[Simon Boag, Niko Tiliopoulos] Personality and In

PSYCHOLOGY RESEARCH PROGRESS Additional books in this series can be found on Nova‘s website under the Series tab.

Additional E-books in this series can be found on Nova‘s website under the E-books tab.

PSYCHOLOGY RESEARCH PROGRESS

PERSONALITY AND INDIVIDUAL DIFFERENCES: THEORY, ASSESSMENT, AND APPLICATION

SIMON BOAG AND

NIKO TILIOPOULOS EDITORS

Nova Science Publishers, Inc. New York

Personality and Individual Differences Ed: Simon Boag and Niko Tiliopoulos

ISBN 978-1-61122-070-4 © 2011 Nova Science Publishers, Inc.

Copyright © 2011 by Nova Science Publishers, Inc. All 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, electrostatic, magnetic, tape, mechanical photocopying, recording or otherwise without the written permission of the Publisher. For permission to use material from this book please contact us: Telephone 631-231-7269; Fax 631-231-8175 Web Site: http://www.novapublishers.com NOTICE TO THE READER The Publisher has taken reasonable care in the preparation of this book, but makes no expressed or implied warranty of any kind and assumes no responsibility for any errors or omissions. No liability is assumed for incidental or consequential damages in connection with or arising out of information contained in this book. The Publisher shall not be liable for any special, consequential, or exemplary damages resulting, in whole or in part, from the readers‘ use of, or reliance upon, this material. Any parts of this book based on government reports are so indicated and copyright is claimed for those parts to the extent applicable to compilations of such works. Independent verification should be sought for any data, advice or recommendations contained in this book. In addition, no responsibility is assumed by the publisher for any injury and/or damage to persons or property arising from any methods, products, instructions, ideas or otherwise contained in this publication. This publication is designed to provide accurate and authoritative information with regard to the subject matter covered herein. It is sold with the clear understanding that the Publisher is not engaged in rendering legal or any other professional services. If legal or any other expert assistance is required, the services of a competent person should be sought. FROM A DECLARATION OF PARTICIPANTS JOINTLY ADOPTED BY A COMMITTEE OF THE AMERICAN BAR ASSOCIATION AND A COMMITTEE OF PUBLISHERS. Additional color graphics may be available in the e-book version of this book. LIBRARY OF CONGRESS CATALOGING-IN-PUBLICATION DATA Personality and individual differences : theory, assessment, and application / editors, Niko Tiliopoulos, Simon Boag. p. cm. Includes index. 1. Personality. 2. Individuality. I. Tiliopoulos, Niko. II. Boag, Simon. BF698.P3584 2010 155.2--dc22 2010036439

Published by Nova Science Publishers, Inc. † New York

CONTENTS Chapter 1

Introduction Niko Tiliopoulos and Simon Boag

1

Part A: Processes and Performance Chapter 2

Chapter 3

Chapter 4

Chapter 5

Chapter 6

Integrating the Study of within- and between-Person Variability in Personality at Work Nadin Beckmann, Amirali Minbashian and Robert E. Wood

11

Deliberate Practice and the Trainability of Fluid Cognitive Functions Damian P. Birney

21

The Effects of Comparison Status and Individual Differences in Social Comparison Tendencies on the Experience of Schadenfreude Jessica E. Franklin and Eugene Chekaluk Processing Style and Social Comparison Outcomes: Assimilation and Contrast Effects in Automatic Behaviour and Self-evaluation Lauren O’Driscoll and Eugene Chekaluk

51

The Excitation Transfer Paradigm and Emotional Intelligence: Predicting Interpersonal Attraction Madeleine Fraser and Simon Boag

67

Chapter 7

Take a Break: A Look at the Incubation Effect on Creativity Gerry Pallier and Niko Tiliopoulos

Chapter 8

The Role of Personality in Learning Processes and Learning Outcomes in Applied Settings Paul Ginns, Gregory Arief D. Liem and Andrew J. Martin

Chapter 9

37

Something about Metacognition: Self-confidence Factor(s) in School-aged Children Sabina Kleitman, Karina Mak, Sarah Young, Phoebe Lau and David Livesey

83

93

103

vi

Contents

Part B: Culture, Religion, and Spirituality Chapter 10

Spiritual-Hypnosis Assisted Therapy: A New Culturally Sensitive Approach to the Treatment and Prevention of Mental Disorders Luh Ketut Suryani

Chapter 11

Militant Extremist Mindset Lazar Stankov, Gerard Saucier and Goran Knežević

Chapter 12

Emotional Intelligence among Church Leaders: Applying the Schutte Emotional Intelligence Scale within Newfrontiers Leslie J. Francis, Andrew Ryland and Mandy Robbins

119 131

141

Chapter 13

Ominous Signs: Schizotypy and Religious Attachment Lisa Hancock

151

Chapter 14

Sham Reasoning, Humpty Dumpty and the Burden of Proof Fiona J. Hibberd

163

Chapter 15

Will Right-brainers Rule the Future? John Reid

171

Chapter 16

Cultural Perspectives on Personality and Personhood Hoyt L. Edge and Cokorda Bagus Jaya Lesmana

181

Part C: Contributions to Clinical Psychology and Assessment Chapter 17

Genetic Variation in DRD4, Criminality, and Conduct Disorder Tara M. Thompson, Marya Getchell, Britt af Klinteberg, Roman A. Koposov, Lars Oreland, Andrew J. Pakstis, Vladislav V. Ruchkin, Carolyn M. Yrigollen and Elena L. Grigorenko

Chapter 18

The Assessment of Depression, Anxiety and Tension/stress in Youth Marianna Szabó

209

Unconscious Schemas, Bulimic Tendencies and the Influence of Subliminal Abandonment and Unification Cues on Eating Kathleen Mairet and Simon Boag

223

Motivational Interviewing Changes the Treatment Trajectory of Group Cognitive-Behavioural Therapy for Anxiety Weiwen Yang and Esben Strodl

237

Chapter 19

Chapter 20

Chapter 21

What do Self-report Measures of Disinhibition Really Measure? Elliroma Gardiner and Chris J. Jackson

Chapter 22

Can Employers Screen Job Applicants for Potential White Collar Fraud Offenders? Jennifer Keen and Julie Fitness

193

249

265

Contents Chapter 23

Chapter 24

Reflections on the Factor Structure of the Occupational Stress Inventory-Revised Richard E. Hicks Introducing the YWEDO Online Cognitive Laboratory Chris J. Jackson

vii

275 283

Part D: Developments in Theory and Structure Chapter 25

Six Solutions to the Circular Nature of Trait Psychology Jonathan P. Gerber

Chapter 26

A Proposition for a Parsimonious Taxonomic Convergence of Personality Traits Niko Tiliopoulos

Chapter 27 Index

The Role of Conceptual Analysis in Personality Research Simon Boag

297

307 321 331

We should fill our life with possibilities, not regrets. Impossibility is a result of an absence of belief in oneself. Learning from experience is the best way to reach success. Our happiness in this life depends on our ability to choose the best possibility to walk on. No need for regrets; everything has an effect on the development of ourselves. There is always a chance, if we are willing to try. Everybody can change; they just have to want to. Before trying to change others, change yourself first.

Luh Ketut Suryani Department of Psychiatry, Faculty of Medicine, Udayana University and The Suryani Institute for Mental Health, Bali, Indonesia

In: Personality and Individual Differences Ed: Simon Boag and Niko Tiliopoulos


Chapter 1

INTRODUCTION Niko Tiliopoulos School of Psychology, University of Sydney, Sydney, NSW 2006, Australia

Simon Boag Department of Psychology, Macquarie University, Sydney, NSW 2109, Australia

ABSTRACT Since Allport and Vernon‘s assessment of the field over eighty years ago (Allport & Vernon, 1930), the field of personality and individual differences research has expanded enormously. This is particularly so in terms of theoretical development and practical relevance, where the research generated by researchers in this field is distinguished for its contributions to both assessment and application. In fact, personality and individual differences research is relevant to practically every facet of human existence. For instance, since theories of persons either explicitly or implicitly guide clinical work, the field contributes to discussions of understanding abnormal psychology and provides a guide for conceptualising best treatment. Additionally, the field is relevant to understanding human development across the lifespan, and our understanding of personality and individual differences impacts upon our views of socialisation and interpersonal relations. Furthermore, business and organisations continues to utilise personality and individual differences research since it provides the possibility of predicting performance. Subsequently, it is difficult to overestimate the importance of the research covered in this field. Concurrently, however, the field of personality and individual differences research is developing in ways that reflect a new appreciation of encompassing a global-context and of possible new approaches to understanding persons. This collection of papers from personality and individual difference researchers highlights social and cultural dimensions of personality and individual differences.



Ph: 61-2-9036 9223; Fax: 61-2-9036 5223; email: [email protected]; The editors would like to acknowledge that the book and chapter represent equal contributions from both authors.

2

Niko Tiliopoulos and Simon Boag

CULTURE AND PERSONALITY Historically, theories of personality and individual differences research have generally emerged primarily out of western thinking and traditions. In fact, Dan Jones (2010) has recently pointed out that since psychology students tend to be the most accessible and ‗willing‘ participants for psychology research, such findings are mostly representative of WEIRDos—Western, educated, industrialised, rich and democratic cultures. It has been further pointed out that such theories and findings have tended to emphasise the individual at the expense of understanding the person in relation (Edge, this volume; Hall, 2003). Consider, for instance, the significance of ‗self‘ in many approaches to personality, and the value placed on ‗self-development‘ and ‗self-actualisation‘, and so forth. While there may be value in such constructs, others have noted that this focus on the self reflects the values of western, industrialised cultures (e.g., North America) and may actually ignore the vast majority of the world‘s cultures that place value on the social collective over and above the individual (Hall, 2004; Markus, 2004). Consequently, some have been critical of personality and individual research, claiming that it reflects (and imposes) a western bias and thus not reflecting the global state of affairs (e.g., Markus, 2004). Some have even claimed that we are each trapped in our own cultural cocoons and are thus prevented from even ever understanding other cultures (e.g., Gergen, 2001). However, the latter claim is itself too extreme since if it is true, then it is itself trapped in its cultural setting and so can make no claim to making any truth statement extending beyond itself. Furthermore, it ignores humanity‘s great capacity to learn and communicate, and while we may struggle to get into another‘s world we can still make the effort, just as someone brought up ‗bi- (or even tri- etc.) culturally‘ has the benefit of appreciating multiple cultural nuances (cf. Hall, 2001). Consequently, while much research has emerged out of western thinking, this is, in itself, not necessarily problematic. Instead, acknowledging the potential for cultural bias raises very interesting questions—empirical questions—about whether findings generated in particular parts of the world generalise to other parts, and whether our theories are as universal as we may wish to believe. Subsequently, there is growing interest in cultural factors and increasing questioning of the universality of theories and findings generated in the predominantly English-speaking, western, industrialised countries. This makes for exciting times, and what is helping to break down cultural barriers is increasing accessibility to the internet. Since the internet‘s inception approximately two decades ago, global communication has become steadily more accessible and this global accessibility and communication has led to the internet being increasingly used in research (see Jackson, this volume; Kraut et al., 2004). More and more, then, we have the technology to help us become global researchers.

GLOBAL PERSPECTIVES ON PERSONALITY AND INDIVIDUAL DIFFERENCES While the greater global focus is both important and necessary for developing comprehensive theories and approaches, it brings with it certain challenges. In particular, coming to terms with different approaches to understanding persons—especially with respect to religious and spiritual dimensions—is an element that many researchers may feel

Introduction

3

uncomfortable with. Since the dawn of modern psychology, it seems psychologists have been sceptical about the value or utility of the psychological study of religion and spirituality. This battering scepticism apparently resulted in a rather uneasy relationship between psychology and religion (for example Esau, 1998; van Till, 1935/1972; Vande Kemp, 1986). Had it not been for people like William James (1902/2002) and Gordon Allport (1950), who through their pioneering ideas helped lift to a great extent that doubt, the field of the psychology of religion may have been far more marginal than it is today. Echoing their words, we argue that religion and spirituality are fundamental, multidimensional, and multilevel aspects of humanity. Since they carry a set of experiences, communal as well as personal belief systems, motives, and codes of values that seem to have a direct, at times powerful, and distinct impact on a person‘s functional and dysfunctional mental processes and behaviour, their investigation considerably overlaps with the domain of psychology, and more specifically with that of personality. Thus, it comes as no surprise that the main psychological body of both theory and empirical evidence on religion and spirituality comes from within the broader area of personology and individual differences. Contextualising the psychological interest, it should be made clear that it is the feeling, thought, or personal interpretation of transcendence, as well as the associated attitudes, behaviours, and personality correlates that are of direct if not exclusive interest to psychology, while the existence of religious objects and the transcendent actual are best addressed through philosophy or theology. It would be sagacious to stress Spilka‘s et al. (2003) disclaimer that states that the object of the psychological study of religion is neither God nor the worldview of religious institutions, nor faith versus reason, nor religion versus science, but people. And religion appears to be important to our understanding of people, not only because it addresses an element of life that seems to be central to most individuals and societies, but also because it seems to affect a plethora of areas, such as individual differences, identity, attachment, well-being, and mental health, that are of principal interest to psychologists. Despite Sigmund Freud‘s (for example 1927/1961) attempts to present it as an anachronistic and disappearing concept, it looks like psychologists, and definitely personologists, are presently yet still gradually acknowledging that the psychology of religion is here to stay.

CLINICAL DIMENSIONS The relevance of the above discussion is all the more pertinent given the application of personality and individual differences research to clinical work. In fact, more than ever is it time to consider such issues given that changes in clinical conceptualisations are beginning and that such changes directly involve personality factors. The American Psychiatric Association (2010a) boldly states on its webpage ―DSM-V: The Future Manual‖, while the official DSM-V site heralds the expected publication date of the manual (May 2013) as ―one the most anticipated events in the mental health field‖ (American Psychiatric Association, 2010b). Bells and whistles aside, what is immediately evident from reading through the published draft of the 5th edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-V) is the triumphal influence of personology and individual differences on the shape of things to come in the clinical assessment, diagnosis, and classification of mental

4


health. This influence is most obvious in the proposed changes to Axis II, and particularly to personality disorders, where terms like personality trait domains, trait facets, personality hierarchy, and dimensional assessment of personality functioning—all rather familiar to psychologists—prominently feature in abundance. According to the currently used clinical and categorical (binary) model of psychopathology (American Psychiatric Association, 2000), mental disorders are conceptualised as distinct health-related (latent) taxa, with specific and differentiated properties (symptoms and aetiologies) that an individual either possesses or not at any given point in their life. This position sits comfortably with at least the ―traditional‖ medical model of illness, and initially it may appear to make intuitive sense. After all, and as a layperson may state, one is either healthy or ill. It has gradually, however, become apparent that the categorical model is problematic. The arguably poorly defined and non-evidence based taxa and subclasses of disorders appear to generate rather unreliable and subjective diagnoses, possess problematic construct validities, result in unrealistically large numbers of co-diagnoses (the problem of comorbidity), and seriously limit scientific enquiry into mental illness (for more see Millon et al., 2004; Stark, 2006). Eventually, even the medical establishment acknowledged that the distinction between normality and abnormality is far fuzzier than it was originally conceived and that the current model is practically incapable of allowing a useful and meaningful differentiation between the two (for example, Helzer et al., 2008). As early as the 1970s (although arguably the seed had been planted decades before), with Hans and Sybil Eysenck‘s bold introduction of psychoticism as a fundamental dimension of ―normal‖ personality (e.g., Eysenck & Eysenck, 1976), personologists and clinical psychologists—also acknowledging the sincere efforts from the fields of behavioural genetics and evolutionary psychology—have been, at times, forcefully and explicitly questioning the value of the medical model of mental health, with a particular focus on personality disorders. The accumulating evidence points to the existence of the same personality traits in both nonclinical and clinical populations, the similar heritability levels of disordered and functional traits, common (if not identical) biological and genetic markers in the formation of both, and a cross-cultural correspondence between them (see also Matthews, Deary, & Whiteman, 2009; O‘ Donohue, Fowler, & Lilienfeld, 2007; Zuckerman, 2005). Thus far, the evidence appears to overwhelmingly suggest that personality traits may function as vulnerability to the aetiology and development of mental illness and that, within certain parameters, personality disorders may be both quantitative and qualitative exaggerated, dysfunctional, and nonadaptive expressions of normal personality variations. The inevitable conclusion is that the adoption of a dimensional approach to mental illness is the viable evidence-based solution to the problematic current state of affairs (see also Widiger, Livesley, & Clark, 2009). Consequently within such framework, the diagnostic and assessment discourse centres around levels of mental health functioning and impairment, and contextual or cultural degrees of adaptation. In the DSM-V the initial brave steps toward this approach are evident, with the complete redefinition of personality (and other) disorders and the introduction of a dimensional traits assessment of personality functioning. As a result, personology and individual differences (officially) gain a new status and acquire a far more central role in mental health, with corresponding levels of responsibility and increased demands for further research in the field. It remains to be seen whether the infusion of such knowledge results in this ―future manual‖

Introduction

5

living up to expectations and allowing for diagnoses to be ―more accurate, valid, and clinically useful‖ (American Psychiatric Association, 2010b).

ORGANISATION AND SCOPE With the above considerations in mind, the purpose of this book is twofold. The first general aim is to draw attention to the rich scientific literature that continues to emerge with respect to personality and individual differences psychology. Here the book addresses the underlying processes associated with known and novel personality factors, as well as current critical developments in theory. The second more specific aim is to draw attention to lesser known cross-cultural considerations that impact upon personality and individual differences research. This latter aim provides the book with a global context which is important for reflecting the growing direction in personality and individual differences research. Here the book addresses cross-cultural considerations and devotes special attention to religious factors, noting a wide-range of views on the matter. At the very least, history demonstrates that humans are attracted to the numinous, and so any account of personality should be able to accommodate this aspect of the human psyche. The book is divided into 4 major sections that address the major directions in personality and individual differences research: Part A: Processes & performance Part B: Culture, religion & spirituality Part C: Contributions to clinical psychology & assessment Part D: Developments in theory & structure Part A (Processes & performance) examines both individual and social processes that impact upon behaviour. Nadin Beckmann and colleagues open this section with a chapter that addresses the important issue of the relationship involving between- and within-person variables and their consideration in research in the workplace. Damian Birney follows this up with a discussion of the possibility of training fluid intelligence, a construct commonly noted for its immutability. Lauren O‘Driscoll and Jessica Franklin then (each with Eugene Chekaluk) focus upon the impact of the social environment on self-concept and selfevaluation. Following this, Madeleine Fraser and Simon Boag‘s chapter examines the interaction of person-variables (emotional intelligence) and situational variables, before Gerry Pallier and Niko Tiliopoulos discuss findings related to ‗nonconscious‘ processing, the incubation effect, and its impact upon creativity (as well as theorising why ‗sleeping on a problem‘ may be effective). Paul Ginns and colleagues discuss the role of traits in learning and performance with respect to academic achievement, workplace performance and physical exercise, and then finally, Sabina Kleitman and colleagues examine the important issue of predicting self-confidence in childhood and its relation to metacognition. Part B (Culture, religion & spirituality) discusses applications and findings relevant to understanding and acknowledging the importance of cultural considerations. This section begins with Luh Ketut Suryani discussing the significance of culture and spirituality, and its application to therapy. At the other end of the scale, Lazar Stankov and colleagues then

6


examine militant extremist mindsets, a mindset often associated with extreme religious views. Following this, Leslie Francis and colleagues look at emotional intelligence in church leaders, finding relatively low scores on emotional intelligence compared to other populations. In a similar vein, Lisa Hancock examines the relationship between religious attachment (an extension from attachment theory) and schizotypal traits, finding a relationship between religious attachment and schizotypy in Christian samples. Fiona Hibberd then addresses the issue of secularism in psychology and whether there is a bias against discussion of religious issues, while John Reid looks at right- and left-brain thinkers and considers possible future issues in organisational psychology in a global economy. Finally for this section, Hoyt Edge and colleagues discuss differences between individualist and collectivist cultures through comparing North American and Balinese cultures. Part C (Contributions to clinical psychology and assessment) discusses clinical findings and evaluates current personality and individual differences assessment tools (including cutting-edge discussion of internet assessment). Tara Thompson and colleagues begin this section with a counterpoint to the influence of culture by examining the genetic variation associated with conduct disorder. Next, Marianna Szabó investigates the commonly used Depression Anxiety Stress Scales and its suitability across different phases of the life-span. Kathleen Mairet and Simon Boag‘s chapter then looks at the potential role of unconscious negative schemas underlying bulimia using subliminal presentation and again finding that cultural consideration appear to be important. Weiwen Yang and Esben Strodl next look at the importance of personality variables and motivational interviewing as an adjunct to successful cognitive behavioural therapy. With respect to assessment, Elliroma Gardiner and Chris Jackson address the noteworthy issue of conceptualising and assessing ‗disinhibition‘, a construct linked to impulse control failure, and then Jennifer Keen and Julie Fitness look at issues related to assessing and detecting white-collar fraud, which is all the more salient in the aftermath of the global financial crisis. Richard Hicks then discusses measures of stress and both disadvantages and advantages associated with different approaches to the issue. Finally, Chris Jackson‘s contribution concerns the use of current technologies in personality and individual research, and will be of interest also to anyone wishing to assess and find new online testing facilities. Part E (Developments in theory & structure) discusses the theoretical and conceptual issues surrounding personality and individual differences research. Jonathan Gerber first discusses the issues of circularity and traits psychology before offering a constructive response to rescuing traits from conceptual problems. Niko Tiliopoulos next discusses an alternative ‗circular‘ conception of traits that arises in contrast to the problematic factor analytic derivation of constructs. Finally, Simon Boag‘s chapter argues for the importance of conceptual research as a supplement for empirical research.

FINAL COMMENTS This book arose from ideas generated at the Australasian Conference on Personality and Individual Differences (ACPID) held at the University of Sydney in November, 2009. ACPID has become a focal point for both personality and individual differences researchers in Australia and increasingly throughout the world. In conjunction with this, ACPID provides a

Introduction

7

constructive forum for the exposition and discussion of personality and individual differences research for both established researchers and also up-and-coming researchers. This spirit is found within this volume. The Editors would especially like to thank Mike Jones (deputy Head of Department at Macquarie University) and Chandera Rajah who made this book possible with financial backing. The Editors would also like to extend their gratitude to Sabina Kleitman, Carolyn McCann and Gerry Pallier for help with chapter reviews and finally, we would also like to extend our gratitude to all of the authors who contributed to this book.

REFERENCES Allport, G. W. (1950). The individual and his religion. New York: MacMillan. Allport, G. W., & Vernon, P. E. (1930). The field of personality. Psychological Bulletin, 27, 677-730. American Psychiatric Association (2000). Diagnostic and statistical manual of mental disorders (4th ed., text revision). Washington, DC: Author. American Psychiatric Association (2010a). DSM-V: The future manual. Retrieved from http://www.psych.org/dsmv.aspx American Psychiatric Association (2010b). DSM-V: The future of psychiatric diagnosis. Retrieved from http://www.dsm5.org/Pages/Default.aspx Esau, T. G. (1998). The evangelical Christian in psychotherapy. American Journal in Psychotherapy, 52, 28-36. Eysenck, H. J., & Eysenck, S. B. G. (1976). Psychoticism as a dimension of personality. London: Hodder & Stoughton. Freud, S. (1961). The future of an illusion. In J. Strachey (Ed. & Trans.), The Standard Edition of the Complete Psychological Works of Sigmund Freud (Vol. 21, pp. 5-56). London: Hogarth Press. (Original work published 1927). Gergen, K. J. (2001). Psychological science in a postmodern context. American Psychologist, 56, 803-813. Hall, G. C. (2003). The self in context: Implications for psychopathology and psychotherapy. Journal of Psychotherapy Integration, 13, 66-82. Helzer, J. E., Kraemer, H. C., Krueger, R. F., Wittchen, H. U., Sirovatka, P. J., & Regier, D. A. (Eds.) (2008). Dimensional approaches in diagnostic classification. Refining the research agenda for DSM-V. Arlington: APA. James, W. (2002). Varieties of religious experience. London: Routledge. (Original work published 1902). Jones, D. (2010). A WEIRD view of human nature skews psychologists‘ studies. Science, 328, 1627. Kraut, R., Olson, J., Banaji, M., Bruckman, A., Cohen, J., & Couper, M. (2004). Psychological research online; report of Board of scientific Affairs‘ Advisory group on the conduct of research on the internet. American Psychologist, 59, 105-117. Markus, H. R. (2004). Culture and personality: Brief for an arranged marriage. Journal of Research in Personality, 38, 75-83.

8


Matthews, G., Deary, I. J., & Whiteman, M. C. (2009). Personality traits (3rd ed.). Cambridge: Cambridge University Press. Millon, T., Grossman, S., Meagher, S., Millon, C., & Ramnath, R. (2004). Personality disorders in modern life (2nd ed.). Int.: Wiley. O‘ Donohue, W., Fowler, K. A., & Lilienfeld, S. O. (Eds.) (2007). Personality disorders: Toward the DSM-V. London: Sage. Spilka, B., Hood, R. W. Jr., Hunsberger, B., & Gorsuch, R. L. (2003). The psychology of religion: An empirical approach (3rd ed.). New York: Guilford. Strack, S. (Ed.) (2006). Differentiating normal and abnormal personality (2nd ed.). New York: Springer. van Til, C. (1972). The psychology of religion. Nutley, NJ: Presbyterian & Reformed Pubs. (Original work published 1935). Vande Kemp, H.(1986). Dangers of psychologism: The place of God in psychology. Journal of Psychology & Theology, 14, 97-109. Widiger, T. A., Livesley, J. W., & Clark, A. L. (2009). An integrative dimensional classification of personality disorders. Psychological Assessment, 21, 243-255. Zuckerman, M. (2005). Psychobiology of personality (2nd ed.). Cambridge: Cambridge University Press.

PART A: PROCESSES AND PERFORMANCE



Chapter 2

INTEGRATING THE STUDY OF WITHIN- AND BETWEEN-PERSON VARIABILITY IN PERSONALITY AT WORK Nadin Beckmann, Amirali Minbashian Accelerated Learning Laboratory, University of New South Wales, Sydney, NSW 2050, Australia

and Robert E. Wood Melbourne Business School, Carlton, VIC 3053, Australia

ABSTRACT Research on personality in the workplace has primarily focussed on differences between individuals in their typical thoughts, feelings and behaviours as represented in the five-factor model of personality (McCrae & Costa, 1999). We argue that a withinperson conceptualisation of personality offers additional meaningful insights into the role of personality in general and at work in particular. In this chapter we present a multi-level framework that integrates between- and within-person approaches to the study of personality in the workplace. We outline questions of interest that arise within this approach, provide examples from the literature that illustrate each question and highlight future research directions.

Keywords: Five-factor model, Experience sampling, Organisational psychology, Personality, Within-person variability



Ph: 61-2-9931 9187; Fax: 61-2-9931 9199; [email protected]

12

Nadin Beckmann, Amirali Minbashian and Robert E. Wood

INTRODUCTION An important but largely unresearched topic in personality research within organisational contexts relates to the variability in personality responses that occurs within a person across time and situations. Although much research has focussed on personality in the workplace, this work has primarily focussed on differences between individuals in their typical thoughts, feelings and behaviours as represented in the five-factor model of personality (McCrae & Costa, 1999), and the relation of these to work outcomes of interest (e.g., Barrick, Mount & Judge, 2001). In trait models, such as the five-factor model, which study between-person effects, variability that occurs within a person is treated as error variance. In this chapter we will present a multi-level framework that integrates between- and within-person approaches to the study of personality in the workplace. We will outline some questions of interest that arise within this approach, provide examples from the literature that illustrate each question and highlight future research directions.

BACKGROUND TO AN INTEGRATED APPROACH Over the past twenty years, the study of personality in organisational psychology has been dominated by traits; that is, relatively stable personality characteristics that influence thoughts, feelings, and overt behaviour (e.g., McCrae & Costa, 1999). The most widelyvalidated trait model—the five-factor model (FFM; Costa & McCrae, 1992)—describes personality in terms of five broad dimensions labelled neuroticism, extraversion, openness to experience, agreeableness, and conscientiousness. The trait approach as represented by the FFM has provided a useful framework for organisational psychologists to describe individual differences on fundamental personality characteristics. In particular, it has allowed us to identify which of these between-person differences in personality are relevant for work. For example, meta-analytic studies have shown that individuals who tend to be more conscientious and/or less neurotic tend to perform better across a wide range of occupations (Barrick et al., 2001). Furthermore, by providing a taxonomy of the major personality factors the FFM has enabled the integration of seemingly heterogeneous findings regarding the role of personality in work settings (e.g., Barrick & Mount, 2005; Borkenau, Egloff, Eid, Henning, Kersting, Neubauer & Spinath, 2005). Although the trait approach has yielded major contributions to the development of personality research in organisational psychology, its emphasis has primarily been on descriptions of people relative to each other on relatively stable characteristics. Less emphasis has been placed on within-person variation, which itself is an important topic within the field of personality (Cervone, 2004). Within-person variability was most prominently studied by Mischel (1973) in the 1970‘s who argued that a considerable amount of variation in cognition, affect, and behaviour occurs within an individual and that this variation can be explained by the impact of the situation. In response, over the past two decades personality research has adopted a model that includes both trait-concepts and situations as sources of variability (see Matthews & Deary, 1998). This has been referred to as an interactionist approach. Within this approach within-person variability is captured by the concept of ‗states‘. In contrast to traits, states are conceptualised as aspects of personality that are

Integrating the Study of within- and between-Person Variability…

13

variable across situations and time. Specifically, traits interact with the situation to influence an individual‘s state at a given point in time. However, within organisational contexts, the emphasis has been on between-person differences, even when an interactionist approach has been adopted. For example, when predicting job performance using conscientiousness as a predictor, Barrick and Mount (1993) took into account the influence of the situation by including job autonomy as the situational moderator. That is, they examined how individuals who are working under different levels of autonomy compare on their job performance; not how an individual‘s job performance varies across situations with different levels of autonomy. Figure 1 illustrates the difference between the two types of variability (i.e., betweenversus within-person) discussed above. The figure plots the mean levels and variation in effort expended by two employees (rated on a scale of 0 to 10) conditional on one situational characteristic, namely who is present. Within-person variability in effort across different situations is represented by the profiles (solid lines), whereas between-person variability is represented by the differences between the mean effort levels (dashed lines). Integrating within-person variability into the study of personality at work is important because of at least three reasons: (1) Previous research within the general personality literature has shown that within-person variability comprises a large part of the total variability in behaviour (Fleeson, 2001). Consequently, in order to get a complete picture of personality at work the two sources of variance (within-person and between-person) need to be simultaneously taken into account; (2) Within-person effects often differ between people and these differences capture meaningful components of personality. For example, it has been argued that situation contingent within-person effects may provide insights into individual differences in adaptability (Fleeson & Jolley, 2006). Assessing these differences and their correlates will provide insights into aspects of personality that are not well captured by a purely between-person trait approach; (3) Between-person relations are often used as proxies for inferences that relate to within-person effects. Specifically, researchers often articulate theories about the psychological processes that occur within an individual, but then collect data to test these theories at the between-person level. However, the two levels are conceptually and statistically distinct, and, therefore, conclusions that are made at one level do not necessarily generalise to the other level (Nezlek, 2001). In recent years the study of within-person variability has been advanced by developments in methodological and data-analytic techniques. These techniques facilitate the collection and analysis of data that involves frequent reassessments of individuals and has a multilevel structure. Specifically, experience sampling methodology (ESM), while known for several decades, has now become a practical option for collecting large-scale within-person data due to advances in computer technology such as the advent of handheld computers (see FeldmanBarrett & Barrett, 2001); additionally, developments in hierarchical linear modelling and associated software provide a means by which researchers can appropriately address the nested structure of such data (see Raudenbush & Bryk, 2002). We discuss experience sampling and hierarchical linear modelling in greater detail below.

14

Nadin Beckmann, Amirali Minbashian and Robert E. Wood 10 9 8

Effort

7 P1 Profile P2 Profile P1 Mean P2 Mean

6 5 4 3 2 1 0 supervisor

peer

subordinate

no one

Pe rs on Pre s ent

Figure 1. Response patterns for effort levels of two individuals (P1 & P2) as a function of people present in the situation for: (a) the average level of effort across situations (dashed lines), & (b) the variation in effort across situations where different people are present (solid lines).

Experience Sampling Method (ESM) To study the within-person aspects of personality we require a design that allows measurement of the individual‘s momentary thoughts, feeling and behaviours, and the variation in these states over time and situations. Experience sampling is a methodology that allows frequent sampling in a natural context without being too intrusive. This method has begun to be used in organisational research (e.g., Miner, Glomb, & Miner, 2005) as it allows for short, frequent measures of feelings, thoughts and behaviours that occur across the workday. By assessing these states as they occur, ESM designs have been argued to be more ecologically valid and less prone to the biases in recall that affect retrospective reports of behaviour (e.g., Fisher, 2002). ESM is typically implemented by sending individuals signals (e.g., via handheld computers or email) at different times during the day over several weeks that request them to report on properties of the situation and their thoughts, feelings and behaviours at that point in time. The sampling design for the timing of the signals and the measures will depend on the specific hypotheses researchers are testing.

Hierarchical Linear Modelling Integrating within-person and between-person variability raises several data-analytic complications that are not present in studies that focus solely on between-person differences (see Hofmann, 1997). In particular, the nested structure of the data (observations are clustered within individuals) creates dependency among the observations that violates the independence of observations assumption associated with the traditional ordinary least squares (OLS) analysis. Consequently, use of OLS to analyse multilevel data results in inaccurate (downwardly biased) estimates of standard error. In contrast, hierarchical linear modelling (HLM) explicitly and simultaneously models each level of analysis—including the residuals


15

at each level—in terms of its own submodel. In this way, HLM accounts for the interdependence between observations that are clustered within individuals. Furthermore, it allows one to disentangle between-person and within-person effects on the outcome of interest, and to examine cross-level interactions. Use of hierarchical linear modelling has been limited in the past due to the practical difficulties associated with estimating the model parameters, however the emergence of several statistical programs for accomplishing this purpose has resulted in an increased use of such models over the past decade (see Raudenbush & Bryk, 2002). Within-person relationships (e.g., between situations and states) can be modelled by parameters at level 1 and between-person relationships (e.g., between traits and mean states) can be modelled by parameters at level 2. Furthermore, cross-level interactions (e.g., the effects of traits on within-person situation-state relationships) can also be represented and tested by parameters at the second level.

QUESTIONS WITHIN AN INTEGRATED APPROACH An integrated approach to studying personality at work lends itself to several types of questions that include but also go beyond the between-person questions that have typically been addressed in studies of personality at work (see Nezlek, 2007). In this section we outline and elaborate on these questions and provide illustrative examples from the literature. 



Between-person relationships: Individuals differ systematically in their typical levels of cognitive, affective and behavioural states as a function of traits Between-person relationships are the typical types of relationships that have been examined in studies on personality at work. They include findings from metaanalyses in which the association between personality variables and cross-sectionally assessed outcomes of interest, such as job performance and job satisfaction, are estimated (e.g., Barrick & Mount, 1991; Judge, Heller, & Mount, 2002). This category also includes studies in which between-person differences in personality dispositions are related to between-person differences in aggregated levels of momentary states, such as positive and negative affect (e.g., Fisher, 2002). Within-person relationships: An individual’s states vary systematically as a function of the nominal and psychological properties of different situations as well as a function of other states In recent years, within-person relationships have received increasing attention in the organisational literature, although this body of research has primarily focused on state constructs that are distinct from personality, such as task performance (Fisher & Noble, 2004), job satisfaction (Ilies & Judge, 2002) and organisational citizenship behaviour (Ilies, Scott, & Judge, 2006). These studies indicate that within-person variability typically accounts for a substantial amount of the total variability in the construct of interest. Moreover, this variability is explicable as a function of momentary situational characteristics such as task difficulty (Fisher & Noble, 2004), as well as affective states (Ilies & Judge, 2002). In a recent study, we examined whether these findings extend to momentary variations in the conscientiousness

16






levels of managers during three working weeks (Minbashian, Wood, & Beckmann, 2009, in press). As argued by Fleeson (2001), this represents a strong test of the viability of studying within-person aspects of personality given the assumptions of cross-situational consistency traditionally associated with the FFM. Nevertheless, we found that within-person variability in conscientiousness was approximately two times greater than between-person difference in conscientiousness, and that psychological properties of the tasks individuals were engaged in (importance, difficulty, urgency) accounted for the majority of this variance. Cross-level interactions: The within-person relationship between situations and states or between states and states, respectively, vary as a function of differences between people in traits In studies of within-person relationships the magnitude and/or direction of within-person effects frequently differ between people. These individual differences are often outcomes of interest in their own right and can be modelled as a function of personality traits and other individual difference variables, thus defining a third type of question that arises within an integrated approach. For example, Ilies et al. (2006) found significant differences between people in the extent to which momentary positive affect was associated with organisational citizenship behaviour (OCB). This effect was moderated by agreeableness such that individuals low on agreeableness tended to display a strong association between positive affect and OCB, whereas highly agreeable individuals tended to display high levels of OCB regardless of their affect levels. In contrast, Ilies and Judge (2002) found that individual differences in the within-person association between affect and momentary reports of job satisfaction were not significantly accounted for by either extraversion or neuroticism, although the latter was related to within-person variability in negative affect and job satisfaction. Such findings help to shed light on the way in which the effects of traits are reflected in everyday work situations. Comparisons of relationships at different levels: The between-person relationship between two constructs is not necessarily equivalent to the within-person relationship between the constructs Findings from studies implementing between-person designs do not necessarily generalise to within-person phenomena. For instance, the between-person finding of a five-factor model of personality does not imply the existence of a five-factor structure that operates within most individuals. This is because, statistically, models of between-person variability are not related to models of within-person variability (Nezlek, 2001; Schmitz, 2006). While the distinctness of the two models has been discussed in the literature (e.g., Cervone, 2005, see also lead paper by Cervone, et al., 2006 and associated commentaries), there has been little systematic empirical investigation of this issue (Borsboom, Mellenbergh, & van Heerden, 2003). Few studies have directly compared findings with regard to the structure of personality at the between-person level with findings at the within-person level. In one study Borkenau and Ostendorf (1998) collected data from 22 students once a day over 90 consecutive days employing a German adjective-based FFM measure. Individual within-person factor structures of FFM relevant states were rather weakly related to the overall between-person reference factor structure of FFM traits as assessed in a different sample. A stronger match, however, was found when the average within-


17

person factor structure was considered. In our own research we have looked at the relationship between two FFM dimensions, neuroticism and conscientiousness, from a between-person and a within-person perspective (Beckmann, Minbashian, & Wood, 2009, in press). We found that while neuroticism and conscientiousness were negatively related at the between-person level (a finding that is well documented in the literature, see Mount, Barrick, Scullen and Rounds, 2005), at the level of the individual this relationship was reversed. Further empirical evidence for the distinctness of within- and the between-person effects can be found in other areas of psychology, for instance, in the field of self-regulation (Vancouver, 2001, 2006).

CONCLUSION Personality has increasingly become an important topic in organisational psychology. Much of this research has relied on between-person comparisons. In this chapter we argued that the field would benefit from an approach that incorporates within-person aspects of personality. We have presented a framework that illustrates the types of research questions that can follow from this integrated approach. Furthermore we outlined some initial studies that fall within each type of research question. In the following paragraphs we list three areas of future research: 





Broader focus. More research is required to evaluate within- versus between-person effects on a broader range of personality related constructs. To date, most of the research has focused on a limited number of broad personality dimensions, such as conscientiousness and neuroticism (Minbashian, et al., 2009; Beckmann et al., 2009). Yet, the personality domain comprises a multitude of constructs, both broad and narrow that are likely to be relevant for a work context, such as openness to experience (Barrick, et al., 2001), or goal orientation (Payne et al., 2007). Conceptual clarity. Research is required to provide theoretical clarity about the types of effects studied at the within- and between-person levels. Questions of interests include: a) whether within-person effects (e.g., situation-state contingencies, such as an individual‘s responsiveness to situational cues) are stable across macro-contexts; b) whether these within-person effects are assessing constructs that are distinct from traditional trait constructs (rather than representing different methods of assessing the same constructs), and if so; c) how they fit in the nomological net of the traditional trait constructs. Malleability. Traditionally, trait approaches to personality have focused on the stable aspects of personality and, consequently, there have been few applications designed to develop personality in a way that facilitates performance in work contexts. It is likely that within-person aspects of personality are more modifiable. Unlike trait measures that have been traditionally routed in biology, within-person effects are thought to reflect cognitive factors that arguable are more amenable to intervention (Mischel & Shoda, 1995). However, this is an empirical question that has received little attention to date in the organisational literature.

18

Nadin Beckmann, Amirali Minbashian and Robert E. Wood 

Implications for work outcomes. Finally, studies are required to evaluate the implications of within-person effects for predicting and explaining work outcomes. In particular, from a practical perspective an important question relates to whether within-person measures provide incremental validity over what is already accounted for by traditional trait measures. Furthermore, a study that addresses this issue can also provide insight into the day-to-day processes by which personality traits influence important work outcomes; that is, one can explore the extent to which within-person constructs mediate the trait-job performance relationships that have traditionally been observed. This knowledge is relevant for organisational psychologists who are interested in identifying the underlying psychological processes that can be targeted as part of employee learning and development programs in order to implement behavioural change.

Integrating within- and between-person variability into the study of personality at work represents an exciting direction for future research. Pursuing this approach will lead to a more comprehensive picture of the individual in their work environment than captured by a purely between-person trait perspective.

REFERENCES Barrick, M. R. & Mount, M. K. (1991). The Big Five personality dimensions and job performance: A meta-analysis. Personnel Psychology, 44, 1-26. Barrick, M. R. & Mount, M. K. (1993). Autonomy as a moderator of the relationships between the Big Five personality dimensions and job performance. Journal of Applied Psychology, 78, 111-118. Barrick, M. R. & Mount, M. K. (2005). Yes, personality matters: Moving on to more important matters. Human Performance, 18, 359-372. Barrick, M. R., Mount, M. K., & Judge, T. A. (2001). Personality and performance at the beginning of the new millennium: What do we know and where do we go next? International Journal of Selection and Assessment, 9, 9-30. Beckmann, N., Wood, R.E., & Minbashian, A. (2009, 25-28 November). It depends how you look at it—on the relationship between conscientiousness and neuroticism. Paper presented at the 8th Australian Conference on Personality and Individual Differences, Sydney, Australia. Beckmann, N., Wood, R.E., & Minbashian, A. (in press). It depends how you look at it: On the relationship between conscientiousness and neuroticism at the within- and betweenperson levels of analysis. Journal of Research in Personality. Borkenau, P. & Ostendorf, F. (1998). The Big Five states: How useful is the Five-Factor Model to describe intraindividual variation over time? Journal of Research in Personality, 32, 202-221. Borkenau, P., Egloff, B., Eid, M., Henning, J., Kersting, M., Neubauer, A. C., & Spinath, F. M. (2005). Persönlichkeitspsychologie: Stand und Perspektiven [Personality psychology: Current state and future prospects]. Psychologische Rundschau, 56, 271-290.


19

Borsboom, D., Mellenbergh, G. J., & van Heerden, J. (2003). The theoretical status of latent variables. Psychological Review,110, 203-219. Cervone, D. (2005). Personality Architecture: Within-person structures and processes. Annual Review of Psychology, 56, 423-452. Cervone, D. (2004). The architecture of personality. Psychological Review, 111, 183-204. Cervone, D., Shadel, W. G., Smith, R.E., & Fiori, M. (2006). Self-regulation: Suggestions and reminders from personality science. Applied Psychology: An International Review, 55, 333-385. Costa, P. T., Jr., & McCrae, R. R. (1992). Revised NEO Personality Inventory (NEO-PI-R) and NEO Five-Factor Inventory (NEO-FFI) manual. Odessa, FL: Psychological Assessment Resources. Feldman Barrett, L. & Barrett, D. J. (2001). An introduction to computerized experience sampling in psychology. Social Science Computer Review, 19, 175-185. Fisher, C. & Noble, Ch. S. (2004). A within-person examination of correlates of performance and emotions while working. Human Performance, 17, 145-168. Fisher, C. D. (2002). Antecedents and consequences of real-time affective reactions at work. Motivation and Emotion, 26, 3-30. Fleeson, W. & Jolley, S. (2006). The challenge and the opportunity of intraindividual variability: A theory of human flexibility. In D. Mroczek & T. D. Little (Eds.). Handbook of personality development (pp.41-60). Mahwah, NJ: Lawrence Erlbaum. Fleeson, W. (2001). Towards a structure- and process-integrated view of personality: Traits as density distributions of states. Journal of Personality and Social Psychology, 80, 10111027. Hofmann, D. A. (1997). An overview of the logic and rationale of hierarchical linear models. Journal of Management, 23, 723-744. Ilies R. & Judge, T. A. (2002). Understanding the dynamic relationships among personality, mood, and job satisfaction: A field experience sampling study, Organizational Behavior and Human Decision Processes, 89, 1119-1139. Ilies, R., Scott, B. A., & Judge, T. A. (2006). The interactive effects of personal traits and experienced states on intraindividual patterns of citizenship behavior. Academy of Management Journal, 49, 561-575. Judge, T. A., Heller, D., & Mount, M. K. (2002). Five-Factor Model of personality and job satisfaction: A Meta-Analysis. Journal of Applied Psychology, 87, 530-541. Matthews, G. & Deary, I. J. (1998). Personality traits. Cambridge, UK: Cambridge University Press. McCrae R. R. & Costa P. T., Jr. (1999). A Five-Factor Theory of personality. In L. A. Pervin and O. P. John (Eds.), Handbook of personality psychology (pp.139-155). New York: Guilford. Minbashian, A., Wood, R. E., & Beckmann, N. (in press). Task-contingent conscientiousness as a unit of personality at work. Journal of Applied Psychology Minbashian, A., Wood, R. E., & Beckmann, N. (2009). Contingent beliefs as predictors of within-person variation in conscientiousness at work. In George T. Solomon (Ed.), Proceedings of the Sixty-Sixth Annual Meeting of the Academy of Management (CD), ISSN 1543-8643.

20


Miner, A. G., Glomb, T. M., & Hulin, C. (2005). Experience sampling mood and its correlates at work. Journal of Occupational and Organizational Psychology, 78, 171193. Mischel, W. & Shoda, Y. (1995). A cognitive–affective system theory of personality: Reconceptualizing situations, dispositions, dynamics, and invariance in personality structure. Psychological Review, 102, 246-268. Mischel, W. (1973). Toward a cognitive social reconceptualization of personality. Psychological Review, 80, 252-283. Mount, M. K., Barrick, M. R., Scullen, S. M., & Rounds, J. (2005). Higher-order dimension of the Big Five personality traits and the Big Six vocational interest types. Personnel Psychology, 58, 447-478. Nezlek, J.B. (2001). Multilevel random coefficient analyses of event- and interval-contingent data in social and personality psychology research. Personality and Social Psychology Bulletin, 27, 771-785. Nezlek, J. B. (2007). A multilevel framework for understanding relationships among traits, states, situations and behaviours. European Journal of Personality, 21, 789-810. Payne, S. C., Youngcourt, S. S., & Beaubien, J. M. (2007). A meta-analytic examination of the goal orientation nomological net. Journal of Applied Psychology, 92, 128-150. Raudenbush, S.W., & Bryk, A. S. (2002). Hierarchical linear models (2nd ed.). Newbury Park, CA: Sage Publications. Schmitz, B. (2006). Advantages of studying processes in educational research. Learning and Instruction, 16, 433-449. Vancouver, J. B., Thompson, Ch. M., & Williams, A. A. (2001). The changing signs in the relationships among self-efficacy, personal goals, and performance. Journal of Applied Psychology, 86, 605-620. Vancouver, J. B. & Kendall, L. N. (2006). When self-efficacy negatively relates to motivation and performance in a learning context. Journal of Applied Psychology, 91, 1146-1153.



Chapter 3

DELIBERATE PRACTICE AND THE TRAINABILITY OF FLUID COGNITIVE FUNCTIONS Damian P. Birney Accelerated Learning Laboratory, University of New South Wales Sydney, NSW 2052, Australia

ABSTRACT Over 100 years of research has failed to produce training programs that lead to demonstrable and sustained increases in fluid intelligence (Gf). We report on the work of Jaeggi, et al. (2008) who showed Gf-gains obtained by training working memory using a deliberate-practice-like approach. While promising, the work is not conclusive and we argue that future research needs to be very clear about what attributes are being modified, through what processes, and via what training-methods. We believe an investigative framework of deliberate practice will be fruitful for such work. Closer investigations of how training in different fluid cognitive functions leads to Gf gains which have a practical impact on everyday functioning are required. Implications for theory and test development, education and training programs, as well as ultimately for socio-political stereotypes are significant. To achieve this, headway will need to be made in overcoming a century of inertia surrounding traditional psychometric assessment and theorising.

Keywords: Deliberate Practice, Fluid Intelligence, Working Memory, Training

INTRODUCTION In this chapter we dissect and investigate the claim made by Jaeggi, Buschkuehl, Jonides, et al. (2008) that they have achieved what over 100 years of education and psychology research has failed to do—to increase fluid intelligence (Gf). The increases in fluid 


22

Damian P. Birney

intelligence were achieved through training working memory (WM) using a variant of the nback task. As will be discussed below, central to their success is a presentation routine that prevented (or at least minimised) the automatisation of solution strategies by participants. It will be argued that this is, in essence, a deliberate practice routine as prescribed by Ericsson and his colleagues (Ericsson, 2003; Ericsson, Krampe & Tesch-Römer, 1993). We outline core issues that future research needs to address and discuss the socio-political importance and implications of this research.

JAEGGI, ET AL.’S RESEARCH In the typical n-back task, participants are shown a series of stimuli (e.g., letters, sounds, spatial locations), and for each stimulus are required to indicate whether it matches the identity of the stimulus that preceded it by n-places in the series (see Figure 1 where n = 2). The n-back task is considered a continuous WM task because participants must maintain and update (and presumably rehearse) a set of items while responding to each current item. The cognitive demands are thus theoretically consistent with the simultaneous storage and processing requirement central to many theories of working-memory (Conway, Jarrold, Kane, Miyake & Towse, 2007). Jaeggi, et al.‘s (2008) training sessions entailed an adaptive presentation of the n-back task that the authors believe minimised automaticity. We will outline their research here in some detail because of the remarkable simplicity of the training design and the significance of its impact on individual Gf gains. Participants were 69 healthy University of Bern students (mean age = 25.6; SD = 3.3). The WM training entailed presentation of varyingly difficult versions of the n-back task contingent on the individual‘s performance on previous trial sets. Difficulty was manipulated in terms of changes in n (1, 2, 3-back). Automaticity was further minimised by mixing the modality of presentation under dual-task conditions. That is, participants had to simultaneously maintain and update different item sets for two different nback tasks—one in an auditory mode and the other in a visuospatial mode. The auditory nback task consisted of one of eight consonants presented in a female voice. The visuospatial n-back task consisted of the presentation of visual stimuli located at one of eight possible screen locations. The rate of presentation of each component was synchronised to coincide and only match responses were required. The difficulty of the task was tailored to the individual. As individual performance changed, n incremented accordingly—up by 1 item for improved performance, and down by 1 item when performance worsened. The amount of nback practice was manipulated across four different groups—8 days (16 participants), 12 days (22 participants), 17 days (16 participants), and 19 days (15 participants). Daily practice consisted of 20 blocks of 20+n trials which equated to a daily training time of approximately 25mins. Change in Gf was evaluated using ‗parallel‘ versions of Raven’s Advanced Progressive Matrices and a short version of the Bochumer Matrizen-Test matrix reasoning task at pre- and post-test (before and after training). The results indicated that relative to a control group, not only did WM training result in greater gains in Gf, but the magnitude of the gains were a function of training ‗dosage‘—the more training, the greater the gains in Gf.

Deliberate Practice and the Trainability of Fluid Cognitive Functions

23

2-Back Task No Match

G

1 No Match

D Match

2

G

3

R

N+1 Lure Trial

4

D

5

Figure 1. The 2-back task. Participants are presented each item (G, D, G… etc) one at a time and are required to indicate whether the current stimuli is the same as the one presented 2-back. Item 2 (‗D‘) is a N+1 lure for Item 5 (‗D‘). In responding to Item 5, participants must inhibit the tendency to incorrectly decide match which is triggered by the temporal location of Item 2’s ‗D‘ to the target location of Item 3 (‗G‘ = no match) & the fact that it primes a match response with the Item 5 ‗D‘.

EVALUATION OF JAEGGI, ET AL.’S FINDINGS Jaeggi, et al.‘s (2008) results are remarkable for a number of reasons. First, their sample consisted of healthy adults in their mid-20‘s. The general consensus regarding the developmental trajectory of Gf is that it rises from birth to plateau in the late teenage years from which time on there is a general, gradual decline (McArdle, Ferrer-Caja, Hamagami & Woodcock, 2002). These trajectory findings are consistent with the common interpretations of neuronal plasticity data (Berlucchi & Buchtel, 2009; Garlick, 2002). Significant neural change, which is presumably happening during Jaeggi, et al.‘s WM training, is more likely to occur in children rather than adults, because neuronal plasticity is reduced in adults. Second, research in educational psychology often report negative correlations between the amount of post-intervention gain and pre-intervention starting point (Sternberg & Grigorenko, 2002). In ability domains, this negative-gain effect is sometimes interpreted as the result of high-ability students simply having less room to improve than students who begin with less ability. In comparison to a general population, university students are very capable and (typically) already have quite high Gf levels. The student population used by Jaeggi, et al. (2008) would be expected to gain less simply because they are starting from an already high level. In short, that Jaeggi, et al. report measurable Gf gains over a relatively short period of time in a high-ability adult sample is impressive.

WHY HAS THIS INTERVENTION WORKED? Jaeggi, et al.‘s (2008) is not the first attempt to develop Gf. There have been numerous programs over the years developed to train WM and attentional resources associated with cognitive control. Most recently, the clear majority of these have been situated in the

24

Damian P. Birney

cognitive neurosciences (Braver, Paxton, Locke & Barch, 2009; Rougier, Noelle, Braver, Cohen & O'Reilly, 2005) and have focused on identifying prefrontal cortex regions associated with various executive process manipulations. Transfer performance has typically been evaluated using untrained tasks that are similar in structure to the trained tasks. There is little research looking at modifiability of higher-order functions like Gf, or at implications for everyday function. A notable exception is the work of Holmes, Gathercole, and Dunning (2009). They instigated an adaptive WM training program (primarily with complex-span like tasks) with a small sample of 10yr old children (n=22) who had been previously assessed to have poor WM (the program was adaptive in the sense that training was targeted at children‘s current span level). These children performed significantly better on other untrained WM tasks compared to children in the non-adaptive condition who trained only on tasks that had a list length of 2 (so as not to tax WM). The trained WM and the untrained (transfer) WM tasks were comparatively similar in structure. It is important to note, the training and criterion tasks used by Jaeggi, et al. (2008) were not structurally similar. Holmes, et al. (2009) found that their adaptive WM training was associated with greater numerical reasoning performance when assessed 6 months later, compared to the control group. However, unlike Jaeggi, et al. there were no measurable increases in Performance IQ of the Wechsler test (the Gf proxy used in the study). In summary, Holmes, et al. (2009) showed improvements on some tasks following WM training, but not on all tasks, and unfortunately, not on Gf tasks. Jaeggi, et al. (2008) argue that their intervention worked because their form of n-back training engaged activities required for the management of executive processes for responding to two n-back tasks simultaneously. Further, and most importantly from our perspective is that the task format ―…discouraged the development of task-specific strategies and the engagement of automatic processes.‖ (p. 6830). The authors speculate that this is because of the tailored, adaptive variation in n-back difficulty and because the two n-back tasks drew on different modalities. We believe this is consistent with the concept of deliberate practice which will be elaborated in the following section (Ericsson, 2003; Ericsson, et al., 1993).

CAN DELIBERATE PRACTICE LEAD TO GF GAINS? Ericsson et al. (1993) used the term deliberate practice as a means to explain how experts in a given domain arrive at a level of performance qualitatively different from people who have been involved in the same domain for just as long, but who have not achieved such high levels. Deliberate practice is differentiated from practice alone in that it is: a) highly structured; b) effortful, and; c) designed to continually stress the individual to improve performance beyond a level most people would consider more than sufficient. As their primary case in point, Ericsson and his colleagues (extending on earlier research, Simon & Chase, 1973) have analysed the practice routines of chess players who differed in levels of expertise (i.e., novice, expert, and grand-masters). They concluded that the maximal levels of performance obtained by grand-masters are not automatically achieved by simply extending the amount of time playing chess. The aspect of experience that most strongly shapes the development of expertise is deliberate practice (Ericsson, 2002, 2005). Within the domain of higher-order problem-solving and knowledge acquisition, deliberate practice seeks to refine


25

knowledge representations while avoiding automatisation. Automatisation here can be thought to refer to the development of eventually over-learnt strategies that might be triggered unthinkingly when certain stimuli in the environment occur. While automatisation of strategies frees resources for other activities, it tends to inhibit further learning improvements (Wood, Beckmann & Birney, 2009). Jaeggi, et al. (2008) make direct reference to the goal of preventing automatisation in their WM training. The question here is whether deliberate practice, as conceptualised by Ericsson (2002, 2005), can be used as a framework for explaining Gf gains reported by Jaeggi, et al. While Ericsson (2003) suggests that performance and its mediating biological mechanisms are modifiable as a result of extended deliberate practice, he presents no clear evidence that this is possible for basic cognitive processes (like Gf). Most of the examples cited are drawn from domains where gains could be obtained by a) better use of putatively unmodifiable, possibly innate, cognitive resources, or b) from physical domains such as sports, where aerobic training might lead to physiological changes that can be integrated with better use of cognitive resources (anecdotally leading to, for instance, superior ―eye-hand‖ coordination). As an example of the former, Chase and Ericsson (1982) conducted a shortterm memory training study where they tested individuals on a simple digit-span task every second day for many weeks and months. After 50h of practice, all subjects improved performance by 200% (~20 digits). After 200-400h of practice, two participants performance had increased by 1000% (~80 digits). Ericsson‘s (2003) interpretation was that through deliberate practice—adaptively extending the to-be-recalled list as individuals improved—the participants had acquired skill for storing and accessing long-term memory. That is, they had expanded their functional WM. The implicit assumption here is that while skills and strategies for using WM became more effective (increasing its function), there is no evidence that the underlying components of WM per se had changed. As an example of expertise development in the physical domain, Ericsson (2003) quotes the typical finding that physically fit adults (e.g., university students) can make around 20 push-ups in a row (with a range of 8-32). In 1966, a record was set by an individual who was able to complete 6000 push-ups. At the time of writing, Ericsson (2003) reported that the then current record, limited by how many push-ups can be achieved in a 24hr period, was 26000. These gains were attributed to deliberate practice. Our conclusion is that the procedure used by Jaeggi, et al. (2008) seems to fit well with Ericsson‘s (2003) deliberate practice framework. However, while deliberate practice research in cognitive and physical domains has shown that it is possible to vastly increase performance on trained tasks, it is not clear (at least in our evaluation of cognitive performance) that what is modified is truly transferable. That is, using the computer analogy of Newell and Simon (1972), it is not clear a priori whether that WM training results in deep structural changes to hardware, or whether it is simply tweaking software. Closer theoretical and empirical consideration of the nature of the attributes Jaeggi, et al. refer to is needed before the deliberate practice framework can be accepted as a plausible general account of the observed Gf gains.

26

Damian P. Birney

WHAT IS BEING MODIFIED? IMPLICATIONS FOR CONSTRUCT DEFINITIONS Jaeggi, et al.‘s (2008) research spanned between 8 and 19 days with no ongoing followup evaluations of the sustainability of the gains observed. It is important for sustainability to be evaluated further because interventions designed to improve intelligence in the past have typically been conducted only with children and any putative Gf-like gains, if they have occurred, have not persisted (Holmes, et al., 2009; Neisser, et al., 1996). For the remainder of this chapter, rather than go into the details of any one study, we reflect on the broader issues that we believe all research in this area needs to be cognizant of—construct definitions issues and socio-political ramifications. Fluid intelligence, working memory, executive functions, and to some extent learning, have all been conceptualised as fluid cognitive functions (FCFs), a higher-order catch-all for a range of cognitive processes that are not contingent on the specific content of the information being processed (Blair, 2006). However, FCF turns out to be a rather nebulous conceptualisation once one begins to drill down to component functions. This is because while it might seem appropriate to include Gf, WM, and executive function under the same overarching umbrella, the constituents of each and the relationships between them are not always apparent theoretically (Birney, Bowman & Pallier, 2006) nor empirically. To understand the nature and broader implications of Jaeggi, et al.‘s results, we need to know what is being modified and whether these can be related to the effects of deliberate practice. In the next section, we provide a brief analysis of the issues that need to be considered in dissecting the claim made by Jaeggi, et al. (2008). Particular emphasis is given to issues associated with the definitions and operationalisations of Gf (including the learning and plasticity debate) and WM (including the executive function debate). We also reflect on some of the potential non-cognitive variables that might moderate WM-training effects through deliberate practice.

FLUID INTELLIGENCE (GF) Definitions of Gf Fluid intelligence could be considered the prototypical FCF. It is frequently considered to be at a conceptually higher level than WM and executive processes. Gf has developed meaning from within the differential psychology domain (Horn & Cattell, 1966) where theoretical constructs are defined by a nomological network of tests. Using correlational techniques (such as structural equation modelling), Gf is empirically defined as the latent trait extracted from a variety of reasoning-dominated psychometric tests. If Raven’s Progressive Matrices (the Gf measure used by Jaeggi, et al.) is included in such a battery, it typically dominates the extracted factor. Gf is considered perhaps the most important, some would say quintessential, aspect of intelligence primarily because it has been linked to many everyday success outcomes, such as schooling and job performance (Jensen, 1998).


27

Gf and Learning A gap remains between how Gf is defined conceptually and how it is often empirically operationalised. For instance, despite a common definition of Gf as the capacity to learn, few measures of it are explicitly structured to provide the opportunity of learning throughout the test (Carlstedt, Gustafsson & Ullstadius, 2000). Instead, research has tended to focus more on the problem-solving aspect of Gf than on the learning ability aspect (Feldman Barrett, Tugade & Engle, 2004). This is an interesting outcome because Gf is also considered empirically distinct but theoretically related to Crystallised Intelligence (Gc). The latter reflects acquired (i.e., learnt) acculturated knowledge. Gc, like Gf, is defined as a latent trait and is typically estimated from tests of verbal based acculturated knowledge. The theoretical link between Gf and Gc is that Gc is thought to develop through the application of Gf (Ackerman, 1996; Ackerman & Beier, 2005; Cattell, 1987). Gf is critical to learning and knowledge acquisition. A possible reason why operationalisation of Gf has focused on reasoning (rather than learning) may be due to the restrictive and theoretically arbitrary assumption of many theories of intelligence—that cognitive abilities are stable and mostly immutable (Birney & Sternberg, 2006). Historically, the immutability of Gf was also supported by early informationprocessing theorists such as Newell and Simon (1972) who suggested that acquired knowledge and skills were like a computer‘s changeable software, and the cognitive processes were like the fixed hardware. Under these types of assumptions, explicitly assessing learning has been thought to have been unnecessary. Rather, the aptitude or ability to learn could simply be inferred from static tasks that assess the capacity to reason. Hence, paradoxically, learning has tended to have been under-represented in the basic theory building research on Gf. In attempting to manipulate task complexity, Carlstedt et al. (2000) intermixed items from a number of different reasoning tasks in the same (i.e., heterogeneous) presentation. Their expectation was that increasing task complexity should result in an increased correlation between task performance and Gf (see Birney & Bowman, 2009, for discussion of psychometric indicators of complexity). However, contrary to their expectations, performance on the homogeneously presented items (items blocked by task type) was actually the better predictor of Gf. The researchers argued that the basis of this finding was that homogeneous presentation allowed strategies to develop and that high Gf individuals benefited more from the opportunity to learn the structure of the items than low Gf individuals. This is a reasonable interpretation. However, if the deliberate practice work of Ericsson (2003) and the WM training approach of Jaeggi, et al. (2008) are considered, a heterogeneous presentation would be expected to be more beneficial to learning in the long term. It may be the case that the best way to assess Gf (i.e., homogeneous presentation, Carlstedt, et al.) is at odds with the way we might develop Gf (i.e., heterogeneous presentation, Jaeggi, et al.).

Gf and Neuronal Plasticity A relationship between learning and Gf has also been implicated at the biological level. Neural plasticity was proposed as a biological account of learning over 100 years ago (Berlucchi & Buchtel, 2009). However the consensus within the psychometric tradition of cognitive abilities is that while it may be possible to better exploit the use of one‘s level of Gf

28

Damian P. Birney

through cognitive training and through the acquisition of reasoning strategies or mnemonics (as supported by Ericsson‘s studies), the underlying basis of Gf is largely immutable (Ackerman & Beier, 2005; Garlick, 2002). Cognitive neuroscience studies tend to support this assertion. That is, while a) brain changes during cognitive training have been observed (McNab, et al., 2009), b) that impressive transfer and long-term maintenance has been reported in untrained tasks of similar structure (Dahlin, Nyberg, Backman & Neely, 2008), and c) that both these are examples of plasticity, before Jaeggi, et al. (2008), evidence for functional behavioural changes that go beyond simple improvements on structurally similar untrained tests has been lacking.

WORKING MEMORY (WM) WM is important to consider because Gf gains were (purportedly) obtained by Jaeggi, et al. (2008) by training on tasks that demand WM processes.

Definitions of WM A clear consensus regarding the nature of WM remains somewhat elusive (Conway, et al., 2007). Part of the confusion results from the varied approaches to the study of cognition. Experimental psychology, neurosciences, and cognitive-individual differences psychology all conceptualise WM processes somewhat differently. As a relevant case in point, the n-back task (as used by Jaeggi, et al., 2008) has received considerable attention, particularly in the cognitive neurosciences literature (Jonides, et al., 1997), however its status as a WM measure has recently been questioned on the grounds that it does not demonstrate construct validity with other (apparently more accepted) measures of WM, such as complex-span tasks (Kane, Conway, Miura & Colflesh, 2007). That is, despite the n-back task having many of the elementary cognitive operations that are hallmarks of WM, such as encoding, storage and rehearsal, and inhibition and continued maintenance of irrelevant items (Gray, 2001; Jonides, et al., 1997), correlations between n-back and complex-span tasks are often low. Jaeggi, et al. focused on WM training because they believed it targeted the executive processes central to WM, and that WM in turn was central to Gf. It is possible that inconsistency in correlations between different WM tasks (e.g., n-back vs complex-span) may be a result of the tasks targeting different executive processes. That is, training on different WM tasks may produce different Gf gains. We explore some of these issues next.

WM as a System of Executive Processes One of the main definitional problems with the executive processes underlying WM is that different theorist often use the same terms for different purposes (Gray, 2001). Oberauer et al. (2007) have argued that the term executive function is used in a ―… rather loose and very encompassing way‖ (p. 59). They prefer to distinguish between primary cognitive operations and executive processes, and argue for a further demarcation of two kinds of


29

executive processes. The first relates to cognitive control processes, the second relates to inhibition processes. A prototypical executive function of the cognitive control type is task-set switching (Oberauer, et al., 2007). In task-set switching studies, the set of task parameters on which decisions are made is changed and a performance cost is generated in the transition to the new set of task parameters. While the exact interpretation of this cost is still open to debate (e.g., time costs or inhibition of proactive interference), Oberauer, et al. report that switching costs have unexpectedly been shown to have low or non-existent correlations with other FCFs such as WM capacity and Gf. This result was unexpected, they argue, because the capacity to maintain different task set parameters and to control when these parameters are activated is something a central executive-like component of WM would do. Finding zero or low correlations with measures of WM brings into question the exact nature of the constructs being assessed. The complexity of the WM/executive process issue can be further appreciated if one considers the work of Gray, Chabris, and Braver (2003). In this work, brain activation was shown to be more correlated with Gf on lure-trials than non-lure trials in the n-back task. Lure trials require inhibition of an inappropriate match-response. Thus, when executive processes are defined by inhibition processes, high correlations are found. It seems that the conceptualisation of the causal link between n-back performance and WM is in part a function of what is emphasised, and thus even within a task, close attention is required regarding just how WM components should be operationalised and trained.

NON-COGNITIVE MODERATORS OF GF GAINS There are a number of non-cognitive factors that are known to impact on reasoning performance and may thus moderate any observed Gf gains. The factors that are of most interest here are: a) implicit theories regarding the malleability of cognitive abilities; b) goal orientations toward learning versus performing, and; c) personality dispositions that favour effortful cognitive activities.

Implicit Theories regarding Malleability of Intelligence Dweck, et al. (1995) argue that the assumptions people make about the malleability of highly valued personal attributes follow either an entity or incremental implicit theory. Entity theorists tend to believe that intelligence is a fixed, non-malleable trait-like entity and that although people can learn new things, their underlying intellectual abilities remain the same. Individuals holding an entity perspective tend to view difficult situations as aversive, are less likely to see intrinsic learning value in attempting difficult tasks, and are more likely to attribute failure to external factors. The threat of potentially exposing one‘s ability limitations is thought to lead to maladaptive behavioural responses. Alternatively, incremental theorists tend to believe that intelligence is a malleable quality that can be changed and developed through effort. Individuals who hold an incremental perspective tend to be more motivated to attempt and persist at difficult problems. Failure is not simply attributed to lack of ability,

30

Damian P. Birney

rather they see challenging tasks as opportunities for learning and as evidence that more effort should be invested (Grant & Dweck, 2003). Investigating the impact of implicit theories on cognitive performance, Birney and Jeltova (2005) presented 92 university students a set of easy to moderately difficult Latin square reasoning problems (Birney, Halford & Andrews, 2006). At the completion of this first set of items, participants were told that the next set would take the same form but would be much more difficult. Individuals holding an entity perspective performed significantly worse on the second set of items than individuals holding an incremental theory, in spite of the fact that the two groups did not differ in Gf. It seems that the perspectives we have regarding the malleability of intelligence in part determines the way we approach cognitively demanding tasks and thus influences subsequent performance. Implicit theories need to be investigated as a potential moderator of WM training effects.

Personality and Goal Orientation Trait-based personality is typically conceptualised by five distinct and independent personality factors: Extraversion, neuroticism, agreeableness, conscientiousness, and openness to experience (Costa & McCrae, 1992). Openness to experience has been linked to need for cognition (Cacioppo & Petty, 1982), a tendency to seek out complex cognitive activities, and thus both are related to implicit theories in that they have the potential to moderate effects on training outcomes. Conscientiousness is also of relevance to training outcomes. Furthermore, the goal orientation one adopts has also been shown to be important for training outcomes (Payne, Youngcourt & Beaubien, 2007). Individuals with learning goal orientations are similar to incremental theorists, in that task engagement is intrinsically motivated. On the other hand, individuals with performance goal orientations are said to be driven by seeking approval from others and they tend to engage in ineffective learning strategies. All these non-cognitive factors are linked in that they influence the motivation to engage in the types of behaviours demanded by deliberate practice.

IMPLICATIONS FOR REAL-WORLD ACHIEVEMENTS We argue that one of the main reasons for why Jaeggi, et al.‘s (2008) findings are important and why they need to be investigated further, has to do with the putative relationship between the fore-mentioned FCFs and everyday life achievements. Although there have been many successful educational programs and psychological interventions developed to improve expertise in almost every area of human endeavour, the general consensus has been that knowledge acquisition is contingent on, and therefore limited by, basic cognitive functions which are largely immutable (Neisser, et al., 1996). As already mentioned, knowledge acquisition in virtually all domains, including vocational ones, is thought to be achieved to a significant extent through the investment of one‘s Gf in appropriate learning activities (Ackerman & Beier, 2005; Cattell, 1987). If Gf is not a malleable human attribute, then a limit on knowledge acquisition concomitant with one‘s Gf


31

would logically follow. Furthermore, if researchers and educators make the presumption that certain cognitive abilities cannot be changed, then the approach to intelligence test development will naturally progress in a way that supports this assumption. If these assumptions are incorrect (even partially), then as argued by Birney and Sternberg (2006), the psychometric tools and the intelligence theories which emerge through their use are necessarily incomplete. Additional research is required to investigate the implications of Jaeggi, et al.‘s findings for Gf testing and how measurement and conceptualisation of the construct changes after training. In this last section, we reflect on the socio-political implications of Jaeggi, et al.‘s (2008) findings to present a case for the importance of continued research in this area. It is not surprising, given the previous discussion, that the standard interpretation of an individual‘s performance on a traditional intelligence test (that is, the meaning assigned to a score) has progressed in a way consistent with the assumption that basic cognitive abilities are largely immutable. It is also not surprising that historically this has had wide, ongoing, social ramifications that have not been missed in the scientific and popular literature. Furthermore, given the close link between intelligence and achievements in everyday activities (e.g., Jensen, 1998), and that the latter are often the responsibility of social institutions (i.e., policy makers at schools, universities, governments, and in commercial industries), it is not surprising there are complex implications. Much has been written and said throughout what has been a long and sometimes sordid history about the educability and intellectual worth of individuals and groups of individuals (the latter almost invariably linked to social, gender or ethnic clustering). A view of the intellectual worth of one such group of individuals 130 years ago was reported by Le Bon in 1879: ―In the most intelligent races, as among Parisians, there are a large number of women whose brains are closer in size to those of gorillas than to the most developed male brain... All psychologists who have studied the intelligence of women… recognize today that they represent the most inferior forms of human evolution... They excel in fickleness, inconstancy, absence of thought and logic, and incapacity to reason. Without doubt, there exist some distinguished women, very superior to the average male, but they are as exceptional as the birth of any monstrosity, as, for example, of a gorilla with two heads; consequently, we may neglect them entirely‖ (cited in, Gould, 1996, p 136).

And further: ―A desire to give them the same education, and, as a consequence, to propose the same goals for them, is a dangerous chimera…. The day when, misunderstanding the inferior occupations which nature has given her, women leave the home and take part in our battles; on this day a social revolution will begin, and everything that maintains the sacred ties of the family will disappear.‖

The contemporary view in comparison is of course considerably more moderate than this extreme example (though of course, the example was not so extreme in 1879). However, similar writings can be found since this time ‗justifying‘ slavery, early immigration policies, and selection decisions regarding who will fight a war at the frontline and who will not. In fact, group intelligence testing arguably emerged in the USA to specifically address just these types of questions (see Gould, 1996, for an excellent review). Controversy persists regarding

32

Damian P. Birney

how to deal with the assumption of cognitive fixedness as a potential success-limiting factor in providing access to educational and/or other opportunities. The debate has at times become polarised with arguments presented for and against slashing social programs because beneficiaries are seen as not being able to benefit because of inborn and fixed cognitive limits (Herrnstein & Murray, 1994). These limits are almost invariably expressed as low IQ scores collected from psychometric measures that were developed under the assumption that cognitive abilities are generally fixed. Proponents of dynamic assessment (Lidz & Elliott, 2000), which philosophically has tailored adaptive assessment at its heart, strongly advocate for a greater acceptance that for at least some individuals, scores on standardised psychometric aptitude tests are not appropriate for making decisions regarding whether they should be permitted access to quality education.

CONCLUSION Specific evaluations of Jaeggi, et al.‘s (2008) work is still somewhat lacking. Sternberg (2008), one of the leading contemporary researchers of intelligence, is optimistic about the findings. He concludes that although several limitations of the study mean that there is a clear need for follow-up research, ―… none of them call into question the obtained results.‖ (p. 6791). We have argued that a fruitful route for further investigation is to use an overarching deliberate practice framework. To do this, one must be very clear about what attributes are being modified, through what processes, and via what training-methods. There are significant scientific and socio-political implications associated with this work. Simply demonstrating that Gf can be measurably improved and that this has a follow-on effect in everyday functioning has substantial implications for theory and test development, education and training programs, as well as ultimately for social stereotypes. To achieve this, significant headway will need to be made in overcoming a century of inertia surrounding traditional psychometric assessment and theorising.

ACKNOWLEDGMENTS I would like to thank Timothy Scriven and Fiona Ang for their assistance in researching the material for this Chapter. This research was supported under Australian Research Council's Linkage Projects funding scheme (project LP0669552). The views expressed herein are those of the authors and are not necessarily those of the Australian Research Council.

REFERENCES Ackerman, P. (1996). A theory of adult intellectual development: Process, personality, interests, and knowledge. Intelligence, 22, 227-257. Ackerman, P., & Beier, M. E. (2005). Knowledge and intelligence. In O. Wilhelm & R. W. Engle (Eds.), Handbook of understanding and measuring intelligence (pp. 125-139). CA: Sage.


33

Berlucchi, G., & Buchtel, A. (2009). Neuronal plasticity: Historical roots and evolution of meaning. Experimental Brain Research, 192, 307-319. Birney, D. P., & Bowman, D. B. (2009). An experimental-differential investigation of cognitive complexity. Psychology Science Quarterly, 51, 449-469. Birney, D. P., Bowman, D. B., & Pallier, G. (2006). Prior to paradigm integration, the task is to resolve construct definitions of gF and WM. Behavioral and Brain Science, 29, 127. Birney, D. P., Halford, G. S., & Andrews, G. (2006). Measuring the Influence of Relational Complexity on Reasoning: The Development of the Latin Square Task. Educational and Psychological Measurement, 66, 146-171. Birney, D. P., & Jeltova, I. (2005). Perceptions of the malleability of intelligence: Culture and influences on reasoning, self-confidence, and tolerance to ambiguity. Paper presented at the 10th International Association for Cognitive Education and Psychology, University of Durham, Durham, UK. Birney, D. P., & Sternberg, R. J. (2006). Intelligence and cognitive abilities as competencies in development. In E. Bialystok & G. Craik (Eds.), Lifespan cognition: Mechanisms of change (pp. 315-330). New York: Oxford University Press. Blair, C. (2006). How similar are fluid cognition and general intelligence? A developmental neuroscience perspective on fluid cognition as an aspect of human cognitive ability. Behavioral and Brain Sciences, 29, 109-160. Braver, T. S., Paxton, J. L., Locke, H. S., & Barch, D. M. (2009). Flexible neural mechanisms of cognitive control within human prefrontal cortex. Proceedings of the National Academy of Sciences of the United States of America, 106, 7351-7356. Cacioppo, J. T., & Petty, R. E. (1982). The need for cognition. Journal of Personality and Social Psychology, 42, 116-131. Carlstedt, B., Gustafsson, J. E., & Ullstadius, E. (2000). Item sequencing effects on the measurement of fluid intelligence. Intelligence, 28, 145-160. Cattell, R. B. (1987). Intelligence: Its structure, growth and action. Amsterdam: Elsevier Science Publishers. Chase, W. G., & Ericsson, K. A. (1982). Skill and working memory. In W. G. Chase (Ed.), The psychology of learning and motivation (Vol. 16, pp. 1-58). New York: Academic Press. Conway, A. R. A., Jarrold, C., Kane, M. J., Miyake, A., & Towse, J. N. (Eds.). (2007). Variation in working memory. New York: Oxford University Press. Costa, P. T., & McCrae, R. R. (1992). Four ways five factors are basic. Personality and Individual Differences, 13, 653-665. Dahlin, E., Nyberg, L., Backman, L., & Neely, A. S. (2008). Plasticity of executive functioning in young and older adults: Immediate training gains, transfer, and long-term maintenance. Psychology and Aging, 23, 720-730. Dweck, C. S., Chiu, C., & Hong, Y. (1995). Implicit theories and their role in judgments and reactions: A world from two perspectives. Psychological Inquiry, 6, 267-285. Ericsson, K. A. (2002). Attaining excellence through deliberate practice: Insights from the study of expert performance. In M. Ferrari (Ed.), The pursuit of excellence through education (pp. 21-55). Mahwah, NJ: Lawrence Erlbaum. Ericsson, K. A. (2003). The search for general abilities and basic capacities: Theoretical implications from the modifiability and complexity of mechanisms mediating expert

34

Damian P. Birney

performance. In R. J. Sternberg & E. L. Grigorenko (Eds.), The psychology of abilities, competencies, and expertise (pp. 93-125). Cambridge, UK: Cambridge University Press. Ericsson, K. A. (2005). Recent advances in expertise research: A commentary on the contributions to the special issue. Applied Cognitive Psychology, 19, 223-241. Ericsson, K. A., Krampe, R. T., & Tesch-Römer, C. (1993). The role of deliberate practice in the acquisition of expert performance. Psychological Review, 100, 363-406. Feldman Barrett, L., Tugade, M. M., & Engle, R. W. (2004). Individual differences in working memory capacity and dual-process theories of the mind. Psychological Bulletin, 130, 553-573. Garlick, D. (2002). Understanding the Nature of the General Factor of Intelligence: The Role of Individual Differences in Neural Plasticity as an Explanatory Mechanism. Psychological Review, 109, 116-136. Gould, S. J. (1996). The mismeasure of man (Revised and expanded ed.). NY: Penguin Books. Grant, H., & Dweck, C. S. (2003). Clarifying achievement goals and their impact. Journal of Personality and Social Psychology, 85, 541-553. Gray, J. R. (2001). Emotional modulation of cognitive control: Approach-withdrawal states double-dissociate spatial from verbal two-back task performance. Journal of Experimental Psychology: General, 130, 436-452. Gray, J. R., Chabris, C. F., & Braver, T. S. (2003). Neural mechanisms of general fluid intelligence. Nature Neuroscience, 6, 316-322. Herrnstein, R. J., & Murray, C. (1994). The bell curve. New York: Free Press. Holmes, J., Gathercole, S. E., & Dunning, D. J. (2009). Adaptive training leads to sustained enhancement of poor working memory in children. Developmental Science, 12, F9-F15. Horn, J., & Cattell, R. B. (1966). Refinement and test of the theory of fluid and crystallized general intelligences. Journal of Educational Psychology, 57, 253-270. Jaeggi, S. M., Buschkuehl, M., Jonides, J., & Perrig, W. J. (2008). Improving fluid intelligence with training on working memory. Proceedings of the National Academy of Sciences of the United States of America, 105, 6829-6833. Jensen, A. (1998). The g factor: The science of mental ability. Westport, CT: Praeger/Greenwood. Jonides, J., Schumacher, E. H., Smith, E. E., Lauber, E. J., Minoshima, S., & Koeppe, R. A. (1997). Verbal working memory load affects regional brain activation as measured by PET. Journal of Cognitive Neuroscience, 9, 462-475. Kane, M. J., Conway, A. R., Miura, T. K., & Colflesh, G. J. H. (2007). Working Memory, attention control, and the N-back task: A question of construct validity. Journal of Experimental Psychology: Learning, Memory, and Cognition, 33, 615-622. Lidz, C. S., & Elliott, J. G. (Eds.). (2000). Dynamic assessment: Prevailing models and applications. New York, NY: Elsevier Science. McArdle, J. J., Ferrer-Caja, E., Hamagami, F., & Woodcock, R. W. (2002). Comparative longitudinal structural analyses of the growth and decline of multiple intellectual abilities over the life span. Developmental Psychology, 38, 115-142. McNab, F., Varrone, A., Farde, L., Jucaite, A., Bystritsky, P., Forssberg, H., et al. (2009). Changes in cortical dopamine D1 receptor binding associated with cognitive training. Science, 323, 800-802.


35

Neisser, U., Boodo, G., Bouchard Jr., T. J., Boykin, A. W., Brody, N., Ceci, S. J., et al. (1996). Intelligence: Knowns and unknowns. American Psychologist, 51, 77-101. Newell, A., & Simon, H. A. (1972). Human problem solving. NJ: Prentice-Hall. Oberauer, K., Suss, H., Wilhelm, O., & Sander, N. (2007). Individual differences in working memory capacity and reasoning ability. In A. Conway, C. Jarrold, M. J. Kane, A. Miyake & J. N. Towse (Eds.), Variation in working memory (pp. 49-75). NY: Oxford University Press. Payne, S. C., Youngcourt, S. S., & Beaubien, J. M. (2007). A meta-analytic examination of the goal orientation nomological net. Journal of Applied Psychology, 92, 128-150. Rougier, N. P., Noelle, D. C., Braver, T. S., Cohen, J. D., & O'Reilly, R. C. (2005). Prefrontal cortex and flexible cognitive control: Rules without symbols. Proceedings of the National Academy of Sciences of the United States of America, 102, 7338–7343. Simon, H. A., & Chase, W. G. (1973). Skill in chess. American Scientist, 61, 394-403. Sternberg, R. J. (2008). Increasing fluid intelligence is possible after all. Proceedings of the National Academy of Sciences of the United States of America, 105, 6791-6792. Sternberg, R. J., & Grigorenko, E. L. (2002). Dynamic testing. New York: Cambridge University Press. Wood, R. E., Beckmann, J., & Birney, D. P. (2009). Simulations, learning and real world capabilities. Education & Training, 51, 491-510.



