite variation in specific targets of study, ''good practice'' in the area of clinical child neuropsychology requires a continual integration of theoretical and applied information from the domains of clinical child, developmental, educational, social, and family psychology. A related point was made by Boll (1985) who described a ''threshold'' movement within the field of neuropsychology toward a more psychological emphasis: In addition to the utilization of behaviors for exquisite neuroanatomical appreciation which represents a continuing and legitimate investigational area, neuropsychologists provide, with increasing sophistication, psychological descriptions designed primarily to help in understanding the whole person, rather than being confined only to the person's neuroanatomy. (p. 474)
As detailed below, mechanisms of coping and adjustment in neurologically impaired children are perhaps best viewed globally from developmental 535
536
CHAPTER 29
and systems-theoretic models of child psychology. of nonbiological systems in the lives of neuAfter a discussion of this conceptual framework, rologically impaired children, however, is virtually general considerations regarding the constructs of mandatory. Children do not exist in isolation from coping and adjustment are presented. Further topics others as adults can choose to do. Rather, they are related to the coping and adjustment of neu- enmeshed in nonself systems to a far greater extent, rologically impaired children are then selectively re- influencing and being influenced by family, peers, viewed and provide a flavor of the complexity of health professionals, and schools. One cannot even related theory and practice. approximate a clear clinical description, however elegant, of a child without reference to relationships between that child and those with whom they are General Systems and Developmental Models bound. An additional dimension of complexity must be A conceptual framework for studying the proadded to the clinical child neuropsychologist's syscesses of coping and adjustment in neurologically tems-theoretic model: the process of development impaired children can be derived from general sysand change. Although the notion that the individual is tems theory, a general science of "wholeness" exan active, developing organism is a fundamental conamining sets of elements standing in interrelationship cept among child development and life-span psychol(von Bertalanffy, 1968). The systems-theoretic ogists, neuropsychological literature has historically model posits that the human organism exists in a neglected this aspect of our functioning (Parsons & hierarchy of systems, ranging from the biological Prigatano, 1978; Smith, 1979). Walter Riese (1977) realm, through cognitive, intrapsychic, and behavcaptured the problem aptly: ioral levels of analysis, to the family and social Overpowered by the ever-increasing intricacies of anaspheres. This is an information flow-through model tomical arrangements, ... yielding in this self-inin which developments at one level theoretically have flicted intellectual emergency to the always threatening ramifications throughout the systems hierarchy. danger of oversimplification, the modern student of Whereas von Bertalanffy has perhaps been most elobrain lesions forgot that every functional disturbance quent in expressing the systems approach, these cenhas its natural or evolutionary history. Whether aftral tenets are not unfamiliar to scientists in general fected by neurosis, psychosis or brain injury, man must and they have appeared in the writings of seminal write the history of his new condition which implies the thinkers in the history of neuropsychology. For inhistory of his whole life. Nobody, healthy or not, can stance, more than 45 years ago Kurt Goldstein ( 1939, escape the law of time and change. (p. 77) 1940) concluded that any particular symptom displayed by a patient could not be easily understood as The particular need for consideration of developmenbeing uniquely the product of a specific lesion or tal issues in child health psychology has been well disease, but instead had to be considered as a man- described by Maddux, Roberts, Sledden, and Wright ifestation of the total organism that behaved as a (1986). This, then, is the model: the child is concepunified whole. tualized on multiple levels standing in interreGeneral systems theory has also been cited in lationship, with the hierarchy of systems set in temsupport of a fundamental reorientation in medical poral motion. It would of course be extravagant to education and practice (Engle, 1977, 1980). Such a shift in thought leads to reconceptualization of ''dis- assert that neuropsychologists can excellently or adeease'' as a biopsychosocial product, and to the study quately conceptualize all of our clients in this fashof disease and medical care as interrelated processes. ion, but the goal of so doing seems worthy. The reliance on such an approach is now particularly evident in the literature on families with illness (Gochman, 1985; Kerns & Curley, 1985; Kerns & Coping and Adjustment Turk, 1985; Leventhal, Leventhal, & Van Nguyen, 1985) and in the field of clinical health psychology In most theoretical models, "coping" is a pro(Millon, Green, & Meagher, 1982). cess that is initiated when an individual perceives or The application of a general systems approach experiences stressful stimuli, such as a change in the to understanding the functioning of neurologically pace of life or a subjective perception of an event as impaired adults seems reasonable and logical: disor- negative or undesirable and beyond the competence ders of biological functioning are likely to affect an of the coping person (Chan, 1977). For example, individual's psychosocial status. The consideration hospitalization, sensorimotor disability, or loss of
COPING AND ADJUSTMENT
cognitive integrity associated with neurological disorder could obviously all be considered untimely, unexpected, and undesirable life changes. Coping presumably leads to adaptation or adjustment to stressful events and perceptions, and successful coping implies successful adjustment. Generally, people cope with daily stressors by responding with habitual and automatic patterns of cognition and behavior (Folkman & Lazarus, 1980). Such coping strategies may involve the cognitive functions of perception, memory, speech, judgment, and reality testing; motor activity; emotional expression; and psychological defenses (Mattsson, 1972). When these customary automatic responses become unavailable, individual attempts at coping will require that old resources be used in new ways. This may be particularly true for children whose increasing cognitive and behavioral abilities continually alter the effectiveness of their previous automatic responses. Coping, then, can be conceptualized as purposeful behavioral and/or intrapsychic activity at either conscious or unconscious levels that serves to ameliorate the experience of stress while facilitating adjustment to stressful stimuli. In relation to coping, the process of "adjustment" allows a return to effective (though not necessarily prior) automatic patterns of behavior, and implies that an individual is functioning effectively. More specifically, for the neurologically impaired child, adjustment includes acceptance and age-appropriate understanding of the disease or handicap; medical compliance; absence of severe psychopathology; age-appropriate interpersonal functioning with family, peers, and in school; and "normal" or age-appropriate personality functioning (Drotar, 1981).
Cautions Although full treatment of the constructs of coping and adjustment is beyond the scope of this chapter, several considerations are in order. First, we have implied that coping is a process and adjustment is an outcome of this process, and this may indeed be the case. However, such a scheme may also be an unfair simplification of the relationship between coping and adjustment. It may also be true, for example, that levels of adjustment influence attempts at coping. For instance, high self-esteem may be an outcome of a child's successful attempts at coping; similarly, it may be that children who already experience high self-esteem will be better able to cope with stressful events.
537
Second, coping is a uniquely individual process and its exposition depends on experiential insight and/or observer inferences; it is not a construct that can be measured directly. Currently, psychosocial science relies on personality and adjustment measures, such as levels of self-esteem, depression, anxiety, and locus of control, among others, to infer both the presence and the effectiveness of the coping process. Cognitive and behavioral components may also be evaluated (Curry & Russ, 1985). Nevertheless, though the processes of coping and adjustment are well understood intuitively, they remain scientifically and empirically vague. Third, it should be noted that coping and adjustment are processes that occur continually; they are not discrete and isolated events. People experience multiple levels of stress simultaneously, from the trials of getting to school or work on time, to fears of being perceived as different or odd, and feelings of unworthiness. Even positive changes in life events, such as promotions, may be experienced as stressful, and the perceived intensity of a stressor at a given time will influence an individual's attempt at coping. Further, what may constitute successful coping behavior at one time may change with age and other intervening variables. Again, this is especially true for children where constant development in their abilities may render previously effective patterns of coping obsolete. When neurologically based disorganization of cognitive and affective functions are superimposed on normal developmental patterns ol organization, the complexity of these processes can be magnified. Finally, coping and adjustment are processes that reflect an interaction between individual and environment. It has long been recognized that people perceive stressful stimuli differently and that individuals will make unique attempts to cope with stress, whether by flight, fight, or inaction, based in part on personality determinants, history with coping experiences, and environmental constraints, including peer and societal expectations (Chan, 1977). The significance of this interaction, especially for children, is reflected in increasing research on the influence of the family on children's abilities to cope with their handicaps.
Societal Influence on Coping and Adjustment A paradoxical dilemma exists when one considers the relationship of coping and neurological disorders. Traditional definitions of coping, including
538
CHAPTER 29
those presented here, suggest that stressors such as neurological deficits are external to the individual and must be adjusted to as unwanted alien agents. This idea originates from the societal doctrine of normalcy, where certain parameters of behavior are acceptable and occurrences outside of these bounds are regarded as deviant. Deviancy is then considered unacceptable, obviating the need for a return (adjustment) to normalcy. This viewpoint can be both unfortunate and not entirely necessary. Children born with a neurological deficit, for example, have always interpreted their worlds through a unique lens, and their "deficits" are a part of their identity as surely as being physically "whole" is a part of most of ours. Often, however, individuals with visible physical differences are considered "deviant" and treated accordingly. The process of "accepting" one's handicap is thus made more difficult when one is continually regarded by peers and society as "different." The problem we are suggesting, then, is that "handicap" is defined normatively and from an "outsider's" perspective (Shontz, 1982). "Insiders," or people who have either always experienced a particular state, such as neurological disorder, or who have come to accept their differences and/ or limitations, do not necessarily view their handicap as something to "overcome" (Massie, 1985). Once an individual has learned to cope with a handicap, the deficit itself no longer remains the focus of attention. Handicapped individuals, like ourselves, must answer the question, how does one achieve satisfaction and happiness in life, given the uniqueness of every individual? Although this question is made no less easy by the presence of a handicapping condition, the burden might be eased if it were not necessary to feel that one had to meet the normative standards of today's society.
Making Meaning of Neurological Disorder "Words, words, words"-Hamlet, Act II,
Scene ii
Whether in the role of consultant or therapist, child neuropsychologists are often called on to provide information to their clients regarding the nature of the neurological disorders. The process of clear and appropriate explanation may be problematic enough with adults. We are reminded of the elderly
gentleman with cerebrovascular disease who had carefully listened to his physician's explanation of regional cerebral blood flow measurement, and who had given his "informed consent" to the procedure. During a later neuropsychological exam he wanted to confirm his understanding of the procedure, and said, "This is something nuclear, right? Like the bomb, right?'' Examples of well-meant yet misjudged attempts to convey the nature of medical problems to children are also present in the literature. Whitt, Dykstra, and Taylor ( 1979) mentioned the potential iatrogenic harm that may come from such casual statements as "the doctor will inject some 'dye' ... "(thus effectively raising the possibility of imminent ''death'') or "epilepsy is excess 'electricity' in the brain" (summoning up parental admonitions regarding wall sockets, shocks, and terminal consequences). Similarly, Perrin and Gerrity (1981) have written on the young pediatric patient's assumption that when the doctor says "there's edema in your belly," the "demon" was sent there to punish him or her for wrongdoing; the notion of a "demonic" brain could evoke even more primitive fears. The pediatric literature advises that health professionals consider the child's level of cognitive development (typically in a Piagetian sense) when conveying information regarding disease processes, medical procedures, and health maintenance (Brodie, 1974; Campbell, 1975; Mechanic, 1964; Neuhauser, Amsterdam, Hines, & Steward, 1978; Palmer & Lewis, 1976). The goals of such consideration include improved regimen compliance, reduced anxiety, and enhanced understanding and acceptance of the condition by the affected child and, if also afforded developmentally appropriate explanations, by healthy peers. According to stage theorists, children functioning at a preoperational level of cognitive development (normally between ages 2 and 7) center on single external events, which are viewed from the child's own perspective without generalization to other situations and without the application of logical operations. Illness prevention and recovery may thus be associated with sets of rigid rules surrounding immediate perceptual experience and concrete action (avoiding the touch of sick friends; staying in bed) (Whittet a/., 1979; Perrin & Gerrity, 1981). Bibace and Walsh (1979) proposed refinements to this general category of cognitive development, as well as to categories described below. For instance, preoperational explanations of disease are divided into the categories of "Phenomenism" (illness caused by single inappropriate, external, and spatially remote
COPINC AND ADJUSTMENT
539
sources) and "Contagion" (reliance on single causes crease in the degree of perceived control the child has of contagion transmitted through mere proximity and over illness, with a concomitant decline in the sense of personal vulnerability. For example, older chilmanifest in a single symptom). At the stage of concrete operations (normally dren in the Internalization phase may believe there between ages 6 or 7 and 11 or 12), the child's thought are things they can do to maintain bodily health, becomes less egocentric and perceptually bound, and whereas young children in the Phenomenistic phase reasoning becomes more logical. Specification of re- may believe themselves to be vulnerable to disease lationships among events or objects, categorical clas- causes that are spatially remote and uncontrollable. sification, and transformation comprehension be- These points are elaborated upon by Maddux et al. come possible. Illnesses at this time may be defined (1986) in an article on developmental issues in child by the child as a set of multiple concrete symptoms, health behavior that focuses on prevention of illness and are often believed to be caused primarily by and injury, and on health promotion. It would be germs. From such a perspective, diseases may im- erroneous, however, to suggest that feelings of conpinge on the body unless sick people are avoided, and trol and reduced vulnerability are inevitable accomcures may consist of passively allowing medicines to paniments of older age or later stages of cognitive act on the body. A clear appreciation of the self- development. Once again, children's conceptualizahealing aspects of bodily functioning is presumably tions of their neurological disorders are likely to lacking in the concrete operational child. Unlike the evolve over time, with general stages of cognition preoperational child, the concrete operational child interacting with particular informational content and can more clearly distinguish between internal and numerous other cognitive and affective variables. The suitable selection of "words" of explanaexternal events, though the focus remains on the latter. Subdivision of this stage by Bibace and Walsh tion and due consideration of the nature of the child's includes "Contamination," in which there is recog- beliefs about disease may still be insufficient to ennition of multiple disease symptoms caused by con- able children to understand neurological disorders, if crete sources such as germs, dirt, or bad behavior. only because neuropathological processes are so The "Internalization" subdivision of concrete oper- often without visible referents. The use of metaphor, ational thought is characterized by the ways in which perhaps aided by drawings, to provide appropriate illnesses are internalized: swallowing or inhaling explanations of medical events may be particularly germs or other contaminants. The body's own re- beneficial for children who have not yet attained forcuperative powers become recognized, and rever- mal operational thought and/ or who are not likely to sibility (the sick person can become well, and vice have a sense of the pathophysiology of the unseen nervous system. For example, in the case of seizure versa) is characteristic. With the emergence of formal operations in late disorders the analogy of a telephone system has been childhood and early adolescence, illness may be con- suggested (Whittet al., 1979). In condensed form, a ceptualized as having complex, interrelated, and discussion with the child might refer to the notion that multiple causes that affect multiple internal systems the brain is like a telephone that sends messages to all and result in multiple external symptoms. Bibace and parts of the body, and just like a telephone, the brain Walsh decompose the explanation of illness in the sometimes gets a "wrong number" by sending mesformal operational stage into two subdivisions: sages to the _ _ (substitute relevant perceptual "Physiological" and "Psychophysiological." In cues, perhaps those related to the aura). In some the former, illness is defined by the child in terms of cases it can also be pointed out that just like a teleinternal organs and structures whose malfunctions phone after a wrong number, the brain works fine are manifest in multiple external symptoms and there again. is a clear departure from previous reliance on conThe same authors have also provided metaphors cretely perceptible reality. In the psychophysiologi- for other neurological conditions. For instance, the cal category, psychological events are included as body can be analogously described as a large city disease symptoms as well as causes of internal dys- made up of many people (cells) with important jobs function. The etiologies of a headache, for instance, (e.g., telephone cells, garbageman cells, doctor may at this developmental point include too much cells, carpenter cells, police cells), and cancer cells worry. may be described as "outlaws" in the system. TreatBased on this six-stage developmental se- ment may then be presented as a means of helping the quence, Bibace and Walsh contend that not only do body's police and medical forces to establish "law conceptions.of illness shift in characteristic ways, but and order.'' Built-up pressure in a garden hose with a that there is also a corresponding developmental in- blocked outlet may serve as a metaphor for children
540
CHAPTER 29
with hydrocephalus, and the swelling and potential bursting of a balloon may foster a better understanding of aneurysms (though we personally find this last example too likely to result in fears of imminent catastrophe to be used with most children). In considering the process of aiding children and their families to better comprehend their neurological or other medical disorder, it should be remembered that the sophistication of children's concepts of illness may differ from their concepts involving different content. Perrin and Gerrity ( 1981) reported that in a sample of normal children, illness-causation concepts (e.g., "How do children get sick?") were slower to develop than concepts to explain physical causality (e.g., "Why does night come?''). Moreover, older children may indeed be able to provide more sophisticated explanations of illness than younger children, but in a conditional sense: younger children with a history of poorer health have the least sophisticated concepts, whereas older healthy children may have less sophisticated concepts than peers of the same age who have been ill more often (Campbell, 1975). In addition, the value of providing illness explanations to healthy peers in order to facilitate acceptance of children with chronic illness has been questioned (Potter & Roberts, 1984). As expected, when groups of healthy preoperational and concrete operational children were provided with either symptom descriptions or metaphorical explanations of diabetes and epilepsy, those receiving the analogous explanations demonstrated significantly more general comprehension of the illnesses, and perceptions of personal vulnerability were reduced. However, these illness explanations did not significantly facilitate ratings of acceptance of a hypothetical child with these diseases. The presence of disease conditions may also alter the normal pattern and pace of cognitive development (Mearig, 1985). Obviously this could be the case in those with brain dysfunctions that change cognitive integrity, and it may also occur in those with chronic illnesses whose intellectual functioning is relatively intact. Myers-Vando, Steward, Folkins, and Hines (1979) reported that although children with congenital heart disease manifest lower levels of cognitive performance on conservation tasks compared with healthy peers (presumably because of the disruptive ''intrusive stress'' of the illness), some of the ill children were capable of thinking formally in the content domain of illness causality, possibly because of the greater affective salience of the topic or greater opportunities for direct education and experience with illness. On the other hand, Carandang, Folkins, Hines, and Steward (1979) reported that
healthy siblings of diabetic children failed to perform at the expected cognitive level in conceptualizing illness causality and treatment when compared with children without ill siblings matched on demographic variables and measures of Piagetian cognitive development. In conclusion, conflicting data in the literature suggest that a child's cognitive understanding of a neurological disorder is not likely to be entirely predictable simply on the basis of their age or measures of their level of intellectual/ cognitive development in nonillness content domains. Attempts should be made to integrate such information with their historical experience with the disorder.
"Facts are the enemy of truth."-Man of La Mancha Quite apart from the strictly cognitive aspects of comprehension, a child's construction of the personal "meaning" of neurological disorder, and thus reactions to it, are likely to involve processes that blend cognition and affect, and that incorporate both past and current experience. Perhaps a clinical anecdote can illustrate this point. A large and powerful adolescent boy with a vague history of seizures entered a children's psychiatric hospital with a diagnosis of paranoid schizophrenia. Precipitating the hospitalization were social isolation, paranoid ideation, verbal threatening, and dangerous behaviors such as jumping out of trees and leaping before slowmoving cars. During the course of therapy, two primary themes emerged sequentially. First, he believed his seizures to be a pervasive and primitive loss of bodily and ''self' control, and that during these episodes he might unknowingly and unwillingly kill the small children for whom he frequently cared at home. Later, a history of physical abuse by the father was revealed. After one incident during which the boy secretly wished his father dead, the father promptly did die of a cardiac arrest. The boy was simultaneously overwhelmed by a sense of omnipotencehe believed he had actually caused his father's death-and by his perception of an organically based lack of control necessary to prevent harming those he most loved in the future. He projected great menace onto his environment, and then attempted to prove that he himself could not be killed by engaging in (but surviving) potentially injurious behaviors. His behavioral symptoms, therefore, seemed indicative of dynamic issues influenced by the objective and subjective realities of seizures and past experience. At
COPING AND ADJUSTMENT
issue, then, are the interactions between the ways in which children and adolescents react to and understand neurological disorder, with the ultimate targets of interest being their construction of "meaning," and daily coping and adjustment.
Defense or Coping? If it is assumed that the onset or diagnosis of neurological disorder represents a threat not only to cerebral integrity but also to the child's "self'' or "ego" in a fundamental manner, then it can be assumed that attempts will be made to control, contain, or minimize that threat. Such actions often lead to distortion, illusion, or self-deception (inaccurate reality testing), and may thus be considered classical defense mechanisms; they may thus seem contrary to mental health. Yet there can be a psychologically positive side to these actions as well, and a number of early papers anticipated current trends in social psychological research in this area. Goldstein (1952) distinguished between "protective" and "defense" mechanisms, suggesting that although both may be employed to protect one from fear and anxiety, the former may arise in a neurologically impaired individual from an inability to function in a shifting environment, whereas the latter may develop in response to psychodynamic conflict. Kroeber (1964) categorized and paired ego defense mechanisms (e.g., isolation, projection, repression) with parallel coping mechanisms (e.g., objectivity, empathy, suppression). Both coping and defense mechanisms may be rooted in common attempts to deal with painful reality, though defenses would be cast in more negative terms reflecting relatively poor adaptability, whereas coping mechanisms may represent active, flexible, and effective attempts to deal with conflict. For instance, if an early adolescent girl hospitalized fot diagnostic tests is playing with dolls, she may be employing mechanisms of time reversal by recapturing experiences, feelings, and ideas of the past. The behavior is not necessarily indicative of the defense mechanism of ''regression'' (i.e., age-inappropriate behavior to avoid responsibility, aggression, or unpleasant demands), but rather of the analogous and healthy coping mechanism of "playfulness" (utilizing feelings and ideas from past experience that are not directly ordered by the immediate elements of the situation). Similarly, the 9-year-old girl with little manifest anxiety during a neuropsychological evaluation on the day prior to surgery for an enormous left frontal tumor was perhaps not refusing to face painful thoughts, percepts, or feelings in the pathological sense of "denial" (or exhibiting frontal lobe signs, as the evaluation itself indicated). Instead she may
541
have been able to recognize and then set aside disturbing thoughts and feelings in order to concentrate on tasks at hand. The point, then, is that some of the behaviors and thoughts of those facing extraordinary levels of disruption in their lives may not be as psychopathological as they might superficially appear. Our evaluation of their actions to contend with severe stress needs to consider the degree (focal or pervasive, flexible or rigid, transient or chronic) of distortion as well as the temporal relationship between crisis moments and defense quality. The role of denial in the coping process has perhaps been most clearly explicated. As Lazarus (1983) suggested, the paradox of self-deception being both adaptationally sound and psychopathological may be resolved by asking the more sophisticated question: "What kinds of self-deceptions are damaging or constructive, and under what conditions?" Lazarus initially distinguishes between classical "denial,'' e.g., the negation of some internal impulse, feeling, or thought, or of an external reality, and "avoidance" or plain "ignorance" of threatening events. He then describes a family of denials. Partial denial, or the temporary and tentative suspension of belief, often takes place among the seriously ill in the context of reassuring social relationships with concerned friends, family, or health care providers. Such a situation is quite common among healthy young children who easily suspend the reality of the moment, particularly when that reality is unpleasant. In addition, Lazarus recommends that psychologists shift their emphasis from considering denial and other coping mechanisms as static states of mind to recognizing them as ongoing process that are often not fixed or consolidated defense mechanisms and that depend on both internal events and the social context. Perhaps most relevant are some of the conclusions Lazarus reaches on the costs and benefits of denial. If direct action to change the relationship between person and environment is adaptationally necessary, denial and subsequent inactivity will be destructive. On the other hand, when direct action is irrelevant to the outcome, denial may reduce distress and afford the individual the possibility for good morale and hope. Note that this position to some extent contrasts with many of the cognitive or rational treatment approaches employed with neurologically impaired clients. An additional time-related principle is that denial may be beneficial early in disease or immediately after severe injury when individuals are actually unable to participate in their own care. Later on, during extended treatment, rehabilitation, or education, it may be more important
542
CHAPTER 29
to contend directly with the insult and to struggle in a problem-focused manner. As an aside, it should be clear that reference is being made here to secondary reactions to neurological events. The existence of neurologically based forms of inaccurate perception and reality testing is not ''denied,'' nor is their significance in case management diminished.
Perception of Competence. Forms of denial may be related to other cognitive and conative attempts to cope with neurological insult. Duchenne muscular dystrophy (DMD) is a neuromuscular disease beginning in early childhood and resulting in relentlessly progressive muscle wasting and weakness, and eventually in death by late adolescence or early adulthood. In part, an understanding of the psychological functioning of children with DMD may be derived from their performances on intellectual and neuropsychological measures (Dubowitz, 1977; Karagan, 1979; Knights, Hinton, & Drader, 1973); however, the literature in this area remains conflicting (Mearig, 1979; Sollee, Latham, Kindlon, & Bresnan, 1985; Whelan, 1987). While we are continuing to research the intricacies of brain-behavior relationships in this population, we are also exploring other aspects of their psychosocial functioning. In this context, one motivational variable, perceived control of events, appears to affect a wide variety of psychological conditions. Indeed, perceptions of personal control, especially inaccurate perceptions, have been seen as central components of problems ranging from depression to paranoia to underachievement (Weisz & Stipek, 1982). Although the conceptualization and measurement of the control dimension have been approached from the perspectives of social learning and attribution theories, the theory and concepts of intrinsic motivation are also important. Competence motivation theoty assumes that humans naturally strive for effective interactions with their environments. Successful mastery of a problem produces pleasurable feelings of efficacy or competence, which, in tum, reinforce and lead the individual to seek out and attempt to master additional tasks (Stipek & Weisz, 1981). Harter (1978) claims that in order for children to experience a feeling of efficacy, they must perceive themselves as responsible for their successful performance. Moreover, she reasons, failures perceived to be caused by a lack of competence or selfworth can lead to anxiety in mastery situations and thus decrease the child's mastery motivation. Children's expectancies and perceptions of efficacy may consequently be particularly important to consider
because they may determine whether coping behavior will be initiated, how much effort will be expended, and how long it will be maintained in the face of obstacles and aversive experiences (Bandura, 1977). Certainly children with DMD experience a particularly acute and reality-based loss of motor control. Their perceptions of motoric, academic, and social competence, and general self-worth have been subjected to empirical investigation (Whelan, 1986). The Perceived Competence Scale for Children (Harter, 1979), which measures perceptions in the four mentioned domains, was administered to 31 boys with DMD. With regard to the central tendencies of the data, mean scores on the scales of cognitive and social competence and on the scale of general self-esteem were approximately at the normative mean. Scores on the scale of physical competence (referring primarily to athletic skills) were about one standard deviation below the mean for normal children (Whelan, 1986). On the surface, these results might suggest that children with DMD maintain relatively accurate perceptions of their own areas of competence and disability, or that the existence of a neuromuscular disorder resulting in motor dysfunction and a reduced sense of efficacy in that domain have not substantially generalized to other measured domains. However, scores on the scale of perceived physical competence were not significantly correlated with any of the neuropsychological measures used in this study, including measures of motor performance. This suggests that perceptions of physical competence or, conversely, of physical disability may vary widely in this population, with little relation to the objective reality of assessed motor performance. That is, some of the mildly physically impaired children may perceive themselves as severely limited, and others with greater actual motor disability may not perceive themselves as so seriously impaired. Other data in this study may contribute to an understanding of the ways in which dystrophic children make sense of their condition. The magnitude of the correlations between scores on the scale of perceived physical competence and those on the scales of general self-worth (0.65) and social competence (0.39) was considerably higher than in the normative population. Together, these data may suggest reasons for the lack of a significant relationship between perceived physical competence and actual motor ability: denial of physical disability in the service of preserving a sense of self-worth may be a prominent coping mechanism in children with neuromuscular disease. An examination of perceptions of competence
COPING AND ADJUSTMENT
in other groups of children with suspected neurological disorder has also proved interesting. For instance, the factor pattern of the Harter scales for a sample of learning-disabled children showed that the physical and social competence factors were retained as in the normal population, although cognitive and self-worth factors did not emerge as discrete entities. Instead, two cognitive-self-worth factors were obtained, the first composed of traitlike descriptions (e.g., being smart, liking yourself as a person) and the second composed of concrete and behavioral items (e.g., feeling it is easy to understand what one reads, thinking the way one does things is fine). Thus, the learning-disabled child's sense of selfworth seems directly tied to scholastic competence (Harter, 1985). Harter recommends that we treat selfconcept as neither epiphenomenal nor as a static construct and concludes that "we cannot simply treat all children with intellectual deficits as a homogeneous group since clearly there are quite different processes influencing the structure and content of their selfperceptions." With necessary modifications, the measurement of domain-specific perceptions of competence and global self-worth in the neurologically impaired population of children may yield important data in the future. For instance, we are interested in determining if the factor structure described by Harter for "learning-disabled" children is truly representative for all those bearing that gross label. Certainly many investigators have recognized that learning-disabled children may frequently have difficulty recognizing and interpreting social cues (Maheady & Maitland, 1982). Based on the subtyping literature (e.g., Rourke, 1985), it seems quite possible that some learning-dis~bled children maintain accurate perceptions of their competencies and areas of disability, whereas others do not. Interventions with children who are accurately perceiving their abilities may consequently differ from those with children who are not. Moreover, the assessment of self-evaluative processes may be critically important to consider in the population of neurologically impaired children: if these processes are more amenable to change than structurally based abilities per se, then school and other performances might be indirectly enhanced through alternative interventions.