Chapter 4

THE EFFECTS OF COMPARISON STATUS AND INDIVIDUAL DIFFERENCES IN SOCIAL COMPARISON TENDENCIES ON THE EXPERIENCE OF SCHADENFREUDE Jessica E. Franklin and Eugene Chekaluk Department of Psychology, Macquarie University Sydney, NSW 2109, Australia

ABSTRACT Little is known of the attributes of an individual that are involved in the emotion of Schadenfreude (pleasure at another‘s misfortune). This study used social comparison theory as a basis for exploring individual differences in experiences of Schadenfreude. Ninety nine first-year psychology students participated in a repeated measures experimental study where they responded to vignettes where a comparison target (upward vs. downward) experienced a misfortune. The participants completed a Schadenfreude Scale that measured their response to the misfortune of each target. They also completed the Iowa-Netherlands Comparison Orientation Measure, which measured their social comparison orientation (SCO), the Upward Comparison Subscale (UCS), and the Downward Comparison Subscale, which measured their directional social comparison tendencies. The results showed that participants reported greater Schadenfreude towards the upward comparison target than the downward comparison target. Participants‘ SCO and UCS scores positively predicted Schadenfreude towards the upward comparison target, but not towards the downward comparison target, whilst participants‘ DCS scores were not predictive of Schadenfreude towards either target. These findings suggest that Schadenfreude is facilitated by upward social comparison processes, and that individual differences in experiences of Schadenfreude can be partially explained by individual differences in one‘s dispositional social comparison tendencies. 


38

Jessica E. Franklin and Eugene Chekaluk

Keywords: Emotion, Negative affect, Schadenfreude, Social comparison theory

INTRODUCTION When another person experiences misfortune, the emotional response directed towards the unfortunate other is typically one of sympathy (Brigham, Kelso, Jackson, & Smith, 1997). However, a somewhat less appropriate, joyous, emotional response may also occur. This emotion is Schadenfreude, or, as it is commonly defined, ‗pleasure at another‘s misfortune‘ (Ben-Ze‘ev, 2000; Hareli & Weiner, 2002). This unusual type of pleasure is characterised by the observer‘s passive role in both causing and sustaining the other‘s misfortune (Ben-Ze‘ev, 2000; Feather, 2008; Leach, Spears, Branscombe & Doosje, 2003). Thus, for one to experience Schadenfreude, a third party or external event must be responsible for the other‘s suffering (Leach et al., 2003). Although the existence of Schadenfreude has been documented by philosophers for centuries (Nietzsche, 1887/1967; Portmann, 2000), empirical studies examining this emotion only began to emerge in the 1990s. Much of the empirical research that has been conducted has emphasised the social nature of Schadenfreude. Schadenfreude is undoubtedly a social emotion: the other person‘s negative experience generates emotion within us, and is subsequently directed towards them (Ben-Ze‘ev, 2000; Hareli & Parkinson, 2008). Schadenfreude is also dependent upon a discordant social appraisal, where the observer positively evaluates a situation or event that the object of the Schadenfreude negatively evaluates (Ben-Ze‘ev, 2000; Hareli & Parkinson, 2008). The discordant social appraisal signifies the presence of a comparative concern between the observer‘s fortune and the other‘s misfortune (Ben-Ze‘ev, 2000). It is due to this comparative concern, which is central to the experience of Schadenfreude (Ben-Ze‘ev, 2000), that Schadenfreude has been theoretically and empirically linked to social comparison theory (e.g., Ben-Ze‘ev, 2000; Feather, 2008; Shamay-Tsoory, Tibi-Elhanany, Aharon-Peretz, 2007; Smith et al., 1996; van Dijk, Ouwerkerk, Goslinga, Nieweg, & Gallucci, 2006). Previous empirical studies that have investigated this link have typically emphasised other-focused social comparison explanations for individual differences in Schadenfreude (Leach & Spears, 2008), and consequently, insight into the dispositional personal attributes of an individual that may contribute to differences in this emotional experience has been lacking. Thus, this study proposes to examine both other- and self-focused social comparison explanations for individual differences in experiences of Schadenfreude to achieve a more comprehensive understanding of the relationship between Schadenfreude and social comparison processes, and to provide insight into the personal characteristics of an individual that contribute to the experience of this complex and under-studied emotion.

LINKING SCHADENFREUDE AND SOCIAL COMPARISON THEORY Social comparison theory (Festinger, 1954) proposes that people seek to gain information and evaluate their own abilities, opinions, and relative standing by comparing themselves with similar others in their social environment. Similarly, Schadenfreude is typically directed

The Effects of Comparison Status and Individual Differences…

39

towards others in our social-emotional environment (Hareli & Weiner, 2002). Furthermore, just as people are more likely to compare themselves with others on attributes that are of personal importance (Festinger, 1954), Schadenfreude is most likely to occur in self-relevant domains (Leach et al., 2003; Nietzsche, 1887/1967), and in situations where there is direct competition between the fortunes of the observer and those of the unfortunate other (BenZe‘ev, 2000). In recent years, social comparison theory has undergone numerous changes (Buunk & Gibbons, 2007), with increasing focus on the affective outcomes of making comparisons with others (Buunk, Zurriaga, Peíró, Nauta, & Gosalvez, 2005). Specifically, there has been an emphasis on the affective consequences of making comparisons with those who are in a superior position to oneself, that is, the process of upward comparison, and, with those who are in an inferior position to oneself, known as downward comparison (Buunk, Zurriaga, et al., 2005; Wills, 1981). Comparing one‘s fortunes with those of a superior or inferior other will lead to differing affective outcomes (Buunk, Kuyper, & van der Zee, 2005; Wills, 1981). Contrast effects are believed to occur when one views another‘s status as unchangeable and substantially different from their own, and thus, an upward comparison will elicit negative feelings such as resentment, envy, inferiority and frustration, and a downward comparison will elicit positive feelings such as relief and pleasure (Buunk, Kuyper, et al., 2005; Buunk, Zurriaga, et al., 2005; Mussweiler, Rüter, & Epstude, 2004).

SCHADENFREUDE AND TARGET COMPARISON STATUS The status of another with whom we compare our fortunes has considerable implications for the elicitation of Schadenfreude. Interestingly, research has found that Schadenfreude is commonly preceded by an upward comparison with another (Ben-Ze‘ev, 2000; Brigham et al., 1997), and is facilitated by the contrastive negative emotions such as envy (Brigham et al., 1997; Smith et al., 1996) and resentment (Feather & Nairn, 2005; Feather & Sherman, 2002) that are elicited by this upward comparison. Pleasure and joy are experienced when the superior other subsequently experiences a misfortune, as the misfortune provides relief from the negative emotions associated with the upward comparison, whilst furthermore emphasising the observer‘s new, comparatively superior position (Ben-Ze‘ev, 2000; Smith et al., 1996). Accordingly, it is believed that Schadenfreude is rarely preceded by a downward comparison, as the downward other poses no threat to the observer‘s comparative fortune (Ben-Ze‘ev, 2000).

SCHADENFREUDE AND SOCIAL COMPARISON ORIENTATION There are considerable individual differences in the extent to which people engage in social comparison (Buunk & Mussweiler, 2001; Gibbons & Buunk, 1999). Gibbons and Buunk (1999) developed the concept of social comparison orientation (SCO), to refer to an individual‘s dispositional tendency to engage in social comparison, and they developed a scale assessing such individual differences. Generally, those who are high in SCO are more prone to engage in social comparison (Gibbons & Buunk, 1999). Researchers have suggested

40


that this variability in SCO may be a function of differing personality characteristics, and indeed, research has indicated that those individuals who are high in SCO are also more likely to have an unstable self concept, lower self-esteem, and be higher in negative affectivity and depression (Buunk & Gibbons, 2006, 2007). Furthermore, those high in SCO are typically more affected by social comparison situations (Buunk & Gibbons, 2006). These results have important theoretical implications for Schadenfreude. Firstly, as an upward social comparison appears to be a necessary antecedent for the elicitation of Schadenfreude (Brigham et al., 1997; Smith et al., 1996), it would be likely that the greater an individual‘s SCO, and thus the more sensitive they are to social comparison situations, the more Schadenfreude they would feel when presented with the misfortune of a superior other in a socially comparative situation. Secondly, as those who are high in SCO are more likely to suffer from low self-esteem and depression (Buunk & Gibbons, 2006, 2007; Gibbons & Buunk, 1999), it is likely that individuals with these characteristics would be particularly sensitive to negative emotions (Gilboa-Schechtman, Foa, Vaknin, Marom, & Hermesh, 2008) emerging from an upward social comparison, and thus experience more pleasure when these negative emotions are alleviated by witnessing misfortune befall a superior other.

SCHADENFREUDE AND DIRECTIONAL SOCIAL COMPARISON TENDENCIES It is also believed that the direction in which people have a dispositional tendency to compare with others varies considerably from one person to the next (Buunk, Ybema, Gibbons, & Ipenburg, 2001; Wills, 1981). This finding has important implications for the elicitation of Schadenfreude. As the experience of Schadenfreude is commonly preceded by an upward comparison (Brigham et al., 1997; Smith et al., 1996), and witnessing a misfortune befall another places the observer in a comparatively superior position to the unfortunate other, effectively producing a downward comparison (Ben-Ze‘ev, 2000), comparison direction appears to play a fundamental role in the experience of Schadenfreude. It is likely then, that the direction in which an observer has a dispositional tendency to compare with others would be associated with their experience of Schadenfreude towards both upward and downward comparison targets.

THE PRESENT STUDY This study aims to provide a more comprehensive understanding of the relationship between Schadenfreude and social comparison processes by examining Schadenfreude within the explicit framework of social comparison theory. Thus, we ask whether social comparison theory can provide an explanation as to why individual variance occurs in experiences of Schadenfreude. Specifically, this study aims to clarify previous findings by investigating whether a social comparison other‘s superior or inferior status may affect one‘s experience of Schadenfreude. Furthermore, we aim to provide a new dimension to the Schadenfreude and social comparison literature by investigating whether individual differences in experiences of


41

Schadenfreude towards upward and downward comparison others can be explained by differences in individuals‘ SCO, and their directional social comparison tendencies. Based on the literature reviewed above, we propose: 1. Overall, participant experience of Schadenfreude will be greater towards the upward comparison target than the downward comparison target. 2. SCO will be a positive predictor of participants‘ Schadenfreude towards the upward comparison target. This relationship will not be found for the downward comparison target. 3. To explore whether participants‘ dispositional directional comparison tendencies are able to predict participants‘ experience of Schadenfreude towards the upward and downward comparison targets.

METHOD Participants and Design One hundred first-year psychology students from Macquarie University, Sydney, Australia, participated in a repeated measures experimental study. The participants were recruited via Macquarie University‘s subject-pool website and were required to be proficient in English. One participant was excluded from the analysis as they failed to meet the eligibility requirements. The remaining sample of 99 participants consisted of 18 males and 81 females whose ages ranged from 17 to 43 years (M = 20.08 years). For their participation, all participants received a time-credit of 30 minutes towards their assessment requirement for an introductory psychology course.

Materials Vignettes Two written vignettes were constructed to induce Schadenfreude in the participants. The participants were asked to imagine themselves as they were described in the vignettes, interacting with a peer student. The peer student was described as having either exceptional social and academic skills (upward comparison target condition ―G.A.‖), or poor social and academic skills (downward comparison target condition ―S.B.‖), whilst in both vignettes the participant was described as having average social and academic skills. The vignettes were written to increase comparison behaviour between the participant and the target by increasing perceived similarity (Festinger, 1954), and were based around characteristics that students typically compare themselves with others on such as grades, performance, and behaviour (Buunk, Kuyper, et al., 2005). The vignettes then described a situation where the target experienced a misfortune. The target was ―caught cheating from notes they had snuck into the final exam‖ and ―consequently failed the unit, jeopardising their chance of getting into a competitive fourth-year program.‖ Aside from the manipulation of the target‘s comparison level, the two vignettes were identical.

42


Schadenfreude Scale Two Schadenfreude Scales were constructed to measure the extent to which the participants experienced Schadenfreude in response to the comparison targets‘ misfortune described in the two vignettes. The Schadenfreude Scales comprised of eight items in total. Schadenfreude was measured by five items that described feelings of satisfaction, pleasure, contentment, joy, and happiness at the target‘s misfortune (e.g., ―I feel satisfied by what happened to G.A.‖). Three filler items that reflected general emotional responses to the target‘s misfortune that are not consistent with the experience of Schadenfreude (e.g., ―I feel upset by what happened to S.B‖) were also included. Participants were asked to rate their level of agreement with each of the statements on a 6-point Likert-type scale that ranged from 1 (strongly disagree) to 6 (strongly agree) from the perspective of the participant they were asked to identify with in each of the vignettes. Overall, a higher mean score on the five items indicated greater Schadenfreude. Aside from being adjusted for the corresponding target in each vignette, the two Schadenfreude Scales were identical.

Social Comparison Tendencies SCO was measured using the Iowa-Netherlands Comparison Orientation Measure (INCOM; Gibbons & Buunk, 1999). The INCOM comprises 11 items (e.g., ―I often compare how I am doing socially (e.g., social skills, popularity) with other people‖), and requires participants to rate their level of agreement with each item on a 5-point Likert-type scale ranging from 1 (I disagree strongly) to 5 (I agree strongly). After reverse coding items 5 and 11, a higher mean score on the 11 items indicates greater SCO. The INCOM‘S two directional subscales, the Upward Comparison Subscale (UCS) and the Downward Comparison Subscale (DCS; F.X. Gibbons, personal communication, March, 2009) were also utilised. The UCS comprises six items measuring dispositional tendency to engage in upward social comparisons (e.g., ―I sometimes compare myself with others who have accomplished more in life than I have‖), and the DCS comprises six items measuring dispositional tendency to engage in downward social comparisons (e.g., ―I sometimes compare myself with others who have accomplished less in life than I have‖). All items on both subscales are rated on the aforementioned 5-point scale, and a higher mean score on each subscale indicates a greater dispositional tendency to engage in upward (UCS) and downward (DCS) social comparisons.

Procedure Participants were randomly allocated a questionnaire booklet containing each of the measures, where the vignettes and related Schadenfreude Scales were counterbalanced across the sample. Instructions on how to complete each questionnaire were presented at the beginning of each measure. The participants then read the first vignette and completed the pertaining Schadenfreude Scale. This was followed by the INCOM, the UCS and the DCS. Next, the participants read the remaining vignette and completed the subsequent Schadenfreude Scale. The study concluded with completion of the demographics questionnaire and a participant debriefing.


43

RESULTS Preliminary Analyses Table 1 presents the descriptive statistics and reliability coefficients for each of the study‘s measures. As can be seen by the Cronbach‘s alpha coefficients, each of the scales demonstrated sufficiently high internal consistency. Table 1. Descriptive statistics & reliability coefficients for the study variables (N=99) Measure Schadenfreude UCTC DCTC Social Comparison INCOM UCS DCS

M

SD

Min.

Max.

Possible Range

Cronbach‘s α

2.76 1.85

0.95 0.66

1.00 1.00

5.60 4.00

1-6 1-6

.84 .80

3.78 3.22 2.89

0.55 0.80 0.82

2.45 1.17 1.00

5.00 5.00 4.67

1-5 1-5 1-5

.82 .82 .85

UCTC = upward comparison target condition; DCTC = downward comparison target condition; INCOM = Iowa-Netherlands Social Comparison Orientation Measure; UCS = Upward Comparison Subscale; DCS = Downward Comparison Subscale.

Data inspection revealed that the Schadenfreude Scale pertaining to the downward comparison target condition was positively skewed, and so in accordance with the method recommended by Tabachnick and Fidell (2007), a log10 data transformation was performed. The resulting transformed data set was normally distributed. All statistical analyses were then conducted twice, once using the untransformed data set and once using the transformed data set. It was found that the actual pattern of significance of the results obtained in the two sets of analyses did not differ. Therefore, for ease of interpretation, the untransformed data set was used to report the results of all analyses. The primary assumptions for all other variables were tested and satisfied and the conventional alpha level of .05 was used in all analyses.

Primary Analyses Hypothesis 1 A paired-samples t-test was conducted to investigate the effect of target comparison status on the participants‘ experience of Schadenfreude. As predicted, participant reported Schadenfreude was greater towards the upward comparison target (M = 2.76, SD = 0.95), than the downward comparison target (M = 1.85, SD = 0.66), as participant scores on the Schadenfreude Scale pertaining to the upward comparison target condition were significantly higher, t (98) = 10.15, p < .0005, η² = .51, 95% CI [0.74, 1.09].

44


Table 2. Results of regression analyses of variables predicting Schadenfreude towards the upward comparison target & Schadenfreude towards the downward comparison target (N=99) Predictor Variable

B

Upward comparison target SCO .45 Upward comparisonª .39 Downward .08 comparisonb Downward comparison target SCO .04 Upward comparisonª .04 Downward -.07 comparisonb

SE B

β

p

% variance explained

.17 .12 .12

.26 .33 .06

.010 .001 .518

7% 10% <.005%

.12 .09 .09

.03 .04 -.08

.772 .689 .435

<.005% <.005% <.005%

ª Controlled for downward comparison scores. b Controlled for upward comparison scores.

Hypothesis 2 To test the second hypothesis for each of the dependent variables, two simple linear regression analyses were conducted with participant scores on the INCOM as the independent variable. As predicted, SCO was a significant positive predictor of participants‘ experience of Schadenfreude towards the upward comparison target, F (1, 97) = 6.95, p = .010, R² = .07, 95% CI [0.11, 0.78], but not towards the downward comparison target, F (1, 97) = 0.09, p = .772, R² < .005. Hypothesis 3 Firstly, to investigate whether participants‘ directional social comparison tendencies could predict participants‘ Schadenfreude towards the upward comparison target, a full multiple regression model was fitted with participant scores on the UCS and DCS as the independent variables. The full model containing the two predictors was significant, F(2, 96) = 6.61, p = .002, R² = .12. An examination of the individual regression coefficients revealed that participant scores on the UCS were a significant positive predictor of Schadenfreude towards the upward comparison target, after controlling for participant scores on the DCS, t (96) = 3.28, p = .001, R² = .10, 95% CI [0.15, 0.62]. However, after controlling for participant scores on the UCS, participant scores on the DCS did not predict Schadenfreude scores towards the upward comparison target, t (96) = 0.65, p = .518, R² < .005. The results of the individual predictor variables for both analyses are summarised in Table 2. Secondly, to investigate whether participants‘ directional social comparison tendencies could predict participant experience of Schadenfreude towards the downward comparison target, a full multiple regression model was fitted with participant scores on the UCS and DCS as the independent variables. The full model was not significant, F (2, 96) = 0.33, p = .721, R² = .01, indicating that participants‘ Schadenfreude towards the downward comparison target could not be predicted from their directional social comparison tendencies. The results of the individual predictors are summarised in Table 2.


45

DISCUSSION Consistent with the first hypothesis, the participants reported experiencing significantly more Schadenfreude in response to the upward comparison target‘s misfortune, than in response to the downward comparison target‘s misfortune. This finding suggests that upward social comparisons may function as stronger elicitors of Schadenfreude than downward social comparisons, and is consistent with previous literature that has suggested that Schadenfreude is frequently preceded by an upward comparison with another (e.g., Ben-Ze‘ev, 2000; Brigham et al., 1997). Furthermore, this finding suggests that the focus of concern central to Schadenfreude is one‘s own comparative fortune. A superior other is perceived as a threat to one‘s situation (Smith, Eyre, Powell, & Kim, 2006), and thus when one witnesses a misfortune befall them, this threat to one‘s fortune is alleviated, giving rise to the pleasurable experience of Schadenfreude. Thus the findings of the current study also appear consistent with those of Leach and Spears (2008) who found Schadenfreude to be more concerned with feelings of inferiority of the self, than with the success or failure of others.

UPWARD COMPARISON PROCESSES AND SCHADENFREUDE Consistent with the second hypothesis, SCO positively predicted participants‘ Schadenfreude towards the upward comparison target. This finding provides the first empirical evidence to suggest that the greater one‘s dispositional tendency to engage in social comparison, the more Schadenfreude they are likely to experience when witnessing misfortune befall a superior other. This finding is in accordance with the idea that those who are more prone to social comparison are particularly sensitive to social comparison situations and information (Buunk & Gibbons, 2006) and thus, may also be particularly sensitive to an emotion emerging from a social comparison situation. Another possible explanation for this result emerges from the finding that those individuals who are high in SCO are also more likely to have unstable self-concepts, low self-esteem and be high in negative affectivity (Buunk & Gibbons, 2006, 2007). Although these variables were not measured in this study, and may need to be included in future research, it may be that individuals with these characteristics are particularly sensitive to negative emotions (Gilboa-Schechtman et al., 2008) elicited by an upward comparison, and thus experience more pleasure when these negative emotions are alleviated by witnessing misfortune befall a superior other.

UPWARD SOCIAL COMPARISON TENDENCIES AND SCHADENFREUDE In terms of the exploratory third hypothesis, participants‘ dispositional upward comparison tendency positively predicted Schadenfreude towards the upward comparison target. This provides the first empirical data to suggest that the greater one‘s tendency to engage in upward social comparisons, the more Schadenfreude they are likely to experience when witnessing misfortune befall a superior other. It is likely that those who have a higher dispositional tendency towards engaging in upward social comparisons may experience more pleasure at a superior other‘s misfortune as it provides relief from the negative emotions

46


associated with frequently comparing upward. If this proposed explanation is correct, it implies that one‘s own negative affectivity, intensified through the social comparison process, may increase the intensity of one‘s Schadenfreude experience.

DOWNWARD COMPARISON PROCESSES AND SCHADENFREUDE Consistent with the second hypothesis, SCO was not predictive of Schadenfreude towards the downward comparison target. Furthermore, in terms of the exploratory third hypothesis, participants‘ dispositional downward social comparison tendencies were not associated with Schadenfreude towards either comparison target. Together with the finding that overall Schadenfreude towards the downward comparison target was very low, it appears that downward comparison processes may have little or no association with one‘s experience of Schadenfreude. The absence of this association appears consistent with theoretical literature that has suggested that we rarely feel Schadenfreude for another who is below us as they pose no threat to our own comparative fortune (Ben-Ze‘ev, 2000), and thus, that one‘s perception of their own negative comparative situation is central to the elicitation of Schadenfreude (Ben-Ze‘ev, 2000).

METHODOLOGICAL LIMITATIONS AND DIRECTIONS FOR FUTURE RESEARCH The current study also has a number of limitations that must be noted. Although the manipulation aimed at increasing Schadenfreude did indeed lead to an increase in this emotion, the participants‘ overall Schadenfreude scores were fairly mild. A possible explanation for this result emanates from the socially undesirable nature of Schadenfreude (Portmann, 2000). Although measures were taken to reduce the likelihood of socially desirable responding, it is likely that the participants under-reported their feelings of Schadenfreude due to the negative social stigma associated with admitting to experiencing pleasure in another‘s pain. An additional concern of the current study is the artificiality of ascribing the participants the role of an ―average student‖ within the vignettes as this may have prevented the obtainment of realistic comparison status data. Future studies may therefore benefit from including measures of actual participant status characteristics such as high school grades, university grades, and popularity. Lastly, it is evident that further research is needed to determine whether the relationship between one‘s negative affect, and their social comparison tendencies, as proposed by the current study, does increase the intensity of Schadenfreude. Therefore, it is suggested that future studies of Schadenfreude and social comparison processes could also include independent measures of participants‘ self-esteem, depression, and negative affect. These studies may also benefit from including measures of narcissism, neuroticism, and stress, which have also been found to correlate with individual‘s social comparison tendencies (Bogart, Benotsch, & Paadovic, 2004; Gibbons & Buunk, 1999).


47

CONCLUSION Social comparison theory provided a useful framework to investigate individual differences in Schadenfreude. Specifically, the current study contributed to the existing literature that has linked Schadenfreude and social comparison theory by finding that one‘s experience of Schadenfreude is largely affected by the comparison status of another. Furthermore, the current study has provided the first evidence that individual differences in the extent and direction of one‘s dispositional social comparison tendencies can partially account for individual differences in experiences of Schadenfreude, and thus has provided much needed insight into the personal attributes of an individual that are involved in this complex emotion. As one‘s dispositional tendency to compare with others is largely shaped by characteristics of one‘s personality (Buunk & Gibbons, 2006), this study has paved the way for future researchers to investigate which specific features of personality may be involved in Schadenfreude. In sum, by examining Schadenfreude within the explicit framework of social comparison theory, the findings of the current study have provided a more comprehensive understanding of the relationship between Schadenfreude and social comparison processes, in finding that Schadenfreude is indeed facilitated by upward social comparison processes.

REFERENCES Ben-Ze‘ev, A. (2000). The subtlety of emotions. Cambridge, MA: MIT Press. Bogart, L. M., Benotsch, E. G., & Paadovic, J. D. (2004). Feeling superior but threatened: The relation of narcissism to social comparison. Basic and Applied Social Psychology, 26, 35-44. Brigham, N. L., Kelso, K. A., Jackson, M. A., & Smith, R. H. (1997). The roles of invidious comparisons and deservingness in sympathy and Schadenfreude. Basic and Applied Social Psychology, 19, 363-380. Buunk, A. P., & Gibbons, F. X. (2006). Social comparison orientation: A new perspective on those who do and those who don‘t compare with others. In S. Guimond (Ed.), Social comparison and social psychology: Understanding cognition, intergroup relations and culture (pp. 15-32). New York, NY: Cambridge University Press. Buunk, A. P., & Gibbons, F. X. (2007). Social comparison: The end of a theory and the emergence of a field. Organisational Behaviour and Human Decision Processes, 102, 321. Buunk, B. P., Kuyper, H., & van der Zee, Y. G. (2005). Affective response to social comparison in the classroom. Basic and Applied Social Psychology, 27, 229-237. Buunk, B. P., & Mussweiler, T. (2001). New directions in social comparison research. European Journal of Social Psychology, 31, 467-475. Buunk, B.P., Ybema, J.F., Gibbons, F.X., & Ipenburg, M. (2001). The affective consequences of social comparison as related to professional burnout and social comparison orientation. European Journal of Social Psychology, 31, 337-351.

48


Buunk, B. P., Zurriaga, R., Peíró, J. M., Nauta, A., & Gosalvez, I. (2005). Social comparisons at work as related to a cooperative social climate and to individual differences in social comparison orientation. Applied Psychology: An International Review, 54, 61-80. Feather, N. T. (2008). Effects of observer‘s own status on reactions to a high achiever‘s failure: Deservingness, resentment, Schadenfreude, and sympathy. Australian Journal of Psychology, 60, 31-43. Feather, N. T., & Nairn, K. (2005). Resentment, envy, Schadenfreude, and sympathy: Effects of own and other‘s deserved or undeserved status. Australian Journal of Psychology, 57, 87-102. Feather, N. T., & Sherman, R. (2002). Envy, resentment, Schadenfreude, and sympathy: Reactions to deserved and undeserved achievement and subsequent failure. Personality and Social Psychology Bulletin, 28, 953-961. Festinger, L. (1954). A theory of social comparison processes. Human Relations, 7, 117-140. Gibbons, F. X., & Buunk, B. P. (1999). Individual differences in social comparison: Development of a scale of social comparison orientation. Journal of Personality and Social Psychology, 76, 129-142. Gilboa-Schechtman, E., Foa, E., Vaknin, Y., Marom, S., & Hermesh, H. (2008). Interpersonal sensitivity and response bias in social phobia and depression: Labelling emotional expressions. Cognitive Therapy and Research, 32, 605-618. Hareli, S., & Parkinson, B. (2008). What‘s social about social emotions? Journal for the Theory of Social Behaviour, 38, 131-156. Hareli, S., & Weiner, B. (2002). Dislike and envy as antecedents of pleasure at another‘s misfortune. Motivation and Emotion, 26, 257-277. Leach, C.W., & Spears, R. (2008). ―A vengefulness of the impotent‖: The pain of in-group inferiority and Schadenfreude toward successful out-groups. Journal of Personality and Social Psychology, 95, 1383-1396. Leach, C. W., Spears, R., Branscombe, N. R., & Doosje, B. (2003). Malicious pleasure: Schadenfreude at the suffering of another group. Journal of Personality and Social Psychology, 84, 932-943. Mussweiler, T., Rüter, K., & Epstude, K. (2004). The ups and downs of social comparison: Mechanisms of assimilation and contrast. Journal of Personality and Social Psychology, 87, 832-844. Nietzsche, F. (1967). On the genealogy of morals (W. Kaufmann & R.J. Hollingdale, Trans.). New York, NY: Vintage Books. (Original work published 1887). Portmann, J. (2000). When bad things happen to other people. New York, NY: Routledge. Shamay-Tsoory, S. G., Tibi-Elhanany, Y., & Aharon-Peretz, J. (2007). The green-eyed monster and malicious joy: The neuroanatomical bases of envy and gloating (Schadenfreude). Brain, 130, 1663-1678. Smith, R. H., Eyre, H. L., Powell, C. A. J., & Kim, S. H. (2006). Relativistic origins of emotional reactions to events happening to others and to ourselves. British Journal of Social Psychology, 45, 357-371. Smith, R. H., Turner, T.J ., Garonzik, R., Leach, C. W., Urch-Druskat, V., & Weston, C. M. (1996). Envy and Schadenfreude. Personality and Social Psychology Bulletin, 22, 158168. Tabachnick, B. G., & Fidell, L. S. (2007). Using multivariate statistics (5th ed.). Boston, MA: Pearson Education.


49

van Dijk, W. W., Ouwerkerk, J. W., Goslinga, S., Nieweg, M., & Gallucci, M. (2006). When people fall from grace: Reconsidering the role of envy in Schadenfreude. Emotion, 6, 156-160. Wills, T. A. (1981). Downward comparison principles in social psychology. Psychological Bulletin, 90, 245-271.



Chapter 5

PROCESSING STYLE AND SOCIAL COMPARISON OUTCOMES: ASSIMILATION AND CONTRAST EFFECTS IN AUTOMATIC BEHAVIOUR AND SELF-EVALUATION Lauren O’Driscoll and Eugene Chekaluk Department of Psychology, Macquarie University, Sydney, NSW 2109, Australia

ABSTRACT The importance of one‘s self-concept has long been recognised in personality research. Social comparison theory suggests that individuals compare themselves with others in order to evaluate their own opinions and abilities (Festinger, 1954). Based on Förster, Liberman and Kuschel‘s (2008) global/local model (GLOMO) of assimilation and contrast, the effect of global versus local processing style on behavioural and selfevaluative social comparison outcomes was investigated. The sample consisted of 93 undergraduates (23 males and 70 females). Participants were first primed with global, local or global/local processing, and were then exposed to either an upward comparison target or a control narrative. Assimilation and contrast effects were assessed using a behavioural measure (Information Subtest of the Multidimensional Aptitude Battery II), and a self-evaluation questionnaire. A 2 x 3 factorial design was utilised. The results demonstrated that although GLOMO was generally unsupported for automatic behaviour and self-evaluation, language and gender were shown to partially moderate the effects of processing style and social comparison on assimilation and contrast.

Keywords: Assimilation, Contrast, Self-concept, Self-evaluation, Social comparison



Ph: 1-647-278-3505; Fax: 61-2-9850 8062; [email protected]

52

Lauren O‘Driscoll and Eugene Chekaluk

INTRODUCTION The self-concept refers to a relatively stable and idiosyncratic knowledge structure encompassing one‘s personal identity, and the significance of the self-concept in personality research has long been recognised (e.g., Bandura, 1989). As Turner (1999) notes, self-concept is a particularly useful construct for social psychologists, since it helps towards explaining individual stability across social contexts (Turner & Onorato, 1999). An individual‘s selfconcept is largely developed in relation to comparison with others, in that evaluation of one‘s personal traits through social comparison can allow for a better understanding of the self (Mendaglio & Pyryt, 2002). As originally proposed by Festinger (1954), humans possess a need for self-evaluation, and that in the absence of objective standards, individuals tend to compare themselves to others in order to determine the correctness of their opinions and the nature of their abilities. Thus, whether people consider themselves to be attractive, intelligent, wealthy or athletic largely depends on how they size themselves up against others. Following Festinger‘s seminal theory of social comparison, a great deal of research has focused on the potential consequences of comparison in terms of whether these consequences are assimilative or contrastive in nature. The present research focused on the consequences of comparing in an upward direction, as when one compares the self to someone who is regarded to be superior on a certain dimension. Within the context of upward comparison, assimilation may be conceptualised as an upward shift in an individual‘s self-evaluation, whereas contrast represents a downward shift in an individual‘s self-evaluation. Notably, research has shown that exposure to the same social comparison information can result in assimilation under some circumstances and contrast in others (Wheeler & Suls, 2007).

THE GLOBAL/LOCAL MODEL OF ASSIMILATION AND CONTRAST Förster, Liberman and Kuschel (2008) proposed the global/local model (GLOMO) in order to explain assimilation and contrast effects. Förster et al.‘s GLOMO was derived from the inclusion-exclusion model (IEM) developed by Schwarz and Bless (1992; 2007). According to the IEM, the same comparison information may lead to assimilation or contrast, depending on how this information is cognitively represented by the individual. The IEM predicts that if such information elicits cognitive inclusion of the comparer‘s self representation in the point of comparison, assimilation results, whereas if the information results in exclusion of the comparer‘s self representation from the point of comparison, contrast results (for an overview of the IEM, refer to Schwarz & Bless, 1992; 2007). Extending from the assumptions of the IEM, Förster et al. (2008) reasoned that because cognition is linked to and derives from perception (Finke, 1985) cognitive inclusion and exclusion might further be influenced by perceptual processing. Processing style can be conceptualised as a way of attending to information in the environment (Förster, 2009; Navon, 1977). Global processing refers to perception of the stimulus as a whole, whereas local processing refers to perception of the parts that make up the stimulus (Navon, 1977). According to GLOMO, whether assimilation or contrast results from a social comparison depends upon whether global or local perceptual processing is initiated at the time of the comparison. Accordingly, Förster et al. proposed that global processing should widen an

Processing Style and Social Comparison Outcomes

53

individual‘s conceptual scope, thus resulting in cognitive inclusion and assimilation toward the social comparison target. Alternatively, local processing should narrow an individual‘s conceptual scope, resulting in cognitive exclusion and thus contrast away from the target.

GLOMO and Self-evaluation Förster et al.‘s (2008) research on self-evaluative consequences of social comparison provided support for GLOMO1. In one study, participants were first presented with one of four comparison targets, ranging from extremely low to extremely high standards of athleticism. They were subsequently primed with global or local processing using the Navon task, in which participants were presented with large characters consisting of smaller characters, and asked to focus on either only the large characters (global processing) or on the small characters (local processing) (Navon, 1977). The participants then compared their own athletic ability to that of the comparison target and completed an objective self-evaluation questionnaire. As predicted by GLOMO, the results demonstrated that after being primed with global processing, participants reported themselves to be significantly more athletic (indicating assimilation), whereas those primed with local processing reported themselves to be significantly less athletic (indicating contrast).

GLOMO and Automatic Behaviour Perception can have inadvertent and unconscious effects on social behaviour. Merely perceiving someone behave in a certain way can increase the likelihood that one will behave in a similar manner (James, 1890). Research has shown that priming an individual with a certain construct can in turn elicit behaviour in line with that construct (Bargh, Chen & Burrows, 1996). For example, priming has been shown to influence even relatively complex behaviour, such as performance on a general knowledge test. Dijksterhuis and van Knippenberg (1998) reported that priming participants with either a ‗professor‘ stereotype or the trait ‗intelligent‘ led them to perform significantly better on a general knowledge test, when compared to those primed with ‗soccer hooligans‘, the trait ‗stupid‘ or no stereotype. Thus, as processing style has been shown by Förster et al. (2008) to affect the self-evaluative consequences of social comparison, it is possible that global versus local processing could result in assimilation and contrast of automatic behaviour in a similar manner.

THE PRESENT RESEARCH The present research was designed to extend on Förster et al.‘s (2008) research on the effects of global versus local processing on social comparison outcomes. Accordingly, the primary aim was to investigate the potential effect of global versus local processing on the 1

Förster et al. (2008) also demonstrated assimilation and contrast effects in social judgement. As the focus of the present study was on behavioural and self-evaluative assimilation and contrast, the social judgement effects of GLOMO are not discussed at length.

54


behavioural and self-evaluative consequences of upward social comparison, in order to determine whether these consequences were assimilative or contrastive in nature. It should be noted that because the potential effect of processing style on the behavioural consequences of social comparison has yet to be examined, the present research represented the first attempt to investigate this possibility.

Hypotheses It was predicted that individuals primed with global processing and presented with an upward comparison target of high intelligence would achieve a significantly higher score on a general knowledge test, and would rate themselves to be objectively and subjectively more intelligent (an assimilation effect). Alternatively, it was predicted that individuals primed with local processing and presented with an upward comparison target of high intelligence would achieve a significantly lower score on a general knowledge test, and would rate themselves to be objectively and subjectively less intelligent (a contrast effect).

Supplementary Investigation A supplementary aim of the present research was to explore the potential moderating roles of language and gender on assimilation and contrast effects within the GLOMO paradigm. Förster et al. (2008) suggested that because processing style has been shown to influence assimilation and contrast, it is possible that the consequences of social comparison could further be moderated by factors that are known to affect processing style. Whereas the effects of language and gender have been studied in relation to processing style2, the potential effects of these individual difference variables on the consequences of social comparison within the GLOMO paradigm have yet to be investigated.

METHOD Participants The sample consisted of 93 (23 male and 70 female) undergraduate students from Macquarie University. Participants ranged in age from 15 to 26 years (M = 19.67, SD = 2.22). In order to ensure the relevance of the social comparison vignettes, participation was further limited to human resource management and psychology students. Of the total number of participants, 24 indicated that English was not their first language and of those, 3 reported that they did not read English fluently. No participants were excluded due to their English language proficiency. Participants gave informed consent before the study began, and each received course credit for their participation. 2

For research on processing style and language refer to Stapel and Semin (2007); and Semin, Görts, Nandum, and Semin-Goossens (2002). For research on processing style and gender refer to Blough and Salvin (1987); Kail, Carter, and Pellegrino (1979); Rilea (2008); and Lawton (2004).


55

Materials Induction of Global/Local Processing Processing style was primed in a manner similar to the procedure used by Förster et al. (2008, Experiment 1). Participants were presented with two colour maps (the first map displayed the U.S. state of Pennsylvania and the second map displayed the U.S. state of Massachusetts), and were instructed to attend to particular attributes of each map for 1 minute3. In the global processing condition, participants focused on the overall shape and outline of each map, in the local processing condition they focused on the details and on a red star at the centre of each map, and in the control condition participants focused on both the overall shape and details of each map. Social Comparison Vignettes The participants in the upward comparison condition read a short vignette describing a highly intelligent and academically accomplished student in a scenario devised by Lockwood and Kunda (1997). In order to ensure that the vignettes were comparable, the target was described to be well-rounded, participating in a range of extracurricular activities. To further maximise the relevance of the description, the comparison vignettes were tailored to match the gender and major of each participant. Control participants were alternatively presented with a neutral description on the subject of killer whales (National Geographic Society, 2009). In order to ensure consistency, the control description consisted of the same approximate word-count as the comparison vignette.

Assessment Materials Objective Self-evaluation Objective self-evaluation was assessed by participant response to three objective intelligence related questions. The first question asked participants to indicate what final percentage grade they believed they would achieve in their first-year psychology course (Seaton, 2002). In the second question participants were required to report how many mathematical problems they believed they could solve in 1 minute (Mussweiler & Strack, 2000). The final question prompted participants to indicate how many minutes they believed it would take them to read the front page of a newspaper in order to reproduce its content (Mussweiler & Strack, 2000). Subjective Self-evaluation Subjective self-evaluation was assessed by each participant‘s rating of his or her own intelligence on a 6-point (1 = extremely unintelligent, 6 = extremely intelligent) Likert scale (Seaton, 2002).

3

All maps were obtained from Friedman, Fishbach, Förster, and Werth (2003).

56


Automatic Behaviour Automatic behaviour was determined by participant performance on the Information Subtest of the Multidimensional Aptitude Battery II (MAB-II)4. The test consisted of 40 general knowledge questions in multiple-choice format (A through E). The MAB-II Information Subtest has been shown to have an internal consistency (as measured by Cronbach‘s alpha) ranging from 0.83 to 0.90 and test-retest reliability of approximately 0.97 (Jackson, 1998). The Information Subtest was selected for administration in this study because its content was designed to be applicable across gender, age, nationality and culture (Jackson, 2003).

Procedure Participants were invited to the laboratory on the pretext of participating in a series of pilot tests for future research. In order to minimise response bias, the true purpose of the research was not initially disclosed. All tasks were completed on the computer using MediaLab in order to limit variability within the sample. Each participant was randomly assigned a global, local or control processing style condition, and to either an upward comparison or control condition. Participants were first instructed to focus on either the overall shape, details, or both the overall shape and details (control) of two maps. They were then asked to compare themselves to an upward comparison standard, or presented with a neutral narrative (control). The final two tasks were behavioural and self-evaluation assessment measures. In the behavioural measure, participants were presented with 40 general knowledge questions from the MAB-II Information Subtest and were given a total of 7 minutes to complete the test. They were then presented with three objective self-evaluation questions (final grade, mathematics and reading time) followed by one subjective self-evaluation question. Participants were also asked to indicate their age, gender, and whether they considered English to be their first language. Finally, participants were probed for suspicion of the true purpose of the study and fully debriefed.

RESULTS Preliminary Analysis All normality assumptions were satisfied except for normality of mathematics, reading time and subjective self-evaluation. A square root transformation resulted in the normal distribution of mathematics. Tests of homogeneity of variance revealed that all of the dependent variables were homogeneous, with the exception of reading time and subjective self-evaluation. A logarithm transformation was successful for both variables. As there were no missing cases in the data, and no significant outliers were detected, the full data set (N = 93) was used in the analyses. 4

Permission to administer the MAB-II Information Subtest was granted by SIGMA Assessment Systems.


57

Main Analysis In order to examine the self-evaluative and behavioural consequences of social comparison, separate 2 (social comparison: upward comparison vs. control) x 3 (processing style: global vs. local vs. control) factorial ANOVAs were conducted for the general knowledge, objective self-evaluation (final grade, mathematics and reading time)5, subjective self-evaluation and automatic behaviour dependent variables. The significance level was maintained at p = .05 for all tests. Notably, the results indicated that the research hypothesis was unsupported for both selfevaluation and automatic behaviour. No main effects of social comparison or processing style proved to be significant (p > 0.05), and the interaction effects were not significant (p > .05) for any of the dependent variables (refer to Table 1), with the exception of a significant main effect of social comparison (F(1,86) = 7.53, p = .007) for the final grade variable (refer to Figure 1). Regardless of processing style, individuals in the upward comparison condition indicated that they anticipated achieving a significantly higher final grade in their first year psychology course (M = 68.50, SD = 7.81), when compared to those in the no comparison control condition (M = 65.57, SD = 7.40). Table 1. Main & interaction effects for dependent variables across conditions Dependent Variable Final Grade Mathematics Reading Time Subjective Self-Evaluation Automatic Behaviour

Processing Style F(1,86) = 0.30 p = .704 F(1,86) = 0.28 p = .761 F(1,87) = 2.56 p = .083 F(1,87) = 2.49 p = .089 F(1,85) = 1.21 p = .304

Social Comparison F(1,86) = 7.53 p = .007* F(1,86) = 0.22 p = .640 F(1,87) = 0.32 p = .573 F(1,87) = 0.61 p = .439 F(1,85) = 0.40 p = .528

Interaction F(2,86) = 0.03 p = .971 F(2,86) = 0.72 p = .488 F(2,87) = 1.26 p = .288 F(2,87) = 1.57 p = .215 F(2,85) = 0.35 p = .704

*

indicates a significant effect (p < 0.05).

5

Although a composite measure was originally considered, the correlations between the three objective selfevaluation measures ranged from r = -.12 to r = .01, indicating that the items were not appreciably related to one another.

58


Figure 1. Final grade as a function of social comparison & processing style.

In order to further clarify the relationships between the independent and dependent variables, the language (English as a first language = EFL, English as a second language = ESL) and gender variables were subsequently analysed with social comparison and processing style for each of the dependent variables.

Language Effects The results revealed a significant main effect of language for the final grade variable (F(1,80) = 9.62, p = .003). EFL participants generally reported anticipating a significantly higher final grade in their first year psychology course (M = 68.62, SD = 7.20), when compared to ESL participants (M = 62.90, SD = 7.72). The results also showed a significant main effect of language (F(1,79) = 32.02, p < .001) for automatic behaviour. In general, EFL participants performed significantly better on the general knowledge test (M = 18.37, SD = 4.88), when compared to ESL participants (M = 11.29, SD = 7.78). These results are to be anticipated, as EFL participants would be expected to be more confident in achieving higher grades while attending an English speaking university and would be expected to perform better on a general knowledge test administered in English.


59

Figure 2. General knowledge as a function of social comparison & processing style for English as a first language.

Additionally, the results for ESL revealed a marginally significant interaction effect between social comparison and processing style (F(2,16) = 3.17, p = .069) for automatic behaviour (refer to Figure 2 & Figure 3). Though no significant main or interaction effects were detected for EFL participants, when primed with local processing ESL participants achieved a higher general knowledge score in the no comparison control condition (M = 17.30, SD = 6.36), when compared to the upward comparison condition (M = 7.86, SD = 5.90).

Gender Effects Finally there was a significant interaction effect between gender and processing style (F(2,79) = 3.56, p = .036) for automatic behaviour (refer to Figure 4 & Figure 5). Males performed significantly better on the general knowledge test in the comparison condition (M = 19.14, SD = 6.05) than in the no comparison control condition (M = 16.93, SD = 7.01). Males also tended to perform significantly better in the local processing condition (M = 21.81, SD = 5.09) when compared to the global processing (M = 15.90, SD = 8.46) and control (M = 14.89, SD = 4.73) conditions. No significant effects were found for females.

60


Figure 3. General knowledge as a function of social comparison & processing style for English as a second language.

Figure 4. General knowledge as a function of social comparison & processing style for males.


61

Figure 5. General knowledge as a function of social comparison & processing style for females.

DISCUSSION Extending from Förster et al.‘s (2008) GLOMO, in the present research it was hypothesised that individuals primed with global processing and presented with an upward comparison target would assimilate automatic behaviour and self-evaluations toward the comparison target, whereas those primed with local processing would contrast their automatic behaviour and self-evaluations away from the target. Notably, the GLOMO hypothesis was not directly supported for automatic behaviour, objective self-evaluation and subjective-self evaluation in the present research.

OBJECTIVE AND SUBJECTIVE SELF-EVALUATION Although the research hypothesis was not upheld for self-evaluation, the results for the objective self-evaluation variable, final grade, indicated an assimilation effect for individuals presented with an upward comparison target, but no effect of processing style. These results corroborate past research on social comparison by Lockwood and Kunda (1997), who demonstrated that assimilation of self-evaluation toward an upward comparison target tends to occur when the target is regarded to be self-relevant and when his or her success is thought to be attainable. Because achieving high grades is expected to be a relevant goal among undergraduate students, engaging in an upward comparison with the target most likely had the

62


effect of motivating participants to achieve high grades themselves, thus resulting in assimilation of their objective self-evaluations toward the comparison target. Accordingly, the absence of findings for the mathematics and reading time variables may have been related to the nature of information that was presented in the upward comparison vignettes, as they did not explicitly mention the mathematical or reading abilities of the target. As these particular domains were not made salient to participants at the time of social comparison, it is conceivable that they did not have enough information to make an appropriate comparison on these dimensions (Lockwood & Kunda, 1997). As discussed by Förster et al. (2008), subjective self-evaluation effects are often difficult to detect. As the participants in the present study consisted of a relatively homogeneous sample of university students, it is conceivable that they generally considered themselves to be relatively intelligent to begin with. Stronger effects for subjective self-evaluation might be detected with a more varied sample.

AUTOMATIC BEHAVIOUR: LANGUAGE AND GENDER EFFECTS Although the GLOMO model was generally unsupported for automatic behaviour, the results suggest that language and gender might partially moderate the effects of processing style on the behavioural outcomes of social comparison. In terms of language effects, the present research demonstrated that the local processing prime evoked a considerable increase in general knowledge performance for those who reported that English was their second language, but only when they were not presented with an upward comparison standard6. It is possible that local processing may be more helpful than global processing in completing a general knowledge test, because general knowledge questions tend to focus on detailed information rather than holistic or global concepts (Navon, 1977). In addition, the finding that participants who reported that English was their second language performed relatively worse on the general knowledge test in the upward comparison condition (when compared to the no comparison control condition) can be explained by research on standard attainability by Lockwood and Kunda (1997). It may be that individuals who indicated that English was a second language might have regarded the academic success of the comparison target to be unattainable, consequently resulting in self-deflation and in turn a compromised performance on the general knowledge test. Although the results from the present research demonstrated no noteworthy effects of processing style for females, males were shown to perform better on the knowledge test after being primed with local processing. Again, it may be that local processing aids general knowledge performance, though further research is required in order to determine the basis of this effect. With regard to the effect of gender on social comparison outcomes, although no effect of social comparison was revealed for females, males were shown to perform better on the general knowledge test after engaging in an upward comparison, indicating a behavioural assimilation effect. Thus, it is possible that the comparison vignettes were regarded to be more relevant (Lockwood & Kunda, 1997) for male participants, or that males may have 6

Admittedly, the interaction effect did not reach significance in the present study (F(2,16) = 3.17, p = .069). Nonetheless, when the analysis was broken down into two two-way interactions, significant effects were revealed for ESL.


63

approached the social comparison in a more competitive way and thus were more affected by the upward comparison standard when compared to their female counterparts.

LIMITATIONS OF THE PRESENT RESEARCH There were several limitations within the present investigation that should be addressed. First, the present research was based on subjective self-reports in response to a hypothetical comparison scenario. It should be noted that individuals may respond differently to social comparisons in the real world. As underscored by Wheeler and Miyake (1992), because social comparison is a dynamic process, it is best studied under naturalistic conditions. In addition, no baseline measures of processing style were taken, which would have allowed for further determination of how individual differences in processing style might influence the effects of priming.

CONCLUSION Investigation into the mechanisms behind the consequences of social comparison is important due to the common occurrence of social comparison in everyday life. Social comparison research may have important implications for personality conceptualisations of self-concept, in so far as comparison with others allows individuals to develop a better understanding of themselves in relation to others. Notably, the present study did not provide direct support for Förster et al.‘s (2008) GLOMO with respect to the self-evaluative and behavioural responses to social comparison. Nonetheless, the present research revealed incidental evidence that individual differences in language and gender may partially moderate the role of processing style on the behavioural consequences of social comparison. Further research is required in order to determine the role of such individual difference variables in social comparison outcomes, and to further clarify how responses to social comparisons vary for individuals from different backgrounds.

ACKNOWLEDGMENT We thank Ladd Wheeler and Julie Sribney for their comments on an earlier draft of this paper.

REFERENCES Bandura, A. (1989). Human agency in social cognitive theory. American Psychologist, 44, 1175-1184.

64


Bargh, J. A., Chen, M., & Burrows, L. (1996). Automaticity of social behaviour: Direct effects of trait construct and stereotype activation on action. Journal of Personality and Social Psychology, 71, 230-244. Blough, P. M., & Slavin, L. K. (1987). Reaction time assessments of gender differences in visual-spatial performance. Perception and Psychophysics, 41, 276-281. Dijksterhuis, A., & van Knippenberg, A. (1998). The relation between perception and behaviour, or how to win a game of Trivial Pursuit. Journal of Personality and Social Psychology, 74, 865-877. Festinger, L. (1954). A theory of social comparison processes. Human Relations, 7, 117-140. Finke, R. A. (1985). Theories relating mental imagery to perception. Psychological Bulletin, 98, 236–259. Förster, J. (2009). Relations between perceptual and conceptual scope: How global versus local processing fits a focus on similarity versus dissimilarity. Journal of Experimental Psychology: General, 138, 88-111. Förster, J., Liberman, N., & Kuschel, S. (2008). The effect of global versus local processing styles on assimilation versus contrast in social judgment. Journal of Personality and Social Psychology, 94, 579-599. Friedman, R. S., Fishbach, A., Förster, J., & Werth, L. (2003). Attentional priming effects on creativity. Creativity Research Journal, 15, 277-286. Jackson, D. N. (1998). Multidimensional Aptitude Battery II (MAB-II). Port Huron, MI: SIGMA Assessment Systems Inc. Jackson, D. N. (2003). Multidimensional Aptitude Battery II (MAB-II) manual, version 2.01. Port Huron, MI: SIGMA Assessment Systems Inc. James, W. (1890). The principles of psychology (Vol. 2). New York: Holt. Kail, R., Carter, P., & Pellegrino, J. W. (1979). The locus of sex differences in spatial ability. Perception and Psychophysics, 26, 182-186. Lockwood, P., & Kunda, Z. (1997). Superstars and me: Predicting the impact of role models on the self. Journal of Personality and Social Psychology, 73, 91-103. Lawton, C. A. (1994). Gender differences in way-finding strategies: Relationship to spatial ability and spatial anxiety. Sex Roles, 30, 765-779. Mendaglio, S., & Pyryt, M.C. (2002, April). Self-concept and giftedness: A multi-theoretical perspective. Paper presented at the Annual meeting of the American Educational Research Association (AERA), New Orleans, Louisiana. Mussweiler, T., & Strack, F. (2000). The ―relative self‖: Informational and judgmental consequences of comparative self-evaluation. Journal of Personality and Social Psychology, 79, 23-38. National Geographic Society. (2009). Killer whale (orca) profile. In Animal profiles. Retrieved April 14, 2009, from http://animals.nationalgeographic.com/animals/mammals/ killer-whale.html. Navon, D. (1977). Forest before trees: The precedence of global features in visual perception. Cognitive Psychology, 9, 353-383. Rilea, S. L. (2008). A lateralization of function approach to sex differences in spatial ability: A reexamination. Brain and Cognition, 67, 168-182. Schwarz, N., & Bless, H. (1992). Constructing reality and its alternatives: An inclusion/exclusion model of assimilation and contrast effects in social judgment. In L. L.


65

Martin & A. Tesser (Eds.), The construction of social judgments (pp. 217-245). Hillsdale, NJ: Erlbaum. Schwarz, N., & Bless, H. (2007). Mental construal processes: The inclusion/exclusion model. In D. A. Stapel & J. Suls (Eds.), Assimilation and contrast in social psychology (pp. 119142). New York: Psychological Press. Seaton, M. (2002). Social comparison, automatic behaviour and self-evaluation: Assimilation and contrast effects. Unpublished honour‘s thesis, Macquarie University, Sydney, New South Wales, Australia. Semin, G. R., Görts, C., Nandum, S., & Semin-Goossens, A. (2002). Cultural perspectives on linguistic representations of emotion and emotion events. Cognition and Emotion, 16, 1128. Stapel, D. A., & Semin, G. R. (2007). The magic spell of language: Linguistic categories and their perceptual consequences. Journal of Personality and Social Psychology, 93, 23-33. Turner, J. C. (1999). Some current issues in research on social identity and self-categorization theories. In N. Ellemers, R. Spears, & B. Doosje (Eds.), Social Identity: Context, commitment, content (pp. 6-34). Oxford, UK & Cambridge, USA: Blackwell. Turner, J. C., & Onorato, R. S. (1999). Social identity, personality, and the self-concept: A self-categorization perspective. In T. R. Tyler, R.M. Kramer, & O.P. John (Eds.), The psychology of the social self (pp. 11–46). Mahwah, NJ: Lawrence Erlbaum Associates. Wheeler, L., & Miyake, K. (1992). Social comparison in everyday life. Journal of Personality and Social Psychology, 62, 760-773. Wheeler, L., & Suls, J. (2007). Assimilation in social comparison: Can we agree on what it is? Revue Internationale de Psychologie Sociale, 20, 31-51.



Chapter 6

THE EXCITATION TRANSFER PARADIGM AND EMOTIONAL INTELLIGENCE: PREDICTING INTERPERSONAL ATTRACTION Madeleine Fraser and Simon Boag Department of Psychology, Macquarie University Sydney, NSW 2109, Australia

ABSTRACT The Excitation Transfer paradigm proposes that externally-caused residual arousal may later intensify unrelated feelings of attraction, and thus emphasising the role of situational variables. On the other hand, Emotional Intelligence (EI) involves an individual‘s internal capacity to interpret and engage with emotionally-relevant information. This study proposed that EI—(as assessed by the Trait-Meta Mood Scale— TMMS)—may moderate the Excitation Transfer effect. A sample of 104 female undergraduates was randomly allocated to complete a cognitive task whilst either seated at a table (control) or riding an exercise bike (aroused condition). Participants then viewed a video recording of a male confederate and rated his attractiveness on a 19-item First Impression Scale. Results indicated no main effects for arousal or EI on any factor of attractiveness. However further analysis demonstrated significant higher-order interactions between EI and arousal upon attractiveness factors. Specifically, individuals with higher EI scores were less likely to be influenced by irrelevant arousal. Results suggest that evaluative judgements of attractiveness are influenced by the interaction of personality and situational variables.

Keywords: Arousal, Emotional Intelligence, Excitation Transfer paradigm, Individual differences



Ph: 61-2-9850 8024; Fax: 61-2-9850 8062; [email protected].

68

Madeleine Fraser and Simon Boag

INTRODUCTION Romantic partner selection can be influenced by a diverse range of variables including both situational factors and internal personality traits. Zillmann‘s (1971) Excitation Transfer paradigm states that residual arousal caused by a previous event may later intensify a diverse range of reactions to other unrelated stimuli, including aggression (e.g., Zillmann, 1971), sexual arousal (e.g., Cantor, Zillmann & Bryant, 1975), or attraction (e.g., White, Fishbein & Rutstein, 1981). In the case of attraction, one might be mildly attracted to a new acquaintance under ordinary circumstances; however this feeling may be intensified when physiologically aroused, thereby eliciting the Misattribution of Arousal (Dutton & Aron, 1974). This phenomenon is based on the premise that a state of physiological arousal and excitation decays slowly over time, often remaining operative even after an individual has cognitively adjusted to a new situation. The term arousal source ambiguity suggests that residual arousal is more likely to be misinterpreted once obvious markers of physical exertion have dissipated (e.g., shortness of breath). Accordingly, misattributions do not usually occur when arousal is explicitly linked to its original source (Foster, Witcher, Campbell & Green, 1998; Gorn, Pham & Sim, 2001). A widely referenced study demonstrating the Excitation Transfer paradigm was conducted by Dutton and Aron (1974) using a high suspension bridge over a canyon to induce a state of physiological arousal. They found that aroused male participants who interacted with a female experimenter soon after crossing the bridge were significantly more likely to later call the experimenter and to have higher sexual content in their Thematic Apperception Test responses than participants with lower levels of arousal. These results suggest that residual physiological arousal was misinterpreted as feelings of attraction toward the experimenter. Subsequently, the Excitation Transfer paradigm has been widely replicated under a number of experimental manipulations ranging from rollercoasters (Meston & Frohlich, 2003) to physical exercise (White & Knight, 1984). However a body of research continues to demonstrate inconsistent results within the Excitation Transfer paradigm. Foster, Witcher, Campbell and Green‘s (1998) metanalysis of 33 Excitation Transfer experiments found that under typical experimental conditions, arousal exerts a small to moderate effect on attraction. These inconsistencies are particularly relevant for female participants. Dutton and Aron‘s (1974) original study failed to replicate their results for female participants using a male confederate. Similarly Kenrick, Cialdini, and Linder (1979) reported four failures to replicate Dutton and Aron‘s (1974) findings, one exact replication, and three replications with minor variations. Similarly, Dienstbier (1979) studied the impact of startle-induced arousal on attraction, finding a significant difference between control and aroused male participants but no difference for female participants. A review of previous Excitation Transfer research conducted on female samples demonstrates conflicting evidence both for and against the Misattribution of Arousal (Ben-Zeev, Fein & Inzlicht, 2003; Bootzin, Herman & Nicassio, 1976; Rickwood & Price, 1988). These disparate results suggest that there may be other factors contributing to the inconsistent Excitation Transfer effects for female populations.

The Excitation Transfer Paradigm and Emotional Intelligence

69

EMOTIONAL INTELLIGENCE, AROUSAL AND ATTRACTION The present study proposes that the misinterpretation of residual arousal does not uniformly effect or even occur to all individuals. Instead, these inconsistencies may be better explained by accounting for Emotional Intelligence (EI). Individuals differ in their ability to decipher which feelings they should trust as a reliable and accurate (Fiori, 2009), and EI refers to individual differences in the recognition, regulation, and expression of emotional information (Nelis, Quoidbach, Mikolajczak, & Hansenne, 2009). Consequently, EI may play a moderating role between perceiving emotional information and subsequently using this information to form an evaluative judgement. Petrides and Furnham (2001) propose a distinction between conceptually and operationally different types of EI. The first, ability models, describe EI as a set of cognitive abilities pertaining to the perception, interpretation and utilisation of affect-laden information (Mayer, Caruso & Salovey, 1999; Osofsky, 1991). In contrast, trait models conceptualise EI as a tendency to behave in a certain way in emotional situations and the frequency with which one uses one‘s emotional abilities and dispositions linked to personality (Petrides & Furnham, 2000). While both approaches have their merits, applying a trait-based understanding of EI in the current study facilitates a clear demarcation between the influences of situational variables (externally caused-levels of arousal) and internally stable EI traits. Trait EI is typically measured using self-report questionnaires such as the widely applied Trait Meta-Mood Scale (TMMS) (Salovey, Mayer, Goldman, Turvey & Palfai, 1995). The TMMS provides an index of individual differences in the mood regulation process termed the ‗meta-mood experience‘, involving monitoring, evaluating and regulating feelings and emotions (Salovey, Mayer, Goldman, Turvey & Palfai, 1995). The TMMS has demonstrated good psychometric properties, including satisfactory to good internal consistency ( range =.66 to .83; Tett, Fox & Wang, 2005) and continues to be applied to Australian samples (e.g., Fitness & Curtis, 2005). The TMMS loads onto three factors including ‗Attention to emotion‘ referring to the degree of focus placed on one‘s own emotions and emotional reactions to others; ‗emotional Clarity’, describing the capacity to notice feelings and accurately recognise their nature and cause, and; ‗emotional Repair’, involving the regulation of feelings and emotions, and capacity to adaptively motivate behaviour (Salovey, Mayer, Goldman, Turvey & Palfai, 1995). A final Global EI factor represents an overall score, indicative of a general capacity to engage with emotional information. Given that the Excitation Transfer paradigm involves a situation whereby individuals fail to distinguish between incidental feelings and feelings elicited by a target, there are several ways that EI may be relevant to understanding the misattribution of arousal. Fiori (2009) suggests that individuals with higher EI may be less influenced by misleading emotions or arousal than individuals with lower EI abilities. A related study by Mikolajczak, Petrides, Coumans and Luminet (2009) found that trait EI moderates the impact of laboratory-induced stress on mood change. Specifically, after inducing stress through an experience of failure, higher trait EI scores were linked to significantly less mood deterioration. Generalising this pattern, it is predicted that individuals scoring higher on EI will be less likely to mistakenly interpret residual arousal as indicative of feelings of attraction toward a target. Furthermore, females tend to receive higher trait-EI scores than males on a variety of measures (Katyal & Awasthi, 2005; Petrides, Furnham & Martin, 2004) and links between EI and uniquely female

70


experiences (e.g., menopause and orgasm frequency) suggest EI may play an important role in everyday functioning and decision-making particularly for females (Bauld & Brown, 2009; Burri, Cherkas & Spector, 2009).

AIM AND HYPOTHESES The aim of the present study was to examine the relationship between arousal, Emotional Intelligence, and attraction, hypothesising that a more precise account of the decisions underlying romantic partner selection may be gained through accounting for both environmentally induced arousal and EI. Firstly, based on the Excitation Transfer paradigm it was predicted that aroused female participants should rate male confederate attractiveness higher than control participants. Secondly, EI scores should moderate the relationship between arousal and attraction. Thirdly and more specifically, it was hypothesised that participants with lower EI scores should rate a confederate as more attractive when in a state of physiological arousal compared to a control.

METHOD Participants The sample consisted of 104 female undergraduate students aged between 17 and 35 years (M = 19, SD = 3) enrolled in a first-year Psychology course.

Materials and Apparatus Trait-Meta Mood Scale (TMMS): The 30 item TMMS questionnaire assesses trait-EI and was administered on a laptop using the MediaLab computer program. Items were individually displayed above a 5-point Likert scale, with 1 = ‗not at all‘ and 5 = ‗very much so‘. In the present study, the overall reliability of the scale was good (Cronbach‘s  = .83), and the reliability of the subscales included Attention  = .65; Clarity  = .86; Repair  = .80. These reliability scales are comparable to those obtained by the original Salovey, Mayer, Goldman, Turvey and Palfau (1995) study (Attention  = .78; Clarity  = .87; Repair  = .76). Each EI subscale was treated as a continuous rather than ordinal variable to maximise the opportunity to identify patterns related to EI scores.

First Impressions Scale To construct an information-rich measurement of confederate attractiveness, two widely applied questionnaires measuring both ‗love‘ and ‗liking were combined to form the First Impressions Scale. Seven items were derived from Foster et al. (1998) including questions such as ―Is he the type of person you would want to get to know better?‖ A further twelve items from White et al. (1981) rated confederate personality traits such as likeability, sincerity and conceitedness. The thirteenth item ―How sexually warm do you think the male in the


71

video is?‖ was excluded from the current study as nine of the ten participants in a video prescreening test expressed confusion concerning its meaning. The nineteen items in the final scale were randomly ordered and used a 10-point Likert scale ranging from 0-9 (higher scores reflecting a more positive evaluation). There was no ‗neutral‘ option so as to optimise scale sensitivity to differences in attraction between conditions, whilst also avoiding mixed interpretations of ‗neutral‘ and minimizing response set tendencies (Kulas, Stachowski & Haynes, 2008). Furthermore, the Likert scale included ‗0‘ as a more appropriate alternative to ‗1‘ to cater for an absence of attraction.

Arousal Manipulation and Manipulation Check To measure the presence of residual physiological arousal, the heart rate of a subsample of 70 participants was measured using a Panasonic wrist monitor at three stages; pre-maze, post-maze completion and eight minutes later at the conclusion of the experiment. Heart-rate has previously been linked to a state of arousal/excitation (Pastor, Mayo & Shamir, 2007).

Video Clip of Male Confederate A video-recording of a male confederate was selected as a standardised yet informationrich stimulus. Based on a pre-test screening of ten participants blind to the experimental hypothesis, one recording of a male confederate was selected from six potential confederates based on a higher and more consistent rating on the First Impressions Scale (M = 5.9 on a 0-9 scale, SD = 1.4). The confederate had a relatively symmetrical face in accordance with the robust finding that ‗average‘ faces are perceived as more attractive universally (Rubenstein, Langlois & Roggman, 2002). The confederate‘s shoulders and head were filmed whilst they delivered a 45 second speech about university life. The content of the script was deliberately generic and broad, devoid of any specific attitudes to avoid the confounding influence of attitudinal similarity-dissimilarity (Byrne, 1997). The video clip was shown to participants on a laptop computer using Windows Media Player software.

Procedure The methodological paradigm was adapted from Cantor et al. (1975) and White, Fishbein and Rutstein (1981), using physical exercise to induce arousal before examining the effects of residual arousal on attraction toward a confederate. This procedure has been widely applied in attraction research (Allen, Kendrick, Linder & McCall, 1989; Meston & Gorzalka, 1996; White & Knight, 1984). Prior to participation, all participants provided written consent and were checked for possible medical conditions. Participants were informed the study was investigating (1) the effect of exercise on concentration and (2) first impressions. This deceptive cover story served to maximise arousal source ambiguity in the later stages of the experiment. After being told they were given a ‗mood assessment‘ test and asked to answer the questions as honestly as possible, participants completed the TMMS.

72

Madeleine Fraser and Simon Boag Table 1. Means & standard deviations of the TMMS & First Impressions Scale

Scale

Subscales

TMMS

Global Attention Clarity Repair Total Scale Traits Behaviour Romantic Negative Traits Excite

First Impressions

Number of Items 30 21 15 12 19 4 5 4 3 3

Total Score

Mean

150 105 75 60 190 40 50 40 30 30

109. 19 48.87 37.19 23.13 120.61 24.67 28.72 28.39 26.54 11.41

Standard Deviation 11.71 4.79 6.10 4.14 18.36 5.50 7.46 4.49 4.53 2.61

Arousal Manipulation Participants were then randomly allocated to either the arousal condition (riding a stationary exercise bike on medium difficulty) or the control condition (seated at a table). Both conditions were given five minutes to find as many correct exits from four maze puzzles as possible. Participants read their answers aloud for the experimenter to record and were not provided with accuracy feedback. All participants were then seated and given three minutes to fill out a crossword puzzle of medium difficulty. On average, 24 of the 56 questions were completed, indicating a relatively consistent level of difficulty for all participants (SD = 12.6; range = 9-48). This filler task served to alleviate the obvious markers of physical exertion (e.g., shortness of breath), thereby optimising the opportunity for residual arousal to be misattributed, whilst also maintaining consistent delay times across conditions, thus circumventing problems involving different delay periods evident in previous studies (e.g., White & Knight, 1984). Attractiveness Assessment Participants then viewed a 45 second video clip of a male confederate and completed the First Impressions Scale. Finally, participants were given a short demographic questionnaire gathering information on potentially confounding variables including age, current relationship status and sexual orientation. To conclude, participants were probed for suspicion regarding the true purpose of the study and debriefed.

RESULTS Based on relationship status, statistical analyses were run separately for participants who were single and those in a romantic relationship to account for the potentially confounding influence of a romantic partner demonstrated in prior research (Meston & Frohlich, 2003; Little, Jones, Penton-Voak, Burt & Perrett, 2002). 52 participants were currently single, and 52 in a relationship. After random allocation, roughly equal proportions were allocated to the control and aroused conditions (58% and 48%). As there were no missing cases in the data, and no significant outliers detected (including 4 participants who were interested in females


73

or both genders), the full data set (N = 104) was used in the analyses. The means and standard deviations of both the TMMS and the First Impressions Scale are displayed in Table 1. Table 2. Five factor solution using a principal components analysis based on the 19 items of the First Impressions Scale First Impressions Items 1 Trait 9. How sympathetic do you think the male in the video is? † 14. How dependable do you think the male in the video is? † 13. How generous do you think the male in the video is? † 10. How sincere do you think the male in the video is? † 12. How much would you like to kiss him? * 17. How much would you like to date him? * 15. How much would you like to work with him? * 16. How well do you think you could get along with him? * 8. How much do you like him? * 1. How physically attractive is the male in the video? * 3. How considerate do you think the male in the video is? † 4. Is he the type of person you would want to get to know better? * 2. How likeable do you think the male in the video is? * 11. How shallow do you think the male in the video is? † 5. How irritating do you think the male in the video is? † 18. How conceited do you think the male in the video is? † 6. How humorous do you think the male in the video is? † 7. How exciting do you think the male in the video is? † 19.How perceptive do you think the male in the video is? † * †

2 Behaviour

Factors 3 4 Roma Negati-ntic ve Traits

5 Excite

Commu -nalities

.89

.70

.85

.73

.78

.73

.72

.60 .91

.79

.83

.78

.58

.75

.56

.74

.39

= items from Foster, Witcher, Campbell & Green, 1998. = items from White, Fishbein & Rutstein, 1981.

.79

.62 .70

.79

.80

.75

.72

.73

.72 .85

.75

.82

.71

.74

.71 .78

.77

.69

.78

.47

.51

74


Table 3. Bivariate correlations among the five factors derived from the 19 itemed First Impressions Scale

1) Trait 2) Behaviour 3) Romantic 4) Negative Traits 5) Excite

SD 5.50 7.50 4.49 4.53 2.61

Factors 1 1.0

2 0.40* 1.0

3 0.42* 0.61* 1.0

4 0.22* -0.09 0.21* 1.0

5 0.46* 0.49* 0.47* -0.31* 1.0

Note: * = Correlation is significant at the 0.05 level (one-tailed).

Factor Analysis of the First Impressions Scale A factor analysis provided evidence of sampling adequacy for the First Impressions Scale (Kaiser-Meyer-Olkin measure = .853; Barlett‘s Test of Sphericity = p <.001) and a principal components analysis revealed the scale loaded onto five factors with eigenvalues exceeding 1, which explain 35.84%, 13.56%, 10.26%, 6.12% and 5.69% of the variance respectively (see Table 2). A promax rotation was applied in the present study as a non-orthogonal method of rotation. For exploratory purposes, the present study used 0.4 as the threshold for determining which item loaded onto which factor (Raubenheimer, 2004), and each factor was allocated according to its highest loading.

Figure 1. Pulse rate differences for control & arousal conditions prior to the maze task, immediately after & at the conclusion of the experiment (including standard error bars).


75

The first factor was comprised of four positive personality traits; each derived from White, Fishbein, and Rutstein (1981) and thus was renamed Traits. The second factor, Behaviour, included five behavioural questions involving interaction between the participant and male target, each derived from Foster et al., (1998). The third factor involved a more eclectic combination of items from the two original scales involving romantic behaviours and personality traits, and was labelled Romantic. The fourth factor was comprised of entirely negative personality traits which were all reverse coded in the scale (labelled Negative Traits). Finally, the fifth factor was renamed Excite and contained three personality trait judgements; exciting, humorous and perceptive. Focus was placed on Trait, Behaviour and Romantic factors due to their higher predictive capacity, accumulatively accounting for 59.66% of total score variability. Table 3 displays the correlation matrix for the five factors derived from the 19 itemed First Impressions Scale. Although each factor retains a low to moderate correlation with other factors, the majority of these are significant at the 0.05 level. Intuitively these correlation coefficients would be expected to share a degree of significance given the factors are measuring similar variables.

Manipulation Check Mean heart rates in the control condition did not significantly differ between Pre-Maze, Post-Maze time periods and at the experiment‘s Conclusion, F(1.74, 58.2) = 2.41, p  .05, as illustrated in Figure 1. Mauchly‘s test indicated that the assumption of sphericity had been violated in the calculation (chi-square = 7.71, p  .05), therefore the degrees of freedom were corrected using Huynh-Feldt estimates of sphericity (epsilon = .87). Contrastingly, in the arousal condition pulse rates significantly differed at each of the three stages of recording (F(1.36, 206) = 22.60, p < .05). Mauchly‘s test indicated that the assumption of sphericity had again been violated (chi-square = 13.43, p  .05), therefore the degrees of freedom were corrected using the Greenhouse-Geisser estimate of sphericity (epsilon = .74). Post hoc tests revealed that the pulse rate of the post-maze participants was significantly higher than the premaze participants (p  .05) and remained higher at the conclusion of the experiment in comparison to the pre-maze pulse rates (p  .05). These results confirm that residual arousal was present at the conclusion of the experiment in the aroused condition. Multiple regressions were conducted to assess the relationship between arousal and attraction, and also EI and attraction. Results demonstrated no significant difference between aroused and control conditions on any of the five factors derived from the First Impressions Scale (all p > .05). This finding did not support the Excitation Transfer hypothesis that aroused female participants would rate the attractiveness of a male confederate higher than non-aroused females. As expected, there were no main effects for any EI scale (Attention, Clarity, Repair or Global) on any of the five First Impressions Scale factors (all p > .05).

76


Investigating the Effect of EI on the Excitation Transfer Paradigm for Currently Single Participants

Confederate mean Trait Score

A series of multifactorial ANCOVAs using the General Linear Model examined the interactions between condition (control and arousal) and EI scores (Global EI, Attention EI, Clarity EI and Repair EI) with each of the five factors derived from the First Impressions Scale as dependent variables (Trait, Behaviour, Romantic, Negative Traits and Excite). This analysis was run separately for participants who were in a relationship and those who were single. Two significant interactions were found for participants who were single. No significant results were found for participants who were in a relationship.

Contro l

Global EI Score

Confederate mean Trait Score

Figure 2. For currently single participants; interaction between Traits & Global EI scores for arousal & control conditions.

Control Arousal

Clarity EI Score Figure 3. For currently single participants; interaction between Traits & Clarity EI scores for arousal & control conditions.


77

The first significant interaction was between condition and Global EI with Traits as the dependent variable (B = 26.43, F (1, 50) = 4.71, p < .05), as illustrated in Figure 2. Tests of simple slopes showed that for the arousal group, the relationship between Global EI and Traits was negative yet not significantly different from zero (B = -0.19, t (50) = -1.81, p = .07). For the control group, the relationship was positive, yet not significantly different from zero (B = .34, t (50) = 1.82, p = .08). The effect of the interaction demonstrated that the difference between the conditions was greater at lower levels of Global EI than at higher levels. For example, at one standard deviation below the mean of Global EI, the model indicated a significant difference between the condition (B = -5.59, t (50) = -2.11, p < .05). At one standard deviation above the mean of Global EI, the difference between the conditions was not significant (B = 2.7, t (50) = 1.14, p = .26). There was a second significant interaction between condition and Clarity EI with Traits as the dependent variable, B = 26.01, F(1, 53) = 5.99, p < .05. This is illustrated in Figure 3. Tests of simple slopes showed that for the arousal group, the relationship between Clarity EI and Traits was negative, but not significantly different from zero (B = -.26, t (50) = -1.64, p = .11). For the control group, the relationship was positive, and also not significantly different from zero (B = .34, t (50) = 1.82, p = .07). The effect of the interaction was that the difference between the conditions was greater at low levels of Clarity EI than at higher levels. For example, at one standard deviation below the mean of Clarity EI, the model indicated a significant difference between the condition (B = -5.51, t (50) = -2.38, p = .02). Conversely, at one standard deviation above the mean of Clarity EI, the difference between the conditions was not significant (B = 2.41, t (50) = 1.09, p = .28).

DISCUSSION The present results suggest that the effect of residual arousal on interpersonal attraction differs depending on the individual‘s capacity to engage with and interpret emotionallyrelevant information, as measured by trait-EI. Firstly, based on the Excitation Transfer paradigm, in the current study it was hypothesised that participants in the aroused condition would rate a male confederate as more attractive than those in the control condition. Although the manipulation check illustrated the presence of residual arousal, no main effect for arousal on attraction was found. This aligned with a small body of research suggesting females are less likely to misattribute arousal than males (Ben-Zeev, Fein & Inzlicht, 2003; Bootzin, Herman & Nicassio, 1976; Dutton & Aron, 1974; Dienstbier, 1979; Rickwood & Price, 1988). Similarly, no main effects were present for EI (Global EI, Attention EI, Clarity EI and Repair EI) on any of the five factors derived from the First Impressions Scale (Traits, Behavioural, Romantic, Negative Traits and Excite). Although accounting for arousal and EI separately produced no main effects, after including both variables in the analyses, two significant higher-order interactions emerged. These interactions provided support for the second and third hypothesis. Specifically, it was hypothesised that participants with the same, lower EI scores would rate a confederate as more attractive when in a state of physiological arousal compared to a control. The first interaction between condition and Global EI with Traits as the dependent variable demonstrated support for this hypothesis. The difference between arousal

78


and control conditions was significantly greater at lower levels of Global EI than at higher levels of Global EI. This may be interpreted to mean that female participants with higher levels of Global EI rate the attractiveness of a male confederate in a similar fashion regardless of whether or not they are influenced by residual arousal. Comparatively, female participants with lower levels of Global EI will rate the attractiveness of a male confederate differently depending on the presence of residual arousal. Thus participants who have previously experienced a physiologically arousing stimulus (operationalised through physical exercise), are more likely to rate the confederate favourably and award higher attractiveness scores due to the presence of residual arousal. The second significant interaction between condition and Clarity EI with Traits as the dependent variable produced similar results. Participants who had lower Clarity EI scores rated the male confederate as significantly more attractive in the aroused rather than control condition. Thus both significant interactions suggest female participants with high Clarity and Global EI scores are more likely to recognise the nature of their inner feelings and emotions and be able to accurately attribute the cause of physiological arousal to exercise rather than interpreting these feelings as attraction toward a target. Conversely, participants with lower Clarity and Global EI are more likely to misinterpret the cause of residual arousal as a positive reaction toward the confederate, reflected by higher ratings on positive Trait factors. This reflects a trend in EI research for EI traits to moderate rather than directly control or direct thought processes and evaluative judgements (e.g., Extremera, Duran & Rey, 2009).

OTHER FINDINGS In both significant interactions the reversed pattern was found in the control condition. In the absence of physiological arousal, participants with higher Clarity or Global EI gave the confederate higher Trait scores than participants with lower Clarity or Global EI. Whilst these scores were not significantly different from high EI scorers in the aroused condition, both interactions demonstrated this trend. A potential explanation for this effect may be that in the absence of residual arousal, higher Clarity or Global EI scores are more likely to demonstrate optimism and charitable feelings toward others. Thus in the presence of potentially misleading residual arousal, higher EI capacities appear to effectively prevent biased judgements; however in the absence of residual arousal, higher EI encourages more charitable and optimistic judgements. This explanation has not been directly tested, but is indirectly supported by previous descriptions of an emotionally intelligent character as well-adjusted, genuine, warm, persistent, and optimistic (Mayer, DiPaolo & Salovey, 1990, pp. 773, 781; Salovey & Mayer, 1990, pp. 199-200).

THE INFLUENCE OF RELATIONSHIP STATUS Importantly, each interaction was only demonstrated for participants who were currently single. The lack of a significant interaction for participants currently in a relationship aligns with Meston and Frohlich‘s (2003) field study which used a rollercoaster to induce physiological arousal. Comparative effects between the participants own romantic partner and


79

male confederate may account for the lack of a significant interaction between arousal, EI and attraction for participants currently in a relationship. Consequently, these comparative effects may override initial judgements of another‘s attractiveness. This particularly being the case as it may be assumed a current partner who has an intimate connection with the participant will be favoured over a brief, limited depiction of a male stranger. Thus, being in a relationship or the presence of a romantic partner was found to moderate the arousal-attraction link. Consequently relationship status appears to also act as a moderating variable.

IMPLICATIONS An important implication from the present findings is that acknowledging both external (environmentally caused arousal) and internal variables (EI) leads to a more comprehensive understanding of why the Excitation Transfer effect is not consistently reflected in all female samples (e.g., Dienstbier, 1979; Dutton & Aron, 1974). EI may be understood to contribute to the underlying the mental processing which translates situational information (externally caused residual arousal) into evaluative judgements (interpersonal attraction). Accordingly, the inclusion of EI within the Excitation Transfer paradigm lends to an increasingly precise understanding of how a situation affects individuals differently depending on their internal dispositions (Mischel, Shoda & Mendoza-Denton, 2002). Thus the present study‘s investigation of personality-situation interactions makes a timely contribution to Excitation Transfer research which was previously dominated by studies focusing primarily on situational factors (Dutton & Aron, 1974; Foster, et al., 1998).

CONCLUSION These findings must be interpreted within the broader results obtained by the study. A larger pattern of non-significant relationships between different combinations of EI subscales and factors suggest EI cannot uniformly predict the effects of arousal for all individuals. Consequently, as the first study to combine EI and Excitation Transfer fields of research, the present results cannot be considered as a definitive conclusion in this area. The present results do, however, suggest that an individual‘s capacity to notice and accurately recognise one‘s own feelings and their cause (Clarity EI) in addition to their general capacity to engage with emotional information (Global EI) moderates the Excitation Transfer effect and the process of using emotional information to form an evaluative judgment. The applied methodology also contributes to EI research by potentially operationalising a situational measure of EI. Future studies may investigate the Excitation Transfer effect as an alternative to self-report measures of trait-EI. A further follow-up study may be a replication for male participants, potentially generalising current results to other situations of Excitation Transfer such as aggression. Thus this investigation of personality-situation interactions contributes to Excitation Transfer research by suggesting that accounting for both external (environmentally caused arousal) and internal variables (EI) leads to a more nuanced understanding of the processes underlying the arousal-attraction link.

80


ACKNOWLEDGMENT We would like to thank Dolf Zillmann and John Reid for their encouragement and Alan Taylor for his invaluable statistical advice.

REFERENCES Allen, J. B., Kendrick, D. T., Linder, D. E., & McCall, M. A. (1989). Arousal and attraction: A response facilitation alternative to misattribution and negative reinforcement models. Journal of Personality and Social Psychology, 57, 261-270. Bauld, R., & Brown, R. F. (2009). Stress, psychological distress, psychosocial factors, menopause symptoms and physical health in women. Maturitas, 62, 160-165. Ben-Zeev, T., Fein, S., & Inzlicht, M. (2003). Stereotype threat and arousal. Journal of Experimental Social Psychology, 41, 174-181. Bootzin, R. R., Herman, C. P., & Nicassio, P. (1976). The power of suggestion: Another examination of the misattribution and insomnia. Journal of Personality and Social Psychology, 34, 673-679. Burri, A. V., Cherkas, L. M., & Spector, T. D. (2009). Emotional intelligence and its association with orgasmic frequency in women. Journal of Sexual Medicine, 6, 19301937. Byrne, D. (1997). An overview (and underview) of research and theory within the attraction paradigm. Journal of Social and Personal Relationships, 14, 417-431. Cantor, J. R., Zillmann D., & Bryant J. (1975). Enhancement of experienced sexual arousal in response to erotic stimuli through misattribution of unrelated residual excitation. Journal of Personality and Social Psychology, 32, 69-75. Dienstbier, R. A. (1979) Attraction increases and decreases as a function of emotionattribution and appropriate social cues. Motivation and Emotion, 3, 201-218. Dutton, D. G., & Aron, A. P. (1974). Some evidence for heightened sexual attraction under conditions of high anxiety. Journal of Personality and Social Psychology, 30, 510-517. Extremera, P., & Durán, A. & Rey, L. (2009). Perceived emotional intelligence and dispositional optimism–pessimism: Analyzing their role in predicting psychological adjustment among adolescents. Personality and Individual Differences, 42, 1069-1079. Fiori, M. (2009). A new look at emotional intelligence: A dual-process framework. Personality and Social Psychology Review, 13, 21-44. Fitness, J., & Curtis, M. (2005). Emotional Intelligence and the Trait Meta-Mood Scale: relationship with empathy, attribution complexity, self-control and responses to interpersonal conflict. E-Journal of Applied Psychology: Social Section, 1, 50-62. Foster, C. A., Witcher, B. S., Campbell, W. K., & Green, J. D. (1998). Arousal and attraction: Evidence for automatic and controlled processes. Journal of Personality and Social Psychology, 74, 86-101. Gorn, G. J., Pham, M. T., & Sim, L. Y. (2001). When arousal influences add evaluation and valence does not (and vice versa).Journal of Consumer Psychology 11, 43-55. Katyal, S., & Awasthi, E. (2005). Gender differences in Emotional Intelligence among adolescents of Chandigarh. Journal of Human Ecology, 17, 153-155.


81

Kenrick, D. T., Cialdini, R. B., & Linder, D. E. (1979). Misattribution under fear-producing circumstances: Four failures to replicate. Personality and Social Psychology Bulletin, 5, 329-334. Kulas, J. T., Stachowski, A. A. & Haynes, B. A. (2008). Middle response functioning in Likert responses to personality items. Journal of Business and Psychology, 22, 251-260. Little, A. C., Jones, B. C., Penton-Voak, I. S., Burt, D. M., & Perrett, D. I. (2002) Partnership status and the temporal context of relationships influence human female preferences for sexual dimorphism in male face shape. Proceedings of the Royal Society of London Series B-biological Sciences, 269, 1095-1100. Mayer, J. D., Caruso, D. R., & Salovey, P. (1999). Emotional intelligence meets traditional standards for an intelligence. Intelligence, 27, 267-298. Mayer, J. D., DiPaolo, M. T. & Salovey, P. (1990) Perceiving affective content in ambiguous visual stimuli: A component of emotional intelligence. Journal of Personality Assessment, 54, 772-781. Meston, C. M., & Frohlich, P. F. (2003). Love at First Fright: Partner Salience Moderates Roller-Coaster-Induced Excitation Transfer. Archives of Sexual Behaviour, 32, 537-548. Meston, C.M., & Gorzalka, B.B. (1996). The differential effects of sympathetic activation on sexual arousal in sexually dysfunctional and functional women. Journal of Abnormal Psychology, 105, 582-591. Mikolajczak, M., Petrides, K., Coumans, N., & Luminet, O. (2009). The moderating effect of trait emotional intelligence on mood deterioration following laboratory-induced stress International Journal of Clinical and Health Psychology, 9, 455-477. Mischel, W., Shoda, Y., & Mendoza-Denton, M. (2002). Situation-behavior profiles as a locus of consistency in personality. Current Directions in Psychological Science, 11, 5068. Nelis, D., Quoidbach, J., Mikolajczak, M., & Hansenne, M. (2009). Increasing emotional intelligence: (How) is it possible? Personality and Individual Differences, 47, 36-41. Osofsky, J. D. (1991). Handbook of infant development (2nd ed.). New York: John Wiley & Sons. Pastor, J. C., Mayo, M., & Shamir, B. (2007) Adding fuel to fire: The impact of followers' arousal on ratings of charisma. Journal of Applied Psychology, 92, 1584-1596. Petrides K.V., Furnham A. & Martin, G. N. (2004) Estimates of Emotional and Psychometric Intelligence: Evidence for Gender-Based Stereotypes. The Journal of Social Psychology, 144, 149-162. Petrides, K. V. & Furnham, A. (2000). On the dimensional structure of emotional intelligence. Personality and Individual Differences, 29, 313-320. Petrides, K. V. & Furnham, A. (2001). Trait emotional intelligence: Psychometric investigation with reference to established trait taxonomies. European Journal of Personality, 15, 425-448. Raubenheimer, J. E. (2004). An item selection procedure to maximize scale reliability and validity. South African Journal of Industrial Psychology, 30, 59-64. Rickwood, L. V. & Price, G. (1988). Hedonic arousal, time intervals, and excitation transfer. Current Psychology, 7, 241-256. Rubenstein, A. J., Langlois, J. H., & Roggman, L. A. (2002). What makes a face attractive and why: The role of averageness in defining facial beauty. In G. Rhodes & L. A.

82


Zebrowitz (Eds.), Facial attractiveness: Evolutionary, cognitive and social perspectives (pp. 1-33). Westport, CT: Ablex. Salovey, P. & Mayer, J.D. (1990). Emotional intelligence. Imagination Cognition and Personality, 9, 185-211. Salovey, P., Mayer, J. D., Goldman, S. L., Turvey, C., & Palfai, T. P. (1995). Emotional attention, clarity, and repair: Exploring emotional intelligence using the trait meta-mood scale. In J. W. Pennebaker (Ed.), Emotion, disclosure and health (pp. 125-154). Washington, DC: American Psychological Association. Tett, R. P., Fox, K. E., & Wang, A. (2005). Development and validation of a self-report measure of emotional intelligence as a multidimensional trait domain. Personality and Social Psychology Bulletin, 31, 859-888. White, G. L., Fishbein, S., & Rutstein, J. (1981). Passionate love and the misattribution of arousal. Journal of Personality and Social Psychology, 41, 56-62. White, G. L., & Knight, T. D. (1984). Misattribution of arousal and attraction: Effects of salience of explanations for arousal. Journal of Experimental Social Psychology, 20, 5564. Zillman, D. (1971). Excitation transfer in communication-mediated aggressive behavior. Journal of Experimental Social Psychology, 7, 419-434.



Chapter 7

TAKE A BREAK: A LOOK AT THE INCUBATION EFFECT ON CREATIVITY Gerry Pallier and Niko Tiliopoulos School of Psychology, University of Sydney Sydney, NSW 2006, Australia

ABSTRACT We outline the somewhat contentions construct of creativity followed by the working definition we used in this chapter. That was the production of novel ideas—also known as ideational fluency—which is more readily amenable to empirical research than some other definitions. An experimental study using that notion and involving both visual and verbal creativity tasks, alongside measures of intelligence, is presented. Included in the study was an assessment of the so-called incubation effect, operationalised via a break in the testing protocol. The outcome indicated that a distinction existed between the visual and verbal creativity tasks and the intelligence measures. Furthermore, there appeared to be a difference between verbal and visual creativity, in that people were able to produce a greater increase in visual responses after the break. This is unusual since, whilst extra new ideas are commonly reported after a break, producing more responses after a break than before it has not previously been reported. This outcome is discussed and possible explanations are conjectured.

Keywords: Break Benefit, Creativity, Incubation effect, Intelligence, Intentional fluency

INTRODUCTION Since the time of Galton, creativity has often been held to be a crucial component of ‗eminence‘ or ‗genius‘; however, the issue of how creativity is to be defined has proven 

Ph: 61-2-9036 9223; Fax: 61-2-9036 5223; email: [email protected]

84

Gerry Pallier and Niko Tiliopoulos

contentious. It is an open question whether creativity is a unitary trait, or whether there exist distinct forms of creativity for various domains of endeavour (Hocevar, 1981). Amongst those who believe that creativity is unitary, there is disagreement over whether it can be captured by the construct of novelty, or whether the appropriateness of the product should also be taken into consideration when determining creativity (e.g., Sternberg & Lubart, 1995). In practice, however, creativity is often conceived of as a form of problem solving, with ‗problems‘ being broadly defined as goals to be reached (Lubart, 2000-2001). In this chapter we focus on that conceptualization by examining an aspect of creativity that would seem to have considerable utility in everyday life, namely the generation of novel ideas.

CREATIVITY AND PROBLEM-SOLVING There is a considerable overlap between the literature on creativity and that on problem solving. Historically, problem solving was a focus of theorising by Gestalt psychologists. Wallas (1926) developed an influential model that involved four stages: (i) Preparation, which involves defining a problem and consciously attempting to solve it; (ii) Incubation, wherein, when a solution has not been forthcoming, conscious work ceases, but continues nonconsciously; (iii) Illumination, which encompasses the moment of insight, and; (iv) Verification, whereby the solution is refined and confirmed. Whilst the preparation and verification stages have been the subject of considerable research within both cognitive and differential psychology, the incubation phase is somewhat more mysterious, and has proven less amenable to scientific investigation.

THE INCUBATION EFFECT Many readers will have experienced the phenomenon of a solution suddenly springing to mind when one has given up working on a problem, and the occurrence of the ―Aha!‖ or ―Eureka!‖ phenomenon is well documented in the creativity literature. Anecdotal accounts of creative ideas or solutions ‗springing to mind‘ after a period of incubation abound, but research on that topic has been sparse, and the few empirical results have been mixed (e.g., Dorfman, Shames & Kihlstrom, 1996). Although some studies have supported the existence of an incubation effect, most of those used a convergent problem-solving task (e.g., Dreistadt, 1969; Murray & Denny, 1969), whereby individuals have been aware that they have not found a solution, and, thus, may consciously have still been seeking a resolution. Fulgosi and Guilford (1968), and Snyder, Mitchell, Ellwood, Yates and Pallier (2004b) are the only researchers to have found evidence for an incubation effect using a divergent thinking task. Snyder, et al. (2004b) undertook preliminary research on the value of a break in a creativity task that required participants to generate as many possible uses for a piece of paper. When participants had seemingly

Take a Break: A Look at the Incubation Effect on Creativity

85

exhausted their supply of ideas, they were told that the task was over, and were interviewed in order to collect biographical information. Afterwards, to their surprise, participants were asked once again to provide uses for paper. Results demonstrated that participants produced a burst of new ideas after the break. The researchers concluded that this might be an instance of nonconscious idea generation. That finding inspired our investigations, which sought to examine further the possibility of an incubation effect by comparing the idea production of individuals on seemingly disparate tasks. In addition, the discriminant validity of the task employed by Snyder et al. (2004b), a measure of so-called Ideational Fluency (IF, see Carroll, 1993), was examined in relation to established measures of intelligence derived from the framework of the theory of fluid and crystallised abilities (Gf/Gc: see, e.g., Horn & Noll, 1994).

EXPLAINING THE INCUBATION EFFECT In cases such as those cited above, where an incubation effect has purportedly been demonstrated, accounts in terms of nonconscious processing have proven contentious (Dorfman, Shames & Kihlstrom, 1996), partly because they rely on unobservables (see, e.g., Feldhusen, 1995). Post-Behaviourism, many psychologists have explained away the incubation effect, claiming that the value of a break lies in enabling recovery from fatigue, or relief from functional fixedness (Posner, 1973). Explanations formulated in terms of recovery from fatigue state that a break relieves neural fatigue within an individual, providing them with renewed energy to attempt a solution. Proponents of functional fixedness claim that a rest allows individuals to break their ‗mindset‘, to restructure the problem and approach it differently, either by ―reducing the ‗recency‘ value of inappropriate strategies‖, or by selectively forgetting less useful memories in favour of more relevant ones (Ochse, 1990, p.198). Several studies involving convergent tasks have failed to find support for an incubation effect (e.g., Dominowski, 1972; Dominowski & Jenrick, 1972; Olton & Johnson, 1976; Olton, 1979). However, implicit in those studies has been the view that incubation can only be evidenced by the achievement of a correct solution. Nevertheless, it may be argued that any ideas produced after conscious work on a problem has ceased could be considered as indicative of the presence of an incubation effect.

THE INCUBATION EFFECT AND NON-CONSCIOUS PROCESSING Despite the mixed nature of the results reported above, theories of creativity that rest on nonconscious processing, and on Wallas‘ model, are still tenable. There exists both anecdotal and empirical evidence for the existence of an incubation effect, and it is a phenomenon in need of explanation. Neither is incubation as intractable to the scientific method as is commonly held. As noted by Fulgosi and Guilford (1968), the employment of a divergent thinking task makes it possible to inform people that they have completed a problem-solving task, introduce a distracting break, and then resume testing, which is a promising

86


methodology for investigating the incubation effect. Presumably, that particular experimental manipulation precludes, to a large extent, the active seeking of solutions to an unresolved problem. Moreover, nonconscious processing need not be disregarded on theoretical grounds. Dorfman, et al. (1996) proposed that implicit memory provides a means by which to reconceptualise nonconscious processes, making possible the readmission of the study of incubation to a scientific psychology. Holyoak (1995) pointed out that the existence of neural networks provides a means of grounding theories of nonconscious aspects of problem solving. Elaborating on that proposition, Gabora (2002) provided a neural network-inspired model of creativity. She based her account on the premises that memory is sparse and distributed, in that each item is stored across many locations, and each location helps to store many items. Additionally, memory is content-addressable, insofar as each memory is characterised by a specific pattern of excitation and inhibition across neurons. Associated memories share activation in common. Individuals differ in the richness of their associations, or the extent to which their memories overlap. According to Gabora (2002), those with greater associative richness have access to a larger, more varied number of responses in creativity tasks.

INDIVIDUAL DIFFERENCES Differential psychologists have been less averse to studying creativity than have modern cognitive psychologists. According to Ochse (1990), Guilford articulated the most influential factor-analytical account of creativity and problem solving, basing his theory on the existence of distinct cognitive abilities, of which creativity is one. Carroll (1993) examined ‗fluency‘ and ‗creativity‘ factors, particularly those associated with what Guilford (1967) referred to as ‗divergent production‘. In his reanalysis of some 477 data sets, Carroll (1993) identified an Ideational Fluency factor, which included ―tasks in which the subject was required to produce, within a fairly liberal time limit, a series of different words or phrases concerned with a specified topic or concept‖ (Carroll, 1993, p.396), which is identical to French, Ekstrom and Price‘s (1963) construct of the same name.

1. [Title]

2. [Title]

1.Pencil

2. Road

Figure 1. (a) Example of the blank Visual Creativity Task & (b) two solutions.


87

The study reported in this chapter investigates the incubation effect from the perspectives of both individual differences and experimental cognitive psychology. This makes possible the examination of whether idea generation is a cognitive ability upon which people differ, which may contribute to successful life-outcomes, and affords a greater appreciation of what incubation might involve. Unlike traditional tests of Ideational Fluency, the tests used included a distracting break halfway through. That made it possible to establish whether a break increases response output when, to the participant‘s surprise, the task is resumed. To the extent that a break enhances performance, it may be that nonconscious processing or incubation is occurring. Numerous past accounts, both anecdotal and scientific, have indicated that taking a break (Dijksterhuis, Bos, Nordgren & van Barren, 2006; Einstein, 1949a; Fulgosi & Guilford, 1968; Gregory, 2004; Hadamard, 1949; Mazzarello, 2000; Poincare, 1913; Snyder, et al., 2004b)— or literally ―sleeping on it‖ (Cohen, Pascual-Leone, Press & Robertson, 2005; Fenn, Nusbaum & Margoliash, 2003; Stickgold, Malia, Maguire, Roddenberry & O‘Connor, 2000; Wagner, Gais, Haider, Vergler & Born, 2004)—can enhance idea production. Commonly, a burst of new ideas follows a break, but never has the number of ideas produced in the post-break session been reported as greater than in the pre-break period. Recently, there has been an upsurge of interest in the, apparently nonconscious, phenomenon of additional solutions appearing after a break from problem solving (see, e.g., Dijksterhuis, et al., 2006; Snyder et al., 2004b). Whilst most reports have considered verbal or mathematical problems, the research we report here concentrates on comparing verbal and visual creativity. We wondered whether a break would differentially advantage creative idea production in the visual over the verbal domain. Wallas‘s (1926) account singled out incubation as essential to the creative process and to problem solving, stating that the benefit of breaks results from nonconscious production of ideas. Whilst this has some empirical support, others have sought to explain the value of a break in more prosaic terms, contending that the benefit lies in enabling recovery from neural fatigue, or relief from functional fixedness (Posner, 1973). We investigated the incubation effect from the perspective of nonconscious idea generation. Most investigations have utilised convergent tasks, with a single correct solution (e.g., Dijksterhuis et al., 2006; Fenn et al., 2003; Hadamard, 1949). However, we contend that convergent problem solving is not necessarily equivalent to creativity. For this reason, our method used tasks with multiple solutions in the verbal and visual domains, which we believe to be superior markers of creativity. Notably, no previous accounts have compared an advantage of verbal versus visual creativity following a break. We did so, with the aim of examining whether or not there is a difference between modalities. First year psychology students from the University of Sydney (N=107, 32 (27%) males and 75 (64%) females), with an average age of 19.79 years (range 18-49, S.D. 4.02), participated in the study. Participants‘ first language may affect performance on verbal tasks and this information was obtained: 86% of people stating that English was their first language. In the outcome, this did not appear to affect scores on the verbal tasks and was thus not considered further.

88


MATERIALS AND PROCEDURE Apart from the Verbal Creativity Test (Snyder et al., 2004b), and the Visual Creativity Test (Pallier, 2005), we administered standardised psychometric tests (derived from within the Cattell-Horn-Carroll framework) of fluid intelligence (Gf), crystallised intelligence (Gc) and visualisation ability (Gv). To help clarify the research protocol the task structure is presented below. Pre-break Verbal Creativity Test: this task was based on Cattell‘s Uses Test (French, Ekstrom & Price, 1963), and required participants to list as many possible different uses for a piece of paper, (paper was chosen as it is presumably equally familiar to all and, thus, required no specialist knowledge, a factor that may account for differences in performance). The Test was administered on computer and ceased automatically when no response was provided for 30 seconds, after which participants were told (by the computer) that the task was over. The 30 second time lag was taken as being indicative of participants being unable to rapidly produce new ideas. Pre-break Visual Creativity Test: the experimenter instructed participants to construct as many varied drawings as they could utilising a series of parallel lines, presented in paper-and-pencil format, adapted from Torrance‘s (Torrance, 1966) test battery. Examples of stimuli and outcomes are presented in Figure 1, below. When a participant had ceased producing drawings for 30 seconds, the experimenter announced that the Test had ended and to begin the computerised cognitive ability tasks. Participants were then given six tests of intelligence, which included an equivalent number of visual and verbal tasks. This was because relief from neural fatigue has been suggested as an alternative to nonconscious idea production (Posner, 1973). Having participants complete verbal and visual intelligence tests during the break maximised the possibility of detecting a genuine incubation effect, rather than a beneficial relief from neural fatigue in either modality. Matrices (Gf): in this Test (Stankov & Roberts, 2001), participants were presented with a series of 3x3 matrices, with all but the bottom right-hand block containing patterns made up of o‘s, +‘s, and blanks. Participants were required to identify the logical series, and to fill in the empty square with the appropriate selection of o‘s, +‘s, or blanks that logically completed the matrix pattern. The task involved two practice and ten test items. Letter Swaps (Gf): this Test (Stankov & Crawford, 1993) presented participants with three letters (J, K and L) in various orders, and required them to carry out (mentally) a number of letter swaps (e.g., ‗Swap the 1st and 3rd letters‘) and identify the resulting order of letters. There were two practice and eight test items, employing either 2, 3, 4, or 5 ‗Swaps‘ operations (i.e., 2 presentations of each difficulty level). Vocabulary (Gc): in this Test (Stankov, 1997), participants were asked to select the most appropriate definition (from four alternatives) for a series of words. The Test comprised two practice and ten test items. Paper Folding Test (Gv): this Test measured the primary mental ability of visualisation (French et al., 1963). It consisted of two rows of figures. The top row represented a


89

square piece of paper being folded, with the final figure indicating where holes had been punched into the paper. The bottom row presented five alternative representations of how the paper would look when unfolded. The task was to pick the correct solution. There were two practise and ten test items. Proverb Matching (Gc): this Test was developed by Stankov and Roberts (2001). Each item showed a typical proverb and a selection of five alternative ―proverbs‖, one of which had a very similar meaning to the test item. The requirement was to identify that ―proverb‖. There were two examples and ten test items. Hidden Figures (Gv): another task developed by Stankov and Roberts (2001), this Test measures the ability to identify a simple figure hidden amongst a complex pattern. Participants were presented with a set of five alternative shapes and a complex set of intersecting lines within a rectangle. The task requirement was to identify which of the alternatives was ―hidden‖ within the rectangle. There was one example and six test items. Post-break Verbal Creativity Test: this was identical to the first trial. Post-break Visual Creativity Test: this was also identical to the pre-break trail. Creativity tests may be evaluated in several ways (see, e.g., Torrance, 1990; Snyder, Mitchell, Bossomaier & Pallier, 2004a; Snyder et al., 2004b), the most germane here being the number of ideas produced (technically known as Fluency). The questions of interest were ―after running out of ideas, does a break induce extra solutions?‖ and ―does this break benefit differ between verbal and visual tasks?‖ (Note that ―breaks‖ in past research have generally been of longer duration, often literally involving a period of sleep; Dijksterhuis et al., 2006; Fenn et al., 2003; Mazzarello, 2005; Wagner et al., 2004).

DISCUSSION OF FINDINGS The scores on the verbal task were as expected from previous studies using that test (e.g., Snyder et al., 2004a, b)—people can immediately produce some extra solutions, even when they seemed to have run of ideas before the break. However, the outcome of the visual creativity test was unexpected—not only did people produce extra ideas after the break but, unlike the verbal test, they produced more ideas than in the first session. Overall, scores for intelligence exhibited means and standard deviations similar to those reported elsewhere using the same (or very similar) tests and participant populations (e.g., Danthiir, Roberts, Pallier & Stankov, 2001; Pallier et al., 2002). Thus, the current sample did not appear to be unusual. To examine the relationship between the creativity and cognitive ability test scores, an exploratory factor analysis was conducted. A parsimonious three factor solution was preferred as it best indicates the aims of the study. The factors were interpreted as follows: Factor 1: clearly represented an ―intelligence‖ or cognitive ability factor. Factor 2: was defined solely by fluency on the visual creativity task and was interpreted as visual creativity. Factor 3: was defined solely by fluency on the verbal creativity task and was interpreted as verbal creativity. The inter-correlations between the factors showed a virtually zero relationship between the

90


factors (i.e., they represent quite distinct abilities). This outcome supports the view that creativity is domain dependent. The difference in break benefit for verbal versus visual creativity cannot, in this case, be easily explained in terms of relief from neural fatigue as the participants worked on cognitively demanding tasks of a similar nature for roughly an hour between creativity sessions. Furthermore, it is unlikely that there is a differential effect of neural fatigue for visual and verbal abilities, particularly since equally demanding tasks were presented in both modalities. Neither is conscious processing of additional solutions a likely explanation because: 1) participants had no reason to believe that they would be asked to produce more ideas at a later time, in fact they were clearly told that the task was over, and; 2) people were asked at the end of the session if they had been thinking about the creativity tasks during the intelligence tests, and they reported that they had not. In fact, they were surprised to be asked to produce more responses. Therefore, we suggest that these results point to nonconscious idea generation taking place in both tasks, but in the visual task this leads to more creative ideas after the break than before it. This is the first time that such a combination of tests has been used in conjunction with the investigation of a break benefit to the creative process. Furthermore, whilst a number of studies have looked at ―sleeping on it‖ in a literal sense (Cohen et al., 2005; Fenn et al., 2003; Stickgold et al., 2000; Wagner et al., 2004)—i.e., being re-tested after an overnight break—our investigation contributes to the relatively sparse knowledge regarding briefer interludes. What essentially differentiates our research from previous studies is the shorter break time employed, alongside our choice of both visual and verbal tasks. We hypothesise that nonconscious processing occurs continuously, independent of the type of break or its duration.

CONCLUSION We can only speculate on the cause of our unexpected outcome. Perhaps what we are seeing is a division of thought processes involving the two modalities (visual and verbal), providing further evidence that the Whorfian hypothesis in its extreme form (Whorf, 1956) is overstated. Not all thought appears dependent on language. Anecdotal accounts, perhaps best thought of as case studies, of the differentiation between verbal and visual processing abound. For example, Einstein stated that his thinking processes were mainly visual ―The words or the language, as they are written or spoken, do not seem to play any role in my mechanism of thought… (the) elements are, in my case, of (the) visual… type.‖ (1949, pp. 142-143). Researchers are yet to fully explore possible differences in the neural underpinnings of incubation for artistic versus verbal creativity. Recently, however, neuropsychological investigations into creative insight have indicated the involvement of the right anterior superior temporal gyrus (Bowden, Jung-Beeman, Flech & Kounios, 2005; Jung-Beeman et al., 2004). There is also an increase in alpha-band followed by gamma-band activity accompanying insightful solutions for verbal tasks (without a break) (Bowden et al., 2005; Jung-Beeman et al., 2004). A fruitful line of research would be to examine the extent to which these purported neural correlates apply during and after breaks and for both verbal and artistic creativity tasks.


91

Even though there is a break benefit to creativity in both modalities, the size of the effect appears reliant on whether verbal or visual creativity is involved, strongly favouring the latter. The results reported here suggest that, whilst meeting deadlines is important, performance could well be enhanced by taking time out during the decision-making processes, particularly for tasks with a visual component.

REFERENCES Bowden, E. M., Jung-Beeman, M., Fleck, J., & Kounios, J. (2005). New approaches to demystifying insight. Trends in Cognitive Science, 9, 322-328. Cohen, D. A., Pascual-Leone, A., Press, D. Z., & Robertson, E. M. (2005). Off-line learning of motor skill memory: A double dissociation of goal and movement. Proceedings of the National Academy of Science, USA, 102, 18237-18241. Danthiir, V., Roberts, R. D., Pallier, G., & Stankov, L. (2001). What the nose knows: Olfaction and cognitive abilities. Intelligence, 29, 337-361. Dijksterhuis, A. P., Bos, M. W., Nordgren, L. F., & van Baaren, R. B. (2006). On making the right choice: The deliberation-without-attention effect. Science, 311, 1005-1007. Einstein, A. (1949). A testimonial from Professor Einstein. In J. Hadamard, (Ed.), An essay on the psychology of invention in the mathematical field (pp. 142-143). Princeton, NJ: Princeton University Press. Fenn, K. M., Nusbaum, H. C., & Margoliash, D. (2003). Consolidation during sleep of perceptual learning of spoken language. Nature, 425, 614-616. French, J. W., Ekstrom, R. B., & Price, L. A. (1963). Manual for kit of reference tests for cognitive ability. Princeton, NJ: Educational Testing Service. Fulgosi, A., & Guilford, J. P. (1968). Short-term incubation in divergent production. American Journal of Psychology, 81, 241-248. Gregory, R. L. (2004). The Oxford companion to the mind. Oxford, UK: Oxford University Press. Hadamard, J. (1949). An essay on the psychology of invention in the mathematical field. Princeton, NJ: Princeton University Press. Jung-Beeman, M., Bowden, E. M., Haberman, J., Frymiare, J. L., Arambel-Liu, S., Greenblat, R., Reber, P. J., & Kounios, J. (2004). Neural activity when people solve verbal problems with insight. Public Library of Science, Biology, 2, 500-510. Mazzarello, P. (2000).What dreams may come? The scientific benefits of eating cheese before bedtime. Nature, 408, 523. Pallier, G. (2005). A test of artistic creativity. Unpublished test, available from the School of Psychology, University of Sydney. Pallier, G., Wilkinson, R., Danthiir, V., Kleitman, S., Knezevic, G., Stankov. L., & Roberts, R. D. (2002). The role of individual differences in the realism of confidence judgements. Journal of General Psychology, 129, 257-299. Poincare, H. (1913). Mathematical creation. In, G. B. Halsted (Trans.), The foundations of science (pp. 383-394). New York, NY: Science Press. Posner, M. I. (1973). Cognition. Glenview, IL: Scott Foresman.

92


Snyder, A., Mitchell, J., Bossomaier, T., & Pallier, G. (2004a). The creativity quotient: An objective scoring of ideational fluency. Creativity Research Journal, 16, 415-420. Snyder, A., Mitchell, J., Ellwood, S., Yates, A., & Pallier, G. (2004b). Nonconscious idea generation. Psychological Reports, 94, 1325-1330. Stankov, L., & Crawford, J. (1993). Ingredients of complexity in fluid intelligence. Learning & Individual Differences,5, 73-111. Stankov, L., & Roberts, R. D. (2001). Matrices Test. Unpublished cognitive test, available from the School of Psychology, University of Sydney. Stankov, L. (1997). The Gf/Gc Quickie Battery. Unpublished test battery, available from the School of Psychology, University of Sydney. Stickgold, R., Malia, A., Maguire, D., Roddenberry, D., & O‘Connor, M. (2000). Replaying the game: Hypnagogic images in normals and amnesics. Science, 290, 350-353. Torrance, E. P. (1990). Torrance tests of creative thinking norms-Technical manual figural (Streamlined) Forms A and B. Bensenville, IL: Scholastic Testing Service. Torrance, E. P. (1966).Thinking creatively with pictures (Booklet A). Lexington, MA: Personnel Press. Wagner, U., Gais, S., H. Haider, Verleger, R., & Born, J. (2004). Sleep inspires insight. Nature, 427, 352-355. Wallas, G. (1926). The Art of Thought. New York, NY: Harcourt Brace. Whorf, B. (1956). Language, Thought, and Reality. Cambridge, MA: MIT Press.



Chapter 8

THE ROLE OF PERSONALITY IN LEARNING PROCESSES AND LEARNING OUTCOMES IN APPLIED SETTINGS Paul Ginns, Gregory Arief D. Liem and Andrew J. Martin Faculty of Education & Social Work, University of Sydney Sydney, NSW 2006, Australia

ABSTRACT Researchers and practitioners working in applied settings, such as schools, workplaces and physical activity/exercise contexts, have identified a range of factors related to learning processes and learning outcomes. In this chapter, we focus on personality factors associated with learning processes and outcomes in each of these settings, drawing on meta-analytic reviews of widely used explanatory frameworks (e.g., the Big Five) and more recent theoretical and empirical developments in applied personality research (e.g., the HEXACO model). We discuss major findings in these areas and the implications of these findings for personality and psycho-educational researchers seeking to more fully account for important learning processes and outcomes in applied settings.

Keywords: Big Five, Personality, Learning processes & outcomes, Traits

INTRODUCTION Learning can be reflected in changes in an individual‘s factual, conceptual, procedural, inter-personal, or meta-cognitive knowledge base (Anderson, et al., 2001) as indicated by performance, achievement, learning-relevant self perceptions (e.g., academic resilience, self


94

Paul Ginns, Gregory Arief D. Liem and Andrew J. Martin

efficacy), and processes that lead to learning. Even more broadly, engagement or disengagement in learning-related processes or settings (e.g., school or workplace retention; adherence to an exercise or diet regime) may also be considered valid indicators of learning. For many years now, researchers and theorists have sought to identify factors relevant to and predictive of these many learning indices. Although individuals vary across many dimensions associated with the process of learning, including prior knowledge and general mental ability (Jonassen & Grabowski, 1993), over the past three decades some consensus has emerged regarding the role of personality factors and the specific facets that comprise them. In this chapter, we review the Big Five Model (McCrae & Costa, 1997) and consider how the Big Five factors relate to learning processes and outcomes (as defined above) across applied settings in education, work, and physical activity/exercise. Because of the voluminous research literature regarding personality in applied settings, we emphasise integrative metaanalytic findings to describe these relationships. In particular, we focus on the constructs of Conscientiousness and Neuroticism (and their component facets), as these constructs appear to play the clearest role across the applied settings examined. Following consideration of evidence for the relations between particular personality traits and learning processes and outcomes, we consider the relative utility of interventions targeted at changing broad personality traits (e.g., Conscientiousness) and those seeking to target related behavioural indicators or processes (e.g., goal-setting, self-control). Lastly, we consider a more recently proposed six-factor model (HEXACO; Ashton & Lee, 2007) that contends an additional factor, Honesty/Humility, which may also impact on a variety of learning processes and outcomes in the settings considered here.

THE BIG FIVE MODEL The Big Five Model (McCrae & Costa, 1997) posits that five broad personality factors— Openness to Experience, Conscientiousness, Extraversion, Agreeableness, and Neuroticism— underlie most of the personality trait descriptors present in any given consideration of the issue. McCrae and Costa note that such traits are abstractions which researchers and theorists infer from observing patterns of behaviour, rather than being directly measurable. These traits may be summarised as follows: Openness to Experience. This trait captures preferences for novelty and complexity versus conventionality in an individual‘s intellectual life. It is captured in descriptors such as ―creative‖, ―intellectual‖, and ―complex‖, versus ―direct‖, ―practical‖, and ―pragmatic‖. Conscientiousness. This trait captures the urge to accomplish goals, and is characterised by the extent of an individual‘s carefulness, planning, and self-discipline. It is captured in descriptors such as ―organised‖, ―systematic‖, and ―neat‖, as compared to ―disorganised‖, ―inefficient‖, and ―untidy‖. Extraversion. This trait describes the extent to which an individual is interested in people and events in the external world, as opposed to preferring one‘s own company and being more focused on one‘s own ―inner life‖. Commonly used descriptors of

The Role of Personality in Learning Processes and Learning Outcomes…

95

extraverts are ―outgoing‖, ―energetic‖ and ―talkative‖, compared to introverts as ―shy‖, ―reserved‖, or ―quiet‖. Agreeableness. This trait captures the degree to which someone feels a need to be part of a larger community, and focuses strongly on interpersonal relationships. It is captured in descriptors such as ―sympathetic‖, ―warm‖, and ―cooperative‖, as compared to ―inconsiderate‖, ―harsh‖, and ―rude‖. Neuroticism. This trait (also referred to as Emotional Stability) describes the character of commonly felt emotions in an individual‘s life. Individuals scoring highly on this trait are typically described as ―anxious‖, ―tense‖, and ―moody‖, while more emotionally stable individuals are often described as ―calm‖, ―poised‖, and ―unenvious‖. As indicated above, of these five personality traits, Conscientiousness and Neuroticism hold particular appeal in relation to learning in school, work, and physical activity/exercise. Discussing these traits in relation to work performance, for example, Barrick, Mount, and Judge (2001) argue that ―…it is hard to conceive of a job where it is beneficial to be careless, irresponsible, lazy, impulsive, and low in achievement striving (low conscientiousness)…similarly, being anxious, hostile, personally insecure and depressed (low emotional stability) is unlikely to lead to high performance in any job‖ (p. 11). Given that success in work and learning settings requires sustained, focused effort closely aligned with Conscientiousness and Neuroticism, we argue that these factors can be expected to hold important explanatory power across a variety of learning settings. Indeed, we review evidence for this assertion below.

PERSONALITY AND LEARNING IN EDUCATIONAL INSTITUTIONS Poropat (2009) used meta-analysis to determine average correlations between Big Five dimensions and academic achievement. Primary studies included eight studies from primary education settings, 24 from secondary settings, and 75 from tertiary settings. Poropat found that the sample-weighted meta-analytic correlations, corrected for reliability, varied substantially by academic level for all dimensions except Conscientiousness. These results are summarised in Table 1. Table 1. Meta-analytic correlations of Big Five personality variables with academic achievement (Poropat, 2009) Academic Level Primary Secondary Tertiary Total sample

Openness to Experience .24 .12 .07 .12

Conscientiousness .28 .21 .23 .24

Extraversion

Agreeableness

Neuroticism

.18 -.03 -.01 .00

.30 .05 .05 .07

.20 .01 -.01 .02

96


The above results indicate, with the exception of Conscientiousness, that the relations between Big Five personality dimensions and academic achievement diminish with academic level, often to negligible amounts. In contrast, the magnitude of the relationship between Conscientiousness and academic achievement remained relatively stable across academic levels. In reviewing these results, Poropat noted that the increasing variability in curriculum as students progress through academic levels, and suggested that this factor might cloud the relationship between overall correlations between personality and achievement. More finegrained matching of specific personality dimensions and their facets with specific academic criteria, including changes over time, may therefore be warranted. We explore several such relationships below using contemporary studies. Lievens, Ones, and Dilchert (2009) investigated the extent to which the criterion-related validities of Big Five personality scales change over the course of study at university/college (in this case, in medicine). They noted that earlier subjects in medical education tended to focus on declarative and procedural knowledge acquisition, whereas later subjects emphasised application and internship performance, with a particular reliance on interpersonal interactions. Given these changes in emphasis, they hypothesised that the pattern of predictive validities might change over time. Using the entire 1997 cohort of students enrolled in Belgian medical degrees, over a seven-year time-frame, Lievens et al. found the validity of Conscientiousness increased from .18 to .45; the validity of Extraversion changed from -.11 to .31; and the validity of Openness changed from .02 to .30. While the above results for Conscientiousness are consistent with Poropat‘s (2009) meta-analysis for tertiary education settings, the results for Extraversion and Openness stand in marked contrast. This set of results, while again highlighting the predominant role of Conscientiousness, raises the potential that additional personality traits should be considered in professional education when the focus of skills under development changes over time.

NEUROTICISM AND ANXIETY IN EDUCATIONAL INSTITUTIONS As noted earlier, each of the Big Five traits is comprised of a number of facets which may be more or less predictive of particular learning outcomes and indicators Anxiety is an example of a specific facet of the Neuroticism dimension with clearly demonstrated implications for student learning. In educational settings, many students experience test anxiety, ―a combination of cognitive and physical responses that are aroused in testing situations or in similar situations in which a person believes that he or she is being personally evaluated‖ (Cizek & Burg, 2006, p. 1). A large body of research has identified a range of negative effects of (test) Anxiety on learning processes and outcomes, including diminished effort during tests, negative self-evaluations, inappropriate test behaviours, reduced motivation for learning, and leaving school (Cizek & Burg, 2006; Hembree, 1988). More recent research has examined the relations of Anxiety to emerging educational constructs such as academic resilience and academic buoyancy, two other factors embedded in the learning process (Martin & Marsh, 2006, 2008). Martin and Marsh (2008) define academic buoyancy as ―students‘ ability to successfully deal with academic setbacks and challenges that are typical of the ordinary course of school life (e.g., poor grades, competing deadlines, exam pressure, difficult schoolwork)‖ (p. 53). In an initial investigation of the


97

construct, Martin and Marsh (2006) used a construct validation approach to examine hypothesised adaptive and maladaptive predictors of this construct, and educationally salient outcomes such as school enjoyment, class participation, and general self-esteem. Selfefficacy, control, planning, low anxiety, and persistence were found to predict academic resilience, with Anxiety (e.g., ―When exams and assignments are coming up, I worry a lot‖) being the strongest negative predictor (r = -.66; partial r = -.60). Martin and Marsh (2008) used longitudinal modelling to examine the causal salience of the predictors described above at Time 2, holding the effects of these predictors and academic buoyancy as measured at Time 1 constant. Consistent with the results of Martin and Marsh (2006), by far the most important predictor of buoyancy at Time 2 was Anxiety, with this variable largely mediating the observed gender difference in buoyancy. Taken together, these findings underscored the salience of facets within Neuroticism (in this case, Anxiety) on the academic learning landscape.

PERSONALITY AND LEARNING IN WORK SETTINGS Whereas mass education (e.g., compulsory attendance of school until at least the age of 15) is a relatively recent phenomenon in the history of human development, workplaces have been a much more longstanding feature of human endeavour. As knowledge, skills and attitudes can provide competitive advantage for organisations, workplace and organisational psychologists have sought to identify individual differences, including personality variables that predict how easily these knowledge, skills, and attitudes might be developed. Colquitt, LePine, and Noe (2000) used meta-analytic path modelling to review the research literature on training motivation, its personological and situational antecedents, and the relationship of these variables with learning outcomes. Personality variables including Conscientiousness and Anxiety (as discussed above, a facet of Neuroticism) were examined for their relations to hypothesised intervening mechanisms (e.g., pre-training self-efficacy; the valence of training outcomes; or job/career variables such as job involvement) as well as motivation to learn, learning outcomes (e.g., declarative knowledge, skill acquisition, posttraining self-efficacy), post-training transfer, and job performance. Hierarchical meta-analytic path analyses, holding antecedent variables constant, revealed variable relationships between personality factors and learning processes and outcomes. Anxiety, but not Conscientiousness, was related to motivation to learn. Across the learning outcomes, Conscientiousness was related to declarative knowledge development, skill acquisition, post-training self-efficacy, and training reactions, while Anxiety was related to declarative knowledge development, skill acquisition, and post-training self-efficacy, but not training reactions. Conscientiousness and Anxiety were also related to training transfer back to the workplace, and ultimately to job performance. Consistent with the results for educational settings described above, these results support arguments for the importance of personality (viz. Conscientiousness and Anxiety) in understanding learning at work. While the above meta-analytic results will be applicable across many organisational settings, idiosyncrasies of particular jobs and/or workplaces may mean other personality traits are also important in some settings. For example, Campbell, Castaneda, and Pulos (2010) performed a meta-analysis of studies using Big Five personality measures to predict military

98


aviation training success. The meta-analysis focused on measures of Big Five constructs (Neuroticism and Extroversion) and component facets (Anxiety). The strongest corrected correlation with training success was for Neuroticism (r = -.25), then Extroversion (r = .17), then Anxiety (-.14). Of particular note for these authors was the similarity in predictive validity of Neuroticism and Anxiety, leading them to conclude that anxiety might be the key facet accounting for the observed association between Neuroticism and training performance.

PERSONALITY AND LEARNING IN PHYSICAL ACTIVITY AND EXERCISE The bulk of research relating personality to physical behaviour and performance has focused on whether, for example, athletes differ from non-athletes, or elite versus non-elite athletes, on key constructs such as Extraversion, Neuroticism, or Tough-mindedness (e.g., Clough, Earle, & Sewell, 2002; Piedmont, Hill, & Blanco, 1999). An additional and relatively understudied line of research has focused on associations between personality factors and physical activity and exercise (e.g., Van den Auweele, De Cuyper, Van Mele, and Rzewnicki, 1993). Physical activity and exercise are a foundation for healthy lifestyles, physical health, and psychological wellbeing (Martin et al., in press). A compelling question, then, is the connection between personality and learning processes, indicators or outcomes in the physical activity and exercise domain. In relation to physical activity, Rhodes and Smith‘s (2006) meta-analysis suggested that individuals higher in Extraversion (r =.23) and Conscientiousness (r = .20) and lower in Neuroticism (r = -.11) were more likely to engage in a physically active lifestyle. Conducting a meta-analysis of the links between Conscientiousness and health-related behaviours, Bogg and Roberts (2004) found that Conscientiousness was a positive correlate of physical activity and a negative correlate of risky health-related behaviours such as unhealthy eating. Among Conscientiousness facets, Industriousness and Traditionalism were the strongest predictors of physical activity and unhealthy eating. Interestingly, some of these findings were moderated by age such that, for example, Conscientiousness significantly predicted physical activity for younger participants (< 30 years), but not so much for older ones (> 30 years). Thus, consistent with findings in relation to learning indicators in education and work, learning indicators and outcomes in the physical domain (e.g., activity levels, diet adherence) are associated with Conscientiousness. More recent research has centered on social cognitive factors, such as motivational beliefs, in mediating and moderating the links between personality traits and health-related processes and outcomes. For example, Hoyts, Rhodes, Hausenblas, and Giacobbi (2009) showed that activity and self-discipline (facets of Extraversion and Conscientiousness, respectively) were significant positive predictors of affective and instrumental aspects of attitudes towards exercise and perceived behavioural control in exercising. Interestingly, Hoyts et al. found a moderating effect suggesting that the relationship between exerciserelated intention and behaviour was stronger for less anxious individuals. In another study, Bogg, Voss, Wood, and Roberts (2008) investigated the associations between values, personal goals, personality traits and health-related behaviours and outcomes. The results showed that, over and above the independent effects of values and personal goals, domaingeneral Extraversion predicted domain-specific physical activity Extraversion which in turn


99

predicted physical fitness. Taken together, these findings demonstrated that Extraversion and Conscientiousness are directly or indirectly predictive of learning outcomes and indicators in the physical activity and exercise domain.

PERSONALITY AND LEARNING: FOCI FOR INTERVENTIONS The associations described above between personality constructs and learning indicators/outcomes across applied settings raise questions about where interventions should be focused. One such question relates to whether it is meaningful and/or desirable to aim to change an individual‘s personality in order to improve learning processes and outcomes, or whether the focus should be placed on (for instance) developing skills for better test anxiety management (cf. Neuroticism) or goal setting (cf. Conscientiousness). While there is substantial evidence for the relative stability of personality over time (e.g., Costa & McCrae, 1992), there is also evidence that intensive psychotherapy may modify personality traits. To the extent that this is the case, such interventions can also impact upon learning outcomes and indicators influenced by the relevant personality trait. Jorm (1989), for example, meta-analysed 63 studies comparing psychotherapy groups with no-treatment controls, where measures of trait Anxiety or Neuroticism were used as a dependent variable. Therapies included psychotherapies (e.g., rational-emotive therapy, group counselling) and skills-based therapies (e.g., muscular relaxation, study skills training). The mean effect across all types of therapy was moderate (d = 0.52), with rational-emotive therapy in particular producing a substantially larger mean effect (d = 1.25). Based on these results, then, it is clearly possible for shifts in traits to take place. At the same time, however, the resource-intensiveness of psychotherapy may preclude its use in many applied settings. In such cases, it may be more realistic to focus on developing specific skills which can be embedded in applied settings (e.g., Hembree, 1988). Similarly, rather than changing the trait of Conscientiousness per se, educators may help students set realistic, achievable goals (Locke & Latham, 2006) that are cognitive-behavioural facets of Conscientiousness. Hence, this is a strategy which addresses a variety of cognitivebehavioural indicators of Conscientiousness (e.g., planning, organisation), and which has been demonstrated to enhance learning across diverse applied settings. For a recent example from tertiary education, see Morisano, Hirsh, Peterson, Pihl, and Shore (2010).

PERSONALITY AND LEARNING IN APPLIED SETTINGS: FUTURE DIRECTIONS FOR RESEARCH Recently, a competing model to the Big Five has emerged. The HEXACO Model developed out of lexical studies of personality structure, which found a six-factor rather than five-factor space was replicated across a variety of Indo-European and non-European languages (for a review of this research, see Ashton & Lee, 2007). While the above Big Five constructs still emerged with varying levels of clarity in these studies, an additional factor, termed Honesty/Humility, was also found. This dimension is captured by terms such as

100


sincere, honest, modest, and fair-minded versus sly, greedy, pretentious, and hypocritical, and appears to capture ―moral conscience‖ (Marcus, Lee, & Ashton, 2007, p. 4). To date, very few studies have examined the relations of Honesty/Humility to learning processes and outcomes, but the nature of the construct suggests an explanatory role in academic misconduct that represents maladaptive orientations to the learning process. Marcus et al. (2007), for example, examined relations between measures of integrity, the HEXACO dimensions, and self-reports of counter-productive academic behaviours (e.g., cheating, low personal standards, duplicity, and misrepresentation) by German and Canadian university students. Using hierarchical multiple regression, they found Honesty/Humility self-reports predicted additional variance in counter-productive behaviours over and above demographic variables, self-reports of integrity, and Big Five constructs. Future research on cheating in academic settings may benefit from simultaneous considerations of personological and situational factors, in order to determine their relative salience.

CONCLUSION Advances in research and theorising regarding the key dimensions of personality over the past three decades have provided a better understanding of the dynamics of learning in applied settings. In particular, major meta-analytic reviews of the educational and organisational training research literatures have concluded that conscientious learners are more likely to engage in and succeed at learning. In contrast, students with higher levels of Anxiety (a facet of the broad personality factor of Neuroticism) will face greater learning challenges than less anxious students. While the research literature on learning in physical activity and exercise contexts is not as comprehensive as that conducted in the other two settings, similar conclusions can be reached if learning indicators such as physical activity, physical fitness, and healthy eating are considered. Across all three applied domains, the impact of Honesty/Humility on learning processes and outcomes is currently poorly understood, but we contend its role in maladaptive learning-related behaviours is likely to be considerable and deserving of further study. Taken together, the findings hold substantive and methodological implications for researchers studying issues relevant to personality and learning in diverse applied settings, and are also relevant to practitioners seeking to enhance learning outcomes that rely in large part on core personality traits and their component facets.

REFERENCES Anderson, L. W., Krathwohl, D. R., Airasian, P. W., Cruikshank, K. A., Mayer, R. E., Pintrich, P. R., Raths, J., & Wittrock, M. C. (2001). A taxonomy for learning, teaching, and assessing: A revision of Bloom’s taxonomy of educational objectives. New York: Longman. Ashton, M. C., & Lee, K. (2007). Empirical, theoretical, and practical advantages of the HEXACO model of personality structure. Personality and Social Psychology Review, 11, 150-166.


101

Barrick, M. R., Mount, M. K., & Judge, T. A. (2001). Personality and performance at the beginning of the new millennium: What do we know and where do we go next? International Journal of Selection and Assessment, 9, 9-30. Bogg, T. & Roberts, B. W. (2004). Conscientiousness and health-related behaviors: A metaanalysis of the leading behavioral contributors to mortality. Psychological Bulletin, 130, 887-919. Bogg, T., Voss, M. W., Wood, D., & Roberts, B. W. (2008). A hierarchical investigation of personality and behavior: examining neo-socioanalytic models of health-related outcomes. Journal of Research in Personality, 42, 182-207. Campbell, J., Castaneda, M. & Pulos, S. (2010). Meta-analysis of personality assessments as predictors of military aviation training success. International Journal of Aviation Psychology, 20, 92-109. Cizek, G. J., & Burg, S. S. (2006). Addressing test anxiety in a high-stakes environment: Strategies for classrooms and schools. Thousand Oaks, CA: Corwin Press. Clough, P. J., Earle, K., & Sewell, D. (2002). Mental toughness: The concept and its measurement. In I. Cockerill (Ed.), Solutions in sport psychology (pp. 32-43). London: Thomson. Colquitt, J. A., LePine, J. A., & Noe, R. A. (2000). Toward an integrative theory of training motivation: A meta-analytic path analysis of 20 years of research. Journal of Applied Psychology, 85, 678-707. Costa, P. T., & McCrae, R. R. (1992). Trait psychology comes of age. In T. B. Sonderegger (Ed.), Nebraska Symposium on Motivation: Psychology and aging (pp. 169-204). University of Nebraska Press, Lincoln. Hembree, R. (1988). Correlates, causes, effects and treatment of test anxiety. Review of Educational Research, 58, 47-77. Hoyts, A. L., Rhodes, R. E., Hausenblas, H. A., & Giacobbi, P. R. (2009). Integrating fivefactor model facet-level traits with the theory of planned behavior and exercise. Psychology of Sport and Exercise, 10, 565-572. Jonassen, D. H., & Grabowski, B. L. (1993). Handbook of individual differences, learning and instruction. Hillsdale, NJ: Lawrence Erlbaum Associates. Jorm, A. F. (1989). Modifiability of trait anxiety and neuroticism: A meta-analysis of the literature. Australian and New Zealand Journal of Psychiatry, 23, 21-29. Lievens, F. Ones, D. S., & Dilchert, S. (2009). Personality scale validities increase throughout medical school. Journal of Applied Psychology, 94, 1514-1535. Locke, E. A., & Latham, G. P. (2006). New directions in goal-setting theory. Current Directions in Psychological Science, 15, 265-268. Marcus, B., Lee, K., & Ashton, M. C. (2007). Personality dimensions explaining relationships between integrity tests and counterproductive behavior: Big Five, or one in addition? Personnel Psychology, 60, 1-34. Martin, A. J., Liem., G. A. D., Coffey, L., Martinez, C., Parker, P., Marsh, H. W., & Jackson, S. (in press). What happens to physical activity behavior, motivation, self-concept, and flow after completing school? A longitudinal study. Journal of Applied Sport Psychology. Martin, A. J., & Marsh, H. W. (2006). Academic resilience and its psychological and educational correlates: A construct validity approach. Psychology in the Schools, 43, 267281.

102


Martin, A. J., & Marsh, H. W. (2008). Academic buoyancy: Towards an understanding of students' everyday academic resilience. Journal of School Psychology, 46, 53-83. McCrae, R. R., & Costa, P. T. (1997) Personality trait structure as a human universal. American Psychologist, 52, 509-516. Morisano, D., Hirsh, J. B., Peterson, J. B., Pihl, R. O., & Shore, B. M. (2010). Setting, elaborating, and reflecting on personal goals improves academic performance. Journal of Applied Psychology, 95, 255-264. Piedmont, R. P., Hill, D. C., & Blanco, S. (1999). Predicting athletic performance using the five-factor model of personality. Personality and Individual Differences, 27, 769-777. Poropat, A. E. (2009). A meta-analysis of the five-factor model of personality and academic performance. Psychological Bulletin, 135, 322-338. Rhodes, R. E., & Smith, N. E. I. (2006). Personality correlates of physical activity: A review and meta-analysis. British Journal of Sports Medicine, 40, 958-965. Van den Auweele, Y, Nys, K., Rzewnicki, R., & Van Mele, V. (2001). Personality and the athlete. In R. N. Singer, H. A. Hausenblas, & C. M. Janelle (Eds.), Handbook of sport psychology (2nd ed., pp. 239-268). Toronto: John Wiley & Sons.



Chapter 9

SOMETHING ABOUT METACOGNITION: SELF-CONFIDENCE FACTOR(S) IN SCHOOL-AGED CHILDREN Sabina Kleitman, Karina Mak, Sarah Young, Phoebe Lau, and David Livesey School of Psychology, University of Sydney, Sydney, NSW 2006, Australia

ABSTRACT Prior research into the area of metacognition has demonstrated evidence of the stable and reliable construct of self-confidence in adult populations. However, few studies have demonstrated evidence of this construct among children. The current chapter discusses the results of several studies which examine the self-confidence construct in primary school children aged 9-12, across several cognitive and physical tasks. The results from these studies demonstrated that there are stable individual differences in confidence ratings. However, there was a clear domain separation for cognitive and physical domains, as two Self-confidence factors were apparent in children. We also examined parental perceptions of their child‘s metacognition and parental confidence ratings. The results suggest that there might be an interesting difference in the strength of the relationship between the child's level of confidence and each of paternal and maternal levels of self confidence. Overall, these results have significant theoretical and practical implications, and provide directions for future research.

Keywords: Decision-making, Individual differences, Metacognition, Self-confidence, Selfregulated learning




104

Sabina Kleitman, Karina Mak, Sarah Young et al.

INTRODUCTION Metacognition refers to the executive processes involved in reflecting on one‘s own thinking: that is, ‗thinking about thinking‘ (Flavell, 1979) or ‗knowing about knowing‘ (Metcalfe & Shimamura, 1994). Most theories distinguish between two major components of metacognition—knowledge about cognition and regulation of cognition—the latter consisting of monitoring and control of cognition (Nelson & Narens, 1994; Schraw & Dennison, 1994). Monitoring of cognition, the focus of this chapter, is defined as the ability to watch, check and appraise the quality of one‘s own cognitive work in the course of doing it (Schraw & Moshman, 1995). Confidence judgments reflect these instances by deliberately evoking subjective feelings of certainty that one experiences in connection with decision-making and action-regulation (Allwood & Granhag, 1999; Stankov, 1999). Understanding the factorial structure underlying confidence judgments across cognitive and non-cognitive domains could provide a powerful window into the taxonomy of metacognitive processes. This chapter provides a broad literature review based on the studies conducted by our Individual Differences laboratory and aims to answer three novel questions: (1) Do primaryschool children display habitual general levels of self-confidence within test and across different cognitive domains? (2) Do these levels generalise across cognitive and physical domains? and; (3) What are the relationships between a child‘s levels of self-confidence with those of his/her parents?

SELF-CONFIDENCE AS AN ASPECT OF METACOGNITIVE SELF-MONITORING This work relies on a definition that captures the main purpose of self-monitoring—the ability to judge the quality of one‘s own performance in the course of doing it. In such an instance, immediately after responding to an item in a test, participants are instructed to give a confidence (or ―sureness‖) rating indicating how confident/sure he or she is that their chosen answer is the correct one. The level of confidence is expressed in terms of percentages and/or verbal statements. The starting point (the lowest confidence) on a rating scale is defined in terms of the number of alternative answers (k) given to a question (100/k). That is, in multiple-choice questions with five alternative answers, 20% is a starting point because 20% is the probability of answering the question correctly by chance. Allwood and colleagues (2006) piloted different types of confidence scales with children aged 11 and 12 years. There were no differences between these scales in calculated calibration indices or in levels of confidence, suggesting equivalence in their ability to capture confidence levels and the adequacy of this procedure with children of this age group. Subsequently, Allwood and co-workers (Allwood, Innes-Ker, Homgren, & Fredin, 2008) demonstrated that children as young as 8–9 years show comprehension of these scales. The studies overviewed in this chapter use two such scales (pictorial and line) (see Figures 1.1 and 1.2).

Something about Metacognition

105

Figure 1.1. Pictorial confidence rating scale used for the four-choice multiple-choice achievement tests items (Kleitman & Moscrop, 2010).

How sure are you that this answer is right? Absolutely unsure 20%

Absolutely sure100%

Figure 1.2. Line confidence rating scale (Allwood, et al., 2006).

The confidence rating procedure follows the cognitive act of providing a response to a typical cognitive test item, rather than relying on a general perception of one‘s own way of acting. Thus, it serves as a more accurate measure of self-confidence than the self-report items such as ―I feel self-assured‖ and ―I‘m self-confident‖ (Stankov, 1999; Kleitman, 2008). Interestingly, studies conducted on adults indicate limited or no relationships between confidence levels and personality factors which include these types of self-report questions (e.g., Extroversion). The only exception to this is the Openness to Experience dimension which shares a positive correlation of low to moderate size (rarely above .30) (see, Kleitman, 2008 for a review). There are numerous findings in relation to confidence judgments in the adult population (see Kleitman, 2008; Stankov, 1999 for reviews). Confidence judgments have high internal consistency in adults (reliability estimates are typically higher than .90) (e.g. Kleitman, 2008; Stankov, 1999) and robust test–retest estimates (Jonsson & Allwood, 2003). There is overwhelming empirical evidence showing individual differences in confidence ratings in adult populations (Kleitman, 2008; Stankov, 1999). The correlations between accuracy and confidence scores from the same test are significant (average between .40 and .50). Nevertheless, correlations between confidence ratings from a broad battery of cognitive tests (reflecting diverse cognitive abilities) have been consistently high enough to define a strong broad Self-confidence factor which reflects the habitual way in which adults assess the accuracy of their cognitive decisions. That is, adults who are more confident on one task, relative to their peers, also tend to be more confident across other tasks. Three studies conducted by our laboratory stemmed from psychological and educational traditions, using the individual differences approach to examine these findings in respect to primary-school children (aged 9-12). In this chapter, we review the results of these studies.

106


OUR EMPIRICAL FINDINGS Self-confidence Traits in Children The studies conducted in our laboratory employed a variety of cognitive and achievement tests with confidence ratings: Raven‘s Progressive Matrices (RPM; Raven, Raven, & Court, 1998); Mathematics, Vocabulary and Reading achievement tests (PAT, Australian Council for Educational Research, 2001; 2005); and those based on a school curriculum (see Kleitman & Moscrop, 2010). Results in all studies clearly indicated that reliability estimates for confidence ratings within each test were uniformly high, ranging between .84 and .96, indicating a high level of consistency in confidence assessments within the task. Employing confirmatory factor analytic techniques, Kleitman and Moscrop (2010) examined the consistency of confidence judgments across different tests. They demonstrated that when three cognitive tests with confidence ratings were utilised (Raven‘s Progressive Matrices, Vocabulary and Mathematics; N = 197, age ranged from 9 years 1 month to 12 years 11 months), two broad factors were apparent: (1) Cognitive Abilities, defined by the accuracy scores from the RPM, Vocabulary, and Mathematics tests; and (2) Self-confidence, exclusively defined by the high loadings of the confidence scores from all three cognitive tests (see Figure 2).

Vocabulary Accuracy

16.92

.44

..64

Vocabulary Confidence

Factor 1 Accuracy

9.48

.81 Mathematics Accuracy

.70

22.17

.56

.61

.69

.97

Mathematics

16.07

Confidence

Factor 2 Confidence .63

Raven‘s PM Accuracy

13.77

.16 Raven’s PM Confidence

Figure 2. Results of the Confirmatory Factor Analysis (Kleitman & Moscrop, 2010).

8.16


107

Table 1. Results of the EFA (ML, Promax Rotation), Gibson & Kleitman (2009, N = 178) Factors Gc

Confidence

Gf

h2

PAT Reading Confidence

.77

.79

PAT Vocabulary Confidence

.95

.86

PAT Maths Confidence

.45

.75

.80

PAT Reading: Accuracy

.73

.61

PAT Vocabulary: Accuracy

.78

.77

PAT Maths: Accuracy Gc: Accuracy

.80 .59

.71 .68

Gf: Accuracy

.63

.54

Factor Inter-correlations Gc Confidence

1

.42

.53

1

.41 1

Gf 2

Note: Only factorial loadings above .30 are presented; h =communalities; Gf=Fluid Intelligence; Gc=Crystallised Intelligence; PAT = Progressive Achievement Test.

Although the majority of the fit indices for this model were within the ranges that signal a good model fit (χ2/df = 3.23, GFI = .97, TLI = .92, and CFI = .97), RMSEA and its CI were greater than the desirable maximum (RMSEA = .11, .05 < 90%CI < .17). This demonstrated that the model might be problematic when generalised to a different sample. The error terms for the items within the same test were allowed to correlate. This indicated that in this study the errors were not random, and were linked to the ‗nature‘ of the test. Thus, given a broad RMSEA estimate and non-random nature of the error terms (albeit within the same test), a replication of these results with a broader selection of cognitive tests was needed. In a subsequent Gibson and Kleitman study (2009) (N = 177; age ranged from 10 years 1 month to 12 years 7 months), confidence ratings were utilised in three standardised achievement tests (Vocabulary, Reading and Mathematics). The study also utilised separate markers of Gf and Gc which did not contain confidence ratings. The results indicated a clear presence of two cognitive ability factors—each defined by the accuracy scores on tests that are markers of Crystallised Intelligence (Vocabulary, Reading and Gc tasks) and Fluid abilities (Gf and Mathematics tests) (see Table 1). There was also a separate broad Selfconfidence factor defined solely by confidence ratings from three achievement tests (see Table 1). Not surprisingly, the two ability factors have positive associations with each other (.53). The Self-confidence factor, although it shared medium size positive correlations with two ability factors (.42 with Gc and .41 with Gf), separated clearly from them. Thus, when measured across different cognitive items, cognitive tests, and knowledge domains, a ‗cognitive‘ Self-confidence factor emerges to reflect the stability of confidence judgments within a cognitive domain. This occurs regardless of the diverse nature of test selection (whether they are markers of Gf, Gc or aptitude tests). However, the question remained: does this stability generalise to domains other than cognitive test-taking?

108


How sure are you that you could do this?

I am very sure

I am pretty sure

I am not very sure

I know that I couldn‘t

Figure 3. Example of the Perceived Movement Confidence item.

Our most recent study (N=189; age ranged from 10 years 4 months to 13 years) aimed to bridge this gap and employed different cognitive tasks as well as a measure of confidence in the physical domain (Mak & Kleitman, 2009). In this study, we employed the Stunt Movement Confidence Inventory measure (SMCI; Griffin & Keogh, 1982). According to these authors, evaluating movement confidence involves examining perceived movement competence (evaluating personal skills and situational demands) and perceived movement sensation (evaluating expected and preferred movement sensation). Movement confidence is argued to represent an outcome state of an interactive cycle among three major variables that constitute movement competence and movement sensation. The three variables are: (1) competence in performing the task; (2) the potential for harm, and; (3) the potential for enjoyment in the performing of the task. The SMCI captures five variables—perceived movement confidence, experience with task, enjoyment, competence and harm—in relation to six movement tasks presented in pictorial format (see Figure 3). The measure utilises Harter‘s (1982) nonstandard response format when presenting statements measuring enjoyment, competence and harm variables. Perceived movement confidence is measured with an item that uses similar language to self-confidence line scales, as seen in Figure 3. This measure differs from a traditional way of capturing confidence levels in the cognitive domain (verifiable responses linked to the actual decision-making act) and relies on a perception of movement competence. Being aware of this limitation, we also examined the convergence pattern between these perceptions and different self-concept measures in an attempt to determine their veracity. Moreover, previous research indicates that by the age of 11, children develop adequate movement competence beliefs. Coote and Livesey (1999) found that while 6-yr-old children demonstrated ‗optimism bias‘ with regards to their ability to perform various motor tasks, 11-year-olds‘ estimates of performance matched their ability. Jacobs, Lanza, Osgood, Eccles and Wigfield (2002) also reported a decline in children‘s sports competence beliefs over this period but also found further declines into adolescence. The results indicated that the Cronbach‘s alpha reliability estimate for Movement Confidence Perceived was .70, while the reliability coefficient for the Experience with Tasks was low (alpha = .68) and ranged between .80–.86 for the other three SMCI scores. The results of the Exploratory Factor Analyses were unambiguous—there was a clear separation in children‘s assessment of confidence across physical and cognitive domains (see Table 2).


109

Table 2. Results of the EFA (ML, Promax Rotation), Mak & Kleitman (2009; N =189) Factors 1 Movement Competence Movement Confidence Perceived Experience with Task Movement Enjoyment Movement Harm PAT Vocabulary Confidence PAT Reading Confidence PAT Math Confidence PAT Vocabulary Accuracy MHVS PAT Reading Accuracy PAT Math Accuracy Raven‘s Progressive Matrices Accuracy Factor Inter-correlations F1: Physical confidence/competence factor F2: Cognitive Self-confidence factor F3: Gc F4: Gf

2

3

4

.88 .88 .80 .78 -.75 .98 .76 .57

.39 .76 .73 .56

1

.08 1

.06 .52 1

.32 .72 .61

h2 .81 .79 .65 .67 .61 .89 .72 .61 .66 .74 .67 .56 .55

0 .43 .55 1

Note: Only factorial loadings above .30 are presented; h2 =communalities; PAT = Progressive Achievement Test; MHVS = Mill Hill Vocabulary Scales; Gc = Crystallised Intelligence; Gf = Fluid Intelligence.

That is, perceived movement confidence loaded together with the assessments of movement competence, experience with the tasks, movement enjoyment and harm (with a negative loading), defining a ‗physical’ confidence/competence factor (Factor 1). Confidence ratings from cognitive tests defined a separate general ‗cognitive’ Self-confidence factor (Factor 2). Consistent with the previous studies, accuracy scores from the cognitive tests defined two ability factors—Gc (Factor 3, defined by accuracy scores of the Vocabulary, Reading and Mill-Hill Vocabulary tests and Gf (Factor 4, defined by the accuracy scores of the Raven‘s Progressive Matrices and Mathematics tests). Both confidence factors were predicted by the relevant self-concept measures and out-ofschool activities: while the ‗cognitive‘ Self-confidence factor was predicted by academic/metacognitive self-concepts, the ‗physical‘ Self-confidence/competence factor was predicted by physical self-concept. These findings suggest that confidence ratings from each domain have meaningful relationships with beliefs that are relevant for each domain. Thus, it demonstrates the veracity of a cognitive and physical split of confidence/competence assessment, likely to be based on the domain of test-taking activities rather than a method of assessment (verifiable performance for the cognitive domain and self-reports for the physical domain).

110


RELATIONSHIP BETWEEN PARENTS’ AND CHILDREN’S METACOGNITIVE PERCEPTIONS, SELF-CONFIDENCE AND ACCURACY We also examined parental perceptions of their child‘s metacognition and parental confidence ratings (189 children, 57 mothers and 23 fathers). This study is the first to attempt to link child and parental levels of self-confidence (Lau & Kleitman, 2009). Parents were asked to complete the Esoteric Analogies Test (EA; Stankov, 1997) which included confidence ratings. In this cognitive test, each item contains a pair of words in a particular relationship, and a third word with its pair word missing. Parents were required to identify a logical relationship between the third word and one of four multiple choice words to complete the second pair. For example, ―CHICK is to HEN as CALF is to: 1. BULL, 2. COW, 3. COAT, 4. ELEPHANT‖ (answer = COW). Parents and children also completed the Metacognition Rating Scale adapted from the Teacher Rating Scale (Sperling, Howard, Miller & Murphy, 2002). This scale, originally designed to measure teachers‘ perceptions of their students‘ level of metacognition, was modified in three ways: (1) to provide a more userfriendly layout; (2) to gauge parents‘ (instead of teachers‘) awareness of their child‘s metacognition, and; (3) to capture the child‘s own perception of his/her metacognition. Five items asked what the parent thinks of the child‘s attention, study planning/learning strategies and performance monitoring accuracy using a five-point Likert scale, ranging from 1=Never, to 5=Always. Table 3 summarises the Pearson zero-order correlations between: (1) parental confidence levels, and; (2) parental perceptions of child‘s metacognition (for both mother and father separately) with the relevant child‘s variables. The most salient correlations between parents‘ and children‘s metacognitive perceptions, self-confidence and accuracy will be reviewed here. Child‘s rating of their own metacognition correlated positively with their levels of confidence and their accuracy of performance on standardised achievement tests (r = .43. p < .05, n = 189; r = .26. p < .05, n = 181). This suggests that one‘s own perception of competency of metacognitive skills predicted higher levels of child‘s performance and relevant confidence levels, and that the strength of these relationships was somewhat stronger for confidence than for accuracy scores. There were also positive correlations between children‘s rating of their own metacognitive abilities and mothers‘ and fathers‘ perception of their child‘s metacognitive abilities (r = .27, p < .05, n = 57; and r = .23, p > .05, n = 23 respectively). While the sizes of these correlation coefficients were similar, the coefficient for fathers did not reach statistical significance. However, this might be due to a limited sample size of fathers who volunteered to take part in this study (23 fathers compared to 57 mothers). Confidence scores for mothers and fathers shared significant positive correlations (r = .52, p < .05, n =15). The correlation between the accuracy scores was also positive, although it did not reach statistical significance (r = .24, p > .05, n = 15). Children‘s accuracy scores were meaningfully but not significantly correlated with parents‘ accuracy scores (r = .23, p > .05, n = 57 and r = .34, n = 23, p > .05). The most intriguing finding, however, was a pattern of correlations between child‘s and parents‘ confidence levels: a child‘s confidence correlated highly and positively with father‘s, but not mother‘s levels of confidence (r = .42, p < .05, n = 23 and r = .03, |p > .05, n = 59 respectively).


111

MRS: Children‘s Rating

r

.27

n

MRS:Mothers‘ r Confidence n Fathers‘ EAT confidence Children‘s EAT Confidence

Fathers‘ EAT Accuracy Children‘s EAT Accuracy

Fathers‘ EAT Accuracy

.23

1

57

23

189

.04

.35

.16

1

19

19

57

.15

.29

.52**

1

23

23

19

23

.03

.42*

1

23

189

.388

.32*

1

19

59

59

.36

.24

1

19

23

.23

.34

59

23

55

r

.46

n

19

*

**

r

-.03

.34

.43

n

57

23

189

Mothers‘ EAT r Accuracy n

Mothers‘ EAT Accuracy

23 *

Children‘s EAT Confidence

19

Fathers‘ EAT confidence

1

Mothers‘ EAT Confidence

.83**

Children‘s MRT Rating

Fathers‘ MRT Rating

MRS: Fathers‘ r Rating n

Mothers‘ MRT Rating

Table 3. Pearson Zero-Order Correlations Between Parents’ & Children’s Metacognitive Perceptions, Self-confidence & Accuracy Scores (Lau & Kleitman, 2009; N =1 89)

57

-.19

.08

.07

.39

57

19

59

57

r

.23

.13

.24

n

19

23

23 **

r

.22

.33

.26

n

57

23

181

*

.130

.59

19

23

**

23 **

-.13

.26

.57

57

23

189

MRS = Metacognition Rating Scale, EAT = Esoteric Analogies Test; n = number of available responses; r = zero-order Pearson correlation coefficient.

DISCUSSION Metacognition is one of the three fundamentals of self-regulated learning, along with cognition and motivation (Schraw et al., 2006). Efficient test-taking behaviour and test-taking outcomes signify academic success. Moreover, the metacognitive confidence judgments which students assign to their on-going performance are at the core of this test-taking behaviour. The studies overviewed in this chapter were the first to examine the existence of the Self-confidence trait in primary school children. We identified the existence of a habitual response pattern of confidence levels, or a trait, which is stable across different cognitive tasks and across movement competence reports, but not across both domains. We were also the first to attempt relating a child‘s self-confidence with the confidence levels of parents. Taken together, our results allow us to draw conclusions as to whether confidence judgments are task or domain-specific in middle childhood. The results support the domain-specific

112


stability of the cognitive Self-confidence factor, and its independence from the physical confidence/competence factor. Thus, within a relative domain, self-confidence exists as a stable and identifiable metacognitive factor in children as young as 9-13 years of age, just as it does in adults. Within each domain, but especially cognitive, confidence scores exhibited high reliability coefficients, suggesting their stability within tasks and across the tasks within the same domain. These novel findings signify that self-confidence is a stable and domainspecific component of a child‘s thinking repertoire in primary school. Thus, these three studies provide a foundation for describing developmental theories of the self-confidence trait. Our most recent study was also the first to examine the relationships between parental evaluation of their child‘s metacognitive abilities and parental levels of self-confidence, and those constructs of their child. Although the sample sizes are small, an interesting pattern emerged. Parental perception of their child‘s metacognitive abilities had a similar strength of positive relationship with the child‘s cognitive and metacognitive performance (positive, similar low size correlations, albeit statistically significant for mothers but not for fathers—a likely result of a smaller sample size of fathers). However, there seems to be an intriguing difference in the strength of the relationships between maternal and paternal levels of confidence and those of a child: moderately strong for fathers and non-existent for mothers. Although further replications with larger sample sizes are required, this study is the first to examine such relationships and this finding is not trivial as it suggests possible different learning roots of self-confidence within a family. Given the obvious limitations of this study, no conclusions could be offered, but this study provides a new direction for future studies into the origins of confidence levels. The results overviewed in this chapter may also assist with improved metacognitive development. For example, knowledge that a child as young as nine is already habitually assessing their own thinking is a crucial and powerful tool, one which can undoubtedly assist parents, teachers, school counsellors and child psychologists to foster self-regulated learning (see Schraw et al., 2006 for a review). Investment in the development of students‘ metacognitive skills could assist in improving performance in academic areas. Moreover, to foster accuracy of performance, many teachers repeatedly tell children to ‗check their work‘. The use of a confidence rating technique (as part of regular in-class practices), could make these teaching practices more efficient. Teachers can also use self-monitoring measures as informal assessment to determine students‘ level of understanding in specific Key Learning Areas—for example, Literacy and Numeracy. In addition, teachers and school counsellors often meet with parents to provide them with strategies they can use to assist their children who struggle with learning. In such meetings, parents‘ understanding of their children‘s metacognitive skills should be promoted to empower parents‘ involvement in their children‘s cognitive and physical development. Furthermore, academic performance is only one part of a child‘s development. With a significant rise in obesity problems, children would undoubtedly benefit from the process of healthy self-reflection within the physical domain, which seems to clearly separate from the cognitive domain. Findings such as those by Wigfield, Eccles, Yoon, Harold, Arbreton, and Blumenfeld (1997) and Jacobs et al. (2002) indicate that changes in self-perceptions of competence are related to changes in value for an activity over time and that this is likely to affect participation in the activity. This is a particular problem that has been noted in children with movement difficulties (Henderson, 1993; Piek, Baynam & Barrett, 2006). Children with


113

movement difficulties avoid situations in which these are displayed, hence missing opportunities to practice and develop such skills, thus exacerbating the problem.

CONCLUSION The exploratory nature of our studies limits the scope of conclusions drawn. A longitudinal research design with larger sample sizes and with a greater control for background factors and common causes (e.g., pre-school levels of cognitive and metacognitive development, previous achievement) could greatly assist in determining developmental trajectories of the Self-confidence traits. It could also suggest the age at which they become both habitual within a domain and also domain-specific. Within such studies, parental metacognitive variables should also be assessed to determine possible links between parental levels of confidence and those of their child. Future research would also benefit from a larger selection of variables to mark each construct. Finally, future studies should utilise the same method of assessment of confidence in cognitive and physical domains, the method which allows for the immediate verification of the veracity of confidence ratings.

REFERENCES Allwood, C. M., & Granhag, P. A. (1999). Feelings of confidence and the realism of confidence judgments in everyday life. In P. Juslin & H. Montgomery (Eds.)., Judgment and Decision Making: Neo-Brunswikian and Process-tracing Approaches. (pp. 123-146). Mahwah, NJ, US: Lawrence Erlbaum Associates, Inc., Publishers. Allwood, C.M., Granhag, P.A., & Jonsson, A.C. (2006). Child witnesses' metamemory realism. Scandinavian Journal of Psychology, 47, 461-470. Allwood, C.M., Innes-Ker, Å., Homgren, J., and Fredin, G. (2008). Children‘s and adults‘ realism in their event-recall confidence in response to free recall and focussed questions. Psychology, Crime & Law, 14, 529-547. Australian Council for Educational Research. (2001). Progressive Achievement Test Reading Comprehension (Revised Edition). Melbourne: Author. Australian Council for Educational Research. (2001). Progressive Achievement TestVocabulary (Revised Edition). Melbourne: Author. Australian Council for Educational Research. (2005). Progressive Achievement TestMathematics (3rd Edition). Melbourne: Author. Coote, K. & Livesey, D. (1999). Optimism bias in children‘s motor performance expectations. The Australian Educational and Developmental Psychologist, 16, 52-61. Flavell, J. H. (1979). Metacognition and cognitive monitoring: A new area of cognitivedevelopmental inquiry. American Psychologist, 34, 906-911. Gibson, J., & Kleitman, S. (2009). The impact of the learning environment on the metacognitive beliefs, self-confidence and use of self-handicapping strategies. Proceedings of the 3rd Redesigning Pedagogy International Conference, Singapore.

114


Griffin N. S., & Keogh, J. F. (1982). A model for movement confidence. In J. A. S. Kelso & J. E. Clark (Eds.). The Development of Movement Control and Co-ordination (pp. 213238). New York: John Wiley. Harter, S. (1982). The perceived competence scale for children. Child Development, 53, 8797. Henderson, S. E. (1993). Motor development and minor handicap. In A.F. Kalverboer, B. Hopkins & R. Geuze (Eds.) Motor development in early and later childhood: Longitudinal approaches. Cambridge, UK: Cambridge University Press. Jacobs, J. E., Lanza, S., Osgood, D. W., Eccles, S. & Wigfield, A. (2002). Changes in children‘s self-competence and values: Gender and domain differences across grades one through twelve. Child Development, 73, 509-527. Jonsson, A. & Allwood, C. M. (2003). Stability and variability in the realism of confidence judgments over time, content domain, and gender. Personality and Individual Differences, 34, 559-574. Kleitman, S. (2008). Metacognition in the rationality debate. Self-confidence and its calibration. Germany: VDM Verlag Dr. Mueller E.K. Inc., Publishers. Kleitman, S. & Moscrop, T. (2010). Self-confidence and academic achievements in primaryschool children: their relationships and links to parental bonds, intelligence, age, and gender. In A. Efklides and P. Misailidi (Ed.). Trends and Prospects in Metacognition Research (Chapter 13). US: Springer. Lau, P. & Kleitman, S. (2009). Predicting achievement and self-confidence: Intrapersonal and interpersonal predictors in school-aged children. Proceedings of the 8th Australasian Conference on Personality and Individual Differences, Sydney, Australia. Mak, K., & Kleitman, S. (2009) Metacognitive regulation in the physical domain: An investigation of school aged children‘s movement confidence and its relationship with self-concept and Big Five personality traits. Proceedings of the 8th Australasian Conference on Personality and Individual Differences, Sydney, Australia. Metcalfe, J., & Shimamura, A. P. (Eds.). (1994). Metacognition: Knowing about knowing. Cambridge, MA, USA: Mit Press. Nelson, T. O., & Narens, L. (1994). Why investigate metacognition? In J. Metcalfe & A. P. Shimamura (Eds.), Metacognition: Knowing About Knowing (pp. 1-25). Cambridge Massachusetts: The MIT Press. Piek, J., Bayman, G. & Barrett, N. (2006). The relationship between fine and gross motor ability, self perception and self-worth in children and adolescents. Human Movement Science, 25, 65-75. Raven, J. C., Court, J. H., & Raven, J. (1982). The Mill-Hill Vocabulary Scale. Manual for Raven's Progressive Matrices and Vocabulary Scales. London: H.K. Lewis & Co. Raven, J., Raven, J. C., & Court, J. H. (1998). Manual for Raven's progressive matrices and vocabulary scales. San Antonio, TX: Harcourt Assessment. Schraw, G., & Dennison, R. S. (1994). Assessing metacognitive awareness. Contemporary Educational Psychology, 19, 460-475. Schraw, G., & Moshman, D. (1995). Metacognitive theories. Educational Psychology Review, 7, 351-371. Schraw, G., Crippen, K. J., and Hartley, K. (2006). Promoting self-regulation in science education: metacognition as part of a broader perspective on learning. Research Science Education, 36, 111-139.


115

Sperling, R. A., Howard, B. C., Miller, L. A., & Murphy, C. (2002). Measures of children‘s knowledge and regulation of cognition. Contemporary Educational Psychology, 27, 5179. Stankov, L. (1999). Mining on the "no man's land" between intelligence and personality. In P.L. Ackerman, P. C. Kyllonen & R. D. Roberts (Eds.). Learning and Individual Differences: Process, Trait and Content Determinants. (pp. 315-367). Washington: DC, USA: American Psychological Association. Stankov, L. (1997). The Gf/Gc Quickie Test Battery. Unpublished test battery. School of Psychology: The University of Sydney. Wigfield, A., Eccles, J.S., Yoon, K.S., Harold, R.D., Arbreton, A.J.A. & Blumenfeld, P.C. (1997). Changes in children‘s competence beliefs and subjective task values across the elementary school years: A three-year study. Journal of Educational Psychology, 89, 451-469.