Attributions. If by definition the word "victim" applies to "anyone who suffers as the result of ruthless design or incidentally or accidentally,'' then the term may be broadly invoked in the context of various life crises, whether accidents, crimes, or diseases (Janoff-Bulman & Frieze, 1983). Because neurologically impaired children surely suffer physical
543
and/or psychological alteration, they may justifiably be considered "victims" in this sense. Even the terms commonly associated with neurological disease or injury reflect this theme: cerebral "trauma" with "loss" of consciousness, brain "insult," vascular "accident." Considerable research on the personal and social consequences of victimization has been conducted by social psychologists, and although there are few available reports concerning those with neurological disorder, the findings are generically relevant. One relatively well-developed domain of research on coping with victimization concerns attributions of causality of undesirable events. In part, the impetus for these investigations came from refinements of learned helplessness theory that were heav. ily based on attribution theory (Wortman, 1983). As part of the learned helplessness reformulation (Abramson, Seligman, & Teasdale, 1978), critical questions on the nature of coping with adverse circumstances shifted from the undesirable events themselves to individuals' interpretations of the events. For example, the type, intensity, and duration of a victim's coping responses to serious accidents may depend less on the precise physiological deficits and more on cognitions regarding the cause of the accident. One of the most relevant studies of this kind examined the relation between the attributions of causality made by adult accident victims with paralysis due to severe spinal cord injury and their subsequent coping patterns (Janoff-Bulman & Wortman, 1977). The findings suggested that those who tended to blame themselves for the accident were rated by medical and rehabilitation staff as coping better than those who blamed others and who felt the accident could have been avoided. Indeed, many respondents (e.g., passengers in cars, people accidentally shot) seemed to attribute more blame to themselves than might seem objectively reasonable. The authors interpreted the findings as reflecting attempts on the part of victims to gain some control over their situations, for blaming oneself may be preferable to the conclusion that random harmful events may occur in a meaningless, chaotic world. The results of the attribution literature concerning victims may be important in the field of clinical child neuropsychology because these and other forms of cognitive distortions may partially determine the quality of coping attempts. Moreover, "real world" findings may be counterintuitive at first glance; many psychologists might not consider, from an outsider's position, self-blame to be particularly adaptive or predictive of good progress in a rehabilitation program. The clinical utility of these forms of cognitions
544
CHAPTER 29
remains to be fully investigated, especially with children and especially over the long term. Attribution theory has also been applied to other areas of child psychology: childhood depression and learning disabilities. According to the reformulated learned helplessness model (Abramson et al., 1978), depressed individuals make more internal, stable, and global attributions forfailure and more external, unstable, and specific attributions for success than nondepressed individuals. Recent research has indicated that, like adults, depressed children have a more "depressed" attributional style than nondepressed children (Kaslow, Rehm, & Siegel, 1984; Blumberg & Izard, 1985), and that attributional style can be used to predict depressive symptoms 6 months later (Seligman et al., 1984). With regard to learning difficulties, the attribution· and learned helplessness literatures are also applicable (Thomas, 1979). It has been reported that children who attribute outcome to ability do not work as long or as hard as those who attribute outcome to effort (Dweck, 1975), and those who attribute failure to ability tend to be less persistent on learning tasks (Hiroto & Seligman, 1975). Diener and Dweck (1978) indicated that "helpless" children attribute failure to lack of ability, and nonhelpless children focused instead on self-monitoring and self-instructions. Compared with average and good readers, poor readers have been found to take less personal responsibility for success, and when they did make internal attributions for success, they were more likely to make effort rather than ability attributions (Butkowsky & Willows, 1980). The potential importance of research in this area is that interventions designed to alter attributional patterns (e.g., to shift attributions for failure from insufficient ability to insufficient effort) may result in improved academic performance (e.g., increased academic task persistence and achievement) (Dweck, 1975; Fowler & Peterson, 1981; Schunk, 1983). Taken together, these lines of theory and results suggest a convergence of information. Attribution patterns affect coping in affective and cognitive domains, and they seem important in adjusting to both acute insults or accidents and long-standing developmental difficulties. Equally importantly, it is possible to modify children's attribution styles through relatively brief interventions. Future research regarding the development and alteration of attributions among children with neurological disorder may consequently prove worthwhile. It may be important, for instance, to investigate not only a child's cognitive understanding of the ''facts'' of the disorder, but also to explore their perceptions of who or what is respon-
sible for the situation, why it happened to them, and to what they attribute their present successes and failures.
Issues in Psychotherapy The neuropsychological literature on professional training models (e.g., Meier, 1981) and intervention procedures (e.g., Edelstein & Couture, 1984; Miller, 1984; Trexler, 1982) generally reflects the position that intervention with neurologically disordered individuals is most commonly cognitive-behavioral in nature. Given the forms of neurological signs and symptoms, such procedures are often warranted, efficient, and effective. In addition, however, some of the challenges to children's mental health described in this chapter might also be addressed within the context of a psychotherapeutic relationship. Moreover, there may be acute (hospitalization for diagnosis or surgery) and chronic (controlled epilepsy, minor head injury, learning disorder, neuromuscular disease) situations in which there are no major behavioral difficulties but in which clients may benefit from, and psychologists may desire, a somewhat different style of intervention. A number of general considerations to be kept in mind by the therapist have been provided by Christ (1978), Geist (1979), and Small (1973). Therapeutic goals may include the provision of nonconfrontal understanding, support, and feedback during periods of confusion, anger, anxiety, and depression. Strengthening of reality testing, learning to select areas of success and to avoid those of failure, and the improvement of relationships with others may also be appropriate targets. Traditional psychotherapeutic emphases and processes may require modification, however. For example, the development of a therapeutic alliance may purposely be extended, allowing greater opportunities for clients to recognize and display their strengths. Primitive and fragile defenses may crumble with mild cognitive or affective stress, leading to catastrophic reactions that seem disproportionate to an outsider's appraisal of the stressor. It may thus be important to concentrate on building a "defensive superstructure," using defenses that are more negotiable than frank denial or projection, such as displacement, rationalization, or intellectualization. The psychologist's concepts of client resistance must be modified in the face of slowly improving or impaired cognitive and integrative capacities, and the inability to recall may obviously reflect faulty memory and not repression of conflict. Finally, it should be remembered that those with neu-
COPING AND ADJUSTMENT
rological disorder do not "work through" a permanent disability as with a neurotic problem, nor do they ··get over it'' as with some normal development hurdles; instead they must continually adjust to the dynamic nature of the disability itself and to its consequences at various levels in the systems hierarchy. For example, realistic limitations in adaptive abilities may prevent the adolescent from taking steps of autonomous action at the same age as most others. Indeed, true termination from therapy may not be desirable, and the option to return at developmentally stressful times may be a sensible alternative. There is another therapeutic issue that deserves comment in order to provoke additional thought or research. A variety of sources suggest that it is important to instill a sense of realistic "hope" in neurologically impaired clients. Travis (1976) recommended that those caring for children and adolescents with progressive muscular dystrophy establish a ''contextforsecurity and an avenue for hope.'' Waddell (1983) discussed the hope that medical and familial people consign to children with life-threatening illnesses, and others considered the role of hope in the process of rehabilitation (Boone, Roessler, & Cooper, 1978; Heinemann, Geist, & Magiera, 1983) and psychotherapy (Erickson, Post, & Paige, 1975; Frank, 1968; Green, 1977; Smith, 1983). Although "hope" is a term used frequently in everyday conversation, and although casual introspection suggests it is a pervasive human construct, there is very little related psychological research. Classical literature provides some insight into the concept. Hope was one of the evils contained within Pandora's box, and indeed, the Greeks viewed hope as an illusion and as mankind's curse because fate was seen as unchangeable. Such sentiment is reflected in lines from Antigone: "We are of the tribe that asks questions, and asks them to the bitter end . . . we are of the tribe that hates your filthy hope, your docile, female hope; hope your whore." On the other hand, the Judea-Christian message is essentially one of hope, and in various cultures the symbol now written in most medical charts for "female" has meant eros, fertility, and hope (Menninger, 1963). Perhaps because of the religious nature of historic tradition and because hope is a difficult construct to operationalize, psychologists may have left the study of hope to theologians and philosophers, and concentrated instead on hopelessness. Still, hope may rightly be classified as a coping phenomenon incorporating a future orientation, optimistic affect, expectant cognition, response to external stress, and resultant motivation (Petiet, 1983). Although multi-
545
dimensional, and although the cognitive and developmental prerequisites of hope remain to be specified for children, the idea that hope is a desirable state during medical recovery and rehabilitation has been investigated with adults (Boone et al., 1978; Brody, 1981; Dubree & Voge'pohl, 1980; Heinemann eta!., 1983; Perley, Winget, & Placci, 1971), and could be explored with chronically ill children.
Psychosocial Adjustment Much ambiguity surrounds the issue of whether chronically ill children, including those with neuropsychological deficits, are maladjusted when compared to their "normal" peers. Findings are contradictory, and diverse methodologies make comparisons between studies difficult. It is generally accepted that children with chronic illnesses are at one and one half to three times greater risk for behavioral, social, and psychological maladjustment than healthy peers (Perrin, 1986; Pless, 1984). Rutter, Graham, and Yule (1970a) reported that the occurrence of psychiatric disorders among the general population of children and adolescents was 6.6%; for children with nonneurological chronic disease, 11.6%; those with epilepsy and no other pathology, 37.5%; and children and adolescents with epilepsy associated with organic brain disease, 58.3%. Professionals and individuals involved with neurologically impaired children should not be misled, however, by the temptation of such figures. For any particular child, the presence of neurological disorder does not necessarily imply lowered psychosocial adjustment. Individual reactions to disability are diverse, and specific disabilities have not been found to be related to specific personality types (Bronheim & Jacobstein, 1984; O'Dougherty, 1983; Roessler & Bolton, 1978).
Psychiatric Symptomatology The range of psychiatric symptoms displayed by children with neuropsychological disorders is similar to the behaviors of their nonhandicapped peers. Results from large-scale epidemiological studies of children with chronic illnesses suggest that these children experience lower academic achievement, greater absenteeism and truancy, and increased behavioral difficulties, nervousness, and aggression (Pless & Roghmann, 1971; Rutter, Tizard, & Whitmore, 1970). In addition, emotional dependence, poor social adjustment, low self-esteem, depression,
546
CHAPTER 29
anxiety, difficulties in sexual adjustment, embarrassment, regression, poor body image, excessive shyness, lifelong feelings of failure and inadequacy, immaturity, exaggerated self-consciousness, shame, and fearfulness have all been used to describe the various experiences of children with spina bifida, epilepsy, muscular dystrophy, cerebral palsy, cancer, and closed head injury. A word of caution is necessary. Although it is true that discrete symptoms or symptom clusters may be manifest by individual children, it should not be construed that this laundry list of psychiatric symptoms uniformly affects all children with neurological handicaps. All children, including those at high risk for developing psychiatric sequelae, will be individually influenced both by the neurophysiological constraints of the disease and by events external to the presence of disease, such as premorbid coping style, family support, and social reaction to disease presentation. For example, social adjustment may be affected when children with spina bifida who are incontinent of bladder and bowel are avoided or teased by their peers because of their ''outhouse syndrome" of smells (Bronheim & Jacobstein, 1984; Shurtleff, 1980). Similarly, the social stigma of epilepsy can increase embarrassment, feelings of shame, and a vigilant need for secrecy for some epileptic children (O'Dougherty, 1983). Depression, which is experienced by some children in all disease categories, may be exacerbated in muscular dystrophy around the time the child becomes wheelchair bound and the relentless nature of the disease becomes less deniable (Lindemann & Stranger, 1981; Pierpont, LeRoy, & Baldfinger, 1984). Expression of psychopathology, then, should be considered in context.
Self-Concept and Self-Esteem
concept and self-esteem are intrinsic to the experience of chronic illness (Geist, 1979). Christ (1978), for example, suggested that neurologically impaired children in psychotherapy view themselves as different, weird, or defective, at least from the time peer comparisons are first made in grade school or preschool. In general, greater agreement exists that children with neurological involvement or deficits are at increased risk for poorer self-concept and lowered self-esteem than their healthy peers (Lindemann & Stranger, 1981; Rutter et al., 1970). Although it is not yet clear precisely why these children may be at increased risk for poorer adjustment, perhaps an understanding can be found in the nature of the relationship of neurological deficit to the development of self-identity. The key question here may be, how does altered brain integrity affect the development of self-concept and self-esteem? For example, administration of a standard measure of perceived competence to a group of educable mentally retarded children suggested that these children did not make the same categorical distinctions of selfcompetence and general self-worth as children in the standardization samples evidenced (Harter, 1982). At one extreme such a question implies that, because of physiological limitations, some children do not develop self-concepts in the same manner that their normal peers do, due perhaps to a physiologically based lack of or unique processing of information. This notion is reflected by parents and teachers who are unsure of how much to expect from their handicapped child and wonder whether the child's behaviors reflect biological limitatioos. Yet, although brain integrity may indeed affect formation of self-concept, the lack of findings correlating one specific emotional or social pattern of behavior with a specific disease or deficit suggests that the relationship between brain functioning and self-concept is complex, mediated by environmental and biological variables, and cannot be subjected to unqualified reductionism.
As empirical studies on the effects of pathoneurological involvement on children's self-concept and self-esteem are sparse, the literature on chronically ill children suggests an equivocal re- Socialization sponse to this issue. Many studies offer findings of Although undoubtedly some people are arrantly lowered self-esteem and poorer self-concepts (e.g., Lineberger, Hernandez, & Brantley, 1984; Tro- satisfied living in relative isolation from family, pauer, Franz, & Dilgard, 1970). In contrast, other friends, and community, most of us recognize the researchers (e.g., Kellerman, Zeltzer, Ellenberg, immeasurable importance of our relations with other Dash, & Rigler, 1980; Simmons etal., 1985; Tavor- people. Indeed, it is notable that children who are mina, Kastner, Slater, & Watt, 1976) report no sig- withdrawn or elect not to participate with their peers nificant differences between various groups of chron- are considered by many to be maladjusted or ically ill children and healthy peers. Anecdotal "pathological. " The relative importance of peer interactions inreports often emphatically suggest that impaired self-
COPING AND ADJUSTMENT
creases with age and growing autonomy. For both normal and neurologically impaired children, the peer group has been described as instrumental in providing confirmation or disconfirmation of children's growing sense of competence and self-esteem, meeting dependency needs, a reference point for growing beliefs about sexuality, and a means of role rehearsal where dimensions of cooperative, competitive, and aggressive behaviors can be expressed (Battle, 1984). Additionally, peer groups are seen as a major source of communication and support, conversation and companionship, and fun and socializing for most adolescents (Resnick, 1984). The importance of peer groups may even be greater for handicapped children. For example, adolescent cancer patients have reported that spending time with their friends is of primary importance in their ability to cope (Zeltzer, LeBaron, & Zeltzer, 1984). Based on a study of survivors of childhood cancer, O'Malley, Koocher, Foster, and Slavin (1979) reported that a decrease in the number of social relationships during diagnosis and treatment had a negative impact on subjects' future adjustment. Minde, Hackett, Killou, and Silver (1972) reported that almost 50% of children with cerebral palsy who did not have a nonhandicapped friend were labeled psychiatrically deviant whereas less than 10% of those with nonhandicapped friends were so labeled. This finding is even more striking when one considers evidence suggesting that nonhandicapped children, especially boys, who initiate contact with a handicapped child generally have less social experience, are more isolated, and adhere less to peer values (Battle, 1984). Although diminished interactions with one's peer group can deprive children of valuable pleasure and experience in their preparation for adulthood, for some children with neurological disorders, gaining access to and acceptance by their peer group can be a formidable task. Hospitalization and requisite medical treatments for some diseases take time away from school attendance and peer activities (Zeltzer et al., 1984). At other times, peers' superstititions and misunderstandings about the nature of the disease can result in cruel teasing and unwarranted ostricism, especially when unfounded fears of contagion are involved (Isaacs & McElroy, 1980). Children who experience a loss of mobility may also face social isolation as their opportunities to participate in the normal activities of childhood and adolescence are restricted. In contrast to the external influences that may limit a handicapped child's full participation in peer group activities, for some children, social isolation
547
and withdrawal are means of coping with their disease. Children who are frightened and embarrassed by a loss of control during a seizure, for example, may consciously or unconsciously remove themselves from the influences of peers in attempts to reduce feelings of being different, unattractive, or socially rejected (Ozuna, 1979). Others have noted that wheelchair-bound children and children with progressive muscular weakness can become isolated and withdrawn, relying heavily on fantasy and imagination (O'Dougherty, 1983; Lindemann & Boyd, 1981). Although children with obvious physical limitations can face rejection from peers because of their visible differences, visible handicaps may also at times be addressed and accepted more openly than deficits with few noticeable manifestations. Indeed, children whose disabilities are less obvious or are better controlled may suffer as much or more than their severely disabled counterparts (Hertzig, 1983; Pless, 1984). The "marginal child" may be teased for being slow, clumsy, or different, and often faces the dilemma of trying to "pass" as a normal peer, meeting the expectations of behavior and ability that such normalcy involves, or choosing to separate from the peer group, enduring consequent ridicule and isolation (O'Dougherty, 1983).
Independence and Autonomy Emotional separateness and independence is recognized as a significant goal of childhood and adolescence, and a hallmark of adult adjustment. Although being "special" may be a plausible role for some handicapped people, American society expects disabled individuals to strive maximally toward independence and autonomy (Parsons, 1964). The influences of neurological disorder, however, can run counter to goals of individuation, as is illustrated in this case description quoted by Resnick (1984): While his age cohorts were arguing with parents over the length of their hair, he needed help washing his; while they were resisting doing assigned chores, he was unable to perform any; while they were battling curfew, he needed not only permission, but physical assistance in order to be out. Instead of sharing his peers' increased independence from parents and others, symbolized by mild acting out behaviors, this patient could merely fantasize his acting out, with his illness providing a constant reminder of his chronic dependent status.
Though not all children with neurological handicaps exhibit the same degree of physical limitation,
548
CHAPTER 29
they all share in an increased dependence on parents, medical staff, and sometimes siblings for physical, financial, and emotional support (Zeltzer et al., 1984). Although disorders that demand large amounts of time and care from parents and family might appear to encourage emotional dependence compared to other diseases, the critical issue remains how to foster developmentally appropriate independence and responsibility within the context of a child's neurological deficit. For "normal" children, autonomy invariably increases with age, and parental control usually decreases proportionally. Although the progression toward adulthood may not always be as smooth as many parents and children would prefer, in most cases autonomy and responsible adult action are considered birthrights. For children with neurological disorders, however, both the pathway to autonomy and the children's "right" to eventually assume traditionally adult responsibilities may be questionable (J(opelman, 1985). Physical and mental abilities that have been compromised by presence of disease may in some cases realistically limit a child's ability to assume such adult activities as driving a car or making important decisions regarding medical treatment. With adults, an assumption is made that everyone over a recognized legal age is competent to make decisions for themselves, unless proven otherwise, and implicit in this supposition is the attainment of a certain, unquantified level of maturity. For children with neurological disorders, the issue of emotional maturity becomes inextricably linked with physical disability. The physical and other limitations that some neurological disorders impose incite some parents to become overly protective of their handicapped child. Such overprotection can be detrimental to the child's quest for autonomy and can too easily create an atmosphere that encourages children to remain overly dependent, both emotionally and physically, on others. At one extreme, children may become complacent, passively accepting the ministries of others. In contrast, overly demanding, noncompliant, acting-out, or intentionally guilt-provoking behaviors may represent attempts to separate from parental domination and establish self-responsibility, while also satisfying certain emotional needs (O'Dougherty, 1983). This secondary gain that many children experience from their dependent roles can be reinforced when parents are reluctant to expect or demand independence from their handicapped child (Resnick, 1984). Noncompliance with medical procedures can become a difficult issue when children and adolescents are unable to assert their autonomy in appropriate ways. Similarly, changing needs during illness
can also complicate the process of separation and autonomy. For example, Zeltzer et al. (1984) reported that immediately following diagnosis and during times of disease relapse, adolescents with cancer prefer a more passive, dependent role, being less involved with the management of their disease than parents and physicians wish them to be.
Impact on the Family Nowhere is systems theory perhaps more useful than in investigating the family. As a unit the family is affected by the presence of chronic illness, whether the illness is of neurological origin or not. One is reminded of John Steinbeck's The Pearl, the story of a poor fisherman who found a pearl so inordinate in its beauty and consequences that the lives of the entire village were altered. The birth or diagnosis of a child with neurological disorder is not unlike Steinbeck's description that time had changed and everything hence would be either before the pearl or after the pearl. Although it is perhaps ubiquitous that a neurological disorder will alter the lives of the family and individuals close to the handicapped child, it is also the case that not all families are similarly affected. Some families report being strengthened by the continuing challenge; other families cannot withstand the stress and become dysfunctional or disintegrate. Presently, no direct cause-and-effect occurrences have been identified that would fully explicate or predict the interaction of chronic illness and family dynamics; rather, the influences are mosaic.
Stages of Family Growth Various theorists have proposed different stages of family growth and development, including marriage, childbirth, early child rearing, child schooling and increasing independence, departure of children from the home, and integration of loss as parents adjust to problems associated with being alone and growing older. This model is influenced both by family subsystems and by groups external to the nuclear family, such as extended family, friends, and community. Stage theories depicting special times of stress may inadequately portray the family of a child with a neurological disorder who must deal with burdens unlike those of their "average" counterparts, and additional crisis points have been suggested for families of chronically ill children: when parents ftrSt become aware of the child's handicap; when the child first becomes eligible for special educational ser-
COPING AND ADJUSTMENT
vices; when the child leaves school; and when parents are aging and can no longer assume responsibility for the care oftheir child (Bailey & Simeonsson, 1984). Stage models of family development are useful in that they provide a framework with which to understand family dynamics; however, variability of family structures due to single parenthood, and ethnic, social, and financial differences make generalizations about the consequences of chronic childhood illness, including neurological disorders, on family life a dangerous task at best. For example, it has been suggested that the birth of a handicapped child is more devastating for lower socioeconomic status families than for middle- and upper-class families; yet, little data are currently available on classrelated coping characteristics of parents (Schilling, Schinke, & Kirkham, 1985). Similarly, anecdotal reports suggest the importance a family's ethnic background can have on family coping and adjustment, and on medical compliance (Hobbs, Perrin, & lreys, 1985). Although systematic investigation of socioeconomic-ethnic variables is sparse, such influences cannot be extricated from the daily lives of neurologically impaired children and must not be forgotten in our quest for greater understanding.
Stages of Parental Adjustment The diagnosis of a chronic illness or neurological disorder marks the beginning of a stressful and confusing time for parents. Even when there has been some suspicion of illness, diagnosis represents an immediate confirmation of parents' fears and a removal of hope. Although each parent may not feel each of these emotions, fear, shock, horror, numbness or detachment, relief, helplessness, denial, sadness, anger or rage, anxiety, depression, and guilt are all likely to be experienced at various times (Drotar, Baskiewicz, Irvin, Kennell, & Kfaus, 1975; Hobbs et al., 1985; McCollum, 1981). During the initial period of diagnosis, many parents experience shock and bewilderment and sometimes feel that the situation is unreal, that it must either be a dream or happening to someone else. They may discuss their child as if he or she were a textbook case rather than their own child (McCollum, 1981). Parents will often have many questions, such as: "How will my child's life be affected?" "Is there a cure?'' "Will my child's life be shortened?" "What does the disease do?" "Could I have done something to avoid this?'' Paradoxically, in their emotional turmoil many parents are unable to remember what professionals say and, although forgetting may be an
549
understandable defense against emotional pain, its consequences can be exacerbated when parents view it as a sign of their own inadequacy. Embarrassed, they may tum to friends or books for information, rather than repeatedly question professionals. Some sources of information, although well-intentioned, may be highly inaccurate, and misconceptions about their child's disease can linger for years, at times to the detriment of effective treatment (Whitten, Waugh, & Moore, 1974). During this initial period parents may also refuse to believe the diagnosis and a period of "shopping around" for second medical opinions may ensue. Sadness, anxiety, and grief, from its raging anger and tears to its heavy numbness and pain, frequently occur next as parents begin to fully experience the unjustness of the situation (Blacher, 1984). Parents grieve for the feared loss of their child through death, for the loss of their "normal" child, and for hopes and aspirations for their child that have been relinquished (Mattsson, 1972). The intensity of emotions experienced and the isolating effects of grief can cause some parents to wonder if their reactions are normal. Customary sources of comfort and solace, such as one's spouse, may be unavailable because they too are grieving. Anger can become a prominent emotion and parents may vent their anger at each other, at other healthy children, at hospitals, physicians, and psychologists, at their church or their God, and at times, at their ill child (McCollum, 1981). Worry and anxiety may also increase as both the demands of care and the family's limitations become more evident. The stress and anxiety of this time may be associated with physical illness or symptoms in the parents and can cause parents to fear that they too are sick. Inevitably, their child's illness confronts parents with their own mortality and eventual death (Isaacs & McElroy, 1980). Progression to final stages in this coping model suggests parental acceptance of the child's handicap, an ability to emphasize positive aspects of the situation, and attenuation of the intensity of earlier feelings (Hobbs et al., 1985). The ability to master guilt, fear, and self-accusatory feelings of responsibility has been suggested as critical in determining parents' acceptance of their child's illness or handicap (Mattsson, 1972). Additionally, Mattsson suggested that the awareness of and ability to verbalize feelings indicates that parents are ready to accept the reality of the illness. Parental coping and acceptance are further facilitated through their use of various defense mechanisms, including rationalization; displacement and projection of feelings onto others such as medical professionals; intellectualization, including educat-
550
CHAPTER 29
ing themselves about medical, physiological, and psychological aspects of the disease; identification with other parents of seriously ill children; and denial and isolation of helplessness and anxious feelings, especially during medical crises. The use of any of these defense mechanisms may at times be exasperating to people who are in contact with the ill child's parents, as is most obviously the case with angry and obstreperous parents. It is important to realize, however, that such "defense" mechanisms may be quite appropriate at different times in the course of· the illness. Although useful as a structure for understanding family adjustment, the utility of stage theories is limited in specific applications. The attendant feelings of a parent toward a chronically ill child may ebb and flow chaotically, and even without apparent crisis or problem they may experience many feelings simultaneously. Parents' grief and the need for coping may reoccur as their ill child reaches chronological and developmental milestones (Schilling et al., 1985). Some parents report that, although they may have learned to live with their child's illness, they do not feel they will ever accept it (Hobbs et al., 1985). This sentiment is reflected in the words of one father: The guilt, like all guilt, has both rational and irrational components. . . . The guilt that parents feel for a handicapped child is much greater than anything you could rationally calculate. It has to do somehow with having the feeling that one has imposed upon a child a kind of permanent burden that the kid dido 't deserve, so there's no way that you can be forgiven for this. I have felt all along that the guilt factor, or by any other name the sense of having imposed on David a lifelong burden that he had no reason to expect and had no choice about and dido 't deserve, that that has affected both (my wife) and me a lot.