PART B: CULTURE, RELIGION, AND SPIRITUALITY



Chapter 10

SPIRITUAL-HYPNOSIS ASSISTED THERAPY: A NEW CULTURALLY SENSITIVE APPROACH TO THE TREATMENT AND PREVENTION OF MENTAL DISORDERS Luh Ketut Suryani Department of Psychiatry, Faculty of Medicine Udayana University, Denpasar, Bali 80114, Indonesia The Suryani Institute for Mental Health, Denpasar, Bali 80237, Indonesia

ABSTRACT The chapter describes the therapeutic approach of Spiritual-Hypnosis Assisted Therapy (SHAT), originally developed in Bali, Indonesia, and its applications to treatment, intervention, and prevention of mental illness among individuals, groups, and communities. A background of the Balinese culture and religion is also presented to show how it was integral to the development of the method. SHAT is being extensively and successfully used in clinical practice in Bali for over a decade now and for a variety of mental disorders, emotional distress (specifically grief and bereavement), sexual, relational, and sexual orientation problems, and PTSD.

Keywords: Cultural psychology, Mental health, Personality change, Spiritual hypnosis assisted therapy



Ph: 62-361-462306; Fax: 62-361-462878; [email protected]. Originally published as: Lesmana, C.B.J., Suryani, L.K., Tiliopoulos, N, & Jensen, G.D. (2010). Spiritual-Hypnosis Assisted Therapy (SHAT): A new culturally sensitive approach to the treatment and prevention of mental disorders. Journal of Spirituality in Mental Health, 12(3), DOI: 10.1080/19349637.2010.498696. Reproduced & adapted with permission.

120

Luh Ketut Suryani

INTRODUCTION Throughout life one faces many problems. Some people can handle most of them well, while others find it difficult to cope successfully. Experiencing and dealing with problems may make one a stronger individual, or in certain cases may negatively influence thought, behaviour, and emotions and potentially result in mental illness. Although the unconscious mind always tries to resolve these problems, the individual may still be unhappy and out of balance, albeit still psychologically functional. Problems can have genetic, organic, experiential, psychological, or a combination of these causes. Past traumas can act as pathological triggers, resulting in the above causes entering dysfunctional or psychopathological states (Bolton et al., 2004). Should one follow the current psychobiological evidence-based approaches (e.g., Canli, 2006; Zuckerman, 2005), one would, arguably, conclude that psychology, and in particular personality, begins developing through proto-experiences in the womb, as soon as the foetal nervous system is formed through the process of neurolation. These proto-experiences influence and shape the biological foundations of thought, emotion, feeling, behaviour, response to stimulation, and adaptation to new situations. Should these proto-experiences be traumatic, they will contribute to the dysfunctional dispositions of the psychobiological systems, and subsequently increase the individual‘s vulnerability to stress and their ability to successfully cope with problems later in life. The purpose of this chapter is to introduce, describe, and discuss the therapeutic approach of Spiritual-Hypnosis Assisted Therapy (SHAT), developed in Bali, Indonesia. The approach focuses on memory and the unconscious mind, and considers the effects of proto-experiences on the (dys)functional psychological development of the individual. It utilises spiritual, psychodynamic (in particular clinical hypnosis), CBT, and humanistic therapeutic elements. I propose SHAT as a new, alternative, and both culturally and spiritually sensitive method of treatment and prevention of mental health problems. A background of the Balinese culture, spiritual beliefs, and religion is also presented, to assist with fully understanding the origins and development of the approach.

BALINESE HINDUISM The island of Bali is geographically located about 8 degrees south of the equator and about 18 degrees north of the western end of Australia. Bali is a small island (5633 square kilometres), one of 13,000 islands that make up the archipelago of Indonesia, the fourth largest nation in the world (Badan Pusat Statistik, 2007a). The total Indonesian population of approximately 220 million is mostly Muslim. Most of the 3.2 million inhabitants of Bali are Hindu (83%), and they constitute less than 2% of the total population of Indonesia (Badan Pusat Statistik, 2007b). The Balinese Hindu religion is unique. It has its roots in India, but was developed largely in Java. It has been influenced by Buddhism, by the aboriginal Balinese culture, and by Balinese pre-Hindu animistic and ancestral cults (Jensen & Suryani, 1992). Furthermore, its relatively isolated nature—being practised almost exclusively on the island of Bali—makes it a unique sect, having both function and content that at times largely deviate from mainstream

Spiritual-Hypnosis Assisted Therapy

121

Hinduism – in fact, it is this cultural uniqueness of the Balinese society, combined with its relativistic/collective structure, that has made it the focus of major anthropological and sociological research (e.g., Bateson & Mead, 1942; Edge, 1996; Geertz, 1973; Suryani & Jensen, 1993).

CENTRAL CONCEPTS IN THE BALINESE RELIGION Panca Srada, the five principal beliefs, and Tri Hita Karana, the concept of harmony and balance, infuse all aspects of the Balinese daily life: (1) The principle of one Supreme God (Sang Hyang Widi Wasa), who is the highest power in one‘s life and a source of life for all people in the world; (2) The principle of the (personal) spirit, the eternal soul (Atman), which is part of God and the source of life and human ability; (3) The principle of Hukum Karma or Karma Pala, which means that current human conditions and behaviours are both a result of past lives and the foundations of the future ones. Karma applies not only to oneself, but also to one's descendants; (4) The principle of reincarnation (Punarbawa) , the continuing cycle of birth, death, and rebirth into the world, which is repeated until one attains the perfect life, at which point it ends, because one is unified with God. (5) The principle of unity with God (Moksha), which may be experienced during one's lifetime or after death in heaven. Once Moksha is achieved, the cycle of birth-rebirth ends. Human relationships are governed by a cosmology that places humans in a ―middle‖ world, between the spirit worlds above and below. As such, life revolves around finding and maintaining harmony and balance among gods, humans, and nature. Traditionally, the Balinese believe that three factors are crucial to a person's well-being, happiness, and both physical and psychological health: (1) the microcosm (Buana alit), which is made up of individual persons; (2) the macrocosm (Buana agung), which comprises the universe; and (3) the supreme God (see Jensen & Suryani, 1992). Peace, both external (interpersonal) and internal (psychological and spiritual), is attained by doing good deeds and by maintaining a holistic self-balance. For these reasons, Balinese strive to maintain good relations among each other, with the environment, their ancestors, the many gods (Dewa) and the supreme God. Harmony in relationships creates a prosperous, peaceful, and happy life (Rahayu). Imbalance and disharmony affect both natural (Sekala) and supernatural (Niskala) elements that in return influence humans, causing both mental disorders and physical illness. The Balinese Hindus believe that ritual ceremonies can redeem karma, and lead to the experience of a peaceful life. The primary Balinese solution for overcoming psychological problems is through ritual ceremonies, with the full understanding that the result ultimately depends on God (Suryani & Jensen, 1993). One accepts whatever happens as the best thing that could happen, treating it either as a test, an opportunity to redeem past misdeeds, or to achieve a better result in the days to come1. God is above all, so if one strives to unite with God, one will be rewarded with the best for the future. 1

The terrorist bombings in 2002 and 2005 were an example of the operation of these principles. While fully cognizant of the horror of so many deaths, grief and much suffering, the Balinese responded by seeking to identify the imbalance in their own actions that may have brought it about. Along with numerous ritual cleansing ceremonies performed at ground zero and across the island, there was an attempt to re-evaluate what they did in

122

Luh Ketut Suryani

For the Balinese the dichotomies of good and evil are constantly present in all conditions of life; it is simply a matter of which one is dominant at any given time. Similar to Western existentialism, good and evil are always present within everyone. Whether one becomes a good or a bad person depends on which of these forces is or becomes dominant in them. Balinese aim to harmonise both of them in their daily life. For example, the demon Buta Kala can cause mischief, if we have bad relationships, but will assist us positively if our relationships are good. Balinese Hindus are taught about these tenants of spirituality from the early days of their life, either by their families, religious leaders, or throughout schooling. There principles are fundamental components of the SHAT approach (Suryani & Wrycza, 1996).

SPIRITUAL-HYPNOSIS ASSISTED THERAPY (SHAT) Biopsycho-Spirit-Sociocultural Approach I have proposed, developed, and successfully been using (and improving) for the past 10 or so years the SHAT approach. I have been treating patients both as a Balinese spiritual healer (Balian) and a psychiatrist (Edge & Suryani, 2002; Edge et al., 2005; Suryani, 1984; Suryani, 2004; Suryani & Entus, 2004; Suryani & Jensen, 1992; Suryani & Stephen, 2000; Suryani et al., 1988; Suryani et al., 2006).The former identity is rooted in the concepts of Balinese culture and Hinduism, while the latter is based on Western concepts and training. In my clinical practice, I integrate these concepts and the ideas behind them in what I call the biopsycho-spirit-sociocultural, and thus holistic, approach (Suryani, 1995)2. Within the context of this term, and the Balinese Hinduism, every person consists of mind, body, and spirit, which largely constitute one's biology and psychology, and are influenced by and in return affect, positively or negatively, one‘s sociocultural environment and religious/spiritual beliefs (see also Lesmana & Tiliopoulos, 2009).

The Concept of SHAT Our personal spirit (atma) is primarily made of our mother‘s, father‘s, and God‘s energy, and it comes into existence at conception. It is the source of our life, ability, knowledge, and aspirations. Personal spirit influences both physical and mental development. Psychological proto-development in the womb is, thus, influenced by the personal spirit, the mother's physical and mental condition, and also by the father, the family, the environment, and God. The quality of the foetus‘ proto-experiences in the womb is largely determined by the pregnant mother‘s psychophysiological states (e.g., feeling happy, calm, stressed, angry, or being physically or psychologically ill; see also, Schwerdtfeger & Goff, 2007). Positive and

2

tourism. It was perceived as a wake-up call, signalling that tourism had gone out of sync with Balinese values, and an opportunity to rethink priorities. Aspects of my treatment were recently shown (30 May 2009 and 14 March 2010) on two documentaries on Australian national television, which can be viewed online at the following addresses: http://www.sbs.com.au/dateline/story/about/id/600049/n/Bali-s-Shame http://www.sbs.com.au/dateline/story/ watch/id/600377/n/Bali-s-Miracle-Worker


123

negative proto-experiences are encoded in the foetus‘s proto-memory, and serve as elements of the biological basis of one‘s personality. Although, with the advent of logical thinking and the increasingly dominant cortical functions our proto-experiences tend to be forgotten rather early in our life, the formed proto-memories can still influence subsequent thought, emotions, learning, behaviour, and adaptive approached to dealing with stress, through acting almost exclusively on unconscious psychological processes. If the foetal experiences were largely negative or aversive, the born individual may not be able to understand them, learn from them, and utilise them (even at the unconscious level) for their personal growth. If that is so, the individual may develop into becoming apathetic, irritable, hostile, antisocial (i.e., highly psychotic), or emotionally disturbed (i.e., highly neurotic), and when or if they are subsequently exposed to stressors, they may be more prone to developing a mental disorder (see also, Lemelson & Suryani, 2006). SHAT brings the patient to an awareness of the feelings and desires of their unconscious mind. It enables them to reframe an experience (e.g., with family, friends, co-workers) or proto-experience, laden with distressing feelings or conflict, into a positive one. The patient is brought to see and feel memories of the past, place them into their present, project them into their future life, and understand and reframe them. The patient actively expresses emotions from their unconscious mind, and develops a different understanding of these memories. The hypnotic state further enables them to modify past situations and feelings, and record these as newly changed in conscious memory. The method does not urge nor imposes change by the therapist. Dealing with their unconscious mind in the hypnotic state helps and allows the patients to overcome their problems, and achieve, or at least work toward, desired personal goals. In hypnosis, the patient is allowed to make their own choices, and follow their desires regarding personal relationships. Finally, hypnotic suggestions may not be given to change sexual orientation. SHAT combines and modifies treatment elements of psychoanalysis, CBT, humanistic therapies, and religious/spiritual counselling. However, it significantly differs from the conventional therapeutic practices of these approaches through: (a) the absence of systematic desensitization; (b) imagery is replaced with re-experiencing and re-evaluating the past; (c) the presence of a non-guided regression process; (d) the modified role of the unconscious, and; (e) the elements of faith, in particular that of the personal spirit, that play a central and constant role in the therapeutic discourse of SHAT and that tend to be absent or of peripheral or eclectic importance and utility in most conventional mainstream hypnotherapies and psychotherapies.

THE METHOD: INDIVIDUAL THERAPY In order for SHAT to be successfully utilised, the patient must trust the therapist, not be delusional or hallucinating, and not have a (comorbid) panic or anxiety disorder. Should the mentioned symptoms be present, SHAT is used in combination with psychotropic medication. In any case, the patient is interviewed in depth; the clinical diagnosis is based on the ICD-10 diagnostic criteria; additional information is also collected, in order to understand their background, their childhood, adolescence, and adulthood history, the main complaint(s)

124

Luh Ketut Suryani

that brought them to seek counselling, and to begin establishing a good relationship with them and gaining their trust. The patient may sit on a chair or lie on the couch. They are asked to feel comfortable and relaxed. After they relax, they are asked to focus only on the therapist‘s voice. Through this process, they are initially instructed to meditate with their eyes closed; after meditating for approximately five minutes, they are instructed to breathe deeply through their nose and slowly exhale through their mouth. These instructions are repeated five times. These deep breaths last for one minute. Subsequently, the participants‘ breathing tends to remain constant and deep, without any further guidance from the therapist, at approximately five breaths per minute. At that stage, the therapist may wish to monitor certain behavioural cues on the patients, such as their arm rigidity and immobilisation, their eye catalepsy, and their inability to communicate, all of which are widely accepted indicators of a hypnotic state—which within the context of SHAT is referred to as a trance state. When in trance, patients are verbally guided, through suggestion, to be sensitive of their personal spirit, and to go back to memories of an unhappy experience, where they felt angry, disappointed, ashamed, sad, scared, worried, or the specific situation of relevance to their problem, if it were clearly identified during the clinical interview. When they remember, feel, or see the experience(s), they are asked to express all unpleasant feelings by, for example, shouting, crying, or even hitting something. The purpose of this is for them to release the emotional burden that has been stored in their memory. They are led to their culturally familiar concept that the power of God would help them gain strength to cope with their emotions. These emotional discharges are different to abreactions, because the experience of catharsis is absent, since the participants are in a hypnotic state, and thus unaware of experiencing any conflictual material. When they have finally expressed their emotions and have returned to a calm state, they are asked to see the experience(s) again, in order to help them understand why they touched and affected them in such a way. They are then instructed to accept this happening in their life and to understand that the experiences were important in developing them to be or become an adult, and that they enabled them to face new challenges in life. In the last step, they need to accept that these experiences belong to the past, and it is made continually clear to them that the past is the past, and that their reality is life today, tomorrow, and the day after. If they are not able to understand, or see the situation again, or are unable to reframe it into a positive context, they are asked to look deeply (even in a figurative sense) into the eyes of their mother, who, if possible, should be present at this stage and feel her eyes. In so doing, a positive meaning of the experience tends to emerge, and subsequently the patients tend not to be influenced by their old emotions. They reframe the experience into having a positive meaning and carrying positive emotions, and accept it as an important element of them becoming a fully psychologically developed individual. Continuing in hypnosis, they may be asked to go to the past, when they were children, in order, once again, to reframe early-life traumas or bad memories. They may feel memories of their mother, father, nuclear family, peers, friends, and environmental surroundings. Negative experiences are reframed to have a positive meaning. An example of a suggested reframing could be ―Although you were angry (or scared) when your father hit you, he did it to help you become a better person‖. For them to feel safe, loved, and respected by their parents it is important to positively influence and guide their thoughts, emotions, behaviour, adaptation, and their ways of communicating and relating to others.


125

Table 1. Treatment statistics of individual SHAT, broken down by disorder diagnosed

*

Main problem diagnosed (ICD-10)

Number of patients treated since 2003

Anxiety disorders Childhood trauma Depression Drug abuse Identity disorders Marital problems Panic disorders Psychoses PTSD Sexual abuse Total number of patients:

50 20 85 10 5 75 10 15 20 25 315

Mean success rate:

Success rate* 80% 80% 76.4% 70% 100% 80% 80% 66.6% 90% 100% 82.3%

Success refers to the absence of observed symptomatology, clinically assessed immediately after the treatment and a month later.

In the final part of the session, they are asked to find the way to happiness in future life— material and spiritual. Toward the end of the session, forgiveness and apology may be suggested. At the end of the session, the patient is guided back to their usual state of consciousness and may be given brief explanations to help them understand the psychological interpretation of the experience(s) expressed and felt while in trance. The entire session usually lasts for over an hour and up to three hours (see Table 1 for specific treatment information and outcomes). For continuing results, the patients are instructed to meditate for 10 to 15 minutes, twice a day, morning and evening, or before going to sleep. Meditation helps with the maintenance of a psychophysiological homeostatic state, and as a result, it allows the self-modification of memories to work faster and be more efficient. Table 2. Demographic information of group-SHAT participation between 1999-2009 (total N = 6880) Attendance: Gender Female: Male: Education Elementary: Junior high: High school: Diploma: University: Religion Buddhist: Christian: Hindu: Muslim:

100-200 people per session Age (in years) 59.3% 40.7%

6.7% 11.9% 47.5% 8.5% 25.5% 4.2% 5.1% 83% 7.7%

M: SD: Range: Marital Status Never married: Married: Divorced: Widowed: Separated: Treatment satisfaction Very satisfied: Satisfied: Somewhat satisfied: Not Satisfied:

38 13 16-70 35% 59.1% 0.8% 4.2% 0.8% 42.2% 37% 17.8% 3%

126

Luh Ketut Suryani

THE METHOD: GROUP-THERAPY The above procedure can also be used in group-therapy setting, although the instructions may need to be more general to apply to all the patients present, and the sessions are usually shorter (around 30 minutes). In addition, for this mode of therapy to be more effective, patients should be matched on symptomatology severity and preferably on the primary diagnosed problem. For example, I used group-SHAT in the treatment of children suffering from posttraumatic stress disorder (PTSD) after the 2002 Bali terrorist attack (Lesmana & Suryani, 2007; Lesmana et al., 2009). The aim of that study was to assess the effectiveness of groupSHAT for treatment of PTSD in children. All children, age 6-12 years (N=226; 52.7% females), who experienced the terrorist bomb blasts in Bali in 2002, and subsequently were diagnosed with PTSD were studied, through a longitudinal, quasi-experimental (pre-post test), single-blind, randomised control design. Of them, 48 received group-SHAT (treatment group), and 178 did not receive any therapy (control group). Statistically significant results showed that SHAT produced a 77.1% improvement rate, at a two-year follow up, compared to 24% in the control group, while at the same time, the mean PTSD symptom score differences were significantly lower in the former group. A similar to the above, but yet unpublished, case occurred in 2005. A male teacher‘s was arrested for serially sexually abusing children in an elementary school in East Bali (Karangasem regency). The victims were 15 girls between the ages of 6-13 years old, and they had been repeatedly abused from weeks to 3 years. Two sessions of group-SHAT were run to treat their trauma. Children were assessed a week after the last session, six months later, and four years thereafter. No residual symptoms relating to their trauma were observed in any of the children. Group-SHAT sessions are being run regularly four times per year since 1999, and descriptive information of the participating individuals is shown in Table 2. The sessions are advertised in the local newspapers, where people, who believe they have (subclinical) mental health problems, are invited to participate. Given that these sessions usually attract large numbers of people, between 100-200, they are not appropriate for clinical cohorts, or delusional or hallucinating individuals. On the treatment day, the participants register and get screened by psychiatrists. However, even if no obvious symptomatology is observed, participants are still allowed to take part in the session. Before the therapy begins, they are introduced to the procedure and the concept of SHAT is briefly explained to them. After the session, which usually lasts for 30 minutes, the participants are asked how satisfied they were by the treatment and are given the option of individual follow-up sessions (see Table 2).

THE METHOD: COMMUNITY-BASED PREVENTATIVE INTERVENTION SHAT may also be, and is being used in Bali as a preventative approach to mental illness through community-based interventions (Suryani et al., 2006). The way this is done in Bali has as follows: A public announcement in the newspapers usually brings more than 200 people to participate in the program, which runs, free of charge, on a weekly basis (this


127

program is independent to the group-SHAT described above). Potential participants in this program should not have any subclinical or clinical mental health problems. Besides the verbal instructions, which are similar to the group-SHAT sessions, recorded music is played to help initiate and maintain the participants‘ trance, dissociation, or meditative state. When in a meditation state, they are asked to feel who they are in terms of their weaknesses and strengths. They are instructed to feel a past unhappy experience, and, while in a dissociative state, draw on a paper with colour pastel of their choice what they are feeling. Subsequent to this exercise, they are asked to stand up, feel the music enter their body, and let their body follow the music and express their emotions through movement. Then, participants are instructed to put their hands in front of their chest and shout loudly while shaking their fists. Next, they sit down in a meditation position, and they are, once again, asked to see their unhappy experiences and to understand why they happened to them. Feeling these experiences is important for them to become functional individuals. They are told that this was the past and that their reality is their life today. After that, they are instructed to lie down, close their eyes, and relax. They are asked to think of happy childhood memories, to feel their mother‘s and father‘s love, respect, and the safety in their presence. They are asked to feel their family, peers, friends, and the surrounding environment. During this stage, they may sleep awhile if they feel sleepy. Next, they are asked to stand up, close their eyes, and, while still recalling a happy experience when they were children, move with the cheerful music that is being played at that stage. Similar to the individual therapy, in the final part of the session, they are guided to find the way to get happiness in future life. Once again, they are instructed to meditate 10 to 15 minutes, twice a day, morning and evening, or before sleep. This session tends to run for approximately 4 hours. At the conclusion of the session, the general meaning of what they have experienced is explained to them. They are asked to rest at home for half to one hour. It is also suggested to them that they should use the rest of the day to relax, go to bed early without watching TV, or otherwise avoid being busy with any personal activities, and the next day resume their activities as usual. Most participants tend to leave these sessions with a serene or happy mood. Summary of the SHAT procedure            

Prior initial detailed interview; establish trust Relax in chair or on couch Enter trance to therapist‘s voice In trance, recall an unhappy experience Feel the emotion and let it out See it again to understand its importance Accept it as in the past Revisit a childhood trauma Feel memories of family and friends Reframe experiences to positive meaning Find the way to get happiness in future life Meditate 10 to 15 minutes, 2 times a day

128

Luh Ketut Suryani

CONCLUSION Future research needs to provide us with an understanding of the specific genetic or otherwise biological vulnerabilities that make certain individuals especially susceptible to the disorganizing effects of traumatic stress. Among the biological markers, the roles of the hippocampal volume and that of certain neurotransmitters in trauma need to be further explored (Bremner et al., 2008). Furthermore, a better understanding of the importance of specific regions of the brain to (proto-) memory may further inform intervention and therapies. More specifically, research on the mechanisms that affect memory storage reversibility, and that seem to work or at least interact with my specific hypnotherapy technique is needed in order to acquire a better understanding of, primarily, the arguably unique therapeutic effects of the spiritual elements in SHAT. Finally, assuming that SHAT is indeed a worthy approach in clinical hypnosis, future studies should assess its combined therapeutic effect with currently practised approaches, while at the same time compare it with other forms of hypnotherapy, to allow for its refinement and eclectic integration into mainstream interventions. It is my conviction that SHAT is relatively easy to master, economic, and it appears to be highly effective. Although, currently the use of the concept of the personal spirit, as presented in this paper, is based on the Balinese Hindu culture, a spiritual-therapeutic connection can be made with any group of patients, if consonant with their religious/spiritual beliefs. If so, SHAT can potentially have application in many cultures.

REFERENCES Badan Pusat Statistik (2007a). Bali in figures. Denpasar: Author. Badan Pusat Statistik (2007b). Statistical yearbook of Indonesia 2007. Jakarta: Author. Bateson, G., & Mead, M. (1942). Balinese character: A photographic analysis. New York: New York Academy of Sciences. Bolton, D., Hill, J., O'Ryan, D., Udwin, O., Boyle, S., & Yule, W. (2004). Long-term effects of psychological trauma on psychosocial functioning. Journal of Child Psychology and Psychiatry, 45, 1007-1014. Bremner, J. D., Elzinga, B., Schmahl, C., & Vermetten, E. (2008). Structural and functional plasticity of the human brain in the posttraumatic stress disorder. Progress in Brain Research, 167, 171-186. Canli, T. (Ed.) (2006). Biology of personality and individual differences. New York: Guilford. Edge, H. (1996). Possession in two Balinese trance ceremonies. Anthropology of Consciousness, 7, 1-8. Edge, H., & Suryani, L. K. (2002). A cross-cultural analysis of volition. Florida Philosophical Review, 2, 56-72. Edge, H., Suryani, L. K., Tiliopoulos, N., & Morris, R. (2005). Two cognitive DMILS studies in Bali. Journal of Parapsychology, 68, 281-321. Geertz, C. (1973). The interpretation of cultures. New York: Basic Books.


129

Jensen, G. D., & Suryani, L. K. (1992). The Balinese people: A reinvestigation of character. Oxford: Oxford University Press. Lemelson, R., & Suryani, L. K. (2006). The spirits, ngeb, & the social suppression of memory: A complex clinical case from Bali. Culture, Medicine and Psychiatry 30, 389413. Lesmana, C.B.J., & Suryani, L.K. (2007). Hipnoterapi-spiritual pada penanganan anak-anak dengan posttraumatic stress disorder setelah bom Bali 2002. Medicina, 38, 132-136. Lesmana, C.B.J., & Tiliopoulos, N. (2009). Schizotypal personality traits and attitudes toward Hinduism among Balinese Hindus. Mental Health, Religion & Culture, 12, 773-785. Lesmana, C.B.J., Suryani, L.K., Jensen, G.D., & Tiliopoulos, N. (2009). A spiritual-hypnosis assisted therapy of children with PTSD after the 2002 Bali terrorist attack. American Journal of Clinical Hypnosis, 52, 23-34. Schwerdtfeger, K. L., & Goff, B. S. (2007). Intergenerational transmission of trauma: exploring mother-infant prenatal attachment. Journal Traumatic Stress, 20, 39-51. Suryani, L. K. (1984). Culture and mental disorder: The case of Bebainan in Bali. Culture, Medicine and Psychiatry, 8, 95-113. Suryani, L. K. (1995). Cultural factors, religious beliefs, and mental illness in Bali: Indonesia. In I. Al-Issa (Ed.), Handbook of culture and mental illness: An international perspective (pp. 203-14). Madison, Connecticut: International Universities Press. Suryani, L. K. (1996). Meditation and the spiritual dimension in psychotherapy. Jiwa, XXIX, 83-90. Suryani, L. K. (2004). Balinese women in a changing society. Journal of the American Academy of Psychoanalysis and Dynamic Psychiatry, 32, 213-230. Suryani, L. K., & Entus, A. K. (2004): Progress in Indonesia: A spirit-based approach. In T. D., Nguyen (Ed.), The Indonesian Dream (pp. 60-70). Singapore: Eastern Universities Press. Suryani, L. K., & Jensen, G. D. (1992). Psychiatrist, traditional healer and culture integrated in medical practice. Medical Anthropology, 13, 301-314. Suryani, L. K., & Jensen, G. D. (1993). Trance and possession in Bali. Singapore: Oxford University Press. Suryani, L. K., & Stephen, M. (2000). Shamanism, psychosis and autonomous imagination. Culture, Medicine and Psychiatry, 24, 5-40. Suryani, L. K., & Wrycza, P. (1996). Moksha a new way of life. Denpasar: Balipost. Suryani, L. K., & Wrycza, P. (2003). Living in the spirit. Denpasar: Balipost. Suryani, L. K, Adnyana, T. A. K., Thong, D., Manik, T. I. R., Putra, D. K. W., Widjaja, W., Tama, W. & Jensen, G. D. (1988). The physical and mental health of elderly in a Balinese Village. Journal of Cross-cultural Gerontology, 3, 105-120. Suryani, L. K., Lesmana, C. B. J., & Jensen, G. D. (2006). Healing the people of Bali through meditation and trance. Natural Medicine, 26,114-115. Zuckerman, M. (2005). Psychobiology of personality (2nd ed.). Cambridge: Cambridge University Press.



Chapter 11

MILITANT EXTREMIST MINDSET Lazar Stankov National Institute of Education, Naynang Technological University, 1 Nanyang Walk 637616, Singapore

Gerard Saucier Department of Psychology, University of Oregon, Eugene Oregon 97403-1227, USA

Goran Knežević Faculty of Philosophy, University of Belgrade, 11000 Belgrade, Serbia

ABSTRACT Militant extremist mindset refers to a fixed set of beliefs that motivate people to engage in violent activities directed toward specific others. In our previous work we have identified three components that together define militant-extremist mindset, and have been identified across different cultures. These are broadly known as nastiness, grudge, and excuse: nastiness represents anti-social and pro-violent attitudes; grudge is a feeling of animosity and resentment typically directed toward some other group which serves as a trigger for action; and excuse represents a higher-order justification for engaging in militant extremism. These ―ingredients‖ of militant-extremist-mindset are defined more specifically according to the context and vary according to culture and setting. For example ―excuse‖ typically represents a particular set of religious beliefs. Further empirical work may reveal whether this mindset is present to some extent in the general population. If so, these measures of militant-extremist mindset might be useful in assessing population-wide attitudinal changes or the effectiveness of anti-terrorist social interventions.

Keywords: Cross-cultural, Extremism, Mindset, Terrorism



Ph: 65 6219 6171; Fax: 65 6316 4787; [email protected]; [email protected]

132

Lazar Stankov, Gerard Saucier and Goran Knežević

INTRODUCTION "Morality is personal. There is no such thing as a collective conscience, collective kindness, collective gentleness, collective freedom. ... To talk of social justice, social responsibility, a new world order, may be easy and make us feel good, but it does not absolve each of us from personal responsibility." Margaret Thatcher, Prime Minister of the United Kingdom from 1979 to 1990 (Draft notes for her address to the October 1979 party conference. In the end these comments were ignored by her speechwriters)

At the dawn of the 21st century, as reminders of the threat of terrorism become part of daily life for many—whether this be in the form of airport security measures, public transport announcements during the daily commute, or a failed bombing attempt on civilian targets— psychologists have turned their attention back to the darker aspects of human nature that attracted a lot of interest in the years following World War II. Positive psychology and the study of happiness and well-being that were of interest to many in the 1980s and 1990s have given way to renewed attempts to understand what lies behind behaviours that lead people to commit acts of violence against their fellow human beings. Our interest has been in the extremist (fanatical) mindset. A dictionary definition of mindset reads: ―A fixed mental attitude or disposition that predetermines a person's responses to and interpretations of situations‖. In our own previous work, it is defined as ―a pattern of beliefs, feelings, thoughts, and motivations that tends to be mobilised under facilitating conditions‖ (see Stankov, Saucier and Knežević, 2010a). In his foreign-policy speech in Cairo, in June 2009, President Obama carefully avoided using the word ―terrorism‖, preferring ―violent extremism‖ 1, thereby emphasising both cognitive (―extremism‖) and behavioural (―violent‖) features of ―terrorists‖, rather than the effects (―terror‖) of their activities. We distinguish two types of extremism—benign and militant—and the focus of our work is on the latter. Much of our effort to date has been motivated by a desire to establish if the construct of militant extremist mindset (MEM) can be measured and, if so, whether: a) it represents an extreme level of the already known constructs and therefore can be measured in the general population, or; b) it is a characteristic of a small group of people who can be classified as standing outside the ―normal‖ range. Since the evidence (described below) favors the first position, what are the main psychological ingredients of militant extremist mindset? In this chapter we shall provide an overview of the approaches used in our studies of MEM and summarise the main findings. Consistent with Lady Thatcher's viewpoint, our approach to the study of MEM is psychological in nature and implies personal responsibility: the sources of MEM are within the individual. We assume that they can be measured by asking people to agree or disagree with statements that describe one's feelings towards other people, institutions and courses of action. Crucial, of course, is the choice of statements. They have been carefully selected to approximate those endorsed by people who have engaged in or supported violent activities.

1

See Nicholas Lemann's piece in the New Yorker: http://www.newyorker.com/arts/critics/books/2010/ 04/26/100426crbo_books_lemann?currentPage=1#ixzz0o2tHXaTm

Militant Extremist Mindset

133

Similar procedures have been successfully used in studies of personality and related constructs.

Nastiness, Grudge and Excuse Our definition of MEM rests on a series of empirical studies that employ procedures of exploratory and confirmatory factor analysis. The outcomes of these studies have been described elsewhere (see Saucier, Akers, Miller, Stankov & Knežević, 2009; Saucier et al., (in press); Stankov, Saucier & Knežević, 2010a; and Stankov, Higgins, Saucier & Knežević, 2010b). The present chapter draws on the report of Stankov, Saucier & Knežević (under review). We refer to the three main factors that emerged from these studies as 'the ingredients of MEM' and name them Nastiness, Grudge, and Excuse.

a) Nastiness In Stankov et al. (2010a), nastiness is captured by the factor labeled ―War: Justification of violent acts‖. In Stankov et al. (2010b; under review) nastiness is captured by the ―Proviolence‖ factor. The ―War‖ and ―Pro-violence‖ factors identified in our studies are conceptually identical—i.e., they both indicate the acceptance, justification and even advocacy of violence in dealing with enemies. We believe that high standing on this factor is not restricted to those prone to committing terrorist acts. The sub-population of people who have committed criminal acts (including those who have never been caught) is likely to have an elevated standing on this factor as well. In general, people do not agree with statements that capture Nastiness. The arithmetic means on a five-point Likert scale with 1 representing ―strongly disagree are slightly over 2 (disagree). Needless to say, there are group differences in the levels of endorsement, with males and some cultural groups (e.g., Korea, China) standing being somewhat higher than other countries included in our samples. b) Grudge The Grudge factors identified in our studies to date differ in terms of specificity. In Stankov et al. (2010b) this factor is somewhat narrow and it is labeled as ―West: Sins of the Western nations‖. Two themes are running through the items that define this factor. First, the West is seen as an aggressor that has committed violence against other countries in the world, implying that revenge against the West is acceptable. As I point out below, its narrow nature derives at least in part from the preponderance of Islamic groups in contemporary writings that reflect militant extremism. Second, the West is seen as morally rotten, and the statements that fall into this category can be interpreted as attempts to demonise the enemy. In Stankov et al. (2010a; under review) this grudge factor is not limited to the West and it reflects general dissatisfaction with the conditions in the world today (e.g., the world is heading for destruction and that human race is facing calamity). It is labeled as ―Vile World‖. In Stankov et al. (under review), however, the vileness of the world is also related to aspects of human nature (competitiveness, crisis of values), not to some general feature of the world we live in. Again, a common feature of all three factors identified in our studies is a grudge that may motivate some people to take action. Criminals (e.g., thieves or those committing homicides) differ from nasty ―revolutionaries‖ with respect to Grudge, not in terms of Nastiness. Whilst

134


criminals may hold a grudge, this might be more personal or localized—in contrast to the ―revolutionary‖, whose grudge is cast wider and is more socially oriented—Robin Hoods are rare in our societies. In agreement with the results from our other studies, the means of the Grudge items are around 3 on a five-point Likert scale and gender differences are not pronounced. Participants from two countries (Serbia and Malaysia) score relatively high and those from China score low on Grudge.

c) Excuse The Excuse factor also takes different forms in our studies. All forms of it represent justifications in people‘s own minds of the nasty and violent things they tend to condone. The items classified under the rubric of excuse bring into focus higher moral principles and therefore imply noble motivation behind such acts. In Stankov et al. (2010b) the excuse factor is labeled as ―God: In the name of God‖. In Stankov et al. (2010a) the factor is labeled as ―Divine Power‖. In Stankov et al. (under review) this factor is labeled ―Utopianism‖. Since we see Excuse not to be restricted to religiosity, Utopianism may accommodate MEM groups with atheistic orientation. The three factors of Excuse are conceptually different, more so than factors of Nastiness and Grudge. Nevertheless, even though they invoke different sources of moral authority such as God or some kind of Divine Power or Utopian ideals, the common thread running through all these sources is a justification or an excuse. In the present context, higher moral principles that derive from spiritual power or from an idealistic goal of creating a better world represent an excuse for being proviolent and having a grudge. Again, in agreement with the results from our other studies, the means of the Excuse items are around 3 (neutral) on a five-point Likert scale. Although significant, the size of gender differences on this factor is small and in favour of females—females are slightly more likely to endorse positively statements with religious and idealistic overtones. The highest standing country on this factor in our studies is Malaysia, with Chile and Guatemala closely behind it. Sometimes psychologists develop a theory about a particular construct and proceed with the development of measures of this construct only to find out that correlations between a new measure and some existing constructs are too high to warrant its usefulness vis-a-vis what is already known. In our work with MEM we have taken care to obtain correlations of our new measures and assessments of personality traits, value orientations, social attitudes and social norms all of which at least in theory, may have elements in common with MEM. Our findings are that the elements of Excuse with clear religious orientation (i.e., God, Divine Power and Utopianism) are substantially the same as traditional measures of religiosity. Measures of Nastiness and Grudge, however, cannot be reduced to other psychological constructs. This is not to say that they do not correlate in a meaningful way with other constructs. Thus, Nastiness (Pro-Violence) factor has moderate correlation with Saucier's Betaism (unmitigated self-interest and dismissal of political correctness) while Grudge (Vile World) scale correlates with measures of Psychoticism and Social Cynicism.

THE USE OF INGREDIENTS OF MEM MEASURES One proposed use of these scales is for the detection of the levels of radicalization in the general population, particularly when a society is experiencing social and political upheavals


135

(Stankov et al., 2010b; under review). It is to be expected that current items within each scale may need to be modified or supplemented by new statements appropriate for a particular society. Another area of potential application will be in assessing the effectiveness of currently popular de-radicalization efforts in several, mostly Muslim, countries that have been plagued by terrorism. Group comparisons such as cross-cultural comparisons between and within countries on three ingredient processes of MEM may be useful for an improved understanding of cultural differences. This information can supplement commonly studied differences in noncognitive traits such as personality traits, values and social attitudes and norms. One area that has already shown promise is the similarity of MEM findings and studies of cross-cultural differences in noncognitive constructs other than those that belong to MEM. Two empirical findings are particularly relevant for our purposes here. Stankov (in press) and Stankov, Lee and van de Vijver (under review) show that countries and cultures can be ranked in terms of their standings on the Harshness vs Softness dimension. This dimension is defined by social attitudes like Toughness and Maliciousness on the Harshness side and by measures of personality like Agreeableness and Conscientiousness on the Softness side. Harsh countries are from the Confucian Asian region and soft countries are from Latin America and East Europe. Importantly, the same differences between Confucian Asian countries and Latin America emerge on the Nastiness factor. Thus, there is substantial agreement between studies of MEM and studies of noncognitive constructs that do not involve MEM measures. Stankov (in press) argues that harsh and unforgiving Confucian culture may be able to explain both the high ranking of Confucian countries on achievement tests used in international comparisons (e.g., PISA and TIMSS projects) and also high test anxiety and self-doubt that is also observed in these countries. At the time of this writing, we have no empirical data to show that those convicted for terrorist activities score higher than the average member of the population on any of the three ingredient processes. We assume that this will be the case since the statements included in our surveys reflect the opinions and are sometimes taken verbatim from writings produced by terrorist groups. Even when such important validity data become available, on their own our measures of Nastiness, Grudge and Excuse will not be suitable for screening potential terrorists for the purpose of psychological profiling. Given that our data indicate only small or moderate correlations between the three MEM ingredients, a single MEM score may not be useful for decision making. Future research may suggest that high standing on all three ingredient processes (a profile score) together with additional clinical data may be indicative of problem behaviour or perhaps of psychopathology in need of intervention.

THE POTENTIAL MILITANT EXTREMIST INSIDE EACH OF US? The approach we have employed for the development of our MEM scales—the selection of statements reflecting MEM and their validation among members of the general population—may lead one to the conclusion that ingredients of MEM exist within all of us. While this is true in theory, the real situation is similar to what happens in medical diagnosis of some diseases that rely on assessments of particular conditions like cholesterol levels. Although we all have cholesterol in our blood stream, only those of us with high levels of

136


―bad cholesterol‖ are candidates for intervention. This analogy may be extended further to include additional ―risk factors‖. Thus, the differing mean levels between groups defined by gender or country of origin, and potentially many others, may provide for a better definition of extremism relative to the society or group to which the person belongs. For example, since Malays as a group score high on the Grudge factor, an extreme score on this factor in Malaysia must be much higher than the extreme score on this factor in, say, China where perhaps the average Malay score may be seen as extreme. Similarly, different cut-off scores for extremism may exist for males and females. If such cut-off scores were to be deemed useful, careful further psychometric studies will have to be carried out in future. Given the three ingredients of MEM, it is reasonable to raise the question whether they are equally important for evaluation or whether differential weights should be assigned to each. Of the three, Excuse is perhaps the least important. This is because it seems to imply an after-the-event activity. Furthermore, in its most common version—God or Divine Power— this is mostly a measure of religiosity and since there are so many religious people, some of whom may be fanatic with respect to their religious belief, high standing on the Excuse factor on its own may not be indicative of MEM and may lead to too many false positive classifications. This is not to say that Excuse is useless; rather, it should be used in conjunction with the other two MEM ingredients2. From among the other two constructs, high standing on the Nastiness factor is the most important if we are interested in identifying those who are likely to spring into action. As mentioned earlier, high standing on this factor is conceptually linked to criminal behaviour. Thus, these are the people who are less opposed to the use of violence, see it as acceptable under certain circumstances and, presumably, would see fewer obstacles to the use of force. Nastiness is best thought off as an example of social attitudes, akin to dogmatism, authoritarianism or conservatism and, like these latter constructs, it is related to personality traits and therefore relatively immutable. The third ingredient, Grudge, represents a trigger for action. If a person is high on Nastiness factor perhaps it can be expected that his/her threshold for action will be lowered whatever the Grudge. Stankov et al. (under review) argue also that Grudge is perhaps the most amenable to intervention. Personal Grudge that may be linked to forms of criminal behaviour may be possible to change by clinical interventions. Grudge in the sense used in our work that is more social in nature and is a precursor of terrorist actions in the world today, can perhaps be affected by media and education campaigns that address misconceptions about particular groups or events in social and political life. Terrorists, even suicide bombers, are best seen as rational actors—that is, what they do is explicable in terms of their beliefs and desires—who respond to the set of incentives that they find before them. Perhaps they can be fought by means of using a cool recalibration of their incentives. It may be argued that, among the three factors identified in our work to date only one, Nastiness, represents a true psychological trait. Grudge and Excuse are defined with respect to outside sources and have a pronounced sociological and political component. This means that psychology on its own cannot provide a satisfactory explanation of militant extremist

2

It is possible to take the opposing view to our interpretation of the role of Grudge factor. For example, as pointed out by Dr. S. Morony, this may be seen as the factor that overcomes any niggling doubt that what you are doing might be ―wrong‖. It also makes the group highly cohesive, and presumably silences any voices of dissent.


137

mindset. Here, as is often the case in life, psychology rubs shoulders with other sciences in order to achieve a better understanding of human behaviour.

MEASURES OF MUSLIM OR ANY TYPE OF MILITANT EXTREMIST MINDSET? From the beginning, our approach to assessing MEM was deliberately developed in order to capture the generalist view of the construct (see Saucier et al., 2009). Needless to say, we were not entirely successful in our endeavor and one of our measures of MEM (Stankov et al., 2010b) still has a flavor that links it to the Muslim world. We refer to this version as ―contemporary‖ to distinguish it from a more general and perhaps timeless versions reported in Stankov et al. (2010a; under review). Items in the contemporary version of MEM were developed using linguistic analysis in the following way. Over the internet we collected texts written by, or about, terrorist groups (target) and also control texts written by members of legitimate political parties in countries where terrorist groups originate. We then carried out linguistic/statistical analyses to determine what words best distinguish target texts from the control texts. The first column in Table 1 contains the first fifteen most critical words. As can be expected, contemporary target texts contain a number of Muslim words (Allah, Muhammad, Islamic, Islam, Muslims, Jihad). Several words (e.g., I, we, you, our) are also not linked to terrorism. In addition to the fifteen listed words, close to 500 additional words did show significant differences between the corpora. We used an extended list of words to select the fifteen critical MEM related words also listed in Table 1). These latter words were then used to select the target statements for the final version of the scale. As can be seen in the second column of Table 1, the sentences chosen contained words that were less Islamic in nature (―Martyr‖ perhaps being an exception). Nevertheless, despite our efforts some statements that survived psychometric analyses and are now included in measures of MEM still have an Islamic flavor. The most notable are two statements that employed ―West‖ as the key word (―It has become clear that the West has an unspeakable hatred for Islam‖ and ―Western leaders have forced their people to believe that Islam is the cancer of the world, thus it must be eliminated‖) and retained ―Islam‖ in the stem. The remaining 22 statements that are a part of the MEM measure in Stankov et al. (2010b) do not make references to Muslims or Islamic themes. The issue that still needs to be addressed is broadness of the measures of MEM. This is related to the frequently discussed distinction between domain specificity and domain generality. Is it better to have statements that are indicative of general MEM or is it better to include statements that reflect the views of a targeted terrorist group? Closely related, of course, is the question of practical vs theoretical utility of the MEM measures. As mentioned above, we have opted for generality/theoretical utility which, in the long run may work better. Practical needs of granting agencies, however, may on occasion dictate the development of more specific measures in future. Of course, a combination of both specific and general approaches may be the optimal solution.

138


Table 1. Most frequent words that distinguish contemporary terrorist writings from language used by legitimate political parties & words used to select statements (Stankov, Higgins, Saucier & Knežević, 2005b) Most Frequent Words in Terrorist Writings 1. Terrorist 2. Terrorism 3. Allah 4. We 5. Muhammad 6. You 7. Suicide 8. I 9. Islamic 10. Islam 11. Our 12. Will 13. Muslims 14. Jihad 15. Kill

Selected Words Bomb Homeland Peace Brother Kill Terror Enemy Leader Terrorism Force Martyr War God (Allah) Path West

WORK STILL TO BE DONE Our approach to the development of MEM measures is frankly empirical. We collected our statements with no theoretical preconceptions about the possible nature of militant extremism. This may be seen both as an advantage and otherwise. The outcome of our work appears meaningful and perhaps a coherent psychological theory that elaborates on our findings may be developed. Of particular importance may be the need to arrive at a better understanding of the Grudge ingredient of MEM. What aspects of our social life makes people angry? And why? In other words, a motivational theory about MEM is needed. Another theoretical aspect is the relationship of the ingredient processes of MEM and a variety of noncognitive constructs that are of interest to social psychologists. Some of these have been related to MEM but many remain unexplored. For example, we know little about the relationship of MEM to measures of individualism and collectivism, social axioms, moral foundations, tightness-looseness and many others. Another area that needs to be covered is the relationship of MEM to demographic variables and especially to criminal behaviour. If, as we believe, Nastiness captured by the MEM scales is similar to that of incarcerated individuals, this will need to be shown empirically. Relatedly, we need to know what categories of convicted criminals agree or do not agree with measures of Grudge and Excuse. Furthermore, while Nastiness has been essentially the same across our studies, Grudge and Excuse have sub-components. It may be useful in future to develop measures of these latter two MEM ingredients that have not been covered in our previous work. For example, the mindset of eco-terrorists may be poorly captured by our current measures.


139

Finally, while we know that there are cross-cultural differences on measures of MEM we still do not know whether these measures will be sensitive enough to detect changes in the overall levels of radicalization in the population. Given that this is one of the main intended uses of our instruments, we need additional data to establish the sensitivity to changed social circumstances. This aspect of the work may be especially hard to accomplish since the effects can be examined only if an extraordinary event that shakes the society at its core were to take place in a particular country.

CONCLUSION This chapter summarises the work carried out by our team over the past few years. The main aim of this research effort was to develop an instrument that can be used for the assessment of militant extremist mindset in the population. Three different rationales and procedures were employed for the selection of items, and two empirical studies based on samples from several countries were conducted in order to examine psychometric properties of the scales. In the outcome, three different ingredient processes of MEM were identified: Nastiness (condoning and advocating violence), Grudge (dissatisfaction with the existing conditions of the group one belongs to) and Excuse (justification of violence in the name of some higher moral power). Nastiness is a psychological trait that is best understood as belonging to the category of social attitudes, similar to dogmatism and authoritarianism. Like most social attitudes, Nastiness is likely to be resilient to attempts to modify it and it is not only characteristic of terrorists but is likely to be present to a significant degree in the minds of persons convicted of criminal activity. The ingredient of MEM we refer to as Grudge has at least two but potentially more than two facets. One of these is the tendency to apportion the blame for contemporary ills in the world to the West or, relatedly, that West is sinful. The other aspect of the Grudge factor is the perception that the world is rotten and devoid of moral principles—i.e., Vile World. Grudge is not a pure psychological trait like Nastiness. Its impact derives from the interaction with the external world and it contains sociological and political overtones. The third ingredient of MEM, the Excuse, is perhaps the least interesting because its main incarnation – religiosity – is not a new construct. The other two subcomponents of Excuse—Divine Power and Utopianism—are approaching separation from God but, in the final analyses, do not change its interpretation. Our measures of MEM are general not group-specific even though a contemporary version of the survey (Stankov et al., 2010b) that is based on linguistic analyses of the published material from different terrorist groups contains statements that relate to Muslim culture. We believe that potential users of the scale will require both generalist and groupspecific versions of the scales. Future work on MEM should focus on providing validity evidence for the measures and, in particular, on their concurrent validity by correlating MEM with measures of terrorist (and criminal) behaviour. Primary uses of our militant extremist mindset measures are anticipated to be in studying general populations not aberrant groups. The main focus may be either on cross-cultural comparisons or on the detection of changes in attitudes and belief systems following dramatic events that affect the society.

140


REFERENCES Saucier, G., Akers, L. G., Miller, S. S., Stankov, L., & Knežević, G. (2009). Patterns of thinking in militant extremism. Perspectives on Psychological Science, 4, 256-271. Saucier, G., Markowitz, E. de Baessa, Y., Gurnakova, J., Knežević, G., Lee, J., Mastor, K. A., Preiss, D. M., Stankov, L., & Zhou, X. (in press). Comparing the cross-cultural invariance of dual and triadic conceptions of religious/spiritual beliefs. Journal of Crosscultural Psychology. Stankov, L. (in press). Unforgiving Confucian culture: A breeding ground for high achievement, anxiety and self-doubt. Learning and Individual Differences. Stankov, L., Saucier, G. & Knežević, G. (2010a). Militant extremist mindset: Pro-Violence, Vile World, and Divine Power. Psychological Assessment, 22, 70-86. Stankov, L., Higgins, D., Saucier, G. & Knežević, G. (2010b). Contemporary militant extremism: A linguistic approach to scale development. Psychological Assessment, 22, 246-258. Stankov, L., Lee, J. & van de Vijver, F. (under review). Country-level dimensions of cultural differences. Stankov, L., Saucier, G. & Knežević, G. (under review). Militant extremist mindset and the assessment of radicalization in general population.



Chapter 12

EMOTIONAL INTELLIGENCE AMONG CHURCH LEADERS: APPLYING THE SCHUTTE EMOTIONAL INTELLIGENCE SCALE WITHIN NEWFRONTIERS Leslie J. Francis1,, Andrew Ryland2 and Mandy Robbins1 1

Warwick Religions and Education Research Unit, Institute of Education, University of Warwick, Coventry, England CV4 7AL, United Kingdom 2 Beacon Church Runnymede, Church Office, 12 The Sainsbury Centre, Chertsey Surrey, England KT16 9AG, United Kingdom

ABSTRACT The present project introduced the notion and measurement of emotional intelligence to research among church leaders. A sample of 154 individuals serving in a range of leadership roles within local churches associated with the Newfrontiers network completed the Schutte Scale of Emotional Intelligence in order to address three research questions. In response to the first question, the data supported the overall internal consistency reliability of the instrument among church leaders. In response to the second question, the data demonstrated that these church leaders recorded lower mean scores of emotional intelligence than those recorded by the participants in Schutte‘s foundation study. In response to the third question, the data drew attention to those aspects of emotional intelligence on which church leaders demonstrated low levels of endorsement, including in particular certain aspects of non-verbal communication. Keywords: Church leaders, Britain, Emotional Intelligence, Psychology, Religion



Ph: 44-0-24-7652 2539; Fax: 44-0-24-7657 2638; [email protected]

142

Leslie J. Francis, Andrew Ryland and Mandy Robbins

INTRODUCTION The notion of emotional intelligence is a relatively recent and highly attractive construct which has been both conceptualised and operationalised in a variety of ways. For example, in the back cover blurb of his pioneering book, Emotional Intelligence, Goleman (1995) offered the following powerful overview of the notion: Emotional intelligence includes self-awareness and impulse control, persistence, zeal and motivation, empathy and social deftness. These are the qualities that mark people who excel: whose relationships flourish, who are stars in the workplace.

In his subsequent publication, Working with Emotional Intelligence, Goleman (1998) argued as follows: Emotional intelligence refers to the capacity for recognising our own feelings and those of others, for motivating ourselves, and for managing emotions well in ourselves. It describes abilities distinct from, but complementary to, academic intelligence—the purely cognitive capacities measured by IQ. Many people who are book smart but lack emotional intelligence end up working for people who have lower IQs than they but who excel in emotional intelligence skills (p.327, italics as in the original).

Goleman‘s (1998, pp. 26-27) classification of the components of emotional intelligence falls into two broad categories: personal competence and social competence. Personal competence includes self-awareness, self-regulation, and motivation. Social competence includes empathy and social skills. In their popular book, Executive EQ: Emotional intelligence in leadership and organisations, Cooper and Sawaf (1997) related emotional intelligence to four key components: emotional literacy, which includes knowledge of one‘s own emotions and how they function; emotional fitness, which includes emotional hardiness and flexibility; emotional depth, which involves emotional intensity and potential for growth; and emotional alchemy, which includes the ability to use emotion to spark creativity. In their review of the construct, Mayer and Salovey (1997) proposed the following four components: perception, appraisal, and expression of emotion; emotional facilitation of thinking; understanding, analyzing and employing emotional knowledge; and reflective regulation of emotions to further emotional and intellectual growth.

MEASURES OF EMOTIONAL INTELLIGENCE Initial measures of emotional intelligence proposed somewhat lengthy and multi-faceted instruments. For example, the Bar-On Emotional Quotient Inventory (Bar-On, 1996), employed 133 items to generate 15 distinct scales, proposing measures of emotional selfawareness, assertiveness, self-regard, self-actualisation, independence, empathy, interpersonal relationships, social responsibility, problem solving, reality testing, flexibility, stress tolerance, impulse control, happiness, and optimism. The Style in the Perception of Affect Scale (Bernet, 1996) employed 93 items to distinguish preferences for three different styles of

Emotional Intelligence among Church Leaders

143

responding to feeling, defined as body-based, evaluation-based and logic-based perception of affect. A constructive and user-friendly contribution to the assessment of emotional intelligence was made by the 33-item Emotional Intelligence Scale proposed by Schutte, Malouff, Hall, Haggerty, Cooper, Golden, and Dornheim (1998). This instrument was grounded in the work of Salovey and Mayer (1990) who suggest that emotional intelligence comprises three categories of adaptive abilities: appraisal and expression of emotion; regulation of emotion; and utilisation of emotions in solving problems. The 33 items were selected from an original pool of 62 items by means of factor analysis conducted on data provided by 346 participants recruited from a variety of settings in a metropolitan area in the south eastern United States. They generated an alpha coefficient of .90. The construct validity of Schutte‘s measure was established alongside a range of other instruments, including the Toronto Alexithymia Scale (Taylor, Ryan, & Bagby, 1985), the Attention, Clarity and Mood Repair subscales of the Trait Meta Mood Scale (Salovey, Mayer, Goldman, Turvey, & Palfai, 1995), the Optimism subscale of the Life Orientation Test (Scheier & Carver, 1985; Marshall, Wortman, Kusulas, Hervig, & Vickers, 1992), the Zung Depression Scale (Zung, 1965), the openness to experience scale of the Revised NEO Personality Inventory (Costa & McCrae, 1992a, 1992b), and the Barratt Impulsiveness Scale (Patton, Stanford, & Barratt, 1995). The mean scale scores also demonstrated predicted differences between men and women (with women recording higher scores of emotional intelligence), and predicted differences between psychotherapists and prisoners (with psychotherapists recording higher scores of emotional intelligence).

EMOTIONAL INTELLIGENCE AMONGST CHURCH LEADERS Against this background the objective of the present study is to assess the application of the Schutte Emotional Intelligence Scale in a study among church leaders. This application will allow three specific research questions to be addressed. The first research question concerns assessing the internal consistency reliability of the instrument among a well-defined population for whom no data currently exist on its psychometric properties. Do the items cohere among a highly religious population in a similar way to that in which they function among other populations? The second research question concerns comparing the mean scale scores of emotional intelligence recorded by church leaders with the scores recorded by other groups: do church leaders (as individuals engaged with ministry among people) overall display higher levels of emotional intelligence? The third research question concerns profiling the emotional intelligence characteristics displayed by church leaders: does close inspection of the ways in which church leaders respond to the emotional intelligence survey provide indicators that may be helpful in shaping educational programmes concerned with their continuing personal and professional development?

144


Table 1. The Schutte Emotional Intelligence Scale: Item endorsement & item rest-of-test correlations r I know when to speak about my personal problems to others When I am faced with obstacles, I remember times I faced similar obstacles and overcame them I expect that I will do well on most things I try Other people find it easy to confide in me I find it hard to understand the non-verbal messages of other people* Some of the major events of my life have led me to re-evaluate what is important and not important When my mood changes, I see new possibilities Emotions are one of the things that make my life worth living I am aware of my emotions as I experience them I expect good things to happen I like to share my emotions with others When I experience a positive emotion, I know how to make it last I arrange events others enjoy I seek out activities that make me happy I am aware of the non-verbal messages I send to others I present myself in a way that makes a good impression on others When I am in a positive mood, solving problems is easy for me By looking at their facial expressions, I recognise the emotions people are experiencing I know why my emotions change When I am in a positive mood, I am able to come up with new ideas I have control over my emotions I easily recognise my emotions as I experience them I motivate myself by imagining a good outcome to tasks I take on I compliment others when they have done something well I am aware of the non-verbal messages other people send When another person fells me about an important event in his or her life, I almost feel as though I have experienced this event myself When I feel a change in emotions, I tend to come up with new ideas When I am faced with a challenge, I give up because I believe I will fail* I know what other people are feeling just by looking at them I help other people feel better when they are down I use good moods to help myself trying in the face of obstacles I can tell how people are feeling by listening to the tone of their voice It is difficult for me to understand why people feel the way they do* alpha *

These items were reverse coded to generate the item rest-of-scale correlations.

r = correlations between individual items and the sum of the remaining items. f = sum of agree and agree strongly responses.

.43 .37

agree % 80 82

.14 .38 .49

64 67 24

.37 .14 .29 .38 .21 .30 .28 .29 -.06 .53 .16 .30 .50

83 50 41 82 66 48 31 75 72 44 59 71 80

.42 .15 .09 .34 .31 .32 .53

62 72 49 70 48 90 68

.38 .27 .21 .49 .48 .31 .40 .44 .81

40 14 4 40 74 43 67 14


145

METHOD Sample A total of 262 questionnaires were distributed to individuals serving in leadership roles within local churches associated with the Newfrontiers network, including elders, staff, volunteer leaders and highly committed members sharing in leadership. The 154 respondents represented the satisfaction response rate of 59%. Of these respondents, 68 were male, 84 were female, and 2 failed to disclose their sex; 15 were under the age of thirty, 27 were in their thirties, 49 were in their forties, 37 were in their fifties, 17 were in their sixties, 7 were aged seventy or over, and 2 failed to disclose their age.

Measure Emotional intelligence was assessed by the 33-item Emotional Intelligence Scale proposed by Schutte, Malouff, Hall, Haggerty, Cooper, Golden, and Dorheim (1998). Each item was assessed on a five-point scale: agree strongly, agree, neither agree nor disagree, disagree, and disagree strongly.

RESULTS Table 1 presents the scale properties for the Schutte Emotional Intelligence Scale in terms of the item endorsement (sum of the agree strongly and agree responses) and the item rest-oftest correlations (the correlations between each item and the sum of the remaining items), together with the alpha coefficient. The first conclusion to be drawn from these data is that the Schutte Emotional Intelligence Scale demonstrates satisfactory internal consistency reliability: the alpha coefficient of .81 is well in excess of the threshold of acceptability of .65 proposed by DeVellis (2003). The item rest-of-test scale correlations, however, demonstrate that not all items contribute well to a unidimensional scale: 6 of the 33 items record an item rest-of-scale correlation below the level of .20. Internal consistency reliability could be enhanced by dropping some of these items, but only at the cost of reducing the breadth of the way in which emotional intelligence was conceptualised by the test constructors. Table 2. Mean scores of emotional intelligence for church leaders compared with Schutte’s standardisation data

Male Female

N 68 84

Leaders mean SD 116.62 10.65 120.41 10.56

N 111 218

mean 124.78 130.94

Norms SD t 16.52 -4.017 15.09 -6.848

p< .05 .05

146


Table 2 presents the mean scale scores of emotional intelligence recorded by male and by female church leaders considered separately Although there are no recognised normative data published for this scale, the mean scores published by the test constructors can at least serve as a helpful reference point. These scores were generated from a heterogeneous sample of individuals recruited from a variety of settings in a metropolitan area in the south eastern United States, including university students, psychotherapists and prisoners. In line with the sex differences reported by the test constructors (Schutte, Malouff, Hall, Haggerty, Cooper, Golden, & Dornheim, 1998), female church leaders recorded higher scores of emotional intelligence in comparison with male church leaders. Table 2 also demonstrates that both male and female church leaders recorded significantly lower scores of emotional intelligence in comparison with the men and women who comprised Schutte‘s sample. A second point of reference can be provided by the study conducted by Zeng and Miller (2003) among 594 undergraduates attending an introductory psychology course at Midwestern University. Although Zeng and Miller (2003) did not report scores separately for men and women, their mean score of 126.04 (SD = 14.8) falls midway between Schutte‘s standardisation of 124.78 for males and 130.94 for females.

EMOTIONAL INTELLIGENCE PROFILE OF CHURCH LEADERS Given the finding that church leaders may not be displaying particularly high levels of emotional intelligence, it is constructive now to return to table 1 in order to chart the profile of church leaders against their responses to the individual items comprising the scale. At least 80% of the church leaders believe that they possess the following positive characteristics: 90% compliment others when they have done something well; 83% recognise that some of the major events of their life have led them to re-evaluate what is important and what is not important; 82% are aware of their emotions as they experience them; 82% say that, when they are faced with obstacles, they remember times when they faced similar obstacles and overcame them; 80% feel that they know when to speak about their personal problems to others; and 80% recognise the emotions people are experiencing by looking at their facial expressions. Over two-thirds of the church leaders arrange events that others enjoy (75%); help other people feel better when they are down (74%); seek out activities that make them feel happy (72%); are able to come up with new ideas when they are in a positive mood (72%); find that solving problems is easy for them when they are in a positive mood (71%); easily recognise their own emotions as they experience them (70%); are aware of the nonverbal messages other people send (68%); believe that other people find it easy to confide in them (67%); and can tell how people are feeling by listening to the tone of their voice (67%). Fewer than half of the church leaders, however, are able to endorse the following characteristics: 31% say that when they experience a positive emotion they know how to make it last; 40% believe that when another person tells them about an important event in his or her life, they almost feel as though they have experienced the event themselves; 40% know what other people are feeling just by looking at them; 41% believe that emotions are one of the things that make life worth living; 43% use good moods to help themselves keep trying in the face of obstacles; 44% are aware of the non-verbal messages they send to others; 48% like


147

to share their emotions with others; 48% motivate themselves by imagining a good outcome to tasks they take on; and 49% feel that they have control over their emotions.

DISCUSSION The present paper set out to address three research questions. The first research question concerned assessing the internal consistency reliability of the Schutte Emotional Intelligence Scale among a group of church leaders. Two main findings emerged from the research data in respect of this question. On the one hand, the alpha coefficient of .81 demonstrated that overall the 31-item Likert-type instrument achieved a sufficiently satisfactory level of internal consistency reliability to commend its application for research among church leaders. On the other hand, the item rest-of-scale correlations drew attention to one item that completely failed to correlate with the sum total of the other 32 items (r = -.06) and five further items that failed to achieve correlations with the sum total of the other 32 items in excess of .2. The item that completely failed to correlate with the sum of the other items was this ‗I seek out activities that make me happy‘. The poor performance of this item could reflect the reluctance of church leaders to seek out personal hedonism in contrast with their commitment to serve God and to serve other people. In this sense the item might function in a distinct manner among a highly religious constituency compared with a broader population survey. The other five items with relatively low correlations may simply reflect the breadth of the construct of emotional intelligence being accessed by the Schutte Scale. While the psychometric properties of the instrument could be improved by the omission of these items, such improvement would only be achieved at the costs of narrowing the width of the construct and of losing comparability with other studies that have used this instrument. On balance, therefore, the recommendation is to commend the application of the instrument among church leaders in its original form. The second research question concerned comparing the mean scale scores of emotional intelligence recorded by church leaders with the scores recorded by other groups. The main finding emerging from the research data in respect of this question was unambiguous. Taking the mean scale scores of the 218 women and 111 men who participated in the foundation study (Schutte, Malouff, Hall, Haggerty, Cooper, Golden, & Dornheim, 1998) as normative data, both male and female church leaders reported significantly lower levels of emotional intelligence. This finding remains vulnerable in light of the non-representative nature of the sample on which the foundation study was based. Nonetheless, the direct comparison made between the two studies raised a proper question regarding the level of emotional intelligence displayed by church leaders. The third research question concerned profiling the emotional intelligence characteristics displayed by church leaders in terms of the item endorsement given to the 33 individual items that comprise the Schutte Scale of Emotional Intelligence. The detailed findings to emerge from the research data in respect of this question provide those who have responsibility for the continuing personal and professional development of church leaders with a detailed agenda on which to work. For example, these data draw particular attention to the difficulties church leaders may experience with non-verbal communication. Well under half of those in

148


the present sample were aware of the non-verbal messages they send to others (44%). It is likely that skills in these areas could be enhanced by effective educational programmes.

CONCLUSION Overall, the present study has demonstrated the value of introducing the notion and measurement of emotional intelligence to programmes of research among church leaders. There are, however, three main limitations with the present study that need to be addressed by future research. The first main limitation concerns the sample, namely a relatively small number of diverse church leaders within one well-defined church context. What is required now are larger samples of clearly defined church leaders recruited from other denominational contexts. The second main limitation concerns the reliance on just one measure of emotional intelligence. Given the diverse ways in which this construct has been conceptualised and operationalised, there would be clear value in some future research among church leaders employing the Schutte Scale of Emotional Intelligence (to ensure comparability with the present study) alongside one or two other well-chosen measures of emotional intelligence (to extend the present study). The third limitation concerns the way in which the present study has discussed emotional intelligence in isolation from other related psychometric constructs, including, for example, indices of personality and empathy (to link with theories shaped by the psychology of religion) or indices of theological belief (to link with themes shaped by empirical theology).

REFERENCES Bar-On, R. (1996). The Emotional Quotient Inventory (EQ-i): A test of emotional intelligence. Toronto: Multi-Health Systems. Bernet, M. (1996). Emotional intelligence: Components and correlates. Toronto: Annual Convention of the American Psychological Association. Cooper, R. K., & Sawaf, A. (1997). Executive EQ: Emotional intelligence in leadership and organizations. New York: Grosset/Putnam. Costa, P. T., & McCrae, R. R. (1992a). Normal personality assessment in clinical practice: The NEO Personality Inventory. Psychological Assessment, 4, 5-13. Costa, P. T., & McCrae, R. R. (1992b). Revised Personality Inventory (NEO PI-R) and NEO Five-Factor Inventory (NEO-FFI): Professional manual. Odessa, FL: Psychological Assessment Resources. DeVellis, R. F. (2003). Scale development: Theory and applications. London: Sage. Goleman, D. (1995). Emotional intelligence. New York: Bantam Books. Goleman, D. (1989). Working with emotional intelligence. London: Bloomsbury. Marshall, G. N., Wortman, C. B., Kusulas, J. W., Hervig, L. K., & Vickers, R. R. (1992). Distinguishing optimism from pessimism: Relations to fundamental dimensions of mood and personality. Journal of Personality and Social Psychology, 62, 1067-1074.


149

Mayer, J. D., & Salovey, P. (1997). What is emotional intelligence? In P. Salovey & S. Sluyter (Eds.), Emotional development and emotional intelligence: Educational implications (pp. 3-31). New York: Basic Books. Patton, J. H., Stanford, M. S., & Barratt, E. S. (1995). Factor structure of the Barratt Impulsiveness Scale. Journal of Clinical Psychology, 51, 768-774. Salovey, P., & Mayer, J. D. (1990). Emotional intelligence. Imagination, Cognition and Personality, 9, 185-211. Salovey, P., Mayer, J. D., Goldman, S. L., Turvey, C., & Palfai, T. P. (1995). Emotional attention, clarity, and repair: Exploring emotional intelligence using the trait Meta-Mood Scale. In J. W. Pennebaker (Ed.), Emotion, disclosure and health (pp. 125-154). Washington, D.C.: American Psychological Association. Scheier, M. F., & Carver, C. S. (1985). Optimism, coping and health: Assessment and implications of generalized outcome expectancies. Health Psychology, 4, 219-247. Schutte, N. S., Malouff, J. M., Hall, L. E., Haggerty, D. J., Cooper, J. T., Golden, C. J., & Dornheim. L. (1998). Development and validation of a measure of emotional intelligence. Personality and Individual Differences, 25, 167-177. Taylor, G. J., Ryan, D., & Bagby, R. M. (1985). Toward the development of a new self-report alexithymia scale. Psychotherapy and Psychosomatics, 44, 195-199. Zeng, X., & Miller, C. E. (2003). Examination of measurements of emotional intelligence. Erogemetika, 3, 38-49. Zung, W. K. (1965). A self-rating depression scale. Archives of General Psychiatry, 12, 6370.



Chapter 13

OMINOUS SIGNS: SCHIZOTYPY AND RELIGIOUS ATTACHMENT Lisa Hancock School of Psychology, University of New South Wales, Sydney, NSW 2050, Australia

ABSTRACT Considering the importance of religious factors in the psychosis spectrum, the unexplored relationship between religious attachment (RA) and schizotypal traits is of strong theoretical and clinical interest. This chapter presents novel research linking these traits in a broad undergraduate Christian sample. Three hierarchical regressions found that RA was a significant positive predictor of cognitive and interpersonal but not disorganised schizotypy traits. There was a specific relationship between anxious RA and cognitive schizotypy, replicating prior findings in a narrow Evangelical sample. Adult peer attachment and neuroticism appeared to be involved in these relationships. The finding highlights RA as a promising new avenue for schizotypy research which may drive important theoretical and clinical developments.

Keywords: Attachment, Christianity, Psychosis, Religion, Schizotypal personality

INTRODUCTION In delusional insanity, paranoia, as they sometimes call it, we may have a diabolical mysticism, a sort of religious mysticism turned upside down. The same sense of ineffable importance in the smallest events, the same texts and words coming with new meanings, the same voices and visions and leadings and missions, the same controlling by extraneous powers; only this time the emotion is pessimistic: instead of consolations we have desolations;




152

Lisa Hancock the meanings are dreadful; and the powers are enemies (William James, 1960/1902, The Varieties of Religious Experience, p.106).

For the psychosis-prone religion seems to be a double-edged sword. On one hand, the terrifying religious delusions to which James alludes represent a pathological extreme, and even in nonclinical populations, schizotypal or psychosis-like personality traits are associated with religious preoccupation (Diduca & Joseph, 1997) and religious experience (Maltby & Day, 2002). On the other hand, for highly functioning schizotypal individuals, religion can sometimes provide an organising framework which helps to make sense of anomalous experiences, perhaps protecting against the associated distress and breakdown in functioning (Jackson, 1997). Psychosis patients report extensive use of religion to understand their experiences, and it seems that this can be both beneficial and harmful (Mohr, Brandt, Borras, Gillieron & Huguelet, 2006). It makes sense, then to investigate psychological factors which might influence religion‘s function as risk or resource for the schizotypal. Religious attachment is one such factor which has remained unexplored. In a rather radical extension of Hazan and Shaver‘s adult attachment theory (1987), Kirkpatrick and Shaver (1990) observed that relationship with God as described in the Bible has features of parent-child and romantic attachment bonds, and that many religious behaviours are attachment-like. For example, prayer and church attendance look like proximity-seeking and these behaviours increase in times of distress or danger. Supporting the concept of God as an attachment figure, implicit affective and cognitive responses to religious attachment-cues (such as ‗God has abandoned me‘) parallel responses to other attachment cues (such as ‗mother has abandoned me‘) (Mikulincer, Dolev & Shaver, 2004). Self-report measures of religious and peer attachment describe two dimensions of attachment insecurity: anxiety about abandonment, and avoidance of intimacy or dependence; these are associated with negative working models of self and other, respectively (Brennan, Clark & Shaver, 1999). Attachment working models are based on past and present attachment experiences and tend to modestly correspond across parent, peer and religious domains; due to a self-confirming influence on information processing, affective experience, and behaviour they are difficult, but not impossible, to revise (Mikulincer & Shaver, 2009). God may be attractive as a figure who can compensate for unmet attachment needs, but the psychological function of such compensatory religious attachments is complex and not yet fully understood (Granqvist & Kirkpatrick, 2008). Unlike many religious variables, insecure religious attachment (RA) reliably predicts low general and spiritual wellbeing (Beck & McDonald, 2004). Rowatt and Kirkpatrick (2002) found anxious RA predicted negative affect, low positive affect, and neuroticism over and above social desirability and other religious variables and suggested that a stress-buffering function of secure RA may underlie religion‘s generally positive association with mental health. Some support for the stress-buffering function is provided by a prospective study in which secure RA prospectively predicted improved adjustment following surgery of a loved one, an effect mediated by positive religious coping (Belavich & Pargament, 2002). This finding requires further confirmation, however, as no reliable multi-item RA measure was available at that time, and the religion (if any) of participants was not known.

Ominous Signs: Schizotypy and Religious Attachment

153

SCHIZOTYPAL TRAITS Schizotypal traits are normally distributed and predict later psychopathology such as schizotypal personality disorder (Raine & Benishay, 1995). They include social withdrawal or flattened affect (interpersonal schizotypy), odd/eccentric behaviour and beliefs (disorganised schizotypy), and unusual perceptual experiences, suspicious, magical or referential cognition (cognitive-perceptual schizotypy). Despite high heritability, schizotypal trait expression is sensitive to a range of short term psychological and environmental influences (see Raine, 2006 for review). A small but growing body of research has linked schizotypal traits with insecure adult peer attachment (AA). A fairly consistent association between anxious AA and cognitiveperceptual schizotypy traits such as paranoia (Berry et al., 2006; Meins, Jones, Fernyhough, Hurndall & Koronis, 2008; Tiliopoulos & Goodall, 2009) has been interpreted within cognitive models proposing that anxious attachment‘s negative schema of self may contribute to paranoid interpretations of hallucinations and other anomalous experiences (Peters et al., 2006). Others propose that the association stems from common biologically based vulnerabilities such as neuroticism (Tiliopoulos & Goodall, 2009), a variable which has not yet been controlled for in studies of schizotypy and AA. In the only study to investigate religious attachment and schizotypy in any form, Hancock and Tiliopoulos (2010) found that less secure RA and specifically the anxious RA dimension predicted cognitive-perceptual, interpersonal and disorganised schizotypy in Evangelical Christians. The relationship between cognitive schizotypy and anxious RA could not be accounted for by peer attachment or neuroticism. This intriguing pilot finding suggests that schizotypy and RA-related difficulties may share common diatheses or be causally related in some fashion. For example, extending the ideas of Rowatt and Kirkpatrick (2002), secure religious attachment might buffer generally against stress and its exacerbating effect on schizotypy. In the absence of any other research or explicit theory linking the two constructs, however, this pilot finding requires replication. A further limitation of this study was the narrow sample characteristics: participants were all active in campus Christian societies of a conservative Evangelical persuasion, a rather narrow and religiously homogenous population which is also likely to select for religiously active and sociable individuals. Evidence exists that indicates that schizotypal traits and their associations with religious variables vary across different religious and cultural groups (Johnstone & Tiliopoulos, 2008; Lesmana & Tiliopoulos, 2009; Tiliopoulos, & Crawford, 2007; Tiliopoulos & Hancock, unpublished manuscript), so it is questionable whether RA relates to schizotypal traits outside this narrow population. It is of clinical interest to know whether this association applies more generally, and if so, whether this reflects correspondence with insecure adult attachment, or is rooted in more general temperamental co-vulnerabilities, such as neuroticism. Accordingly, the present study aimed to extend knowledge about religious attachment by determining whether RA is related to schizotypy in a broader religious population. It was expected that as for Evangelicals, schizotypal traits will be predicted by less secure RA, and more specifically, by the anxious RA dimension. A secondary aim was to explore the role of two potential covariates, adult attachment and neuroticism.

154

Lisa Hancock

METHOD Participants Participants were 106 first year psychology students who identified themselves as Christian and reported experiencing a relationship with God (an assumption of the RA measure). 60.6% prayed at least weekly. Mean age was 18.95 (SD: 1.55). Two participants did not specify gender; of the remaining 104, 77 (74.0%) were female. There were no other missing values in the dataset. All relevant variables met the assumptions for multivariate analysis.

Materials Schizotypal Personality Questionnaire—Brief (SPQ-B) (Raine & Benishay, 1995): A brief self-report instrument based on the DSM-IV-TR criteria for SPD, assessing cognitiveperceptual (8 items), interpersonal (8 items), and disorganised schizotypy (6 items). Experiences in Close Relationships—Revised (ECR-R) (Fraley, Waller & Brennan, 2000): A self-report measure of adult peer attachment assessing avoidance of intimacy or dependence (18 items) and anxiety about abandonment (18 items). Attachment to God Inventory (AGI) (Beck & McDonald, 2004): A self-report measure based on the ECR-R and assessing two dimensions of RA insecurity: anxiety about abandonment (14 items), and avoidance of intimacy and dependence (14 items). International Personality Item Pool (IPIP) Emotional Stability/Neuroticism subscale (Goldberg, 1999): This subscale consists of 10 items chosen from 2036 IPIP items as having high loadings on the Big Five Neuroticism factor. Higher scores indicate lower emotional stability (i.e., higher neuroticism). The Eysenck Personality Questionnaire—Revised, short version (EPQ-R-S) Lie scale: A valid 12-item index of lying or ‗faking good‘ (Eysenck, Eysenck & Barrett, 1985). Higher scores represent lying.

Procedure Students were invited to participate in an ―Online religion, personality and relationships study‖ for course credit. Participants were blind to the constructs being assessed, and order of presentation of questionnaires was randomised. Demographic and religious indicators were also assessed.

RESULTS Descriptive statistics and reliabilities for continuous variables are summarised in Table 1. The SPQ-B cognitive subscale and EPQ lie scale showed relatively poor reliability and should be treated with caution. For total SPQ-B score, 9 students scored over 17, the 90th


155

percentile cut-off suggesting high risk of SPD in the original US college sample (Raine & Benishay, 1995). Cognitive schizotypy was elevated relative to scale norms. One-way ANOVA found more avoidant RA (large effect of group, ƞ 2 = .23, df = 200) and cognitive schizotypy (moderate effect of group, ƞ 2 = .07, df = 200) than in the prior Evangelical sample, and prayer was less frequent (r = .54, p<.05).

Table 1. Descriptive statistics & reliabilities Scale

Observed α [norms]

M (SD)

.58 [.72]

4.13 (1.91) a

Interpersonal

.69 [.76]

3.04 (2.11)

Disorganised

.71 [.73]

2.65 (1.83)

.82 [.87]

2.53 (.58)

. 78 [.86]

2.92 (.58)

.91 [.91]

3.72 (1.01)

.92 [.94]

3.00 (.92)

Subscale

SPQ.B Cognitive

AGI

Anxiety Avoidance

ECR-R Anxiety Avoidance EPQ

Lie

.58 [.73-.77]

2.66 (2.06)

IPIP

Neuroticism

.82 [.86]

2.86 (.76)

Note. Normative reliabilities are in square brackets. a Mean discrepant from scale norms at a = .005, two-tails.

Table 2. Pearson correlations (upper diagonal) & partial correlations (lower diagonal) controlling for age, gender, & lying 1

2

1. Cognitive 2. Interpersonal 3. Disorganised 4. RA anxiety 5. RA avoidance 6. AA anxiety

.27 .28

**

.29

**

.30

**

-.17 .37

**

4 **

.30

**

.38 **

.38 .15

*

.20

**

.32

**

5 **

.31 .16 .15

.13 .05 **

.36

**

.38

-.16 .36 .17 .31

**

.06 .37

**

-.12 .40

**

.08

11

.26

**

.43

**

-.13 -.02

<|.01| .27

**

-.16 <|.01| -.17

**

.42

**

.26

**

.35 **

.27

8. Neuroticism

.38**

.28*

.44** .26** -.03 .56** .23*

p < .05. **p <.01.

10

.38

.14

.44

11. Lie scale

9 **

.18

.09 .14 .36

8

.19

10. Gender

.02

7 **

-.08 **

6

7. AA avoidance

9. (log of) Age

*

3 **

-.06 .56 .19

**

-.08 -.05

-.15

.02 <|.01| .01

-.10 -.10 -.20

*

-.07 .15

.08

-.09

-.11

.07

.23*

-.83

.18

-.21*

-.09

.21* -.14

156

Lisa Hancock

Table 3. Hierarchical multiple linear regressions predicting schizotypy dimensions Cognitive-perceptual SE β ∆R2 Block1 .11** (.17**) ** * Anxious RA .29 (.42 ) .31 Avoidant RA -.13 (-.04) .31 .08* (.02) Block2 Anxious RA .17 (.33**) .33 Avoidant RA -.18 (-.04) .30 Anxious AA .29** (.20) .20 Β

Avoidant AA Block3 Anxious RA Avoidant RA Anxious AA Avoidant AA Neuroticism

.07 (-.05)

Interpersonal SE β ∆R2 .06* (.10**) * .18 (.23 ) .35 .19 (.16) .35 .17**(.12**) .11 (.09) .35 .11 (.10) .33 .14 (.21) .22 Β

.35** (23*) .22

.20 *

.03 (.00) *

.16 (.32 ) -.15 (-.04) .16 (.19) .06 (-.05) .23* (.03)

.32 .30 .23 .20 .30

Disorganised SE β ∆R2 .03 (.13**) ** .16 (.34 ) .31 .07 (.04) .31 Β

.13**(.06*)

.03 (.21) .01 (.07) .31** (.33**) .15 (-.18)

.32 .30 .20 .20 .08**(.00)

.01 (.00) .11 (.09) .13 (.10) .06 (.21) .35**(.23*) .14 (.01)

.35 .33 .25 .22 .33

.01 (.23) .04 (.07) .12 (.36**) .15 (-.19) .33**(-.08)

.31 .29 .22 .19 .29

Note. Corresponding statistics from an Evangelical sample (Hancock & Tiliopoulos, 2010) are presented in brackets and bolded where regression coefficients were outside the 95% confidence interval for the current sample. The VIF results for all predictors were below 2, indicating absence of multicolinearity. *p < .05; ** p < .01.

Three hierarchical multiple linear regression analyses were conducted to evaluate predictors of schizotypy dimensions (see Table 3). Anxious and avoidant RA were entered in the first step. To explore the role of potential mediators, AA dimensions were entered in the second step, and neuroticism in the third step. As shown in Table 2, age, gender and lying did not correlate with any of the schizotypy or RA dimensions or moderate their inter-relationships, and therefore were not included in the regression model. Overall, RA dimensions accounted for a significant 11% and 6% of the variance in cognitive and interpersonal schizotypy respectively, but only 3% of disorganised. For cognitive schizotypy, there was a unique relationship with Anxious RA. This appeared to be mediated by adult attachment and neuroticism, both of which added to the model. For interpersonal schizotypy, AA added a substantial additional 17% of the variance, and avoidant AA was the strongest predictor in the final model. Neuroticism did not add to the model. For disorganised schizotypy, AA again added to the model, and anxious AA was a moderate positive predictor of disorganised schizotypy. However, this relationship appeared to be mediated by the shared influence of neuroticism, which was the only unique predictor in the final model.


157

DISCUSSION The positive association between schizotypal traits and insecure religious attachment in Christians replicates pilot findings in a narrower Evangelical sample (Hancock & Tiliopoulos, 2010). There were interesting discrepancies in the mediating role of adult peer attachment and Neuroticism relative to the pilot. The pattern of relationship between RA and schizotypal traits in our broad undergraduate sample was generally similar to that in Evangelical society members, but rather weaker. Anxious RA was a unique predictor as expected, but only for cognitive schizotypy, suggesting that Christians with anxious RA are more likely to also experience paranoid ideation, ideas of reference, and/or unusual perceptions. The anxious RA—cognitive schizotypy association replicates findings in Evangelicals and also parallels findings for AA anxiety. RA anxiety may contribute to schizotypy in similar ways to AA anxiety, such as by promoting threat expectancy and more paranoid religious attributions of meaning, potentially bringing together cognitive (Peters et al., 2008) and religious (Jackson, 1997) theories. Unfortunately the SPQ-B does not allow subscales for paranoia and hallucinations to be generated, but future studies using more detailed schizotypy measures could explore whether theories about paranoia development should be extended to include religious attachment attribution biases. It is interesting that AA and Neuroticism appeared to be of greater relative importance in understanding RA‘s role in the current sample. This highlights the importance of attending to the religious characteristics of the sample in studies of RA. Considering the greater religious involvement, lower RA avoidance, and greater frequency of RA-relevant activities such as prayer among the Evangelical society members, one could speculate that in some intensely religious populations the religious attachment bond may be so important to experience and cognition that it overwhelms the role of temperament in the development of cognitive schizotypy. For those who believe God is intimately involved in ordering the details of their lives, and have an anxious RA, almost any every day event might be taken as an ominous sign which activates religious attachment distress. The resultant attachment system hyperactivation could escalate schizotypal tendencies by increasing vigilance for further signs confirming their inability to elicit God‘s care or protection. In contrast, for those with more secure RA a sense of special significance or presence in the everyday could be interpreted as a comforting and calming affirmation of a secure bond with God. Further examinations of RA characteristics across religious subcultures are required, perhaps particularly with regard to a group‘s experiential focus and emphasis on personal relationship with the deity.

INTERPERSONAL SCHIZOTYPY Interpersonal schizotypy was also predicted by insecure RA. The pattern of specific findings was remarkably similar to the pilot finding, with adult attachment but not neuroticism accounting for additional variance, and avoidant AA the most important unique predictor. This is consistent with adult attachment findings (Tiliopoulos & Goodall, 2008). There is certainly conceptual overlap between the schizotypal aloof interpersonal style and avoidant attachment, but causal explanations are also possible: AA could increase relationship

158

Lisa Hancock

distress and exacerbate schizotypy; alternatively, and perhaps more likely, early schizotypal deficits could interfere with development of secure peer attachments. This might then transfer to RA difficulties.

DISORGANISED SCHIZOTYPY Disorganised schizotypy was not predicted by RA and anxious RA was not a unique predictor, in contrast to evangelicals. Adult attachment added to the model, and anxious AA was a moderate positive predictor, just as in the Evangelicals. However, in a broader sample this relationship appeared to be explained by the shared influence of neuroticism, which was the strongest predictor of disorganised schizotypy in clear contrast to its negligible role for evangelicals. Again, possibly RA is of lower relative importance among less intensely religious Christians, but it is more difficult to explain why anxious AA should also be of lower importance relative to neuroticism and avoidant AA a positive rather than negative predictor of disorganisation. It would be interesting to explore whether for some Evangelical society members disorganisation items such as ‗Some people think that I am a very bizarre person‘ (item 6) actually tapped perceived marginalisation by peers on the basis of visible group identity or evangelistic outreach activities on campus, perceptions which might be biased by adult attachment anxiety (negative working model of self). Adult attachment related to schizotypy in expected ways, consistent with prior findings linking interpersonal schizotypy with avoidant AA, and cognitive and disorganised schizotypy with anxious AA. There was modest correspondence across religious and peer domains for attachment anxiety but not for avoidance, in contrast to US norms using the same measures in college and community samples (Beck & McDonald, 2004) but consistent with the prior Australian study (Hancock & Tiliopoulos, 2010). This is puzzling; possibly RA avoidance, or reporting thereof, is more reflective of a religious culture‘s teachings on intimacy than an individual‘s attachment working models, at least in Australia.

DIFFICULTIES IN INTERPRETATION There are several reasons why the present findings were difficult to interpret. Firstly, there is no theory explicitly linking RA and schizotypy to guide interpretation. Even theories linking other forms of attachment and schizotypy rely on cross-sectional data and are not well developed in terms of causal processes. Secondly, the SPQ lacks rigorous validation and norming in homogenous religious populations, so it is not known whether particular religious views or practices confound measurement of schizotypy and hence its association with RA. Distinguishing between normative and pathological religious experiences is not easy, and clinician administered measures may be preferable. The SPQ-B cognitive schizotypy subscale showed rather poor reliability in both samples and in the current sample it was elevated relative to both scale norms and Evangelicals. Perhaps particular cognitive schizotypy items were endorsed on the basis of religious beliefs or experiences which are normative within a more experiential Christian subculture, such as those tapping experiences of guidance or presence.


159

Thirdly, very different relationships with RA might be found in clinical populations, and this needs to be investigated. Related research questions are RA‘s relationships with religious distress, information processing biases, and clinical outcomes in schizotypal patients. If future research finds RA difficulties to be prevalent clinically, promoting or priming secure RA might buffer against religion-related distress and paranoia by encouraging more adaptive interpretation of schizotypal experiences and improving use of religious resources and coping strategies. Attachment difficulties and impairment in relationship functioning with God might translate more naturally to therapeutic intervention than do religious beliefs and values. Furthermore, according to the ‗broaden and build‘ model (Mikulincer & Shaver, 2009), secure RA might also have positive effects beyond the religious domain, such as by enhancing use of social or religious community support or promoting emotion regulation and prosocial behaviour in schizotypal individuals.

CONCLUSION Finally, the terms ‗prediction‘ and ‗mediation‘ should not be taken to suggest that these cross-sectional analyses tested causal association. In early adulthood both schizotypy and religious attachment may be in flux, meaning tests of mediation could be hopelessly biased by unmodelled auto-regressive effects (Maxwell & Cole, 2007), and there is insufficient theory to assume temporal precedence or rule out the possibility of important omitted variables. Although RA‘s status as a predictor tends to be assumed, presumably because of its roots in infant behaviour, both schizotypy and insecure attachment in adults are sensitive to past and present psychosocial experiences, while themselves impairing relationship functioning and information processing. It is therefore entirely possible that schizotypal deficits and cognitions interfere with secure RA development, or that aspects of schizotypy and insecure RA develop in mutual influence with one another and are sustained by their negative effects on peer, parental and religious experiences. Careful longitudinal and experimental studies of RA and schizotypy are clearly required to resolve these fascinating questions. Regardless of these limitations, the discovery of a replicable association between insecure RA and schizotypy among Christians is an important development which challenges us to adapt our understanding of other religious and attachment processes and their relationship to schizotypy. This line of enquiry could help illuminate religion‘s multifaceted functioning within personality and potentially even stimulate new therapeutic approaches for psychosis spectrum disorders.

REFERENCES Beck, R., & McDonald, A. (2004). Attachment to God: The Attachment to God Inventory, tests of working model correspondence, and an exploration of faith group differences. Journal of Psychology and Theology 32, 92-103. Belavich, T. G. & Pargament, K. (2002). The role of attachment in predicting spiritual coping with a loved one in surgery. Journal of Adult Development, 9, 13-29.

160

Lisa Hancock

Berry, K., Wearden, A. J., Barrowclough, C., & Liversidge, T. (2006). Attachment styles, interpersonal relationships and psychotic phenomena in a non-clinical student sample. Personality and Individual Differences, 41, 707-718. Bowlby, J. (1969/1999). Attachment (2nd ed.): Attachment and loss (vol. 1), New York: Basic Books. Brennan, K.A., Clark, C.L., & Shaver, P.R. (1998). Self-report measurement of adult romantic attachment: An integrative overview. In J. A. Simpson & W. S. Rholes (Eds.), Attachment theory and close relationships (pp.46-76). New York: Guilford. Fraley, R. C., Waller, N. G., & Brennan, K. A. (2000). An item-response theory analysis of self-report measures of adult attachment. Journal of Personality and Social Psychology, 78, 350-365. Goldberg, L. R. (n.d.). International Personality Item Pool. Retrieved June 10, 2008 from http://ipip.ori.org/ Hancock, L. & Tiliopoulos, N. (2010). Religious attachment dimensions and schizotypal personality traits. Religion, Mental Health and Culture, 13, 261-265. Hazan, C., & Shaver, P. R. (1987). Romantic love conceptualized as an attachment process. Journal of Personality and Social Psychology, 52, 511-524. Jackson, M. (1997). Benign schizotypy? Mental Health, Religion and Culture The case of religious experience. In G. Claridge, (Ed.), Schizotypy, implications for illness and health (pp 227-250). Oxford:OUP. James, W. (1960/1902). The varieties of religious experience. London: Collins. Johnstone, J. & Tiliopoulos, N. (2008). Exploring the relationship between schizotypal personality traits & religious attitude in an international Muslim sample. Archive for the Psychology of Religion, 30, 241-253. Granqvist, P., & Kirkpatrick, L. A.(2008). Attachment and religious representations and behavior. In J. Cassidy & P. R. Shaver (Eds.), Handbook of attachment theory and research (2nd ed.) (pp.906-933). NY: Guilford Press. Lesmana, C.B.J., & Tiliopoulos, N. (2009). Schizotypal personality traits and attitudes toward Hinduism among Balinese Hindus. Mental Health, Religion and Culture, 12, 773-785. Kirkpatrick, L. A., & Shaver, P. R. (1990). Attachment theory and religion, childhood attachments, religious beliefs, and conversion. Journal for the Scientific Study of Religion, 29, 315-334. Maltby, J., & Day, L. (2002). Religious experience, religious orientation and schizotypy. Mental Health, Religion and Culture, 5, 163-174. Maxwell, S. & Cole, D. (2007). Bias in cross-sectional analyses of longitudinal mediation. Psychological Methods, 12, 23-44. Meins, E., Jones, S. R., Fernyhough, C., Hurndall, S., & Koronis, P. (2008). Attachment dimensions and schizotypy in a non-clinical sample. Personality & Individual Differences, 44, 1000–1011. Mikulincer, M., & Shaver, P. (2009). An attachment behavioural systems perspective on social support. Journal of Social and Personal Relationships, 26, 7-19. Mikulincer, M., Dolev, T., & Shaver, P.R. (2004). Attachment-related strategies during thought-suppression: Ironic rebounds and vulnerable self-representations. Journal of Personality and Social Psychology, 71, 230-244.


161

Mohr, S., Brandt, P. Y., Borras, L., Gillieron, C. & Huguelet, P. (2006). Toward an integration of spirituality and religiousness into the psychosocial dimension of schizophrenia. American Journal of Psychiatry, 163, 1952-1959. Peters, E., Williams, S., Greenwood, K., Kuipers, E., Scott, J. & Garety, P. (2008). Appraisals, psychotic symptoms and affect in daily life. Schizophrenia Research, 98, 180-181. Raine, A., & Benishay, D. (1995). The SPQ-B: A brief screening instrument for schizotypal personality disorder. Journal of Personality Disorders, 9, 346-355. Raine, A. (2006) Schizotypal personality: neurodevelopmental and psychosocial trajectories. Annual Review of Clinical Psychology, 2, 291-326 Rowatt, W. & Kirkpatrick, L. (2002). Two dimensions of attachment to god and their relation to affect, religiosity, and personality constructs. Journal for the Scientific Study of Religion 41, 637-651. Tiliopoulos, N., & Crawford, G. (2007). Three-factor model of schizotypal personality in a British Christian sample. Mental Health, Religion & Culture, 10, 563-569. Tiliopoulos, N., & Goodall, K. (2009). The neglected link between adult attachment and schizotypal personality traits. Personality and Individual Differences, 47, 299-304. Tiliopoulos, N. & Hancock, L. (unpublished manuscript). Religiosity and schizotypal traits in Australian Christians.



Chapter 14

SHAM REASONING, HUMPTY DUMPTY AND THE BURDEN OF PROOF Fiona J. Hibberd School of Psychology, University of Sydney, Sydney, NSW 2006, Australia

ABSTRACT Slife and Reber ask of psychologists that they recognise their prejudice against theism and the incompatibility between theistic and naturalistic worldviews. Yet, the subtext of their paper is that theism and naturalism are equally valid and that psychology‘s secularism is a mistake. Given that theism is not beyond reason, the only sufficient ground for charging psychologists with prejudice is if: (i) theism has survived serious attempts at conceptual and empirical test, and; (ii) psychology ignores or disguises this fact. So, the grounds for believing in the reality of a supernatural existent are highly relevant to the authors‘ allegation. However, their concept of God affords no such grounds. They disavow the logic involved in the ordinary meaning of the term ‗incompatible‘ and they ignore a crucial distinction between conditions of existence and qualities of things or processes. The consequence is that either God is this-worldly and there is no ontological basis to Slife and Reber‘s incompatibility thesis, or God‘s mode of existence is quite different from the mode of existence of ordinary things and all the problems of dualism follow. Either way, their charge of prejudice is without foundation. The burden of proof rests with the defender of theism and, for now, psychology‘s secularism can be considered a mark of the discipline‘s (sometimes questionable) rationality.

Keywords: Dualism, Logic, Rationality, Relations, Theism



Ph: 61-2-9351 2867; Fax: 61-2-9036 5223; [email protected]. From Hibberd (2009a). Copyright © 2009 by the American Psychological Association. Reproduced with permission.

164

Fiona J. Hibberd

INTRODUCTION Slife and Reber‘s (2009) central thesis is that psychologists exercise an unintentional and pervasive prejudice against theism. They think they are bias-free because their discipline remains non-committal on the topic of God. Yet introductory texts omit discussion of theistic concepts, psychologists assume that their findings, explanations, interpretations and theories apply to atheists and theists alike, their methodologies presume atheism, and they ignore the theism present in theories that they seek to appropriate. Therein lies the prejudice and, according to Slife and Reber, it exists because many psychologists continue to incorrectly assume that theism is compatible with naturalism. The authors claim not to object to a secular psychology, but merely to the compatibility assumption and to the lack of any warning given to theists who may wish to access psychological research and services. Slife and Reber‘s paper is marked by conceptual looseness, contradiction and unsupported empirical claims1. However, I shall bypass much of this and go straight to the crux of the matter—the paper‘s subtext. Consider the following statements: there is a ―... subtle discrediting of theism as a serious interpretative and explanatory framework‖ (p. 2); naturalism ―... makes the pre-investigatory assumption that the supernatural is not needed ... in psychological science‖ (p. 4); most psychologists mistakenly think that their naturalistic findings are equally relevant to both theists and naturalists (p. 5); given the incompatibility between theism and naturalism, psychology ―... violate[s] the assumptions and values of the former‖ (p. 6); ―... the systematic biases against theism in psychology are part of the ―social imaginary‖ of the discipline‖ (p. 23); researchers ―... still do not consider ... that God, as one of several constitutive necessary conditions, had something to do with the participants ―images of God‖ (p. 25); ―... many psychologists ... have attempted to understand and apply his [Buber‘s] philosophy of the I-thou to psychology without theism‖ (p. 26). Despite the authors‘ conditional support for a secular psychology (p. 30), the subtext of their paper is that theism is as valid as naturalism and psychologists err in their disregard of the former; psychology’s neglect of theism is a mistake. Slife and Reber do not say this, of course. They are marked by philosophy‘s twentieth century ‗linguistic turn‘, its preoccupation with the supposed futility of metaphysics and classical logic, and postmodernism‘s intrusion into theoretical psychology2. As a consequence, their grip on the concepts of truth, error and rational inquiry is marginal. Instead, their focus is ―... the worldview—or world of meanings in the Heideggerian sense— and not the physical world per se ...‖ (p. 10), thereby avoiding the rather difficult matter of why secular psychology in general, and the atheist psychologist in particular, is wrong. Heidegger‘s theory of meaning involves a conception of language as the universal medium or ‗house of Being‘ (Kusch, 1989). This undermines the aboutness of language and, 1

Of the latter, the following are noteworthy: ―many psychologists ... believe in God‖ (p. 6) and most ―... have conducted their research and formulated their theorizing with this understanding [the compatibility of theism and naturalism]‖ (p. 9). These claims are central to Slife and Reber‘s argument but the reader cannot discern whether they refer only to American psychologists or to some more general class. Moreover, if the authors‘ analysis of research practices in the second half of their paper is right, some psychologists are behaving not at all as compatibilists but, conceivably, as atheists who acknowledge the incompatibilities between naturalism and theism. In fact, psychologists‘ recognition of the incompatibilities is implied by the authors when they invoke this as motivation for the exclusion of theism from psychology (p. 19). 2 Of the twentieth century, Strawson (2000) notes that it was ―... the silliest of the centuries, philosophically speaking‖ (p. 12).

Sham Reasoning, Humpty Dumpty and the Burden of Proof

165

although Slife and Reber provide a definition of theism (its sense), they are not concerned to address whether its key concepts actually refer, i.e., whether what theism purports to be about is true. They recognise that semantic matters are underpinned by ontological assumptions (p. 3), but prefer the unsuccessful strategy of certain social constructionists, viz., frame your analysis in terms of semantics, forget that signifiers must at some stage in the chain of signification be grounded in uninterpreted reality, and remain ontologically mute (e.g., Gergen, 1994). Unsurprisingly, then, the spirit of Slife and Reber‘s paper is relativist. It assumes from the outset that theism and naturalism are equally valid worldviews and it offers the reader no means of evaluating this assumption, i.e., no means of determining its truth or epistemic merit. One is reminded of Peirce‘s observations about the demise of conceptual rigour when philosophy is in the hands of theologians: ―... it is no longer the reasoning which determines what the conclusion shall be, but it is the conclusion which determines what the reasoning shall be. This is sham reasoning‖ (1931-1958, paragraph 57, emphasis added). I appreciate that every argument rests on assumptions that are not always argued for and that it was never the authors‘ intention to address the truth of theism. Their manifest intent is to bring out the incompatibility between the naturalist and theist worldviews and psychologists‘ prejudice against the latter. However, the only sufficient reason for charging psychologists with prejudice is if (i) theism has survived serious attempts at conceptual and empirical test, and (ii) psychology ignores or disguises this fact. It follows that the grounds for believing in the reality of a supernatural existent are highly relevant to their allegation. Slife and Reber might object to (i) because it assumes that faith or ‗knowledge from revelation‘ should be subject to the same probatory considerations (conceptual, empirical) as scientific knowledge-claims. Recall, for example, their (invalid) inference that ―... attempts to use the insights gleaned from a naturalistic worldview to inform or correct the understandings of a theistic worldview constitute a significant prejudice against the latter‖ (p. 3). This may be their attempt to limit the scope of reason. It has all the features of postmodernism‘s late twentieth century criticisms of science: science, as the embodiment of Enlightenment reason, is a presumptuous, dogmatic, grand narrative that seeks to exclude other equally valid ‗voices‘; its belief in a perspectiveless, God‘s-eye view of the world is a conceit. Arguments against the alleged defects of a universalistic conception of rationality (e.g., Siegel, 1987) are typically ignored and the misrepresentation continues—science is just one narrative or system of meanings amongst many. And perhaps we should not be surprised if Slife and Reber do object to (i) because history tells us that when reason and faith are at odds, the theist is more inclined to conclude that something is wrong with reason. The sagacity of John Locke (1706/1924, bk. IV, ch. XVIII) is relevant: ―I find every sect, as far as reason will help them, make use of it gladly; and, where it fails them, they cry out, It is a matter of faith, and above reason‖ (p. 355). However, if the history of philosophy is anything to go by, religion does not belong solely to the province of faith (e.g., Aquinas, Descartes, Leibniz and Berkeley presented what they thought were proofs of God‘s existence), and Slife and Reber do gesture towards the importance of critical inquiry (p. 9). Moreover, the idea that examining any notion of God involves standing outside the theist‘s system of meanings and, thereby, missing the point, is misguided. First, it mistakenly attributes ‗meanings‘ to non-symbolic phenomena, viz., to what people take to be the case. What theists believe is either true or false and such beliefs can, like any other belief, be tested. Second, whilst we cannot stand outside, or above, society and cultural/historical contexts, this does not entail that our statements cannot be universal

166

Fiona J. Hibberd

and either true or false. Whether what we state is true depends on whether the world is as described, not on where we are (spatio-temporally) located when we do the stating (Hibberd, 2005).

A WORD OF CLARIFICATION Before I return to Slife and Reber‘s paper, it may be prudent to make some pre-emptive clarifications. I assume that: 1. as inquirers we are fallible. No knowledge-claim is incorrigible; 2. philosophy and science are neither always rational nor always moral and so are not themselves immune from criticism; 3. reasons can provide justification for various knowledge-claims and philosophical inquiry or conceptual analysis can provide answers that are either true or false; 4. religion is an inherently valuable subject; 5. to comment on theism is not to comment on the broader subject of religion, the importance of texts (such as the Bible or the Quran), nor on how religious practices and interpretations continue to shape our behaviour and values.

THE AUTHORS’ CONCEPT OF GOD Consider what Slife and Reber mean by ‗theism‘. It includes non-natural (supernatural) events and processes ( ―... God for theists‖, p. 4) that are not mere epiphenomena (they ―... exist in a difference-making way ― (p. 4). ―... [A] currently active God (or Gods) is necessary for understanding the world; the world is ―indissolubly connected to God,‖ including the psychological world ... . The good life ―is inconceivable without God‖ ... . ―God is already and always intimately acting in nature which depends from moment to moment ... upon divine activity‖ (p. 4). It is ―... a functionally existent God, one that relates to the events of the world ... in a practical sense‖ (p. 13). God has the ability to ―... act otherwise ...‖ (p. 13) and acts ―... concurrently in nature so that everything that happens is both fully a divine action and a fully natural action‖ (p. 14); ―... God ... [plays] a role in the development of peoples ― images of God (p. 26); ―God ―s constant involvement confers meaning on even the smallest of subatomic events‖ (p. 31). These statements are not to be interpreted as figures of speech. They purport to be about God‘s existence and God‘s activities. Typically (though not always) in philosophy, such statements are said to involve a form of dualism between a supernatural world and the natural world. Although Slife and Reber are aware of the problems that arise from dualism‘s ‗twoworlds‘ argument, they appear to think that these problems are not relevant to their characterisation of God, presumably because they believe it makes no appeal to two worlds. Their explicit appeal is only to a functionally existent God actively involved in this world. Even so, they do invoke the concepts ‗supernatural‘ and ‗divine‘ and, therefore, the logically prior ontological question is: can the natural and supernatural co-exist or co-occur? One interpretation of Slife and Reber‘s concept of God is that the supernatural is a this-worldly


167

kind of process, quite different from natural processes, but one that interacts with them (see p. 14), i.e., the natural and super-natural can co-exist or co-occur. Yet, if this is the case, there are no ontological grounds to their incompatibility thesis. Alternatively, we can interpret the authors as treating the supernatural not as a kind of process but as a kind of existence quite different from existence as it is ordinarily understood. This is consistent with the Christian theists‘ usual understanding of God (as transcendent and independent of the natural world). However, the problem here is that it unwittingly invokes another realm because existence is said to be of a different type. This yields all the difficulties associated with dualism‘s ‗two-worlds‘ argument. Although these difficulties are usually well understood, here they are worth repeating because, if Slife and Reber do mean that God‘s existence is of a different type, they have not (contrary to their belief) avoided dualism. Dualism is the thesis that that some things have a distinct type of existence from others, that there are two realms or two different ways of being. Yet, entities in one realm are said to ‗participate in‘, ‗imitate‘, ‗share in‘, or ‗interact with‘ entities in the other. The consequence is that the original distinction cannot be maintained. Passmore (1970) refers to this as ‗the Humpty Dumpty argument‘ because: ... once we break up any system in a certain kind of way, it becomes quite impossible to put the pieces together again in a single situation: and yet, unless they can be so put together, the whole point of the breaking-up is lost (p. 40).

Opposing dualism, then, is not to endorse a monism that makes all differences unreal. It is to endorse only a monism about existence. That is, there are not types, classes, levels, modes or orders of existence because ‗existence‘ is not a predicate3. It makes no sense to say, for example, ‗some kangaroos exist and some do not‘. Contra Heidegger, then, existence is not a characteristic shared or possessed by existing objects (Edwards, 2004). Existence is not a quality or process. There are many different kinds of things or processes, but only a single way of being or existing. This is a pluralistic conception of a single reality. If God‘s mode of existence is different from the natural mode of this world, then God, as non-natural, is not of this world. Another realm or world has been invoked, albeit unwittingly. The difficulty now lies in explaining the interaction between realms or modes or worlds. There has to be some kind of relation between God (not located in this world) and God‘s effects (which are said to occur in a ‗difference-making way‘ in this world). Regardless of what that relation is, if it obtains in the non-natural world (God‘s realm) then so must the other term of the relation, God‘s effects. This is because items or entities have to be located in the same realm for there to be a connection between them. It is not logically possible (and, therefore, not empirically possible either) for a connection or relation between God and God‘s effects to obtain across different realms or different orders of existence. To be connected is to be in a spatio-temporal relation within the one realm of being. The upshot is that if God‘s effects are real, God is not transcendent or supernatural because transactions between the natural and supernatural are a logical impossibility. The distinction between the natural and the supernatural, as different modes of existence, goes the same way as the distinction between mind and body or that between universals and 3

Existence is signified in the use of the copula ‗is‘. ‗Is‘ is an occurrence or something being the case. Logically, there is no alternative to either ‗is‘ or ‗is not‘ or to either ‗being‘ or ‗not being‘, i.e., there are no types or modes of existence.

168

Fiona J. Hibberd

particulars—it collapses4. Alternatively, there is no God, just as there is no Cartesian mind and no pure universals. Passmore (1970) notes that ―… the rejection of dualism is indeed one of the few points on which almost all the creative philosophers of modern times have agreed‖ (p. 38). Perhaps, then, we should return to the possibility that God is this-worldly and immanent. Such a notion contradicts Slife and Reber‘s incompatibility thesis and I imagine that, to many theists, a thisworldly God is no God at all. Yet it is worth more attention because the authors‘ concept of God is sometimes pantheistic. God is said to be constantly involved in this world, always ―... acting in nature ...‖ (p. 4), a ―functionally existent God ... that relates to the events of the world ... in a practical sense‖ (p. 13). These descriptions are at odds with their general description of theism because a this-worldly God rules out a ‗divine‘ reality, a ‗worldly connection to God‘, and a God ‗concurrent with nature‘. Each of these assumes a dualism: between the divine and its opposite; between the non-natural and the natural; between the worldly and something outside of the worldly, and so on. Still, if we set these contradictions aside, we can focus on the notion of a this-worldly God. First, Slife and Reber‘s expression ‗functionally existent‘. What it means is not clear. Perhaps the idea is that God is constituted through what he does? If so, this ignores the logic of relations, in particular the ontological distinction between relations and the items that stand in those relations (see Hibberd, 2005, 2009b). A simple example will suffice to make the point. You (that is, your physical being) are not constituted by your interactions with colleagues, students, members of your family, etc. It is logically necessary that you exist independently of those interactions in order for those interactions to occur. Likewise with a this-worldly God. But Slife and Reber refer to God as ―uniquely active‖ (p. 14, emphasis added) suggesting perhaps that God is active in a way that no existent from this world is. If this is the case, the authors are again invoking conditions of existence that are different for God which, again, leaves their account mired in the problems of dualism5. Avoiding dualism, just means that God is subject to the same conditions of existence as everything and everybody else, and this means that God‘s existence is not constituted through God‘s interactions in nature. Second, if we allow a this-worldly God, then God exists in the ordinary sense and is just one finite being among others, albeit one with unique qualities that no other existent possesses. Now we are considering a scientific claim as opposed to a philosophical one. If God belongs to this world, then God‘s activities can, at least in principle, be understood in terms of a material being that engages with other worldly events and processes. As a scientific claim, it would presumably be immune from empirical testing. What testable hypothesis can be derived from a claim of God‘s existence as opposed to a claim of belief in ‗God‘, or an experience of ‗God‘, or knowledge of ‗God‘ through revelation? Of course, some theists (or should that be ‗many theists‘?) are unconcerned about the issue of God‘s existence (logically independent of epistemic considerations). But the ontological matter remains: to have faith in me presupposes that I exist and, if the problems of dualism are to be avoided, that same logic 4

A solution is not to be found in postulating another metaphysical entity (e.g., a soul) which is said to have ‗a foot in both worlds‘ because this (i) obviates the need for two worlds, and (ii) only repeats the dilemma about the new entity‘s relations to both worlds (Passmore, 1970). 5 Dualism is also present in the authors‘ reliance on the concept of agency (Slife & Reber, p. 13). Space prevents me from developing this point. Suffice it to say, a concept of agency is mistakenly thought to be compatible with the doctrine of determinism. On the contrary, ‗agency‘ presupposes an alternative set of conditions of existence.


169

holds for God. Receiving communications from God, or experiencing God, or praying to God, presupposes that God exists. The theist‘s claim that God, Jesus and the Holy Spirit are persons united into one Godhead is just an obfuscation. A strict identity cannot obtain because the same individual cannot be both corporeal and incorporeal. Similarly, the term ‗God‘ cannot be said to refer to the historical figure Jesus, for that makes theism redundant and all talk of an eternal, infinite God particularly strange. The fact that people employ a term intending it to refer to some ‗thing‘, does not suffice to show that the intended reference is or could be achieved. In short, a this-worldly God must exist under the same conditions as any other existent if dualism is to be avoided. It cannot be constituted by its activities in this world and it cannot be incorporeal. Its status is that of any hypothetical entity in science and the question is ‗Does the term ‗God‘ successfully refer?‘ The burden of proof lies with the defender of theism. Finally, a comment about the authors‘ use of the term ‗incompatible‘. ‗Incompatibility‘ is not a recognised logical relation and all that Slife and Reber offer in making clear what they mean by this term is a statement about what they do not mean. They do not mean ‗incommensurable‘ or ‗incomparable‘ (p. 9). In the absence of any positive definition, the reader is left to assume that the word ‗incompatible‘ is intended in its ordinary sense, meaning ‗incapable of being held together‘, ‗opposite in character‘, ‗incongruous‘, ‗irreconcilable‘. It is then a point of logic that claims which distinguish one worldview from its incompatible opposite must stand in either contrary or contradictory relations to each other. They cannot both be true and may both be false, i.e., at least one of the worldviews is wrong. Framing the discussion in terms of ―...meanings from the world of the naturalist [or theist] ...‖ (p. 10) doesn‘t sidestep the logical point. If the theists‘ claims are true, any contrary claim is false. So much for naturalism! And so much for the authors‘ presumption that theism and naturalism are equally valid.

CONCLUSION It has been variously said of theism that it violates the rules and conditions of intelligible discourse, that it is contrary to not only science and philosophy but also to common-sense, that it cannot be reconciled with rational thinking, that there are no grounds for a belief in God. Slife and Reber‘s paper does nothing to counter these perceptions. Instead of explaining why theism is as valid as naturalism and, therefore, why psychologists err in ignoring it, the authors take the intellectually easier and less confrontational option—assume theism‘s validity from the outset and demonstrate only the incompatibility between the two ‗worldviews‘ and psychology‘s allegedly prejudicial practices. However, Slife and Reber‘s notion of ‗incompatibility‘ is sufficiently vague as to render their ‗demonstration‘ implausible. And when we examine theism‘s validity through their concept of God, we find their definition of theism self-contradictory. The authors‘ ignore a crucial distinction—that between conditions of existence and qualities of things or processes. They seem unaware that existence is not a predicate and, therefore, not something of which there are types. Their notions of ‗supernatural‘ or ‘divine‘ could be interpreted as qualities of something this-worldly, in which case the authors have offered a concept of God at odds with theism and a scientific hypothesis which renders implausible their

170

Fiona J. Hibberd

incompatibility thesis. On the other hand, these notions could be interpreted as a mode of existence distinct from a natural mode, in which case they have unintentionally waded into the Humpty Dumpty consequence of dualism. The upshot of Slife and Reber‘s paper is that it nicely illustrates the conceptual difficulties that theism faces. It also shows their charge of prejudice to be groundless— psychologists are not making prejudicial judgements when claims about God‘s existence as divine or supernatural are contrary to reason6. And it serves to demonstrate that psychology‘s secularism is a mark of the discipline‘s commonsense. The burden of proof lies with the theist. Until then, it is only authentic for psychologists who are agnostic or atheist to behave in a manner consistent with their beliefs7. Contra Slife and Reber, this does not involve practising with ―intentional prejudices‖ (p. 9) and discriminating against those who think differently. Consider, finally, the education of psychology‘s students. Helping to develop in students the disposition and ability to think critically is a fundamental educational aim and moral obligation. Slife and Reber do not lead by example. Their paper supports an acquiescence to theism, not an active inquiry into it.

REFERENCES Edwards, P. (2004). Heidegger's confusions. Amherst, NY: Prometheus Books. Gergen, K. J. (1994). Realities and relationships: Soundings in social construction. Cambridge, MA: Harvard University Press. Hibberd, F. J. (2005). Unfolding social constructionism. New York: Springer. Hibberd, F. J. (2009a). Sham reasoning, Humpty Dumpty, and the burden of proof. Journal of Theoretical and Philosophical Psychology, 29, 91-96. Hibberd, F. J. (2009b). Anderson's development of (situational) realism and its bearing on Psychology today. History of the Human Sciences, 22, 1-30. Kusch, M. (1989). Language as calculus vs. language as universal medium: A study in Husserl, Heidegger and Gadamer. Dordrecht: Kluwer. Locke, J. (1706/1924). An essay concerning human understanding (5th ed.). Oxford: Oxford University Press. Passmore, J. (1970). Philosophical reasoning (2nd ed.). London: G. Duckworth & Co. Peirce, C. S. (1931-1958). Collected papers of Charles Sanders Peirce. In C. Hartshorne & P. Weiss (Eds.) (Vol. 1). Cambridge, MA: Harvard University Press. Siegel, H. (1987). Relativism refuted: A critique of contemporary epistemological relativism. Dordrecht: D. Reidel Publishing. Slife, B. D., & Reber, J. S. (2009). Is there a pervasive implicit bias against theism in psychology? Journal of Theoretical and Philosophical Psychology, 29, 63-79. Strawson, G. (2000). Esprit de core. A new way of viewing "the movie-in-the-brain". Times Literary Supplement (October 27), 12-13.

6

Relevant, too, are the arguments that theism is not necessary in explaining how the world works and that theism can itself be explained by psycho-social-cultural theory. 7 I appreciate that a theistic commitment is unlikely to be moved by reason.



Chapter 15

WILL RIGHT-BRAINERS RULE THE FUTURE? John Reid School of Psychology, University of Tasmania, Hobart, Tasmania 7001, Australia

ABSTRACT During the 20th century as the technological age developed, there was high demand for logical, left-brained thinkers such as engineers and scientists. This was the golden era of IQ testing, where IQ scores were excellent predictors of performance on the job. However, the trend in the West since the downsizings of the 90s has been to outsource expensive cognitive left-brain tasks, like electronic engineering and computer programming, to low-cost countries like China and India. However, by moving these activities offshore, the question arises as to what will constitute viable career options in future in the West. This study explores Pink‘s (2005) claim that individuals that have right-brain hemisphericity will be better adapted to the future work environment in the West. The participants in this study were 145 psychology students and their family (mean age 33.5, SD = 14.5) who completed a test battery which included hemisphericity, emotional intelligence, verbal reasoning and big five factor personality questionnaires. It was found that older participants were less right-brain oriented than the younger participants. The general characteristics of right-hemisphericity participants included lower conscientiousness, and higher openness, neuroticism and attention to emotion scores. These results have implications for organisational psychology in the future.

Keywords: Big Five, Emotional intelligence, Hemisphericity, Left-brain, Right-brain

INTRODUCTION Hemisphericity, especially right-brained thinking preference, has been proposed as a major success factor for individuals for future careers (Pink, 2005). Pink argues that the 


172

John Reid

period from the industrial era through the information age required left-brained thinkers who specialised in logical and linear thought processes. These logical thinkers who made major contributions during the last century were employed in a diverse range of positions such as engineers, computer programmers, accountants, auditors, actuaries and so on. The leftbrained or logical thinking period reached its zenith with the introduction of the personal computer and the internet and the myriad of applications that they both spawned. Computer applications that previously required an expensive mainframe could now be developed in the home on a low cost personal computer. This enabled third world communities to become part of the information age. During the same period, the World Wide Web allowed information to be sent simply and cheaply to any point in the globe. With the corporate downsizings and cost reductions in the West in the mid-90s, and the rising skill base in China and India, organisations sought to reduce costs by sending work offshore for processing in countries that paid employees lower wages. The range of jobs outsourced overseas in the early part of the 21st century include a diverse range of applications such as engineering, computer programming, auditing of company records, and processing tax returns for individuals and corporations. The countries accepting this work are mainly China and India with their vast resources of university-educated and skilled engineers, IT specialists, and accountants (Pink, 2005). This trend has continued to the point where the local industries perform very little engineering design or application development for computers. In addition, intellectually-intensive tasks such as auditing and tax return preparation can be performed cheaper in India. The thrust of Pink‘s thesis (2005) is to draw attention to the trend of sending work overseas thereby reducing the range of work-options available in Western countries, such as Australia and the USA. Having identified the problem inherent in outsourcing complex work offshore, Pink developed a rationale for future jobs and activities that are potentially resistant to being outsourced to lower-paid labour forces elsewhere. He proposes that activities that are not susceptible to being easily outsourced are right-brain skills that include: design, storytelling, complex pattern recognition, empathy, play, interpreting and providing meaning in today‘s world. He concludes by predicting that right-brainers will rule the future. It is generally believed that the right-hemisphere provides a dominant and holistic focus on the task at hand, while the left-hemisphere deals with detail and logical processes after the right-hemisphere has holistically surveyed the big picture (Jackson, 2009; Martindale, 1999). These findings generally divide brain hemispherical functions into global, parallel and holistic processing in the right-hemisphere, with logical, verbal, numerical, sequential and analytical processing occurring in the left-hemisphere. Creativity as a right-brain activity including Pink‘s design and story telling skills finds considerable support in the literature (e.g., Morton, 2002; Martindale, 1999; Hermann, 1975, 1989). Many hemisphericity researchers have used dichotic hearing preference as an indication of right versus left-brain preference (e.g., Jackson, Furnham, & Miller, 2001; Jackson, 2002, 2005, 2008; Morton, 2002, 2003; Williams, 1986). However, this study was designed to be delivered over the internet, and the dichotic listening test could not be controlled, so was therefore omitted. Further, we assumed that a hemisphericity test based on a questionnaire would be more suitable for organisational use; an area where we planned future studies. An alternative potential method of determining hemisphericity was the rotating figure (Times, 2007). The rotating figure has appeared in popular psychology articles for some

Will Right-brainers Rule the Future?

173

years, and would be a quick and valuable test if it provided a reliable indicator of hemisphericity. The rotating figure test was therefore included in the internet survey.

EMOTIONAL INTELLIGENCE Since the publication of Goleman‘s (1995) book on Emotional Intelligence (EI) there has been a growing recognition of the value of EI in the workplace (e.g., Caruso, Mayer & Salovey, 2002; Cherniss, 2001; Matthews, Zeidner & Roberts, 2004). There is a general assumption that EI is a right-brain holistic activity; this is supported by neurological evidence (e.g., Damasio, 1994; Lane, 2000; Springer & Deutsch, 1998). Jackson (2002) also found that the right hemisphere provided facilitation of emotional expression. This study provided an opportunity to test the assumption that right-brain orientation is related to emotions and EI; therefore, it included the necessary scales to examine whether hemisphericity and EI were convergent or divergent constructs. To this end the Trait-Meta Mood Scale (TMMS) EI scale (Salovey et al., 1995) was employed in the study.

FUTURE EMPLOYMENT ISSUES Since the data was not available to determine employability of participants and thus include it in the statistical analysis it was decided that the study would focus on the personality differences between individuals categorised as having a right or left-brain preference. If Pink‘s hypothesis (2005) is correct that ―right-brainers will rule the future,‖ then it behoves organisational psychologists to seek an insight into the personality issues at play in the workplace of the future. In order to examine personality differences as a function of hemisphericity, a range of questionnaires including hemispheric preference, verbal and numerical skills, personality and emotional intelligence were included in a single battery delivered to participants over the internet.

STUDY OVERVIEW The study first determined the hemispheric preference of participants using the Zenhausern preference questionnaire (Zenhausern, 1978). The hemisphericity score was then compared with personality, emotional intelligence and verbal/numerical cognitive markers to identify significant personality traits related to hemisphericity. The overall purpose of the study was to examine the relationships between hemisphericity and personality characteristics, with a view to understanding the psychological issues that may arise in the workplace of the future.

174

John Reid

Hypotheses Hypothesis 1: Participants with higher right hemispheric preference scores will have higher emotional intelligence scores. Hypothesis 2: Participants with higher right hemispheric preference scores will have lower verbal and numerical reasoning scores. Hypothesis 3: The rotating figure test will demonstrate convergent validity with the Zenhausern hemisphericity preference questionnaire. Hypothesis 4: There will be notable differences between personality factors for right and left-brain oriented participants.

METHOD Participants The sample of 145 participants (mean age 33.5, SD = 14.5, 63% female) was made up of 55 students enrolled in an advanced 4th year research methods class, plus parents and friends recruited by the students. The participants provided their age, sex and country of birth and then proceeded to complete a web-based battery consisting of the following questionnaires.

Rotating Figure This putative test of hemisphericity has been widely circulated on the internet (e.g., Times, 2007) and was included to determine its veracity. The figure is an animated graphics file (.gif) of a slowly rotating dancer. If the participant perceives it to be rotating in a clockwise direction they are assumed to use more of their right-brain, and alternatively, more of their left-brain if they perceive anti-clockwise rotation. Zenhausern’s Preference Questionnaire This hemisphericity questionnaire consisted of 20-items scored on a 10-point Likert scale ranging from ―1 = Very unlike me‖ to ―10 = Very like me‖ (in Morton, 2002; Zenhausern, 1978). Examples of typical items are: My decisions based upon objective facts rather than feelings (left hemisphericity); I am artistically or musically creative (right); My thinking consist of mental pictures or images (right). Three questions that were directly related to verbal skills, such as I am good at solving crossword puzzles, were removed to reduce method variance with the verbal reasoning skills test. Numeric and Verbal Reasoning Markers These brief cognitive tests (Stankov, 1997) were included to determine the relationship between IQ and hemisphericity scores. The numeric reasoning test consisted of spotting the odd-one-out in a set of numeric sequences, while the verbal reasoning consisted of items such as: Horse is to Animal as Chair is to _______. Big Five Personality Inventory A brief personality inventory (Saucier, 1994) was included to determine the relationships between the Big Five and hemisphericity.


175

Trait Meta-Mood Scale This is a brief EI questionnaire (Salovey et al., 1995) used to determine the relationship between EI and the hemisphericity scores. It includes three subfactors: Attention to Emotions, Clarity of Emotions, and Repair of Emotions.

RESULTS Descriptive statistics are presented in Table 1 below. Correlations between variables are shown in Table 2. Table 1. Descriptive statistics for variables used in this study (N=145) Variable Zenhausern right-brain score Numeric reasoning score Verbal reasoning score TMMS Attention to emotions score TMMS Clarity of emotions score TMMS Repair of emotions score Age Openness Conscientiousness Extraversion Agreeableness Neuroticism

Minimum 50.00 .20 .40 2.46 2.27 2.00 18.00 18.00 17.00 15.00 17.00 7.00

Maximum 120.00 1.00 .95 5.00 5.00 5.00 71.00 40.00 40.00 40.00 40.00 33.00

Mean 83.72 .74 .79 3.88 3.56 3.74 33.46 30.39 29.93 27.01 32.59 19.55

Std. Deviation 15.30 .16 .12 .52 .64 .75 14.46 4.65 5.62 5.89 4.55 5.55

Table 2. Correlations between the variables used in this study Variable 1. Zenhausern Rightbrain score 2. Numeric Reasoning score 3. Verbal Reasoning score 4. TMMS Attention 5. TMMS Clarity 6. TMMS Repair 7. Age 8. Openness 9. Conscientiousness 10. Extraversion 11. Agreeableness 12.Neuroticism

1 1

2

3

4

5

6

-.02

1

-.16

.14

1

.42** -.07 .03

.00 .07 .08

.07 .09 .06

1 .14 .11

1 .32**

1

-.34** .31** -.39** .06 .10 .34**

-.05 -.17* -.03 -.11 .00 -.01

.17* .03 .18* -.04 .24** -.17*

-.13 .21** .18* .14 .40** .22**

.22** .07 .18* .10 .23** -.27**

-.07 .00 .01 .30** .30** -.36**

7

8

9

10

11

1 -.17* .25** -.22** .06 -.24**

1 -.06 .06 .17* .13

1 -.06 .25** -.07

1 .08 -.03

1 -.17*

176

John Reid

Tests between Means An independent samples t-test showed that there was a significant relationship between the perception of the direction of rotation of the dancer and the Zenhausern test of hemisphericity (t(143)=2.27, p=.02). Unfortunately, the significant difference was in the opposite direction to that expected (Times, 2007), with anti-clockwise rotation being associated with higher right-brain scores. An independent samples t-test did not indicate a significant positive relationship between EI (TMMS total score) and hemisphericity (t(143)=1.24, p=.22); however, the right-brain score was significantly related to the TMMS EI Attention to Emotions subfactor (r(144)=.42, p<.01).

DISCUSSION It was proposed in hypothesis 1 that participants with higher right hemispheric preference scores would be significantly higher in total EI scores. This was not confirmed; however, participants did have significantly higher Attention to Emotion subfactor scores (p<.01). This hypothesis was therefore partially supported. It appears that hemispheric preference is somewhat convergent with the Attention to Emotions EI subfactor (see Table 2). However, this translates into an overlap of only 18% of shared variance between the two variables, suggesting that the two constructs are different. Clearly, the Attention to Emotions scale assesses a different construct to the Zenhausern hemisphericity scale and could not be used as a proxy for hemisphericity. The implication of this finding is that there is a place for both hemisphericity and the EI concept, and the two can potentially contribute to a better understanding of human behaviour. In hypothesis 2 it was proposed that participants with higher right hemispheric preference scores would have significantly lower verbal and numerical reasoning scores. While the rightbrain score negatively correlated with both the verbal and numerical reasoning scores, verbal reasoning scores just missed out on reaching significance (p = .054), while the numeric reasoning relationship was nowhere near significant (p = .79). The general assumption that left-brained individuals perform better in numeric and verbal reasoning was apparent in the results, and verbal reasoning came close to reaching significance, and would have reached significance if the sample size had been larger. It can therefore be assumed that logical verbal operations are related to left-brain preference, with the caveat that statistical significance was just missed. Numeric reasoning was not found to be significantly correlated to hemisphericity (p = .79). This finding supports those by Szirony et al. (2007), where mathematical ability was found to be not significantly related to left brain function. The overall conclusion was that hypothesis 2 was partially supported, and there was some evidence that verbal reasoning was a left-brain function, but there was little support for numeric reasoning being a left-brain function. Perhaps a broader battery of verbal and numeric reasoning items could be used in a future study to confirm their putative left-brain association. In hypothesis 3 the effectiveness of the rotating figure as a measure of hemisphericity was examined. Although there was a weak (r = .21) but significant relationship between the rotating figure and the Zenhausern preference questionnaire, the direction of rotation was in


177

the opposite direction to that expected. Using binary logistic regression it was found that the rotating figure would only correctly predict hemisphericity in 59% of the cases, even with the rotation reversed. Therefore, since the rotating figure predicted at only slightly above chance, and in the wrong direction, this pop psychology indicator of hemisphericity must be rejected as a scientifically useful tool. With regard to Hypothesis 4, it was proposed that there would be notable differences between personality factors for right and left-brain oriented participants. This study found that Openness and Neuroticism were positively related, and Conscientiousness negatively related with right-brain oriented participants (p < .01, see Table 2). Further, the age of participants was negatively related to the right-brain score. This indicated that younger participants were significantly more right-brained than the older participants (Table 2). These findings are quite robust with all correlations being significant at the p < .01 level. One implication is that if you want to hire right-brain oriented individuals then the younger the better. However, if the hiring preference was biased towards hiring younger employees, then an organisation would need to ensure that they did not transgress any age discrimination laws.

CONCLUSION Considering the personality factors, it would appear that if you hire more right-brained individuals, they would be more open to experience, which hopefully would translate into them being more creative than an average group. On the other hand, our results suggest that these right-brain individuals would be less mentally stable, and less conscientious than average. Overall this appears to be a somewhat mixed blessing attendant with right-brainers. Based on Pink‘s (2005) assumption that future job opportunities in the West will favour right-brain thinkers, one implication for organisations arising from this study is that future workers will score higher on Neuroticism. This may lead to more conflict in the workplace, increased absenteeism, and more need for Employee Assistance Programs (EAP). Organisations will need to plan and adjust to these anticipated changes. Another finding was that right-brained individuals showed increased Openness to Experience; this would be valuable in enhancing the creativity of an organisation. However, on the downside there is a concern that the reduced Conscientiousness of right-brained individuals could reduce the productivity of an organisation and contribute to absenteeism. A final caution should be added noting that using age as a hiring indicator, with youth as a predictor of right-brain candidates, could be found to contravene age discrimination laws. Hiring based on handedness—with left handedness indicating more right-brain activation— could also fall foul of anti discrimination laws. It would appear that right-brainers provide a mixed blessing for an organisation with increased openness and creativity, but with the dark side of increased neuroticism and lower conscientiousness. Further studies in this area, especially employment success versus hemisphericity, would be valuable in confirming or disproving Pink‘s thesis. The confirmation of a general shift to right-brained employees could have significant implications for organisations and organisational psychology.

178

John Reid

REFERENCES Caruso, D. R., Mayer, J. D., & Salovey, P. (2002). Emotional Intelligence and emotional leadership. In R. E. Riggio & S. E. Murphy (Eds.), Multiple intelligences and leadership (pp. 55-74). Mahwah: Erlbaum. Cherniss, C. (2001). The Emotionally intelligent workplace. San Francisco: JosseyBass/Wiley. Damasio, A. R. (1994). Descartes error: Emotion, reason and the human brain. New York: Putnam. Goleman, D. (1995). Emotional Intelligence: Why it can matter more than IQ. London: Bloomsbury Publishing Plc. Herrmann. (1975). The Herrmann Brain Dominance Instrument. Retrieved from: http://www.hermann.com.au/hbdi-profile.htm March 16, 2010. Herrmann, N. (1989). Herrmann Brain Dominance Instrument. 720 Maple Lane, Sewickley, PA 15143: Hope Unlimited. Jackson, C. (2002). How preferred-ear for listening moderates emotional cognitions in the prediction of personality. Retrieved from: http://www2.psy.uq.edu.au/~chrisj/index.html on March 16, 2010. Paper presented at the European Conference on Personality, Frederich Schiller University, Jena, Germany. Jackson, C. J. (2005). How preferred ear for listening moderates emotional cognitions in the prediction of personality. Laterality, 10, 305-320 Jackson, C. J. (2008). When avoidance leads to approach: How ear preference interacts with neuroticism to predict disinhibited approach. Laterality: Asymmetries of Body, Brain and Cognition, 13, 333-373. Jackson, C. (2009). Do left and right asymmetries of hemispheric preference interact with attention to predict local and global performance. Paper presented at the Australian Conference on Personality and Individual Differences (ACPID09), Sydney University, NSW, Australia. Jackson, C., Furnham, A. & Miller, T. (2001). Moderating effect of ear dominance on personality in the prediction of sales performance. Laterality, 6, 133-140. Lane, R. D. (2000). Levels of emotional awareness. In R. Bar-On & J. D. A. Parker (Eds.), The handbook of emotional intelligence (pp. 171-191). San Francisco: JosseyBass/Wiley. Martindale, C. (1999). Biological basis of creativity. In R. J. Sternberg (Ed.), Handbook of creativity (pp. 137-152). Cambridge: Cambridge University Press. Matthews, G., Zeidner, M., & Roberts, R. D. (2004). Emotional Intelligence: Science and myth. Cambridge, Massachusetts: MIT Press. Morton, B. E. (2002). Outcomes of Hemisphericity Questionnaires Correlate with Unilateral Dichotic Deafness. Brain and Cognition, 49, 63-72. Morton, B. E. (2003). Asymmetry questionnaire outcomes correlate with several hemisphericity measures. Brain and Cognition, 51, 372-374. Pink, D. H. (2005). A whole new mind: Why right-brainers will rule the future. New York: Riverhead Books. Salovey, P., Mayer, J. D., Goldman, S. L., Turvey, C., & Palfai, T. P. (1995). Emotional attention, clarity, and repair: Exploring emotional intelligence using the Trait Meta-Mood


179

Scale. In J. W. Pennebaker (Ed.), Emotion disclosure and health (pp. 125-154). Washington, DC: American Psychological Association. Saucier, G. (1994). Mini-markers: A brief version of Goldberg's unipolar Big-Five markers. Journal of Personality Assessment, 63, 506-516. Springer, S. P., & Deutsch, G. (1998). Left brain, right brain (5th Edition). New York: Freeman and Company. Stankov, L. (1997). The Gf/Gc Quickie test battery: (unpublished test battery). The University of Sydney: School of Psychology. Szirony, G. M., Pearson, L. C., Burgin, J. S., Murray, G. C., & Elrod, L. M. (2007). Brain hemisphere dominance and vocational preference: A preliminary analysis. Work: A Journal of Prevention, Assessment and Rehabilitation, 29, 323-329. Times. (2007). Rotating Dancer. Sunday Times, Retrieved 31st March 2010 from: http://www.perthnow.com.au/fun-games/left-brain-vs-right-brain/story-e2016frg2046u1111114517613?from=mostpop. Williams, S. (1986). A group test of auditory lateral advantage. Cortex, 22, 319-324. Zenhausern, R. (1978). Imagery, cerebral dominance, and style of thinking: A unified field model. Bulletin of the Psychonomic Society, 12, 381-384.



Chapter 16

CULTURAL PERSPECTIVES ON PERSONALITY AND PERSONHOOD Hoyt L. Edge Department of Philosophy & Religion, Rollins College Winter Park, Florida 32789 USA

Cokorda Bagus Jaya Lesmana Department of Psychiatry, Faculty of Medicine, Udayana University, Denpasar, Bali 80114, Indonesia

ABSTRACT The study of personality in psychology developed within the Western tradition and has been criticised as embodying implicit Western assumptions. The popularity of a culturally sensitive approach to personality has seen a recent resurgence because of research in cross-cultural psychology and social psychology. In this chapter we discuss three of our studies comparing Americans and Balinese, representing an individualist and a collectivist culture, respectively. The first one focuses on the cultures‘ understanding of ―volition‖ and supports the view that aspects of this concept seem to hold cross-culturally while other aspects are culturally sensitive. The following two studies lend further support for the view that there are elements of personality that are intimately tied to cultural differences. The second study points to culturally different cognitive preferences in responding to an ethical scenario, with Americans responding in an abstract manner while Balinese respond in a concrete manner. The third study probes aspects of individualism and collectivism, illuminating differences in aspects of personality traits. We argue that psychologists need to develop a more culture-sensitive psychology because there are aspects of the person that are culture-sensitive. As we expand our notions of person through cross-cultural studies, we will concomitantly expand our notion of personality to incorporate traits that are located in interrelationships.



Ph: (407) 646-2178; Fax: (407) 646-2517 email: [email protected]

182

Hoyt L. Edge and Cokorda Bagus Jaya Lesmana

Keywords: Bali, Cognition, Collectivism, Cross-cultural, Determinism

INTRODUCTION As Markus and Kitayama (1998) have effectively argued, the study of personality developed within the Western tradition and embodies to an unknown degree implicit Western assumptions. As a result, it is far from certain that theories of personality apply crossculturally. It may be the case that certain personality traits can be found cross-culturally while other traits cannot; indeed, it may even be the case that the concept of traits is itself a Western construct and not one that is easily transportable cross-culturally. Some researchers have recognised this complexity, and more culturally sensitive approaches to personality have been proposed. The popularity of these approaches has risen and fallen over the last 80 years, but there has been a recent resurgence of them because of research in cross-cultural psychology and social psychology (see Matsumodo, 2001). In this chapter we will argue that personality theorists need to engage in cross-cultural research and incorporate cultural factors. The Western view of the person as an integrated, bounded, stable, autonomous, and free entity is not a view that is fully consistent with the experience of non-EuroAmericans. Based on our research comparing responses to surveys in the US and in Bali, Indonesia, we will argue for a culture psychology in which one seeks persistent patterns both across cultures and within cultures. We view the research discussed here as providing evidence for the necessity of such an approach. Our approach and perspective makes use of research from several disciplines.

CULTURE PSYCHOLOGY As others note (e.g., Markus, 2004), personality theorists traditionally have assumed that their theories apply to all people and that foundational features of personality apply crossculturally. Employing an essentialist approach to knowledge, psychologists and philosophers have focused on what they assumed were essential features of human personality, and they believed that any cultural differences were secondary and derivative. These thinkers believed that since the person was an integrated, bounded, and stable entity, they could focus on this entity and ignore cultural concerns as not being central to personality. Let us be clear from the outset that we are not suggesting that the concept of personality should be abandoned, or that there is no aspect of the personality that is universal; in fact, we will present some evidence from our own research to suggest cross-cultural consistency in an important aspect of personality. Nevertheless, we believe that cross-cultural conceptions of the person are so radically different and so fundamental in how people in various cultures experience themselves that they complicate any notion of a unified understanding of personality cross-culturally. Rather, the evidence that we and others have developed suggest that a culture psychology (Markus & Kitayama, 1998), which is sensitive to cultural differences in personality, should be pursued.

Cultural Perspectives on Personality and Personhood

183

CROSS-CULTURAL PERSONALITY A definition of personality is itself not a straight-forward task. Pervin (1996) defines personality as ―the complex organization of cognitions, affects, and behaviours that gives direction and pattern (coherence) to the person‘s life‖ (p. 414). Carver and Scheier (2000) similarly define it as ―a dynamic organization, inside the person, of psychophysical systems that create a person‘s characteristic patterns of behaviour, thoughts, and feelings‖ (5). For our purposes, we will employ the general view that personality refers to distinctive patterns in how a person thinks, behaves and feels. Given the focus of this anthology on personality and intelligence, we will turn our attention to research that we have carried out on patterns of thinking, or cognition, in the US and in Bali, Indonesia. We believe that there are aspects of the person that seem to be found cross-culturally. Although inconsistent in their results, a number of researchers have had success employing the five-factor model (see McCrae, et al., 1998; McCrae & Terracciano, 1979; Yamagata, et al., 2006). Our own research (Edge & Suryani, 2002) focused on the concept of volition, or willing, a concept central in Western philosophy; it describes the ability of humans to act intentionally in the world. Volition designates the agency through which people affect the world; it is the active aspect of the person as contrasted with the more receptive aspect of receiving information. In our study we surveyed 162 Americans and 282 Balinese using a questionnaire of 82 statements, and through factor analysis we found four factors that indicated high volitional competence in both cultures: Persistence/Initiative, Helping and Influencing Others, Positive Self-Image, Handling Outside Influence. These identical factors were also found in a sample in Edinburgh, Scotland. Those individuals possessing high volitional competence displayed similar traits. Thus, this research suggests that one may find patterns of personality that are similar cross-culturally. However, while we believe that lines of research that seek to find cross-cultural similarity in patterns of personality are important, we want to spend the rest of the paper suggesting that such an approach to personality, while necessary, is not sufficient to develop robust theories of personality that apply crossculturally. Notions of the person are so fundamentally different between EuroAmerican and non-EuroAmerican cultures that any personality theory must take these differences into account. We present both empirical data and theoretical reasons in support of this view. A number of psychologists and anthropologists have written about the distinction between two characteristic views of the person. The Western view of the person has been called variously independent (Markus & Kityama, 1991), individuated (Marsella, 1985), egocentric (Schweder and Bourne, 1984) or atomistic (Edge, 1994). Clifford Geertz (1983) captured this Western approach to the person with an emphasis on individualism in his famous description: ―The Western conception of the person [is] a bounded, unique, more or less integrated motivational and cognitive universe, a dynamic center of awareness, emotion, judgment, and action organised into a distinctive whole and set contrastively both against other such wholes and against the social and natural background…‖ (p. 59). Geertz goes on to point out that this Western conception of the person ―is, however incorrigible it may seem to us, a rather peculiar idea within the context of the world‘s cultures‖ (p. 59). This view of the individuated, independent self can be juxtaposed to the dominant nonEuroAmerican view, which has variously been described as sociocentric (Shweder & Bourne 1984), unindividuated (Marsella,1985), interdependent (Markus & Kitayama, 1991), and

184


relational (Edge, 1994). As opposed to the idea of a bounded, independent personality, a nonEuroAmerican approach to the person emphasises the person in relationship with others, intimately part of social structures, and they cannot be fully differentiated separate from these relationships. Research has shown that this difference between the two cultures‘ views of the person results in behavioural differences. Nisbett and his research team (Nisbett 2003; Nisbett & Peng 2001; Nisbett & Masuda 2003) have produced a significant body of research displaying these differences in cognition, attention, and perception. Our focus in this paper will be in the area of cognition.

DIFFERENCES IN COGNITIVE PREFERENCES IN THE US AND BALI To study cultural aspects of personality, one should seek to understand how patterns of responses differ systematically cross-culturally. Of all the areas involved in personality, it may seem that cognition has the best chance of being displayed consistently cross-culturally. Western philosophy has assumed that rationality is universal and is captured by the axioms of logic. Going back at least to the Greeks, the dominant strain of Western thinking has emphasised a universal approach to truth, assuming that rationality applied cross-culturally. Just as any cultural aspects of the person were viewed as secondary and derivative of the integrated, stable person, so any local cognitive differences were relegated at best to colorful but insignificant variations, but at worst it was argued that whole cultures were designated as pre-logical. Nesbitt has argued, however, that what is accepted as rational differs fundamentally in EuroAmerican and non-EuroAmerican cultures. The analytic tradition of the West views cognition as making abstract distinctions apart from context, and putting arguments in syllogistic form, while the holism of the East emphasises relationships and works within concrete contexts. This distinction can be found in early work of Luria (1971), Scribner (1997), and Cole, Gray, et al. (1971), who investigated why non-industrialised people (in Siberia and in Africa) often gave quite different answers to syllogisms from those of the West. For instance, Scribner asked a non-literate Kpelle (African) farmer: ―All Kpelle men are rice farmers. Mr. Smith (a Kpelle name is used) is not a rice farmer. Is he a Kpelle man?‖ A Kpelle might say in response, ―I don‘t know the man in person. I have not laid eyes on the man himself‖ (p. 107). When asked to just think about the question, the man responds: ―…if you do not know the person, if a question comes up about him, it‘s hard for you to answer it‖ (p. 107). When given a series of classical syllogisms, to which they must respond either ‗yes‘ or ‗no,‘ Kpelle gave syllogistically correct answers only 53% of the time, at chance level, versus 86% solution rate of third grade students in the US (Hill, 1961). The researchers argued that these differences were not due to cognitive deficits but to different characteristic modes of cognition. The pattern of concrete, context-dependent thinking was found consistently in this research, and this type of cognition paralleled their understanding of themselves as persons—related individuals within specific contexts. Thus, people in Western cultures display both abstract, context-free notions of self and employ cognition emphasizing the same characteristics, while non-EuroAmerican concepts of self and cognitive preferences can be described as concrete and context-dependent. How they think about themselves as persons and their characteristic patterns of cognition are


185

systematically related. In one research project among the Kpelle, of the 171 wrong answers to syllogisms, there was ―not a single case in which a theoretical [abstract] reason is given for a wrong answer‖ (Scribner 135). The Kpelle displayed a cognitive preference for thinking concretely, but they were quite capable of thinking abstractly. The same difference in cognition was found in our research in US and Balinese cultures. We gave American and Balinese the following vignette (in each sample, half were university students and half were not students). Question 1: Imagine a universe in which everything that happens is completely caused by whatever happened before it. This is true from the very beginning of the universe, so that what happened in the beginning of the universe caused what happened next, and so on right up to the present. This is true not only of large phenomena such as the movement of the stars, but it is also true of even the smallest physiological response on your body and brain. In this universe Susan accidentally found out, during a doctor visit, that Alice‘s mother has contracted a dangerous virus. Susan tells this to her friend Mary but makes her promise that she will not tell Alice. Now Mary is facing a situation where she has promised Susan that she will not relay this bad news to Alice, but Mary knows that Alice would be better off if she received the news—she would be able to focus on keeping her mother‘s spirits up in this difficult time. Mary makes a decision to tell Alice. In this situation (check one): _____ is the decision that Mary makes necessarily caused by the past? _____ or does Mary have the possibility of making a decision in this situation without it being caused solely by the past?

We found that there was a statistically significant difference between the choice of the two cultures, x2 (1, N=169)=28.2, p < .001. Although the American preferences were almost split in half between the two choices (55.7% for choice 1 vs. 44.3% for choice 2), 83.3% of the Balinese went for the 2nd choice.1 What accounts for this highly significant difference? We believe we find the answer in the different ways that the cultures responded when we asked them to comment on their answers. 44 of the 80 American participants (or 55%) offered a usable comment. The main pattern we found in these comments (from 72% of these comments) indicated that they thought abstractly, in syllogistic form: given the premise in the first paragraph that everything is completely caused, then the conclusion has to follow that the decision was necessarily caused by the past. In doing so, the respondents essentially ignored the concrete information given in the second paragraph. Examples of their responses are:  



1

If everything that happens is completely caused by whatever happened before it, then there is no possibility of making a decision without it being caused solely by the past. This question seems rather obvious. You already said that everything was caused by the thing before it. So naturally, Mary had no choice. It was predestined by her past, according to your stated prerequisites. Well if the universe is one in which everything happens by whatever happened before it then Mary‘s decision is caused from the past.

We wish to thank Dr. Niko Tiliopoulos for his statistical help in studies two and three.

186


On the other hand, the Balinese reasoned differently. Of the 44 comments among those who responded that Mary‘s decision was not necessarily caused by the past, the majority of them (24) explained their decision by referring to some moral aspect of the situation, employing information found in the second paragraph. These comments essentially ignored the first paragraph, the abstract information, the exact opposite cognitive strategy found in the American sample. Not one response offered any of the syllogistic responses that were seen in the American responses. Here are some examples (the names are Balinese):   

Made [Mary] wants to help Mrs. Asti [Alice] recover quickly. I think the decision is taken to a greater level. Although he has pledged not to notify [her], it may be better to break the promise for the safety of Asti‘s mother. Because Made has compassion with Asti‘s mother so she tells Asti about the situation so that Asti can maintain and encourage her mother to be strong in the face of the illness.

Why is there this stark difference? We want to argue that there is a parallel between how Westerners and non-Westerners conceive of themselves as persons and their cognitive responses. Insofar as the self is viewed as bounded and abstract and separate, one tends to employ a cognitive strategy that is abstract, defaulting to syllogistic reasoning. The difference in cognitive patterns between Americans and Balinese mirrors the atomistic and relational definitions of the person. EuroAmericans focus on the person who is a bounded, unitary, enduring entity, while non-EuroAmericans view the person as context-dependent. While Americans typically respond by ignoring the concrete, contextual information, the Balinese respond in the opposite way. Each cognitive preference, therefore, is appropriate given the different conceptions of the self. There is a systematic interrelationship between views of themselves as persons and patterns of cognition; the two are intimately related. Such a view has implications for a conception of personality: those traits that are sensitive to context are emphasised in non-EuroAmerican contexts, with more individual and abstract traits being emphasised in the US. As the sense of the person is sensitive to contexts and relations with other people (and those relationships also dependent on context), then it is hypothesised that traits will be understood as more context-sensitive in non-EuroAmerican cultures.

OTHER EMPIRICAL FINDINGS To pursue this hypothesis, we administered a survey to 60 American and 60 Balinese university students, seeking to investigate more fully what these differing notions of self may entail. We hypothesised that if the Balinese view themselves as defined relationally, in the context of relationships with others in community, and the Americans view themselves as bounded and independent2, we would find a significant difference in how people in these cultures find it important to fit in with others‘ expectations and how willing they were in doing so. We asked them to indicate their response to the statement: ―I am willing to align my actions if it fits better with the expectations of the community,‖ and gave them a 7 point 2

We take Locke‘s (1980/1690) notion of the social contract as the paradigm of the American view of how self and others are connected.


187

Likert scale ranging from ―no willingness‖ to ―a lot of willingness.‖ As expected, to a significant degree Balinese responded that they were more willing to align their actions to the expectation of the community. On average, US individuals (M = 4.12, SD = 1.34) tended to be less willing to align their actions with the expectations of their community than Balinese ones (M = 5.59, SD = 1.28), t(164)=7.22, p < .0005, two-tails. Then we focused on one specific aspect of this difference in the notions of self. We hypothesised that Americans, who sense themselves as independent and bounded, would feel more comfortable in being alone than Balinese. Based on theory and our own experience within these cultures, we hypothesised that persons in individualist cultures value being alone, having a need to retreat to privacy to find their inner, real selves—being alone charges the batteries of the individualist person, allowing ―space‖ to re-center and revitalise themselves as separate individuals. On the other hand, people in relational cultures value being in relationship and being with others. Thus, it was hypothesised that Americans would find being alone as important to their sense of the individual self and even revitalizing to them, while Balinese would find being alone as uncomfortable. We asked: ―How do you feel when you are alone?‖ giving them a 7 point Likert scale from ―extremely uncomfortable‖ to ―mentally revitalised.‖ As anticipated, we found a significant difference in the responses in the expected direction. On average, US individuals (M = 4.77, SD = 1.44) tended to feel far more comfortable and mentally revitalised when alone than the Balinese (M = 3.66, SD = 1.54), t(164)=4.78, p < .0005, two-tails. It is important to understand that these are fundamentally different approaches to persons. Balinese do not simply feel themselves to be connected to others in a close way, an experience that is well known in the West, but they feel themselves to be importantly defined by those relationships. The connection they have with the community is so fundamental that they cannot conceive of themselves apart from the community. Aristotle made an analogous suggestion when he said that the connection between a person and the community is so strong that an individual apart from a community might be conceived of as a god or a beast but not as a person. The ancient Greeks also defined themselves as part of a community. The difference between being defined by relationships within the community and simply being associated with other people (e.g., through voluntary association a la Locke) can be seen in the following results. We asked these Americans and Balinese to respond to the statement: ―I am open to the influence of other people in my life,‖ with responses on the 7point Likert scale ranging from ―No openness‖ to ―A lot of openness.‖ US individuals (M = 5.24, SD – 1.05) tended to be more open to the influence of other people in their lives than the Balinese (M = 4.32, SD = 1.22), t(164)=5.21, p < .0005, two-tails). This result might be surprising if a person does not make the distinction between Balinese, who define themselves in terms of their community, and EuroAmericans who view themselves as bounded wholes that choose to associate with others. In other words, Balinese do not identify with ―other people,‖ but they identify with the community of which they are a part. They would be open to the community influence, but not from ―others‖ (who may or may not be part of the community). Hence, there was no identification of the self with an undifferentiated ―other people,‖ and so they were even less open to their influence than individualist Americans. The trait of openness can be understood only within the differing understandings of personhood.

188


IMPLICATIONS The differences found in this study are robust and suggest that there are aspects of indigenous views of the person that differ significantly. Western theories of personality have tended to valorise individualist traits, assuming that the person is an autonomous, bounded, individuated entity. Action that was inner directed, signalled a mature person, and Westerners have applauded persons creating situations of separateness, of being alone, where individuals could focus on this internal self and develop it. However, our data suggests that these personality traits are not held in the same esteem in Bali, and by implication by other nonEuroAmerican cultures. Hence, we may find different personality traits as valorised in different cultures, and even if the same traits are valorised cross-culturally, it is likely that they are understood in fundamentally different ways in different cultures, given their differing definitions of personhood. It may even be the case that traits are viewed in non-Western cultures as describing relationships among people rather than attaching to individuals (see Sampson, 1977, for an argument for this position).

CONCLUSION To develop a robust culture psychology in which we understand personality in nonEuroAmerican cultures adequately, we will need to continue research into how nonEuroAmerican patterns of cognition, feeling, and behaviours differ from EuroAmerican ones. We have presented evidence that it is likely that a culturally-sensitive understanding of personality will find patterns of cognition, emotion, and values that are consistent crossculturally, on the one hand, but also patterns that differ cross-culturally, on the other hand. Our conception of personality is grounded in our conception of the person, and much research indicates that there are fundamentally different views of personhood in EuroAmerican and non-EuroAmerican cultures. Concomitant with these different conceptions of personhood are different patterns of cognitive responses; such differences suggest that any attempt at understanding personality cross-culturally by examining only universal traits will be insufficient to establish a robust cultural psychology.

REFERENCES Carver, C. S., & Scheier, M. F. (2000). Perspectives on personality. Boston, MA: Allyn and Bacon. Cole, M. J. (1971). The Cultural context of learning and thinking. New York, NY: Basic Books. Edge, H.L. (1994). A Constructive postmodern perspective on self and community: From atomism to holism. Lewiston, NY: The Edwin Mellen Press. Edge, H. L., & Suryani, L.K. (2002). A Cross-cultural analysis of volition. Florida Philosophical Review, 2, 56-72.


189

Geertz, C. (1983). "From the Native's Point of View": On the Nature of Anthropological Understanding. In C. Geertz (Ed.), Local Knowledge: Further Essays in Interpretive Anthropology (pp. 55-70). New York: Basic Books. Hill, S. (1961). A Study of the Logical Abilities of Students (Doctoral dissertation), Stanford University, Stanford, CA. Locke, J. (1980/1690). The second treatise of government. Indianapolis, IN: Hackett Publishing. Luria, A.R. (1971). Towards the problem of the historical nature of psychological processes. International Journal of Psychology, 6, 259-272. Markus, H. R. (2004). Culture and personality: Brief for an arranged marriage. Journal of Research in Personality, 38, 75-83. Markus, H. R., & Kitayama, S. (1991). Culture and the self: Implications for cognition, emotion, and motivation. Psychological Review, 98, 224-253. Markus, H. R., & Kitayama, S. (1998). The Cultural psychology of personality. Journal of Cross-Cultural Psychology, 29, 63-87. Marsella, A. (1985). Culture, self, and mental disorder. In A. Marsella, G. DeVos, & F. Hsu (Eds.), Culture and Self: Asian and American Perspectives (pp. 281-308). New York: Tavistock Publications. Matsumoto, D. (Ed.). (2001). The handbook of culture and psychology. New York, NY: Oxford University Press. McCrae, R. R., Costa Jr., P. T., Del Pilar, G. H., Rolland, J. P., & Parker, W. D. (1998). Cross-cultural assessment of the five-factor model: the revised neo personality inventory. Journal of Cross-Cultural Psychology, 29, 171-188. McCrae, R. R., Terracciano, A., & 79 Members of the Personality Profiles of Cultures Project (2005). Personality profiles of cultures: Aggregate personality traits. Journal of Personality and Social Psychology, 89, 407-425. Nisbett, R.E. (2003). The geography of thought: How Asians and westerners think differently … and why. New York, NY: The Free Press. Nisbett, R.E., & Masuda, T. (2003). Culture and point of view. Proceedings of the National Academy of Sciences of the United States of America, 100, 11163-11170. Nisbett, R. E., Peng, K., Choi, I., & Norenzayan, A (2001). Culture and systems of thought: Holistic versus analytic cognition. Psychological Review, 108, 291-310. Pervin, L. A. (1996). The science of personality. New York, NY: John Wiley. Sampson, E. E. (1977). Psychology and the American ideal. Journal of Personality and Social Psychology, 35, 767-782. Shweder, R. A., & Bourne, E. J. (1984). Does the concept of the person vary cross-culturally? In R. A. Shweder & R. A. LeVine (Eds.), Culture theory essays on mind, self, and emotion (pp. 158-199). New York, NY: Cambridge University Press. Scribner, S. (1997). Recall of classical syllogisms: A cross-cultural investigation of error on logical problems. In A. E. Tobach, R. J. Falmagne, M. Parlee, M., M. L. Dapelman (Eds.), Mind and social practice: Selected writings of Sylvia Scribner (pp. 106-124), New York, NY: Cambridge University Press. Yamagata, S., Suzuki, A., Ando, J., Ono, Y., Kijima, N., Yoshimura, K., Ostendorf, F. Angleitner, A., Riemann, R., Spinath, F., Livesley, W. J., & Jang, K. L. (2006). Is the genetic structure of human personality universal? A cross-cultural twin study from North America, Europe, and Asia. Journal of Personality and Social Psychology, 90, 987-998.

PART C: CONTRIBUTIONS TO CLINICAL PSYCHOLOGY AND ASSESSMENT



Chapter 17

GENETIC VARIATION IN DRD4, CRIMINALITY, AND CONDUCT DISORDER Tara M. Thompson Child Study Center, Yale University, USA

Marya Getchell School of Public Health, Harvard University, USA

Britt af Klinteberg Institute of Psychology, Stockholm University, Sweden

Roman A. Koposov Department of Psychology, University of Tromsø, Norway

Lars Oreland Department of Neuroscience & Pharmacology, Uppsala University, Sweden

Andrew J. Pakstis Department of Genetics, Yale University, USA

Vladislav V. Ruchkin Karolinska Institute & Skonviks Psychiatric Clinic, Sweden Yale University, USA

Carolyn M. Yrigollen Department of Genetics, University of California, Davis

Elena L. Grigorenko

Child Study Center, Department of Psychology, Department of Epidemiology & Public Health, Yale University, New Haven, CT 06519-1124, USA



Ph: 203-737-2316; Fax: 203-785-3002; [email protected]. Elena L. Grigorenko is also associated with Columbia University, USA, & Moscow State University, Russia.

194

Tara M. Thompson, Marya Getchell, Britt af Klinteberg et al.

ABSTRACT Conduct disorder is the most commonly referred psychological problem in child psychiatric and mental health services (Sholevar, 1995). It is characterised by externalizing behaviours such as aggression and antisocial behavior, and is often associated with impulsivity and hyperactivity. Research has shown that externalizing behaviour has a sizable genetic component and is thought to be associated with the dopaminergic system, in which the dopamine D4 receptor gene (DRD4) plays a role. The current study explores whether detainees or conduct disorder individuals can be differentiated from controls based on the variation in genetic polymorphisms of DRD4. Allele, genotype, and haplotype frequencies for the exon 3 VNTR and the C_1611535 (rs58688157), rs9366, rs11246226 SNPs in the DRD4 gene were compared in 179 incarcerated adolescent delinquents and 60 matched controls. Results indicated that genetic variation in DRD4 is associated with criminality (i.e., incarceration) and conduct disorder. Understanding the influence of DRD4 on externalizing behaviour problems associated with conduct disorder is important for understanding etiological mechanisms of conduct problems.

Keywords: Conduct disorder, Dopamine, Dopamingeric system, DRD4 gene, Endophenotypes

GLOSSARY Allele: One of the variant forms of a gene at a particular locus, or location (site), on a chromosome. Candidate Gene: A gene, located in a chromosome region suspected of being involved in a disorder or trait, or whose protein product suggests that it could have a role in the pathophysiology of the disorder. Chromosome: One of the threadlike "packages" of genes and other DNA in the nucleus of a cell. Different kinds of organisms have different numbers of chromosomes. Humans have 23 pairs of chromosomes, 46 in all: 44 autosomes and two sex chromosomes. Each parent contributes one chromosome to each pair, so children get half of their chromosomes from their mothers and half from their fathers. Conduct Disorder: a behavioural and emotional disorder of childhood and adolescence. Children with conduct disorder act inappropriately, infringe on the rights of others, and violate the behavioural expectations of others. Deoxyribonucleic Acid (DNA): The nucleic acid that carries the genetic information in the cell and is capable of self-replication and synthesis of RNA. DNA consists of two long chains of nucleotides twisted into a double helix and joined by hydrogen bonds between the complementary bases adenine and thymine or cytosine and guanine. The sequence of nucleotides determines individual hereditary characteristics. Detainee: A person held in custody or confinement. DRD4: The Dopamine Receptor 4 gene, one of five receptor subtypes for dopamine. Dopamine: A neurotransmitter in the brain that has both excitatory and inhibitory functions related to motor control, motivation, learning, and reward pathways.

Genetic Variation in DRD4, Criminality, and Conduct Disorder

195

Endophenotype: Measurable components connecting disease and distal genotype. Exon: The region of a gene that contains the code for producing the gene's protein. Each exon codes for a specific portion of the complete protein. In some species (including humans), a gene's exons are separated by long regions of DNA (called introns) that have no apparent function. Externalizing behaviours: Behaviours that are overt, disruptive, and often involve the violation of societal norms, the destruction of property, and harm towards others. G protein: Any of a class of cell membrane proteins that function as intermediaries between hormone receptors and effector enzymes, and enable the cell to regulate its metabolism in response to hormonal changes. Gel electrophoresis: The process in which molecules (such as proteins, DNA, or RNA fragments) can be separated according to size and electrical charge by applying an electric current to them. The current forces the molecules through pores in a thin layer of gel, a firm jelly-like substance. The gel can be made so that its pores are just the right dimensions for separating molecules within a specific range of sizes and shapes. Gene: The functional and physical unit of heredity passed from parent to offspring. Genes are pieces of DNA, and most genes contain information for making a specific protein. Genetic variation: A phenotypic variance of a trait in a population attributed to genetic heterogeneity. Genotype: The combination of alleles at a particular locus (e.g., CC, CT or TT at a ‗CT‘ polymorphism). Haplotype: A group of alleles within a gene or across different genes on a single chromosome that are closely enough linked to be inherited as a unit. Hybridization: The process of joining two complementary strands of DNA. Neurotransmitter: A brain chemical that transmits signals between neurons. PCR: Polymerase chain reaction; a technique for copying the complementary strands of a target DNA molecule simultaneously for a series of cycles until the desired amount is obtained. Phenotype: Observable characteristics of an organism produced by the organism's genotype interacting with the environment. Polymorphism: A common variation in DNA in which alternate sequences occur in populations. Proband: An individual or member of a family being studied in a genetic investigation. SNP: A single nucleotide polymorphism, a variation in the gene sequence that occurs at a single locus (one nucleotide base). VNTR: Variable number tandem repeats.

INTRODUCTION Conduct disorder (CD) is the most commonly referred psychological problem in child psychiatric and mental health services (Sholevar, 1995). It is characterised by externalizing behaviours (e.g., persistent aggression, theft, vandalism, violent behaviour, defying authority) that challenge social rules and regulations and, therefore, are often referred to as ―anti-social‖. CD, like other externalizing disorders (e.g., Oppositional Defiant Disorder, ODD), is also

196


characterised by a propensity to express distress outwardly (Krueger, et al., 2002), so that affected individuals are prone to anger, impulsivity, and low attentional and behavioural regulation. Unfortunately, CD has proven to be highly resistant to psychosocial interventions; as a result, youth diagnosed with this disorder may develop other serious behavioural and emotional problems in adulthood (e.g., substance abuse, antisocial personality disorder, criminal behaviour, and alcoholism). It has been observed that CD is familial, i.e., the risk for first degree relatives of individuals with CD appears to be higher than for unrelated individuals (Sholevar, 1995). Both genetic and environmental factors are thought to underlie this familiality (Slutske, et al., 1997). The biological machinery underlying CD is unclear, but genetic factors are considered to be important (Rhee & Waldman, 2002). One such factor is the genetically controlled functional variation in the dopaminergic (DA) system (Moffitt, et al., 2008). Here we explore the association between selected genetic characteristics of the DA system and CD. Children and adolescents with CD exhibit a persistent behaviour pattern in which major age appropriate societal norms or rules, or the basic rights of others are repeatedly violated. These behaviours constitute four main groupings (as per DSM-IV): aggressive behaviour that either threatens or causes harm to others; non-aggressive property destruction; covert aggressive behaviours of deceitfulness or theft; and rule violation. For a diagnosis to be made, a child must show at least three criterion behaviours during the past 12 months, with one behaviour present in the last 6 months. The DSM-IV criteria have been shown to have internal consistency, test-retest reliability, and validity (Lahey, et al., 1994). In a longitudinal twin study, researchers tested the concurrent, convergent, and predictive validity of DSM-IV conduct disorder in children 5 years of age (Kim-Cohen, et al., 2005). They found that 5 year olds diagnosed with CD were significantly more likely than comparison participants to have behavioural and educational difficulties 2 years later despite their apparent remission from CD. Long-term studies have also shown that true recoveries from CD are exceedingly rare because, although not all children diagnosed with CD grow up to have antisocial personality disorder, virtually all children with CD have mental disorders and poor functioning in adulthood (Kim-Cohen, et al., 2005). Yet, the literature indicates that interventions for preventing and remediating CD can be effective if applied early in life (Kim-Cohen, et al., 2005). Thus, it is important to identify, if possible, genetic risk markers associated with CD that can be used in prevention and remediation practices.