When asked if he had come to terms with his guilt, this father reported that it wasn't until he was about 50 years old that he could finally ''accept that life had to be the way it was so that you no longer grieve for the way it wasn't or feel a failure because of how your life turned out. " The consequences of unresolved guilt and anguish, or the inability of parents to adequately accept and cope with their child's chronic illness can negatively affect their relationship with their ill child and other relations within the family. At one extreme, parents may reject or severely neglect their disabled child by denying the presence of illness or the need for treatment, or by blaming abandoned careers, financial ruin, and much inconvenience on their ill child (Hobbs et al., 1985). More frequently, prolonged parental overconcern leads to indulgence and
overprotection. Family members become more loving toward the ill child, and normal rules and discipline are suspended. Although allowances need to be made according to the realistic limitations imposed by the disease or illness, changes in family attitude can be confusing and sick children are likely to gain a sense of their own vulnerability through the fears and reactions of their parents and siblings (Mattsson, 1972). Mattsson described four situations that may predispose parents to overprotection or rejection: the child is afflicted with a hereditary disorder found among relatives; the child was unwanted; the child was not expected to live at birth or as an infant; and emotional conflicts around the death of a close relative are aroused by the child's illness.
Parental Differences in Coping Style The coping styles of parents may differ according to sex. Findings suggest that women tend to employ interpersonal and cognitive coping strategies and men more frequently use cognitive coping patterns. Using their own health inventory, McCubbin, McCubbin et al. (1983) factor analyzed the scaled responses of 100 families of children with cystic fibrosis. Mothers' coping efforts were directed at the interpersonal dimensions of family cohesiveness, support, and expressiveness; fathers placed more emphasis on maintaining the family through cooperation and minimizing conflicts in family interactions through the use of rigid rules and procedures. Similar coping profiles were reported by McCubbin, Nevin et al. (1983) in a study of parents of children with cerebral palsy. Such findings are consistent with available research on developmentally disabled children that suggests that parents of handicapped children tend to be more traditional in terms of sex roles than other families. Within traditional families the father's role is most frequently as provider first and parent second, and for mothers the reverse is true (O'Donnell, 1982). In a Colorado statewide survey by Linder and Chitwood (1984), fathers of handicapped infants and preschoolers reported that their time with their handicapped child was limited by job and other family demands, even though they desired to become more involved with their child. Mothers were found to be the primary source of information about their child for fathers, though fathers indicated that newsletters or training in working with their child would be helpful. Additionally, survey replies indicated that fathers were least interested in "someone to talk to about my child'' as a means of information or source
COPING AND ADJUSTMENT
of comfort and solace. Such responses are consistent with findings in other studies, and may be partially comprehensible when one considers that husbands tend to rely on their wives for intimate support, whereas wives report turning to other women and friends for support. Women in general report more dissatisfaction with family life, less freedom and opportunity to develop self interests, worse health, and less positive moods. This is perhaps not surprising, as wives and mothers are called on to balance the needs of their handicapped or ill child, unaffected children, and spouse, with their own needs.
Dyadic Relationships It has already been suggested that the diagnosis of chronic illness or handicap in a young child may contribute to maternal overprotection (Mattsson, 1972). Such overprotection may result in the formation of an intense dyadic relationship, usually between mother and handicapped child, that isolates the dyad from other family interactions (Shapiro, 1983). This relationship then becomes an axis around which other family relations develop, especially other children's resentment of the special closeness between mother and handicapped child. Paradoxically the handicapped child may also develop feelings of being outside the family, participating primarily as an observer who is never fully accepted by other siblings or is fully a part of family life. Although such an intense relationship may represent a mother's conscious or unconscious efforts to atone for the guilt she may feel, its effects on the family can be severe. Psychodynamic theory clearly posits the insult to emergent self-identity and resultant psychopathology in response to prolonged and stage-inappropriate affective symbiosis. Spousal and sibling jealousies can also arise within the family system, and consequent emotional alliances that demarcate the family may actually only represent attempts at emotional connection and survival between members excluded from the dyad. For example, the birth of a chronically ill second child may leave the mother little time for her first child, who soon may exhibit a clear preference for the father. Such alliances can readily exacerbate an already stressful family or marital situation.
Siblings Although it is reasonable to assume that brothers and sisters of children with neurological disorders will be affected by their sibling's illness, little consensus exists regarding what those effects will be.
551
Some reports indicate that siblings of chronically ill children are more likely to have adjustment, behavioral, and academic difficulties (Allan, Townley, & Phelen, 1974; Lavigne & Ryan, 1979). Others suggest that although general mental health may remain stable, social adaptation may be compromised. Still other studies report no significant differences between comparison groups on measures of adjustment or sociability (Drotar, Crawford, & Bush, 1984). Again, although generalizations are to be made with caution as methodologies, patient groups, developmental levels, and comparison groups vary across studies, it may be reasonable to suggest that, like their siblings, brothers and sisters of neurologically impaired children are at increased risk for psychosocial maladjustment. In some families the needs of healthy children can take second place to those of the ill sibling, especially during times of stress and crisis, and throughout the course of the illness parental adjustment and coping styles will directly influence healthy siblings. For example, depleted emotional reserves and lowered ability to communicate may make parents seem unavailable or rejecting. Younger children who are not yet cognitively able to interpret their parents' feelings or understand what is happening with their ill sibling will tend to effect individual interpretations of the family situation. They may feel guilty, or blame themselves for their sibling's illness. Children may also fear they are susceptible to the same fate, and older children may wonder if they are potential genetic carriers (McCollum, 1981). Distribution of labor may change in the family and researchers have suggested that older female siblings perform a disproportionate share of extra chores. In other comparisons, younger male siblings have been reported to be more sensitive to peer's comments about the illness (Hobbs et al., 1985).
Discussion It has been assumed in this chapter that the human organism is a complex web of interaction, with normal and pathological developments taking place at multiple levels, from the biological to the social. Under this assumption, the understanding and significance of neurologically based changes in sensorimotor functioning, in cognitive and executive capacities, or in emotion and behavior are enhanced by placing these alterations in a social and historical context. When neuropsychologists listen to their clients, they may hear expected questions about the
552
CHAPTER 29
brain and the consequences of its disorder. Yet in our experience, these questions do not often end with anatomy and physiology, or with strict brain-behavior relationships per se. Instead, the concerns of clients, both children and adults, extend to attempts to ''make sense'' of their condition and to the ramifications of their neurological disorder in the context of family, school, and social settings, and in terms of past experience and future expectations. Consideration of these temporal and ecological dimensions may thus lead to a richer understanding of the implications of neurological disorder in the lives of children, and may suggest additional directions for assessment and intervention in clinical child psychology. Some of the topics included here have been selected because of their wide mention in the related literature; others were chosen because they reflect questions generated in our own research and practice, and for their utility in provoking additional research. In most instances, the topics purposefully suspend the trend toward reductionism in the social sciences and advocate the application of theory and methods from other domains of psychology to the field of neuropsychology.
References Abramson, L. Y., Seligman, M. E. P., & Teasdale, J.D. (1978). Learned helplessness in humans: Critique and reformulation. Journal of Abnormal Psychology, 87, 49-74. Allan, J., Townley, R., & Phelen, P. {1974). Family response to cystic fibrosis. Australian Pediatrics Journal, 10, 136-146. Bailey, D. B., & Simeonsson, R. J. (1984). Critical issues underlying research and intervention with families of young handicapped children. Journal ofthe Division for Early Childhood, 8, 38-48. Bandura, A. {1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84, 191-215. Battle, C. U. (1984). Disruptions in the socialization of a young, severely handicapped child. In R. P. Marinelli & A. E. De10rto (Eds. ), The psychological and social impact ofphysical disability. Berlin: Springer. Bibace, R., & Walsh, M. E. (1979). Developmental stages in children's conceptions of illness. In G. C. Stone, F. Cohen, & N. E. Adler (Eds.), Health psychology. San Francisco: Jossey-Bass. Blacher, J. (1984). Sequential stages of parental adjustmentto the birth of a child with handicaps: Fact or artifact? Mental Retardation, 22, 359-371. Blumberg, S. H., & Izard, C. E. (1985). Affective and cognitive characteristics of depression in l 0 and ll year old children. Journal ofPersonality and Social Psychology, 49, 194-202. Boll, T. (1985). Developing issues in clinical neuropsychology. Journal of Clinical and Experimental Neuropsychology, 7, 473-485.
Boone, S. E., Roessler, R. T., & Cooper, P. G. (1978). Hope and manifest anxiety: Motivational dynamics of acceptance of disability. Journal of Counseling Psychology, 25, 551-556. Brodie, B. (1974). View of healthy children toward illness. American Journal of Public Health, 64, ll56-ll59. Brody, H. (1981). Hope. Journal of the American Medical Association, 246, 1411-1412. Bronheim, S. M., & Jacobstein, D. M. (1984). Psychosocial assessment in chronic and fatal illness. In P. R. Magrab (Ed.), Psychological and behavioral assessment: Impact on pediatric care. New York: Plenum Press. Butkowsky, I. S., & Willows, D. M. (1980). Cognitive-motivational characteristics of children varying in reading ability: Evidence for learned helplessness in poor readers. JourRDI of Educational Psychology, 72, 408-422. Campbell, J. D. (1975). Illness is a point of view: The development of children's concepts of illness. Child Development, 46, 92-100. Carandang, M. L.A., Folkins, C. H., Hines, P. A., & Steward, M. S. (1979). The role of cognitive leveland sibling illness in children's conceptualizations of illness. American Journal of Orthopsychiatry, 49, 474-481. Chan, K. B. (1977). Individual differences in reactions to stress and their personality and situational determinants: Some implications for community mental health. Social Science and Medicine, II, 89-103. Christ, A. E. (1978). Therapy of the child with true brain damage. American Journal of Orthopsychiatry, 48, 505-515. Curry, S. L., & Russ, S. W. (1985). Identifying coping strategies in children. Journal of Clinical Child Psychology, 14, 6169. Diener, C. 1., & Dweck, C. S. (1978). An analysis of learned helplessness: Effects of aversive stimulation on subsequent performance. Journal of Experimental Psychology: Human Perception and Performance, 1, 411-417. Drotar, D. (1981). Psychological perspectives in chronic childhood illness. Journal of Pediatric Psychology, 6, 211-228. Drotar, D., Baskiewicz, A., Irvin, N., Kennell, J., & Klaus, M. (1975). The adaptation of parents to the birth of an infant with a congenital malformation: A hypothetical model. Pediatrics, 56, 710-717. Drotar, D., Crawford, P., & Bush, M. (1984). The family in context of chronic childhood illness: Implications for psychological intervention. In M. Eisenberg (Ed.), Chronic illness and disability through the life span: Effects on selfandfamily. Berlin: Springer. Dubowitz, V. (1977). Mental retardation in Duchenne muscular dystrophy. In L. P. Rowland (Ed.), Pathogenesis of human muscular dystrophies. Amsterdam: Excerpta Medica. Dubree, M., & Voge'pohl, R. (1980). When hope dies-so might the patient. American Journal of Nursing, 80, 2046-2049. Dweck, C. S. (1975). The role of expectations and attributions in the alleviation of learned helplessness. Journal ofPersonality and Social Psychology, 31, 674-685. Edelstein, B. A., & Couture, E. T. (Eds.). (1984) Behavioral assessment and rehabilitation of the traumatically braindamaged. New York: Plenum Press. Engle, G. (1977). The clinical application of the biopsychosocial model. American Journal of Psychiatry, 137, 535-544.
COPING AND ADJUSTMENT Engle, G. (1980). The need for a new medical model: The challenge for biomedicine. Science, 196. 129-136. Erickson, R. C., Post, R. D., & Paige, A. B. (1975). Hope as a psychiatric variable. Journal of Clinical Psychology, 31. 324-330. Folkman, S., & Lazarus, R. S. (1980). An analysis of coping in a middle-aged community sample. Journal of Health and Social Behavior, 21,219-239. Fowler, J. W., & Peterson, P. L. (1981). Increasing reading persistence and altering attributional style of learned helpless children. Journal of Educational Psychology, 73, 251-260. Frank, J. ( 1968). The role of hope in psychotherapy .International Journal of Psychiatry, 5, 383-395. Geist, R. A. (1979). Onset of chtonic illness in children and adolescents: Psychotherapeutic and consultative intervention. American Journal of Onhopsychiatry, 49, 4-23. Gochman, D. S. (1985). Family detenninants of children's concepts of health and illness. In D. Turk & R. Kerns (Eds.), Health, illness, and families. New York: Wiley. Goldstein, K. (1939). The organism. New York: American Book Co. Goldstein, K. (1940). Human nature in the light of psychopathology. Cambridge, MA: Harvard University Press. Goldstein, K. (1952). The effect of brain damage on personality. Psychiatry, 15, 245-260. Green, M. R. (1977). Anticipation, hope, and despair. Journal of the American Academy of Psychoanalysis, 5, 215-232. Harter, S. (1978). Effectance motivation reconsidered: Toward a developmental model. Human Development, 1, 34-64. Harter, S. (1979). Perceived competence scale for children, Manual. Boulder, CO: University of Denver. Harter, S. (1982). The perceived competence scale for children. Child Development, 53, 87-97. Harter, S. ( 1985). Processes underlying the construction, maintenance, and enhancement of the self-concept in children. In J. Suls & A. Greenwald (Eds.), Psychological perspectives on the se/f(Vol. 3). Hillsdale, NJ: Erlbaurn. Heinemann, A., Geist, C., & Magiera, R. (1983). The role ofhope in spinal cord injury rehabilitation. Paper presented at the 91st annual conference, American Psychological Association, Anaheim, California. Hertzig, M. E. (1983). Temperament and neurological status. In M. Rutter (Ed.), Developmental neuropsychiatry. New York: Guilford Press. Hiroto, D. S., & Seligman, M. E. P. (1975). Generality of learned helplessness in man. Journal of Personality and Social PsychohJgy, 31, 311-327. Hobbs, N., Perrin, J. M., & Ireys, H. T. (1985). Chronically ill children and their families. San Francisco: Jossey-Bass. Isaacs, J., & McElroy, M. R. (1980). Psychosocial aspects of chronic illness in children. Journal of School Health, 50, 318-321. Janoff-Bulman, R., & Frieze, I. H. (1983). A theoretical perspective for understanding reactions to victimization. Journal of Social Issues, 39, 1-17. Janoff-Bulman, R., & Wortman, C. B. (1977). Attributions of blame and coping in the "real world": Severe accident victims react to their lot. Journal ofPersonality and Social Psychology, 35, 351-363.
553
Karagan, N. J. ( 1979). Intellectual functioning in Due benne muscular dystrophy. Psychological Bulletin, 86, 250-259. Kaslow, N. J., Rehm, L. P., & Siegel, A. W. (1984). Socialcognitive and cognitive correlates of depression in children. Journal of Abnormal Child Psychology, 12, 605-620. Kellennan, J., Zeltzer, L., Ellenberg, L., Dash, J., & Rigler, D. ( 1980). Psychological effects of illness in adolescence. Journal of Pediatrics, 97, 126-131. Kerns, R., & Curley, A. (1985). A biopsychosocial approach to illness and the family: Neurological diseases across the life span. In D. Turk & R. Kerns (Eds.), Health, illness, and families. New York: Wiley. Kerns, R., & Turk, D. (1985). Behavioral medicine and the family: Historical perspectives and future directions.ln D. Turk & R. Kerns (Eds.), Health, illness. and families. New York: Wiley. Knights, R. M., Hinton, G., & Drader, D. (1973). Changes in intellectual ability with Duchenne muscular dystrophy. Research Bulletin #8, Department of Psychology, Carleton University, Ottawa, Canada. Kopelman, L. (1985). Paternalism and autonomy in the care of chronically ill children. InN. Hobbs & J. M. Perrin (Eds.), Issues in the care of children with chronic illness. San Francisco: Jossey-Bass. Kroeber, T. C. (1964). The coping functions of ego mechanisms. In R. W. White (Ed.), The study of lives. New York: Atherton. Lavigne, J., & Ryan, M. (1979). Psychological adjustment of siblings of children with chronic illness. Pediatrics, 63, 616-627. Lazarus, R. S. (1983). The costs and benefits of denial. In S. Breznetz (Ed.), Denial of stress. New York: International Universities Press. Leventhal, H., Leventhal, E., & Van Nguyen, T. (1985). Reactions of families to illness: Theoretical models and perspectives. In D. Turk & R. Kerns (Eds.), Health, illness, and families. New York: Wiley. Lindemann, J. E., & Boyd, R. D. (1981). Myelomeningocele. In J. E. Lindemann (Ed.), Psychological and behavioral aspects of physical disability. New York: Plenum Press. Lindemann, J. E., & Stranger, M. E. (1981). Progressive muscle disorders. In J. E. Lindemann (Ed.), PsychohJgical and behavioral aspects of physical disability. New York: Plenum Press. Linder, T. W., & Chitwood, D. G. (1984). The needs offathers of young handicapped children. Journal ofthe Division for Early Childhood, 9, 133-139. Lineberger, H. P., Hernandez, 1. T., & Brantley, H. T. (1984). Self-concept and locus of control in hemophiliacs. International Journal of Psychiatry in Medicine, 14, 243-251. Maddux, J. E., Roberts, M. C., Sledden, E. A., & Wright, L. (1986). Developmental issues in child health psychology. American Psychologist, 41, 25-34. Maheady, L., & Maitland, G. E. (1982). Assessing social perception abilities in learning disabled students. Learning Disabilities Quarterly, 5, 363-370. Massie, R. K. (1985). The constant shadow: Reflections on the life of a chronically ill child. In N. Hobbs & J. Perrin (Eds.), Issues in the care of children with chronic illness. San Francisco: Jossey-Bass.
554
CHAPTER 29
Mattsson, A. (1972). Long-term physical illness in childhood: A challenge to psychosocial adaptation. Pediatrics, 50, 801811. McCollum, A. T. (1981). The chronically ill child: A guide for parents and professionals. New Haven, CT: Yale University Press. McCubbin, H. I., McCubbin, M. A., Patterson, J. M., Cauble, A. E., Wilson, L. R., & Warwick, W. (1983). CHIP-coping health inventory for parents: An assessment of parental coping patterns in the care of the chronically ill child. Journal of Ma"iage and the Family, 45, 359-370. McCubbin, H. I., Nevin, R. S., Cauble, A. E., Larsen, A., Comeau, J., &Patterson, J. M. (1983). Familycopingwithchronic illness: The case of cerebral palsy. In H. I. McCubbin, A. E. Cauble, & J. M. Patterson (Eds.), Family stress, coping, and social support. Springfield, IL: Thomas. Mearig, J.S. (1979). The assessment of intelligence in boys with Duchenne muscular dystrophy. Rehabilitation Literature, 40, 262-274. Mearig, J. S. (1985). Cognitive development of chronically iii children. InN. Hobbs & J. M. Perrin (Eds.), Issues in the care of children with chronic illness. San Francisco: JosseyBass. Mechanic, D. (1964 ). The influence of mothers on their children's health attitudes and behaviors. Pediatrics, 33, 444-453. Meier, M. J. (1981). Education for competency assurance in human neuropsychology: Antecedents, models, and directions. InS. Filskov & T. Boll (Eds.), Handbook of clinical neuropsychology. New York: Wiley. Menninger, K. (1963). The vital balance. New York: Viking Press. Miller, E. (1984). Recovery and management of neuropsychological impairments. New York: Wiley. Millon, T., Green, C., & Meagher, R. (1982). Handbook of clinical health psychology. New York: Plenum Press. Minde, K. K., Hackett, J. D., Killou, D., & Silver, S. (1972). How they grow up: 41 physically handicapped children and their families. American Journal of Psychiatry, 128, 15541560. Myers-Vando, R., Steward, M.S., Folkins, C. H., & Hines, P. (1979). The effects of congenital heart disease on cognitive development, iiiness causality concepts, and vulnerability. American Journal of Orthopsychiatry, 49, 617-625. Neuhauser, C., Amsterdam, B., Hines, P., & Steward, M. (1978). Children's concepts of healing: Cognitive development and locus of control factors. American Journal of Orthopsychiatry, 48. 335-341. O'Donnell, L. (1982). The social worlds of parents. Marriage and Family Review, 5, 9-36. O'Dougherty, M. M. (1983). Counseling the chronically ill child: Psychological impact and intervention. Lexington: Lewis Publishing. O'Malley, J. E., Koocher, G., Foster, D., & Slavin, L. (1979). Psychiatric sequelae of surviving childhood cancer. American Journal of Orthopsychiatry, 49, 608-616. Ozuna, J. (1979). Psychosocial aspects of epilepsy. Journal of Neurosurgical Nursing, II. 242-246. Palmer, B., & Lewis, C. (1976). Development of health attitudes and behaviors. Journal of School Health, 46, 401-402. Parsons, 0., & Prigatano, G. (1978). Methodological considera-
tions in clinical neuropsychological research. Journal ofConsulting and Clinical Psychology, 46, 608-619. Parsons, T. (1964). Socialstructureandpersonality. Glencoe, IL: Free Press. Perley, J., Winget, C., & Placci, C. (1971). Hope and discomfort as factors influencing treatment continuance. Comprehensive Psychiatry, 12, 557-563. Perrin, E. C., & Gerrity, P. S. (1981). There's a demon in your belly: Children's understanding of illness. Pediatrics, 67, 841-849. Perrin, J. M. (1986). Chronically ill children: An overview. Topics in Early Childhood Special Education, 5, 1-11. Petiet, C. A. (1983). Hope: The major predictor ofpositive resolution after marital loss. Paper presented at the 91st annual conference, American Psychological Association, Anaheim, California. Pierpont, M. E., LeRoy, B., & Baldfinger, S. (1984). Genetic disorders. In R. W. Blum (Ed.), Chronic illness and disability in childhood and adolescence. New York: Grune & Stratton. Pless, I. B. (1984). Clinical assessment: Physical and psychological functioning. Pediatric Clinics ofNorth America, 3/, 3345. Pless, I. B., & Roghmann, K. J. (1971). Chronic illness and its consequences: Observations based on three epidemiologic surveys. Journal of Pediatrics, 79. 351-359. Potter, P. C., & Roberts, M. C. (1984). Children's perceptions of chronic illness: The roles of disease symptoms, cognitive development, and information. Journal ofPediatric Psychology, 9, 13-27. Resnick, M. (1984). The teenager with cerebral palsy. In R. W. Blum (Ed.), Chronic illness and disability in childhood and adolescence. New York: Grune & Stratton. Riese, W. (1977). Dynamics in brain lesions. In Y. Lebrun & E. Buyssens (Eds.), Selected papers on the history of aphasia. Amsterdam: Swets & Zeitlinger. Roessler, R., & Bolton, B. (1978). Psychosocial adjustment to disability. Baltimore: University Park Press. Rourke, B. P. (1985). Neuropsychology of learning disabilities: Essentials of subtype analysis. New York: Guilford Press. Rutter, M., Graham, P., & Yule, W. (1970). A neuropsychiatric study in childhood (Clinics in Developmental Medicine Nos. 35 & 36). London: Spastics International Medical Publications/Heinemann Medical Books. Rutter, M., Tizard, J., & Whitmore, K. (1970). Education, health and behavior: Psychological and medical study of childhood development. London: Longman. Schilling, R. F., Schinke, S. P., & Kirkham, M.A. (1985). Coping with a handicapped child: Differences between mothers and fathers. Social Science and Medicine, 21, 857-863. Schunk, D. H. (1983). Abilityversuseffortattributional feedback: Differential effects on self-efficacy and achievement. Journal of Educational Psychology, 76, 1159-1169. Seligman, M. E. P., Peterson, C., Kaslow, N.J., Tanenbaum, R. L., Alloy, L. B., & Abramson, L. Y. (1984). Journal of Abnormal Psychology, 93, 235-238. Shapiro, J. (1983). Family reactions and coping strategies in response to the physically ill or handicapped child: A review. Social Science and Medicine, 17, 913-931. Shontz, F. C. (1982). Adaptation to chronic illness. InT. Millon,
COPING AND ADJUSTMENT
C. Green, & R. Meagher (Eds. ), Handbook of clinical health psychology. New York:: Plenum Press. Shurtleff, D. B. (1980). Myelodysplasia: Management and treatment. Current Problems in Pediatrics, 10, 1-98. Simmons, R. J., Corey, M., Cowen, L., Keenen, N., Robertson, J., & Levison, H. (1985). Emotional adjustment of adolescents with cystic fibrosis. Psychosomatic Medicine, 47, 111122. Small, L. (1973). Neurodiagnosis in psychotherapy. New York:: Brunner/Mazel. Smith, A. (1979). Practices and principles of clinical neuropsychology: Focusing on the hole rather than the doughnut. International Journal of Neuroscience, 9. 233-238. Smith, M. B. (1983). Hope and despair: Keys to the socio-psychodynamics of youth. American Journal of Onhopsychiatry, 53, 388-399. Sollee, N.D., Latham, E. E., Kindlon, D. J., & Bresnan, M. J. (1985). Neuropsychological impairment in Duchenne muscular dystrophy. Journal of Clinical and Experimental Neuropsychology, 7, 486-496. Stipek:, D., & Weisz, J. R. (1981). Perceived personal control and academic achievement. Review ofEducational Research, 51, 101-137. Tavormina, J. B., Kastner, L. S., Slater, P.M., & Watt, S. L. (1976). Chronically ill children: A psychologically and emotionally deviant population? Journal of Abnormal Child Psychology, 4, 99-110. Thomas, A. (1979). Learned helplessness and expectancy factors: Simplifications for research in learning disabilities. Review of Educational Research, 49, 208-221. Travis, G. (1976). Chronic illness in children: Its impact on child and family. Stanford: Stanford University Press. Trexler, L. E. (Ed.). (1982). Cognitive rehabilitation: Conceptualization and intervention. New York:: Plenum Press. Tropauer, A., Franz, N. M., & Dilgard, V. W. (1970). Psycho-
555
logical aspects of the care of children with cystic fibrosis. American Journal of Diseases in Childhood, 119, 424. von Bertalanffy, L. (1968). General systems theory. New York:: Braziller. Waddell, C. (1983). Faith, hope and luck: A sociological study of children growing up with a life-threatening illness. Washington, DC: University Press of America. Weisz, J. R., & Stipek, D. J. (1982). Competence, contingency, and the development of personal control. Human Development, 25, 250-281. Whelan, T. (1986). Neuropsychological performance, reading achievement, and perceptions of competence in boys with Duchenne muscular dystrophy. Paper presented at the 14th annual meeting, International Neuropsychological Society, Denver. Whelan, T. (1987). Neuropsychological performance of children with Duchenne muscular dystrophy and spinal muscle atrophy. Developmental Medicine and Child Neurology, 29, 207-211. Whitt, J. K., Dykstra, W., & Taylor, C. (1979). Children's conceptions of illness and cognitive development. Clinical Pediatrics, 18, 327-339. Whitten, C., Waugh, D., & Moore, A. (1974). Unmet needs of parents of children with sickle cell anemia. In J. Hercules (Ed.), Proceedings of the first national symposium on sickle cell disease. Bethesda: National Heart, Lung, and Blood Institute. Wortman, C. B. (1983). Coping with victimization: Conclusions and implications for future research. Journal ofSocial/ssues, 39, 195-221. Zeltzer, L., LeBaron, S., & Zeltzer, P. (1984). The adolescent with cancer. In R. W. Blum (Ed.), Chronic illness and disability in childhood and adolescence. New York:: Grune & Stratton.
30 Child Neuropsychology in the Private Medical Practice ERIN D. BIGLER AND NANCY L. NUSSBAUM
Child Neuropsychology in the Private Medical Practice
will focus on the role of the child neuropsychologist in the context of a general pediatric practice.
Utilizing a broad definition of child neuropsychological practice, the scope of such a practice would encompass all children with developmental and acquired disorders that affect cognition, behavior, and/or sensory and motor skills. National statistics (Department of Education) indicate that for 1982 and 1983, approximately 11% of all children received some form of special education (Weiner, 1983). More specific to pediatric practice, Dworkin (1985) presented statistics indicating that approximately 10% of a pediatrician's practice involves children with learning disabilities (LD), attention deficit hyperactivity disorder (ADHD), speech-language disorder, mental retardation, cerebral palsy, and related disorders (see Table 1). Equally pertinent are the findings of Burnett and Bell (1978) who reported that the greatest increase in pediatric practice in terms of new referrals came in the area of school and related problems. Accordingly, with such an incidence there is a clear indication for the need and important role that the child neuropsychologist can play in the pediatric and general medical setting. The most common setting for child neuropsychology to interface with the private medical practice will be the general practice of pediatrics or in pediatric neurology. Although there are other settings (i.e., family practice) or outlets for the practice of clinical child neuropsychology, this chapter
Identification of the Patient
ERIN D. BIGLER • Department of Psychology, University of Texas at Austin, Austin, Texas 78712; and Austin Neurological Clinic, Austin, Texas 78705. NANCY L. NUSSBAUM • Learning Diagnostic Center/ Austin Neurological Clinic, Austin, Texas 78705.
In the private medical setting, the pediatrician is the individual who plays the pivotal role in identifying children with a potential neuropsychological problem. In that considerable expertise is involved in the full evaluation of such children, the role of the pediatrician should be one involved in screening with appropriate referral when a potential problem is identified. Dworkin (1985) outlined a model of an office-based approach to the child with school and developmental problems. An adaption of this model is depicted in Figure l. In this approach the pediatrician constitutes the first line of contact with the child, and the neuropsychologist functions as the subspecialist providing specific evaluation, assessment, and possible treatment for the referral problem. With this model, potential medical problems (e.g., thyroid dysfunction, hypoglycemia) that can influence behavior and mimic neurobehavioral disorders can be addressed directly by the pediatrician. Likewise, certain conditions, such as ADHD, which may require ongoing medical management (i.e., stimulant medication), can also be dealt with directly by the physician. In addition to a diagnostic role, the neuropsychologist is in the position (has the expertise) to assist with behavioral management, family and school intervention, as well as individual supportive psychotherapy.
Identification of the Problem As portrayed in Figure 2, there is a considerable overlap between the medical and the neuropsycholo557
SS8
CHAPTER 30
TABLE 1. Pediatricians' Estimates of the Number of Children in Their Practices with Handicapping Conditionsa Reponed no. of children Disability
Mean
Range
Specific learning disability (as defined by physician or school) Hyperactive/minimal brain dysfunction (as defmed by physician) Language/speech impainnent (excluding developmental articulation problems) Mental retardation (mild to profound) Cerebral palsy (mild to severe) Hearing impairment (nontransient conductive or sensorineural-mild to profound) Serious emotional disturbance (as defined by physician) Legally blind
50.6 33.3 29.3 27.0 14.1 10.6 8.2 2.1
0-300 0-400 0-100 0-250 0-150 0-250 0-100 0-50
8
Modified from Shonkoff et al. (1979).
gical sphere regarding the child with developmental related to the accurate diagnosis and definition of the problems. Accordingly, this requires that there be a presenting problem. systematic and integrated approach toward the detecThe overlapping area in Figure 2 illustrates the tion, evaluation, and treatment of developmental dis- type of information that the child neuropsychologist orders. A crucial dimension of such an approach is is uniquely suited to provide the medical practitioner Yisn 1
Direct Treatment Route
History Physical Examination Vision-Heertng Screen Disperse! of Questionnaires
Possible Treatment Route
1
VIsit 2
Neurodevelopmental Assessment ---------------------------
Parent Conference Review Test Results Objectives -------------------------Recommendations
I Medicel Intervention I
1
I Follow-up ~- ---
---
Behavioral Management Family Intervention ------- School Consultation Indi vi due I Psychother11py
FIGURE I. An office-based appro~K:h to children with developmental problems. (Modified from Dworkin et al., 1981.)
CHILD NEUROPSYCHOLOGY IN THE PRIVATE MEDICAL PRACTICE
Gross & Fine Motor Viouel/Spetiel Processing Sequential Proceni 114 Auditory/Verbal Processing Body/Sensory A'Wareness Cognitive Depelopment Personality/Behavioral
559
Physico! Development
Hearing/Vision
FIGURE 2. Components of the assessment of children. (Modified from Dworkin & Levine, 1980.)
(this topic area has been more completely addressed by Fletcher & Taylor, 1984). Information pertaining to these particular areas of functioning can be considered critical for the accurate diagnosis and definition of childhood developmental problems. Using a carefully chosen battery of assessment techniques, the neuropsychologist is able to thoroughly define the child's ability structure. The neuropsychologist can provide information regarding the child's strengths and weaknesses through the selection of a broad range of developmentally structured tests. It is important to emphasize that these measures should be chosen and interpreted within a developmental context; therefore, the use of normative data is essential in evaluating the individual child (Fletcher & Taylor, 1984). Next, some specific methods for obtaining this developmental information will be discussed. The battery of assessment techniques given in Table 2 has been developed at the Austin Neurological Clinic in order to provide a comprehensive evaluation of neuropsychological functioning in children (Nussbaum, Bigler, & Koch, 1986). It was felt that the measures included in this battery provided the necessary information to the medical practitioner and other professionals for the accurate diagnosis and remediation of the child with developmental dysfunction. The measures included in the Comprehensive Austin Neuropsychological Assessment Battery for Children (CAN-ABC) have been found to be useful in providing the pediatrician with the data needed to accurately diagnose and plan remediation for the child with developmental problems. The measures in this battery were selected to provide comprehensive information in the areas shown in the overlapping zone in Figure 2. As shown in Table 3, information obtained from the CAN-ABC can be organized in terms of these overlapping areas. In addition, the
organization of the measures into these areas can be useful when the referral question calls for a more restricted rather than a comprehensive evaluation. For example, if the child's pediatrician has a specific question regarding only the child's personality/behavioral functioning, then the battery may be modified to focus on this referral question. The overlapping among the areas presented above is indicative of the point that very few measures of ability are pure, but are more often interdependent. This is especially true for more complex or higher level areas of functioning such as cognition and memory. Thus, it is the job of the trained child neuropsychologist to interpret the results from a battery of tests in a meaningful way. For example, if a child does poorly on the Digit Span subtest of the WISC-R, is it due to an attentional problem, a memory deficit, an auditory processing problem, or some combination of these three areas? One must attempt to integrate information gathered from a wide variety of sources to provide a complete understanding of the individual child. The meaningful interpretation of information gathered through neuropsychological assessment is the truly unique capacity that the neuropsychologist brings to a private medical setting.
Communication of Assessment Results The final stage in the process of consultation by the neuropsychologist is the communication of the assessment results. The importance of this stage cannot be overemphasized for it is at this point where it is determined whether or not the information gathered during the evaluation can be used efficaciously by the pediatrician. The clear and effective communication of assessment results is crucial for the accurate diagnosis and appropriate intervention for the child with developmental problems.
TABLE 2. The Comprehensive Austin Neuropsychological Assessment Battery for Children (CAN-ABC)a General physical features Physical measures -Height -Weight -Visual acuity -Head circumference
Lateral dominance -Hand -Foot -Eye
Physical anomalies -Facial features -Epidennal features (e.g., cafe au lait spots) -Hands -Other anomalies (e.g., steepled palate)
The Halstead Neuropsychological Test Battery for Children and the Reitan-lndiana Neuropsychological Test Battery for Children (selected subtests)b Motor 1. Strength of grip 2. Finger oscillation -Electric-5- to 8-year-olds -Manual-9 and older Tactual Performance Test 1. 6 fonn/horizontal-5- to 8-year-olds 2. 6 fonn/vertical-9- to 14-year-olds 3. 10 fonn/vertical-14 to adult 4. Sequin/Goddard (Anastasi, 1969)-poor cooperation, or under age 5 Sensory-Perceptual Exam l. Tactile -Single, double simultaneous, ipsilateral, and contralateral 2. Auditory -Single, double simultaneous 3. Visual -Upper, middle, lower visual fields -Single, double simultaneous -Visual fields 4. Finger recognition 5. Finger graphesthesia -5- to 8-year-old, symbols X's and O's -9 to adult, numbers 6. Fonn recognition -Included if the child scores one standard deviation below the mean on tactile, finger recognition, and/or finger graphesthesia Reitan-Aphasia Screening Battery (Halstead & Wepman, 1959; Selz & Reitan, 1979) 1. 5- to 8-year-old fonn 2. 9 and older fonn Wide Range Achievement Test (Jastak & Jastak, 1965) 1. Reading (primarily a test of reading recognition) 2. Spelling (provides quantitative spelling level) 3. Arithmetic 4. Preacademic tasks as indicated Bader Test of Reading and Spelling Patterns (Boder & Jarrico, 1982) 1. Provides qualitative information on spelling/reading Durrell Analysis of Reading Difficulty (selected subtests) (Durrell & Catterson, 1980) 1. Silent reading (measure of reading comprehension) 2. Oral reading (measure of visual/auditory processing and reading comprehension) 3. Listening comprehension (measure of auditory/verbal comprehension) Beery Test of Visual/Motor Integration (Beery, 1982) 1. Measure of perceptual motor ability Raven's Coloured Progressive Matrices (Raven, 1965) 1. Measure of nonverbal abstract reasoning
CHILD NEUROPSYCHOLOGY IN THE PRIVATE MEDICAL PRACTICE
561
TABLE 2. (Continued) Wechsler Intelligence Scale for Children-Revised (Wechsler, 1974) I. Verbal IQ subscales (infonnation, vocabulary, similarities, arithmetic, comprehension) 2. Perfonnance IQ subscales (picture completion, block design, object assembly, picture arrangement, coding) 3. Digit span Family history questionnaire I. Background infonnation 2. Pregnancy history 3. Birth history 4. Developmental history 5. Medical history 6. School history Child Behavior Checklist-Revised (Achenbach & Edelbrock, 1983) I. Provides an easily reviewed list of possible behavior problems 2. Provides quantitative scores on such personality scales as Depression, Aggression, etc. Personality Inventory for Children-Revised (Wirt, Lachar, Klinedinst, & Seat, 1982) 1. Provides quantitative scores on such personality scales as Depression, Aggression, etc. Projective Drawings I. House/tree/person (Buck, 1948), kinetic family drawings (Burns & Kaufman, 1972) -provide qualitative infonnation on self-concept, family dynamics, etc. Behavioral observation inventory I. Provides a short informal assessment of behaviors observed during !he evaluation Additional measures included as needed Benton Visual Retention Test (Benton, 1974) -included if: 1. questionable ADHD problems 2. deficient visuomotor performance (Beery) leads to questions about deficits in visual memory versus visuomotor coordination Kaufman Assessment Banery for Chikken (Kaufman & Kaufman, 1983) -included if: 1. marginal or questionable LD 2. more in-depth information is needed concerning child's nonverbal intellectual abilities 3. particularly useful subscales Hand Movements (useful attentional measure) Gestalt Closure (useful visual processing measure) Matrix Analogies (useful nonverbal reasoning measure) Spatial Memory (useful visual memory measure without a motor confound) Other Halstead Neuropsychological Test Battery for Children subtests Administered to 9- to 14-year-olds as indicated 1. Trails A (measure of sequential visual processing, attention) 2. Trails B (measure of sequential visual processing, attention, cognitive flexibility) 3. Seashore Rbylhm Test (measure of sequential auditory processing, attention, auditory memory) 4. Speech Sounds Perception Test (measure of auditory processing, attention, sight/sound matching) aMeasures are scored according to individual norm tables provided with each specific test or according to nonnative infonnation provided by Spreen and Gaddes (1969) or Knights and Norwood (1980). ~>Reitan & Davison, 1974.
findings so that the presenting questions may be answered. Sufficient background history should be reported to answer any questions concerning pregnanThe exemplary report should always start with a cy, birth and delivery, developmental milestones, specific presenting or identified problem. This per- and medical history that may be salient variables remits focusing the results of the consultation and test lated to the presenting problem(s).
The Neuropsychological Report
562
CHAPTER 30
TABLE 3. The Organization of Neuropsychological Test Results Motor (fine and gross) Strength of grip Finger oscillation Tactual Performance Test Beery Test of Visual/Motor Integration WISC-R (Block Design, Object Assembly, Coding) Visllllll spatial processing Visual acuity Sensory-perceptual exam (visual exam) Beery Test of Visual/Motor Integration Benton Visual Retention Test Trails A & B Tactual Performance Test Reitan-Aphasia Screening Battery (Visual Constructional tasks) K-ABC (Hand Movements, Gestalt closure) WISC-R (performance IQ subscales) Body awareness Sensory-Perceptual Exam errors (tactile, finger recognition, finger graphesthesia, fonn recognition) Tactual Performance Test Auditory verbal processing Sensory-Perceptual Exam errors (Auditory) Seashore Rhythm Test Speech Sounds Perception Test Durrell (listening comprehension) WISC-R (verbal IQ subscales) Peabody Picture Vocabulary Test-Revised (1981) Sequential processing WISC-R (Digit span, picture arrangement) K-ABC (Hand Movements) Trails A & B Seashore Rhythm Test Memory WISC-R (Digit span) Benton Visual Retention Test K-ABC (Spatial memory, Hand Movements) Durrell (silent reading-unstructured story recall; listening comprehension-structured story recall)
Tactual Performance Test (memory for objects and location) Seashore Rhythm Test Cognitive development (knowledge, reasoning) Raven's Coloured Progressive Matrices WISC-R (Information, Vocabulary, Similarities, Arithmetic, Comprehension, Picture Arrangement) K-ABC (Matrix Analogies) Auention Sensory-Perceptual Exam (tactile, auditory, visual, finger recognition, finger graphesthesia) Benton Visual Retention Test Trails A & B Seashore Rhythm Test Speech Sounds Perception Test Durrell (listening comprehension) WISC-R (Arithmetic, Picture Completion, Coding, Digit Span) K-ABC (Hand Movements) Child behavior checklist Academic skills Reitan-Aphasia Screening Battery (reading, spelling, arithmetic tasks) Wide Range Achievement Test (reading, spelling, arithmetic, preacademic tasks) Boder Test of Reading and Spelling Patterns Durrell (silent reading, oral reading, listening comprehension) WISC-R (Arithmetic) Personality/behavioral Child Behavior Checklist Personality Inventory for Children-Revised Projective drawings Behavioral observation inventory Psychosocilll factors Parent interview Family history questionnaire Child behavior checklist
In the next section of the neuropsychological report, a listing of the tests administered during the evaluation should be provided. This informs the reader of the specific measures that were used to obtain information regarding the child's ability structure or behavioral characteristics. As shown in Table 4, the section containing the assessment results has been divided into subsections dealing with the child's intellectual/cognitive functioning, academic abilities, neuropsychological functioning, and personality /behavioral characteristics. We have found that it is useful to have the evaluation results divided into these subsections in order to provide an organized picture of the child's func-
tioning, and to provide the reader with ready access to pertinent information. The next section containing the evaluation summary and clinical impression is of critical importance. It is in this section that the results of the assessment are summarized and integrated in order to provide a holistic understanding of the child's functioning. In addition, this section contains the child's DSM-III-R classification when such a categorization is appropriate. Finally, appropriate recommendations should be made based on the results of the assessment. These recommendations should contain general as well as specific information that may be helpful in treatment
CHILD NEUROPSYCHOLOGY IN THE PRIVATE MEDICAL PRACTICE
563
TABLE 4. Format of the Neuropsychological Report Presenting problem Referral question (e.g. , presenting seizures) Background history Genetic history (e.g., Down's Syndrome, epilepsy) Pregnancy (e.g., complications-alcohol use, etc.) Birth and delivery (e.g., complications-forceps, etc.) Neonatal history (e.g. , birth weight) Medical history (e.g. , significant head injuries) Family history (e.g., parents' education, LD in the family) Tests administered Assessment results Intellectual/ cognitive functioning IQ scores Subtest scores Clinical description/ interpretation Academic functioning Achievement scores Clinical description/ interpretation Neuropsychological test findings , Physical stigmata/physical measurements Hand, eye, and foot dominance Motor functioning Fine motor Gross motor Praxic ability Visuomotor copying Sensory perceptual functioning Vision (acuity/fields) Hearing Tactile (double simultaneous) Graphesthesia Stereognosis Finger gnosis
planning for the child. The important point that must be emphasized here is that the recommendations should clearly follow from the results of the evaluation.
Language Articulation Receptive Expressive Naming Spelling Reading Calculations Memory Verbal Visual/spatial General cognitive PersonalityI emotional functioning Behavioral observations-subjective findings Projective test results Projective drawings Thematic testing Rorschach (1942) Objective personality scores/patterns Clinical summary Clinical impression Summary and integration of assessment results DSM·lli·R fonnat followed when appropriate Recommendations: 1. To referring doctor, including therapists who the child may be seeing 2. To school 3. To parents 4. When to follow up 5. Miscellaneous
Case Study Material
Next, a case study will be presented to illustrate the type and format of information typically obtained in a neuropsychological assessment. First, the raw data will be presented in terms of how they can be Follow-up Conference conceptually organized as outlined in Table 3. SecThe parents play a focal role in assuring that ond, the results of the evaluation will be presented in appropriate recommendations are followed; accord- the report format shown in Table 4 in order to ingly, considerable effort should be directed at illustrate the communication of neuropsychological providing a clear understanding on the parents' part. findings. Typically, this will require at least an hour conThe case of JB presented in Table 5 and 6 illusference to review test results and outline potential trates the way in which results from the neuropsytreatment options where appropriate. Also it is most chological evaluation can be communicated to the helpful to have various reading lists available to the physician in report format. The goal of the neuropsyparents so that they can pursue on their own further chologist in writing the neuropsychological report is enlightenment pertaining to the nature of their child's to: (1) address the specific referral question(s); (2) problems. thoroughly present relevant test results; (3) pro-
564
CHAPTER 30
TABLE 5. Case Presentation for JB: Results of the Comprehensive Austin Neuropsychological Assessment Battery for Children Name: JB Education: 2nd grade Date: 1124/84 Age: 8 years, 1 month Race: Caucasian Examiner: MH Sex: Male Location: Austin Neurological
Motor (fine and gross) Strength of grip Dominant hand= 15; Nondominant Finger oscillation Dominant hand SS* = 106
= 39.8;
=
Nondominant SS = 106
12.2
= 35.2
Tactual performance test Dominant = 5.9; Nondominant = 4.1;
ss
= 99
ss = 99
Both= 1.5;
ss =
100
Total= 11.5
ss =
100
Beery Test of Visual/Motor Integration SS = 10; Percentile = 60; Age Equivalent = 7,9 WISC-R (Block Design, Object Assembly, Coding) BD = II, OA = 13; Cod = 12
Visual/ spatial processing Visual acuity Uncorrected: Right = 20/50; Left = 20/30 Corrected: Right = 20/20; Left = 20/20 Sensory-perceptual exam (Visual exam) Visual fields = Normal to simple confrontation DSS errors: Right = 1; Left = 0 ss = 76 ss = 106 Beery Test of Visual/Motor Integration SS = 10; Percentile= 50; Age equivalent= 7,9 Benton Visual Retention Test (normed IQ = 91, age = 8) Expected correct = 2; Obtained correct = 5 Expected errors = 12-13; Obtained errors = 8 Errors: Right visual field = 4; Left visual field = 3 Trails A and B Trails A = 43 sec, 0 errors (9 y.o. X = 21 sec)
Trails B = 52 sec, 0 errors (9 y.o. X = 49)
Tactual performance test Memory = 3; Location = 0 ss = 84 ss = 74 Reitan-Aphasia Screening Battery (visual constructional tasks) Within normal limits K-ABC (Hand Movements, Gestalt Closure) HM=8 GC=7 WISC-R {performance IQ = 102) (Picture Completion, Picture Arrangement, Block Design, Object Assembly, Coding) PC = 8 PA = 8 BD = 11 OA = 13 Cod = 12
CHILD NEUROPSYCHOLOGY IN THE PRIVATE MEDICAL PRACTICE
TABLE 5. (Continued) Body awareness Sensory-perceptual exam errors (tactile, finger recognition, finger graphesthesia, form recognition) Tactile: DSS, Right = 0; Left = I ss = 114 ss = 95 Finger Graphesthesia: Right = 0; Left = 0 = 121 = 119 Finger Recognition: Right = 3; Left = 3 ss 70 = 80
ss
=
ss
ss
Tactual performance test Dominant= 5.9; Nondominant = 4.1; Both = 1.5; Total 11.5 ss = 99 ss = 99 ss = 100 ss = 100
=
Auditory verbal processing Sensory-perceptual exam errors (auditory) Right = 0; Left = 0 ss = 105 ss = 108 Seashore Rhythm Test Not applicable for this age Speech Sounds Perception Test Not applicable for this age Durrell (listening comprehension) LC = second grade level WISC-R (information, similarities, arithmetic, vocabulary, comprehension) Info = 5; Sim = 6; Arith = ll; Voc = 7; Verbal IQ = 82 Peabody Picture Vocabulary Test-Revised (1981) Mental Age Score 5 years, 11 months; ss = 75
=
Sequential processing WISC-R (Digit span, Picture Arrangement) PA = 8 DS = 6 K-ABC (Hand Movements) HM= 8 Trails A and B Trails A = 43 sec, 0 errors; Trails B (SS, see above)
= 52 sec, 0 errors
Seashore Rhythm Test Not applicable for this age
Memory
WISC-R (Digit Span) DS = 6
Benton Visual Retention Test (norrned IQ = 91, age = 8) Expected correct 2; obtained correct 5 Expected errors = 12-13; obtained errors = 8
=
=
K-ABC (Spatial Memory, Hand Movements)
SM=7
HM=8 (continued)
565
566
CHAPTER 30
TABLE 5. (Continued) Dunell (silent reading-unstructured story recall; listening comprehension-structured story recall) LC = Fair SR = Poor Tactual performance test (memory for objects and location) Memory = 3; Location = 0 ss = 74 ss = 84 Seashore Rhythm Test Not applicable for this age Cognitive development (knowledge, reasoning) Raven's Coloured Progressive Matrices 50 Percentile
=
WISC-R (information, vocabulary, similarities, arithmetic, Arith = II Sim = 6 Voc = 7 Info = 5
comprehension, picture arrangement) PA = 8 Comp = 7
K-ABC (matrix analogies) MA = 11 Anention Sensory-perceptual exam (tactile, auditory, visual, finger recognition, finger graphesthesia) Tactile: DSS, Right = 0; Left = I Auditory: Right = 0; Left = 0 Visual, DSS: Right = I; Left = 0 Finger Recognition: Right = 3; Left = 3 Finger Graphesthesia: Right = 0; Left = 0 (For SS, see above) Benton Visual Retention Test Expected correct = 2; Obtained correct = 5 Expected errors = 12-13; Obtained errors = 8 Trails A & B Trails A = 43 sec, 0 errors; Trails B (SS, see above)
= 52 sec, 0 errors
Seashore Rhythm Test Not applicable for this age Dunell (Listening Comprehension) LC = 2nd grade level, fair structured recall Digit Span) WISC-R (Arithmetic, Picture Completion, Coding, Cod = 12 DS = 6 Arith = 11 PC = 8 K-ABC (Hand Movements) HM = 8 Child Behavior Checklist Mild attentional problems noted Behavioral observation inventory No attentional problems noted on informal observation Academic skills Reitan-Aphasia Screening Battery (reading, spelling, arithmetic tasks) Reading errors noted Wide Range Achievement Test Standard Score 89 Reading 99 Spelling 105 Arithmetic
Grade Equivalent 2.4
Percentile
23
2.8
47
3.1
63
CHILD NEUROPSYCHOLOGY IN THE PRIVATE MEDICAL PRACTICE
567
TABLE 5. (Continued) Boder Test of Reading and Spelling Patterns Reading/Spelling Pattern = Dysphonetic Type Durrell (silent reading, oral reading, SR = Middle I st OR = Lower 1st
listening comprehension) LC = 2nd grade
WISC-R (arithmetic) Arith = II
Personality! behavioral Child Behavior Checklist (scale elevations above 70) Aggressive, Internalizing, Anxious, Depressed Personality inventory for children-revised (scale elevations above 70) Adjustment, depression, withdrawal, anxiety Projective drawings House/Tree/Person: sparse, vacant, small, human stick figures Behavioral observation inventory Poor eye contact, withdrawn, no spontaneous speech, brief verbal response
Psychasociol factors Parent interview (significant points) Slow progress in school, father reported similar learning problems, school phobia, temper outbursts Family history questionnaire (significant points) Normal pregnancy, induced labor, forceps delivery, developmental milestones within normal limits, allergies Child behavior checklist Noted to have poor peer and family interactions
• SS = Standard score.