ENDOPHENOTYPES Endophenotypes are vital to understanding the relationship between genes and complex traits, externalizing behaviour notwithstanding. The conceptual purpose of endophenotype is to divide behavioural symptoms into more stable ―componential‖ phenotypes that have a clear genetic connection (etiology). Endophenotypes have a potential to increase the power of genetic linkage and association analyses by identifying individuals with particular genetic susceptibilities across samples, subtypes, situations, levels of severity and the presence of comorbidity (McBurnett & Pfiffner, 2007). Externalizing problematic behaviour is related to a variety of mental health disorders such as CD, attention deficit/hyperactivity disorder (ADHD), and anti-social personality disorder. In turn, these externalizing behaviours have


197

been associated with such endophenotypes as impulsivity, aggression, and flawed decisionmaking. For example, there is a considerable overlap between CD and ADHD both at the genotypic level and the phenotypic level. Some researchers argue that this overlap may result from common, underlying core psychopathological processes that are strongly influenced genetically (Krueger, 1999). Thus, candidate genes for ADHD may also be relevant candidates for CD. Genes related to dopamine turnover are plausible candidates for ADHD, based on biological and pharmacological evidence; these genes might also be of relevance to CD (Krueger & Tackett, 2006).

DOPAMINE AND THE DOPAMINERGIC SYSTEM Dopamine (DA) is an important neurotransmitter in the brain that controls various functions, including motor activity, cognition, motivation and emotion. Genes in the dopaminergic system have been systematically and increasingly associated with externalizing behaviours. Thus, molecular-genetic investigations have identified associations between ADHD and a number of dopamine-related genes, most notably, the dopamine transporter gene (DAT1) and two dopamine receptor genes, DRD4 and DRD5 (Dawei, Sham, Owen, & He, 2006). These and other (Genro, Kieling, Rohde, & Hutz, 2010) findings strongly implicate the involvement of brain DA systems in the pathogenesis of externalizing behaviours associated with ADHD. Moreover, clues to DA‘s role in externalizing behaviours have come from many animal studies (e.g., Arias-Carrión & Pöppel, 2007; Hejjas, et al., 2007; James, et al., 2007; Peciña, Cagniard, Berridge, Aldridge, & Zhuang, 2003). DA signaling in the brain is supported by a number of genes, including the genes producing proteins responsible for synthesizing, delivering, binding, and metabolizing DA. Here we focus on a particular DA receptor, receptor 4. DA receptors belong to a class of metabotropic G protein coupled receptors. In the human brain dopamine can bind to five DA receptor subtypes (D1 to D5), which are split into two families: the D1-like family (excitatory) or D2-like family (inhibitory). The D1 and D5 receptors belong to the D1-like family and the remaining receptors (D2, D3, and D4)—to the D2-like family. Intracellularly, the dopamine D4 receptor interacts with inhibitory G-proteins. This interaction inhibits adenylate cyclase, an enzyme that can catalyse the production of cyclic AMP (cAMP), one of the most important messengers in the cell. When stimulated, cAMP exerts several biochemical changes such as activating genes and influencing the opening and closing of calcium and potassium (i.e., ion) channels; corresponding inhibitory interactions with cAMP have the opposite effect. The D4 receptors are mainly found in the regions of the brain that are part of the limbic system, which is thought to be an underlying substrate for emotion and memory; correspondingly, it has long been assumed that variation in DRD4 is associated with variation in attention, motivation and memory (DeYoung, et al., 2006).

THE DRD4 GENE The DRD4 gene is located on chromosome 11p15.5; its size is 3400bp and it is considered to be one of the most variable human genes known (Dawei, et al., 2006). Both due

198


to the role of D4 in the human brain and the variation in DRD4, polymorphisms in this gene have received considerable attention both in studies of individual differences in related psychological functions in typical individuals and in studies of various mental health disorders. The most studied polymorphisms within the DRD4 gene are a variable number tandem repeat (VNTR) in exon 3, a polymorphic deletion of 12bp in exon 1, and single nucleotide polymorphisms resulting in amino acid substitutions in exon 1 and exon 3 (Seaman, Fisher, Chang, & Kidd, 1999). The gene contains numerous other less studied polymorphisms. To illustrate, the association between the variation in the DRD4 exon 3 VNTR and the variation in inattention, response inhibition, impulsivity, and monitored activity level was examined in a study with 133 children ranging in age from 6 to 13 years (Langley, et al., 2004). A particular variant in this polymorphism, the so-called 7-repeat variant, was of special interest. Based on the presence/absence of this allele, the children were divided into two groups: those with one or more DRD4 exon 3 7-repeat allele(s) and those without. It was reported that the group of children possessing one or two copies of this allele was characterised by higher rates of ODD and CD compared to children who had other variants at this polymorphism. Based on these convergent lines of the literature, we explored in this study instances of genetic association between the variation in the DRD4 gene, criminality (i.e., having committed a crime and serving a sentence in a residential facility), and the diagnosis of CD.

METHOD Participants The detainee group (N=179) consisted of a group of male adolescent inmates who were court ordered to the only juvenile detention facility in the Arkhangelsk region of northern Russia, an area with a population of 1.5 million. In this ethnically homogeneous region, approximately 98% of the population is of Russian ancestry. The convictions of the participants varied, with the most severe violations including property crimes (51%), violence-related crimes (38%), rape or other forms of violence (6%), and murder (5%). Normative participants (N=60) were recruited from one of the public schools in the city of Arkhangelsk. In both groups, the participants were matched by gender and age.

Grouping The actual status at the time of data collection (i.e., incarcerated vs. not) was used as the first grouping scheme. The second grouping scheme utilised the diagnosis of CD. The diagnoses were determined using the Schedule for Affective Disorders and Schizophrenia for School-Age Children (K-SADS-PL), a widely used, extensively validated, semi-structured psychiatric interview (Kaufman, et al., 1997), carried out by two psychiatrists who had received standard K-SADS training from the author of the instrument. K-SADS data were available for 174 offenders, but for only 39 controls because of attrition (psychological assessments were carried out following specimen collection but took substantially more time,


199

which decreased the participation rate). Diagnoses were based exclusively on information collected from the participants. Participants were placed in the CD group if they received DSM–IV diagnoses of either current or past CD. Among the incarcerated youth, 130 individuals (74.7%) received the diagnosis; in the control group, there were 11 (28.2%) individuals with the CD diagnosis. This difference in frequencies was significant (φ = .38, p < .000; χ²(1) = 140.7, p < 0.001).

DNA Sampling, Processing, and Analyses After completing the assessments and reports, blood samples were obtained by two nurses from the participants‘ arm veins. The DNA was extracted from the samples and collected via 5ml vacutainer tubes containing ethylenediaminetetraacetic acid (EDTA), and it was then subsequently amplified with Repli-G technologies and genotyped using the ABI TaqMan platform. Once a sufficient amount of DNA was available for each sample, SNPs within the DRD4 gene were genotyped. DNA from the blood samples were purified using the Qiagen FlexiGene DNA kit. The blood sample in each tube was transferred, along with any residual liquid, into a 50ml centrifuge tube and 25ml of buffer FG1 was added and mixed by inverting the tube 5 times. The tubes were then centrifuged (2000g) for 5mins and the supernatants were discarded leaving DNA pellets. Buffer FG2 (5ml) and Qiagen protease (50µl) were added to the tube and the tubes were immediately vortexed until the pellets were completely homogenised. The tubes were inverted three times and placed in a water bath to incubate at 65ºC for 10mins. The tubes were vortexed again for 5s to complete homogenization and 5ml of 100% isopropanal was added to the mixture. The tubes were mixed thoroughly by inversion until the DNA precipitate became visible as a clump. The tubes were then centrifuged (2000g) for 3mins and the supernatants were discarded leaving the pellets. The DNA pellets were rinsed with 5ml of 70% ethanol followed by a 3 min centrifuge to remove any remaining salts. The ethanol was carefully discarded from each tube and the DNA pellets were air dried until all the liquid was evaporated. The pellets were resuspended in 1ml of buffer FG3, vortexed for 5s at low speed and dissolved by incubating for 1hr at 65ºC in a water bath. The concentration of DNA in each sample was analysed using a NanoDrop® ND-1000 spectrophometer. The nanodrop was calibrated to zero absorbance using samples of TE buffer and distilled water. The absorbance of a 1µl aliquot of each sample was analysed and the concentration of DNA was calculated based on the constant: 1OD (Optic Distance) =50µg/ml of DNA. The DNA bands were visualised using E-Gel® 96 High Range DNA Markers. These markers are used as a molecular weight standard in agarose electrophoresis. The marker is supplied in gel loading buffer and can be applied directly to an agarose gel. These fragments can be visualised by ethidium bromide staining. Samples were loaded (96 in each plate) with 10µl of water, 2µl of loading dye and 1µl of each DNA sample. The gels were run for 12 mins and then digitally photographed to visualise the bands. Four polymorphisms, C_1611535 (12,220 bp from 5' end), rs936461 (809 bp from 5' end), 48-bp tandem repeat (exon 3), and rs11246226 (488 bp from 3' end) were genotyped in DRD4. Polymorphisms were detected through methods based on the TaqMan technology used for allelic discrimination. Allelic discrimination detects different forms of the same gene that differ by nucleotide substitution, insertion, or detection (Applied Biosystems). The Allelic discrimination assay combines PCR amplification and detection into a single step.

200


This allelic discrimination is based on a traditional PCR in which two fluorescent probes directed to the polymorphism locus studied are additionally included. The probes were labelled with FAM (6-carbonyl-fluorescein) or VIC fluorescent dyes (Applied Biosystems) and designed to match perfectly either one or other alleles. Fluorescent signals are only produced when the probe matches the DNA genomic sequence, indicating the presence of one or more SNPs. The exon 3 VNTR was genotyped using agarose gel technology. Genotyping was carried out both in Drs. Oreland and Grigorenko‘s laboratories.

Statistical Analyses Chi-square analyses were used to analyze the genotype, allele and haplotype frequencies and determine statistically significant associations. For all of the statistical tests, results were considered significant at p< 0.05. All of the analyses were performed with SPSS for Windows (version 15; SPSS, Inc., Chicago).

RESULTS A total of 239 participants were included in this study. These participants‘ data were analysed by means of group-based comparative analyses (i.e., incarcerated individuals vs. controls and individuals with and without CD) with regard to the variation in DRD4. The allele, genotype, and haplotype frequencies for each polymorphism in each group were examined to determine any statistically significant differences between groups. In a nutshell, for the incarcerated versus control group comparison, the C_1611535 polymorphism was the only polymorphism differentiating the groups. In the CD versus no CD group comparison, the same polymorphism, C_1611535, as well as the rs11246226 polymorphism were found to be statistically informative for group differentiation. For the haplotype analyses, 15 haplotypes were investigated and they revealed a number of interesting, statistically significant results in differentiating both the incarcerated versus control and CD versus no CD groups. Finally, there were some statistically significant findings for YSR. In summary, the variation in the DRD4 gene appears to be associated with risk for incarceration, CD, and externalizing behaviours. The specifics of this conclusion are presented below.

Group Analyses 1: Incarcerated versus Control Samples Analysis of Allele Frequencies The allele frequencies for the three polymorphisms are given in Figure 1 (a-d) below. There were no differences in allele frequencies for any of the genotyped polymorphisms. Although no statistically significant associations were found, Figure 1 shows slight differences in the frequencies between the detainees and controls. Genotype frequencies for each polymorphism in each group are shown in Table 1 and Figure 2. Of note is that, for the exon 3 VNTR, we grouped all but the 7-repeat alleles into a


201

Detainees Controls

Number of Alleles

Number of Alleles

single group. The genotype frequencies did not differ significantly between the two groups for the rs9366, rs11246226, or exon 3 polymorphism. There were, however, statistically significant differences in genotype frequencies at the C_1611535 (rs58688157) polymorphism (X2(2) =6.63, p=0.036, with the genotype GG being present in incarcerated and absent in control individuals. Figure 2 shows that there appears to be no drastic difference between the frequencies of each genotype, with the exception of the GG genotype located at C_1611535.

(b)

Detainees

Controls

(c)

Controls

Number of Alleles

Number of Alleles

(a)

Detainees

Detainees Controls

(d)

Figure 1. (a-d). Allele frequencies at the three SNPs C_1611535(a), rs9366(b), rs11246226(c) & Exon 3 VNTR (d).

202

Tara M. Thompson, Marya Getchell, Britt af Klinteberg et al. Table 1. Counts for DRD4 genotypes in incarcerated vs. control groups N

Detainees Controls

179 60

C_1611535 GG AG AA 11 62 103 0 29 29

rs9366 GG AG 80 63 3 29

Exon 3 VNTR X-X X-7 53 126 23 37

AA 34 8

rs11246226 CC AC AA 38 75 55 14 15 14

Detainees Controls

Individual SNPS

Figure 2. Genotype frequencies in incarcerated vs. control groups.

Table 2. Estimated DRD4 haplotype frequencies with their standard errors for groups of detainees & controls for C_1611535, rs9366, the exon 3 VNTR polymorphisms, & rs11246226 Haplotype 1 (AAXA) 2 (AGXA) 3 (GAXA) 4 (GAXC) 5 (AAXC) 6 (GA7C) 7 (AGXC) 8 (GG7A) 9 (GG7C) 10 (GA7A) 11 (AA7C) 12 (AG7C) 13 (GGXC) 14 (AG7A) 15 (GGXA)

Controls Frequencies Standard Error .132 .031 .292 .042 .041 .018 .047 .020 .052 .020 .073 .024 .237 .039 .038 .018 .033 .017 .004 .006 .024 .014 .015 .011 .013 .010 0 0 0 0

Detainees Frequencies Standard Error .104 .016 .380 .026 .022 .008 .050 .011 .118 .017 .058 .012 .110 .017 0 0 .063 .013 .010 .005 .007 .005 .019 .007 .022 .008 .021 .008 .016 .007


203

Analysis of Haplotype Frequencies Haplotype analysis was carried out with the HAPLO program (Hawley & Kidd, 1995). As indicated above, 15 haplotypes with frequencies >5% in at least one out of the two groups were revealed (see Table 2). Significant differences in frequencies of DRD4 haplotypes across the C_1611535 (rs58688157), rs9366, exon 3 VNTR, and rs11246226 polymorphisms were registered (X2(14)=40.2, p<0.001). It is interesting to note that the 8th haplotype was present in the control group but not in the detainee group. The 14th and 15th haplotypes, however, were present in the detainee group but not in the control group.

Figure 3. Allele frequencies at the three SNPs C_1611535(a), rs9366(b), rs11246226(c) & Exon 3 VNTR (d).

204


Group Analyses 2: Conduct Disorder vs. no Conduct Disorder Samples Analysis of Allele Frequencies The allele frequencies for three SNP polymorphisms are provided in Figure 3 (a-d). Similar to the group analyses presented above, there were no differences in allele frequencies for any of the genotyped SNPs. Although no statistically significant associations were found, Figure 3 shows slight differences in the frequencies between the participants with and without CD. Analysis of Genotype Frequencies Genotype frequencies for each polymorphism in each group are presented in Table 3 and Figure 4. These frequencies, mostly, did not differ significantly between the two groups for genotypes at rs9366 However, there was a significant difference at rs11246226 (X2(2)=6.39, p=0.041). Analysis of Haplotype Frequencies Haplotype analysis revealed significant differences in patterns of DRD4 haplotype frequencies across the four studied polymorphisms among the individuals diagnosed with CD and those who were not (X2(14)=37.7, p<0.001). Table 4 shows that the 8th haplotype was present in the no CD group, but absent in the CD group. The 12th and 13th haplotypes, however, were absent in the no CD group and present in the CD group. Table 3. Counts for DRD4 genotypes in groups with & without CD N

Conduct Disorder NO Conduct Disorder

C_1611535

rs9366

141

GG 10

AG 49

AA 80

GG 65

AG 47

AA 27

Exon 3 rs1124622 VNTR X-X X-7 CC AC 46 95 28 65

72

1

33

37

25

34

13

22

50

20

20

AA 36 24

CD No CD

Individual SNPS

Figure 4. Genotype frequencies in groups with & without Conduct Disorder.


205

Table 4. Estimated DRD4 haplotype frequencies with their standard errors for groups with & without conduct disorder for C_1611535, rs9366, the exon 3 VNTR, and rs11246226 Haplotype 1 (AAXA) 2 (AGXA) 3 (GAXA) 4 (GAXC) 5 (GA7C) 6 (GA7C) 7 (AGXC) 8 (GG7A) 9 (GG7C) 10 (GA7A) 11 (AA7C) 12 (AG7C) 13 (GGXC) 14 (AG7A) 15 (GGXA)

No Conduct Disorder Frequencies Standard Error .145 .029 .282 .037 .035 .015 .062 .020 .074 .022 .059 .019 .199 .033 .011 .009 .038 .016 .004 .005 .041 .016 0 0 0 0 .010 .008 .042 .017

Conduct Disorder Frequencies Standard Error .095 .018 .376 .029 .014 .007 .055 .014 .111 .019 .064 .015 .124 .020 0 0 .076 .016 .015 .007 .006 .005 .015 .007 .018 .008 .026 .010 .005 .004

DISCUSSION In a sample of 179 male adolescent delinquents and 60 matched controls, it has been found that variation in DRD4 is associated with risk for externalizing problems, for both incarceration and CD. In the present analyses, Incarceration and CD were studied separately; nonetheless, differential patterns of genotype and/or haplotype frequencies across the DRD4 gene indicate an association between the variation in this gene and broadly defined externalizing problems (Krueger, et al., 2002; Krueger & Johnson, 2008). Yet, in interpreting the results, it is important to discuss a number of issues. First, although our data supports a connection between the variation in DRD4 and externalizing problems, they do not explain the particular role of the 7-repeat allele of the exon 3 VNTR. The literature on the association between this allele and externalizing problems is contradictory (e.g., DeYoung, et al., 2006; Leung, et al., 2005); this particular allele might or might not substantiate the role of DRD4 in these behavioural problems. Second, the results presented here feature the C_1611535 (rs58688157) and rs11246226 polymorphisms as interesting markers to consider in further analyses. To our knowledge, these SNPs have not been previously investigated in the context of studying externalizing behaviour and, thus, might be of interest in future studies. Third, it is important to note that the analyses of allele frequencies did not reveal any significant differences for either of the two grouping analyses. This highlights the importance of analysing polymorphisms of DRD4 or any candidate gene at different levels, covering allele, genotype, and haplotype frequencies. Fourth, haplotype frequencies showed significant associations with both

206


criminality and conduct disorder. A haplotype refers to a cluster of polymorphisms within a gene (or across multiple genes) that might form a unit, transmitted across a generation. The haplotype analyses presented here revealed the most interesting findings. Analysing haplotypes (which is not common for psychological research) is vital because it can provide insight on factors influencing the dependency among genetic markers (Liu, Zhang, & Zhao, 2008). It is important, in conducting analyses of genetic bases of complex human behaviours, to acknowledge the complexity of genes as structures. Genes cannot be adequately captured by single polymorphisms but can be represented by multiple measurements, similar to what is used for the representations of complex latent variables. These results demonstrate the advantage of using multiple markers clustered into haplotypes in considering group differences. Haplotypes appear to be more informative in differentiating groups of interest, seem to result in a higher power of differentiation, and are less vulnerable to the adjustments of p-values since they require fewer comparisons. Finally, this work has a number of limitations to consider. First, genetic association studies have a tendency to lack the power to detect a statistically significant association (Salanti, Sanderson, & Higgins, 2005) and, thus, replications are essential (Fonseca & Denton, 2007). Despite the fact that information provided by genetic association studies is valuable, we must interpret these associations with great caution.

CONCLUSION In conclusion, the present study adds to the research on the dopamine D4 receptor genes‘ relationship with externalizing behaviours, re-assures the presence of this association, and provides some of the first support for the role of the C_1611535 (rs58688157) and rs11246226 sites in contributing to these relationships.

ACKNOWLEDGMENT This research was supported by funds from the American Psychological Foundation and Successful Intelligence to Elena L. Grigorenko; by funds from the Swedish Royal Academy of Sciences to Britt af Klinteberg and Elena L. Grigorenko; by funds from the Soderstrom— Koenig Fund to Lars Oreland and Britt af Klinteberg; by funds from the Mobilizing Against Drugs Committee, Ministry of Health and Social Affairs, Sweden, to Britt af Klinteberg and Lars Oreland; by funds from Swedish Medical Research Council Grant 4145 to Lars Oreland; and by funds from the Rutgers Aresty Research Grant to Tara Thompson. We express our gratitude to Ms. Mei Tan for her editorial assistance. We are also grateful to the participants in this research, for their willingness to collaborate, and for their time and efforts.


207

REFERENCES Arias-Carrión, O., & Pöppel, E. (2007). Dopamine, learning and reward-seeking behaviour. Act Neurobiol Exp, 67, 481-488. Dawei, L., Sham, P. C., Owen, M. J., & He, L. (2006). Meta-analysis shows significant association between dopamine system genes and attention deficit hyperactivity disorder (ADHD). New York, NY: Oxford University Press. DeYoung, C. G., Peterson, J. B., Seguin, J. R., Mejia, J. M., Pihl, R. O., Beitchman, J. H., et al. (2006). The dopamine D4 receptor gene and moderation of the association between externalizing behaviour and IQ. Archives of General Psychiatry, 63, 1410-1416. Fonseca, C., & Denton, C. P. (2007). Genetic association studies in systemic sclerosis: more evidence of a complex disease. Journal of Rheumatology, 34, 903-905. Genro, J. P., Kieling, C., Rohde, L. A., & Hutz, M. H. (2010). Attention-deficit/hyperactivity disorder and the dopaminergic hypotheses. Expert Review of Neurotherapeutics, 10, 587601. Hawley, M. E., & Kidd, K. K. (1995). HAPLO: a program using the EM algorithm to estimate the frequencies of multi-site haplotypes. Journal of Heredity, 86, 409-411. Hejjas, K., Vas, J., Topal, J., Szanthai, E., Ronai, Z., Szekely, A., et al. (2007). Association of polymorphisms in the dopamine D4 receptor gene and the activity-impulsivity endophenotype in dogs. Animal Genetics, 38, 629-633. James, A. S., Groman, S. M., Sev, E., Jorgensen, M., Fairbanks, L. A., & Jentsch, J. D. (2007). Dimensions of Impulsivity are associated with poor spatial working memory performance in monkeys. Journal of Neuroscience, 27, 14358-14364. Kaufman, J., Birmaher, B., Brent, D., Rao, U., Flynn, C., Moreci, P., et al. (1997). Schedule for Affective Disorders and Schizophrenia for School-Age Children—Present and Lifetime Version (K–SADS–PL): Initial reliability and validity data. Journal of the American Academy of Child and Adolescent Psychiatry, 36, 980-988. Kim-Cohen, J., Arseneault, L., Caspi, A., Tomas, M. P., Taylor, A., & Moffitt, T. E. (2005). Validity of DSM-IV conduct disorder in 4 1/2 and 5 year old children: A longitudinal epidemiological study. American Journal of Psychiatry, 162, 1108-1117. Krueger, R. F. (1999). The structure of common mental disorders. Archives of General Psychiatry, 56, 921-926. Krueger, R. F., Hicks, B. M., Patrick, C. J., Carlson, S. R., Iacono, W. G., & McGue, M. (2002). Etiologic connections among substance dependence, antisocial behaviour, and personality: Modeling the externalizing spectrum. Journal of Abnormal Psychology, 111, 411-424. Krueger, R. F., & Johnson, W. (2008). Behaviour genetics and personality: A new look at the integration of nature and nurture. In O. P. John, J. R. Robbins & L. A. Pervin (Eds.), Handbook of personality (pp. 287-310). New York, NY: Guilford Press. Krueger, R. F., & Tackett, J. L. (2006). Personality and psychopathy. New York, NY: Guilford. Lahey, B. B., Applegate, B., McBurnett, K., Biederman, J., Greenhill, L., & Hynd, G. W. (1994). DSM-IV field trials for Attention Deficit Hyperactivity Disorder in children and adolescents. American Journal of Psychiatry, 151, 1673-1685.

208


Langley, K., Marshall, L., van den Bree, M., Thomas, H., Owen, M., O'Donovan, M., et al. (2004). Association of the dopamine D4 receptor gene 7-repeat allele with neuropsychological test performance of children with ADHD. American Journal of Psychiatry, 161, 133-138. Leung, P. W. L., Lee, C. C., Hung, S. F., Ho, T. P., Tang, C. P., Kwong, S. L., et al. (2005). Dopamine receptor D4 (DRD4) gene in Han Chinese children with attentiondeficit/hyperactivity disorder (ADHD): increased prevalence of the 2-repeat allele. American. Journal of Medicine, 133B, 54-56. Liu, N., Zhang, K., & Zhao, H. (2008). Haplotype-association analysis. Advances in Genetics, 60, 335-405. McBurnett, K., & Pfiffner, L. (Eds.). (2007). Attention Deficit Hyperactivity Disorder: Concepts, controversies, new directions New York, NY: Informa. Moffitt, T. E., Arseneault, L., Jaffee, S. R., Kim-Cohen, J., Koenen, K. C., Odgers, C. L., et al. (2008). Research review: DSM-V conduct disorder: research needs for an evidence base. Journal of Child Psychology & Psychiatry & Allied Disciplines, 49, 3-33. Peciña, S., Cagniard, B., Berridge, K., Aldridge, J., & Zhuang, X. (2003). Hyperdopaminergic mutant mice have higher "wanting" but not "liking" for sweet rewards. Journal of Neuroscience, 23, 9395-9402. Rhee, S. H., & Waldman, I. D. (2002). Genetic and environmental influences on antisocial behaviour: A meta-analysis of twin and adoption studies. Psychological Bulletin, 128, 490-529. Salanti, G., Sanderson, S., & Higgins, J. P. (2005). Obstacles and opportunities in metaanalysis of genetic association studies. Genetics in Medicine, 7, 13-20. Seaman, M. I., Fisher, J. B., Chang, F., & Kidd, K. K. (1999). Tandem duplication polymorphism upstream of the dopamine D4 receptor gene (DRD4). American Journal of Medical Genetics. Part B, Neuropsychiatric Genetics, 88, 705-709. Sholevar, P. G. (1995). Conduct disorder in children and adolescents. Washington, DC: APA. Slutske, W. S., Heath, A. C., Dinwiddie, S. H., Madden, P. A. F., Bucholz, K. K., Dunne, M. P., et al. (1997). Modeling genetic and environmental influences in the etiology of conduct disorder: A study of 2,682 adult twin pairs. Journal of Abnormal Psychology, 106, 266-279.



Chapter 18

THE ASSESSMENT OF DEPRESSION, ANXIETY AND TENSION/STRESS IN YOUTH Marianna Szabó School of Psychology, University of Sydney, Sydney, NSW 2006, Australia

ABSTRACT The Depression Anxiety Stress Scales (DASS; S. H. Lovibond & Lovibond, 1995) is a widely used measure of negative affect in adults, both in normal and clinic-referred populations. It was originally developed to assess the full range of core symptoms of anxiety and depression while also providing maximal differentiation between these two affective states. During its development, a third specific factor emerged empirically, and was labelled ‗stress‘ or ‗tension/stress‘ by the researchers. This third emotional state has subsequently been shown to be uniquely associated with excessive and uncontrollable worrying in adults. A measure similar to the DASS is not yet available to assess negative emotional states in children and adolescents. Nevertheless, extant research suggests that there are similarities in the structure of negative affect across different developmental stages. For example, the core symptoms of depression in primary-school aged children and in adolescents appear to be highly similar to those previously identified in adults by S. H. Lovibond and Lovibond. However, anxiety and tension/stress has not yet been successfully differentiated from each other as two specific emotional states in respondents younger that 15-18 years of age. Further research is now underway to identify the age at which these two emotional states are first observable via self-report questionnaires, and to develop a new measure of anxiety, depression and tension/stress for children and adolescents.

Keywords: Adolescence, Anxiety, Children, Depression, Depression Anxiety Stress Scales, Stress




210

Marianna Szabó

INTRODUCTION Anxiety and mood disorders are among the most prevalent psychiatric disorders in children and adolescents (Sawyer, Arney, Baghurst, Clark, Graetz, Kosky, et. al., 2000), as well as in adults (Andrews, Hall, Teesson, & Henderson, 1999). These disorders often persist into adulthood and increase young people‘s risk for other mental health problems, for example, drug and alcohol abuse, eating disorders, and suicide (Bulik, Sullivan, Fear, & Joyce, 1997; Hale, Raaijmakers, Muris, Van Hoof, & Meeus, 2008; Kim-Cohen, Caspi, Moffitt, Harrington, Milne, & Poulton, 2003). To improve mental health and reduce the burden of disease, a better understanding, prevention and treatment of anxiety and depression in youth is crucial. A critical issue hindering progress in this field, however, is our limited knowledge concerning the conceptualisation and valid assessment of negative affective states in children and adolescents (Andrews, Szabó, & Burns, 2002; Cartwright-Hatton, 2006; Schniering, Hudson, & Rapee, 2000). As in many other areas of research, developments in our understanding and assessment of negative emotion in youth have lagged behind advances in adult research (Clark & Watson, l991; S. H. Lovibond & Lovibond, 1995). Such advances have initially been stimulated by reports of a limited discriminant power of popular assessment tools. Scales purporting to assess anxiety frequently correlated as highly with depression scales as with other anxiety scales, and depression scales showed a similar lack of specificity. Very high rates of comorbidity between anxiety and depressive disorders were also seen as partly reflecting a difficulty differentiating between these two constructs in clinical settings (Clark & Watson, 1991). Mirroring similar findings in adults, popular child self-report instruments such as the State-Trait Anxiety Inventory for Children (STAIC; Spielberger, Edwards, Montuon, & Lushene, 1970) or the Children's Depression Inventory (CDI; Kovacs, 1981) have been shown to have a limited discriminant power and to assess a mixture of symptoms of both affective states as well as symptoms of general distress (Cole, Truglio, & Peeke, 1997; Finch, Lipovsky, & Casat, 1989; Joiner, Catanzaro, & Laurent, 1996; Szabó & Lovibond, 2006).

IDENTIFYING THE SPECIFIC SYMPTOMS OF ANXIETY AND DEPRESSION To explain such findings in adult populations, Clark and Watson (1991) suggested that anxiety and depression share certain characteristics with each other, but that they also have unique symptoms, specific to each. To better differentiate anxiety and depression from each other, assessments of these two emotional states need to de-emphasise their common characteristics and emphasise their specific symptoms. In a series of factor analytic studies, these researchers (e.g., Watson, Clark, Weber, Assenheimer, Strauss, & McCormick, 1995) identified symptoms of Physiological Hyperarousal (PH; e.g., racing heart, sweaty palms, dry mouth) as specific to anxiety, and low Positive Affect (PA; i.e., a lack of pleasure or enthusiasm) as specific to depression. They suggested that symptoms of irritability, tension, worry, sad or anxious mood, concentration or decision making difficulties, or sleep and appetite disturbances are common to both anxiety and depression, and they reflect Negative Affect (NA) or psychological distress in general.

The Assessment of Depression, Anxiety and Tension/Stress in Youth

211

While this ‗tripartite model‘ of negative affective states (Clark & Watson, 1991) has made important contributions to current knowledge, it has also been criticised for its overly restrictive definition of anxiety and depression. For example, Burns and Eidelson (1998) pointed out that physiological arousal on its own does not assess the core aspect of anxiety, since it can also occur in many other states, such as intense anger or excitement. Similarly, low positive affect is likely to be only one of several specific features of depression. Other such features also need to be identified and included in any comprehensive assessment tool.

THE DEPRESSION ANXIETY STRESS SCALES (DASS) Consistent with Burns and Eidelson‘s (1998) critique, S. H. Lovibond and Lovibond (1995) aimed to develop a self-report instrument that is able to assess the full range of core symptoms of anxiety and depression, while obtaining maximal discrimination between the two affective states. The results of a series of factor analyses in more than 30 samples indicated that the unique symptoms of anxiety include a subjective awareness of anxious affect and escape or avoidance tendencies, as well as physiological hyperarousal. The unique symptoms of depression were found to include hopelessness, devaluation of life, self deprecation and inertia, in addition to low positive affect. The item content of the Depression Anxiety Stress Scales (DASS, S. H. Lovibond & Lovibond, 1995) reflects these additional symptoms. Similar to Watson et al.‘s (1995) findings, Lovibond and Lovibond reported that several other symptoms, including appetite and sleep disturbances, fatigue, or poor concentration, had no specific loading on either the Depression or the Anxiety factor. These items were considered to indicate emotional distress or negative affect in general and were excluded from the DASS in order to increase its specificity and discriminant power. A unique feature of the DASS is its inclusion of a Stress scale in addition to the Depression and Anxiety scales. This scale emerged empirically during the development of the Depression and Anxiety scales, via an aggregation of such items as difficulty relaxing, tension, impatience, irritability and agitation. It was initially labelled ‗Stress‘ or ‗Tension/Stress‘ by the researchers (S. H. Lovibond & Lovibond, 1995). Knowledge about the precise nature of the construct assessed by this scale is still developing. Some researchers (e.g., Brown, Korotitsch, Chorpita, & Barlow, 1997) argue that scores on the Stress scale largely reflect the construct of general distress or Negative Affect (NA), as described within the framework of the tripartite model (Clark & Watson, 1991). However, others have provided compelling evidence that the Stress scale in fact measures a specific, independent emotional syndrome, empirically distinguishable from a common general negative affect construct as well as from anxiety or depression (Henry & Crawford, 2005; Lovibond, 1998). For example, Henry and Crawford (2005) conducted a stringent test of the specificity of the three DASS scales in a large adult sample by testing whether a ‗quadripartite‘ structure underlies the DASS. This quadripartite model builds upon the original DASS model first proposed by S. H. Lovibond and Lovibond (1995), which involved the three correlated factors of Depression, Anxiety and Stress (see Figure 1).

212

Marianna Szabó

Figure 1. The correlated three-factor model of Depression, Anxiety & Stress underlying the 21-item brief version of the DASS, originally proposed by S. H. Lovibond & Lovibond (1995). Note: Error terms are omitted for reasons of clarity.


213

Figure 2. Graphical representation of the ‗Quadripartite‘ model originally proposed by Crawford & Henry (2005). Note: Error terms are omitted for reasons of clarity.

In the quadripartite model, these three factors are assumed to be orthogonal, and a fourth orthogonal factor, labelled ‗General Psychological Distress‘, is added. The Anxiety, Depression and Stress factors are each defined by the items originally specified by S. H. Lovibond and Lovibond. However, each item is also allowed to load on the fourth, common

214

Marianna Szabó

General Distress factor. Thus, the quadripartite model implies that scores on each item of the DASS reflect a common underlying construct as well as their specific Anxiety, Depression or Stress constructs (see Figure 2). To further explore whether Stress is in fact a specific emotional state or, alternatively, it is synonymous with the construct of ‗general distress‘, Henry and Crawford compared the quadripartite model with an alternative model in which no specific Stress factor was included and all Stress items were instead set to load on the common ‗General Distress‘ factor only (see Figure 3). Results showed that all three scales of the DASS index a substantial proportion of common variance, explaining the correlations among their manifest scores. However, items on each scale were also found to contain a significant amount of specific variance, unique to their underlying factor. The quadripartite model was shown to be superior to the alternative model, and was reported as the optimal model to summarise adults‘ responses to the DASS-21. These results provide strong supportive evidence for the specificity of the items comprising the DASS, and for the specificity of the Stress construct in particular. Concerning the nature of this construct, it has recently been shown that scores on the DASS Stress scale are strongly associated with a diagnosis of DSM-IV Generalised Anxiety Disorder (GAD, American Psychiatric Association, 2000) in clinical samples and with excessive and uncontrollable worrying in general populations (Brown et al., 1997; Szabó, in press; Huang, Szabó, & Han, 2009). Moreover, items that form the DASS Stress scale have uniformly strong associations with worrying (Szabó, in press). These studies are the first to demonstrate that GAD and worrying are associated with a specific emotional syndrome. Emphasising the symptoms reflected in the DASS Stress items in future revisions of DSM may lead to increased discriminant validity of the diagnostic category of GAD, and to a more clear delineation of its boundaries from depression and from the other anxiety disorders (cf. Andrews, Hobbs, Borkovec, Beesdo, Craske, Heimberg, et al., 2010). Since its first publication, several independent research groups have reported on the favourable psychometric properties of the DASS, both in its original 42-item version and its brief 21-item version, in clinical and in non-clinical samples (Antony, Bieling, Cox, Enns, & Swinson, 1998); Brown et al., 1997; Clara, Cox, & Enns, 2001; Crawford & Henry, 2003; Henry & Crawford, 2005; Lovibond, 1998; P. F. Lovibond & Lovibond, 1995; Taylor, Lovibond, Nicholas, Cayley, & Wilson, 2005). It has been translated into 24 languages and shown to be a reliable and valid measure in several cultures (de Beurs, Van Dyck, Marquenie, Lange, & Blonk, 2001; Daza, Novy, Stanley, & Averill, 2002; Norton, 2007). Its factor structure is stable and essentially the same in clinical and nonclinical samples (Antony et al., 1998; Brown et al., 1997; Clara et al., 2001), supporting the idea that clinical disorders represent an extreme or maladaptive manifestation of basic emotional syndromes. Thus, the DASS is a unique self-report measure in that it is able to assess the full range of the specific symptoms of both anxiety and depression, as well as of stress or ‗tension/stress‘ in one brief instrument. It measures pure emotional states without reference to stimuli that evoke these states (i.e., social situations, animals, contamination) making it useful as a measure of anxiety or depression associated with a broad range of DSM-defined disorders. At the same time, because it has been developed empirically, independent of DSM-derived constructs, its use is unaffected by any future changes to DSM diagnostic criteria.


215

THE ASSESSMENT OF NEGATIVE AFFECT IN CHILDREN AND ADOLESCENTS Similar to the adult instruments from which they have been developed, popular child selfreport measures of negative emotion have been shown to have limited discriminant validity (e.g., Finch et al., 1989). Recent attempts to compensate for the limitations of these measures have taken two different directions. Some researchers (e.g., Chorpita, Yim, Moffitt, Umemoto, & Francis, 2000; Spence, 1998) constructed new self-report scales for 7-18-yearolds to assess dimensions of anxiety and mood disorder symptoms as defined by DSM-IV-TR (APA, 2000). Others (e.g., Chorpita, Daleiden, Moffitt, Yim, & Umemoto, 2000; Laurent, Catanzaro, & Joiner, 2004; Laurent, Catanzaro, Joiner, Rudolph, Potter, Lambert, et al., 1999) developed measures of pure emotional states independent of DSM-defined constructs, typically using Clark and Watson‘s (1991) tripartite model of negative affect as a guide to item selection. In spite of its central role in the assessment of negative emotional states in adults, no selfreport questionnaire similar to the adult DASS (S. H. Lovibond & Lovibond, 1995) is as yet available to assess negative emotion in children and adolescents. Such an instrument would fill an important need by providing assessments of pure emotional states that transcend DSMderived constructs, as well as by complementing the more restrictive item content of measures based on the tripartite model. Importantly, none of the existing measures of childhood negative affect is able to assess the ‗stress‘ or ‗tension/stress‘ construct in youth. Establishing the age at which a specific tension/stress syndrome is first distinguishable from anxiety, depression and general negative affect would have important implications for our better understanding of the development of increasingly differentiated emotional states. In addition, it would represent a crucial first step towards identifying the emotional symptoms associated with excessive worrying and generalised anxiety disorder in youth, as this question has remained unanswered to date (Kendall & Pimentall, 2003; Tracey, Chorpita, Douban, & Barlow, 1997; Szabó & Lovibond, 2004).

DEVELOPING AGE-APPROPRIATE ITEMS TO ASSESS DEPRESSION, ANXIETY AND TENSION/STRESS So far only one published study attempted to assess depression, anxiety and tension/stress, as defined by the DASS, in primary-school-aged children (Szabó & Lovibond, 2006). A sample of 577 7-14-year-olds completed a 76-item exploratory questionnaire that included the item content of established measures of childhood anxiety and depression (i.e., STAIC, CDI) as well as new items developed by the authors. A series of exploratory factor analyses was conducted. The discriminant validity of such symptoms as fatigue, indecision, and appetite or sleep changes was first evaluated. Although these symptoms are included in popular measures of childhood negative affect, it was hypothesised that, as in adults, they would reflect general distress and have no strong specific association to either anxiety or depression. Findings provided strong support for the hypothesis and suggest that, as in adults,

216

Marianna Szabó

measures of negative affect in children should de-emphasise such symptoms to maximise specificity and discriminant power. The analyses further indicated that the core symptoms of depression in 7-14-year-olds are highly similar to those previously identified in adults and reflected in the item content of the DASS. In addition to using items from existing scales (i.e., STAIC, CDI) we developed new items to assess symptoms that are not included in current youth depression scales but are core symptoms of depression as defined by the DASS (e.g., devaluation of life, hopelessness, inertia). Each item was found to have strong specific associations with the underlying ‗depression‘ construct in our analyses. These results are in accordance with earlier suggestions (S. H. Lovibond & Lovibond, 1995) that in adults depression is defined by a broader range of symptoms than low Positive Affect alone (c.f. Watson et al., 1995), and extend this finding to children and adolescents. To date, no existing child self-report measure of depression provides a comprehensive coverage of such core symptoms while also deemphasising non-specific symptoms. Although a Depression factor was successfully identified, we were unable to differentiate the anxiety and tension/stress constructs in this study. Instead of a 3-factor model of Depression, Anxiety and Stress, a 2-factor model involving Depression and a combined Anxiety-Stress factor appeared to best summarise the data. These results are consistent with suggestions that negative emotional states become increasingly differentiated as young people mature (Cole, et al., 1997). However, our inability to differentiate the anxiety and tension/stress constructs may have also reflected the limitations of the draft items developed for this study. For example, the adult DASS items concerning autonomic arousal point it out to the respondents that they refer to autonomic arousal experienced in the absence of physical exercise (e.g., I was aware of the action of my heart in the absence of physical exertion). Such information was omitted from the simplified items to increase their readability and comprehension (e.g., ‘My heart was beating fast’). However, omitting this information may have reduced the items‘ validity and precision. Similarly, items intended to assess the irritability and impatience aspects of tension/stress may have reflected anger or externalising tendencies in youth, needing further refinement.

THE LATENT STRUCTURE OF THE DASS IN ADOLESCENT SAMPLES Reflecting the need to provide a psychometrically sound self-report measure of negative affect for youth, researchers have begun to explore the construct validity of the adult DASS in adolescents. First, Duffy, Cunningham, and Moore (2005) conducted Confirmatory Factor Analyses (CFA) to test whether the underlying constructs of depression, anxiety and tension/stress are observable in 11-15-year-olds, using the brief 21-item version of the adult DASS. Results did not support the hypothesised three-factor structure. After testing several alternative models, the authors concluded that young adolescents do not differentiate between depression, anxiety, and stress when responding to the adult DASS-21; instead they report on physiological arousal and generalised negative affective states. However, Duffy et al.‘s data may have reflected the characteristics of a small sample recruited from a single school. Tully, Zajac, and Venning (2009) included larger and more diverse samples of both younger (aged 12-14 years) and older (aged 15-18 years) adolescents in a similar study. These


217

researchers found that a model involving a Depression and an Anxiety factor, with all Stress items loading on a common factor, fit the data best (see Figure 3). Although the authors claim that their results provide evidence for Clark and Watsons‘s (1991) tripartite model, they in fact did not select items to assess low Positive Affect and Physiological Hyperarousal as defined by that model. Instead, Tully et al.‘s results appear to show that while Depression and Anxiety, as defined by the DASS, may be assessable via the adult DASS-21 in adolescents, no specific Stress factor is observable. However, such a conclusion may be premature, as the authors‘ omission to test the ‗quadripartite‘ model (Henry & Crawford, 2005; see Figure 2) in that study greatly limits the interpretability of their findings. So far, only two studies have tested the quadripartite model in adolescent samples. Szabó (2010) explored the underlying structure of young adolescents‘ responses to the DASS-21 in a sample of 481 11-15-year-olds. A correlated three-factor model comprising Depression, Anxiety and Stress, as originally proposed by S. H. Lovibond and Lovibond (1995; see Figure 1) was tested first and found to fit the data adequately. However, correlations among the three factors, especially between Anxiety and Stress, were very high. Several alternative models were tested to explain such correlations. These included a one-factor model, a two-factor model involving a Depression and a combined Anxiety-Stress factor (Szabó & Lovibond, 2006), and the quadripartite and ‗general distress‘ models first offered by Henry and Crawford (2005, see Figures 2 and 3, respectively). As in adults, the quadripartite model was found to have superior fit to all other models. Specific factor loadings in the quadripartite model were less clear than those previously found in adult samples, however. Six of the seven Depression items and five of the seven Anxiety items had significant loadings on their own factor, in addition to the common factor. The only item that originally intended to assess Depression but failed to load on the specific Depression factor reflected a lack of initiative (I found it difficult to work up the initiative to do things). Interestingly, this item has been found to perform poorly in adult samples as well (e.g., Antony et al., 1998). Both items that failed to load on the Anxiety factor were originally intended to assess physiological hyperarousal (‘I experienced breathing difficulty (e.g., excessively rapid breathing, breathlessness in the absence of physical exertion)’; and ‘I was aware of dryness of my mouth’). The poor performance of these items may reflect either comprehension difficulties or a limited awareness of the somatic symptoms of anxiety among young adolescents. Importantly, the majority of the items originally aimed to assess tension/stress failed to have significant positive loadings on their allocated Stress factor, in addition to their loadings on the common factor. These results question whether 11-15-yearolds experience the specific emotional state of tension/stress, or whether they are able to report on this experience via the items comprising the adult DASS. Again, it is likely that the findings have been influenced by the complex wording of some of the items on the adult DASS Stress scale. Adolescents asked for clarification for several of these items, and missing values were most common when items included such words as ‗tolerate‘, ‗agitated‘ or ‗irritable‘, most of which belong to the Stress scale (Szabó, 2010). We conducted a follow up study to test whether older adolescents‘ responses to the adult DASS reflect a clearer underlying symptom structure (Szabó, 2009 November). A sample of 690 15-18-year-olds completed the DASS-21 and Confirmatory Factor Analyses were used to test alternative models. Although the original correlated three-factor DASS model (S. H. Lovibond & Lovibond, 1995) achieved adequate fit, correlations among the factors were again very high. Therefore, several other models were tested, following the procedures

218

Marianna Szabó

reported by Szabó (2010). As expected, results showed that the quadripartite model provided the best summary of the underlying structure of 15-18-year-old adolescents‘ responses to the adult DASS-21. Importantly, extending Tully et al.‘s (2009) findings, the quadripartite model provided superiour fit to a model implying that DASS Stress items assess a common ‗General Psychological Distress‘ factor only (see Figure 3).

Figure 3. Graphical representation of the ‗General Distress‘ model originally proposed by Crawford & Henry (2005). Note. Error terms are omitted for reasons of clarity.


219

An inspection of item loadings in the quadripartite model indicated that each of the 7 items originally allocated to the Depression factor had significant positive loadings on this factor, in addition to the common factor. As expected, the item reflecting ‗a lack of initiative’ again had a very low, albeit statistically significant, loading. All seven ‗Anxiety‘ items loaded as expected, suggesting that the symptom structure of Anxiety is highly similar in 15-18-yearolds to that previously found in adults. However, only five of the seven ‗Stress‘ items had significant positive loadings on their allocated factor. Two items that were hypothesised to reflect the Stress construct but failed to contribute significant amount of variance to the specific Stress factor were ‗I felt that I was using a lot of nervous energy’ and ‘I found it difficult to relax’ (Szabó, 2009 November).

CONCLUSION While new research is developing to provide improved assessment of negative affect in children and adolescents, our knowledge concerning the specific core symptoms of depression, anxiety and tension/stress across developmental stages is still limited. Extant data to date suggest that there are age-related differences in the extent to which the constructs of Depression, Anxiety and Stress are observable via self-report. It appears that the core symptoms of Depression are highly similar in adults (S. H. Lovibond & Lovibond, 1995), older adolescents (Szabó, 2009 November), younger adolescents (Szabó, 2010), and primary school aged children (Szabó & Lovibond, 2006). In each of our studies, a clear Depression factor was found, defined by symptoms of hopelessness, devaluation of life, anhedonia, self deprecation and (albeit less strongly), inertia. In contrast, clear anxiety and stress constructs were not yet distinguishable in primaryschool-aged samples. Although an ‗anxiety‘ construct similar to that previously found in adults was observable in young adolescents (Szabó, 2010), we were only able to assess ‗stress‘ via self report in individuals older than 15 years of age (Szabó, 2009 November). The age at which this specific emotional state first emerges is currently unknown. Further work towards identifying the specific core symptoms of depression, anxiety and stress in youth is now needed. In addition to its theoretical importance, such work would lead to improved assessment of depression and anxiety, as well as providing the first self-report measure to assess tension/stress in youth. An ability to assess tension/stress would enable us to document any age-related changes in the emotional correlates of excessive worrying and Generalised Anxiety Disorder, and would contribute to a better understanding of the development of this chronic disorder. Our research group is currently examining the psychometric properties of original and modified versions of the DASS at different ages. These studies will contribute to establishing the lower age limit for the reliable and valid use of the adult DASS in adolescent populations, and will inform our work towards constructing age-appropriate items to assess depression, anxiety and tension/stress in children. Ultimately, improved assessment will facilitate detection, prevention and treatment of a wide variety of emotional disturbances involving depression, anxiety, stress and excessive worrying in youth.

220

Marianna Szabó

REFERENCES American Psychiatric Association (2000). Diagnostic and statistical manual of mental disorders (4th Ed.) Text Revision (DSM-IV-TR). Washington D.C.: American Psychiatric Association. Andrews, G., Hall, W., Teesson, M., & Henderson, S. (1999). The mental health of Australians. Canberra: Mental Health Branch, Commonwealth Department of Health and Aged Care. Andrews, G., Hobbs, M. J., Borkovec, T. D., Beesdo, K., Craske, M. G., Heimberg, R. G., Rapee, R. M., Ruscio, A. M., & Stanley, M. A. (2010). Generalized worry disorder: A review of DSM-IV generalized anxiety disorder and options for DSM-V. Depression and Anxiety, 27, 134-147. Andrews, G., Szabó, M., & Burns, J. (2002). Preventing major depression in young people. British Journal of Psychiatry, 181, 460-462. Antony, M. M., Bieling, P. J., Cox, B. J., Enns, M. W., & Swinson, R. P. (1998). Psychometric properties of the 42-item and 21-item versions of the Depression Anxiety Stress Scales (DASS) in clinical groups and a community sample. Psychological Assessment, 10, 176-181. Brown, T. A., Korotitsch, W., Chorpita, B. F., & Barlow, D. H. (1997). Psychometric properties of the Depression Anxiety Stress Scales (DASS) in clinical samples. Behaviour Research and Therapy, 35, 79-89. Bulik, C. M., Sullivan, P. F., Fear, J. L., & Joyce, P. R. (1997). Eating disorders and antecedent anxiety disorders: a controlled study. Acta Psychiatrica Scandinavia, 96, 101107. Burns, D. D., & Eidelson, R. J. (1998).Why are depression and anxiety correlated? A test of the tripartite model. Journal of Consulting and Clinical Psychology, 66, 461-473. Cartwright-Hatton, S. (2006). Worry in childhood and adolescence. In G. L. C. Davey & A. Wells (Eds.), Worry and its psychological disorders: Theory, assessment and treatment (pp. 81-97). Chichester: Wiley. Chorpita, B. F., Daleiden, E. L., Moffitt, C., Yim, L., & Umemoto, L. A. (2000). Assessment of tripartite factors of emotion in children and adolescents I: Structural validity and normative data of an affect and arousal scale. Journal of Psychopathology and Behavioral Assessment, 22, 141-160. Chorpita, B. F., Yim, L., Moffitt, C. E., Umemoto, L. A., & Francis, S. E. (2000). Assessment of symptoms of DSM-IV anxiety and depression in children: A Revised Child Anxiety and Depression Scale. Behaviour Research and Therapy, 38, 835-855. Chorpita, B. F., Tracey, S. A., Brown, T. A., Collica, T. J., & Barlow, D. H. (1997). Assessment of worry in children and adolescents: An adaptation of the Penn State Worry Questionnaire. Behaviour Research and Therapy, 35, 569-581. Clara, I. P., Cox, B. J., & Enns, M. W. (2001). Confirmatory factor analysis of the Depression Anxiety Stress Scales in depressed and anxious patients. Journal of Psychopathology and Behavioral Assessment, 23, 61-67. Clark, L. A., & Watson, D. (l991). Tripartite model of anxiety and depression: Psychometric evidence and taxonomic implications. Journal of Abnormal Psychology, 100, 316-336.


221

Cole, D. A., Truglio, R., & Peeke, L. (1997). Relation between symptoms of anxiety and depression in children: A multitrait-multimethod-multigroup assessment. Journal of Consulting and Clinical Psychology, 65, 110-119. Crawford, J. R., & Henry, J. D. (2003). The Depression Anxiety Stress Scales (DASS): Normative data and latent structure in a large non-clinical sample. British Journal of Clinical Psychology, 42, 111-131. Daza, P., Novy, D. M., Stanley, M. A., & Averill, P. (2002). The Depression Anxiety Stress Scale-21: Spanish translation and validation with a Hispanic sample. Journal of Psychopathology and Behavioral Assessment, 24, 195-205. de Beurs, E., Van Dyck, R., Marquenie, L. A., Lange, A., & Blonk, R. W. B. (2001). The DASS: A questionnaire for the measurement of depression, anxiety, and stress. Gedragstherapie, 34, 35-53. Duffy, C. J., Cunningham, E. G., & Moore, S. M. (2005). The factor structure of mood states in an early adolescent sample. Journal of Adolescence, 28, 677-680. Finch, A. J., Lipovsky, J. A., & Casat, C. D. (1989). Anxiety and depression in children and adolescents: Negative affectivity or separate constructs? In P. Kendall & D. Watson (Eds.), Anxiety and depression: Distinctive and overlapping features(pp. 171-202). New York: Academic Press. Hale III, W. W., Raaijmakers, Q., Muris, P., Van Hoof, A., & Meeus, W. (2008). Developmental trajectories of adolescent anxiety disorder symptoms: a 5-year prospective community study. Journal of the American Academy of Child and Adolescent Psychiatry, 47, 556-564. Henry, J. D., & Crawford, J. R. (2005). The short-form version of the Depression Anxiety Stress Scales (DASS-21): Construct validity and normative data in a large non-clinical sample. British Journal of Clinical Psychology, 44, 227-239. Huang, K., Szabó, M., & Han, J. (2009). The relationship of low distress tolerance to excessive worrying and cognitive avoidance. Behaviour Change, 26, 223-234. Joiner, T. E. Jr., Catanzaro, S. J., & Laurent, J. (1996). Tripartite structure of positive and negative affect, depression, and anxiety in child and adolescent psychiatric inpatients. Journal of Abnormal Psychology, 105, 401-409. Kendall, P. C. & Pimentell, S. S. (2003). On the physiological symptom constellation in youth with Generalized Anxiety Disorder (GAD). Journal of Anxiety Disorders, 17, 211221. Kim-Cohen, J., Caspi, A., Moffitt, T. E., Harrington, H., Milne, B. J., & Poulton, R. (2003). Prior juvenile diagnoses in adults with mental disorder: Developmental follow-back of a prospective-longitudinal cohort. Archives of General Psychiatry, 60, 709-717. Kovacs, M. (1981). Rating scales to assess depression in school-aged children. Acta Paedopsychiatrica, 46, 305-315. Laurent, J., Catanzaro, S. J., & Joiner, T. E. (2004). Development and preliminary validation of the Physiological Hyperarousal scale for children. Psychological Assessment, 16, 373380. Laurent, J., Catanzaro, S. J., Joiner, T. E., Rudolph, K. D., Potter, K. I., Lambert, S., et al. (1999). A measure of positive and negative affect for children: Scale development and preliminary validation. Psychological Assessment, 11, 326-338. Lovibond, P. F. (1998). Long-term stability of depression, anxiety, and stress syndromes. Journal of Abnormal Psychology, 107, 520-526.

222

Marianna Szabó

Lovibond, P. F., & Lovibond, S. H. (1995). The structure of negative emotional states: Comparison of the Depression Anxiety Stress Scales (DASS) with the Beck Depression and Anxiety Inventories. Behaviour Research and Therapy, 33, 335-343. Lovibond, S. H. & Lovibond, P. F. (1995). Manual for the Depression Anxiety Stress Scales. (2nd Ed.) Sydney: Psychology Foundation. Norton, P. J. (2007). Depression Anxiety and Stress Scales (DASS-21): Psychometric analysis across four racial groups. Anxiety, Stress & Coping, 20, 253-265. Sawyer, M. G., Arney, F. M., Baghurst, P. A., Clark, J. J., Graetz, B. W., Kosky, R. J., et. al. (2000). The mental health of young people in Australia. Canberra, ACT: Commonwealth Department of Health and Aged Care. Schniering, C. A., Hudson, J. L., & Rapee, R. M. (2000). Issues in the diagnosis and assessment of anxiety disorders in children and adolescents. Clinical Psychology Review, 20, 453-478. Spence, S. H. (1998). A measure of anxiety symptoms among children. Behaviour Research and Therapy, 36, 545-566. Spielberger, C. D., Edwards, C. D., Montuon, J., & Lushene, R. (1970). State-Trait Anxiety Inventory for Children. Palo Alto, CA: MindGarden. Szabó, M. (2009, November). The latent structure of the Depression Anxiety Stress Scales (DASS) in adolescent samples. Paper presented at the 8th Australasian Conference on Personality & Individual Differences, November 25-28, 2009, Sydney, NSW, Australia Szabó, M. (2010). The short version of the Depression Anxiety Stress Scales (DASS-21): Factor structure in a young adolescent sample. Journal of Adolescence, 33, 1-8. Szabó, M. (in press). The emotional experience associated with worrying: Anxiety, depression or stress? Anxiety, Stress, & Coping. Szabó, M., & Lovibond, P. F. (2006). Anxiety, depression and tension/stress in children. Journal of Psychopathology and Behavioral Assessment, 28, 195-205. Szabó, M. & Lovibond, P. F. (2004). The cognitive content of worry in clinic-referred anxious and non-anxious children. Journal of Clinical Child and Adolescent Psychology, 33, 613-622. Taylor, R., Lovibond, P. F., Nicholas, M. K., Cayley, C., & Wilson, P. H. (2005). The utility of somatic items in the assessment of depression in chronic pain patients: A comparison of the Zung Self-rating Depression Scale (SDS) and the Depression Anxiety Stress Scales (DASS) in chronic pain and clinical and community samples. Clinical Journal of Pain, 21, 91-100. Tracey, S. A., Chorpita, B. F., Douban, J., & Barlow, D. H. (1997). Empirical evaluation of DSM-IV Generalized Anxiety Disorder criteria in children and adolescents. Journal of Clinical Child Psychology, 26, 404-414. Tully, P. J., Zajac, I. T., & Venning, A. J. (2009). The structure of anxiety and depression in a normative sample of younger and older Australian adolescents. Journal of Abnormal Child Psychology, 37, 717-726. Watson, D., Clark, L. A., Weber, K., Assenheimer, J. S., Strauss, M. E., & McCormick, R. A. (1995). Testing a tripartite model: II. Exploring the symptom structure of anxiety and depression in student, adult and patient samples. Journal of Abnormal Psychology, 104, 15-25.



Chapter 19

UNCONSCIOUS SCHEMAS, BULIMIC TENDENCIES AND THE INFLUENCE OF SUBLIMINAL ABANDONMENT AND UNIFICATION CUES ON EATING Kathleen Mairet and Simon Boag Department of Psychology, Macquarie University, Sydney, NSW 2109, Australia

ABSTRACT Recent research suggests that negative schemas (i.e., core beliefs) that are not directly related to eating, shape and weight may play an important role in bulimia‘s manifestation (Jones, Harris & Leung, 2005). Since these schemas may be prevented from awareness due to defensive processes, subliminal stimulation may be an effective means for treatment and serve as an adjunct to current therapies. To examine this, a sample of one hundred and twenty undergraduate women was divided into one of three visual subliminal stimuli groups: an abandonment related cue ―lonely‖, a unification cue ―friendship‖ or a neutral cue ―gallery‖. After subliminal stimulus presentations participants were then left alone for 5 minutes with a bowl of biscuits to eat and the weight eaten was recorded. Contrary to what was predicted, women who received the subliminal abandonment cue did not eat significantly more than women in the control group. Furthermore, mixed support was found for Waller and Barter‘s (2005) suggestion that women presented with a subliminal unification cue would eat significantly less than the women in the control group. While an overall difference was not found, women in the unification group who spoke English as a second language ate significantly less than those women in the control group. The implications of the findings for both theory and treatment concerning bulimia are discussed, along with suggestions for future research.

Keywords: Abandonment, Bulimia, Defenses, Early maladaptive scheme, Subliminal, Unconscious processes 


224

Kathleen Mairet and Simon Boag

INTRODUCTION Eating disorders are notorious for their persistent course and poor prognoses (Jones et al., 2005). Cognitive-Behavioural Therapy (CBT) has been the most widely studied treatment for bulimia (Bowers & Andersen, 2007) and is generally considered the most effective form of therapy (Wilson, Grilo & Vitousek, 2007). This model emphasises the manifestation of dysfunctional attitudes and behaviours originating from societal expectations (sociocultural influences) concerning food, body shape and weight (Leung & Price, 2007). Recently, however, the effectiveness of CBT has been called into question with research indicating that, on average, only 50% of patients cease binge eating and purging behaviours subsequent to treatment (Halmi et al., 2002; Meyer et al., 2005; Wilson, 1996). Consequently, CBT appears to be valuable but not always sufficient for treatment. As such, building upon Young‘s (1999) Schema-Focused Therapy, Waller, Kennerley and Ohanian (2007) have recently proposed a Schema-Focused Cognitive-Behavioural Therapy which accentuates both the behavioural traits and underlying maladaptive beliefs (or schemas) associated with the disorder. There has been much debate within eating disorder literature as to whether individuals with bulimia are categorically different from sub-clinical bulimic and non-eating disordered individuals (the discontinuity hypothesis) or whether they represent the end-point of an eating disorder continuum (the continuity hypothesis) (Stice, Killen, Hayward & Taylor, 1998). While several authors have generally supported the discontinuity hypothesis (Dykens & Gerrard, 1986; Garfinkel et al., 1995; Ruderman & Besbeas, 1992), recent studies tend to advocate the continuity hypothesis which places undisturbed eating at one end of the continuum (i.e., asymptomatic group), clinical eating disorders at the other end (i.e., eating disordered group) and milder forms of disturbed eating at an intermediate point (i.e., symptomatic group) (e.g., Dingemans et al., 2006; Stice et al., 1998; Tylka & Subich, 1999).

BULIMIC BEHAVIOURS AND DEFENSE It has been suggested that bulimic behaviours (i.e., binging and purging) serve a defensive function by either reducing the individual‘s awareness of negative affects (McManus, Waller & Chadwick, 1996; Meyer & Waller, 1999), a process Luck et al. (2006) have termed secondary avoidance of affect, or alternatively by narrowing the focus of attention to the immediate stimulus environment and sensations (i.e., cognitive narrowing) (Heatherton & Baumeister, 1991). For instance, research suggests that feelings of loneliness precede the onset of bulimia, as well as help to maintain the disorder (Lacey, Coker & Birtchnell, 1986). In support of this contention, Pyle, Mitchell and Eckert (1981) report that 88% of individuals with bulimia recall a significant event involving the separation from, or loss of, a significant relationship prior to the onset of their disorder. The loss of a significant other is likely to result in the development of automatic thoughts associated with that event as well as increased abandonment concerns (Meyer & Waller, 1999). Young, Klosko and Weishaar (2003) propose that these early maladaptive schemas (EMS) are primarily the result of dysfunctional experiences with significant caregivers during childhood. In an attempt to differentiate the core schemas conditional to binge eating disorder, bulimia, and control women, Waller (2002) had female participants complete the Young

Unconscious Schemas, Bulimic Tendencies and the Influence…

225

Schema Questionnaire (Young, 1994). Whilst the two clinical groups had broadly similar levels of negative core schemas, they differed significantly in the nature of the schemas that most clearly characterised the groups. When compared to both the control and binge eating samples, women with bulimia were distinguished by heightened fears of abandonment. Congruent with the above results, a study conducted by Jones et al. (2005) investigating the cognitive factors contributing to eating disorder recovery reports that women with current bulimic eating patterns displayed significantly higher fears of abandonment than individuals with anorexia. Of significance was the finding that fears of abandonment were equally high in both currently eating-disordered and ―recovered‖ women. As such, these fears may not only represent a risk for purging behaviours but also for relapse.

UNCONSCIOUS MALADAPTIVE SCHEMAS Maladaptive schemas can occur without conscious awareness (Meyer & Waller, 1999). This poses a potential problem for traditional CBT as individuals are unlikely to be able to identify maladaptive schemas in order to challenge them (Waller & Barter, 2005). It has been suggested that the most valid and reliable methodology used to examine the impact of preconscious cues is the subliminal visual processing task (Meyer & Waller, 1999). Such tasks involve brief presentations of visual stimuli below the level of awareness, usually for a duration less than 5 milliseconds (ms), using a tachistoscope to control the necessary illumination levels and speed to render the stimuli undetectable. It is believed that perceiving a stimulus without awareness is, in part, effective because it bypasses defences and as such can influence later behaviour (Silverman, 1983). These effects, however, are typically not observed when the test stimulus is presented so that it is easily recognised (Weinberger & Silverman, 1990). In an attempt to assess the impact of preconscious abandonment cues on women‘s eating behaviour, Patton (1992) presented either abandonment (i.e., ―Mama is leaving me‖) or neutral (i.e., ―Mama is loaning it‖) stimuli to 40 ―high eating disordered‖ and 40 ―low eating disordered‖ women. During a subsequent 10 minute bogus taste discrimination task, women with unhealthy eating attitudes ate significantly more after exposure to the subliminal abandonment cue. Although these findings are compatible with other models of bulimic psychopathology, several methodological issues need to be addressed. First, the amount eaten by participants may not reflect the individual‘s avoidance of threats, but rather their difficulty in taste differentiation or decision-making. Second, it is possible that overeating reflected a response to threat processing in general, rather than the specific threat of abandonment. The first two of these methodological limitations were addressed in a later related study. Waller and Barter (2005) examined whether the behavioural impact (eating) of abandonment cues can be countered by the subliminal presentation of contradictory (or counterschematic) information (unification cues). Ninety-six non-clinical women were presented with a subliminal abandonment cue (―lonely‖) that either preceded or followed a neutral (―gallery‖) or unification (―friendship‖) cue. After excluding participants with an Eating Disorder Inventory (EDI) score above 30 participants were then left alone in a room with a bowl of crackers and the opportunity to eat (rather than implementing a taste discrimination task). Relative to the impact of the neutral cues, presenting the subliminal unification stimuli before

226


or after the subliminal abandonment cue significantly reduced the amount eaten (p < .001). Only one other study has implemented a unification cue when assessing eating symptomology. Hallings-Pott, Waller, Watson and Scragg (2005) exposed participants to a subliminal neutral (―gallery‖) as well as a threat (―lonely‖) cue to assess state dissociation and then later used the unification cue ―friendship‖ to successfully reduce their participant‘s affective arousal to baseline levels.

IMPLICATIONS AND LIMITATIONS OF PAST RESEARCH While these results have important implications for the understanding and treatment of bulimia, several limitations to Waller and Barter‘s (2005) study necessitate mention. First, the number of participants in each group was relatively small (n = 24). Second, with evidence suggesting that eating disorder behaviours and traits exist along a continuum it was questionable to remove participants with scores above 30 on the Eating Disorder Inventory (EDI). As noted earlier, related research has inconsistently dichotomised groups into high and low EDI groups using a controversial method known as the extreme groups approach (EGA) (Preacher, MacCallum, Rucker & Nicewander, 2005). As a result, the current study will employ a dimensional as opposed to categorical, approach to eating disorders. Further, Fariburn et al. (2003) suggests that there is a pressing need for more treatment research that extends beyond adolescent and young women to include older individuals. Finally, each of the four groups in Waller and Barter‘s (2005) study included an abandonment related cue. As such, there was no control group to allow for baseline comparisons. Subliminal presentations have been found to have behaviourally positive effects (e.g., Silverman & Weinberger, 1985; Waller & Barter, 2005) and this study may indicate the possibility of using subliminal cues in conjunction with other forms of therapy, such as CBT or Schema-Focused Cognitive-Behavioural Therapy, to have a therapeutic effect for individuals with bulimia. As such, the following question will be investigated: Is there an association between bulimic tendencies, subliminal presentations of abandonment and unification cues, and eating behaviours? It was hypothesised that women presented with a visual abandonment cue will eat significantly more when compared to individuals in the control group, while women presented with a visual unification cue will eat significantly less when compared to individuals in the control group. Furthermore there will be an increase in significance when treating group differences as a categorical rather than continuous variable, and scores on the Eating Disorder Inventory will moderate the relationship between groups and the amount eaten, with women high on bulimic tendencies eating relatively more within each group.

METHOD Design The study employed a between-subjects experimental design, which compared the amount eaten by the participants in the 5 minutes after exposure to one of the three visual


227

subliminal cues. Participants were randomly allocated to one of three conditions only (n = 40 in each group), to avoid cue contagion and expectancy effects. Each group was defined by the subliminal cue presented (neutral, unification or abandonment). Participants completed a selfreport measure concerning eating psychopathology.

Participants The participants were 120 female first year university students who received course credit for their participation. Ages within the sample ranged from 18 to 62, with a mean age of 21 (SD = 5.66). Although the majority of the sample (68.3 %) was aged between 18 and 20, a univariate analysis of variance (ANOVA) showed that the age differences among the groups did not approach significance. In terms of relationship status, 52.5% were single, 41.7% were dating, 3.3% were married, 1.7% were in a defacto relationship and 0.8% identified themselves as a widow. Of the participants, 67.5% affirmed that English was their first language.

Measures and Procedure The Eating Disorder Inventory-3 (EDI-3) (Garner, 2004) is a self report measure assessing the psychological and behavioural constructs shown to be clinically pertinent to individuals with eating disorders. The EDI-3 is an expansion and improvement upon the earlier inventory, the Eating Disorder Inventory-2 (Garner, Olmstead & Polivy, 1983), a well established measure of eating disturbances, such as bulimia and anorexia. As an improvement upon the EDI-2‘s 11 scales, the EDI-3 implements 12 scales (3 eating disorder specific scales and 9 psychological scales). Although revised, there is considerable overlap between the EDI2 and EDI-3‘s 91 items, both conceptually and in content, allowing comparisons to be made across inventories (Cumella, 2006). For the purposes of this study, and in line with previous research (e.g., Meyer & Waller, 1999; Patton, 1992; Waller & Barter, 2005), only the 3 scales directly assessing eating behaviours were used (Drive for Thinness, Bulimia and Body Dissatisfaction). Due to the assessment of eating attitudes using a self-report inventory, the Marlowe-Crowne Social Desirability Scale (M-C Form C) (Crowne & Marlowe, 1960) was employed as a measure to control for biased self-presentation. The M-C Form C contains 13 items, 4 of those items were incorporated into the EDI-3 to disguise their purpose.

Subliminal Stimuli and Task Three cues were used for the stimulus presentation. Following Meyer and Waller (1999), Waller and Barter (2005) and Hallings-Pott et al. (2005) the neutral cue used was ―gallery‖ and the abandonment cue was ―lonely‖. Similarly to Waller and Barter (2005) and HallingsPott et al.‘s (2005) studies the unification cue presented was ―friendship‖. These three words were matched for frequency of use in the English language, using the criteria of Johansson and Hofland (1989).

228


Participants were tested individually and resembling previous methodologies (e.g., Waller & Barter, 2005), a cover story was provided to participants to disguise the purpose of the study until the debriefing at the completion of the study. Participants enrolled in a study titled ―Subliminal Messages and the Effects that Mild States of Hunger have on Visual Sensitivity‖. As such, they were asked to abstain from eating for 2 hours prior to undertaking the task. Participants were first asked to complete a demographics questionnaire. Having completed these brief questions, the participants completed the tachistoscopic presentations. Unlike previous studies (e.g., Patton, 1992; Waller & Barter, 2005), participants were given an eyesight test via the PsyTec tachistoscope (Persona AB, Stockholm, Sweden) to control for any vision difficulties and concerns which would confound a participant‘s ability to read the cue. In each of the conditions a single cue was presented 10 times for 5ms and each cue was separated by 5s. Having completed the stimuli exposures, participants were asked to identify the cue presented using a free-recall test. Signal detection research indicates that second guesses are more accurate than chance (Patton, 1992). As such, each participant was asked to provide a first and second guess. Similarly to previous research (Patton, 1992; Waller & Barter, 2005), no participant reported being able to correctly identify the cue. Previous research indicates that binge eating behaviour often occurs in secrecy and so it was decided that the experimenter would leave the room during the 5 minute eating phase (Loeb, Pike, Walsh & Wilson, 1994). Upon leaving, the researcher said ―In case you are hungry after not eating for so long, there are some biscuits for you. Please feel free to help yourself‖. Fantastics’ Delites Sour Cream and Chives biscuits (calorific value = 451 kcal per 100g) were chosen because they are small and of equal size and weight. The researcher then ate a predetermined biscuit to augment the cover story and reduce ambivalence towards eating before leaving the room. The participants were then given the opportunity to complete the survey. Finally two short questions devised for this study concerning the biscuits provided were 1. Did you like the brand of biscuits left for you to eat? Yes/No 2. Do you often eat biscuits that are similar to the ones that were left? Yes/No. If not, why not e.g., allergies. While previous research (e.g., Meyer & Waller, 1999; Patton, 1992; Waller & Barter, 2005) did not ask these two questions, biscuit preferences were considered to be a potential confound to the amount eaten. Ninety-four percent of participants liked the biscuits presented. Participants were then debriefed. The researcher then weighed the remaining biscuits (roughly 42grams per bowl to begin with).

RESULTS Although the amounts eaten by participants in the different conditions were in the predicted direction, with women receiving the abandonment stimuli eating the most and women receiving the unification stimuli eating the least, an analysis of co-variance (ANCOVA) suggested that these differences were not significant (F(1,110) = 1.192, p = .31). Unlike the variables age (F(1,114) = .253, p = .62), relationship (F(4,109) = 1.123, p = .34) and EDI total (F(1,114) = 1.007, p = .32), the variables like biscuit (whether one likes the biscuits provided) and language (whether English is your first or second language) were entered into the analysis because they displayed significant main effects, F(1,113) = 17.39, p


229

< .0005 and F(2,114) = 3.13, p < .05, respectively. As previously mentioned, evidence suggests that language considerably influenced the amounts eaten by women. As such, follow-up univariate and pairwise comparisons were carried out. There were no significant differences between condition and language for women whose first language was English (n = 81). However, there were significant group differences (F(2,114) = 3.99, p = .02) when comparing women who spoke English as a second language (n = 39) placed in the control (M = 3.69) group to those in the unification (M = 2.59) group (F(2,114) = 3.99, p = 0.25). This suggests that women in the unification group whose first language was not English ate significantly less when compared to women who speak English as their first language (see Figure 1).

Figure 1. Amounts eaten by participants in each group for whom English is either their first or second language.

An analysis was carried out in order to test whether the significant results found in previous studies were influenced by the transformation of eating disorder scores from a continuous to categorical variable. There did not appear to be an increase in significance (F(2,114)=1.73, p = .18). Finally, it was predicted that scores on the Eating Disorder Inventory (EDI) would moderate the relationship between groups and the amount eaten, with women high on bulimic tendencies eating relatively more within each group. A GLM analysis was carried out with language entered as a fixed factor. Contrary to what was predicted, this interaction was not significant, F(2, 111) = .884, p =.42. A second GLM was carried out to

230


test whether priming effects may have influenced the results with participants receiving the subliminal presentations and eating before completing the EDI. This analysis did not indicate any priming effects (F(2,120) = 1.87, p = .83).

DISCUSSION The aim of the current experimental study was to examine the relationship between subliminal abandonment and unification stimuli, early maladaptive schemas and women‘s eating behaviours. Unlike previous research, where the subliminal presentation of the cue ―lonely‖ resulted in increased compensatory eating behaviours when compared to women in the control group (Meyer & Waller, 1999; Patton, 1992; Waller & Barter, 2005; Waller & Mijatovich, 1998), the present study did not find a significant difference between the amounts eaten by women in the control and abandonment groups. The present study did, however, provide partial support for Waller and Barter‘s (2005) contention that women who receive the unification cue ―friendship‖ will eat significantly less than women who received the neutral cue ―gallery‖ with the unexpected finding that English as a first or second language plays an important role in the amounts eaten. Here, women who spoke English as a second language ate significantly less than those women in the control group. Contrary to what was predicted, the aforementioned results held irrespective of whether Eating Disorder Inventory (EDI) scores were formalised as a continuous or categorical variable. Finally, the present study did not find any evidence that the psychological and behavioural constructs shown to be clinically pertinent to women with bulimia moderate the relationship between group and the amounts eaten. As such, it does not appear that an interaction exists between EDI scores, group and the amounts eaten for this sample. A possible explanation for the finding that women in the abandonment group did not eat significantly more than those women in the control group may come from Weiss‘ distinction between the loneliness of social isolation (i.e., an experienced lack of social contacts, intimacy or support—van Baarsen, Snijders, Smit & van Duijn, 2001) and that of emotional isolation (DiTommaso & Spinner, 1997). While social isolation is thought to result from the absence of an engaging social network and results in feelings of boredom and aimlessness, the loneliness of emotional isolation stems from the absence or loss of a close attachment relationship, such as a parent, and produces a sense of utter aloneness and feelings of abandonment (DiTommaso & Spinner, 1997). However, as DiTommaso and Spinner (1997) contend, research relating loneliness to mental health rarely distinguishes between these two different forms of loneliness. This finding is noteworthy given that the subliminal cue presented, ―lonely‖, may not have adequately reflected Young‘s (1999) conceptualisation of fears of abandonment by significant others, but rather a sense of loneliness due to the lack of friendships. Further, one can be lonely but not necessarily abandoned by significant others. As such, prospective research examining the effect that subliminal abandonment cues have on eating behaviours should distinguish the differing forms of loneliness (social and emotional) and their implication for subliminal stimuli. Comparisons of present findings with current conceptualisations of ‗abandonment‘ with cues that might reflect the construct more accurately would be particularly valuable. For instance, the cue ―mother/father abandoned


231

me‖ may interact with fears of abandonment by significant others more specifically than the present cue ―lonely‖. Although this distinction provides avenues for future research and may have contributed to amounts eaten by women in the abandonment group, the present study did not replicate previous research using the identical cue (i.e., ―lonely‖). A possible explanation for this might be that approximately half (46.7%) of the women in the present study were in some form of relationship. While other studies do not report the relationship status of their participants (e.g., Meyer & Waller, 1999; Patton, 1992; Waller & Barter, 2005; Waller & Mijatovich, 1998) it is possible that being in a romantic relationship mitigates feelings of being abandoned by friends or family. Prospective research would do well to explore the effects that being in some form of romantic relationship has on the women who are subliminally presented with abandonment cues.

CULTURAL CONSIDERATIONS It was discovered that women in the unification group whose second language was English ate significantly less when compared to women in the control group. Bulimia has long been considered to have important cultural determinants (King, 1993). However, the Diagnostic and Statistical Manual of Mental Disorders (4th ed., or DSM-IV-TR) (American Psychiatric Association [APA], 2000) criterion for the disorder is based almost exclusively on research with Caucasian North American and European samples (Herzog & Delinksy, 2001). Similarly, friendships (i.e., the voluntary association between individuals) invariably occur within a cultural context (French, Bae, Pidada & Lee, 2006). Triandis et al. (1988) suggest that friendships can be best understood within the framework of individualistic and collectivist cultures. While the aforementioned authors suggest that individuals from more individualistic cultures interact with many groups and easily develop new friendships, individuals from more collectivist cultures are thought to develop friendships that are more intimate and long lasting. For instance, within the United States the label ―friend‖ is used to describe a wide variety of relationships, ranging from acquaintances to individuals who are as close as family members, whereas the same label within Chinese, South Korean (French et al., 2006) and Greek (Malikiosi-Loizos & Anderson, 1999) populations is limited to fewer, more intimate friends. As such, it is possible that the meaning of the unification cue presented, ―friendship‖, may have influenced the amount eaten by women depending upon the culture from which they came. Related subliminal research (Meyer & Waller, 1999; Patton, 1992; Waller & Barter, 2005; Waller & Mijatovich, 1998) has not accounted for the cultural differences within the predominantly Western samples, nor discussed the implications of such disparities. These findings accentuate the need to look at how culture influences the aetiology of eating disorders as well as how subliminal stimuli are conceptualised in disparate cultures. It also highlights the need for prospective research to assess and report upon differential and individualised responses to treatment based on client diversity. While supporting Waller and Barter‘s (2005) finding that unification cues can have ameliorative effects for some women with bulimic tendencies, it is worth prospective researchers who are examining unification cues within the context of Young‘s (1999) schema

232


theory to find counterschematic stimuli that originate from the perceived healthy bonding with parents (e.g., ―mother/father loves me‖). Similar to research concerning the use of ‗abandonment‘ cue ―lonely‖, it would be valuable to compare these findings with present subliminal research. In line with the current conceptualisation of eating disorders as existing along an eating disorder continuum, it was predicted that there would be an increase in significance when group differences were considered categorical (using methods such as EGA) rather than continuous, and that scores on the EDI would moderate the relationship between group and the amount eaten by each woman. While these predictions were not supported, it is hoped that this study will provoke future thought about the use of EGA in eating disorder literature given that this approach often reduces individual differences into simple binary codes. Researchers often consider that a given construct is dichotomous by nature despite evidence suggesting that the instrument used to assess it, such as the EDI, displays essentially continuous scores (Preacher et al., 2005). In addition, there is also the possibility that scores on the EDI were not reflective of the pathological attitudes and behaviours of women with the disorder or that women with bulimic tendencies may have denied or distorted their responses due to a lack of insight or shame, despite the use of social desirability items. While this is the first study assessing the moderating effects of EDI on the relationship between group and amounts eaten within subliminal research, it would be worth assessing these effects within a clinical sample where effects may be more pronounced.

IMPLICATIONS FOR FUTURE RESEARCH The findings of the present study have a number of implications. First, the present study lends some support to Waller and Barter‘s (2005) contention that counterschematic subliminal presentations can have an ameliorative effect on some women with eating disturbances. While the overall sample did not show a significant decrease in the amounts eaten following the unification cue ―friendship‖, women who spoke English as their second language ate significantly less than those women in the control condition. Helping women become aware of their concerns about relationships can allow them to cope more effectively when there are real or perceived threats to friendships, for example, by extending upon existing social networks. To date, there does not appear to be any subliminal research conducted on women with anorexia. Previous research indicates that women with anorexia display comparatively higher scores on self-sacrifice indices when compared to women with binge eating disorder and bulimia (Unoka, Tölgyes & Czobor, 2007). As such, the visual subliminal paradigm might also be extended to develop our current understanding of other eating disorders and the implication this might have for alternate forms of treatment. Future research would also benefit from determining whether abandonment and unification cues have similar effects on other forms of escape behaviour, such as alcohol abuse (Waller & Barter, 2005).