TABLE 6. Neuropsychologica l Assessment Report (Based on Data in Table 5) Child's name: JB Clinic number: 99999 Assessment date: 1/24/84 Presenting problem This child was referred by Dr. K., pediatric neurologist, for a comprehensive neuropsychological evaluation due to questions concerning a possible learning disorder. Background history JB was an 8-year-Qid boy who was in the second grade at the time of assessment. He had been referred for neuropsychological evaluation due to inadequate academic progress and behavioral problems. This child's medical history showed that he was the product of a full-term, normal pregnancy. InJB's birth history, it was reported that labor was induced and he was delivered using forceps. His developmental milestones were reported to be within normal limits. He crawled at approximately 7 months, walked at 12 months, and started saying his first words at about I year of age. JB's parents were reported to have high school educations. His father was self-employed as a plumber and his mother was not employed outside the home. Also, JB's father reported that he may have experienced similar learning problems as a child. In addition, as part of his evaluation, JB received an electroencephalogram that showed sharp wave activity over the left temporal region which suggested left temporal lobe dysfunction. No electroencephalographic seizure activity was noted. There were no other significant findings in JB's medical history except that he was noted to have an allergy to milk and pollens.
(continued)
568
CHAPTER 30
Table 6. (Continued) Tests administered Comprehensive Austin Neuropsychological Assessment Battery for Children (CAN-ABC) • Wechsler Intelligence Scale for Children-Revised • Kaufman Assessment Battery for Children • Wide Range Achievement Test • Boder Reading-Spelling Test • Durrell Analysis of Reading Difficulty • Peabody Picture Vocabulary Test-Revised • Reitan-Indiana Neuropsychological Test Battery for Children • Raven's Coloured Progressive Matrices • Benton Visual Retention Test • Beery Visuo-Motor Integration Test • Behavioral Observations Inventory • Personality Inventory for Children • Child Behavior Checldist Test results Cognitive /intellectual functioning
WISC-R results Full scale IQ = 91 Verbal IQ score= 82 Information 5 6 Similarities Arithmetic II Vocabulary 7 Comprehension 7 (Digit Span) 6
Performance IQ score = 102
8 Picture completion Picture arrangement 8 Block design II 13 Object assembly Coding 12
K-ABC results Mental processing composite = 83 Simultaneous processing Sequential processing = 74 Gestalt closure 8 Hand movements Triangles 5 Number recall Spatial memory 4 Word order Photo series
93
7 13 7 7
Raven's CPM Results: 50th percentile Results of intellectual assessment indicated m 's level of intellectual functioning to be in the average to low average range. On the WISCR, m was found to have severely discrepant performance between the PIQ and VIQ scales, with his VIQ score over one standard deviation below his PIQ score. He exhibited a great deal of scatter in his subtest scores, with particularly deficient performance on the Information, Similarities, and Digit Span subtests of the VIQ scale. Similarly, on the K-ABC, m·s score on the Sequential scale was over one standard deviation below his Simultaneous scale score. He scored especially low on the Number Recall and Word Order subtests of the Sequential scale. In addition, m scored at the 50th percentile on Raven's CPM, a test of perceptual discrimination and abstract reasoning. Also, m showed adequate performance on the Beery Test of Visual/Motor Integration. His score on this test showed his visual/motor skills to be at age level. Likewise, m·s performance on the Benton Visual Retention Test was also within normal limits. In general, it appeared that m•s overall intellectual functioning was in the normal range, but he exhibited an abnormal pattern of performance. His intellectual profile showed significantly more impaired verbal and sequential abilities than visual/constructional abilities.
CHILD NEUROPSYCHOLOGY IN THE PRIVATE MEDICAL PRACI'ICE
S69
Table 6. (Continued) Academic functioning WRAT results Reading Spelling Arithmetic
Grade level equivalent 2.4
2.8 3.1
Standard score 89 99 105
Percentile 23
47 63
Boder Test of Reading and Spelling Patterns Reading/Spelling Pattern: Dysphonetic Durrell Analysis of Reading Difficulty Silent Reading Middle 1st; Oral Reading = Lower 1st Listening Comprehension = 2nd
=
Peabody Picture Vocabulary Test Age equivalent = 5 years, II months; Standard Score = 75 On the WRAT, JB was found to have grade level academic skills, with slightly better arithmetic than reading and spelling abilities. However, on the Durrell, JB's reading and language scores were below grade level. He was noted to have particular difficulties with sound/symbol associations, and his reading comprehension was found to be quite poor. Similarly, JB scored over one and a half standard deviations below the mean on the Peabody Picture Vocabulary Test-Revised (1981). Also, JB was reported to sometimes make semantic substitutions in his reading. For example, when the stimulus word was "house" he responded with "home" and "her" was read as "she," also "horse" was read as "pony." These types of reading and spelling errors have been associated with a dysphonetic type of dyslexia (Boder & Jarrico, 1982). In general, findings from achievement testing indicated that JB had some basic academic skills, such as word identification skills and basic calculation skills. However, he also exhibited marked deficiencies in other academic areas, such as reading comprehension and vocabulary development. Neuropsychological functioning At the time of the examination, this child stood 50 inches in height and weighed 54 pounds. His head circumference was measured at 53 centimeters. IB was right hand, foot, and eye dominant. No abnormal morphological physical characteristics were noted on a brief physical examination.
Reitan-lndiana Neuropsychological Test Battery for Children Test results for the Reitan-Indiana Battery were essentially within normal limits. His motor findings on the finger oscillation and strength of grip test were in the normal range. However, on the Sensory Perceptual Exam, JB did exhibit mild, bilateral finger dysgraphesthesia, but these results in isolation were not found to be clinically significant. Language screening
On the Reitan-Aphasia Screening Test, JB was found to make numerous reading errors. He also exhibited marked dysnomia on the Peabody Picture Vocabulary Test with an age equivalent score of only 5 years II months. (See the Intellectual and Academic sections for a review of other pertinent information.) Visual/perceptual tests Graphomotor ability on the Beery VMI was approximately at age level. JB scored at the 7 year 9 month level with a standard score of 10, which placed him at the 50th percentile. Memory assessment Test results from a wide variety of sources indicated that JB was having some difficulties with both verbal and visual/spatial memory. He exhibited deficient performance on the recall of verbal information from the WISC-R, K-ABC, and the Durrell. Similarly, he showed deficits in visual/spatial memory on the K-ABC and the Tactual Performance Test.
(continued)
570
CHAPTER 30
Table 6. (Continued) Summary of neuropsychological test results Results from the neuropsychological assessment were in general agreement with other test results reviewed to this point. JB seemed to be showing greater problems in the area of verbal language functioning than visual/spatial functioning. He also exhibited some memory difficulties which may have been related attentional deficits. Personality/ behavioral functioning
Objective data JB was found to have significant elevations on the following scales from the Personality Inventory for Children (PIC) and the Child Behavior Checklist (CBC): Depression, Anxiety, and Aggression. Also, his scale score on PIC-Withdrawal was over one and a half standard deviations above the mean, which indicated significant problems with withdrawal. It was reported that JB had apparently had some difficulty separating from his parents. He developed some form of school phobia during kindergarten, and would become physically ill prior to school. This behavior apparently abated after several weeks of school attendance. His teacher at the time of assessment noted that JB was a tense, shy child. He was also reported to have occasional temper tantrums with his parents and siblings. During assessment, the examiner noted that JB was very quiet and reserved throughout the examination. He was observed to be cooperative, but in a passive way in that rapport was never established and he did not appear to become engaged during the evaluation. He was reported to initiate no spontaneous conversation, and he had poor eye contact. Similarly, his affect was noted to be somewhat flat and depressed. Subjective data JB's projective drawings were found to be somewhat vacant and sparse. He drew a very small, simple tree. Likewise, his human figure drawings were sticldike. They were also quite small and placed at the bottom edge of the page. These findings may have indicated significant self-esteem and adjustment problems. Summary of personality/behavioral findings In summary, JB appeared to have been experiencing general problems in the area of social interaction. Specifically, he seemed to be showing internalizing-type problems with features of depression, anxiety, and withdrawal. Clinical impression In summary, this child performed poorly on a number of verbal/language/sequential tasks typically thought to be dependent on lefthemisphere functioning; whereas he exhibited relatively intact visual-spatial functioning. Such findings suggested a probable lefthemisphere-based learning disorder. This interpretation was supported further by abnormal EEG findings in the left temporal region.
In addition, at the time of the evaluation, JB appeared to be experiencing a number of significant emotional and adjustment problems related to depression and impaired socialization. DSM-III-R classification Axis I: (309.28) possible childhood adjustment disorder with mixed emotional feature Axis ll: (315.9) developmental disorder (learning disability) Axis III: abnormal EEG (left temporal, sharp wave activity) Recommendations Given the results of JB's neuropsychological assessment, it is recommended that he receive special education services. With respect to the special education curriculum, this child would probably benefit from a remediation program that focused on verbal/language training. Some of JB 's verbal expression problems may be related to anxiety and a lack of spontaneity in social discourse. Thus, there may be some improvement in his level of functioning as he becomes more relaxed and comfortable in communicating with others. In addition, because of his verbal/language difficulties, JB may not have acquired the necessary interpersonal skills for good social interaction, which may have led to depression and withdrawal. Compounding these problems were also left-hemisphere-based academic difficulties, which may have aggravated self-esteem and adjustment problems. It is therefore recommended that he participate in a language enrichment program with a qualified speech/language therapist in order to increase his confidence and social appropriateness in communication.
CHILD NEUROPSYCHOLOGY IN THE PRIVATE MEDICAL PRACI'ICE
5'71
Table 6. (Continued) In addition, JB appeared to show some significant problems with emotional adjustment. These findings indicated that he may benefit from some supportive counseling that would focus on social skills training to improve the quality of his peer and family relationships. This intervention should also take advantage of the strengths JB has in some nonverbal areas. Furthennore, his parents should encourage his participation in other nonacademic activities that would help promote positive self-esteem. Finally, JB 's progress should be closely monitored, and he should return for follow-up testing on an annual basis for the next two to three years.
vide a clinicai interpretation of the data; (4) present pertinent recommendations; (5) communicate these points clearly and effectively.
Benton, A. (1974). The Revised Visual Retention Test (4th ed.). New York: Psychological Corporation. Boder, E., & Jarrico, S. (1982). The Boder Test of ReadingSpeUing Patterns: A diagnostic screening test/or subtypes of reading disability. New York: Grone & Stratton. Buck, J. (1948). House-tree-person test. Journol ofClinical PsySummary chology, 4, 151-159. Burnett, R., & Bell, L. (1978). Projecting practice patterns. Pedialrics, 62(Suppl.), 625-680. The role of the child neuropsychologist in the R., & Kaufman, S. (1972). Actions, styles and symbols in Bums, consulprivate medical practice is that of an expert kineticftunily drawings. New York: Brunner/Mazel. tant. Due to the increasing identification and referral Picture Vocabllof children with developmental disorders (Dworkin, Dunn, L. M., & Dunn, L. M. (1981). Peabody lary Test-Revised (PPVT-R). Circle Pines, Minnesota: 1985) to pediatricians, there is a growing recognition American Guidance Service. of the need for better diagnostic and remediation Durrell, D., & Catterson, J. (1980). D11rrellAnalysis of Reading techniques. The child neuropsychologist is uniquely Difficllity. New York: Psychological Corporation. suited to provide relevant information to the child's Dworkin, P. (1985). Learning and behllvior problems of schoolphysician to aid in the delineation of problem areas children. Philadelphia: Saunders. and to assist in formulating an appropriate interven- Dworkin, P., & Levine, M. (1980). The preschool child: Prediction and prescription. In A. Scheiner & I. Abroms (Eds.), The tion strategy. practical mJZnagement ofthe developmentally disabled child. Specific information was presented earlier in St. Louis: Mosby. this chapter outlining one test battery that has been developed to provide a comprehensive evaluation of Dworkin, P., Woodrum, D., & Brooks, K. (1981). Pediatricbased assessment: Children with school problems. J0117111ll of the child's neuropsychological functioning (see School Health, 51, 325-329. Table 2). The CAN-ABC is just one battery of tests Fletcher, J., & Taylor, H. ( 1984). Neuropsychological assessment that can be used to obtain this type of information. of children: A developmental approach. Texas Psychologist, The important feature is that a comprehensive and 36(3), 14-20. integrative approach be used in the assessment of the Halstead, W., & Wepman, J. (1959). The Halstead-Wepman individual child. Aphasia Screening Test. Journol ofSpeech and Hearing Disorders, 14, 9-15. In addition, a report format was presented to illustrate the organization and communication of the Jastak, J., &Jastak, S. (1965). The Wide Range Achievement Test mJZnual. Wilmington, DE: Guidance Associated. evaluation results (see Table 4). Finally, a case study A., & Kaufman, N. (1983). KtlllfmonAssessrMntBatKaufman, was provided to exemplify the type of information Children: Administralion and Interpretive Manual. for tery gathered and the way in which it can be communiCircle Pines, MN: American Guidance Service. cated to the child's physician. Knights, R., & Norwood, J. (1980). Revised smoothed normative datil on the ne11ropsychological test battery for children. Ottawa, Canada: Robert M. Knights Psychological References Consultants. Nussbaum, N., Bigler, E., & Koch, W. (1986). Neuropsychological1y derived subgroups of learning disabled children: PerAchenbach, T., & Edelbrock, C. S. (1983). The Child Belulvior sonality/behavioral dimensions. Journal of Research and Checklist-Revised. University of Vennont: Burlington, Development in Ed~~ealion, 19, 51-61. Vermont. Anastasi, A. (1969). Psychological testing (3rd ed.). London: Ravens, J. C. (1965). Guide to using the colored progressive mJZtrices. London: H. K. Lewis. Macmillan & Co. Beery, K. (1982). Revised administralion, scoring and teaching Reitan, R., & Davison, L. (1974). Clinical neuropsychology: Current status and application. New York: Hemisphere. mJZnual for the developmental test of visual-motor integraRorschach, H. (1942). Psychodi4gnostics: A dillgnostic test based tion. Cleveland: Modem Curriculum Press.
572
CHAPTER 30
on perception (P. Lemkau & B. Kranenburg, trans.). Bern: Huber (U.S. distributor: Grone & Stratton). Selz, M., & Reitan, R. ( 1979). Rules for neuropsychological diagnosis: Classification of brain function in older children. Journal of Consulting and Clinical Psychology, 47, 258-264. Shonkoff, J., Dworkin, P., & Leviton. A. (1979). Primary care approaches to developmental disabilities. Pediatrics, 64, 506-514. Spreen, 0., & Gaddes, W. (1969). Developmental norms for fif-
teen neuropsychological tests for ages 6 to 15. Cortex, 5, 171-191. Wechsler, D. (1974). Wechsler Intelligence Scale for ChildrenRevised. New York: Psychological Corporation. Weiner, R. (1983). Number of handicapped students leveling off, ED official says. Educational Handicaps, 9, 1-3. Win, R., Lachar, D., Klinedinst, J., & Seat, P. (1982). The Personality Inventory for Children, Revised. Los Angeles: Westem Psychological Services.
31 Public Policy and Legal Issues for Clinical Child Neuropsychology ROBERT HENLEY WOODY
Clinical child neuropsychology has a myriad of forces that create countervalences for its definition, identity, and placement in human services. While clinical neuropsychology, broadly defined, is still striving to establish itself as capable of fulfilling a function on behalf of human welfare, as would be distinct from neurology, it must also justify being a specialization within clinical, counseling, and school psychology. In tum, clinical child neuropsychology must deal with the same issues, but must also carve out its uniqueness from clinical adult neuropsychology.
Public Policy. Clinical neuropsychology, be it for children or adults, is subject to the vicissitudes of public policy. Public policy represents society's views, preferences, expectations, and demands. Authority to achieve the objectives of public policy is vested in the governmental system: Public policies are developed by governmental institutions and officials through the political process (or politics). They are distinct from other kinds of policies because they result from the actions of the legitimate authorities in a political· system. (Bullock, Anderson, & Brady, 1983, p. 3)
Therefore, the political framework, which is constantly being altered (literally on a day-to-day basis) has a major influence on public policy, which in turn has a major influence on professional practices.
ROBERT HENLEY WOODY • Department of Psychology, University of Nebraska at Omaha, Omaha, Nebraska 68182.
Public Health Policy Of relevance to clinical child neuropsychology, every person's health is influenced by society, and anything that connotes "welfare" seems to meet with ambivalent attitudes from the public (AuClaire, 1984). That is, although the public wants to have a healthy society, there is commonly a reticence to grant social-health-welfare services total "support." As might be obvious, "support" goes beyond endorsement by policy-it requires funding, and this provokes conflicting priorities: The psychology of illness, and the importance that consumers give to their own medical care, make policy fonnulation particularly difficult. . . . There is agreement that frivolous utilization and expenditures should be discouraged, but few patients ever think their own problems frivolous or unworthy of the best care available. (Mechanic, 1981, p. 82)
As detailed elsewhere (Woody, I985b), this is an era wherein: (1) funding for human services has diminished; (2) cost containment is the password to entering the domain of public policy; and (3) health policy planning is plagued by competing and conflicting objectives.
Governmental Regulation One outcome is increased governmental regulatory responsibility and authority over all professionals: "The atmosphere within which perennial issues of access, quality, and costs are considered now involves formal public policies expressed in regulatory programs that are mandated and operated by governments" (Bice, 1981, p. 12). Consequently, all psychological practices, including (of course) by clinical child neuropsychologists, must receive rec573
574
CHAPTER 31
ognition, endorsement, and approval by the regulato- and the Divisions of (among others) Clinical Psyry arm of the body of public policy relevant to health chology, Counseling Psychology, School Psycholcare. ogy, and Clinical Neuropsychology will surely expeAs clinical child neuropsychology develops, a dite the professionalization process for clinical child first order of business must be to attain consonance neuropsychology. Specialty associations, such as the with public policy. Fundamentally, it must be accept- National Academy of Neuropsychologists, will also ed that the preexisting recognition, endorsement, and be facilitative. Nonetheless, public policy has not yet approval assigned to clinical, counseling, or school embraced clinical neuropsychology to a satisfactory psychology may not be automatically applied to degree. clinical neuropsychology. This will be exemplified later in the chapter in the discussion of the negativism toward expert clinical neuropsychological testimony Professional Specialty Credentials that has occurred in certain legal instances (Schwartz, Another means for promoting recognition, en1987). dorsement, and acceptance of a specialty by public policy is through a professional association's grantDeterminations by the Profession ing some sort of specialized credential. In psycholAt this point, the emphasis should be on the ogy, there is the time-honored American Board of profession's inability to self-ordain a psychological Professional Psychology (1984), which has herefunction, role, or specialty. There is no inalienable tofore administered an examination process for right to operate as, say, a clinical child neuropsychol- awarding the status of Diplomate in the areas of ogist-it must be accorded by society. To date, there clinical psychology, counseling psychology, school is some doubt as to just how solidified are the recog- psychology, and industrial/ organizational psycholnitions, endorsements, and approvals via public pol- ogy. More recently, the Board (ABPP) has added icy for neuropsychology generally and clinical child forensic psychology and clinical neuropsychology. The ABPP' s program to award the status of Dipneuropsychology specifically. To earn this positive reception, a profession, lomate in Clinical Neuropsychology is conducted in discipline, or specialty must have a proven track re- conjunction with the American Board of Clinical cord of critiquing the would-be function, definition, Neuropsychology (ABCN). It is still in a fledgling state, and has not received uniform acceptance, even or role: among clinical neuropsychologists. For example, a As society recognizes a profession, it imposes upon rival group, the American Board of Professional that discipline a concomitant responsibility or duty-a Neuropsychology (ABPN), also grants Diplomate set of expectations as to what should and should not status (it is not, however, affiliated with the ABPP). occur in professional practice. In other words, the quid Regrettably, the two neuropsychology boards have pro quo for societal recogniti!)n of professionalism is repeatedly been at odds about standards. Some psyprofessional accountability to society. (Woody, 1985a, chologists interpret these opposing views, allegedly p. 509) in service to establishing standards, as reflecting a This mandate is typically fulfilled by a code of ethics battle for political power over clinical neuropsycholand a set of standards, as would be promulgated by a ogy and/or a skirmish between self-serving personprofessional association. alities. The notion of having a specialty credential in clinical neuropsychology to gain favor from public Professional Standards policy is not necessarily hallowed. Aside from the This introduces a problem for the clinical child possibility that competing politically motivated neuropsychologist. Despite the many (too many?) forces within clinical neuropsychology could potenprofessional associations, there has yet to be a tially damage the specialty to the point of disarray ''homebase'' association for the clinical child neuro- and condemnation from the public and the profession psychologist that provides the needed standards that of psychology (as well as the other health care professions) alike, it could well be that too much specializaare prescribed by public policy. The American Psychological Association has tion could prove to be detrimental to the profesmeritoriously dealt with the issues of ethics (1981a) sionalization of clinical neuropsychology. First, it is feasible that overspecialization could and specialty guidelines for the delivery of clinical psychology (and other specialty) services (1981b), lead to faulty general underpinnings, both in academ-
PUBLIC POLICY AND LEGAL ISSUES
575
ics and in practices. Sarason (1987) raised the ques- . representative for clinical neuropsychology, one tion ''what price do the student and the field pay for noted clinical neuropsychologist said, in effect, specialization," regardless of level. He urged that "Let's face it, having a credential in clinical neurothis question be discussed, and cautioned that such psychology means money, that's why there is such analysis and discussion may be preempted by a competition to become the source of the specialty ''marketplace mentality.'' Sarason sagely asserted: diplomate.'' The wrong motives could backfire on the speThe climate that provides no fonun for serious and cialty of clinical neuropsychology (or any other spesustained discussion is one that encourages early and cialty). Public policy usually reacts to battles within a undue specialization. By undue I mean a degree of profession as reason not to give endorsement. For specialization that phenomenologically makes a part of example, discord between psychological specialpsychology the whole of it, that makes the student an ities, such as clinical psychologists' resisting the inisolationist in the world of psychology. This happens clusion of school psychologists in licensing laws, has unrcflectively and with the best of intentions. For me the question is not whether specialization should be a historically led to legislators' backing away from any postdoctoral affair. The important question is: what governmental support, such as "sunsetting" a licenshould be the core of the identity of a psychologist, sure statute. More certainly, motives that are more regardless of special interest? (p. 37) self-serving for the professional than benefiting to the In fairness to the previously mentioned programs for public could easily usher in a rejection within public the status of Diplomate, there is a prerequisite of policy (such as, hypothetically, having language being a psychologist first and a clinical· neuropsy- within the statutory rules of evidence that would rechologist second (in other words, there are certain strict or negate courtroom testimony by a core areas of psychological training that must predate neuropsychologist). training in clinical neuropsychology), but Sarason's question remains to be answered. The wrong answer could be a harbinger of rejection by public policy. The Interface between Professional Second, it seems feasible that the profession of Ethics and the Law psychology could be damaged by too many specialty credentials. At the risk of sounding like an alarmist, a It is a myth to believe that professional ethical proliferation of credentials for the wrong motive principles supersede the law. In fact, the ethical princould result in a public policy condemnation of the profession overall. As (past) president of the APA ciples of a particular professional association apply Division of Clinical Psychology, Sechrest ( 1985) be- only to members of the association. An exception lieves that the creation of specialties is occurring at a may occur if a state statute codifies a code of ethics, "frightful rate." He reports 30 different groups' pro- such as a statute for the licensing of psychologists posing themselves as specialties (apparently aligned mentioning that the ethics of the American Psychowith clinical psychology). Sechrest acknowledges logical Association (1981a) must be upheld by lithe rationale that specialization credentials can assure censed psychologists in that state. Relatedly, it is a myth to believe that a long-time quality control to safeguard the public, but he notes professional, such as the chair of a state psychologithe motives that special certificates accommodate adcal association's ethics committee, can give auvertising to increase income. He states: thoritative conclusions about an omission or commission . in professional practice. The experienced One suspects also that just sheer ego has something to do with the problem of specialization. Most diplomas, psychologist, such as the chair of an ethics commitat least in psychology, are probably of very little matetee, can offer guidance, but in no way does that opinrial value; they may serve no greater purpose than perion create a legal ruling. It is feasible that, should a suading their possessors that they are in some way spelegal action occur over a professional practice, such a cial. Presumably some warm feeling flows through the voice would be an important contribution to estabpractitioner who can gaze upon the large array of neatly lishing the standard of care-but nothing more framed documents decorating his or her office wall. legally. One of my colleagues has suggested that we go into Professional ethics are important. For the psybusiness manufacturing diploma wallpaper that would chologist, there must be strict adherence to the tenets simplify the whole thing. (p. 1) of the American Psychological Association's ( 1981a) When talking about the furor about which group ethics code. Sanders ( 1983) provided a practical treashould be recognized by the ABPP as the legitimate tise on values and ethics in clinical psychology, which
576
CHAPTER 31
has direct relevance to clinical child neuropsychology. In terms of sanctions, however, an ethics code is, as stated, only applicable to the members of the association.
Standard of Care As the practice of clinical child neuropsychology expands, there will be increasing legal liability. That is, from justifying third-party payments (i.e., getting a patient's clinical neuropsychological services reimbursed by his/her health insurance policy) to allegations of malpractice, the clinical child neuropsychologist will be expected and required to maintain an acceptable standard of care. The term "acceptable" is defined by both the profession and society. As will be discussed shortly, the profession of psychology and, moreover, the specialty of clinical neuropsychology and! or clinical child neuropsychology will have to decide what are or are not to be the endorsed procedures. As mentioned, there is still doubt about what professional source, if any, can speak definitively to the issue of standards for the practice of clinical child neuropsychology. Society remains the final arbiter for the choice of standards.
The Legal Test Society expresses its decision through laws. Legally, it is well established that a profession's endorsed conduct will get the benefit of the doubt, but an entire profession could be negligent, for example, have a standard of care that was unacceptable to public policy. The legal test is as follows: It thus is fundamental that the standard of conduct which is the basis of the law of negligence is usually determined upon a risk-benefit form of analysis: by balancing the risk, in the light of the social value of the interest threatened, and the probability and extent of the harm, against the value of the interest which the actor is seeking to protect, and the expedience of the course pursued. For this reason, it is usually very difficult, and often simply not possible, to reduce negligence to any definite rules; it is ''relative to the need and the occasion,'' and conduct which would be proper under some circumstances becomes negligence under others. (Keeton, Dobbs, Keeton, & Owen, 1984, p.
173)
Stated differently, just because the entire profession does or does not do a particular act (service), does not justify it.