233

CONCLUSION In conclusion, this study extends current understandings of bulimia‘s aetiology, and provides confirmatory evidence for several fundamental aspects of Young‘s (1999) schema theory and Waller et al.‘s (2007) Schema-Focused Cognitive-Behavioural Therapy. The findings, and their implications for the treatment of bulimia, subliminal research, and current theories of negative schemas associated with the disorder, appear to warrant further investigation outside the realms and exclusivity of Cognitive Behavioural Therapy.

ACKNOWLEDGMENT The authors would like to thank Dr Alan Taylor for his invaluable statistics advice and Dr Wayne Warburton for his thoughtful input.

REFERENCES American Psychiatric Association (1994). Diagnostic and statistical manual of mental health disorders (4th ed). Washington DC: Author. Bowers, W. A., & Andersen, A. E. (2007). Cognitive-behavioural therapy with eating disorders: The role of medications in treatment. Journal of Cognitive Psychotherapy: An International Quarterly, 21, 16-27. Cumella, E. J. (2006). Review of the eating disorder inventory-3. Journal of Personality Assessment, 87, 116-117. Dingemans, A. E., Spinhoven, P., & van Furth, E. F. (2006). Maladaptive core beliefs and eating disorder symptoms. Eating Behaviors, 7, 258-265. DiTommaso, E., & Spinner, B. (1997). Social and emotional loneliness: Re-examination of Weiss‘ typology of loneliness. Personality and Individual Differences, 22, 417-427. Dykens, E. M., & Gerard, M. (1986). Psychology profiles of purging bulimics, repeat dieters, and controls. Journal of Consulting and Clinical Psychology, 54, 283,288. Fairburn, C. G., Stice, E., Cooper, Z., Doll, H. A., Norman, P. A., & O‘Connor, M. E. (2003). Understanding persistence in bulimia nervosa: A 5-year naturalistic study. Journal of Consulting and Clinical Psychology, 71, 103-109. French, D. C., Bae, A., Pidada, S., & Lee, O. (2006). Friendships of Indonesian, South Korean, and U.S college students. Personal Relationships, 13, 69-81. Garfinkel, P. E., Lin, E., Goering, P., Spegg, C., Goldbloom, D. S, Kennedy, S., Kaplan, A. S., & Woodside, D. B. (1995). Bulimia nervosa in a Canadian community sample: Prevalence and comparison of sub-groups. American Journal of Psychiatry, 152, 10521058. Garner, D. M. (2004). Eating Disorder Inventory-3 (3rd Ed). USA: Psychological Assessment Resources. Garner, D. M., Olmstead, M. A., & Polivy, J. (1983). Development and validation of a multidimensional eating disorder inventory for anorexia nervosa and bulimia. International Journal of Eating Disorders, 2, 15- 34.

234


Hallings-Pott, C., Waller, G., Watson, D., & Scragg, P. (2005). State dissociation in bulimic eating disorders: An experimental study. International Journal of Eating Disorders, 38, 37-41. Halmi, K. A., Agras, S. W., Mitchell, J., Wilson, T. G., Crow, S., Bryson, S. W., & Kraemer, H. (2002). Relapse predictors of patients with bulimia nervosa who achieved abstinence through cognitive behavioral therapy. Archives of General Psychiatry, 59, 1105-1109. Heatherton, T. F., & Baumeister, R. F. (1991). Binge eating as escape from self-awareness. Psychological Bulletin, 110, 86-108. Herzog, D. B., & Delinsky, S. S. (2001). Classification of Eating Disorders. In L. Smolak (Ed,), Eating disorders: Innovative directions in research and practice (pp. 31-50). Washington, DC: American Psychological Association. Johansson, S., & Hofland, K. (1989). Frequency analysis of English vocabulary and grammar (vol.1). Oxford: Claredon Press. Jones, C., Harris, G., & Leung, N. (2005). Core beliefs and eating disorder recovery. European Eating Disorders Review, 13, 237-244. King, M. B. (1993). Cultural aspects of eating disorders. International Review of Psychiatry, 5, 205-216. Lacey, J. H., Coker, S., & Birtchnell, S. A. (1986). Bulimia: Factors associated with its etiology and maintenance. International Journal of Eating Disorders, 5, 475-487. Leung, N., & Price, E. (2007). Core beliefs in dieters and eating disordered women. Eating Behaviors, 8, 65-72. Loeb, K. L., Pike, K. M., Walsh, T. B., & Wilson, T. G. (1994). Assessment of diagnostic features of bulimia nervosa: Interview versus self-report format. International Journal of Eating Disorders, 16, 75-81. Luck, A., Waller, G., Meyer, C., Ussher, M., & Lacey, H. (2006). The role of schema processes in the eating disorders. Cognitive Therapy and Research, 29, 717-732. Malikiosi-Loizos, M., & Anderson, L. R. (1999). Accessible friendships, inclusive friendships as related to social and emotional loneliness in Greece and the USA. European Psychologist, 4, 165-178. Marlowe, D., & Crowne, D. P. (1960). A new scale of social desirability independent of psychopathology. Journal of Consulting Psychology, 24, 349-354. McManus, F., Waller, G., & Chadwick, P. (1996). Biases in the processing of different forms of threat in bulimic and comparison women. Journal of Nervous and Mental Disease, 184, 547- 554. Meyer, C., & Waller, G. (1999). The impact of emotion upon eating behavior: The role of subliminal visual processing of threat cues. International Journal of Eating Disorders, 25, 319-326. Meyer, C., Serpell, L., Waller, G., Murphy, F., Treasure, J., & Leung, N. (2005). Cognitive avoidance in the strategic processing of ego threats among eating-disordered patients. International Journal of Eating Disorders, 38, 30-36. Patton, C. (1992). Fear of abandonment and binge eating: A subliminal psychodynamic activation investigation. Journal of Nervous and Mental Disease, 180, 484-490. Preacher, K. J., MacCullum, R. C., Rucker, D. D., & Nicewander, W. A. (2005). Use of the extreme groups approach: A critical reexamination and new recommendations. Psychological Methods, 10, 178-192.


235

Pyle, R. L., Mitchell, J. E., & Eckert, E. D. (1981). Bulimia: A report of 34 cases. Journal of Clinical Psychiatry, 42, 60-64. Ruderman, A. J., & Besbeas, M. (1992). Psychological characteristics of dieters and bulimics. Journal of Abnormal Psychology, 101, 383-390. Silverman, L. H. (1983). The subliminal psychodynamic method: Overview and comprehensive listing of studies. In J. Masling (Ed.), Empirical studies of psychoanalytic theory (Vol. 1) (pp. 69-103). Hillsdale, NJ: The Analytic Press. Silverman, L. H., & Weinberger, J. (1985). Mommy and I are one: Implications for Psychotherapy. American Psychologist, 40, 1296-1308. Stice, E., Killen, J. D., Hayward, C., & Taylor, C. B. (1998). Support for the continuity hypothesis of bulimic pathology. Journal of Consulting and Clinical Psychology, 66, 784790. Triandis, H. C., Bontempo, R., Villareal, M. J., Asai, M., & Lucca, N. (1988). Individualism and collectivism: Cross-cultural perspectives on self-ingroup relationships. Journal of Personality and Social Psychology, 54, 323-333. Tylka, T. L., & Subich, L. M. (1999). Exploring the construct validity of the eating disorder continuum. Journal of Counseling Psychology, 46, 268-276. Unoka, Z., Tölgyes, T., & Czobor, P. (2007). Early maladaptive schemas and body mass index in subgroups of eating disorders: A differential association. Comprehensive Psychiatry, 48, 199-204. van Baarsen, B, Snijers, T. A., Smit, J. H., & van Duijn, M. A. (2001). Lonely but not alone: emotional isolation and social isolation as two distinct dimensions of loneliness in older people. Educational and Psychological Measurement, 61, 119-135. Waller, G. (2002). Schema-level cognitions in patients with binge eating disorder: a case control study. International Journal of Eating Disorders, 33, 458-464. Waller, G., & Barter, G. (2005). The impact of subliminal abandonment and unification cues on eating behaviour. International Journal of Eating Disorders, 37, 156-160. Waller, G., & Mijatovich, S. (1998). Preconscious processing of threat cues: Impact on eating among women with unhealthy eating attitudes. International Journal of Eating Disorders, 24, 83-89. Waller, G., Kennerly, H., & Ohanian, V. (2007). Schema-focused cognitive-behavioral therapy for eating disorders. In L. Riso, P. L. du Toit, D. J. Stein, & J. E. Young (Eds.), Cognitive schemas and core beliefs in psychological problems: A scientist-practitioner guide (pp. 139-175). Washington, DC, US: American Psychological Association. Weinberger, J., & Silverman, L. H. (1990). Testability and empirical verification of psychoanalytic dynamic propositions through subliminal psychodynamic activation. Psychoanalytic Psychology, 7, 299-339. Wilson, T. G. (1996). Treatment of bulimia nervosa: When CBT fails. Behaviour Research Therapy, 34, 197-212. Wilson, T. G., Grilo, C. M., & Vitousek., K. M. (2007). Psychological treatment of eating disorders. American Psychologist, 62, 199-216. Young, J.E. (1994). The Young Schema Questionnaire-Short Form. http:// www.schematherapy.com.

236


Young, J. E. (1999). Cognitive therapy for personality disorders: A schema focused approach (3rd ed.). Sarasota, FL: Professional Resource Exchange. Young, J. E., Klosko, J. S., & Weishaar, M . E. (2003). Schema therapy: A practitioner’s guide. New York: Guilford Press.



Chapter 20

MOTIVATIONAL INTERVIEWING CHANGES THE TREATMENT TRAJECTORY OF GROUP COGNITIVE-BEHAVIOURAL THERAPY FOR ANXIETY Weiwen Yang and Esben Strodl School of Psychology & Counselling, Queensland University of Technology Brisbane, Queensland, Australia

ABSTRACT Anxiety disorders have been viewed as manifestations of broad underlying predisposing personality constructs such as neuroticism combined with more specific individual differences of unhelpful information processing styles. Given the high prevalence of anxiety and the significant impairment that it causes, there is an important need to continue to explore successful treatments for this disorder. Research indicates that there is still room for significantly improving attrition rates and treatment adherence. Traditionally Motivational Interviewing (MI) has been used to facilitate health behaviour change. Recently MI has been applied to psychotherapy and has been shown to improve the outcome of CBT. However, these studies have been limited to only considering preand post-treatment measures and neglected to consider when changes occur along the course of therapy. This leaves the unanswered question of what is the impact of pretreatment MI on the treatment trajectory of therapy. This study provides preliminary research into answering this question by tracking changes on a weekly basis along the course of group CBT. Prior to group CBT, 40 individuals with a principal anxiety disorder diagnosis were randomly assigned to receive either 3 individual sessions of MI or placed on a waitlist control group. All participants then received the same dosage of 10 weekly 2 hour sessions of group CBT. Tracking treatment outcome trajectory over the course of CBT, the pre-treatment MI group, compared to the control group, experienced a 

Correspondence concerning this chapter should be addressed to Esben Strodl, School of Psychology and Counselling, Queensland University of Technology, Victoria Park Road, Kelvin Grove 4059, Queensland, Australia; email: [email protected]; phone: +61 7 3138 8416

238

Weiwen Yang and Esben Strodl greater improvement early on in the course of therapy in their symptom distress, interpersonal relationships and quality of life. This early advantage over the control group was then maintained throughout therapy. These results not only demonstrate the value of adding MI to CBT, but also highlight the immediacy of MI effects. Further research is needed to determine the robustness of these effects to inform clinical implications of how to best apply MI to improve treatment adherence to CBT for anxiety disorders.

Keywords: Anxiety Disorders, CBT, Motivational Interviewing, Personality, Treatment trajectory

INTRODUCTION Twenty thousand Australians are found to have a mental illness every year, of which anxiety disorders are most common, affecting one in ten Australian adults (AIHW, 2008). Anxiety disorders have been associated with pervasive economic, social, and physical consequences (Martin-Merino et al., 2010). The reduced quality of life reported in individuals with anxiety disorders has been found to be comparable to and in some instances even worse than other major medical illnesses, such as cancer (Somers et al., 2006). Given the high prevalence of anxiety and the significant impairment that it causes in those who experience it, there is an important need to continue to explore successful treatments for this disorder. The Five-Factor Model of Personality has provided a meaningful framework for formulating and testing hypotheses and has helped advanced the understanding of how adult personality relates to psychopathology (Krueger & Tackett, 2003). A recent meta-analysis on the relationship between the Five-Factor Model and symptoms of clinical disorders found that measures of clinical disorders are typically associated with high Neuroticism, low Conscientiousness, low Agreeableness and low Extraversion (Malouff, Thorsteinsson, & Schutte, 2005). Research into evolutionary personality psychology has proposed that social anxiety disorder may be due to personality vulnerabilities (maladaptive rank styles) to perceptions of people being inferior to others (Michalski & Shackleford, 2010). Anxiety disorders have long been seen as manifestations of broad underlying predisposing personality constructs such as neuroticism and negative emotionality combined with more specific individual differences of unhelpful information processing styles (e.g., anxiety sensitivity, attribution style etc.) (Clark, Watson & Mineka, 1994). Cognitive-Behavioural Therapy (CBT) is widely recognised as the current most effective treatment for anxiety disorders (Barlow, 2008; Haby, Donnelly, Corry, & Vos, 2006; Westen & Morrison, 2001). However in spite of its recognised status as a gold standard intervention, there is also clear evidence of problems with treatment adherence (Issakidis & Andrews, 2004), treatment completion (McKay, Taylor, & Abramowitz, 2009) and efficacy in that a significant portion of participants reach clinical improvement at the end of therapy (Federici, Rowa, & Antony, 2009). There is emerging evidence that motivation to engage in therapy is an important predictor of the outcome of psychotherapy. Moreover there is preliminary evidence that individual differences such as learning goal orientation (Dweck, 1986, 2000) may mediate motivation to engage in psychotherapy (Michalak & Holtforth, 2006; Yang & Strodl, 2010a; Zweig & Webster, 2004). As such interventions designed to modify these

Motivational Interviewing Changes the Treatment Trajectory…

239

individual differences should change motivation to engage in therapy which in turn should change the outcome of CBT.

MOTIVATIONAL INTERVIEWING Motivational Interviewing (MI) has been demonstrated to be an effective short-term treatment for enhancing treatment outcome (Bisono, Manuel, & Forcehimes, 2006; Burke, Arkowitz, & Melanchola, 2003; Hettema, Steele, & Miller, 2005). Motivational Interviewing can be defined as a ―client-centred, directive method for enhancing motivation to change by exploring and resolving ambivalence‖ (Miller & Rollnick, 2002, p.25). It is centred on the client in that it focuses on the current concerns of the client and respects the client‘s choices. Unlike CBT, it does not focus on teaching new coping skills or trying to influence patterns of thoughts or behaviours. MI is directive in the manner of therapist‘s selective responding to client speech in a way that is tuned to resolve ambivalence and move the client towards change. The decision was made to utilise MI as a pre-treatment to CBT given recent emerging research that has suggested that it may be effective as an adjunct intervention with CBT for anxiety (Murphy et al., 2004; Maltby & Tolin, 2005; Westra, 2004, Westra & Dozois, 2006; Westra, Arkowitz, & Dozois, 2009). While these studies have provided promising evidence that MI can enhance the effectiveness of CBT for anxiety, they have all only relied on examining change using pre and post measures. Although this provides useful information that the addition of MI to CBT does result in greater changes in anxiety over the course of therapy, it does not provide more detailed information on when change may occur during the course of therapy. Such information is important because it provides guidance on how to maximise MI‘s effectiveness (i.e., at what stage of CBT MI may be most appropriately added). For example, if the effects of MI are seen at the start of therapy with no ongoing enduring effects, then this would support an argument for an additional MI session mid-therapy. If the effects of MI continue to show additive effects consistently throughout therapy, then such an argument would not be sustained. Another example would be if the effects of MI are delayed and only seen towards the end of therapy. In this scenario, it would be helpful to know how many sessions are needed before the effects of MI are manifested, and it may also suggest that MI should be provided at least a minimum time period before the onset of CBT. As such we believe that measuring the trajectory of change of anxiety during the course of therapy is an important improvement over previous research in this area. That is, it will extend our knowledge beyond just answering the question of whether MI enhances CBT to also answering important questions about how best to apply MI to best enhance CBT for anxiety. So far no study has specifically looked at the trajectory of MI upon CBT for anxiety, however a meta-analysis from 72 outcome studies showed that MI tends to have an early effect upon health behaviour change and diminish across a year of follow-up (Hettema, Steele, & Miller, 2005). As such this study aims to test the hypothesis that MI would indeed have an early effect upon anxiety change during CBT and that this will be sustained through treatment. The full content of the intervention is described in detail elsewhere (Yang & Strodl, 2010b) and is summarised below. This paper reports part of the larger study conducted and narrows the scope examining the impact of MI on treatment trajectory of group CBT.

240

Weiwen Yang and Esben Strodl

METHOD This paper draws its methodology from a pilot study by Westra and Dozois (2006) who conducted a randomised controlled trial with pre-treatment MI for CBT with Anxiety.

Procedure Participants were recruited from media releases and treatment was carried out within the university psychology clinic. Potential participants were first screened over the telephone and then in person to determine eligibility for treatment. The face-to-face screening was conducted using the SCID-IV structured clinical interview. The study population were adults aged 18-65 who met criteria for a principal Axis I Anxiety Disorder. Individuals were excluded from consideration for the study if they had: (a) current psychotic symptoms; (b) concurrent psychological treatment; (c) any major deficits in neurocognitive functioning (e.g., learning disability, illiteracy); (d) active substance abuse; and/or; (e) any evidence of suicidal or homicidal intent. Individuals were not excluded if they met criteria for another DSM-IVTR diagnosis as long as their principal diagnosis was for an anxiety disorder. Individuals who met criteria DSM-IV-TR Axis II disorders were assessed on a case-by-case basis. Participants were randomly assigned to receive either a pre-treatment MI intervention consisting of 3 individual weekly hour-long MI sessions (pre-MI) or no pre-treatment (control) prior to participating in 10 weekly 2 hour-long group CBT sessions. Each CBT group consisted of 6-8 participants and were facilitated by two therapists (1 primary therapist and 1 co-therapist). The primary therapists were registered psychologists with extensive knowledge and experience in CBT, whilst co-therapists were postgraduate clinical psychology students. In administering pre-treatment MI, there was a single research therapist in the study who was a final-year doctoral level clinical psychology student with over 5 years of clinical experience. This therapist was trained in MI1 and was supervised closely by the second author who has over 14 years of clinical experience. To avoid any clinician bias, the research therapist delivering MI was not directly involved in the group CBT administration and CBT therapists were blind to participant group membership (pre-MI or control). All sessions were videotaped and then randomly assessed for treatment integrity and fidelity by an external assessor.

Treatment The MI pre-treatment was based on the principles of MI (Miller & Rollnick, 2002) generalised to be applicable specifically to anxiety. This manual was adapted from a treatment manual for MI of Anxiety by Westra and Dozois (2006)2. The treatment embodied all of the core strategies and principles of MI (expressing empathy, rolling with resistance, 1

Training included (a) multiple viewings of MI training DVDs by Miller, Rollnick & Moyers (1998); (b) videotaped role-plays between the student and 1st year Clinical Psychology postgraduate students; (c) extensive use of MI in student‘s own placements with clients; and (d) continued MI supervision with research supervisor. 2 In personal communication with H. A. Westra, who has written extensively on various applications of MI, she had generously provided us with the use of her treatment manual (Westra & Dozois, 2003).


241

developing discrepancy, and enhancing self-efficacy) and focused on ambivalence about both anxiety change and treatment procedures to manage anxiety (e.g., doing exposure, reducing worry, and reducing avoidance). The MI intervention was conceptualised as having two phases, with each phase consisting of specific objectives and strategies that may overlap depending on the circumstances (Westra & Dozois, 2006). Motivation for change was cultivated in the first phase (importance phase) by examining and resolving ambivalence, enhancing the importance of change, and bolstering self-efficacy. The second phase (confidence phase) was designed to strengthen the client‘s commitment to change by considering change options, setting goals, and making plans. Progression from the first to the second phase of the intervention was determined by the individual‘s readiness to change. The CBT treatment manual was based on well-evaluated treatments for anxiety (Craske & Barlow, 2008) and modified to suit the context of this study (Strodl, 2008). The group CBT protocol included four general components: (a) education and corrective information regarding the aetiology and maintenance of anxiety; (b) mental strategies such as cognitive restructuring and problem solving; (c) physiological strategies such as breathing and relaxation exercises; and (d) behavioural strategies such as interoceptive exposure and behavioural experiments.

Measures A principal DSM-IV-TR Anxiety Disorder diagnosis was assigned to each participant at baseline assessment through the use of the Structured Clinical Interview for DSM-IV Axis I Disorders (SCID-IV; Spitzer, Williams, & Gibbon, 1994). Given the scope of this paper to consider treatment trajectory, we used the Outcome Questionnaire 45 (OQ45; Lambert et al., 1996) which is a brief 45-item, self-report outcome measure that is designed for repeated measurement of client progress through the course of therapy and following termination. It is broadly applicable to adults in behavioural health treatment. Specifically, the OQ45 contains 3 subscales: (a) symptom distress; (b) interpersonal relations; and; (c) social role. Test-retest reliability has been found to be 0.84, with internal consistency of alpha value 0.93 (Lambert et al., 1996). Participants were asked to fill in the OQ45 at the start of every group CBT session; beginning from the second session through to the last session (10th).

RESULTS Table 1 shows the characteristics of the group participants (n=40) who met criteria for an anxiety disorder (ascertained following a telephone and face-to-face clinical interview). As a whole, the study sample was predominantly female (67.5%) and all except one participant (Asian) were Caucasian (97.5%). The mean age of participants was 44 years of age (SD 13.5, range 20 to 68 years). All anxiety disorders in the sample were based on DSM-IV-TR Axis I diagnoses; the most common were GAD (25%) and Panic Disorder with agoraphobia (30%). As indicated, the pre-treatment MI and control groups did not differ significantly on baseline demographic characteristics. As such, none of these variables were included in the main analyses.

242


Figure 2 shows the OQ45 treatment trajectory across all 10 CBT sessions separated by pre-treatment grouping. An independent t-test confirmed that there were no significant differences between the pre-treatment MI and waitlist control group at session 1 (baseline), t (38) = .283, p = .778.

93 interested participants

Enrolment

9 dropped out whilst on waitlist 15 ineligible from telephone screening 12 dropped out from face-to-face screening

55 randomised to pre-tx MI or control group

Pre-tx MI group 28 allocated to pre-tx MI. 24 received intervention.

2 excluded due to non-compliance of baseline measures

Waitlist Control group

Pre-treatment: 3weeks

2 dropped out before intervention (1 disinterest, 1 no response)

27 allocated to control. 24 remained in control. 3 dropped out during waitlist (2 disinterest, 1 transport)

2 dropped out during intervention (1 scheduling, 1 parental preference)

Group CBT: 10weeks 24 allocated to intervention. 21 received intervention.

48 allocated to group CBT

24 allocated to intervention. 19 received intervention.

40 received intervention 3 dropped out (1 disinterest, 1 scheduling, 1 no response)

Figure 1. Participant flow through the study. Figure 1. Participant flow through the study.

5 dropped out (3 disinterest, 2 no response)


243

Table 1. Demographic & DSM-IV-TR Axis I diagnostic characteristics of participants Variable n Demographic characteristics Gender – female Mean Age (SD) Education ≤ Yr 12 TAFE University Marital status Married Single Divorced Had previous tx before Currently on medication Diagnostic characteristics Panic disorder w agora Panic disorder wo agora Social phobia GAD OCD PTSD

pre-tx MI (N = 21) %

n

Control (N = 19) % Pdiff

13 43.6 (14.8)

61.9 -

14 44.74 (12.3)

73.7 -

7 6 8

33.3 28.6 38.1

6 5 8

31.6 26.3 42.1

15 5 1 7 6

71.4 23.8 4.8 33.3 28.6

13 5 1 9 6

68.4

5 4 5 5 1 1

23.8 19 23.8 23.8 4.8 4.8

7 0 4 5 2 1

36.8 0 21 26.3 10.5 5.3

.427 .229 .704

.979

47.4 31.6

.366 .836 .455

Note. The Pdiff column shows significance levels based on t and χ² on group differences.

To determine whether there were any main effects across groups, a repeated measures ANOVA was performed with time (baseline, post-CBT) by group (pre-MI, control) as independent variables and scores on the individual‘s OQ45 as dependent variables. A repeated measures ANOVA reflected a significant main effect of time, F(1, 39) = 8.79, p = .005, suggesting that there was a change in OQ45 scores over the course of treatment. A nonsignificant time by group interaction, F(1, 39) = 1.69, p = .202, however indicated that OQ45identified symptom distress, interpersonal relationships and quality of life does not significantly change over time across groups. Viewing the lined trajectory of the pre-MI group‘s OQ45 scores also indicates a dip in session five. Reviewing the scatterplot of the individual participants‘ data did not show any strong outliers and there were no significant differences found between control and pre-MI group. It was identified that most MI participants experienced a drop in OQ45 after session 5 and that this mean was not pulled down by any outliers. Although no significant time-group interactions were found, line analysis revealed a consistent OQ45 trend between participants who did/ did not receive pre-treatment MI. Participants who received pre-treatment MI appeared to gain greater improvement early on in the course of therapy in their symptom distress, interpersonal relationships and quality of life. This early advantage over the control group was then maintained throughout therapy.

244


85

OQ45 scores

80 75 70 65 60 55 s1

s2

s3

s4

s5

s6 session

s7

s8

s9

s10

MI Control

Figure 2. OQ-45 treatment trajectory for participants in pre-treatment MI & waitlist control group.

DISCUSSION Broadly, these results demonstrate the value of adding MI to CBT for anxiety disorders, confirming the growing research pointed in this direction. The synergistic effect that MI has with CBT can be conceptualised as follows: MI directed at increasing motivation and resolving ambivalence about change blends nicely as a prelude to CBT which is directed at helping the client achieve the desired changes. Whilst MI strives to increase one‘s motivation to change, CBT works by providing the tools for one to engage in that change. Our results also confirm the immediacy of effect of MI. MI showed a strong influence on CBT treatment outcomes within the first half of therapy. This is important to note considering the fluctuating levels of motivation that have often been cited throughout therapy (Miller, 1996; Miller & Rollnick, 2002; Resnicow et al., 2002). Based on a conservative interpretation of these results, considering the immediacy and maintenance of MI effects, it is reasonable to venture that pre-treatment MI works optimally with CBT. However, if we considered the increase in OQ45 scores after session five to be due to the wearing off of MI effects, then a mid-treatment booster MI session on top of pre-treatment sessions may be beneficial. This has been confirmed in other randomised controlled trials that have utilised booster MI sessions at various follow-up periods (Kemp et al., 1998, McCambridge & Strang, 2004).

LIMITATIONS AND FUTURE RESEARCH Results from completer analyses (data including only subjects who completed the treatment) may have been biased by the tendency of participants who were doing either very poorly or very well to drop out of the trial. Follow-up data from drop-outs could have provided a clearer picture as to possible reasons for drop-out. While the sample size was respectable for a pilot study, a larger sample size in future studies will provide more power to allow greater confidence in the generalisability of the results. There also needs to be the


245

consideration that the use of MI with anxiety is a new development. Further research is needed to determine the robustness of MI‘s additive effects on CBT to inform clinical implications of how to best apply MI to improve treatment outcomes to CBT for anxiety. Despite MI‘s widespread efficacy across diverse populations, there are instances in which MI may not be ideal. For clients who indicate readiness to change, MI may be less useful, and some findings indicate that it can be counterproductive (Hettema, Steele, & Miller, 2005; Project MATCH Research Group, 1997). From the perspective of the trans-theoretical model, clients who are in the ―action stage‖ may not benefit from MI (DiClemente & Valesquez, 2002). It has been suggested that attempting to engage clients to explore ambivalence and develop discrepancy may only serve to annoy clients who are ready to implement a change (Moyers & Waldorf, 2006). In this context, it may have further indicated to the client that any more therapy that they receive would not be helpful, which may have led to their premature termination of treatment. In order to be consistent with previous similar studies, this study did not screen participants on their level of motivation prior to allocation to MI or waitlist control—instead relying on a random allocation. However it is possible that the effect size of the MI may be even greater if only participants who were screened to have low motivation to change were included in the study. As such, screening to determine how ready individuals were to change could provide a more appropriate randomisation to the pre-treatment MI allocation which could benefit clients (Heather et al., 1999). Finally, as highlighted by Westra and Dozois (2006) in their pilot study, this present study did not provide a comparison pre-treatment group to MI. As such, the better outcomes of the pre-treatment MI group could merely have been due to some participants receiving additional treatment and not from MI per se. It is not possible to discern whether specific ingredients of the intervention were critical to its effectiveness; further controlled studies comparing other forms of pre-treatments to enhance motivation would be helpful to determine the mechanisms of change of effective pre-treatments.

CLINICAL IMPLICATIONS The present study demonstrates that MI holds promise as a pre-treatment for group CBT. Research presented is preliminary considering the need for further large scale randomised controlled trials examining the efficacy of MI for psychological problems. Nevertheless, in the case of anxiety disorders, the results from this study support growing evidence that indicates that MI enhances treatment outcomes (Westra & Dozois, 2006; Westra, Arkowitz, & Dozois, 2009). The importance of motivation for the success of any psychotherapeutic treatment cannot be understated. Given that motivation in treatment is an important predictor of psychotherapy outcome, enhancing our participant‘s motivation to engage in therapy is important to improve and broaden response rates to treatment. The results from this study indicate that the gains in improving outcome in psychotherapy occur early in therapy and are maintained through the course of therapy. As such, Motivational Interviewing pre-treatments provide one possible answer to enhancing treatment effectiveness. Given the high prevalence and burden of disability of anxiety disorders, and the current limitations in CBT interventions, it would seem prudent to consider pre-treatment motivational interviewing as a useful adjunct to standard CBT interventions for anxiety.

246


CONCLUSION This study aimed to explore the impact of a standard intervention for motivation to engage in therapy (i.e., motivational interviewing) on improving the outcome of CBT. Seeing that this link has been established, the next step is to clarify changes in which possible individual difference factors (arguably learning goal orientation) may mediate the impact of motivational interviewing upon CBT outcome.

REFERENCES Australian Institute of Health and Welfare (2008). Australia’s Health 2008. Canberra: AIHW. Barlow, D. (2008). Clinical handbook of psychological disorders (4th ed.). New York: Guilford. Bisono, A., Manuel, J., Forcehimes, A. (2006). Promoting treatment adherence through motivational interviewing. In W. O‘Donohue & E. Levensky (Eds.), Promoting treatment adherence: a practical handbook for health care providers (pp. 71-84). Thousand Oaks, CA: Sage Publications. Craske, M. G., & Barlow, D. H. (2008). Panic disorder and agoraphobia. In D. H. Barlow (Ed.), Clinical handbook of psychological disorders (4th ed., pp. 1-64). New York: Guilford Press. DiClemente, C. C., & Velasquez, M. M. (2002). Motivational interviewing and the stages of change. In W. Miller & S. Rollnick (Eds.), Motivational interviewing: Preparing people for change (2nd ed., pp. 201-216). New York: Guilford Press. Dweck, C. S. (1986). Motivational processes affecting learning. American Psychologist, 41, 1040-1048. Dweck, C. S. (2000). Self-theories: Their role in motivation, personality, and development. Philadelphia: Psychology Press. Fedirici, A., Rowa, K., Antony, M. (2009). Adjusting treatment for partial or nonresponse to contemporary cognitive-behavioural therapy. In D. McKay, J. Abramowitz, & S. Taylor (Eds.), Cognitive-Behavioural Therapy for refractory cases: turning failure into success (pp. 11-38). Washington: American Psychological Association. Haby, M., Donnelly, M., Corry, J., & Vos, T. (2006). Cognitive behavioural therapy for depression, panic disorder and generalized anxiety disorder: A meta-regression of factors that may predict outcome. Australian and New Zealand Journal of Psychiatry, 40, 9-19. Heather, N., Luce, A., Peck, D., Dunbar, B. & James, I. (1999). The development of a treatment version of the Readiness to Change Questionnaire. Addiction Research, 7, 6368. Hettema, J., Steele, J., & Miller, W. (2005). Motivational Interviewing. Annual Review of Clinical Psychology, 1, 91-111. Issakidis, C., & Andrews, G. (2004). Pre-treatment attrition and dropout in an outpatient clinic for anxiety disorders. Acta Psychiatrica Scandanavica, 109, 426-433. Kemp, P., Kirov, G., Everitt, B., Hayward, P., & David, A. (1998). Randomised controlled trial of compliance therapy. British Journal of Psychiatry, 172, 413-419.


247

Krueger, R., & Tackett, J. (2003) Personality and psychopathology: working toward the bigger picture. Journal of Personality Disorders, 17, 109-128. Lambert, M. J., Hansen, N. B., Umphress, V., Lunnen, K., Okiishi, J., & Heufner, J. C., & Reisinger, C. R. (1996). Administration and scoring manual for the Outcome Questionnaire (OQ-45.2). Wilmington: American Professional Credentialing Services. Malouff, J., Thorsteinsson, E., & Schutte, N. (2005). The relationship between the five-factor model of personality and symptoms of clinical disorders: a meta-analysis. Journal of Psychopathology & Behavioural Assessment, 27, 101-114. Maltby, N., & Tolin, D. T. (2005). A brief motivational intervention for treatment-refusing OCD patients. Cognitive Behaviour Therapy, 34, 176-184. McCambridge, J. & Strang, J. (2004). Deterioration over time in effect of Motivational Interviewing in reducing drug consumption and related risk among young people. Addiction, 100, 470-478. McKay, D., Taylor, S., & Abramowitz, J. (2009). Cognitive-behavioural therapy and refractory cases: what factors lead to limited treatment response? In D. McKay, J. Abramowitz, & S. Taylor (Eds.), Cognitive-Behavioural Therapy for refractory cases: turning failure into success (pp. 3-10). Washington: American Psychological Association. Michalak, J., & Holtforth, M. (2006). Where do we go from here? The goal perspective in psychotherapy. Clinical Psychology Scientific Practice, 13, 346-365. Michalski. R., & Shackelford, T. (2010). Evolutionary personality psychology: Reconciling human nature and individual differences, Personality and Individual Differences, 48, 509-516. Miller, W. (1996). Motivational interviewing: research, practice and puzzles. Addictive Behaviours, 21, 835-842. Miller, W. & Rollnick, S. (2002). Motivational interviewing: preparing people for change. (2nd ed.). New York: Guilford. Moyers, T., & Waldorf, V. (2006). Motivational Interviewing: Destination, direction and means. In F. Rotgers, J. Morgenstern, & S. Walters (Eds.), Treating substance abuse: theory and technique (2nd ed., pp. 298-314). New York: Guilford Press. Murphy, R., Rosen, C., Thompson, K., Murray, M., Rainey, Q. (2004). A readiness to change approach to preventing PTSD treatment failure. In S. Taylor (Ed.), Advances in the treatment of posttraumatic stress disorder (pp. 67-93). New York: Springer. Project Match Research Group. (1997). Matching alcoholism treatments to client heterogeneity: Project MATCH Posttreatment drinking outcomes. Journal of Studies on Alcohol, 58, 7-29. Resnicow, K., DiIorio, C., Soet, J., Borelli, B., Hecht, J., & Ernst, D. (2002). Motivational Interviewing in health promotion: it sounds like something is changing. Health Psychology, 21, 444-451. Spitzer, R. L., Williams, J. B. W., Gibbon, M., & First, M. (1994). Structured Clinical Interview for DSM-IV. New York: New York State Psychiatric Institute. Strodl, E. (2008). Treatment Manual for Group CBT treatment of Other Anxiety Disorders Study. Unpublished manual. Westen, D. & Morrison, K. (2001). A multidimensional meta-analysis of treatments for depression, panic, and generalised anxiety disorder: an empirical examination of the

248


status of empirically supported therapies. Journal of Consulting and Clinical Psychology, 69, 875-899. Westra, H., & Dozois, D. (2006). Preparing clients for cognitive behavioural therapy: Randomised pilot study of motivational interviewing for anxiety. Cognitive Therapy Research, 30, 481-498. Westra, H., Arkowitz, H., Dozois, D. (2009). Adding a motivational pre-treatment to cognitive behavioural therapy for generalised anxiety disorder: A preliminary randomised controlled trial. Journal of Anxiety Disorders, 23, 1106-1117. Yang, W. & Strodl, E. (2010a). What Motivates MI: Learning Goal Orientation mediates changes in motivation levels. Submitted for publication. Yang, W. & Strodl, E. (2010b). Motivational Interviewing enhances group CognitiveBehavioural Therapy for Anxiety. Submitted for publication. Zweig, D., & Webster, J. (2004). What are we measuring? An examination of the relationships between the big-five personality traits, goal orientation, and performance intentions. Personality and Individual Differences, 36, 1693-1708.



Chapter 21

WHAT DO SELF-REPORT MEASURES OF DISINHIBITION REALLY MEASURE? Elliroma Gardiner School of Psychology, University of Queensland, St Lucia, QLD 4072, Australia

Chris J. Jackson Australian School of Business, University of NSW, Sydney, NSW 2052, Australia

ABSTRACT We compare and contrast the use of various self-report instruments to measure Disinhibition. Disinhibition, as conceptualised by Patterson and Newman's (1993) response modulation model, characterises disinhibited individuals as having a tendency towards rapidly approaching situations of conflict and also as being prone to having deficits in passive avoidance learning. According to this model, Disinhibition occurs only in situations of competing approach/avoidance goals and is also thought to be driven by Neuroticism. Different subjective measures such as Zuckerman‘s Sensation Seeking Scale, Dickman‘s Impulsivity Inventory and the Barrett Impulsivity Scale have been used interchangeably to measure Disinhibition. However, little consideration has been directed towards understanding how these scales map on to definitions of Disinhibition. Based on a comprehensive literature review, the current research will highlight the tendency of many popular self-report measures to be heavily focused on reporting the resultant disinhibited behaviour, rather an interest in understanding its underlying mechanisms. The current research will also demonstrate how this bias can influence the reporting and meaning of this construct. This 'construct confusion' has important implications for Disinhibition research.

Keywords: Disinhibition, Impulsivity, Self-Report measures, Sensation Seeking




250

Elliroma Gardiner and Chris J. Jackson

INTRODUCTION Despite the lack of a uniform definition, disinhibition is generally thought to occur as a result of a cognitive-attention deficit that impairs an individual‘s ability to inhibit behaviours and to reflect upon and learn from the consequence of those behaviours (Gorenstein & Newman, 1980; Patterson & Newman, 1993). For example, upon approaching a yellow traffic light a person running very late for an important meeting will need to make a quick decision about whether to run the yellow light or stop. An individual high in Disinhibition is more likely to continue with the approach behaviour; that is, they are more likely to run the yellow light rather than stop. Disinhibited approach has been associated with a number of negative outcomes such as alcoholism (Earleywine & Finn, 1991), substance abuse (Kosson, Steuerwald, Newman, & Widom, 1994; McGue, Slutske, Taylor, & Iacono, 1997; Sher & Trull, 1994; Thompson, Whitmore, Raymond & Crowley, 2006), and problem gambling (Gupta, Derevensky & Ellenbogen, 2006; McCormick, 1993). Given the many different ways in which disinhibited tendencies can be observed it is unsurprising that this construct is largely considered to be ‗multifaceted‘ (e.g., Colder & O‘Connor, 2002; Dindo, McDadeMontez, Sharma, Watson & Clark, 2008; Nigg, 2000). Unfortunately, whilst this is a true assertion, it has become an overused and overly simple solution to the more complex issue of actually defining Disinhibition. Conceptualizing Disinhibition as multifaceted has also led to blanket applications of methodologically different measures by researchers investigating this behaviour. The aim of this chapter is to compare and contrast subjective measures of Disinhibition to demonstrate how the theoretical framework and item content of these tools influence the reporting and conceptualisation of Disinhibition. This aim will be achieved by firstly discussing the methodological implications of incorporating a self-report methodology in study design. Following this, we will highlight and discuss two causes of the construct confusion surrounding Disinhibition. The first issue is concerned with divergent conceptualisations of the same self-report tools across studies and the second issue focuses on the ubiquitous use of terms such as, impulsivity, to also refer to Disinhibition and vice versa. We argue that these issues stem from an over focus of self-report measures on the resulting end behaviour of Disinhibition rather than its underlying mechanisms and the situational aspects which elicit this behaviour.

A CASE FOR SELF-REPORT MEASURES There are a number of good reasons to employ self-report measures in experimental design. For instance, self-report scales are inexpensive, easy to administer and interpret and generally have very good psychometric properties (Kline, 1993; Paulhus & Vazire, 2007; Westen & Rosenthal, 2005). However, this method has been criticised for having little overlap with behavioural and physiological measures (Barratt & Patton, 1983; Brenner, Beauchaine & Sylvers, 2005; Gomez & McLaren, 1997) and for being too reliant on the participant possessing adequate self-awareness (Enticott & Ogloff, 2006; Judd & McClelland, 1998; Kagan, 1988).

Table 1. A selection of studies which have utilised self-report measures to investigate disinhibited behavior Measure Barrett Impulsivity Scale (BIS; Patton et al., 1994)

Authors Keilp et al. (2005) Loxton et al. (2008)

Rosval et al. (2006) Ledgerwood et al. (2009) Hanson et al. (2008)

Flory et al. (2006)

Dickman‘s Impulsivity Scale (DII; Dickman, 1990)

Stahl & Rammsayer (2007) Jackson (2008)

Study Aim Determine the types of tasks and cognitive functions that are associated with impulsiveness ratings. Investigated the relationship between biologically based dimensions of impulsivity in problem gamblers with and without alcohol problems, and non problem gamblers. Investigate role of response inhibition, planning and attention in eating disorders. Determine levels of impulsivity in pathological gamblers with and without substance use disorders versus health controls. Examine reward-related decision-making and impulsivity among methylenedioxymethamphetamine users while controlling for polydrug use via appropriate comparison groups. Examine the structure of dimensional impulsive personality traits in a large normative sample

Investigate Extraversion-related individual differences in response orientation. Test the utility of aural attentiveness x Neuroticism in the prediction of Disinhibition.

Conceptualisation Impulsivity/ Inhibition

N 58

M:F 21:37

Sample Community

Impulsivity/ Disinhibition

110

62:38

Clinical

Response Inhibition/ Impulsivity Impulsivity/ response inhibition

182

0:182

102

51:51

Clinical (121) Community (61) Clinical (61) Community (41)


81

49:32

Clinical (52) Community (29)

Impulsivity/Novelty seeking/Behavioural Disinhibition/ Nonplanning Impulsivity/Inhibition

421

235:186

Clinical

50

0:50

Student

Disinhibited Approach/ Impulsivity

145

39:106

Student

Table 1. Continued Lansbergen et al. (2007)

Disinhibition scale from Zuckerman‘s Sensation Seeking Scale (ZSSS, Form V: Zuckerman et al., 1978)

McAlister et al. (2005) Smillie & Jackson (2006) Vigil-Colet & MoralesVives (2005) Foster & Trimm (2008) Gupta et al. (2006) Reio & SandersReio (2006) Ledgerwood et al. (2009) Hanson et al. (2008)

Linnet et al.

Test whether impulsivity within the normal population is associated with weaker interference control Explored predictors of young people‘ propensity to engage in unfaithful activities while dating. Determine the fit of Dickman‘s (1990) descriptive concept of Functional Impulsivity with Gray‘s (1970) Reinforcement Sensitivity Theory. Investigated the relationship between impulsivity, intelligence, and academic failure in a sample of secondary students. Demonstrate the value of incorporating approach–avoidance motivation into the study of narcissistic behaviour. Determine personality characteristics and risktaking characteristics among adolescent gamblers. Investigate sensation seeking as an inhibitor of job performance.

Impulsivity/ Interference control/ Response inhibition Impulsivity

435

78:357

Student

123

46:77

Student

Impulsivity/ Behavioural Inhibition Impulsivity

299

105:194

Student

241

107:134

Student

Impulsivity/ Uninhibited

193

49:144

Student

Disinhibition/ Impulsivity Disinhibition/ Sensation seeking

817

417:400

Student

233

152:81

Workers

Determine levels of impulsivity in pathological gamblers with and without substance use disorders versus health controls. Examine reward-related decision-making and impulsivity among methylenedioxymethamphetamine users while controlling for polydrug use via appropriate comparison groups. Investigate differences in episodic chasing

Impulsivity/ response inhibition

102

51:51



81

49:32


Impulsivity/Sensation

100

65:35

Clinical (61)

(2006) Leeman et al. (2009) Beck et al. (2009)

Impulsiveness and Venturesome scale from Eysenck Impulsiveness Scale (EIS; Eysenck et al., 1984; Eysenck & Eysenck, 1980

Fallgatter & Herrmann (2001) Kane et al. (2004) Lansbergen et al. (2007) Lijfijt et al. (2004) Marsh et al. (2002) Reynolds et al. (2006)

between pathological gamblers and nonpathological gamblers. Identify prospective predictors of problem drinking in undergraduates. Evaluate the psychometric properties of the BIS/BAS Scales and the Sensitivity to Punishment and Sensitivity to Reward Questionnaire (SPSRQ) in eating disordered patients.

Seeking

Community (49)

Disinhibition/ Impaired Control Behavioural Inhibition/ Impulsivity

337

137:205

Student

103

0:103

Clinical

Determine whether the electrophysiological correlates of cognitive response control are related with I7-impulsivity in healthy participants. Determine the role of impulsivity in women with eating disorders. Test whether impulsivity within the normal population is associated with weaker interference control Investigate whether there is an association between trait impulsivity in the normal population and inhibitory motor control. Compare women with high and low trait impulsivity to test response-disinhibition and reward-choice models of Disinhibition. Compare behavioural with self-report measures of impulsivity.

Impulsivity/Inhibition

22

12:10

Workers

Impulsivity

23

0:23

Clinical

Impulsivity/ Interference control/ Response inhibition Impulsivity

435

78:357

Student

62

17:45

Student

Impulsivity/ Response Disinhibition Impulsivity/Inhibition

86

na

Community

70

na

Student

254


Using the earlier example of the individual running the yellow light, Disinhibition occurs in situations of high emotional arousal and in this instance, it was the anxiety about being late for the meeting which motivated the individual to continue with their approach goal despite the potential negative consequences (i.e., having an accident or getting a traffic fine). Given that Disinhibition is elicited in circumstances of intense emotional conflict, self-report measures have been widely perceived as inadequate in capturing the high emotional valence required to elicit a true disinhibited response (Bechara, 2004, 2005; Enticott & Ogloff, 2006; Thompson et al., 2006). However, it can also be argued that Disinhibition, similar to other personality traits is perceptual in nature and therefore, appropriately measured by self-report (e.g., Howard, 1994; Schmitt, 1994; Spector, 1994). In spite of these issues, self-report remains to be the most popular method of measurement for personality researchers (Kagan, 2007; Vazire, 2006).

DEFINING DISINHIBITION Disinhibition has been a construct of central interest in psychopathology research, although there has also been much work linking together cognitive and neuropsychological perspectives (Nigg, 2000). Table 1 lists the details of a selection of studies that have utilised popular self-report measures of disinhibited types of behaviour. This table also demonstrates that typical participant samples employed in Disinhibition research include those who are pathological gamblers (e.g., Ledgerwood, Alessi, Phoenix, & Petry, 2009; Linnet, Rojskjær, Nygaard, & Maher, 2006), substance users (e.g., Hanson, Luciana, & Sullwold, 2008; Kane, Loxton, Staiger, & Dawe, 2004), or clinical (e.g., Beck, Smiths, Claes, Vandereycken, & Bijttebier, 2009; Flory et al., 2006; Rosval et al., 2006). Due to these taxonomies and varied samples, self-report measures of Disinhibition have been largely focused on linking Disinhibition with a wide range of related constructs such as urgency (Cyders & Smith, 2008), impulsivity (Clark, 1993; Clark & Watson, 1999), lack of restraint (Watson & Clark, 1993), sensation seeking (Zuckerman, 1979; Zuckerman, Kuhlman & Camac, 1988), effortful control (Moriya & Tanno, 2008) and Eysenck and Eysenck‘s (1975) Big Three factors (Markon, Krueger, & Watson, 2005). These tools have focused on the resulting disinhibited behaviour and not on differentiating between Neuroticism and Impulsivity as the mechanisms underscoring Disinhibition. Thus, despite Disinhibition being considered a multifaceted construct there still remains a lack of a uniform agreement as to what Disinhibition is and is not (Harnishfeger, 1995; Nigg, 2000).

DISINHIBITION AND CONSTRUCT CONFUSION Research focused on surface level descriptions of Disinhibition has contributed to construct confusion by failing to adequately contribute to the development of the theoretical basis of this behaviour (e.g., Clark, 1993; Dindo et al., 2008). Definitions of Disinhibition have been broad and inconsistent across authors. For instance, some define Disinhibition as acting impulsively and without control (e.g., Clark & Watson, 1999; Colder & O‘Connor, 2002), acting with no restraint (e.g., Dindo et al., 2008), as a preference for smaller-sooner

What do Self-report Measures of Disinhibition Really Measure?

255

reward versus larger-later rewards (e.g., Dickman, 1993), as a form of high risk, sensation seeking or carelessness (e.g., Dindo et al., 2008; Markon et al., 2005; Schalling, 1978; Zuckerman, 1979) and as a combination of low Agreeableness and low Conscientiousness (e.g., Costa & McCrae, 1995; Watson, Clark & Harkness, 1994). Ambiguous definitions of constructs make it difficult to draw conclusions and to ensure consistency in purpose, conceptualization and measurement (Ruddell, 2002). Informed by Patterson and Newman‘s (1993) response modulation model, we view Disinhibition as occurring only in situations of conflicting approach and avoidance goals. Also in line with Patterson and Newman‘s (1993) model, we define Disinhibition as a lack of inhibition resulting from Neuroticism. The first part of our definition is based on evidence that individuals high in Disinhibition demonstrate poor response modulation when they are presented with conflicting stimuli (Avila, Moltó, Segarra, Torrubia, 1995; Avila & Parcet, 2000; Avila & Torrubia, 2005; Gorenstein, & Kelsey, 1983; Gorenstein & Newman, 1980; Newman, Windom & Nathan, 1985; Newman & Kosson, 1986; Newman, Scerbo et al., 1990). Using the previous example, the individual running late was presented with a choice of either running the yellow light to achieve their goal (getting to their meeting on time) or halting their approach behaviour (and risk being late for their meeting). Under these conflicting conditions individuals high in Disinhibition tend to pay too much attention to reward stimuli at the expense of paying too little attention to punishing cues. Again, using this example, this individual was focused on making the meeting on time rather than the possible financial or safety costs associated with running a yellow light. This is further supported by a lack of significant differences between individuals high in Disinhibition and comparison groups in reward only or punishment only tasks (e.g., Newman et al., 1985; Newman et al., 1990). Similar to other popular conceptualisations of Disinhibition we view this behaviour as being strongly driven by Neuroticism or other components of the Avoidance system (e.g., Patterson & Newman, 1993; Robinson, 2007). Empirical evidence identifying Neuroticism as an important predictor of increased sales performance (e.g., Jackson, 2008; Smillie, Yeo, Furnham & Jackson, 2006), impaired response modulation (i.e., an inability to switch response sets), disinhibited behaviour (e.g., Wallace, Bachorowski, & Newman, 1991), and faster reaction times on a reversal learning task (e.g., Gullo, Jackson & Dawe, 2010) further support our inclusion of Neuroticism in our definition. We also explicitly exclude the traits of ‗riskiness‘, ‗recklessness‘, ‗sensation seeking‘ and ‗impulsivity‘ from our definition. Impulsivity has often formed part of definitions of Disinhibition (e.g., Clark & Watson, 1999); however, given the lack of an agreed upon definition of impulsivity (e.g., Arnett, 1994; Carrillo-de-la-Pena, Otero, & Romero, 1993; Enticott & Ogloff, 2006; Leshem & Glicksohn, 2007), its basis in the approach as opposed to the avoidance system, and the almost non-existent argument that impulsivity has a basis in conflict, the utility of this construct when researching Disinhibition is questionable. Nevertheless, as illustrated in table 1, the terms Impulsivity and Disinhibition have been used interchangeably in a number of different studies. Our definition is necessarily prescriptive and narrow and is similar to other conceptualisations of Disinhibition (e.g., Gullo & Dawe, 2008). Whilst some researchers may be at odds with our conceptualisation (e.g., Sher, Bartholow & Wood, 2000; Watson & Clark, 1999; Zuckerman, 1979) and others may advocate for more vague and descriptive definitions (e.g., Avila, 2001; Gorenstein & Newman, 1980) where Impulsivity forms part of the definition of Disinhibition, there is some evidence which suggests that differentiation of

256


Disinhibition from other related traits is well overdue (e.g., Enticott & Ogloff, 2006; Nigg, 2000).

ISSUES IN MEASURING DISINHIBITION Table 1 lists a number of self-report measures that have been used to measure Disinhibition; however, although many instruments are used to measure this behaviour and its underlying mechanisms, few are actually specifically designed to do so. This problem raises two concerns—one of construct validity and one of extrapolating beyond the data. Table 1 also demonstrates how researchers typically measure other constructs, like Impulsivity (e.g., rash impulsivity, Loxton, Nguyen, Casey & Dawe, 2008; rapid response impulsivity, Thompson et al., 2006) to make inferences about disinhibited behaviour. The Barrett Impulsivity Scale (BIS, Patton, Stanford & Barratt, 1995), Dickman’s Impulsivity Inventory (DII; Dickman, 1990) and Eysenck’s Impulsiveness Scale (EIS; Eysenck et al., 1985) are three examples of Impulsivity measures listed in table 1 which have been used to investigate disinhibited behaviour. The BIS purports to measure impulsive thoughts and behaviours and was originally developed to relate anxiety and impulsivity to psychomotor efficiency (Barratt, 1959; Patton et al., 1995; Stanford, Greve, Boudreaux, Mathias & Brumbelow, 1996). Some example items include, ‗I act on impulse‘, ‗I act on the spur of the moment‘ and ‗I buy things on impulse‘. Analogous items can be found in Impulsiveness scale of the Eysenck Impulsiveness Scale (EIS; Eysenck et al., 1985) such as, ‗Do you usually make your mind up quickly?‘ and ‗Do you often do things at the spur of the moment?‘. Similarly, the EIS and the Dysfunctional Impulsivity scale from the DII (Dickman, 1990) are widely used self-report measures that claim to tap into unconscious risk and failure to assess risk and undirected impulsive behaviour. Representative items from DII include, ‗I often get into trouble because I don‘t think before I act‘ and ‗I would enjoy working at a job that required me to make a lot of split-second decisions‘. Whilst the focus of each scale is slightly different, it is interesting to note that at the core of each measure is Impulsivity and manifestations of rash, maladaptive and nonplanned behaviour. We argue that Impulsivity and Disinhibition are different constructs, disagree with the interchangeable use of this terminology and advocate for Impulsivity and Disinhibition to be considered independently. We acknowledge that both individuals high in Disinhibition and Impulsivity are liable to act without sound judgement; however, we maintain that different situational constraints and mechanisms underlie this behaviour. Explicitly, we claim that individuals high in Impulsivity are likely to display rapidly expressed behaviour in any situation. In contrast, individuals high in Disinhibition are likely to exhibit rapidly expressed behaviour only in situations of opposing approach-avoidance goal conflict. Using our earlier example of the yellow traffic light, whilst an individual high in Disinhibition and an individual high in Impulsivity are both as likely to run the light rather than stop, we argue that the individual high in Impulsivity would have continued to drive regardless of whether they were running late for a meeting or not.


257

DISINHIBITION AND NEUROTICISM Based on theoretical models of Disinhibition (e.g., Patterson & Newman, 1993; Robinson, 2007) and some emerging empirical evidence (e.g., Jackson, 2008; Moeller & Robinson, 2010; Robinson, Wilkowski & Meier, 2008), we argue that Disinhibition is the result of mechanisms related to Neuroticism. Therefore, in situations of competing approach and avoidance stimuli, individuals high in Disinhibition act without restraint to avoid punishment rather than to seek reward. Thus, Impulsivity and Disinhibition are not synonymous. Impulsivity can be regarded as either a direct result of Disinhibition or as a byproduct of a lack of inhibition and the resulting lack of reflection (Enticott & Ogloff, 2006; Newman & Wallace, 1993). Table 1 also demonstrates the popularity of Zuckerman‘s (1979) Sensation Seeking Scale as a measure of Disinhibition. Zuckerman (1994) characterised sensation seeking as ―a trait defined by the seeking of varied, novel, complex and intense sensations and experiences and the willingness to take physical, social, legal and financial risks for the sake of such experience‖ (p.27). Sensation seeking can be viewed as similar to the dimensions related to impulsivity in Eysenck‘s and Cloninger‘s models (Acton, 2003). Zuckerman‘s Sensation Seeking Scale (SSS-V, Zuckerman, Eysenck & Eysenck, 1978) consists of four subscales, Thrill and Adventure Seeking, Disinhibition, Experience Seeking and Boredom Susceptibility. For the purposes of this review we will restrict discussion to the Disinhibition scale. The Disinhibition subscale assesses social sensation seeking through drinking, sex, and parties (Zuckerman, 1994). Some example questions include ‗I like ‗wild‘ uninhibited parties‘, ‗I enjoy the company of real ‗swingers‘‘, and ‗I often like to get high (drinking liquor or smoking marijuana)‘. We view this measure as inappropriate for investigating Disinhibition as it is so heavily focused on risk and sensation seeking. This measure fails to take into account the underlying mechanisms and prerequisite situational aspects of Disinhibition. Disinhibition occurs in situations of competing conflict (e.g., Arnett, Smith & Newman, 1997; Patterson & Newman, 1993). Individuals high in Sensation Seeking actively seek and find themselves in highly stimulating conflict situations more often than individuals low in Sensation Seeking. Therefore, they may engage in more disinhibited behaviour; however, we claim that this is a result of their situational exposure and their underlying high Neuroticsm which leads them to experience conflict rather than their actual level of Disinhibition per se. This position is partially supported by Roberti‘s (2004) conclusion that ―the relation between behavioural disinhibition and drinking habits appears to be the result of a connection with each other and a preference for sensation seeking‖ (p.265). Furthermore, claims that the Disinhibition scale of the SSS-V suffers from criterion contamination make its use problematic (Action, 2003). Close examination of table 1 also demonstrates that despite utilising the same self-report measure, labelling of the focal construct varies across studies with some researchers labelling the behaviour ‗response inhibition‘, others ‗Disinhibition‘ and several ‗Impulsivity‘. It is also worth noting that labelling not only varies across studies, but also within studies. This further contributes to construct confusion and makes it difficult to disentangle Disinhibition from other related behaviours.

258


CONCLUSION The aim of this chapter was to compare and contrast subjective measures of Disinhibition to demonstrate how the framework and item content of these tools influenced the reporting and conceptualisation of Disinhibition. We then discussed the issues surrounding the lack of a definition of Disinhibition, put forward our own prescriptive and narrow definition of Disinhibition and presented a summary of studies investigating disinhibited approach to demonstrate some causes of the construct confusion surrounding this behaviour. The most important issue this chapter has highlighted has been the inability of these instruments to measure the underlying aspects of Disinhibition. These measures are heavily focused on the resulting end behaviour and have not been concerned with understanding what mechanisms are responsible for the elicitation of Disinhibition. Collectively, these self-report instruments do not determine whether the motivation towards disinhibited behaviour is driven by Impulsivity or Neuroticism. Exacerbating this issue is the failure of instruments to integrate situational aspects relevant to Disinhibition into the item content and overall instrument conceptualisation. An item such as ‗When I feel anxious I often make quick decisions‘ would be a more meaningful and conceptually coherent way to measure Disinhibition via self-report. By imbedding items within a conceptual framework the focus of the instrument becomes more closely aligned with Disinhibition and its underlying mechanisms rather than being subsumed by other related behaviours. This requires only a small semantic change in item content yet will have large effects with regards to differentiating Disinhibition from other related constructs. Another possibility is to use two self report measures at the same time. Jackson (2008) proposes that disinhibition can be measured by the interaction between neuroticism (which provides a measure of anxiety and likely predisiposition to conflict) and ear preference which is argued to be a biological marker for hemispheric asymmetries related to lack of restraint. Nevertheless, in spite of these issues, we expect that self-report will remain to be a popular option for personality researchers. We believe that future research should aim to correct the issues mentioned in this paper and highlight the merits of enlisting a multi-method approach to take advantages of the positive aspects of both self-report and behavioural measures of Disinhibition. There are a number of benefits to enlisting self-report and by taking into account the underlying and situational aspects of Disinhibition this methodology will remain to be a viable option for Disinhibition researchers.

REFERENCES Acton, G. S. (2003). Measurement of impulsivity in a hierarchical model of personality traits: Implications for substance use, Substance Use & Misuse, 38, 67-83. Arnett, J. (1994). Sensation seeking: A new conceptualization and a new scale. Personality & Individual Differences, 16, 289-296. Arnett, P. A., Smith, S. S., & Newman, J. P. (1997). Approach and avoidance motivation in psychopathic criminal offenders during passive avoidance. Journal of Personality and Social Psychology, 72, 1413-1428.


259

Avila, C. (2001). Distinguishing BIS-mediated and BAS-mediated Disinhibition mechanisms: A comparison of Disinhibition models of Gray (1981, 1987) and of Patterson and Newman (1993). Journal of Personality and Social Psychology, 80, 311-324. Avila, C., Moltó, J., Segarra, P., & Torrubia, R. (1995). Sensitivity to primary or secondary reinforcers, what is the mechanism underlying passive avoidance deficits in extraverts? Journal of Research in Personality, 29, 373-394. Avila, C., & Parcet, M. A. (2000). The role of Gray‘s impulsivity in anxiety-mediated differences in resistance to extinction. European Journal of Personality, 14, 185-198. Avila, C., & Torrubia, R. (2005). Personality differences in suppression of behaviour as a function of the probability of punishment. Personality and Individual Differences, 41, 259-260. Barratt, E. S. (1959). Anxiety and impulsiveness related to psychomotor efficiency. Perceptual and Motor Skills, 9, 191-198. Barratt, E. S., & Patton, J. H. (1983). Impulsivity: Cognitive, behavioural, and psychophysiological correlates. In M. Zuckerman (Ed.), The Biological Basis of Sensation Seeking, Impulsivity, and Anxiety (pp. 71-117). Hillsdale, NJ: Erlbaum. Bechara, A. (2004). The role of emotion in decision making: Evidence from neurological patients with orbitofrontal damage. Brain and Cognition, 55, 30-40. Bechara, A. (2005). Decision making, impulse control and loss of willpower to resist drugs: A neurocognitive perspective. Nature Neuroscience, 8, 1458-1463. Beck, I., Smiths, D. J. M., Claes, L., Vandereycken, W., & Bijttebier, P. (2009). Psychometric evaluation of the behavioural inhibition/behavioural activation system scales and the sensitivity to punishment and sensitivity to reward questionnaire in a sample of eating disordered patients. Personality and Individual Differences, 47, 407-412. Brenner, S. L., Beauchaine, T. P., & Sylvers, P. D. (2005). A comparison of psychophysiological and self-report measures of BAS and BIS activation. Psychophysiology, 42, 108-115. Carrillo-de-la-Pena, M. T., Otero, J. M., & Romero, E. (1993). Comparison among various methods of assessment of impulsiveness. Perceptual and Motor Skills, 77, 567-575. Clark, L. A. (1993). Manual for the Schedule for Nonadaptive and Adaptive Personality (SNAP). Minneapolis: University of Minnesota Press. Clark, L. A., & Watson, D. (1999). Temperament: A new paradigm for personality. In L. Pervin & O. John (Eds.), Handbook of personality (2nd ed., pp. 399-423). New York: Guilford. Colder, C. R., & O‘Connor, R. (2002). Attention boas and disinhibited behaviour as predictors of alcohol use and enhancement reasons for drinking. Psychology of Addictive Behaviours, 16, 325-332. Costa, P. T., & McCrae, R. R. (1995). Domains and facets: Hierarchical personality assessment using the Revised NEO Personality Inventory. Journal of Personality Assessment, 64, 21-50. Cyders, M. A., & Smith, G. T. (2008). Emotion-based Dispositions to rash action: Positive and negative urgency. Psychological Bulletin, 134, 807-828. Dickman, S. J. (1990). Functional and Dysfunctional Impulsivity: Personality and cognitive correlates. Journal of Personality and Social Psychology, 58, 95-102. Dickman, S. J. (1993). Impulsivity and information processing. The impulsive client: Theory, research and treatment. In W. G. McCown, J. J. Johnson, & M. B. Shure (Eds.). The

260


impulsive client: Theory, research, and treatment. (pp. 151-184). Washington, DC, US: American Psychological Association. Dindo, L., McDade-Montez, E., Sharma, L., Watson, D., & Clark, L. A. (2008). Development and initial validation of the Disinhibition Inventory: A multifaceted measure of disinhibition. Assessment, 16, 274-291. Earlywine, M., & Finn, P. R. (1991). Sensation seeking explains the relation between behavioural disinhibition and alcohol consumption. Addictive Behaviour, 16, 123-8. Enticott, P. G., & Ogloff, J. R. P. (2006). Eludication of impulsivity. Australian Psychologist, 41, 3-14. Eysenck, H. J., & Eysenck, S. B. G. (1975). Manual of the Eysenck Personality Questionnaire. London: Hodder & Stoughton. Eysenck, S. B. G., Pearson, P. R., Easting, G., & Allsopp, J. F. (1985). Age norms for Impulsiveness, Venturesomeness and Empathy in adults. Personality and Individual Differences, 6, 613-619. Fallgatter, A. J., & Herman, M. J. (2001). Electrophysiological assessment of impulsive behaviour in healthy subjects. Neuropsychologia, 39, 328-333. Flory, J. D., Harvey, P. D., Mitropoulou, V., New, A. S., Silverman, J. M., Siever, L. J., & Manuck, S. B. (2006). Dispositional impulsivity in normal and abnormal samples. Journal of Psychiatric Research, 40, 438-447. Foster, J. D., & Trimm, R. F. (2008). On being eager and uninhibited: Narcissism and approach avoidance motivation. Personality and Social Psychology Bulletin, 34, 10041017. Gomez, R., & McLaren, S. (1997). The effects of reward and punishment on response Disinhibition, moods, heart rate and skin conductance level during instrumental learning, Personality and Individual Differences, 23, 305-316. Gorenstein, E. E., & Newman, J. P. (1980). Disinhibitory psychopathology: A new perspective and a model of research. Psychological Review, 87, 301-315. Gullo, M. J., & Dawe, S. (2008). Impulsivity and adolescent substance use: Rashly dismissed as ‗all bad‘? Neuroscience and Biobehavioural Reviews, 32, 1507-1518. Gullo, M.J., Jackson, C. J., & Dawe, S. (2010). Impulsivity and reversal learning in hazardous alcohol use. Personality and Individual Differences,48, 123-127. Gupta, R., Derevensky, J. L., Ellenbogen, S. (2006). Personality Characteristics and RiskTaking Tendencies Among Adolescent Gamblers. Canadian Journal of Behavioural Science, 38, 201-213. Hanson, K. L., Luciana, M., & Sullwold, K. (2008). Reward-related decision-making deficits and elevated impulsivity among MDMA and other drug users, Drug and Alcohol Dependence, 96, 99-110. Harnishfeger, K. K. (1995). The development of cognitive inhibition: Theories, definitions and research evidence. In K. K. Harnishfeger, N. Frank, & C. J. Brainerd (Eds.). Interference and inhibition in cognition. (pp. 175-204). San Diego, CA: Academic Press. Howard, G. S. (1994). Why do people say nasty things about self-report? Journal of Organisational Behavior, 15, 399-404. Jackson, C. J. (2008). When avoidance leads to approach: How ear preference interacts with neuroticism to predict disinhibited approach. Laterality: Asymmetries of Body, Brain and Cognition, 13, 333-373.


261

Judd, C.M., & McClelland, G. (1998). Measurement. In D. Gilbert, S. Fiske, & G. Lindzey (Eds.), The handbook of social psychology (4th ed). Cambridge: Cambridge University Press. Kagan, J. (1989). Temperamental contributions to social behaviour. American Psychologist, 44, 668-674. Kagan, J. (2007). A trio of concerns. Perspectives on Psychological Science, 2, 361-376. Kane, T. A., Loxton, N. J., Staiger, P. K., & Dawe, S. (2004). Does the tendency to act impulsively underlie binge eating and alcohol use problems? An empirical investigation. Personality and Individual Differences, 36, 83-94. Kelip, J. G., Sackeim, H. A., & Mann, J. J. J. (2005). Correlates of trait impulsiveness in performance measures and neuropsychological tests. Psychiatry Research, 135, 191-201. Kline, P. (1993). The Handbook of Psychological Testing. London: Routledge. Kosson, D. S., Steuerwald, B. L., Newman, J. P., & Widom, C. S. (1994). The relation between socialization and antisocial behavior, substance abuse, and family conflict in college students. Journal of Personality Assessment, 63, 473-488. Lansbergen, M. M., van Hell, E., & Kenemans, J. L. (2007). Impulsivity and conflict in the Stroop Task. Journal of Psychophysiology, 2, 33-50. Ledgerwood, D. M., Alessi, S. M., Phoenix, N., & Petry, N. M. (2009). Behavioural assessment of impulsivity in pathological gamblers with and without substance use disorder histories versus healthy controls. Drug and Alcohol Dependence, 105, 89-96. Leeman, R. F., Toll, B. A., Taylor, L. A., Volpicelli, J. R. (2009). Alcohol-induced Disinhibition expectancies and impaired control as prospective predictors of problem drinking in undergraduates. Psychology of Addictive Behaviors, 23, 553-563. Leshem, R., & Glicksohn, J. (2007). The construct of impulsivity revisited. Personality and Individual Differences, 43, 681–691. Lijffijt, M., Bekker, E. M., Quik, E. H., Bakker, J., Kenemans, J. L., & Verbaten, M. N. (2004). Differences between low and high trait impulsivity are not associated with differences in inhibitory motor control. Journal of Attention Disorders, 8, 25-32. Linnet, J., Rojskjær, S., Nygaard, J., & Maher, B. A. (2006). Episodic chasing in pathological gamblers using the Iowa gambling task. Scandinavian Journal of Psychology, 47, 43-49. Loxton, N. J., & Dawe, S. (2007). How do dysfunctional eating and hazardous drinking women perform on behavioural measures of reward and punishment sensitivity? Personality and Individual Differences, 42, 1163-1172. Loxton, N. J., Nguyen, D., Casey, L., & Dawe, S. (2008). Reward drive, rash impulsivity and punishment sensitivity in problem gamblers. Personality and Individual Differences, 45, 167-173. Markon, K. E., Krueger, R. F., & Watson, D. (2005). Delineating the structure of normal and abnormal personality: An integrative hierarchical approach. Journal of Personality and Social Psychology, 88, 139-157. Marsh, D. M., Dougherty, D. M., Mathias, C. W., Moeller, F. G., & Hicks, L. R. (2002). Comparison of women with high and low trait impulsivity using laboratory impulsivity models of response-disinhibition and reward-choice. Personality and Individual Differences, 33, 1291-1310. McAlister, A., Pachana, N., Jackson, C. J. (2005). Predictors of young dating adults‘ inclination to engage in extradyadic sexual activities: A multi-perspective study. British Journal of Psychology, 96, 331-350.

262


McCormick, R. A. (1993). Disinhibition and negative affectivity in substance abusers with and without a gambling problem. Addictive Behaviors, 18, 331-336. McGue, M., Slutske, W., Taylor, J., & Iacono, W. G. (1997). Personality and substance use disorders: Effects of gender and alcoholism subtype. Alcoholism: Clinical and Experimental Research, 21, 513-520. Moeller, S. K., & Robinson, M. D. (2010). Sex differences in implicit punishment sensitivity: Evidence from two cognitive paradigms. Personality and Individual Differences, 48, 283287. Moriya, J., & Tanno, Y. (2008). Relationships between negative emotionality and attentional control in effortful control. Personality and Individual Differences, 44, 1348-1355. Newman, J. P., & Kosson, D. S. (1986). Passive avoidance learning in psychopathic and nonpsychopathic offenders. Journal of Abnormal Psychology, 95, 252-256. Newman, J. P., Patterson, C. M., Howland, E. W., & Nichols, S. L. (1990). Passive avoidance in psychopaths: The effects of reward. Personality and Individual Differences, 11, 1101-1114. Newman, J. P., Widom, C. S., & Nathan, S. (1985). Passive avoidance in syndromes of Disinhibition: Psychopathy and extraversion. Journal of Personality and Social Psychology, 48, 1316-1327. Newman, J. P., & Wallace, J. F. (1993). Diverse pathways to deficient self-regulation: Implications for disinhibitory psychopathology in children. Clinical Psychology Review, 13, 690-720. Nigg, J. T. (2000). On inhibition/disinhibition in developmental Psychopathology: Views from Cognitive and Personality Psychology and a Working Inhibition Taxonomy. Psychological Bulletin, 126, 220-246. Patton, J. H., Stanford, M. S., & Barratt, E. S. (1995). Factor structure of the Barratt Impulsiveness Scale. Journal of Clinical Psychology, 51, 768-774. Patterson, C. M., & Newman, J. P. (1993). Reflectivity and learning from aversive events: Toward a psychological mechanism for the syndromes of Disinhibition. Psychological Review, 100, 716-736. Paulhus, D. L., & Vazire, S. (2007). The self-report method. In R. W. Robins, R. C. Fraley, and R. Krueger (Eds.), Handbook of research methods in personality psychology (pp. 224-239). New York: Guilford Press. Reio, T. G., & Sanders-Reio, J. (2006). Sensation seeking as an inhibitor of job performance. Personality and Individual Differences, 40, 631-642. Reynolds, B., Ortengren, A., Richards, J. B., & de Wit, H. (2006). Dimensions of impulsive behaviour: Personality and behavioural measures. Personality and Individual Differences, 40, 305-315. Robinson, M. D. (2007). Lives lived in milliseconds: Using cognitive methods in personality research. In: R. W. Robins, R. C. Fraley, & R. Krueger (Eds.), Handbook of research methods in psychology (pp.345–359). New York: Guilford Press. Robinson, M. D., Wilkowski, B. M., & Meier, B. P. (2008). Approach, avoidance, and selfregulatory conflict: An individual differences perspective. Journal of Experimental Social Psychology, 44, 65-79. Ruddell, E. (2002). Constructing constructs: Some thoughts on how to write workable definitions for research. Issues and Innovations, 5-8.


263

Roberti, J. W. (2004). A review of behavioural and biological correlated of sensation seeking. Journal of Research in Personality, 38, 256-279. Rosval, L., Steiger, H., Bruce, K., Israel, M., Richardson, J., & Aubut, M. (2006). Impulsivity in women with eating disorders: problem of response inhibition, planning, or attention? The International Journal of Eating Disorders, 39, 590-593. Scerbo, A., Raine, A., O‘Brien, M., Chan, C., Rhee, C., & Smiley, N. (1990). Reward dominance and passive avoidance learning in adolescent psychopaths. Journal of Abnormal Child Psychology, 18, 451-463. Schalling, D. (1978). Psychopathy-related personality variables and the psychophysiology of socialization. In R.D. Hare & D. Schalling (Eds.), Psychopathic behaviour: Approaches to research (pp. 85-106). Chichester: Wiley. Schmitt, N. (1994). Method Bias: The importance of theory and measurement. Journal of Organisational Behaviour, 15, 393-398. Sher, K. J., Bartholow, B. D., & Wood, M. D. (2000). Personality and substance use disorders: A prospective study. Journal of Consulting and Clinical Psychology, 68, 818829. Sher, K. J., & Trull, T. J. (1994). Personality and disinhibition in psychopathology: Alcoholism and antisocial personality disorder. Journal of Abnormal Psychology, 103, 92-102. Smillie, L., & Jackson, C. (2006). Functional impulsivity and reinforcement sensitivity theory. Journal of Personality, 74, 47-83. Smillie, L., Yeo, G., & Furnham, A., & Jackson, C. J. (2006). Benefits of all work and no play: The relationship between neuroticism and performance as a function of resource allocation. Journal of Applied Psychology, 91, 139-155. Spector, P. E. (1994). Using self-report questionnaires in OB research: A comment on the use of a controversial method. Journal of Organisational Behaviour, 15, 385-392. Stahl, J., & Rammsayer, T. (2007). The point of no return in motor response: Extraversionrelated differences. Psychology Science, 49, 58-67. Stanford, M. S., Greve, K. W., Boudreaux, J. K., Mathias, C. W., & Brumbelow, J. L. (1996). Impulsiveness and risk-taking behavior: Comparison of high school and college students using the Barratt Impulsiveness Scale. Personality and Individual Differences, 21, 10731075. Thompson, L. L., Whitmore, E. A., Raymond, K. M., & Crowley, T. J. (2006). Measuring Impulsivity in Adolescents with Serious Substance and Conduct Problems. Assessment, 13, 3-15. Vazire, S. (2006). Informant reports: A cheap, fast and easy method for personality assessment. Journal of Research in Personality, 40, 472-481. Vigil-Colet, A. & Morales-Vives, F. (2005). How impulsivity is related to intelligence and academic achievement. The Spanish Journal of Psychology, 8, 199-204. Wallace, J. F., Bachorowski, J., & Newman, J. P. (1991). Failures of response modulation: Impulsive behavior in anxious and impulsive individuals. Journal of Research in Personality, 25, 23-44. Watson, D., & Clark, L. A. (1993). Behavioral disinhibition versus constraint: A dispositional perspective. In D. M. Wegner & J. W. Pennebaker (Eds.), Handbook of mental control (pp. 506-527). New York: Prentice-Hall.

264


Watson, D., Clark, L. A., & Harkness, A. R. (1994). Structures of personality and their relevance to psychopathology. Journal of Abnormal Psychology, 103, 18-31. Westen, D., & Rosenthal, R. (2005). Improving construct validity: Cronbach, Meehl, and Neurath‘s ship: Comment. Psychological Assessment, 17, 409-412. Zuckerman, M. (1979). Sensation seeking: Beyond the optimal level of arousal. Hillsdale: Lawrence Erlbaum. Zuckerman, M. (1994). Behavioral expressions and biosocial bases of sensation seeking. NY: Cambridge University Press. Zuckerman, M., Eysenck, S. B., & Eysenck, H. J. (1978). Sensation seeking in England and America: Cross-cultural, age and sex comparisons. Journal of Consulting and Clinical Psychology, 46, 139-149. Zuckerman, M., Kuhlman, D. M., & Camac, C. (1988). What lies beyond E and N? Factor analyses of scales believed to measure basic dimensions of personality. Journal of Personality and Social Psychology, 54, 96-107.



Chapter 22

CAN EMPLOYERS SCREEN JOB APPLICANTS FOR POTENTIAL WHITE COLLAR FRAUD OFFENDERS? Jennifer Keen and Julie Fitness Department of Psychology, Macquarie University, Sydney, NSW 2109, Australia

ABSTRACT High profile business failures resulting from white collar fraud have prompted calls for employers to identify potential offenders among internal and external job applicants. However, the success of any intervention is dependent on the availability of a reliable and valid tool for this purpose. The question of whether employers can or should identify job applicants who are likely to engage in white collar fraud is thus addressed in several stages. First, the context within which these calls are being is made is described. Next, the validity of different types of screening tools currently being used by employers is reviewed. Evidence about the stability and pervasiveness of personality traits assumed to typify people who are likely to engage in white collar fraud is also examined. The chapter concludes with a brief description of research being undertaken to develop a conceptual framework that can be used to address the limitations of existing research in this field, improve existing fraud reduction strategies and reduce the significant the costs associated with these behaviours.

Keywords: Employment screening, Fraud, Integrity tests, Narcissism, Psychopathy, Validity

INTRODUCTION White collar fraud comprises diverse behaviours with diverse and frequently devastating consequences. Corrupt employees deprive their employers of the benefits they were expected to negotiate on their behalf. Insider trading reduces confidence in the stock market, 


266

Jennifer Keen and Julie Fitness

threatening the value of pension funds, investment portfolios and the capacity of organisations to raise funds from investors to expand their business activities. Finance providers who lend their customers money on the basis of falsified financial statements stand to lose both the loans advanced and the profits they would have made on the transactions when their customers are unable to repay the money advanced to them. Cash and inventory stolen from employers and their customers in the U.S have been estimated as amounting to 7% of their annual revenues (i.e., US$994 billion—Association of Certified Fraud Examiners, 2008). A survey of organisations in Australia and New Zealand (KPMG, 2009) revealed that little, if any, of amounts stolen are recovered. These losses may be further compounded by the costs of establishing how much has been stolen, reconstructing the accounting records to correctly state the organisation‘s financial position and/or pursuing the offender to recover the theft proceeds. Estimates of between five and thirty percent of business failures being attributed to fraud (Murphy, 1993) highlight the diversity of stakeholders adversely impacted by fraud, with losses including colleagues‘ employment, investors‘ income and capital, and debts owed to suppliers and taxation authorities. Victims of white collar fraud have reported symptoms of depression, anxiety and suicidal ideation comparable to those experienced by victims of violent crime (Ganzini, McFarland, & Bloom, 1990; Spalek, 2007). These costs have prompted calls from several sources for employers to reduce the risk of white collar fraud by screening job applicants for potential offenders. Corporate governance legislation introduced in America in the wake of high profile corporate collapses resulting from white collar fraud requires many organisations to establish and maintain procedures to reduce the likelihood of fraud (―Sarbanes Oxley Act,‖, 2002). While such measures were a commercial imperative before the enactment of this legislation, the scope of processes to which these measures apply has been expanded from implementing accounting policies and procedures to reduce these risks to include the people who execute those policies and procedures (Waller, 2006). PricewaterhouseCoopers (2003) developed a framework for implementing this legislation that includes a recommendation that background checks be conducted for job applicants. Employers themselves appear keen to embrace screening, with 87% of organisations surveyed in Australia and New Zealand either currently conducting screening checks on job applicants or intending to do so (KPMG, 2009). Perceptions that the risks of fraud may be increasing rather than decreasing in the face of corporate governance regulations (Smith, 2007; Standards Australia, 2008; Zahra, Priem, & Rasheed, 2005) give further weight to calls to screen for potential offenders among job applicants. Increasing complexity in the business environment, the speed with which transactions are processed and a tendency to allocate resources to income generation at the expense of business administration, have been cited as factors increasing those risks (Holmes, Langford, Welch, & Welch, 2002). Given the imperatives for reducing the likelihood of white collar fraud and employers‘ willingness to embrace screening, now is a good time to evaluate whether it is currently possible to accurately screen job applicants for potential white collar fraud offenders.

Can Employers Screen Job Applicants…

267

TOOLS CURRENTLY BEING USED TO SCREEN JOB APPLICANTS FOR POTENTIAL WHITE COLLAR FRAUD OFFENDERS Background Checks PricewaterhouseCoopers (2003) recommend that background checks include an examination of educational background, employment history and criminal records. While confirming education and employment history provided by an applicant in their resume or during an interview may reveal how truthful they have been about these disclosures, their usefulness in exposing potential offenders may be limited. Exaggerating previous work experience to improve one‘s prospects of employment is not necessarily predictive of fraud. Glowing references from previous employers may reflect applicants‘ ingenuity in making career moves before their behaviours are uncovered. Suspicions of fraud held by a previous employer may not be disclosed to a reference checker because the amounts involved were minor, for fear of being sued for defamation, or because they do not want to reveal that their organisation has been the victim of such behaviour. The existence or absence of a criminal record also provides no guarantee that applicants will or will not commit white collar fraud in future. Most fraud offenders are first time offenders (Australian Institute of Criminology and PricewaterhouseCoopers, 2003; Kapardis & Krambia-Kapardis, 2004) and thus are unlikely to have a record. Applicants may have committed fraud but their crimes may not have been detected. Offences may have been detected but criminal charges not brought or pursued; or charges may have been brought but a conviction not secured. Offenders may conceal their criminal record from an employer by changing their name by deed poll. Applicants with criminal records may not reoffend and if they do, they may not necessarily be capable of, inclined to, or occupy a position in which committing white collar fraud is possible.