As a hypothetical (but quite likely) example, suppose that all (or most) of the practitioners of clinical child neuropsychology did only a short form of a particular neuropsychological test battery, even though it was known to reduce substantially the reliability and validity of the results. Assuming the expenditure (e.g., professional time and costs to the patient was not "excessive" (which has no predetermined definition) for and the benefits were substantially greater from the long form of the test (e.g., reduced risk of diagnostic error), the practitioners of clinical child neuropsychology could potentially all be negligent by using the short form of the neuropsychological test battery. Common practice within a profession does not establish the standard of care.
The Reasonable Practitioner Test When determining the .standard of care, the court applies the "reasonable person" test. In this instance, the test becomes the "reasonable clinical child neuropsychologist.'' There is no exact prototype to which the clinical child neuropsychologist can compare his/her qualities. Often the term "prudence" is integrated in the conceptualization of "reasonableness." "Prudence'' is but a short step away from ''conservative,'' which in tum could move to "traditional," which could lead to the notion that the standard of care is counter to innovation. In legal reasoning, there is a certain truth to the notion that tradition is favored over innovation. Innovation cannot be foolhardy or create an unreasonable risk to the patient. In opposition, public policy holds that a scientist-practitioner should, for the welfare of society, strive for advancement in competency, and thus innovation should be encouraged and legally protected-within reason. Generally speaking, if the professional has performed an innovative technique or procedure that was predicated upon a reasonably sound theoretical basis, executed it with good intentions and logic, maintained precautions and safeguards, and subjected his/her work to professional scrutiny (e.g., a research-review committee), there will be an effort to uphold the innovative dimension. The "reasonable person" test does not require superiority per se: ''Professional persons in general, and those who undertake any work calling for special skill, are required not only to exercise reasonable care in what they do, but also to possess a standard minimum of special knowledge and ability'' (Keeton
PUBLIC POLICY AND LEGAL ISSUES
et al., 1984, p. 185). If the issue is breach of the standard of care, such as in a malpractice action: The formula under which this is usually put to the jury is that the doctor must have and use the knowledge, skill and care ordinarily possessed and employed by members of the profession in good standing; and a doctor will be liable if harm results because he does not have them. Sometimes this is called the skill of the ''average'' member of the profession; but this is clearly misleading, since only those in good professional standing are to be considered; and of these it is not the middle but the minimum common skill which is to be looked to. If the defendant represents himself as having greater skill than this, as where the doctor holds himself out as a specialist, the standard is modified accordingly. (Keeton et at., 1984, p. 187)
Perfection and/or superiority are not required, but this legal formula does require that the clinical child neuropsychologist be in the mainstream of competency for the specialty. The foregoing introduces three issues unique to clinical child neuropsychology: (1) the neuropsychological theory or procedure; (2) the group of child neuropsychologists with whom there would be a comparison; and (3) the effects of having a specialist designation, such as considering oneself a "child neuropsychologist" as opposed to, say, a "clinical child psychologist" or "school psychologist."
Neuropsychological Theory or Procedure Public policy is tolerant of disagreements between experts, as long as the differing views have a reasonable semblance of being birthed by professional seeds. In clinical neuropsychology, there is rampant disagreement about: ( 1) the theoretical explanation of neuropsychological structures and processes; and (2) the assessment and treatment techniques most suitable for particular neuropsychological conditions. In his discussion of the development of theories of brain function, Golden (1978) provided a historical trace up to localization theory and equipotential theory, and noted: "Unable to accept either the localizationist or equipotentialist models of brain function, psychologists and neurologists have searched for other models" (p. 8). When considering the diverse views expressed at the typical conference of clinical neuropsychologists, it is tempting to assert that there is a model unique to each practitioner. Legally, although the clinical child neuropsychologist will be supported in his/her professional right to be aligned with a given theory of brain func-
577
tion: "This does not mean, however, that any quack, charlatan or crackpot can set himself up as a 'school,' and so apply his individual ideas without liability" (Keeton et al., 1984, p. 187). By legal prescription, a professional will only receive deference from public policy on this matter if he/ she espouses a theory or "school" that has earned professional respect, has a set of definite principles, and is based on a line of thought that a reasonable number of qualified professionals would share. The latter means that any clinical child neuropsychological view must be based on research, as would be compatible with the scientist-practitioner model. Although apparently not yet subjected to litigation in the area of clinical child neuropsychology, a legal analysis of cases relevant to alleged psychotherapy malpractice supports that nontraditional approaches would likely be tested against more traditional schools of therapy (Glenn, 1974). Going from theory to procedures, public policy requires that psychologists be qualified by, among other factors, academic training according to standards maintained for the profession. For example, state licensing statutes consistently require a doctorate from a regionally accredited institution of higher education. Some states require more; for example, Florida requires would-be psychologists to have: Submitted proof satisfactory to the board that he has received a doctoral degree with a major in psychology from a university or professional school that has a program approved by the American Psychological Association or that he has received a doctoral degree in psychology from a university or professional school maintaining a standard of training comparable to the standards of training of those universities having programs approved by the American Psychological Association or the doctoral psychology programs of the state universities. (Florida Statutes, 1985b, p. 1510)
To be deemed a clinical child neuropsychologist, the principle would seemingly be having fulfilled training and practice experiences compatible with specialty. As mentioned, there is no definite source to date, but the various specialty groups, such as (but not limited to) the ABPP, hold the potential for delineating specific training and practice experiences that would be necessary if the clinical child neuropsychologist is to anticipate a protective legal framework. Again there is professional debate about clinical neuropsychology procedures. For example, there is contradictory research evidence and clinical views about many (all?) neurological conditions (Walsh, 1978). Despite all the hullabaloo about the left brain versus the right brain, the research is, by far, in-
578
CHAPTER 31
conclusive about numerous issues (Springer & Deutsch, 1985). Perhaps the two most famous (or infamous?) debates center around: (1) whether the clinical neuropsychologist should make use of an individualized set of tests or a standardized neuropsychological battery; and (2) if preference is given to a standardized battery, whether it should be given to the Nebraska-Luria or the Halstead-Reitan. Often it seems that there is a tendency to base one's preference on the shortcomings of the alternatives, which hardly seems the most professional way of garnering support from public policy.
criteria by which one psychologist would be compared to another. The prevailing approach is to: (1) consider the qualities of the targeted practitioner (e.g., holding a Ph.D. degree. licensed as a psychologist in the state, X number of years of clinical experience, and so on); and (2) attempt to pinpoint a range of qualities that could be reasonably expected. Obviously such a comparison is based on a fiction: each practitioner is unique; but public policy holds that this exercise must be fulfilled in legal proceedings.
Specialization Comparison to Other Clinical Child Neuropsychologists
An exception to the foregoing occurs when the practitioner asserts, or his/her patient has a reasonable basis for believing, that he/she has special expertise. When this happens, the determination for whether or not an appropriate standard of care has been maintained tends to based on a nationwide comparison, with the comparison group being comprised of experts of the same ilk. For declaring oneself a "clinical child neuropsychologist," as opposed to, say, a "clinical child psychologist" or "school psychologist," the title alone would likely elevate the standard of care (or competency) that could be reasonably expected by recipients of the services. Although there may not be an explicit assertion of "greater skill" by the practitioner, the legal perspective would probably be that use of a special title makes an implicit promise of special or greater expertise, as compared to those practitioners in the same discipline who lay no claim to a special title. By legal principle, any certificate, such as being a Diplomate in Clinical Neuropsychology, is a definite statement of expertise. If a comparison with other clinical neuropsychologists is to be made, it would be with those who have the same (or comparable) certificate. It is important to note that in some instances an implied credential, if reasonably derived by the patient, can cast the standard of care to a credential that the practitioner does not hold. Therefore, the prudent approach is to studiously avoid fostering any notion in the.mind of a patient (or anyone else) that is not truly compatible with one's qualifications.
In the "good old days," any professional comparison was made at the local level: ''Formerly it was generally held that allowance must be made for the type of community in which the physician carries on his practice, and for the fact, for example, that a country doctor could not be expected to have the equipment, facilities, libraries, contacts, opportunities for learning, or experience afforded by large cities" (Keeton et al., 1984, pp. 187-188). Today, with the advances in communication, travel, sources of research (e.g., the plethora of professional journals), and so on, the "locality rule" is far less important. Instead, a comparison is based on the qualifications and practices of the particular professional. It is important to recognize, however, that states (and even particular courts therein) differ in comparative criteria. For example, some states still require that a psychologist testifying in a malpractice case have a reasonable knowledge of the psychological standards and practices in the particular state or jurisdiction. Today the foremost source for deriving a comparative framework would likely be the standards promulgated by the American Psychological Association, especially as they receive more regional or local endorsements through state licensing boards of psychological examiners. For example, the Florida Psychological Association recommended that the Florida Board of Examiners (of psychologists) adopt the provisions of the American Psychological Association's (1981b) "Specialty Guidelines for the Delivery of Services.'' In so doing, a national standard Malpractice was used to create a standard of care for psycholThe standard of care is the critical component of ogists in the state of Florida; these standards essentially prescribe and proscribe the contents for the a malpractice legal action. This chapter is not
PUBLIC POLICY AND LEGAL ISSUES
579
intended to be a treatise on malpractice. An elabora- chological test battery and/or the inept interpretation on and guidelines for avoiding malpractice in tion). Neither of these faults would necessarily have a mental health services are available elsewhere causal connection to an injury. Suppose, however, (Woody, 1983, 1988a,b). Whatisappropri ateherein that a more astute diagnostic evaluation would have is to specify the criteria that are used for determining led to a treatment intervention that would have prewhether or not a malpractice action is appropriate, vented an exacerbation of the brain-related problem. noting the current malpractice scene for psychol- In other words, the fact that the evaluation did not ogists, and considering the relevance to clinical child detect the neuropsychological condition led to a deneuropsychology. lay in treatment intervention, and the passage of time led to incre8sed severity. Thus, there would be a causal connection to injury. Yet, there must be proof The Negligence Formula of the fourth element: actual loss or damage. Damages are supposed to be compensatory. The ''Malpractice'' is the popular term for professional negligence. In order for a cause of action to be injured patient is to be restored his/her preinjury confounded on negligence, there must be four elements dition by the award of a monetary remedy. Unless prohibited by a state statute, some courts will allow present: punitive damages, such as to teach an entire profes1. A duty, or obligation, recognized by the law, resion a lesson. If there is proof of an injury, damages quiring the person to confonn to a certain standard of are assumed, such as pain and suffering (past, preconduct, for the protection of others against unreasonsent, and future). Damages cannot be unreasonably able risks. speculative, there must evidence, such as in the form 2. A failure on the person's part to conform to the of expert testimony, of the nature and extent of the standard required: a breach of the duty. . .. injury and how much damage there has been, is cur3. A reasonably close causal connection between the conduct and the resulting injury. This is what is comrently being, and/or will be inflicted upon the pamonly known as "legal cause," or "proximate tient's life. As an example, it is possible that a cause,'' and which includes the notion of cause in fact. clinical child neuropsychologist could have a duty to 4. Actual loss or damage resulting to the interests of his/her patient, breached the standard of care by another.... (Keeton et al., 1984, pP. 164-165) faulty services, and caused the patient to suffer longIn applying this formula to clinical child neuropsy- er than was ideally necessarily-an d still the patient chology: ( 1) it seems obvious that a clinical child would not experience nor be able to prove damages to neuropsychologist has a "duty, or obligation" to the point of receiving more than, say, a nominal each of his/her patients to conform to the standard of amount. conduct or care relevant to professional practice; (2) a breach of that duty is established by proof, such as by Malpractice Actions against Psychologists testimony from another clinical child neuropsychologist, that there was an omission or commission of a Whereas psychologists once enjoyed a certain procedure that did not meet the standard; (3) having a kind of relationship with their patients that served to duty and a breach are not enough-the breach must minimize legal liability, their increased identity as have a ·'causal connection'' to the alleged injury; and health care providers leads to greater patient willing(4) even with the three previously mentioned ele- ness to file a legal action (Knapp & Vandecreek, ments, the patient must have incurred damages. 1981). The medical context in which clinical neuroFor example, suppose that the clinical child psychologists function would certainly lead to inneuropsychologist allegedly failed to diagnose a par- creased legal liability. ticular brain-related problem with a patient. There Further, the principle of vicarious liability holds was surely a duty to the patient. Assume that the that: ''One who is free from all moral blame or legal clinical child neuropsychologist did, in fact, fail to fault is held liable for the tort of another, and this may administer certain subtests of a neuropsychological be described as a form of liability without fault" test battery that would have revealed the brain-related (Keeton et al., 1984, p. 593). For the clinical child problem, or that he/she had the test data, but failed to neuropsychologist, this means that being a neuinterpret the data properly. Other clinical child rological "team member" could result in being neuropsychologists migltt well be willing to provide named a defendant, even though the direct cause of expert testimony about the poor quality of the diag- the injury was allegedly due to the negligence of, say, nostics (the faulty administration of the neuropsy- the neurologist or another health care provider.
580
CHAPTER 31
Psychologists are being subjected to an increasing number of lawsuits. The causes of action are many, including (but not limited to) "involuntary servitude, false arrest, trespass, malicious infliction of emotional distress, abuse of process, loss of liberty, misrepresentation, libel, assault and battery, malicious prosecution, and false imprisonment" (Hogan, 1979, p. 7). Wilkinson (1982) categorized malpractice cases involving psychiatrists as patientinflicted injuries and suicide, harm by the patient to third persons, errors in judgment, faulty treatment methods, drug-related liability, and sexual misconduct. Fisher (1985) indicated that "the number of claims against psychologists have risen faster in the past three years than for any other mental health profession" (p. 6). There is no reason to believe that clinical child neuropsychologists would be able to find any exemption from or exception to the increasing liability; indeed, as stated, the medical context probably increases the liability for clinical neuropsychologists beyond the level of liability generally attributed to psychologists.
will lead to liability for both the employing organization and the professional-it is highly improbable that either defendant could find immunity because of the context in which the service was performed. The concept of immunity is important to psychologists in the schools (as well as in other nonprofit organizations). Often an employer, say a school administrator, will glibly assure the employees, say school psychologists, that they will be covered by the organization's liability insurance, when the fact of the matter is that only the employer is covered by the insurance for the commissions or omissions of the employee. Any relevant legal action can potentially name the employee individually, and a judgment could be rendered against the employee that would not be covered by the employer's insurance. Therefore, every clinical child neuropsychologist, wherever employed, would be well advised to obtain a carefully drafted statement of liability coverage and indemnification for any claims or legal actions that are associated with the employing source (this would, hopefully, cover the vicarious liability issue as well).
Immunity Because many practitioners of clinical child neuropsychology are employed as school psychologists, there might be the notion that any malpractice legal action would be directed at the school system/ employer, and that there would be no personal liability. Although an attorney would likely draft the pleadings to include the employer, under the common-law principle of "Master/Servant" liability (whereby the master/ employer may be held accountable for the actions or inactions of his/her servant/employee), this does not, in any manner, exempt the employee from personal liability. In the past, public policy supported that certain persons or organizations should be immune from suit or liability. In other words, the public policy reasoning held that there would be benefits from society's granting immunity to a designated class, for example, governmental officials or charitable organizations. Without going into a detailed exposition of the legal theory and cases, suffice it to say that, at best, the concept of immunity has been eroded. Indeed, many legalists would probably assert that immunity no longer exists, certainly not for a health care practitioner because of the employment source. For example, the duty to warn others of the physical dangerousness of a patient supersedes privileged communication, and failure to fulfill the duty to warn
Expert Clinical Child Neuropsychological Testimony To provide expert testimony, a witness must have qualifications that will allow him/her to offer special information to the trier of fact (the judge and/or jury). Although there may be slight differences between state statutes, consider the definition from Florida: Testimony by experts. If scientific. technical, or other special knowledge will assist the trier of fact in understanding the evidence or in detennining a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training, or education may testify about it in the fonn of an opinion; however, the opinion is admissible only if it can be applied to evidence at trial. (Florida Statutes, 1985a, p. 367)
At first glance, it would seem like surely the testimony of a clinical child neuropsychologist would fulfill the requirements for being "expert" in nature. Such may or may not be the case. For many years, only a medical physician could testify about the human condition. As the behavioral sciences developed, public policy was altered to accommodate professional disciplines other than medicine. Now it is commonplace for, among others, psychologists to be qualified for rendering expert
PUBLIC POLICY AND LEGAL ISSUES
581
For the foregoing and other reasons, certain courts have excluded testimony by clinical neuropsychologists on neurological conditions; or if the clinical neuropsychologist has been qualified to testify, the judicial opinion points toward very little weight being accorded to the neuropsychological data. Schwartz (1987) analyzed two court decisions in the Florida Court of Appeals that negated and/or restricted the testimony proferred by a clinical neuropsychologist. As there are only limited cases that are on point (and the rulings may be idiosyncratic to the case, the judge, orthejurisdiction), areviewofthose cases seems premature for purposes of deriving legal principles. Suffice it to say that there has not been unreserved judicial support for clinical neuropsychological testimony. It seems probable that childhood development, with its individual differences, and the malleability of children will result in the reservations about expert testimony on adult neuropsychology being as staunch (or more so) for expert testimony on child neuropsychology. The principal message for the child clinical neuropsychologist would certainly be: "Forensic cases have demanded an increasing sophistication on the part of the neuropsychologist beyond those issues encountered with other types of psychological testimony" (Golden, 1986, p. 1). To the contrary, there have been many instances where the testimony of the clinical neuropsycholoAll that produces cognitive dysfunction is not necesgist, be it for an adult or child, has been readily sarily, primarily, or even secondarily, neurological in accepted and accorded decisive weight by the court. nature. Antecedent factors, such as disruptive environGalaski (1985) described the present-day relaments, poor opportunity, primary emotional pathology tionship between the attorney and clinical neuropsynot tied to neurological dysfunction, all can result in chologist, and indicated: ''Cases in which the neurocompromised mental performance. These deficits psychologist can be of greatest practical help to the may, in some circumstances, appear strikingly similar attorney include cases of personal injury, especially to those which are likely to be produced by neuwhen there has been head trauma caused by a fall or rological events themselves. (p. 480) sustained in a motor vehicle accident'' (p. 10). lncagA clinician must be attuned to both neurologically noli (1985) described how clinical neuropsycholobased and nonneuro1ogically based diagnoses. gists can fulfill a valuable role in litigation, and she Related to the preceding point, there is often a recommended that: ''A clinical neuropsychologist fine line between scientific assertions and proselytiz- should be called as an expert witness to evaluate any ing for a cause that will benefit the professional. With client with suspected or known brain injury'' (p. 60). any emerging strategy, and clinical neuropsychology Related to the previous comments about the resis no exception, self-serving motives may lead to ervation of courts to admit the testimony of clinical ''professional enthusiasm'' that is unjustified. Le- neuropsychologists, the ever-expanding recognition galists are aware of this tendency in professionals. of the usefulness of clinical neuropsychological data One trial lawyer commented about a deposition that for legal determinations creates a press for spehe had taken of a well-published clinical neuropsy- cialized research on applying neuropsychological chologist, saying, "It should be easy to impeach his procedures to legal issues. Lanyon (1986) emphatestimony during the trial because he comes across as sized that psychological assessment, in general, for a 'True Believer' in the infallibility of the neuropsy- legal cases must go beyond the typical clinical prochological tests.'' cedures. He provided a useful review of specialized testimony (Jenkins v. United States, 1962). Having said this, however, the area of clinical neuropsychology has not yet received unequivocal acceptance. Part of the reservation comes from the belief of many legalists that a patient's neuropsychological condition can only be adequately evaluated, diagnosed, described, and/or treated by a neurologist. Another way of phrasing the problem is that neuropsychological tests have yet to receive adequate documentation for reliability and validity. In point of fact, there is still a dearth of supportive research for clinical neuropsychological judgments. For example, Shordone and Rudd ( 1986) had psychologists examine clinical vignettes for purposes of recognizing neurological disorders, and found that one-third of the psychologists failed to recognize the underlying neurological disorder. Likewise, the use of individualized test batteries allows for personal preferences by the clinical child neuropsychologist. Consequently, the subjectivity in the choice of methods or instruments can lead to questionable validity for the decisions derived from the data, which may be compounded by the limitations (e.g., inadequate standardization) of the tests (Golden, 1986). As another weak point, misdiagnosis can result from an overcommitment to clinical neuropsychological strategies. As Boll (1985) stated:
582
CHAPTER 31
procedures/instruments developed for such legal issues as competency, insanity, dangerousness, child custody evaluations, homicide, and sex offenders. Regrettably, it appears that there has been precious little research on the application of clinical neuropsychology to legal issues per se, and few, if any, specialized procedures/instruments within the realm of clinical neuropsychology that are tailored to legal issues. For example, an appropriate legal query could be: What research findings have been obtained to connect certain data from children's neuropsychological tests to a specific degree/amount of damages? The answer would likely be: None.
Summary Public policy has issued an invitation to clinical child neuropsychology to enter into human service. In so doing, there will be exacting expectations, yet the practitioner has, to date, ill-defined guidelines. The nebulous framework for practice combines with this litigious era to sound a caveat to clinical child neuropsychologists for reasonable ethical and legal functioning. It would be foolhardy to plunge recklessly into a new specialty-instead, the clinical child neuropsychologist must predicate all practices on a carefully studied rationale based on academics, research, and human need.
References American Board of Professional Psychology. ( 1984). Policies and Procedures for the Creation of Diplomates in professional psychology. Columbia, MO: Author. American Psychological Association. (1981a). Ethical principles of psychologists. American Psychologist, 36, 633-638. American Psychological Association. (1981 b). Special guidelines for the delivery of services. American Psychologist, 36, 639685. AuClaire, P. A. (1984). Public attitudes toward social welfare expenditures. Social Work, 29, 139-144. Bice, T. W. (1981). Social science and health services: Contributions to public policy. In J. B. McKinlay (Ed.), Issues in Health Care Policy (pp. 1-28). Cambridge, MA: MIT Press. Boll, T.J. (1985). Developing issues in clinical neuropsychology. Journal of Clinical and Experimental Neuropsychology, 7, 473-485. Bullock, C. S., Ill, Anderson, J. E., & Brady, D. W. (1983). Public Policy in the Eighties. Belmont, CA: Wadsworth. Fisher. K. (1985, May). Charges catch clinicians in cycle of shame, slip-ups. APA Monitor, 16(5), 6--7. Florida Statutes. (1985a). Chapter 90, Evidence (see 90.704), 367.
Florida Statutes. (l985b). Chapter 490, Psychological Services (see 490.005 (I) (b)), ISlO. Galaski, T. (1985). The neuropsychologist: Key member of the doctor-lawyer team. Case & Comment, 90(4), 10, 12-14. Glenn, R. D. (1974). Standard of care in administering non-traditional psychotherapy. University of California, Davis Law Review, 7, 56-83. Golden, C. J. (1978). Diagnosis and Rehabilitation in Clinical Neuropsychology, Springfield, IL: Thomas. Golden, C. I. (1986). Forensic neuropsychology: Introduction and review. In C. 1. Golden & M. A. Strider (Eds.), Forensic Neuropsychology (pp. 1-47). New York: Plenum Press. Hogan, D. B. (1979). The Regulation of Psychotherapists (Vol. ll)). Cambridge, MA: Ballinger. Incagnoli, T. (1985). Clinical neuropsychologists: Their role in litigation. Trial, 21(6), 60, 62-63. Jenkins v. United States. (1962). 307 F.2d 637. Keeton, W. P., Dobbs, D. B., Keeton, R. E., & Owen, D. G. (1984). Prosser and Keeton on the Law ofTorts (5th ed.). St. Paul, NM: West. Knapp, S., & Vandecreek, L. (1981). Behavioral medicine: Its malpractice risks for psychologists. Professional Psychology, 12, 677-683. Lanyon, R. I. (1986). Psychological assessment procedures in court-related settings. Professional Psychology. 17, 260-
268. Mechanic, D. ( 1981 ). Some dilemmas in health care policy. In J. B. McKinlay (Ed.), Issues in Health Care Policy (pp. 8094). Cambridge, MA: MIT Press. Sanders, 1. R. (1983). Values and ethics in clinical psychology. In C. E. Walker (Ed.), The Handbook of Clinical Psychology (Vol. II, pp. 1328-1350). Homewood, IL: Dow JonesIrwin. Sarason, S. B. (1987, January). Is our field an inkblot? APA Monitor, 18(1), 37. Schwartz, M. L. (1987). Limitations on neuropsychological testimony by the Florida appellate decisions: Action, reaction, and counteraction. Clinical Neuropsychologist, 1, 5160. Sechrest, L. B. (1985). Specialization. Who needs it. Clinical Psychologist, 38, I, 3. Shordone, R. 1., & Rudd, M. (1986). Can psychologists recognize neurological disorders in their patients? Journal of Clinical and Experimental Neuropsychology, 8. 285-291. Springer. S. P., & Deutsch, G. (1985). Left Brain, Right Brain (rev. ed.). San Francisco: Freeman. Walsh, K. W. (1978). Neuropsychology: A Clinical Approach. Edinburgh: Churchill Livingstone. Wilkinson, A. P. (1982). Psychiatric malpractice. Identifying areas of liability. Trial, 18(10), 73-77, 89-90. Woody, R. H. (1983). Avoiding malpractice in psychotherapy. In P. A. Keller & L. B. Ritt (Eds.), Innovations in Clinical Practice: A Sourcebook (Vol. II, pp. 205-216). Sarasota, fl.: Professional Resource Exchange. Woody, R. H. (and Associates). (1984). TheLawandthePractice of Human Services. San Francisco: Jossey-Bass. Woody, R. H. (l985a). Public policy, malpractice law. and the mental health professional: Some legal and clinical guidelines. In C. P. Ewing (Ed.), Psychology, Psychiatry, and the
PUBLIC POLICY AND LEGAL ISSUES Law (pp. 509-525). Sarasota, FL: Professional Resource Exchange. Woody, R. H. (1985b). Techniques for handling psycholegal cases. In C. E. Walker (Ed.), The Handbook of Clinical Psychology (Vol. II, pp. 1420-1439). Homewood, IL: Dow Jones-Irwin.
583
Woody, R. H. (1988a). FiftyWaystoAvoidMalpractice:A Guidebook for Mental Health Professionals. Sarasota, FL: Professional Resource Exchange. Woody, R. H. (1988b). Protecting Your Mental Health Practice: How to Minimize Legal pnd Financial Risk. San Francisco: Jossey-Bass.
32 Training and Credentialing in Child Neuropsychology LAWRENCE C. HARTLAGE AND CHARLES J. LONG
Background Although neuropsychology as a scientific field of inquiry has origins dating at least as far back as the late 19th century, it is only during the past 20 years that neuropsychology has enjoyed widespread recognition and acceptance as a formal applied professional specialty area. Until recently, neuropsychology was primarily identified with diagnostic testing of adults with verified brain injury. With the increasing recognition of neuropsychological substrates of learning and adaptive behavior problems in adults with brain injury, there developed a progressive interest in some possible central processing dysfunctions as being etiologic in a wide variety of children's learning problems (e.g., Chalfant & Scheffelin. 1969). Given impetus and support by the focus of "The Great Society" programs on identification, description, and treatment of childhood learning problems, neuropsychology increasingly was involved with the assessment of exceptional children. The growing involvement of neuropsychology with children's problems raised a number of scientific and professional questions and issues. As the body of research relating known brain damage to specific learning and behavior problems had for the most part involved adults, one obvious scientific question involves the extent to which this research could be applied to children. Stemming from this scientific question arose a professional issue, namely, which tests or diagnostic approaches are appropriate for use with children. If findings from adults LAWRENCE C. HARTLAGE • Department of Psychology, University of Arkansas, Fayetteville, Arkansas 72701. CHARLES J. LONG • Psychology Depanment, Memphis State University, Memphis, Tennessee 38152.
could be applied directly to children, then presumably a downward extension of a battery appropriate for use with adults might be adequate for this purpose. Conversely, if findings from adult neuropsychology could not be applied to children, it would be necessary to develop a new data base for application to child neuropsychology. Another scientific question dealt with whether findings from individuals with known brain damage verified on neurological, neurosurgical, or neuroradiological criterion measures, could be applied to children who were presumed to have neuropsychological impairments on the basis of neuropsychological assessment, but for whom there was no definitive evidence of structural or physiological damage. This scientific question translated into obvious professional issues. Because for many children whose neuropsychological examination findings suggested a clear central nervous system dysfunction, there was no external criterion that could validate such an impression, the misclassification of such children as "brain injured" could adversely influence their educational programming and management.
Assessment Approaches. In response to the demand for neuropsychological services for children, and in attempts to address the scientific and professional issues raised by this demand, two diverse approaches to provision of neuropsychological services to children emerged. One approach involved modified versions of traditional neuropsychological batteries such as the Halstead-Reitan (Reitan, 1955; Reitan & Davison, 1974; Selz, 1981) and the LunaNebraska Neuropsychological Battery (Golden, 1981; Plaisted, Gustavson, Wilkening, & Golden, 1983; Golden, Hammeke, & Purisch, 1980), which standardized the adult battery items on a child sample. For the most part, this standardization took the 585
586
CHAPTER 32
form of deleting from the adult battery those items that were too difficult for children. There is reportedly good congruence between the adult and child batteries on classificatory accuracy, and also between the Reitan-Indiana Children's Battery and the Luna-Nebraska Neuropsychological Battery for Children (Geary, Schultz, Jennings, & Alper, 1984; Berget al., 1984; Golden et al., 1981). Even among proponents of a standardized battery approach, there is disagreement concerning which battery is best for which population of patients (e.g., Adams, 1980a,b; Spiers, 1981). The second emphasis is on interpretation of standard psychometric tests from a neuropsychological perspective, augmented by some measures of sensory and motor function, using relevant age-appropriate tests for children of given ages, ranging from preschool through adolescent ages (Hartlage, 1981, 1984; Hartlage & Telzrow, 1983; Telzrow & Hartlage, 1984). This approach uses standardized behavioral tests and interprets them according to the individual's strengths and weaknesses and in some cases makes inferences regarding neurological integrity. Such an approach is popular with psychologists working in school settings, and in many cases may be adequate for child neuropsychological assessment. Although there is little evidence that one approach is clearly superior, "turf skirmishes'' often center on the issue of qualifications. Psychologists who have developed expertise in the use of a given neuropsychological test battery tend to support the view that the only legitimate neuropsychologists are those with a similar background and expertise. Psychologists who espoused diagnostic approaches involving traditional psychometric tests counter by questioning the relevance of a standardized battery developed for adults with known brain lesions for assessing children who often do not have evidence of brain lesions. They also question the redundancy involved in adding a standard neuropsychological battery to the array of psychometric instruments required by most school districts for psychoeducational assessment. The second approach appears to be preferred by most professionals. A survey of internship training programs suggests that most professionals prefer the second approach (in- . terpretation of standard psychometric tests from a neuropsychological perspective) (Goldberg & McNamara, 1984). In such settings, 78% employ nonstandardized assessment strategies, 63% the Halstead-Reitan, and 35% the Luna-Nebraska. Even those individuals employing a neuropsychological test battery frequently augment the battery with common psychological tests.
Professional Context of Child Neuropsychology It has been argued that neuropsychodiagnosis has little or no relevance to education and/ or rehabilitation; however, with the advent of the CAT scan, NMR, and other instruments, neuropsychological assessment has shifted away from simple yes/no "organic" diagnosis as a primary endeavor and has moved toward comprehensive assessment of cognitive skills relevant to planning for intervention.
Levels of Inference An important issue in training and credentialing in child neuropsychology involves the purposes for which neuropsychologically relevant data are to be used. A comparatively low level of inference involves a conclusion that impaired brain function may be etiologic or at least contributory to a given problem. An example of this level of inference might be a conclusion reached by a school psychologist that a child's failure to acquire a given academic skill is likely related to brain damage or dysfunction. At a considerably higher level of inference are diagnostic statements indicating specific localizing and etiologic phenomena. An example of this level of inference might be a statement, reached by a clinical child neuropsychologist working in a neurological setting, that a child appears to have an astrocytoma confined to anterior portions of the nondominant cerebral hemisphere. Perhaps the highest level of inference involves statements concerning some irreversible intervention. An example of this type of inference might involve a clinical child neuropsychologist working in conjunction with a pediatric neurosurgeon, who concludes that removal of a major portion of a child's hippocampus will not impair memory or other mental function. Between these low and high levels of inference occur many intermediate levels involving such matters as optimal instructional mode, referral to a neurological specialist, prognostic statements based on inferred level of cortical integrity, or conclusions concerning whether (or the extent to which) a child's impaired cognitive performance may be due to an injury for which legal action is pending. It is possible that a well-trained clinical child psychologist or school psychologist, with only moderate training (or credentials) in child neuropsychology, may make appropriate lower-level inferences concerning brain-behavior relationships. For exam-
TRAINING AND CREDENTIALING
pie, a school psychologist may, by training, experience, and clinical skill, be quite adequately prepared to develop perfectly appropriate academic intervention programs for a child with a chronic or acquired neurological impairment, and precluding such individuals from such practice on the grounds that they are not sophisticated in brain-behavior relationships may serve to deprive a child of such a valuable professional resource. Conversely, it is not reasonable to expect such a professional to detect manifestations of an early stage neurodevelopmental disorder or a neoplasm of some slowly progressive type. On the one hand, it can be argued that, until the proper diagnosis is made, it is not possible to determine what level of inference may be required: this might suggest that all questions concerning possible brain involvement in children require the involvement of a qualified child neuropsychologist. On the other hand, in a typical school population, the base rate of neurodegenerative or slowly progressive neoplastic disorders is sufficiently low that such a requirement may be considered to be unrealistic. Interactive with the level of inference is the issue of potential harm to the child. Some chronic neurological conditions, such as might be represented by chronic cerebral hemispheric functional asymmetry, can conceivably be overlooked without necessarily causing major problems. In cases where appropriate educational and counseling services are provided, overlooking the neurological substrates of uneven levels of academic performance may be only minimally handicapping to the child. Conversely, labeling the child ''brain damaged'' may deprive the child of needed educational support. Similarly, the mismatch between a child's neuropsychologically mediated abilities and deficits in an ongoing educational program that does not take these factors into account may cause harm to the child, both in terms of frustration and failure to achieve academically at ability levels. Obviously, higher-order inferences regarding brain-behavior relationships should only be made by individuals whose training, experience, and clinical skills qualify them for such inferences. Although guidelines concerning training and credentialing can and should address these issues, it is not reasonable to hope that such guidelines can resolve them all.
Credentialing of Psychologists Although the study of developmental brain-behavior relationships is a relatively recent endeavor in
587
neuropsychology (Dean, 1982), it has already been argued that there is a need for some type of credentialing and certainly for more specialized training if one is to provide appropriate neuropsychological services to children. Clinical psychologists have traditionally tended to function as generalists-setting few limits regarding credentialing and developing no formal method for identifying a particular area of expertise. They tend not to limit their practice to a specific problem area or specific age group (VandenBos, Stapp, & Kilburg. 1981). This state of affairs no longer appears appropriate for the current practice of psychology due to the dramatic change in the knowledge base. Certainly it is clear that neuropsychological assessment requires specific knowledge, not gener~ any obtained in traditional clinical psychology training programs. Furthermore, the techniques and issues in child psychology cannot simply be deduced from knowledge of adults. Specialty training in school psychology and specialization in child psychology also speak to the changes in training promoted to meet the needs of the child. Although only four specialty areas initially were recognized by the American Psychological Association, a recent review in credentialing activities by Sales ( 1985) identified 31 specialty credentialing boards. Even though psychologists are identifying areas of specialization and devising procedures for membership inclusion, clinical psychologists seem reluctant to limit their practice by establishing formal specialties within clinical psychology. In the absence of credentialing, control is left to licensing activities, done at the state levels, resulting in a wide variety of requirements for practice in a specified area such as neuropsychology, and a tendency to rely on the individual practitioner with respect to not making professional judgments at levels of inference for which the practitioner is not qualified. As has just been noted, however, in cases involving some neuropsychological problems, an otherwise well-trained clinican may not recognize the neuropsychological nature of the problem, and at the same time feel justified by avoiding any inferential statements concerning CNS involvement. Although it could be argued that, in such a case, making no inferential statement concerning CNS deficit may in fact be inferring something about CNS integrity, such activities are extremely difficult to control within limitations of generic state licensing laws. A national credentialing board, not limited by whims and caprices of legislators who enact and amend licensing laws at the state level, is an accepted
588
CHAPTER 32
approach toward ensuring some level or degree of competence among practitioners who have met the requirements of that board. With credentialing requirements set by professionals, this obviously represents an approach with considerable potential for helping ensure such competence. As participation in the activities required for credentialing is entirely voluntary (and can entail a fair amount of energy, frustration, and money), there is no assurance that the only qualified neuropsychology practitioners are those who are board certified. As with generic state licensure, board certification in neuropsychology does not necessarily guarantee expertise in all areas of neuropsychology. Unlike the American Board of Neurology and Psychiatry, which adds a ''with special competence in child neurology" (or psychiatry) for practitioners who satisfy the required training and experience for this endorsement, neuropsychology issues only generic endorsement. Further complicating the issue of board certification, usually designated as "diplomate" status, is the pervasive level of inference issue. Because only the best neuropsychological clinicians-for instance, those qualified to make the highest levels of neuropsychological inference-are likely to receive "diplomate" status, who is to do the lower level of inference work? As has been mentioned, whereas it might be considered optimal practice to have all children with any problem seen by a skilled child neuropsychologist, to ensure that no problems of a neurological nature are overlooked, this is obviously not realistic. One credentialing approach that attempts to treat this issue in a realistic way is that endorsed by the American Board of Professional Neuropsychology (ABPN), which recognizes competence at two levels. With the diplomate representing the highest level of recognition (with requirements generally very similar to those of the American Board of Professional Psychology), the ABPN also issues certification for competencies in neuropsychology at levels somewhat below those required for the diplomate status. Presumably a number of levels of neuropsychological inference could be made by such certified practitioners, while helping ensure that the child (or adult) was being evaluated by an individual sufficiently trained to recognize most neurological conditions that might require further evaluation. This model of a two-tier diplomate/ certification appears to have promise, but the relative newness of all credentialing approaches in neuropsychology precludes the accumulation of such data concerning successes or problems involved in such credentialing. It appears that specialty credentialing is unlikely
to be accepted in the near future by the vast majority of existing clinical psychologists. Therefore, training must be designed and offered to best prepare these individuals for their designated area of clinical service. Such needs are being met by universities offering specialty training in school psychology, child psychology, and/ or neuropsychology.
General Issues in Child Clinical Training Clinical child neuropsychology may best be viewed as a subarea of clinical child psychology, and it is relevant to preface a review of issues in clinical child neuropsychology training with an overview of training issues in clinical child psychology. Presently, although there are seven formal predoctoral training programs in neuropsychology (Lubin & Sokoloff, 1983), none are specifically designed for child neuropsychology. Thus, much of the specialty training in clinical child neuropsychology currently is provided by postdoctoral positions. The report of the task force from Division 40 recommends that in the absence of formal accredited educational programs: (l) The entry level credentials for the practice of clinical neuropsychology shall be predicated on the license to practice at the independent professional level in the state or providence in which the practitioner resides; (2) In addition, 1600 hours of clinical neuropsychological experience, supervised by a clinical neuropsychologist at the pre- or post-doctoral level, shall be required; (3) Persons receiving a doctoral degree in psychology before 1981 may substitute 4800 hours of post-doctoral experience in a neuropsychology setting involving a minimum of 2400 hours of direct clinical service. (Newslener 40, 1984)
In the absence of formal training programs in child neuropsychology, specialization in child neuropsychology must either combine two existing areas or lead to even greater specialization. Due to the changing nature of the nervous system in the child and the impact of nonneurological factors on the child's behavior, the child neuropsychologist needs to be trained in basic psychological, developmental, and neuropsychological issues. In addition, the role of psychological assessment in clinical child neuropsychology needs to be well understood. The reliance on standardized tests increases
with decreasing experience of professionals in any discipline. Of primary importance is the issue to be
TRAINING AND CREDENTIALING
addressed or the question to be answered. If the primary question relates to whether there is cerebral dysfunction, then regardless of the test employed, the evaluators' effectiveness depends on their training in brain-behavior relationships and their understanding of the nervous system and its contributions to behavior. Without such training, effective interpretation of behavior leading to decisions regarding brain dysfunction cannot be reached. Ifleaming disability is of primary interest; then the evaluator needs to understand the relationship between test behavior and learning disability. The same argument holds for developmental delays, emotional disorders, retardation, and so on. New graduates, individuals shifting their area of basic training, or researchers tend to depend on a fixed battery or evaluation strategy and rigorously defend it against all others. They thus exhibit a strong tendency to become method oriented, rather than problem oriented. With further education on the part of the professional and understanding of the relationship between areas of primary importance, less reliance is made on a specific test battery and a broad range of assessment devices may be employed in order to effectively assay the behaviors in question and outline an effective treatment plan. Clinical neuropsychology as a specialty within psychology is a very new area that is continuing to undergo change and self-analysis in order to outline clinical courses most appropriate to the practice of neuropsychology. The data base on neuropsychology has also served to shift psychologists into a designated specialty area as the knowledge base required to pursue neuropsychological assessment is sufficiently broad to make it difficult for traditionally trained psychologists to pursue effectively such clinical activities without extensive training or experience. In 1977 it was recognized that a conference dealing with training in clinical child psychology was needed, and a preliminary working conference was held in 1983 with the principle conference held in May 1985. In general, the recommendations included three features involving general clinical psychology training, involving requirements for training in normal development; experience with normal children; and minimal competencies in assessment, psychopathology, and intervention with children (Tuma, 1986; Johnson & Tuma, 1986). Specific to clinical child psychology graduate training were seven recommendations, the first of which endorsed the Boulder Model for clinical child psychology. Another recommendation endorsed the APA Division 27 task force documented Guidelines for Training Psychol-
589
ogists to work with Children, Youth and Families (Roberts, Erickson, & Tuma, 1985). In general, the other recommendations specific to clinical child psychology training dealt with such issues as recognizing cultural diversity and the multiple contexts in which psychologists working with children, youth, and families must function. Internship training was recommended as involving at least two thirds of the training experience in child clinical activity, with research incorporated into the internship program. Postdoctoral and continuing education training in clinical child psychology was recommended, although specific guidelines concerning required background prerequisities or context areas were not proposed. With respect to recognition of proficiencies and specialty areas in psychology, the APA Board of Professional Affairs (BPA) appointed a Committee on Specialty Practice from 1970 to 1980 to explore such issues. Specialty guidelines for clinical, counseling, industrial/organizational, and school psychology were approved by the APA Council in 1980, marking APA's first detailed public statement concerning service provisions in specialty areas. The BPA appointed a Subcommittee on Specialization in 1980 to address the issues involved in criteria for specialty areas not covered by these four major areas, and in 1983 a second draft manual for the identification and continued recognition of proficiencies and new specialty areas in psychology was published (Sales, Bricklin, & Hall, 1983). Differentiation was made between proficiencies and specialties, on the basis of several major criteria. A specialty was recommended as involving a body of knowledge with (1) unique client populations, (2) specific techniques and technologies, (3) problems addressed, and (4) settings wherein the knowledge is applied. A proficiency, on the other hand, would involve a body of knowledge and skills that provide the basis for services in one of these four parameters. The requirements for the identification of a specialty area involved (1) a formal organization, recognized in the field, that is responsible for managing the development of a specialty; (2) a definition of the specialty, including knowledge and skills required; and (3) an educational sequence of training and experience. Requirements for the identification of a proficiency involved ( 1) a formal organization, (2) a definition, (3) evidence of need and parameters ()f practice, (4) demonstrated efficiency, and (5) uniqueness. In this context, neuropsychology could be viewed as representing either a specialty or an area of proficiency, with clinical child neuropsychology a subarea of either a specialty or a proficiency.
590
CHAPTER 32
In a related and somewhat parallel area, the APA Task Force on Education and Credentialing ( 1985) published a recommendation concerned with educational content required for designation as a psychology program. Although related in only a tangential way to clinical child neuropsychology, the designation system tends to discourage the graduate education of clinical child neuropsychologists in academic settings without a clear identification as part of a psychology program (e.g., freestanding clinical child neuropsychology programs in medical schools or professional schools would have difficulty meeting the designation criteria).
Focus on Training in Clinical Child Neuropsychology Where does training in clinical child neuropsychology fit into this broader context? Training in clinical child neuropsychology is generally provided in one of three ways: graduate coursework; internship/practicum training; and postdoctoral training fellowships. Graduate course offerings show considerable variability. Approximately seven programs offer a terminal degree in neuropsychology; 40 clinical programs offer some coursework in neuropsychology; and some half-dozen clinical programs offer lectures on neuropsychology but no formal coursework (Golden & Kuperman, 1980). Thus, among the 60 or more APA-approved clinical programs that indicate they provide offerings in neuropsychology, these offerings may range from formal coursework to practica or even possible work placements. Division 40 of the APA (Neuropsychology), aware of the need for establishing guidelines for neuropsychology training, has formed a task force to develop such guidelines. A preliminary report of their efforts was published in Newsletter 40 (1984). According to those guidelines the major function of the clinical neuropsychologist is to assess current behavioral disturbances associated with neurological impairment. The report suggested that neuropsychological assessment should include measures of ( 1) abstract reasoning and categorical thinking, (2) cognitive flexibility and planning, (3) language communication, (4) learning and memory, (5) sensation and perception, (6) fine and gross motor functions, (7) initiation and attention, (8) affect and mood, and (9) psychosocial adaptation. In order to effectively pursue these assessment goals, the diagnostician needs training in (1) func-
tional neuroanatomy, (2) clinical diseases, (3) child development, (4) changes in behavior as a function of aging, (5) behavioral psychopharmacology, (6) psychophysiological principles underlying pathologies, (7) sociocultural factors, (8) personality assessment and interviewing skills, (9) principles of test construction and validation, and (IO) test administration and interpretation. Properly trained neuropsychologists should be able to outline treatment plans and consult with family members, educators, employers, and so on, in order to aid in improving the behavioral adjustment of the individual in specific situations. Remediation by a clinical neuropsychologist focuses primarily on disability associated with cerebral dysfunction and secondarily on emotional or other maladaptive behaviors that are a consequence of the individual's primary disability. That same report outlines the needs of the child neuropsychology training to include much of the above with adjustment in training suggested to incorporate bodies of knowledge as well as techniques and resources specific to clinical child neuropsychology. Major issues such as child development, CNS plasticity, and the nature of the referral questions are seen as primary additional areas of competence. One of the primary distinctions between child and adult neuropsychology is the emphasis on description of processes in children, because the focus on process helps delineate specific treatment plans. More so than with adults, children are often evaluated by a multidisciplinary team; thus, child neuropsychologists must have knowledge of related professions so that they may effectively interface their findings in developing the final treatment plan. Among practicum offerings that indicate child neuropsychology as an area of training, these offerings in many cases exist as ancillary options, such as being available on a limited basis within a child therapy practicum. Even in practicum or internship settings wherein neuropsychology is mentioned as an area of training emphasis, there is considerable variability. This variability appears to reflect both the differing concepts of neuropsychology as a specialty area within clinical neuropsychology, and the unique backgrounds of the faculty who provide such training. In one grouping of 28 graduate settings that offered neuropsychology training, for example, Golden and Kuperman (1980) found that the tests used most frequently were the Wechsler and Bender Gestalt. Postdoctoral training programs in clinical child psychology are relatively rare. However, a number of postdoctoral programs in clinical neuropsychology offer some exposure to child neuropsychology,
TRAINING AND CREDENTIALING
and a few provide some segment of the program devoted to work with children. Informal surveys of postdoctoral trainees who have had at least some postdoctoral training in clinical child neuropsychology reveal a rather wide range of backgrounds. Some "retread" postdoctoral fellows, whose doctoral training is in nonclinical areas such as physiological psychology, have very little background in either child development or the special skills needed to evaluate children. Others with backgrounds in areas like school psychology may have excellent skills in child assessment and good knowledge of developmental phenomena, but little expertise in functional neuroanatomy or basic brain-behavior relationships. Yet others enter postdoctoral child neuropsychology training programs with good assessment skills involving both children and adults, with coursework in neuroanatomy and physiology, and prior exposure to neurologically impaired children from practicum or work experiences. Thus, the content of the "ideal" postdoctoral experience in clinical child neuropsychology may relate to the unique backgrounds that such postdoctoral fellows bring to the program.
591
Unlike the adult brain, which is assumed to be developmentally static with fixed effects associated with injury, the child's brain is characterized by growth and differentiation that extends from conception until young adulthood (Renis & Goldman, 1980; Rourke, Bakker, Fisk, & Strang, 1983). The effects of neurological damage are influenced by age, the locus of the injury, the nature of the damage, the sex and socioeconomic status of the individual, as well as the emotional adjustment, coping and adaptive skills of the individual (Bolter & Long, 1985). Thus, even our limited understanding of chronogenetic localization can improve the assessment and remediation of neurologically impaired children. Neurological damage during the developmental years may produce permanent deficits, temporary deficits, and/or delayedonset deficits (Teuber & Rudel, 1962). Understanding the neurological contribution to the overall behavioral complex is necessary in order to effectively identify barriers and plan for remediation.
Professional Relationships
All psychologists view behavior from a systems perspective; however, problems are viewed somewhat differently depending on the specialization. School psychologists focus primarily on acaNeuropsychologists assume that understanding demic problems and secondarily on how nonbrain-behavior relationships is necessary for both academic factors influence this performance (e.g., diagnosis and treatment planning. Such knowledge is emotional, situational, neurological, genetic, develnot, however, sufficient; consequently, few neuro- opmental). Child psychologists focus primarily psychologists focus on the brain as the only contrib- on emotional/behavioral problems with secondary uting variable. Child neuropsychological assessment focus on other areas. The child neuropsychologist must include measures of personalityI emotional focuses primarily on brain-behavior relationships well-being and identification of environmental influ- with other factors being viewed as secondary. ences. Given such a broad "systems" analysis, the The approach of child neuropsychologists has child neuropsychologist can provide information of been challenged by professionals in other specialties. benefit to a number of other disciplines. For exam- School psychologists have argued that understanding ple, the interpretation of neurological dysfunctions in neurological systems is not important for effective the context of situational, learning, emotional, and treatment (Senf, 1979). It is further argued that neuother important dimensions provides the neu- rological labeling connotes irreversibility and mitirosurgeon with a more comprehensive picture of the gates responsibility for remediation (Sandoval & role that a lesion or area of damage might exert on the Haapanen, 1981). In fact, Hynd (1982) suggested child's behavior. This can assist teachers in the class- that the neuropsychological evaluation may provide room and parents at home by identifying strengths information that reduces the need for referral for exand weaknesses and identifying those factors that pensive and nonproductive neurological evaluations. appear to be most amenable to modification. The There remain many unresolved issues regarding assumption is that one needs to identify factors that training and practice of clinical child neuropsycholcontribute to aberrant behaviors and prioritize them ogy. As outlined in this chapter, the clinical child regarding those that would appear to require primary neuropsychologist must possess a knowledge base assistance as well as those that are most likely to that cuts across many existing areas of specialization. change with remediation. Perhaps for this reason, individuals from a number of
Professional Context of Clinical Child Neuropsychology
592
CHAPTER 32
specialty areas may function in the assessment and treatment of children with neurological dysfunction in the future. Hopefully, with improved awareness and education, effectiveness of communication will be enhanced across these specialties. This may lead us to recognize the requisite combination of broad skills in general child clinical areas and specific skills in child neuropsychology, as constituting clinical child neuropsychology; both a specialty and an area of proficiency.
References Adams. K. M. (1980a). In search of Luria's battery: A false start. Jourf!lll of Consulting and Clinical Psychology, 48, 511516. Adams, K. M. (1980b). An end of innocence for behavioral neurology? Adams replies. Journal of Consulting and Clinical Psychology, 48, 522-523. American Psychological Association Task Force on Education and Recommendations for a Designation System. (1985). Washington, DC: American Psychological Association. Berg, R. A., Bolter, I. F., Ch'ien, L. T., Williams, S. 1., Lancaster, W., & Cumming, J. (1984). Comparative diagnostic accuracy of the Halstead-Reitan and Luria-Nebraska Neuropsychological Adult and Children's Batteries. International Journal of Clinical Neuropsychology, 6(3), 200-204. Bolter, J. F., & Long, C. 1. (1985). Methodological issues in research in developmental neuropsychology. In L. C. Hartlage & C. F. Telzrow (Eds.), Neuropsychology of individual differences: A developmental perspective (pp. 41-59). New York: Plenum Press. Chalfant, I. C., & Scheffelin, M. A. (1969). Central processing dysfunctions in children: A review of research. Bethesda: U.S. Department of Health, Education and Welfare. Dean, R. S. (1982). Focus on child neuropsychology. The National Academy of Neuropsychologists Bulletin, 2(2), 5. Geary, D. C., Schultz, D. D., Jennings, S.M., & Alper, T. G. (1984). The diagnostic accuracy of the Luria-Nebraska Neuropsychological Battery-Children's Revision for 9 to 12 years old learning disabled children. School Psychology Review, 13(3), 375-380. Goldberg, A. L., & McNamara, K. M. ( 1984). Internship training in neuropsychology. Professional Psychology: Research and Practice, 15(4), 509-514. Golden, C. 1. (1981). The Luna-Nebraska Children's Battery: Theory and formulation. In G. W. Hynd & I. E. Obrzut (Eds.), Neuropsychological assessment and the school-age child: Issues and procedures (pp. 277-302). New York: Grune & Stratton. Golden, C. J., Hammeke, T. A., & Purisch, A. D. (1980). The LuriD-Nebraska Neuropsychological Battery. Los Angeles: Western Psychological Services. Golden, C.1., Kane, R., Sweet, J., Moses, J. A., Cardellino, J. P., Templeton, R., Vicente, P., & Graber, B. (1981). Relationships of the Halstead-Reitan Neuropsychological Battery
to the Luria-Nebraska Neuropsychological Battery. Journal of Consulting and Clinical Psychology, 49(3), 410-417. Golden, C. J., & Kuperman, S. K. (1980). Graduate training in clinical neuropsychology. Professional Psychology, 11(1), 55-63. Hartlage, L. C. (1981). Clinical application of neuropsychological data. School Psychology Review, 10(3), 362-366. Hartlage, L. C. (1984). Neuropsychological assessment of children. In P. Keller & L. Ritt (Eds.), Innovations in clinical practice (Vol. III, pp. 153-165). Sarasota. FL: Professional Resource Exchange. Hartlage, L. C., & Telzrow, C. F. (1983). Assessment of neurological functioning. In B. Bracken & F. Paget (Eds.), Psychoeducational assessment of preschool and primary aged children (pp. 1-68). New York: Grune & Stratton. Hynd, G. W. (1982). Neuropsychological consultation in the schools. The National Academy of Neuropsychologists Bulletin, 2(2), II. Johnson, I. H., & Turna, 1. M. (1986). The Hilton Head conference: Recommendations for clinical child psychology training. The Clinical Psychologist, 39(1), 9-11. Lubin, B., & Sokoloff, R. M. (1983). An update of the survey of training and internship programs in clinical neuropsychology. Journal of Clinical Psychology, 39(1), 149-152. Newsletter40. (1984). American Psychological Association, Division of Clinical Neuropsychology, Vol. 11(2). Plaisted, J. R., Gustavson, J. L., Wilkening, J. G. N., & Golden, C. 1. (1983). The Luria-Nebraska Neuropsychological Battery-Children's Revision: Theory and current research findings. Journal of Clinical Child Psychology, 12, 13-21. Reitan, R. M. (1955). An investigation of the validity of Halstead's measures of biological intelligence. AMA Archives of Neurological Psychiatry, 73, 28-35. Reitan, R. M., & Davison, L. (1974). Clinical neuropsychology: Current status and applications. Washington, DC: Winston. Renis, S., & Goldman, I. M. (1980). The Development of the Brain. Springfield, IL: Thomas. Roberts, M. C., Erickson, M. T., & Tuma,1. M. (1985). Addressing the needs: Guidelines for training psychologists to work with children, youth and families. Journal of Clinical Child Psychology, 14, 70-79. Robinson, E. A. (1985). Specialization: Who needs it? Newslener of the Society of Pediatric Psychology, 9, 5-6. Rourke, B. P., Bakker, D. J., Fisk, J. L.; & Strang,1. 0. (1983). Child neuropsychology: An introduction to theory, research, and clinical practice. New York: Guilford Press. Sales, B. (1985). Specialization: Past history and future alternatives. The Clinical Psychologist, 38, 48-52. Sales, B., Bricklin, P., & Hali,J. (1983). Second draft: Manual for the identification and continued recognition ofproficiencies and new areas in psychology. Washington, DC: American Psychological Association. Sandoval, J., & Haapanen, R. M. (1981). A critical commentary on neuropsychology in the schools: Are we ready? School Psychology Review, 10, 381-388. Selz, M. (1981). Halstead-Reitan neuropsychological test batteries for children. In G. W. Hynd & J. E. Obrzut (Eds.), Neuropsychological assessment and the school-age child: Issues and procedures. New York: Grune & Stratton.
TRAINING AND CREDENTIALING Senf, G. M. (1979). Can neuropsychology really change the face of special education? The Journal of Special Education. 13(1).
Spiers, P. (1981). Have they come to praise Luria or to bury him? The Luria-Nebraska controversy. Journal ofConsulting and Clinical Psychology, 49, 331-341. Telzrow, C. F., & Hartlage, L. C. (1984). A neuropsychological model for vocational planning for learning disabled students. In W. Cruickshank & J. Kleibhan (Eds.), Early adolescence to early adulthood (pp. 143-156). Syracuse: Syracuse University Press.
593
Teuber, H. L., & Rudel, R. G. (1962). Behavior after cerebral lesions in children and adults. Developmental Medicine and Child Neurology, 4, 3-20. Tuma, J. M. (1986). Clinical child psychology training: Report of the Hilton Head conference. Journal of Clinical Child Psychology, 15(1), 88-96. VandenBos, G. R., Stapp, J., & Kilburg, R. R. (1981). Health service providers in psychology: Results of the 1978 APA human resources survey. American Psychologist, 36, 13951418.
Index Absence seizures, 409 Academic skills, 298 tests of, 562 Acetylcholine, 49 Adjustment, 536 psychosocial, 545 Adjustment reaction, 479 Adolescents, epilepsy and, 424 Aerosols, 315 Affective disorders, 97 Agenesis of corpus callosum, 25 Aggression, 529 Agnosia, visual, 296 Agraphia, 299 Akathisia, 478 Alcohol, 322 Alexia, 299 Alpha process, 267 American Board of Clinical Neuropsychology, 574 American Board of Professional Neuropsychology, 574 American Board of Professional Psychology, 574 American Journal of Psychiatry, 5 American Psychological Association, 8, 574 Amitriptyline, (Elavil), 482 Amphetamine, 280, 321 Anemia, 110 Anencephaly, 25 Aneurysms, 252 Anorexia, 463 Anoxia, 259 Anterior metencephalon, 43 Antianxiety drugs, 476, 482 Anticonvulsants, 361,409 Antidepressants, 432, 480, 481 Antiepileptic drugs, 419 . Antimania drugs, 419, 476 Antipsychotic drugs, 477 Anxiolytics, 475 Aphasia, 297 Aphasia Screening Test, 186, 360 Aptitude by Treatment Interaction, 514 Archives of Clinical Neuropsychology, 5 Aristotle, 3
Arithmetic disabilities, 137, 368 Assessment approaches, training in, 585 Assessment methods, 172 Association for Advancement of Behavior Therapy, 521 Astrocytes, 47 Astrocytoma, 253, 586 Asymmetry, 70, 294 neural, 70 morphological, 72 somatic, 70 Atrophy, cortical, 252 Attention Deficit Disorder, 95, 140, 361, 443-468 behavior and, 450 conduct disorders and, 452 definitions of, 451 ecological measures of, 461 learning disabilities and, 450 Attention span, 459 Attention, selective, 458 Attention, tests of, 562 Attributions, 543 Auditory verbal processing, tests of, 562 Aura, 539 Autism, 259 Autogenic training, biofeedback, 390 Autonomic nervous system, 42, 458 Autonomy, 547 Axonal development, 21 Background history of a child, 567 Balance, 462 Bamazepine, 411 Barbituates, 322, 429 Basal ganglia, 43 Base rates, 512 Bayley Scales of Infant Development, 228 Behavior, 26 aggressive, 430 direct observations of, 460 Behavior checklists, 460 Behavior modification, 383 Behavior rating scales, 460 Behavioral engineering, 398
Behavioral neuropsychology, 521 Behaviorist model, 522 Bender Gestalt, 384, 462 Benton's Sentence Memory Test, 360 Benton Visual Retention Test, 431 Benzedrine,443,476 Benzodiadzepine, 433 Benztropine mesylate (Cogentin), 479 Biofeedback, 377 Bipolar disorders, 477, 482 Blind, 558 Boder Test of Reading and Spelling, 149, 236 Body awareness, tests of, 562 Boston Naming Test, 431 Bradykinin, 389 Brain and Language, 5 Brain Electrical Activity Mapping, 302 Brain, 5, 37, 108 Brain injury, 35, 507 early, 35 frontal lobe, 36 Bronchial Asthma, 117 Buccolingual-masticatory movements, 478 Caffeine, 444 Carbamazepine, 420 Carboxylic acid, 433 Case study materials, 563 Category Test, 182 Caudality, 527 Cellular differentiation, 41 Central information processing, 58 Central nervous system, 458 Centrencephalic epilepsy, 426 Cephalic flexure, 42 Cerebellar agenesis, 25 Cerebellum, 46 Cerebral aqueduct, 43 Cerebral asymmetry, 30 Cerebral cortex, 43 Cerebral hemispheres, 47 Cerebral palsy, 378, 476, 558 Cerebrospinal fluid, 42 Certification, 9 Cervical flexure, 42
595
596
INDEX
Child Behavior Checklist-Revised, 239, 561 Child clinical neuropsychology, training in, 588 Childhood autism, 274 Childhood schizophrenia, 93, 275; see also Schizophrenia Chlorpromazine (Thorazine}, 476 Circumstantiality, 431 Classroom behavior, 429 Classroom instruction, 367 Clinical judgment, 170 Clinical neuropsychology, 5, 14 Clinical-inferential methods, 167 Cocaine, 313 Cogentin, 479 Cognitive development, 57, 59 tests of, 562 Cognitive processing, 347 Cognitive rehabilitation, 397 behavioral engineering and, 398 computer programs and, 401 developmental approaches to, 400 neurobehavioral approaches to, 402 programs, 399 psychometic model of, 397 strategies of, 404 theoretical models of, 397 Color Fonn Test, 186 Comprehensive Assessment Report, 490 Comprehensive Austin Neuropsychological Assessment Battery, 559 Computer programs, 401 Concentration, 431 Concrete operations, 539 Conduct disorders, 96, 452 Congenital anomalies, 259 Connective tissue system, 121 Consequent variables, 525 Contamination, 539 Context updating, 272 Contingent Negative Variation, 270 Continuing Education, 6 Continuous Perfonnance Test, 481 Controlled Word Association Test, 431 Coping, 536 Corpus callosum, 299 Corpus striatum, 43 Cortex, 5 regions of, 18 Cortical dysplasia, 259 Cranial Cerebral Trauma, 255 Credentialing, 9, 574 CT scanning, 347 Cylert, 445 Cystic Fibrosis, 118 Cytoarchitectonic areas, 21 Daily living skills, 399 Deanol, 444
Dean's Test of Lateral Dominance, 360 Defense mechanisms, 541 Deficit training, 363 Degenerative syndromes, 249 Delirium, 477 Delusional thinking, 477 Dementia, 477 Dendrite development, 21 Denial, 541 Depakane (valproic acid), 433 Depression, 424, 480 Descartes, 3 Designer drugs, 325 Development axonal, 21 cognitive, 57 dendrite, 21 glial, 23 neurochemical, 23 perceptual, 57 speech, 55 white matter, 52 Developmental dyslexia, 335; see also Dyslexia Developmental Test of Visual Motor Integration, 238 Dexedrine, 445 Dextroamphetamine, 444, 445 Diabetes mellitus, 112
Diagnostic and Statistical Manual of Mental Disorders, 3rd edition, 476 Diazepam (Valium), 452 Dichhaptic recognition, 30 Diphenhydramine (Benadryl), 411 Dilantin (Phenytoin), 433 Dissociative states, 475 Dopamine, 49, 445 Dorsality, 529 Down's syndrome, 503 Draw-a-Person, 477 Drug abuse, 311 Drugs antianxiety, 482 anticonvulsants, 432 antidepressants, 475, 481 antiepilepsy, 419 antimania, 476 antipsychotic, 477 Duchenne muscular dystrophy, 542 Durrell Analysis of Reading Difficulty, 560 Discalculia, 300 Dysdiadochokinesis, 89 Dyseidetic readers, 370 Dyskinesias, 478 Dysphoria, 457 Dyslexia, 269, 279 developmental, 335 dysmetric, 337 subgroups of, 369
Dysmetric dyslexia, 337 Dysphonetic readers, 370 Dysthymic disorder, 480 Ear dominance, 339 Echolalia, 477 Ecological measures, 461 Educational deficits, 357 Educational treatment, 363 EEG, 54,425 abnonnal activity in, 266 maturation, 266 training in, 380, 425 Ego,541 Elavil, 482 Electric shock, 531 Electronic ear, 337 EMG Biofeedback, 383 Emotional disturbance, 361, 497 Encephalitis, 108, 260 Endocrine system, ll2 Enuresis, 480, 482 Epilepsy, 482 adolescents and, 421 age of onset, 426 assessment of, 420 biofeedback and, 434 cognitive ability and, 424 emotional disorders and, 420 fears of, 423 general intelligence and, 427 memory and, 430 myoclonic, 422 personality changes and, 422 psychosocial variables and, 420 stress and, 43 Ependymoma, 254 Epilepsy, 261, 360, 379,409 Equipotentiality, 291 Ergotamine, 389 Ethics, 574 Ethosuximide (Zarontin), 422 Evoked potentials, 458 Executive functions, 301 Expert testimony, neuropsychological, 580 External locus of control, 423 Eye movement patterns, 340 Eye tracking, 462 Face perception, 32 Face recognition, 30 Failure aversion, 403 Family history questionnaire, 561 Family, impact of chronic illness on, 561 Finger Twitch Test, 462 Fecal incontinence, 381 Feedback circuit, 42
INDEX Fernald method, 368 Filter, 271 Finger Oscillation, 184 Fingertapping Test, 182 Fingertip Number Writing Test, 186 Fissures, 44 Follow-up conference, 563 Food and Drug Administration, the, 453 Fornix, 43 Freedom form distractibility factor, 461 Frontal lobe, 32, 301, 363 Frostig Developmental Test of Visual Perception, 450 Functional circuits, 42 Galen, 3 Gasoline inhalation, 316 Gastrointestinal system, 122 Gastrointestinal distress, 464 Gates Reading Test, 451 General Aptitude Test Battery, 4ll Gerstmann's syndrome, 295 Governmental regulation, 573 Glial cells, 21, 47 Glial development, 23 Glioblastoma, 253 Glucose utilization, 24 Glue sniffing, 315 Graduate programs, 590 Grand mal seizures, 409 Graphesthesia, 462 Gray matter, 249 Great Britain, neuropsychological information and, 4 Great Society, the, 585 Growth, family, 548 Habilitation, 337 Haldol, 478 Hallucinations, 465 auditory, 422 Haloperidol (Haldol), 478 Halstead-Reitan Neuropsychological Test Battery for Children, 181, 220,365,431,432,477,578 Handicap, definition of, 538 Hashish, 312 Head injury, closed, 529; see also Brain injury Headaches, 464 Hearing impairments, 558 Hemispheres, cerebral, 47 Hemispheric specialization, 70 Heroin, 321 Herpes encephalitis, 296 Hetertopia, 25 Hippocampus, 43 Hippocrates, 3 Histamine, 389 Hodgkin's disease, 464
Holoprosencephaly, 25 Huntington's Chorea, 252, 529 Hydantoin, 433 Hydrocephalus, 259 Hydroxyzine (Atarax), 476 Hyperactivity, 89, 361,443,449, 479 Hyperkinesis, 276 Hyperkinetic impulse disorder, 443 Hypotension, 120 Hypothalamus, 45 Hypothesis formation, 169 Idiopathic seizures, 425 Illinois Test of Psycholinguistic Abilities, 384 lminostilbene, 433 Imipramine (Tofranil), 476 Impulsivity, 443, 530 In-service training, 493 Incontinence, neurogenic, 393 Independence, 547 Individual approach, 4 Individual Education Plan, 491 Individualization of treatment, 401 Infantile autism, 93, 477 Infantile hemiplegia, 259 Infectious disorders, 258 Inference, models of, 171 qualitative, 173 quantitative, 173 Information processing, 58 Inhalants, 315 Insomnia, 463 Interindividual comparisons, 525 Internalization, 539 International Journal of Clinical Neuropsychology, 5 International League Against Epilepsy, 414 Intervention techniques, 525 IQ, 158 Irritability, 457 Journal of Clinical and Experimental Neuropsychology, 5 Journal of Clinical Neuropsychology, 5 Journal of Consulting and Clinical Psychology, 5 Kagan's Matching Familiar Figures Test, 461 K-SOS, 364, 365 Kaufman Assessment Battery for Children, 149, 205, 296, 229, 230, 524 Kaufman Sequential or Simultaneous Remedial Program, 364, 365 Klonopin (Cionazepan), 433 Kluver-Bucy syndrome, 422
597
Language Experience approach, 368 Language, 297 development, 28 impairment, 558 Lateral ventricles, 43 Laterality, 73, 430, 527 Lateralization central, 75 cerebral, 59, 209 degree of, 77 of function, 74 Learning disabilities, 132, 261, 335, 503 biofeedback and, 386 classifications of, 504 goals and, 507 intervention strategies and, 506 Learning, 362, 459 state dependent, 460 Leiter International Performance Scale, 231 Leukemia, Ill Leukodystrophy, 252 Levels of inference, 512, 586 Lewinsohn model, 524 Limbic system, 43, 419 Linguistic reading approach, 369 Linguistic rules, 192 Lip-smacking, 478 Lissencephaly, 25 Lithium carbonate, 280, 482 Localization, techniques of, 291 Locus of control, 387 Low birth weight, 53 Luminal (Phenobarbital), 433 Lucia-Nebraska Neuropsychological Battery: Children's Revision, 193, 217, 360, 525, 578 clinical scales, 202 critical levels, 196 developmental issues and, 198 interpretation of, 195 qualitative analysis and, 202 scales, 194 scale interpretation, 197 Luria, Alexander, 4, 207, 343 Lysergic Acid Diethylamide, (LSD), 320 Macrogyria, 25 Magnetic Resonance Imaging, 247 Malpractice, 578 Marching Test, 187 Marijuana, 312 Matching Familiar Figures Test, 384, 449 Matching Pictures Test, 188 Maturational lag, 268 McCarthy Scales of Children's Abilities, 232
598
INDEX
Medulla oblongata, 43 Medullary plate, 42 Mellaril, 476 Memory, 256 epilepsy and, 430 nonverbal, 431 tests of, 562 Meningitis, 109 Mental Processing Composite, 210 Mental processing, 208 Mental retardation, 259, 269, 273, 452, 558 Merrill-Palmer Performance Tests, 425 Methylphenidate, 280 Micropolygyria, 25 Midrin, 389 Migraine headaches, 378 biofeedback treatment of, 389 Minimal brain dysfunction, 89, 451 Minnesota Multiphasic Personality Inventory, 477, 523 Minnesota Rate of Manipulation Test, 411 Monoamine oxidase, 445 Mirror movements, 294 Morphological asymmetry, 72 Motor development, 57 Motor skills, assessment of, 293, 562 Movement disorders, 378 MRI. See Magnetic Resonance Imaging Multisolvent inhalation, 316 Muscle spasms, 478 Muscular dystrophy, 109 Music, 30 Myelin development, 23, 48 Myelomeningocele, 381 Mysoline (Primidone), 433 National Advisory Committee on Handicapped Children (1967), 504 National Institute of Mental Health, 443 National Institute for Neurological and Communicative Diseases and Stroke (NINCDS), Epilepsy Branch of, 414 National Joint Committee for Learning Disabilities, 505 Nebraska Neuropsychological Children's Battery, 193-204, 217, 360, 525,578 Negligence, 579 Neoplasms, 249 Neural asymmetry, 70 Neural tube, 18 Neurochemical development, 23 Neurocognitive approaches, 357 Neurodevelopmental issues, 358 stages, 359 Neurohormones, 49
Neuroleptic medications, 476 Neurological reflexes, 52 Neuromuscular diseases, 109 Neuropsychologia, 5 Neuropsychology, 5, 14 behavioral, 521 definition of, 573 future of, 155 Neuropsychological assessment, 490, 511 constructs, 505 diagnoses, 504 models, 336 reports, 561 strengths, 506 theory, 577 Neuropsychologist, role of, 570 Nonstimulant psychotropic drugs, 476 Nonverbal memory, 431 Normative data, 148 Nutrition, 53 Nystagmus, 338 Oculogyric crisis, 478 Oligodendrocytes, 18 Optic vesicles, 42 Organic personality syndrome, 421 Organismic variables, 525 Orton-Gillingham method, 368 Overcorrection, 530 Papilledema, 295 Paradoxical effects of medication, 450 Paraphasia, 297 Parasympathetic nervous system, 42 Parent interview, 567 Parental adjustment, 549 coping styles, 550 Parents, 491 Parkinson's features, 478 Partial complex seizures, 409 Pathognomonic signs approach, 189 Pavlov, 4 Peabody Picture Vocabulary Test-Revised, 237, 432 Peer interactions, 546 Pemoline, 444, 446, 448 Perceived Competence Scale for Children, 542 Perceptions, 542 Perception, visual, 295 Perceptual and motor skills, 5 Perceptual development, 57 Peripheral nervous system, 42 Personality Inventory for Children, 561 Pervasive developmental disorder, 93 PET: see Positron Emission Tomography
Petit mal seizures, 409 Phencyclidine (PCP), 318 Phenobarbital, 411, 420, 432 Phenytoin (Dilantin), 4ll Phonematic analysis, 349 Phonemes, 30 Phonemic awareness, 349 Physical and Neurological Examination for Soft Signs, 462 Piaget, Jean, 24 Piperidine, 445 Placebo, 460 Placebo treatment in biofeedback training, 387 Planning, 346 Plasticity, 18, 291, 358 Polycythemia vera, 110 Polydrug therapy, 436 Polysensory intergration, 42 Porencephaly, 25 Porteus Mazes, 449 Positron Emission Tomography (PET), 302 Postdoctoral training programs, 590 Preacademic tasks, 560 Prematurity, 53 Primidone, 411 Problem solving, 31, 373 Professional enthusiasm, 581 organizations, 6 relationships, 591 standards, 574 Progressive Figures Test, 188 Projective drawings, 561 Pronominal reversal, 477 Prosopagnosia, 296 Psychological processing, 515 Psychopathology, 87 Psychopharmacology, 476 pediatric, 444 Psychopharmacology Bulletin, 476 Psychosocial adjustment, 545 Psychosocial factors, tests of, 562 Psychostimulants, 444; see also Stimulants Psychotherapy, issues in, 544 Psychotic disorders, 93 Psychotropic medication, 475 Public policy, 573 Public Schools, 492, 573 Pyriform lobes, 43 Qualitative inference, 173 Quantitative inference, 173 Raven's Progressive Matrices, 560 Reading disabilities, 133 Reading skills, 368
INDEX
Reasoning and planning deficits, 373 Reinforcement, 531 Reitan-Aphasia Screening Battery, 560 Rehabilitation, 397 REHABIT, 364 Reinforcement, 404 Relaxation training, 385 Remedial programs, 357 Remediation, 501 deficit model, 223 strength model, 223 Report format, 563 Response variables, 525 Renal system, 123 Revised Behavior Problem Checklist, 240 Rey-Ostereith Complex Figure Test, 431 Rhinencephalon, 43 Ritalin, 445 Ritalinic acid, 445 Rorschach, 477 Russia, neuropsychological infonnation and, 4 Schemata, higher order and, 450 Schizoaffective disorder, 480 Schizophrenia, childhood, 93 Schizophrenia, 259, 477 Schoolsettings,489 Seizures absence, 409 age of onset, 410 diagnosis of, 413 emotional aspects of, 412 generalized tonic-clonic, 409 intellectual impairment and, 410 medications and, 411 partial complex, 409 psychomotor, 413 receptive language and, 410 social aspects of, 412 Seashore Rhythm Test, 183 Sedatives, 322 Selective attention, 458 Selective awareness, 45 Self-concept, 546 Self-control, loss of, 540 Self-esteem, 90, 546 Self-help skills, 399 Self-monitoring, 531 Sensory Perceptual Exam, 185, 560 Separation anxiety, 60, 480 Septeum pellucidum, 43 Sequential processing, 206, 562 Serotonin, 49 Siblings of children with neurological disorders, 551 Side-effects of stimulants, 447
Simultaneous processing, 207 Social skills training, 450, 477, 530 Socialization, 546 Socioeconomic status, 359 Sodium valproate, 420 Soft signs, 507 Somatic asymmetry, 70 Sound Neuropsychological Substructure for a Public School, 498 Spatial/visual processing, tests of, 562 Special Education, 491, 557 reform; 516 · SPecial interest group, 521 Specialization, 578 Specific learning disability, 505; see also Learning disabilities Speech abnormalities, 478 development, 55 zones, 18 Speech Sounds Perception test, 182 Spelling disabilities, 135, 357, 370 Sperry, Roger, 4 Spina Bifida, 381, 546 Spinal cord fonnation, 44 Spinal medulla, 46 Spongioblasts, 47 Standard of Care, 576 Standard scores, 158, 210 Stanford-Binet Intelligence Scale, 233 State-dependent learning, 460 Statistics, 160 Stimulants, 314, 361, 443 adolescents and, 453 adults and, 453 administration of, 467 cardiovascular effects of, 458 clinical evaluation of, 461 iatrogenic effects of, 463 learning and, 459 learning disabilities and, 450 mental relaxation and, 453 motor effects of, 443 neuropsychological evaluation of, 462 preschoolers and, 453 psychological testing and, 461 side-effects of, 442, 462 Stimulus antecedents, 525 Strength of Grip Test, 185, 564 Stroop Color Word Test, 462 Student benefits, 496 Substance abuse, 311 Succinimide, 433 Sulci, 44 Suprastriatal region, 43 Sympathetic nervous system, 42 Symptomatology, psychiatric, 545 Synaptic development, 22 Synergy, 462
599
Syntactic deficits, 292 Synthesis, 342 System of Multicultural Pluralistic Assessment (SOMPA), 238 Tactile difficulties, 295 Tactile Finger Localization Test, 186 Tactual Performance Test, 182, 426, 560 Target Test, 188 Task analysis, 399 Teaching, 363 Teacher-child interactions, 560 Tegretol (Carbamazepine), 433 Temperature biofeedback, 387 Temporal lobe epilepsy, 263 Temporal lobe, lesions of, 405 Tennessee Self-Concept Scale, 387 Test of Adolescent Language, 238 Test of Language Development, 237 Test of Written Language, 237 Thematic Apperception Test, 477 Theta filter, 435 Thioridazine (Mellaril), 280 Thorazine, 476 Threshold, 435 Tics, 464 Time out procedure, 531 Tofranil, 476 Token economies, 521 Token Test for Children, 238 Tornatis Listening Test, 338 Tourette's Syndrome, 98, 462, 464, 478 Trails A, 183 Trails B, 185 Training, 577 Tranquilizers, 475 Transfer of functions, 359 Trycyclic antidepressants, 433 Tumors, brain, 109 Valium, 452 Valproic acid (Sodium valproate), 411 Variables, inferred, 522 Vascular disorders, 252 Ventricle fonnation, 44 Ventrobasa nuclear complex, 435 Verbal memory, 431 Verbosity, 431 Veterans Administration, 5 Video games, 402 Visual acuity, 564 Visual evoked responses, 282, 315 Visual perception, 295 Visual/spatial processing, tests of, 562 Vocational training, 399 Vygotsky, L. S., 205
600
INDEX
Wechsler Adult Intelligence Scale (WAIS), 425 Wechsler Intelligence Scale for Children-Revised (WISC-R), 150, 205,230,296,426,430, 461
Wechsler Memory Scale, 430 White matter, development of, 52 Wide Range Achievement Test, 295, 387,432,560 Wisconsin Card Sorting Test, 301
Woodcock Johnson Psychoeducational Battery, 234 World War II, 527 Zarontin (Ethosuximide), 433 Zone of proximal development, 221