Polygraph Tests Although polygraphs have generally been outlawed in most private organisations in the United States, they may be used for recruitment purposes in federal, state and local government bodies in that country (United States Department of Labor). Polygraphs have also been proposed as a screening tool in South Africa (Polygraph Institute of South Africa) by an organisation reported as conducting twenty to thirty thousand polygraph tests a year on behalf of employers (Sibali, 2008). However, criticisms about the degree of subjectivity involved in interpreting polygraph test results and the interpretation of those results without reference to normative standards (Goldberg, Grenier, Guion, Sechrest, & Wing, 1991) undermine their reliability and validity for any purpose. Murphy‘s (2000) observation that polygraphs are intended to detect deception at a particular point in time rather than predicting honesty in general further undermines the case for their use in employment screening.

268


Integrity Tests Integrity or honesty tests are based on the premise that individual differences in honesty or integrity are stable and predictable (Murphy, 2000). They comprise overt and personality based measures (Murphy, 2000; Sackett, Burris, & Callahan, 1989). Overt tests measure admissions of and attitudes towards theft and other counterproductive workplace behaviours (e.g., absenteeism, tardiness, and substance abuse). Past transgressions are assumed to predict future behaviour. Personality based integrity tests measure personality constructs thought to predict counterproductive behaviours. Earlier research suggested that conscientiousness was such a predictive construct (Murphy, 2000), although agreeableness and neuroticism have also been identified as additional relevant constructs (Ones & Viswesvaren, 1998). Some authors (Camara & Schneider, 1995; Lilienfeld, Alliger, & Mitchell, 1995) have questioned whether conformity and conventionality are being measured rather than conscientiousness. Opinions surrounding the validity of integrity tests are divided (Goldberg, et al., 1991; Ones, Viswesvaren, & Schmidt, 1995; U.S. Congress: Office of Technology Assessment, 1990). These appear to be underpinned by differences in methodologies and in the level of detail available to researchers. Ones and Viswesvaren (1998) claim integrity tests are useful for predicting counterproductive workplace behaviours on the basis of the correlation coefficients arrived at in their meta-analysis of 27 integrity tests. These correlations were derived from both peer reviewed studies and data provided by test providers. The American Psychological Association (Goldberg, et al., 1991) and the U.S. Congress (U.S. Congress: Office of Technology Assessment, 1990) both commissioned teams to evaluate evidence about the validity of integrity tests used in pre-employment screening. Several data sources were used including literature reviews, publishers‘ test manuals, test booklets, unpublished research reports, marketing materials, and interviews with subject matter experts or interested parties. The Congress report concluded that ―the existing research is insufficient as a basis for supporting the assertion that these tests can reliably predict dishonest behaviour in the workplace‖ (U.S. Congress: Office of Technology Assessment, 1990, p.10). The APA report was not quite as restrictive in its findings, observing that a few integrity tests are supported by adequate documentation about their validity (Goldberg, et al., 1991). Test publishers‘ materials were criticised for several reasons. These included a lack of independent research to corroborate publishers‘ claims, restrictions in what could be reported about the material because it was provided on condition that it not be identified in the task force reports, and observations that for some tests, the only validation data available were claims by the publishers that their tests were valid, with some of these claims being described as bordering on fraudulent (Camara & Schneider, 1995; Goldberg, et al., 1991; U.S. Congress: Office of Technology Assessment, 1990). Goldberg et al.‘s (1991) observation that insufficient information about cutting scores for integrity tests is available provides a means for reconciling the disparate views on integrity testing. While Ones and Viswesvaren‘s (1998) results indicate that higher scores on integrity tests are associated with lower ratings of counterproductive behaviours, this does not provide evidence that the cut-off scores on specific tests are appropriate.


269

ARE INTEGRITY TESTS VALID? Although it has been suggested that research undertaken subsequent to the APA and U.S. Congress reports has strengthened the case for the validity of integrity tests (Murphy, 2000; Sackett & Wanek, 1996), their validity does not necessarily extend to reliably predicting white collar fraud for several reasons. As Lilienfeld et al. (1995) suggested, people who score poorly on overt integrity tests are likely to be those who are honest about their previous dishonesty. While Murphy (2000) argues that people who respond to tests in this way may do so because such dishonesty was normal in their work environment, people who commit white collar fraud often conceal their transgressions to avoid discovery. This suggests that they recognise that their behaviours may be subject to challenge. It also suggests that they may have sufficient insight to identify the purpose of these tests and to respond in a way that presents them as attractive to a potential employer. The inclusion of absenteeism and drug abuse in integrity tests and the validation of test results against supervisor ratings of job performance has raised questions about whether honesty is being measured by these tests (Goldberg, et al., 1991; U.S. Congress: Office of Technology Assessment, 1990). In the absence of evidence linking absenteeism, drug abuse and poor job performance to white collar fraud, the validity of these tests in predicting fraud remains unestablished. Ones and Viswesvaren (1998) suggest that the validity of integrity tests for predicting major embezzlement has not been established because of the low base rates of these behaviours while at the same time providing validity statistics for theft. This highlights assumptions about possible qualitative and quantitative differences between offence types and offenders that need to be explored further. Despite this, the validity statistic for theft was low (.13), casting doubt on the utility of integrity tests regardless of how fraud is conceptualised. The description of a psychometric tool as an integrity test provides obvious appeal to employers eager to embrace diverse strategies for reducing their exposure to fraud. However, using a tool that has not been established as a reliable and valid predictor of white collar fraud may provide a false sense of security about the protection afforded by such a measure. This risk highlights the need for decisions about the use of any psychometric tool to be made by people with: (i) detailed knowledge about the requirements of the role and the environment in which the incumbent is to function; (ii) sufficient skill and objectivity to be able to identify tools that may reliably and validly predict the behaviours of interest, and; (iii) knowledge of employment and other relevant laws in the prevailing jurisdiction to ensure that the choice and use of tool is made within these parameters. Although background checks, polygraphs and integrity tests have been described above as being limited in their usefulness in predicting white collar fraud, it is worth exploring the possibility that increasing the accuracy with which perpetrators are identified may be possible if screening tools include more appropriate constructs.

270


PSYCHOLOGICAL FACTORS POTENTIALLY ASSOCIATED WITH WHITE COLLAR FRAUD Psychopathy and Narcissism White collar fraud offenders have been described as being psychopathic (Cleckley, 1955; Haapasalo, 1994; Hare 1999) or narcissistic (Blickle, Schlegel, Fassbender, & Klein, 2006). Such people are supposedly characterised by superficial charm, manipulation of people for their own ends and a lack of empathy for victims. However, some degree of caution needs to be exercised in the application of these labels to fraud offenders because of the validity of the constructs and the varying characteristics of offenders. Not everyone who is charming and manipulative necessarily commits white collar fraud, and not everyone who commits fraud is necessarily charming and manipulative. Empathy may be irrelevant to white collar fraud if, as Farrell and Healy (2000) observed, perpetrators consider their crimes victimless. As fraud is a crime of deception and success in perpetrating it is dependent on victims‘ ignorance of their status, victims will not be displaying suffering that perpetrators could fail to empathise with at the time the offences are being perpetrated. A pervasive pattern of anti-social behaviour and offending is a key feature of psychopathy (Kapardis & Krambia-Kapardis, 2004; Williams & Paulhus, 2004). However, as most white collar fraud offenders do not have criminal records (Australian Institute of Criminology & PricewaterhouseCoopers, 2003), they do not appear to demonstrate the pattern of offending that typifies psychopaths. Williams and Paulhus (2004) argue that white collar criminals are not psychopaths because their offences fall outside the scope of those that define psychopaths i.e., drug abuse, violent assault, and bullying. Research by Hare and Jutai (1983) suggests that fraud is as likely to be committed by non-psychopaths as psychopaths. The superficial charm and manipulation of others that characterises narcissistic personality disorder (NPD) is intended to elicit the admiration that such individuals believe they are entitled to because of their grandiose sense of self-importance (American Psychiatric Association, 2000). While some individuals with NPD may begrudge others their possessions, this is not a central feature of the disorder. Such thoughts are an example of envy of others that results from a fragile self-esteem, not a sense of entitlement to other people‘s money. While some white collar fraud offenders may well display superficial charm and manipulate others, this could reflect diversionary tactics engaged in to prevent discovery of their intentions or behaviours. Such perceptions could also be post hoc evaluations that reflect a victim‘s reaction to having been deceived by a fraud perpetrator. Disagreement about what comprises psychopathy, overlaps between psychopathy and narcissism, and the potential for confusion about what a sense of entitlement attaches to highlights the need to adequately formulate and differentiate these constructs before attempting to categorise anyone using these constructs. Perhaps research exploring the ‗Dark Triad‘ of psychopathy, narcissism and Macchiavellianism (Paulhus & Williams, 2002) may contribute to such a solution. Until then, Farrell and Healy‘s (2000) description of perpetrators as trusted, innocuous employees who do not draw attention to themselves suggests the need to expand the criteria explored for predicting potential white collar fraud offenders.


271

STATES OR TRAITS? Screening job applicants for potential white collar fraud offenders carries an assumption that all the variability in who commits these offences lies with the individual. This approach contrasts with research and practices that focus on organisational factors associated with fraud. Mumford et al. (2001) suggest that situational variables are better predictors of counterproductive behaviours than individual variables. Within the accounting literature, the focus of fraud prevention has typically focused on establishing policies and processes that remove opportunities for fraud and that increase the likelihood of discovery if fraud does occur (i.e., internal controls). Within the organisational justice literature, Greenberg (2007) proposes a need to redress perceptions of injustices committed by an organisation against its employees as a motivating factor for workplace theft. As not all people with the opportunity to steal do so, and some people go to considerable effort to commit fraud, the occurrence of these behaviours is likely to reflect an interaction of both characteristics of individuals and the circumstances in which those individuals find themselves. Fine et al. (2010) explored the relationship of both situational and individual factors to scores on integrity tests and counterproductive workplace behaviours (CWBs). They found that the relationship between internal controls and employee engagement, and CWBs was moderated when integrity was low but not when integrity was high. Although these findings support the validity of integrity tests for predicting CWBs—and such validity has been questioned above—the value of Fine et al.‘s (2010) work lies in their recognition of the need to identify and explore the full range of organisational and individual factors that may impact on fraud and the relationships between those factors.

CONCLUSION Some current screening tools may identify some potential offenders some of the time. As with any psychometric tool, the risk of false negatives and false positives exist. The costs of engaging a person who defrauds their employer have been described. Evidence that the majority of candidates failing integrity tests are not detected as stealing from their employer (U.S. Congress: Office of Technology Assessment, 1990) suggests the false positive rate as being high. In a skills-short labour market, false positives may also impose costs on employers. Employee recruitment costs may be unnecessarily inflated as the candidate pool is expanded or searched more thoroughly than is necessary to find a suitable applicant. If a suitable applicant cannot be found because the candidate pool has been unnecessarily restricted, employers may lack the skills sets essential for their businesses. The limitations of these tools and the costly consequences of false negatives and false positives posed by these tools support the need for further research to explore whether these false hit rates can be improved. Duffield and Grabosky‘s (2001) observation that no single psychological feature characterises fraud offenders, and literature highlighting gambling as a growing factor in fraud (Standards Australia, 2008) raises several issues for consideration. Perhaps some latent characteristic that typifies these offenders is yet to be identified. Perhaps organisational factors are a better predictor of who perpetrates fraud. Perhaps white collar fraud is best

272


predicted by a taxonomy of offenders that incorporates both characteristics of perpetrators and the contexts in which their behaviours occur. The development of such a taxonomy is currently being tackled by the authors via two studies. The first comprises a content analysis of judgements handed down in white collar fraud court cases. A diverse range of offender characteristics, the behaviours they engaged in and the circumstances in which the offences were committed are described in those judgements. Preliminary analyses appear to support the development of a taxonomy of offenders who commit different types of offences under different circumstances rather than a stereotypical offender who engages in a unitary offence. Results from this study are being used to develop an attributions framework to explore fraud offences from the perspective of the perpetrators. It is hoped that integrating observer perspectives on white collar fraud with those of perpetrators and existing literature will contribute to reducing the likelihood of fraud occurring by reducing the opportunities for fraud within organisations and increasing the likelihood of predicting people who are likely to take advantage of opportunities if they do arise.

REFERENCES American Psychiatric Association (2000). Diagnostic and statistical manual of mental disorders: DSM-IV-TR. Washington, DC: American Psychiatric Association. Association of Certified Fraud Examiners (2008). Report to the nation on occupational fraud and abuse. Australian Institute of Criminology and PricewaterhouseCoopers (2003). Serious fraud in Australia and New Zealand. Blickle, G., Schlegel, A., Fassbender, P., & Klein, U. (2006). Some personality correlates of business white-collar crime. Applied Psychology: An International Review - Psychologie Appliquee: Revue Internationale, 55, 220-233. Camara, W. J., & Schneider, D. L. (1995). Questions of Construct Breadth and Openness of Research in Integrity Testing. American Psychologist, 50, 459-460. Cleckley, H. (1955). Inadequately motivated antisocial behavior. St Louis, MO: C V Mosby Co. Duffield, G., & Grabosky, P. (2001). The psychology of fraud: Trends and issues in crime and criminal justice (No. 199). Canberra: Australian Institute of Criminology. Farrell, B. R., & Healy, P. (2000). White Collar crime: A Profile of the Perpetrator and an Evaluation of the Responsibilities for its Prevention and Detection. Journal of Forensic Accounting, 1, 17-34. Fine, S., Horowitz, I., Weigler, H., & Basis, L. (2010). Is good character good enough? The effects of situational variables on the relationship between integrity and counterproductive work behaviors. Human Resources Management Review, 20, 73-84. Ganzini, L., McFarland, B., & Bloom, J. (1990). Victims of Fraud: Comparing Victims of White Collar and Violent Crime. Bulletin of the American Academy of Psychiatry and the Law, 18, 55-63. Goldberg, L. R., Grenier, J. R., Guion, R. M., Sechrest, L. B., & Wing, H. (1991). Questionnaires used in the prediction of trustworthiness in pre-employment selection


273

decisions: An A.P.A. Task Force Report. Washington, D.C.: American Psychological Association. Greenberg, J. (2007). The top ten reasons why everyone should know about, and study, organisational justice. In A. I. Glendon, B. M. Thompson & B. Myors (Eds.), Advances in organisational psychology (pp. 323-343). Bowen Hills: Australian Academic Press. Haapasalo, J. (1994). Types of Offense among the Cleckley Psychopaths. International Journal of Offender Therapy and Comparative Criminology, 38, 59-67. Hare, R. D. (1999). Without conscience: The disturbing world of the psychopaths among us. New York: Guilford Press. Hare, R. D., & Jutai, J. W. (1983). Criminal history of the male psychopath: some preliminary data. In K. T. Van Dusen & S. A. Mednick (Eds.), Prospective studies of crime and delinquency (pp. 225-236). Boston: Kluwer-Nijhoff. Holmes, S. A., Langford, M., Welch, O. J., & Welch, S. T. (2002). Associations between internal controls and organizational citizenship behavior. Journal of Managerial Issues, 14, 85-99. Kapardis, A., & Krambia-Kapardis, M. (2004). Enhancing fraud prevention and detection by profiling fraud offenders. Criminal Behaviour and Mental Health Special Issue: Criminological and Legal Psychology in Europe, 14, 189-201. KPMG (2009). KPMG Forensic Fraud Survey 2008. Sydney: KPMG. Lilienfeld, S. O., Alliger, G., & Mitchell, K. (1995). Why integrity testing remains controversial. American Psychologist, 50, 457-458. Mumford, M. D., Connelly, M. S., Helton, W. B., Strange, J. M., & Osburn, H. K. (2001). On the construct validity of integrity tests: Individual and situational factors as predictors of test performance. International Journal of Selection and Assessment 9, 240-257. Murphy, K. R. (1993). Honesty in the workplace. Pacific Grove, California: Brooks/Cole Publishing. Murphy, K. R. (2000). What constructs underlie measures of honesty or integrity? In R. D. Goffin & E. Helmes (Eds.), Problems and solutions in human assessment: Honoring Douglas N. Jackson at seventy (pp. 265-283). New York: Kluwer Academic/Plenum Publishers. Ones, D. S., & Viswesvaran, C. (1998). Integrity testing in organizations. In R. W. Griffin, A. O'Leary-Kelly & J. M. Collins (Eds.), Dysfunctional behavior in organizations: Violent and deviant behavior (pp. 243-276): Connecticut: Elsevier Science/JAI Press. Ones, D. S., Viswesvaran, C., & Schmidt, F. L. (1995). Integrity Tests: Overlooked facts, resolved issues, and remaining questions. American Psychologist, 50, 456-457. Paulhus, D. L., & Williams, K. M. (2002). The Dark Triad of personality: Narcissism, Machiavellianism, and psychopathy. Journal of Research in Personality, 36, 556-563. Polygraph Institute of South Africa. The Polygraph Explained. Retrieved 9 February 2010, from http://www.polyinstitute.co.za/faq.htm. PricewaterhouseCoopers (2003). Key elements of antifraud programs and controls: A White Paper. Sackett, P. R., Burris, L. R., & Callahan, C. (1989). Integrity testing for personnel selection: An update. Personnel Psychology 42, 491-529. Sackett, P. R., & Wanek, J. E. (1996). New developments in the use of measures of honesty, integrity, conscientiousness, dependability, trustworthiness, and reliability for personnel selection. Personnel Psychology 49, 787-829.

274


Sarbanes-Oxley Act of 2002, Pub. L. No. 107-204, 116 Stat. 745 (codified as amended in scattered sections of 15 U.S.C.). Sibali, L. (2008, 21 August 2008). Dishonest White-Collar Workers On The Increase Worldwide. Resource Investor. Retrieved 10 November 2009, from http://www.resourceinvestor.com/News/2008/8/Pages/Dishonest-White-Collar-WorkersOn-The-Increase.aspx. Smith, J. K. (2007). Evaluating the boundaries of SEC regulation. Journal of Corporate Finance, 13, 189-194. Spalek, B. (2007). Knowledgeable consumers? Corporate fraud and its devastating impacts. London: Centre for Crime and Justice Studies. Standards Australia (2008). AS 8001-2008 Fraud and corruption control. Sydney. U.S. Congress: Office of Technology Assessment (1990). The use of integrity tests for preemployment screening (OTA-SET-442). Washington D.C.: U.S. Government Printing Office. United States Department of Labor. The Employee Polygraph Protection Act (EPPA) Retrieved 5 February 2010 from http://www.dol.gov/compliance/laws/comp-eppa.htm Waller, A. (2006). Reaping the rewards: Corporate governance. Accounting and Business Magazine, June. London: The Association of Chartered Certified Accountants. Williams, K. M., & Paulhus, D. L. (2004). Factor structure of the Self-Report Psychopathy scale (SRP-II) in non-forensic samples. Personality and Individual Differences, 37, 765778. Zahra, S. A., Priem, R. L., & Rasheed, A. A. (2005). The antecedents and consequences of top management fraud. Journal of Management, 31, 803-828.



Chapter 23

REFLECTIONS ON THE FACTOR STRUCTURE OF THE OCCUPATIONAL STRESS INVENTORY-REVISED Richard E. Hicks Department of Psychology, BondUniversity, Gold Coast, QLD 42299, Australia

ABSTRACT Many studies have identified jobs or occupations where incumbents are most at risk of occupational stress, though stress appears endemic across the whole spectrum of work. The oft cited jobs whose members are most ‗at risk‘ include those of air traffic controllers, nurses and other health professionals, the police, those in management, teachers, and more. But trying to understand the factors involved that lead to the stressful outcomes, even when people are selected for their jobs on relevant criteria, remains a problem. A difficulty lies in the fact that findings from studies of the relationships between stress and well-being at work across different work groups are confused because different instruments are used to measure stress and well-being. Specially designed questionnaires for the particular work group are often cited as being necessary. This may well be the case for some aspects within given jobs. However, using different measuring instruments both clouds and compromises the ability to do cross-group comparisons. One measuring instrument, the Occupational Stress Inventory-Revised (OSI-R), shows promise for use across different occupational areas and if validated would thus allow comparisons on a common base of stress levels, strain experienced and the coping resources used across the various groups. Reflections on the OSI-R and its use in this way and its value for assessing factors associated with stress are given in this paper.

Keywords: Occupational stress, Occupational Stress Inventory-Revised, Stress assessment




276

Richard E. Hicks

INTRODUCTION The incumbents of jobs such as those in air traffic control, defence and police services, health services and teaching (among many) experience high levels of workplace stress, though all incumbents in most jobs experience some level of stress at least, from time to time. Reported studies of workplace stress suggest stress is endemic; such studies include those emphasizing general workplace stress (Beehr, 1995; Byrne, 2003; Caulfield, Chang, Dollard, & Elshaug, 2004; Cooper, 1998; Daniels, 2006; Johnson, Cooper, Cartwright, Donald, Taylor, & Millet, 2005; Murphy, 1996; Osipow, 1998; Zapf, Dormann, & Frese, 1996), stress in management (cf. Cooper, 1998; Donaldson-Feilder, Yarker, Lewis, & Flaxman, 2007; La Montagne, Keegel, Louie, Ostry, & Landsbergis, 2007), stress among educationists (cf. Boyle, Borg, Falzon & Baglioni, 1995; Hicks, Fujiwara, & Bahr, 2006; Kinman, 2001; Kyriacou, 1998; Manthei & Gilmore, 1996; Rudow, 1999), stress in the health and service industries (e.g., Barton, 2009; De Jonge, Le Blanc, Peeters, & Noordam, 2008; Dollard, Winefield, & Winefield, 2003; La Montagne, Keegel, & Vallance, 2007; Lewig & Dollard, 2003; Peterson, 2005; Snow, Swan, Raghavan, Connell, & Klein, 2003; Toch, 2002), and among students (e.g., Christopoulos & Hicks, 2008; Paspaliaris & Hicks, 2010).

STRATEGIES FOR MEASURING STRESS AT WORK The ability to make comparisons of levels and kinds of stress experienced across the different occupational and student groups is limited because most studies use different strategies to gather their data: including individual interview or focus group open discussion methods, open written response methods, and different surveys or short-answer/ multiplechoice surveys. There is also a common emphasis on ‗specially designed surveys‘ for the group being studied (e.g., educationists, entrepreneurs, the police, nurses and so on; cf. in education, Kyriacou, 1998).

Need for a Common Questionnaire in Cross-occupational Comparisons of Stress It would be useful to have available questionnaires or surveys that could enable comparisons across all variables that relate to and impact on stress in workplaces (cf. Boyle et al., 1995; Dollard & de Jonge, 2003; LaMontagne, Keegel, & Vallance, 2007). There have been some attempts to build models that incorporate the variables of importance and relate stress to the environment, personal experiences and personal coping capacities (Cooper‘s extensive work in this field can be cited—cf. Cooper, 1998), but there are few models in practice along with associated measuring instruments that operationalise the full context. Recent work in the UK using the ASSET model (cf. Johnson et al., 2006 cited above) may yield fruit, though that model fails to emphasise personal coping skills and attributes in its survey measurement. Most other stress models are limited mostly to one or two related personal stress dimensions as in assessments in mental health using the biopsychosocial model of stress (e.g., the Depression Anxiety & Stress Scales of Lovibond & Lovibond,

Reflections on the Factor Structure…

277

1995), medical emphases as in the Personal Stress Navigator (Miller & Smith, 1999), and other occupational stress inventories including the Job Stress Survey (Spielberger & Vagg, 1999), and the Occupational Stress Indicator (Cooper, Sloan, & Williams, 1988). Very common in stress research in specific occupations is the use of open-ended questions (interview or written). These questions involve personal or group responses as to how (say) teaching is for the teachers, and their reactions to the issues they face (cf. Fujiwara, 2004; Kyriacou, 1998). Lists of issues or concerns are also often used and the respondents mark those issues that are relevant. However, one model of stress that appears potentially attractive as covering many areas in common across occupations is that of the Occupational Stress Inventory-Revised (OSI-R, Osipow, 1998). This model assesses interaction between the role demands of the job, the experience of stress (strain) and the personal resources of the incumbents. The model uses the OSI-R to assess these three areas, across 14 scales. This paper reflects on how well this inventory operates as a comparative model across different occupational groups. The inventory-model itself is described next and Table 1 defines the 14 scales. Table 1. Scale descriptions of the OSI-R: 14 scales with 10 items per scale Scale What each scale measures Occupational Roles Questionnaire (ORQ) Role Overload Job demands, resources, and ability to complete tasks Role Insufficiency Fit between skills and job, needs for recognition, boredom Role Ambiguity Awareness of one‘s work expectations Role Boundary Level of conflict in loyalties and priorities in workplace Role Responsibility Feeling of responsibility for the welfare and performance of others Physical Environment Exposure to stressful work environment (e.g., heat, noise) Personal Strain Questionnaire (PSQ) Vocational Strain Attitudes towards work, problems in work quantity andor quality Psychological Strain Whether employees are experiencing psychological problems Interpersonal Strain Degree of disruptions in interpersonal relationships Physical Strain Symptoms of physical illness and worries about personal physical health Personal Resources Questionnaire (PRQ) Recreation Level of recreational and leisure activities engaged in Self-Care Participation in stress-reducing habits (e.g., adequate sleep) Social Support Feeling of having support and help from others Rational/Cognitive Coping Knowledge and use of cognitive techniques to deal with stress

278

Richard E. Hicks

BACKGROUND TO THE OCCUPATIONAL STRESS INVENTORY— REVISED: ITS DIMENSIONS The OSI-R was built on a broad theoretical base and has normative data across several professional groups. The OSI-R assesses three inter-related dimensions or factors assessing aspects of occupational adjustment: occupational role stress, personal strain and coping resources. Occupational role stress includes six sub-scales, while Personal strain and Personal resources each include four subscales (see Table 1). The OSI-R theoretical model hypothesises that: (i) work environment stressors and coping resources influence perceptions of work roles, and; (ii) that interactions between work stressors and stress-inducing work roles produce personal or psychological strain, with that strain influenced also by the variety, strength and level of personal resources of the individual. This model therefore provides the basis for meaningful comparisons across a variety of groups, on the three main dimensions and the 14 scales of the questionnaire. Does the questionnaire meet expectations? First the professional manual (Osipow, 1998) of the OSI-R shows all 14 scales (each of ten items) possess high levels of internal consistency (Cronbach Alpha coefficients from .70 to .89). However, would the three-factor (questionnaire) model be reproduced in studies? This aspect is handled next.

THE STRUCTURAL INTEGRITY OF THE OCCUPATIONAL STRESS INVENTORY-REVISED A series of confirmatory factor analyses have been conducted and further studies are under way as part of a programmatic study examining the OSI-R and its structural integrity, resilience across samples, and ability to highlight similarities and differences across different groups. The first of the studies examined 141 secondary school teachers (Fujiwara, 2004; Hicks, et al., 2006) and partially supported the model with confirmation of the 14 scales but suggested the underlying structure of the model needed further study. This early study, however, raised other issues of importance supporting the concern to find a questionnaire that assessed stress reliably and in such a way as to enable comparisons across occupations. Just how stressed were the teachers in the study and were they more stressed or less stressed than other professionals? We were able to use the data from the professional manual (Osipow, 1998) to make some beginning comparisons showing that teachers in this sample were indeed more stressed in at least one of the 14 scales (Role Overload) than the great majority of professionals in other occupations listed in the Manual. This indeed underlined an important value in the use of the OSI-R as a cross-occupational comparison instrument. However, the integrity of the underlying factor or questionnaire structure of the inventory needed more extensive study. The second of the studies revisited the inter-correlational data given in the professional manual of the OSI-R (see Hicks, Bahr, & Fujiwara, 2010) using more than 900 respondents. The results indicated that the underlying structure of the OSI-R was confirmed for the Personal Strain and the Personal Resources questionnaires, but not for the Occupational Roles questionnaire: this questionnaire or dimension could be divided sensibly into two or three different factors (Role Clarity; Role Workload, and perhaps


279

Environmental Impacts). Role Clarity involves the three scales of Role Ambiguity, Role Boundary, and Role Insufficiency. Role Workload includes the Role Overload and the Role Responsibility scales, while the third factor would be the remaining scale regarding the physical work environment itself. Further research is being conducted into the Occupational Roles Questionnaire of the OSI-R and the results will be reported in due course.

POSITIVE PRACTICAL AND THEORETICAL OUTCOMES The technical results overall for the 14 separate scales involved in the OSI-R are all positive (Hicks et al., 2006) and there is therefore support for the use of the 14 scales or subscales of the OSI-R in both practical (clinical, organizational) and theoretical (research) contexts. The 14 scales provide substantial information about stressors (role and environmental pressures), experienced personal strain, and personal resources of individuals. In practice, organisations have been helped to identify which areas seem to be most related to stress in the workplace (through examining grouped section or departmental results from completion by individuals of the OSI-R); they are thus able to use the information for decision processes within the organisation to alleviate stress points. Individuals themselves have been helped to identify their own strengths and weaknesses in personal resources and responses (levels of strain) to the various work role areas assessed by the OSI-R. Thus the OSI-R appears to be working in practice.

CONCLUSION Further data collection across a variety of employed and unemployed groups and student groups has occurred under support from Bond University‘s Research and Consulting Services and the Faculty of Humanities and Social Sciences. More than 1000 Australian students, educational administrators, managers, unemployed persons and others have now completed the questionnaires and several student theses have been completed on how the OSI-R questionnaires and sub-scales relate to a variety of psychological and social dimensions (such as well-being, depression, anxiety, emotional intelligence, and personality factors). Further studies are being conducted with colleagues and data gathered in other countries including in Turkey where studies of teachers are being completed. The data on the individual groups, and the combined data on the OSI-R, will help extend our understanding of how the OSI-R model might be improved in practice. There are good indications that the OSI-R‘s 14 scales are useful, and reasonably comprehensive and effective. Identifying and reporting on what would be an optimal explanatory model for these scales (four or five factors) and what are the main health, wellbeing, and personality correlates of the scales of the OSI-R, based on the data already gathered from the programmatic studies, are the next steps currently being given attention.

280

Richard E. Hicks

REFERENCES Barton, R.(2009). It’s your call: Improving Australian call centres for workers; 2009 survey report. Carlton South, Victoria: Australian Services Union. Beehr, T. (1995). Psychological stress in the workplace. London:Routledge. Boyle, G. J., Borg, M. G., Falzon, J. M., & Baglioni, A. J. (1995). A structural model of the dimensions of teacher stress. British Journal of Educational Psychology, 65, 49-67. Byrne, D. (2003). Occupational stress in Australia in the twenty first century: The health and social costs. Workshop Report, Academy of the Social Sciences. Dialogue, 22, 44-48. Caulfield, N., Chang, D., Dollard, M. F., & Eishaug, C. (2004). A review of occupational stress interventions in Australia. International Journal of Stress Management, 11, 149166. Christopoulos, M. & Hicks, R. E. (2008). Occupational stress, perfectionism and depression among university students. In S. Boag (Ed.), Personality Down Under: Perspectives from Australia (pp. 201-209). New York: Nova Science. Cooper, C. L. (1998). Theories of organisational stress. Oxford: Oxford University Press. Daniels, K. (2006) Rethinking job characteristics in work stress research. Human Relations, 59, 267-291. De Jonge, J., Le Blanc, P., Peeters, M., & Noordam, H. (2008). Emotional job demands and the role of matching job resources: A cross-sectional survey study among health care workers. International Journal of Nursing Studies, 45, 1460-1469. Dollard, M. F., & de Jonge, J. (2003). Measurement and methodological issues in work stress research. In M. F. Dollard, A. H. Winefield, & H. R. Winefield (Eds.), Occupational stress in the service professions (pp. 75-101). London: Taylor & Francis. Dollard, M. F., & Winefield, A. H. (1996). Managing occupational stress: A national and international perspective. International Journal of Stress Management, 3, 69-83. Donaldson-Feilder, E., Yarker, J., Lewis, R., & Flaxman, P. (2007). Understanding the stress management-line management link. Paper presented at the Division of Occupational Psychology Annual Conference of the British Psychological Society, held 10-12 January, in Bristol, UK. Fujiwara, D. (2004). Occupational stress, personal strain and coping resources in rural and urban secondary public school teachers in Australia. D Psychology thesis: Bond University. Hicks, R. E., Bahr, M., & Fujiwara, D. (2010) The Occupational Stress Inventory-Revised: Confirmatory factor analysis of the original inter-correlation data set and model. Personality and Individual Differences, 48, 351-353. Hicks, R. E., Fujiwara, D., & Bahr, M. (2006). Confirmatory factor analysis of the Occupational Stress Inventory-Revised, among Australian teachers. Social Science Methodology Conference, Australian Consortium for Social & Political Research Inc. Published in Proceedings (online) of the Conference held December, 2006, Sydney University. Johnson, S., Cooper, C., Cartwright, S., Donald, I., Taylor, P., & Millet, C. (2005). The experience of work-related stress across occupations. Journal of Managerial Psychology 20, 178-187.


281

Kinman, G. (2001). Pressure points: A review of research on stressors and strains in UK academics. Educational Psychology, 21, 473-492. Kyriacou, C. (1998). Teacher stress: Past and present. In J. Dunham & V. Varma (Eds.), Stress in teachers: Past, present and future (pp.1-13). London: Whurr. LaMontagne, A. D., Keegel, T., & Vallance, D. (2007). Protecting and promoting mental health in the workplace: developing a systems approach to job stress. Health Promotion Journal of Australia, 18, 221-8. La Montagne, A., Keegel, T., Louie, A., M., Ostry, A., & Landsbergis, P. A. (2007). A systematic review of the job stress intervention evaluation literature 1990-2005. International Journal of Occupational & Environmental Health, 13, 268-280. Lewig, K., & Dollard, M. F. (2003). Can I help you? Emotional dissonance, emotional exhaustion and job satisfaction in call centre workers. Special Issue. European Journal of Work and Organisational Psychology, 12, 366-392. Lovibond, S. H. & Lovibond, P. F. (1995). Manual for the depression anxiety stress scales (2nd Ed.). Sydney: Psychology Foundation. Manthei, R., & Gilmore, A. (1996). Is stress among New Zealand teachers increasing? New Zealand Journal of Educational Studies, 29, 73-87. Miller, L. H., & Smith, A. D. (1999). The Personal Stress Navigator. (Copy made available through personal correspondence). Boston, Massachusetts: Stress Directions Inc. Murphy, L. R. (1996). Stress management in work settings: A critical review. American Journal of Health Promotion, 11, 112-135. Osipow, S. H. (1998). Occupational Stress Inventory-Revised Edition (OSI-R): Professional Manual. USA: Psychological Assessment Resources, Inc. Paspaliaris, T., & Hicks, R. E. (in press: 2010). Coping Strategies Employed by University Students in Handling their ‗Occupational Role Stress‘. In R. E. Hicks (Ed.), Personality and individual differences: Current directions. Brisbane: Australian Academic Press. Peterson, C. L. (2005). Risk associated with stress in the health care industry. In C. Mayhew & C. L. Peterson (Eds.), Guide to Managing OHS Risks in the Health Care Industry (pp. 193-210). Sydney: CCH Publishers. Rudow, B. (1999). Stress and burnout in the teaching profession: European studies issues, and research perspectives. In R. Vandenberhe & A. M. Huberman (Eds.), Understanding and preventing teacher burnout: A sourcebook of international research and practice (pp. 3858). Cambridge: Cambridge University Press. Snow, D. L., Swan, S. C., Raghavan, C., Connell, C. M., & Klein, I. (2003). The relationship of work stressors, coping and social support to psychological symptoms among female secretarial employees. Work & Stress, 17, 3, 241-263. Spielberger, C. D., & Vagg, P. R. (1999). Job Stress Survey: Professional manual. Lutz, Florida: Psychological Assessment Resources. Toch, H. (2002). Stress in policing. Washington DC: American Psychological Association. Zapf, D., Dormann, C., & Frese, M. (1996). Longitudinal studies in organizational stress research: A review of the literature with reference to methodological issues. Journal of Occupational Health Psychology, 1, 145-169.



Chapter 24

INTRODUCING THE YWEDO ONLINE COGNITIVE LABORATORY Chris J. Jackson School of Organisation & Management, University of new South Wales Sydney, NSW 2052, Australia

ABSTRACT YWEDO (Jackson, 2007; available at: http://www.ywedo.com/experiments/) is an on-line psychology laboratory that offers interactive experiments, data collection, and easy data output. The laboratory is designed to be an instructional, educational and research resource for academics. The aim of this chapter is to introduce YWEDO to the academic audience as it is likely to prove useful to other researchers. The chapter discusses advantages and disadvantages of online data collection and outlines the modules available for research via the YWEDO laboratory. Strengths of the YWEDO laboratory and suggestions for avoiding problems with online research are discussed.

Keywords: 2D:4D, Baddeley Reasoning Test, Balloon Analog Reaction Task (BART), Bisecting lines, Chimeric faces, Colour Stroop, Global & specific biases, Navon, Online Laboratory, Time estimation, Towers of Hanoi, Trail Making Test, YWEDO, Work performance

INTRODUCTION For the last 20 years, I have worked hard collecting data from applied settings and relating these dependent variables to psychological mechanisms. Each study has been unique and required a large amount of effort to generate just a small amount of data. In about 2005, I




284

Chris J. Jackson

realised that there might be an easier way of going about things. From this line of thinking, the YWEDO (―why-we-do‖) online laboratory was born in 2007. YWEDO provides a generic platform to deliver a range of psychological tests and surveys to participants with a sophisticated online configuration which includes recordings of reaction times, a pre-configured database, and a script for writing and retrieving data from the database. Table 1 gives a list of the modules currently running at YWEDO. YWEDO is modular such that researchers can pick and choose whichever modules are required for their research. Moreover, it is reasonably easy to add new modules. At the moment, YWEDO follows a translational model of internet use in which previously used ideas are translated for internet use (Skitka & Sargis, 2006).

ADVANTAGES The main advantages of conducting online experiments can be classified as follows: large sample size and diversity, increased participants‘ motivation and experimenter‘s absence (Reips, 2002; Skitka & Sargis, 2006). In terms of sample size and diversity, YWEDO provides fast access to participants from across the world (in contrast to laboratory studies where local participants are only useable and only small numbers can use a laboratory at any one time). The range of possible participants at YWEDO is only limited by online access and peoples‘ online habits such that access to very specific target groups is possible and limited only by the creativity of the researcher. Many people might think that participants‘ motivation in online studies is likely to be limited in comparison to participants in laboratory studies. However in some situations this seems unlikely to be the case. My use of YWEDO involves participants who are monitored and only rewarded for providing meaningful, complete and non-random data. In such cases, participants are likely to be motivated. When I have collected data from ―surfers‖ then such participants are unlikely to be motivated and the data are untrustworthy. The idea that good management of participants leads to highly motivated participants fits in with the literature (Rickman, Kiesler, Weisband & Drasgow, 1999). Table 1. Modules of the YWEDO online laboratory

           

Colour Stroop Trail marking Test Towers of Hanoi Time estimation Balloon Analog Reaction Task (BART) Navon style global and specific biases of perception Bisecting Lines Chimeric Faces Baddeley Reasoning Test Self reported digit ratio and digit length Supervisor and third party ratings of work performance Survey data collection

Introducing the YWEDO Online Cognitive Laboratory

285

The absence of an experimenter at YWEDO suggests that a study will be free from experimenter bias and, after set up costs, will be much easier and cheaper to run. The absence of an experimenter may also enable collection of sensitive data although at the same time it seems likely that online participants will need extensive reassurance regarding the security of the data and the uses that will be made of the data. Universities require ethical approval and usually demand detailed descriptions of how research data will be used so university research generally can be expected to provide such reassurance. Moreover the use of University information sheets will generally reassure participants, and the use of instructions to convey the need for honesty usually leads to improved honesty in participants.

DISADVANTAGES The strong advantages YWEDO need to be set against the disadvantages of data collection using this online methodology. Disadvantages include having to use internetconnected samples which may not be representative of the actual population (Lenhart, Horrigan, Rainie, Allen & Boyce, 2003), lack of environmental control, dropout and possible multiple submission (and see Birnbaum, 2004; Reips, 2002). Problems of lack of environmental control for example lead to the absence of the ability to verify instructions being followed and the absence of equipment leads to any number of likely effects resulting from background noise, lighting, viewing angle, music, TV, third parties etc. Data collected from the YWEDO online laboratory is limited by these environmental variables. However, it is important to note that generally the effect of these variables is to reduce the chance of finding significant results not increase it. YWEDO does not minimise environment differences by preloading experimental materials as it was thought that this might lead to more problems than it would solve (such as increasing doubts over participation). Instead, file sizes of experimental materials are minimised and reaction times are measured as accurately as possible given the constraints. This leads to potential concerns about precision and control including whether participants can load web pages, speed of loading, speed of internet access, type of browser being used and type and size of monitor. At YWEDO precise measurement is not possible but in most cases precise measuring of response times to the nearest millisecond is not required. Again, the effect of imprecision is likely to reduce the chance of finding significant effects, but such limitations need to be acknowledged. Unlike lab experiments participants, participants are known to frequently dropout in online experiments and this can have an adverse effect upon the quality of data obtained. First, and mostly obviously, it reduces the sample size and hence decreases likelihood for detecting an effect. However, large sample sizes are easily obtained in online experiments hence the reduction in sample size due to dropout is not in itself a serious problem. Nevertheless dropout can be a problem if there is a non-random reason for dropout as this might adversely affect results. In my research work using the YWEDO, dropout is not a problem because I have only paid participants for completion of a research study. Completion of each module and overall completion of a research study is measured in the YWEDO online laboratory.

286

Chris J. Jackson

Multiple submission is a very important problem for online research which is not faced by laboratory research. YWEDO records IP address of each participant. Although this is not an infallible way of checking identity, it is a reasonably effective way of identifying multiple submission. Another problem faced by YWEDO is that of filtering meaningless data provided by participants. Researchers can often spot participants who are not completing a project properly in a laboratory but this is not possible at YWEDO. To help offset this problem, YWEDO records completion times for each project such that a researcher can identify participants who are too slow or too fast at completing test modules. Moreover, YWEDO was designed to export data easily into SPSS such that SPSS can be used to spot meaningless data. A final problem faced by YWEDO is the increasing and real problem of hacking. Despite password security, some participants may find ways to fake participation, download confidential data, or find other ways of undermining the YWEDO laboratory. The reality is that no online system is 100% secure, yet YWEDO does incorporate many modern methods of reducing the likely successful outcome of a hacking operation. Given that there are more interesting and lucrative targets for hacking, YWEDO is unlikely to be hacked.

THE YWEDO MODULES As already noted, the YWEDO online laboratory consists of a growing collection of modules representing tasks, surveys, experiments or questionnaires. Data are currently stored in an MS Access database and the script is asp or similar.

Colour Stroop The Colour Stroop (derived from Stroop, 1935) is one of the best known classic psychology experiments and is quite widely measured online (e.g., Krantz & Dalal, 2000). It involves the presentation of the names of colours presented in different colours (e.g., the word ―red‖ presented in red or presented in a different colour such as green). Beneath this are presented four different colour options (red, blue, green, pink, grey) in four boxes for the participant to select their response. In some trials, the words are presented in the colour associated with the word, and in other trials the words are presented in a different colour to the word. Participants are asked to identify not the word but the colour of the word. The Colour Stroop task indexes the ability to inhibit well-learned responses. This ability to inhibit is measured by the difference in reaction time to respond between the simple or congruent condition (where the spelling of the word matches the colour of the word) and the complex or incongruent condition (where the spelling of the word does not match the colour of the word). This measure of interference control (inhibition of practiced response in favour of a less practiced response) is a widely used measure of executive function and attention (e.g., Semrud-Clikeman, Steingard, Filipek, Biederman, Bekken, Renshaw, 2000). Also measured and recorded is the number of incorrect responses on each colour and the overall time to complete the Colour Stroop.


287

Trail Marking Test The Trail Making Test (derived from Reitan, 1958) consists of 20 squares distributed across the monitor. Again there is a simple or congruent task and a complex or incongruent task. In the simple task, the squares are numbered 1 – 20, and the participant clicks on each square in ascending numerical order. In the complex task, the squares consist of numbers (1 – 10) or letters (A – J) and the participant is asked to click on the squares in ascending order but with the added task of alternating between the numbers and letters (i.e., 1-A-2-B-3-C, etc.). Reaction time to completing the two trails is recorded. Errors are notified to the participant and the participant must correct the error before continuing. Errors therefore affect the completion time for the task. This is standard method of dealing with errors for this task (Bowie & Harvey, 2006). Similarly to the Colour Stroop Test, the difference between the two tasks is seen as a measure of interference control and therefore a measure of executive function. Others see the results as being related to overall intelligence and neurological deficits (Bowie & Harvey, 2006). Also recorded in the YWEDO online laboratory is the overall time taken to complete the Trail Marking Test.

Towers of Hanoi The Towers of Hanoi is a mathematical game invented in the late 19th Century and again is often used to investigate executive functions related to planning, organising and problem solving. The YWEDO version consists of three rods, and four disks of different sizes which are initially all placed in a stack in ascending order of size such that the smallest is at the top and the largest at the bottom. The objective of the puzzle is to move all the disks from one rod to another rod whilst following these rules:   

Only one disk may be moved at a time. Each move consists of taking the upper disk from one of the rods and moving it to another rod such that it can only go on top of any disks already present. No disk may be placed on top of a smaller disk.

The YWEDO version of the Towers of Hanoi task instructs a participant to complete as many 4 disk solutions as possible within a 2 minute period. The participant moves disks by clicking on a disk and then by clicking the rod where it is to be placed. A message informs the participant if a move is illegal and a countdown provides time information. Feedback is provided to the participant and the opportunity to see how the task is performed is offered. The Towers of Hanoi Task provides a measure of executive function which is different to interference control as measured by the Colour Stroop and the Trail Marking Test. This is because the Towers of Hanoi task requires that a sequence of moves is planned, executed, monitored, and revised in advance of action (Wright, Hardie & Rodway, 2004). A variety of summary measures are recorded in the Tower of Hanoi Task including: (a) number of solutions achieved; (b) total number of moves; (c) number of moves per solution achieved; (d) average time to pick up a disk, and; (e) average time to put down a disk. Also recorded are the details of every move and the time taken to make the move.

288

Chris J. Jackson

Time Estimation Time estimations are made by participants based on the well established prospective time paradigm in which participants are informed in advance that they will be asked to judge the duration of a time interval. Participants are instructed to estimate time at the following intervals and in the following order (5, 10, 15, 20, 25, 30, 35, 40 seconds). Participants press a start button clearly labelled with the time to be estimated. Once the perceived time to be estimated had passed the participant presses the button again. At present the YWEDO time estimation module consists of three conditions. In condition 1, participants are simply presented with the amount of time to estimate; in condition 2, participants are distracted by reading aloud randomised numbers which are presented, and which range in presentation time to prevent subvocal counting; and in condition 3, participants are asked to count according to their natural rhythm (but not to use artificial counting methods to mark the passage of time). Condition 1 is always presented first and then the remaining conditions are presented in random order. Time Estimation is regarded as a key cognitive skill as the ability to estimate short time intervals plays an important role in everyday life in which a response is required at an appropriate point in time. Moreover, it appears to be a cognitive skill related to attention and learning as it is important to monitor and check the progression of time (Taatgen, van Rijn & Anderson, 2007). Accurate perception of time in the presence of a distractor is often conceived as being an executive control task in which people who attend to the task are likely to be the better performers. Data from each time interval are collected as well as overall time taken to complete the module. SPSS code standardises the time estimates and they are then added together to provide a measure of total time estimated. Standardization is necessary so that each time interval equally contributes to the total time estimated. The data are also converted into Absolute Errors by subtracting absolute differences between the estimated time interval and actual time interval. Absolute error shows the proportional difference between objective clock time and judged time, and it is used to assess overall level of accuracy of time judgment (Brown, 1985); an absolute error of zero indicates perfect performance.

Balloon Analogue Risk Task (BART) In the BART a participant is presented with a picture of a small balloon that can be inflated incrementally with pumps. A pump wins points. After each balloon pump, the participant can choose to stop and collect the points or can choose to collect more points by pumping again. If the balloon pops then the money is lost and the participant begins again with another balloon. If the participant chooses to collect the points then the points are saved and the participant begins with another balloon. There are 20 balloons in the YWEDO implementation of BART. Participants undertake one of two conditions. In one condition of the BART, the probability of the balloon bursting increases with each pump such that after 20 pumps there is a 100% chance of the balloon bursting. The probability of the balloon bursting after the first pump is zero, the second pump was 1/19, after the third pump was 1/18, and so on, until the twentieth pump, where a balloon burst was certain.


289

In the second condition of the BART, the task is divided up into first a ―reward‖ condition of 10 balloons and then a ―punish‖ condition of 10 balloons. In the reward condition the probability of the balloon bursting is ½ the chance and in the punish condition the probability of the balloon busting is twice the chance of the first condition. Summary measures which are recorded include number of balloon pops, total points won, total points lost, points finished, and time taken across 20 conditions and across the reward and the punish conditions. Also recorded is information about each balloon including points obtained, number of pumps, total time taken to pump and average pump time. The BART is thought to provide insights into the underlying mechanisms of poor decision-making (Lejuez et al., 2002). Each pump can be interpreted as a risky decision that can result in a gain of points or a loss of all points if the balloon pops. The BART is therefore thought to model real world risk-taking and provide a measure of an individual‘s tendency to engage in risky behaviours (i.e., make risky decisions). The BART has been associated with measures of safety and health risk behaviours (e.g., drug use, gambling, stealing, unprotected sexual intercourse), which are real-world manifestations of poor decision making (Lejuez et al., 2002). The reward and punish condition is interesting as it transforms the BART into a measure of disinhibition such that it is possible to see how people learn to respond in a reversal learning task in which people are first conditioned to respond and then punished for making that response.

Navon Type Measure of Global and Spatial Attention Letter stimuli (C, H, I, S, T, U, O, D, X) were employed for this version of the Navon task (Navon, 1977) in which larger letters are constructed from a series of smaller letters. There are 36 presentations and each letter is presented four times. On half the occasions, trials are congruent (e.g., large A made of smaller A‘s) and on the other half then the trials are incongruent (e.g., large A made up of small C‘s). Navon (1977) reported a global precedence effect was apparent in which there was a bias towards attending to the global as opposed to the local and slower processing of the local as a result of interference from the global. A participant‘s performance on the Navon task seems therefore to be influenced by local versus global and by attentional control mechanisms. Participants are instructed to report the smallest sized letter that they see by choosing from one of five buttons and the time taken to make the decision is recorded as well as the accuracy of the decision. Buttons are only available once the picture of the letter has been fully downloaded and the order of the buttons is random. Results can be used to compare congruent accuracy and time taken with incongruent accuracy and time taken. The measures provide indexes of how much a person focuses on the global or the specific image. There are a number of uses of this task which include using it as a measure of executive control (i.e., how much an incongruent task interferes with a congruent task), using it as a measure of global versus specific analysis, or using it as a measure of hemispheric asymmetry given that there are known global versus local differences in hemispheric functioning.

290

Chris J. Jackson

Bisecting Lines Visual line bisecting is quite often used as a measure of relative hemisphere activation. It is frequently used to measure the balance of activation between the two cerebral hemispheres (Bradshaw, Nathan, Nettleton,Wilson, & Pierson, 1987). Like EEG measures of lateral asymmetry of activation, it appears to be stable over time (Tomarken, Davidson, Wheeler, & Doss, 1992). Jewell and McCourt (2000) have provided a meta-analytic review of 73 studies of line bisecting. Eight horizontal lines are randomly selected between 4 in. (10.16 cm) and 6 in. (15.24 cm) in length and presented together, with their centres offset from a common centre. The participants are asked to use their mouse and click on the centre of each line such that they divide it into two equal segments without measuring. Participants are also instructed to keep their heads towards the centre of the screen and shoulders parallel to the screen. YWEDO records time taken to finish the task, the length of the line and the distance from the left of the line that the participant clicks on the line to represent the centre. Bisecting a line to the right of its true centre represents more attention toward the smaller (right) line segment, and is interpreted as a sign of relatively greater contralateral (left) activation.

Chimeric Faces Test The Chimeric Faces Test (derived from Levy, Heller, Banich, & Burton, 1983) involves 36 presentations of 9 photographs. In each presentation, pairs of faces are presented one on top of the other (the top is an exact mirror image of the one on the bottom). Half of the time the chimera on top have the smile to the participant's left, and the other half have a smile to the participant‘s right. Participants must choose which of the faces (top or bottom) is happier and are not provided with a "can't decide" response. Left is considered indicative of a left-visual-field bias (LVF) and is interpreted as a sign of right-hemisphere control of face perception (and vice versa). In the original study by Levy et al. (1983), most right handers chose happy features to the left; a result interpreted as use of right side of the brain and which is thought to be better at recognizing faces and emotional expression in faces. In YWEDO, a laterality quotient (LQ) is computed as the number of times a participant reports a face to the right is happier minus the number of times the face with the smile to the left is happier, divided by the total number of face pairs (36). A positive number indicates a participant chose the face with the smile to the right more and a negative number means the smiling face to the left was chosen more. Reaction times for each presentation are also computed. The counter starts when the picture is fully downloaded reduce differences in download times as a result of environmental variables such as internet access.

The Baddeley Reasoning Test (Baddeley, 1968) This 64-item ability test is administered in 3 min and measures logical reasoning. Each item is presented in the form of a grammatical transformation and is answered with ‗true‘/‗false‘, e.g., ‗A precedes B-AB‘ (true) ‗A does not follow B-BA‘ (false). The test has


291

been employed extensively in the past (see Chamorro-Premuzic & Furnham, 2006) to obtain a quick and reliable indicator of people‘s intellectual ability as it represents one the quickest ways of providing a reliable measure of gf. YWEDO reports the overall score and the accuracy of each answer. Scores can range from 0 to 64.

Self-reported Digit Ratio and Digit Length The relative lengths of the index finger or 2nd digit and the ring finger or 4th digit (2D:4D) are thought to be sexually dimorphic and the dimorphism is argued to be a correlate of prenatal testosterone and oestrogen. Many gender-related traits have been claimed to correlate with 2D:4D. YWEDO asks people to self-report their digits. In my laboratory research, I have asked research assistants to use rulers and callipers to report digit lengths of participants to the closest millimetre. YWEDO also includes a ―virtual ruler‖ presented on the monitor which aims to report millimetre level accuracy but given different sized monitors this has not in fact proved too easy to achieve. Participants also report whether they self-report digit lengths based on use of rulers or the online virtual ruler.

Supervisor and Third Party Ratings of Performance This module emails a link to third parties such as work supervisors that enables them to provide survey assessments of the participant. In my research, I have achieved a 50% success rate with workers. Once the third party provides the assessment then the personal details of the third party are deleted from the database in the interests of privacy.

Survey Data Collection The current library of self-report questionnaires and surveys available on YWEDO is quite extensive. Any self-report questionnaire can be administered from within YWEDO. Questionnaires presently enabled include the learning styles profiler (Jackson, 2008) and the Jackson 5 measure of revised Reinforcement Sensitivity Theory (Jackson, 2009).

DEBRIEFING AND OUTPUT Many of the modules in YWEDO provide brief feedback to participants regarding their results—the aim being to maintain interest and provide a service to participants. Skitka and Sargis (2006) report that only 41% of online studies to date make specific mention of providing debriefing to participants. A key feature of the YWEDO is that data are output as comma separated variables which are easily input into analytical packages such as SPSS. Prior to input it is important to: (a) delete any multiple submissions by test takers, and; (b) identify any exponential numbers

292

Chris J. Jackson

in the output which SPSS cannot process. An automated module deletes multiple submissions by test takers and should be run prior to exporting the data.

CONCLUSION YWEDO is designed to be a modular online laboratory and the aim of YWEDO is to continually expand and develop the available modules to match current research interests. YWEDO is also developing an online research panel of participants to support research being conducted using the online laboratory or elsewhere. Please contact Chris J. Jackson for access ([email protected]; [email protected]).

REFERENCES Baddeley, A. D. (1968). A three minute reasoning test based on grammatical transformation. Psychonomic Science 10, 341-342. Bowie, C. R., & Harvey, P. D. (2006). Administration and interpretation of the trail making test. Nature Protocols, 1, 2277-2281. Bradshaw, J. L., Nathan, G., Nettleton, N. C., Wilson, L., & Pierson, J. (1987). Why is there a left side underestimation in rod bisection? Neuropsychologia, 25, 735-738. Birnbaum M. H. (2004). Human research and data collection via the Internet. Annual Review of Psychology, 55, 803-832. Brown, S. W (1985). Time perception and attention: The affects of prospective versus retrospective paradigms and task demands on perceived duration. Perception and Psychophysics, 38, 115-124. Chamorro-Premuzic, T., & Furnham, A. (2006). Personality, intelligence and approaches to learning as predictors of academic performance. Personality and Individual Differences, 44, 1596-1603. Jackson, C. J. (2007). The YWEDO online Cognitive Laboratory. Sydney, Australia: Cymeon. Jackson, C. J. (2008). Measurement issues concerning a personality model spanning temperament, character and experience. In G. Boyle, G. Matthews, & D. Saklofske (Eds.), Handbook of Personality and Testing (pp. 73-93). Thousand Oaks, CA: Sage Publishers. Jackson, C. J. (2009). Jackson-5 Scales of revised Reinforcement Sensitivity Theory (r-RST) and their application to dysfunctional real world outcomes. Journal of Research in Personality, 43, 556-569. Jewell, G., & McCourt, M. E. (2000). Pseudoneglect: A review and meta-analysis of performance factors in line bisection tasks. Neuropsychologia, 38, 93-110. Krantz, J. H., & Dalal, R. (2000). Validity of Web based psychological research. In N. M. Birnbaum (Ed.), Psychological Experiments on the Internet (pp. 35-60). Orlando, FL: Academic. Lejuez, C. W., Read, J. P., Kahler, C. W., Richards, J. B., Ramsey, S. E., Stuart, G. L., Strong, D. R., & Brown, R. A. (2002). Evaluation of a behavioral measure of risk-taking:


293

The Balloon Analogue Risk Task (BART). Journal of Experimental Psychology: Applied, 6, 75-84. Lenhart, A., Horrigan J., Rainie L., Allen K., Boyce A., et al. (2003). The Ever-Shifting Internet Population: A New Look at Internet Access and the Digital Divide. Washington, DC: Pew Internet Am. Life Proj. Levy, J., Heller, W., Banich, M. T., & Burton, L. A. (1983). Asymmetry of perception in free viewing of chimeric faces. Brain and Cognition, 2, 404-419. Martinez, B. B. (1994). The role of cognitive changes in immediate and remote prospective time estimation. Acta Psychologica, 85, 99-121. Navon, D. (1977). Forest before trees: The precedence of global features in visual perception. Cognitive Psychology, 9, 353-383. Reips, U. D. (2002). Standards for internet based experimenting. Experimental Psychology, 49, 243-256. Reitan R. M. (1958). Validity of the Trail Making test as an indicator of organic brain damage. Perception and Motor Skills, 8, 271-276. Richman W., Kiesler S., Weisband S., & Drasgow, F. (1999). A meta-analytic study of social desirability distortion in computer-administered questionnaires, traditional questionnaires, and interviews. Journal of Applied Psychology, 84, 754-755. Semrud-Clikeman, M. Steingard, R. J., Filipek, P., Biederman, J., Bekken, K. & Renshaw, P. (2000). Using MRI to examine brain behavior relationships in males with attention deficit disorder with hyperactivity. Journal of American Academy of Child Adolescent Psychiatry, 39, 477-484. Skitka, L. J., & Sargis, E. G. (2006). The internet as psychological laboratory. Annual Review of Psychology, 57, 529-555. Stroop, J. R. (1935). Studies of interference in serial verbal reactions. Journal of Experimental Psychology, 18, 643-662. Taatgen, N. A., van Rijn, H., & Anderson, J. (2007). An integrated theory of prospective time interval estimation: The role of cognition, attention and learning. Psychological Review, 114, 577-598. Tomarken, A. J., Davidson, R. J., Wheeler, R. E., & Doss, R. (1992). Psychometric properties of resting anterior EEG asymmetry: Temporal stability and internal consistency. Psychophysiology, 29, 576-592. Wright, L., Hardie, S, M., & Rodway, P. (2004). Pause before you respond: handedness influences response style on the Tower of Hanoi Task. Laterality, 9, 133-147.

PART D: DEVELOPMENTS IN THEORY AND STRUCTURE



Chapter 25

SIX SOLUTIONS TO THE CIRCULAR NATURE OF TRAIT PSYCHOLOGY Jonathan P. Gerber Department of Psychology, Macquarie University, Sydney, NSW 2109, Australia

ABSTRACT This chapter explores non-circular explanations of personality. Trait psychology has sometimes used illogical, circular reasoning when trying to explain behaviour. Circularity is seen when behaviour is used to define a trait and then this trait being used to ‗explain‘ the same behaviour. However, personality theory has also attempted to move beyond pseudo-explanation by offering many ways to jump outside the trait-behaviour circle. This chapter collates and critiques non-circular accounts of personality, focusing on biology, evolution, human development, learning, factor analytic accounts, and perception. It is recommended that researchers avoid substituting self-report measures for biological or developmental measures, that factor account analytic accounts not be considered definitional, and that greater care is taken to understand the items involved in measuring a trait.

Keywords: Circularity, Personality, Rejection sensitivity, Trait psychology

INTRODUCTION Nigel Tufnel: … the numbers all go to eleven. Look… right across the board… eleven, eleven, eleven and... Marty DiBergi: Oh, I see. And most amps go up to ten? Nigel Tufnel: Exactly. 

Jonathan P. Gerber is now at Gordon College, MA. Ph: 1-978-867-4821; Fax: 1-978-867-4671; [email protected].

298

Jonathan Gerber Marty DiBergi: Does that mean it's louder? Is it any louder? Nigel Tufnel: Well, it's one louder, isn't it? It's not ten. (This Is Spinal Tap: The official companion, 2002, pp. 29-30)

Much like Nigel Tufnel chasing his tail explaining the volume of his amplifier, some (e.g., Skinner, 1965, Boag, in press) have argued that trait explanations of personality are sometimes flawed because they commit a key logical fallacy: circularity. If such a charge is true then personality psychology offers little in the way of explanation and instead much posthoc meandering. The purpose of this chapter is to explore whether the charge of circularity is correct when applied to personality psychology and, in particular, trait theories of personality. We will find that trait psychology is often actively engaged in seeking non-circular explanations of behaviour.

CIRCULARITY Circularity occurs when ―the construct used to explain some effect (the explanans) is equivalent to the effect that it is said to explain (the explanandum)‖ (Boag, in press, p.6). In trait psychology terms, it is when we define a trait on the basis of a group of behaviours and then use the resulting trait to explain that same group of behaviours. A recent example of circularity is the Big Five model of personality. For example, if someone likes to go to parties they are assumed to be extraverted. If we then ask why someone likes to go to parties, it is assumed to be due to their extraversion. The explanation (extraversion) is equivalent to that which is purportedly explained (liking for parties). A simple way of putting this is illustrated in Figure 1. A set of behaviours is used to define and trait and then the trait is used to explain the same set of behaviours. The definition and the explanation are circular because they refer to the same set of behaviours. There is nothing ‗outside the circle‘. 1

Trait

causes

Behaviour

is used to define Figure 1. Conceptual diagram of circularity. Note: Circularity: behaviours are used to define a trait (the black arrow) & then that same set of behaviour is used as an explanation for the behaviour (the grey arrow). 1

Closely related to circularity is a second logical fallacy called reification. Reification occurs when a relationship between two entities is taken to be an entity in itself. For example, mathematical ability, a relationship between each test taker and the test, is sometimes reified and used to explain performance on a maths test. This reification is circular because the definition and explanation involve the same two terms, even though the two terms (maths ability and test performance) are disguised by a reified relationship (Passmore, 1935). Given that most cases of reification in personality psychology also involve circular explanations, this chapter will focus only on the challenge of circularity and not reification.

Six Solutions to the Circular Nature of Trait Psychology

299

Are there dangers in circularity? In many cases, circularity does not matter. Circularity is fine when we wish to offer descriptions. Even a great circle-breaker like Skinner noted that circularity is no big deal when defining and describing phenomena (Skinner, 1954). However, when circular reasoning is involved in explanations then there are two reasons why it is destructive. Firstly, as outlined above, circular explanations are redundant. To define a trait in terms of observed behaviour and then use that trait to explain behaviour adds nothing to the original phenomena. There is nothing added to the circle, so to speak. Second, circular explanations are untestable. If maths ability is axiomatically defined by performance on a maths test then maths ability will always (by virtue of its definition) perfectly explain maths performance, with no chance for it to be proven right or wrong. Circularity obscures knowledge and impedes progress in understanding the trait in question. For the remainder of this chapter I will explore the means by which trait psychology has attempted to overcome circularity and show that trait psychology is often non-redundant and testable.

REJECTION SENSITIVITY To illustrate non-circular explanations I will refer to a relatively new personality construct, rejection sensitivity (Mehrabian, 1976; Downey & Feldman, 1996). Rejection sensitivity is defined as the extent to which an individual expects rejection and interprets ambiguous social situations as instances of rejection. Highly rejection sensitive individuals expect rejection and interpret ambiguous situations as rejection. Low rejection sensitive individuals are less likely to expect rejection and do not interpret ambiguous situations as rejection. Rejection sensitivity is typically, although not exclusively, assessed using the Rejection Sensitivity Questionnaire (Downey & Feldman, 1996). Respondents are given a variety of ambiguous situations (e.g., ―You ask your parents for extra money to cover living expenses‖ or ―You ask someone you don‘t know well out on a date‖), then answer two questions for each situation: how concerned or anxious they would be about the situation, and the likelihood of this situation leading to rejection. The situations are relatively ambiguous; for example, parents could be upset or accepting of the request for extra money. Those high in rejection sensitivity are likely to expect and perceive rejection (i.e., believe their parents will not help them) while those low in rejection sensitivity are likely to interpret the situation as non-threatening (i.e., believe their parents will probably help them out). The causes and consequences of rejection sensitivity are being actively explored. High rejection sensitivity is associated with specific brain activity when viewing pictures of rejection (Kross, Egner, Ochsner, Hirsh & Downey, 2007). Rejection sensitive individuals tend to have elevated startle reflexes (Gyurak & Ayduk, 2007) and are more likely to have conflict in close relationships (Downey & Feldman, 1996). Paradoxically, rejection sensitivity is self-fulfilling; highly rejection sensitive individuals tend to react defensively to ambiguous situations which in turn elicits hostility from those around them (Downey & Feldman, 1996). Early rejection episodes are associated with later rejection sensitivity (Butler, Doherty & Potter, 2007; London, Downey, Bonica & Paltin, 2007). Given this variety of findings, rejection sensitivity can illustrate the different approaches to non-circular explanation in trait psychology.

300

Jonathan Gerber

SOLUTION 1: BIOLOGICAL SUBSTRATES Trait psychologists have long used biological substrates as a non-circular mode of personality explanation and Boag (in press) notes that this is an obvious and promising direction for non-circular trait accounts (pp. 27-8). Eysenck‘s ARAS (Eysenck, 1967) and the BIS/BAS model of personality (Gray, 1972, 1982) both posit underlying biological individual differences that explain differences in behaviour. The biology exists outside the circle to define the trait of interest with the behaviour an outcome of the biologically defined traits (Figure 2). Note that the behaviour itself does not come back to redefine the biology, removing the circularity. Biological explanations are non-redundant, they specify additional biological systems, and are testable (e.g., do individual differences in biology correlate highly with observed behaviour?). However, for ease of conducting research, researchers often substitute selfreport measures for biological measures of the trait. For this to be valid, the convergent validity between biology and self-report needs to be strong. This is clearly not the case with rejection sensitivity, the RSQ only correlates 0.5 - 0.6 with measures of brain activity (Kross et. al., 2007). The substitution of self-report measures for biological measures where convergent validity is not demonstrated undermines attempts to break the circularity of trait psychology. However, the possibility of poor convergent validity also shows the testable and non-circular nature of the biological approach.

Trait

causes

Behaviour

defines Biology Figure 2. Biological trait theories. Note: Behaviour does not come back to define biology, & hence the definition is non-circular.

SOLUTION 2: EVOLUTION Evolutionary accounts of personality (e.g., Buss, 1991) also attempt to break circularity by stressing distal factors in individual differences (see Figure 3) and their distal nature makes them only partially useful. Evolutionary explanations are also comparatively hard to test, and have difficulty explaining the different expressions of behaviour between individuals (Skinner, 1956). Alertness to rejection could be evolutionarily adaptive because it helps individuals stay in social groups, providing a survival advantage (Spoor & Williams, 2007). Similarly, rejection sensitivity may be adaptive because it develops in those most likely to be excluded from the group and hence suffer losses due to exclusion. In any event, individual differences cannot be measured evolutionarily and researchers must resort to other measures. Therefore, the evolutionary account, while logically well-intentioned, is unable to provide a practical solution for assessing individual differences.


Trait

causes

301

Behaviour

defines

Evolution Figure 3. Evolutionary trait theories. Note: Although non-circular, it is hard to define traits by evolution.

SOLUTION 3: LIFESPAN DEVELOPMENT Developmental accounts of personality offer more proximal explanations of personality traits, suggesting that traits develop in an individual‘s past, encompassing both genes and environmental influences (Figure 4). Attachment theory (Ainsworth, Blehar, Waters & Wall, 1978), for example, argues that early experiences with caregivers can lead to lasting individual differences in relationship outcomes (Hazan & Shaver, 1987). Developmental accounts suggest traits arise in two ways, from genes and environmental influences on biology, and from learning due to the environment. Genes and environmental influences have been dealt with under Solution 1, while learning approaches will be considered more fully in Solution 4. Learning approaches to development of traits are strong because the prior experiences that lead to a trait may not have involved any behaviour on part of the individual. For example, early rejection experiences are said to lead to the development of rejection sensitivity. But we note that it is the amount of rejection experiences, and not the reaction of the individual, that is the developmental cause. Given that rejection sensitivity is a perception of ambiguous situations and defensive reactions, the developmental definition (i.e., rejection experiences) is logically distinct from the trait itself (i.e., reactions to ambiguous situations).

Trait

causes

Behaviour

defines Development Figure 4. Developmental trait accounts. Note: Development comprises both genes, environmental influences on biology, and classical and instrumental conditioning.

These earlier behaviours can be viewed directly using longitudinal designs or recounted via self-report, allowing for testable definitions of the construct of interest. The testing of the link between definition and later outcome demonstrates the non-circular nature of

302

Jonathan Gerber

developmental trait accounts. However, as in biological measures, the correlation between the longitudinal measure and the later self-report scales are essential to ensure that one can be substituted for the other. In the case of rejection sensitivity, although rejection sensitivity could be measured by early rejection experiences, little direct observation has been made of these early experiences and the correlation between the widely used RSQ and early experiences remains untested, making the substitution of the RSQ unreasonable.

SOLUTION 4: LEARNING Learning accounts of traits suggest that differential reinforcement of behaviour leads to individual differences in traits. The trait is a sum of previous learning experiences (Figure 5). Where developmental accounts suggest that an individual‘s experience impacts upon later traits, learning accounts explicitly note the importance of the behavioural response to experience.

Trait

causes

Learning

Figure 5. Learning trait theories. Note: Past experiences (i.e., learning) are separated in time from the later behaviour, providing a weak form of non-circularity.

Rejection sensitivity can be explained by learning. Ambiguous situations are associated with defensive reactions in rejection-sensitive individuals. These negative reactions are punished by hostile reactions from relational partners (Downey & Feldman, 1996), paradoxically reinforcing the perception that ambiguous situations will lead to rejection. Learning accounts are not technically circular. To say that an individual in the past responded to a situation in one way and then did so again in the future is not circular because the situations are unique in time. However, personality psychology tends to assume that situations are functionally and psychologically equivalent, so an explanation of a future behaviour from a similar past behaviour is not very compelling, even though it is noncircular.

SOLUTION 5: COMPONENTS AND HIERARCHY Trait theories often appeal to components of underlying constructs to explain higher-level traits (Figure 6). Factor analytic approaches to intelligence (e.g., Carroll, 1991) and personality (Costa & McCrae, 1991) typify these approaches. Within the factor analytic


303

tradition, component behaviours are combined into broader factors via an analysis of covariance. In intelligence, this has resulted in a high level factor called g. Among many other criticisms of g, it has been criticised for being undefined. The lack of definition of g is actually a strength because if g is defined by the components then circularity follows: the behaviour we wish to explain by g is contained in the totality of the definition of g. Less fear about leaving things undefined can help avoid circularity. To avoid circularity, g (and the Big Five) need to be defined apart from their constituent parts. Circularity arises when we then combine these components into our measures of the trait and then operationalise the trait. In general, it is better when the behaviour used to measure a construct is of a different type to the overarching construct. In addition to providing nonredundant information, defining the construct without reference to the component parts allows for a test of the definition, difficult though such a test may be. Rejection sensitivity has avoided this trap by only including one component of the definition, interpreting ambiguous situations, in its operationalisation in the RSQ. Defensive reactions to rejection are not assessed by the RSQ. This leaves the broader concept slightly larger than the operational definition. We have here a situation of partial circularity, although the link between defensive reactions and the RSQ is testable and not circular, the perception of ambiguous situations is largely circular.

A, B, & C provide evidence for

Overarching construct

Trait

causes

Behaviour A, B, & C

defines

Figure 6. Factor analytic trait accounts. Note: The higher level trait is commonly not defined by the components but is typically stated at a more molar level uncapturable by the individual components.

SOLUTION 6: INTRA-INDIVIDUAL COGNITIVE PROCESSES The last solution is to look at intra-individual psychological processes that may underlie personality traits (Figure 7). For example, highly anxious individuals tend to have biased attention; they pay attention to threatening stimuli in different ways to non-anxious individuals (Benoit, McNally, Rapee, Gamble & Wiseman, 2007). In the attention fixation and lexical decision tasks, people are presented with words (singly or in pairs) and then asked to indicate which words are words or non-words while also measuring the participant‘s attention via eye movement data. Alertness to threat would be tested by giving subjects threatening words such as ‗fear‘ or ‗danger‘. Words such as these are not directly related to the behaviour in question nor explicitly to any social situation and hence the issue of circularity is removed.

304

Jonathan Gerber

Cognitive process

Trait

causes

Behaviour

defines Figure 7. Cognitive processes accounts of personality. Note: Cognitive processes can be measured independently of the later behaviour.

Rejection sensitivity might similarly be construed in attentional terms, with individuals perceiving rejection where there is none. The correlations between these attentional biases and behaviour could later be tested, demonstrating the non-circular nature of the explanation.

CONCLUSION Trait psychology has offered many logical, coherent explanations of personality. While some explanations turn out to be empirically unsupported, the ability for these explanations to be tested shows the non-circular nature of the explanation. In the case of rejection sensitivity, nearly all the approaches to breaking circularity have been pursued. Biological underpinnings, evolution, infant development and learning accounts have all been explored to understand the basis of the rejection sensitivity trait. Rejection sensitivity may be the product of early rejection experiences, or there may be no association. Rejection sensitivity may involve heightened startle reflexes, or it may not. Rejection sensitivity might arise from early experiences and develop into a consistent perceptual bias, adding further information to the definition. This research on rejection sensitivity shows that the explanations used in trait psychology do not show the redundant and untestable nature of circular explanations. Rejection sensitivity research illustrates the paradox of circular explanations: failure to find support for an explanation in psychology is disheartening at an empirical level but also heartening because it proves the logical nature of the explanatory attempt. One of the key problems of circularity is that the explanation is essentially non-testable. Explanations of rejection sensitivity have not always been supported (e.g., the moderate correlations between brain behaviour and RSQ) but at least the explanation could be disproven. Being able to be wrong empirically is a good way to ensure being correct logically. I suggest that much of trait psychology may similarly be logically correct but empirically untrue. To suggest that trait psychology is necessarily circular is misleading and unfair to the breadth and depth of much trait psychology. While some explanations are circular (e.g., Big Five, as identified by Boag, in press) researchers are pushing for explanations of personality that do not involve mere re-descriptions of the behaviour themselves. We should remain vigilant to ensure that circularity does not creep in, we should always be aware of the detail of our assessment items, we should avoid blindly substituting self-report for biological or developmental measures, we should avoid converting factor-analytic accounts into definitions, we should be unafraid of empirical failures, but we should not be deluded into believing that trait theory has made little progress since the critiques of Skinner and Passmore three generations ago.


305

REFERENCES Ainsworth, M. D., Blehar, M., Waters, E., & Wall, S. (1978). Patterns of attachment: A psychological study of the strange situation. Hillsdale NJ: Lawrence Erlbaum Associates. Benoit, K. E., McNally, R. J., Rapee, R. M., Gamble, A. L., & Wiseman, A. L. (2007). Processing of emotional faces in children and adolescents with anxiety disorders. Behaviour Change, 24, 183-194. Boag, S. (in press). Explanation in personality psychology: ‗Verbal magic‘ and the FiveFactor Model. Philosophical Psychology. Buss, D. M. (1991). Evolutionary personality psychology. Annual Review of Psychology, 42, 459-491. Butler, J. C., Doherty, M. S., & Potter, R. M. (2007). Social antecedents and consequences of interpersonal rejection sensitivity. Personality and Individual differences, 43, 1376-1385. Downey, G. & Feldman, S. I. (1996). Implications of rejection sensitivity for intimate relationships. Journal of Personality and Social Psychology, 70, 1327-1343. Eysenck, H. J. (1967). The biological basis of personality. Springfield, IL: Charles C. Thomas. French, K. (Ed.) (2000). This is Spinal Tap: The official companion (2000). London: Bloomsbury. Gray, J. A. (1972). The psychophysiological basis of introversion-extra-version: A modification of Eysenck's theory. In V. D. Nebylitsyn & J. A. Gray (Eds.), The biological bases of individual behaviour (pp.182-205). San Diego, CA: Academic Press. Gray, J. A. (1982). The neuropsychology of anxiety: An enquiry into the functions of the septo-hippocampal system. New York: Oxford University Press. Gyurak, A., & Ayduk, O. (2007). Defensive physiological reactions to rejection: The effect of self esteem and attentional control on startle responses. Psychological Science, 18, 886892. Hazan, C., & Shaver, P. (1987). Romantic love conceptualized as an attachment process. Journal of Personality and Social Psychology, 52, 511-24. Kross, E., Egner, T., Ochsner, K., Hirsh, J., & Downey, G. (2007). Neural dynamics of rejection sensitivity. Journal of Cognitive Neuroscience, 19, 945-956. London, B., Downey, G., Bonica, C. & Paltin, I. (2007). Social causes and consequences of rejection sensitivity. Journal of Research on Adolescence, 17, 481-506. Mehrabian, A. (1976). Questionnaire measures of affiliative tendency and sensitivity to rejection. Psychological Reports, 38, 199-209. Passmore, J. A. (1935). The nature of intelligence. Australasian Journal of Psychology and Philosophy, 13, 279-289. Skinner, B. F. (1965). Science and human behavior. New York: Free Press Spoor, J. R., & Williams, K. D. (2007). The evolution of an ostracism detection system. In J. P. Forgas, M. Haselton, & W. von Hippel (Eds.), The evolution of the social mind: Evolutionary psychology and social cognition (pp. 279-292). New York: Psychology Press.



Chapter 26

A PROPOSITION FOR A PARSIMONIOUS TAXONOMIC CONVERGENCE OF PERSONALITY TRAITS Niko Tiliopoulos School of Psychology, University of Sydney, Sydney, NSW 2006, Australia

ABSTRACT Personality traits‘ taxonomies are normatively based on the pre-structuring of testcomponents and factor analytic (FA) interpretations. These procedures have been criticised, especially FA‘s limitation to accurately identify the underlying structure of constructs. This chapter examines whether the five factor model of traits, as measured through the NEO-PI-R, contains the same trait structures when analysed through alternative statistical procedures. Nonmetric Multidimensional Scaling (NMDS) was employed alongside FA techniques (N = 384). FA indicated the presence of five factors, but NMDS suggested a more parsimonious arrangement, comprising three (similar to Eysenck‘s PEN model), or even two ―super structures‖. Furthermore, a circular trait arrangement was a reasonable alternative, suggesting that traits can be meaningfully (re)classified based on their kind and their integrative centrality. Over-reliance on FA models might not be the most appropriate approach for a better understanding of personality.

Keywords: Circular regression, Big-3, Big-5, Multidimensional scaling, NEO, Radex, Traits



Ph: 61-2-9036 9223; Fax: 61-2-9036 5223; email: [email protected]; Reprinted from Personality and Individual Differences, 48/1, Tiliopoulos, N., Pallier, G., & Coxon, A.P.M., A circle of traits: A perceptual mapping of the NEO-PI-R, 34-39, Copyright 2010, with permission from Elsevier.

308

Niko Tiliopoulos

INTRODUCTION Attempts to uncover the ―structure‖ of personality have been a focal point in psychology for over 50 years. The trait approach to personality is arguably one of the currently dominant perspectives in the field, despite the fact that it has experienced both theoretical and empirical falls and rises in popularity through the years. From within this approach it has been repeatedly argued that an understanding of traits is essential for understanding personality as a whole (e.g., Mathews, Deary, & Whiteman, 2003). However, a variety of, at times polemic, traits‘ typologies, taxonomies, models, and theories have been and are being proposed, following primarily the lexical or the psychobiogical approach to personality (for an overview see Boyle, Matthews, & Saklofske, 2008; John, Robin, & Pervin, 2008). One of the most popular and widely used such typologies is the five factor model (FFM) (Costa & McCrae 1988; Goldberg, 1993). Over the past 50 or so years, countless analyses of lexical traits markers, or assessments of such, resulted in the same number of five dominant factors (e.g., Bagby et al., 1999; Egan, Deary, & Austin, 2000; Fiske, 1949; Guilford, 1975; Norman, 1963; Tupes & Christal, 1961). This led to Goldberg‘s (1981, p. 160) comment that ―any model for structuring individual differences will encompass these big-five dimensions‖ and to Digman (1990) noting that the FFM is capable of explaining nearly all accounts of personality traits. Arguably, the most used instrument for assessing the FFM is the NEO Personality Inventory (NEO-PI-R, Costa & McCrae, 1992). The NEO-PI-R comprises 240 statements, to which the participants respond on a Likert-scale by indicating the degree of their agreement to whether a particular statement is relevant to themselves. The NEO-PI-R provides an overall measure of the five broad dimensions of Neuroticism (N), Extraversion (E), Openness to experience (O), Agreeableness (A), and Conscientiousness (C). Each of these domains comprises 6 facet-scales (with 8 items per scale). Thus the NEO-PI-R is capable of giving a fine-grained analysis of personality, and possesses very good validity and reliability indices. Two of the more fundamental criticisms of the FFM relate to the necessity (if not fundamental existence) of the five factors and the assumed orthogonality across them. Indeed, Eysenck (1992; and elsewhere) is an acknowledged critic of the FFM (and by implication the NEO). Accordingly, he argued that, whilst the N and E dimensions are present in his own three factor model (often called the PEN model or the Big-3), A, C and O are not. In Eysenck‘s view A and C are subsumed by his Psychoticism dimension, while the O dimension is to be dismissed altogether as a reflection of intelligence and hence, not really a part of the structure of personality. Such proposals have implications for both the measurement and the underlying theoretical structure of the FFM. Additionally, if one considers the typical procedures of test-construction then these implications become readily apparent. When a new questionnaire is devised it is usual to begin with many more items than are intended for the final version. By examining the intercorrelations and reliability data of individual items in comparison to other questions, it is possible to identify those which do not appear to be similar to the proposed scale. These items are then eliminated until the test constructers arrive at a selection of questions which appear to ―conform to their expectations‖. Thus, a somewhat arbitrary process takes place in which the subjective interpretations of the researchers are fulfilled, including the (desired) orthogonality of the expected factor structure. Indeed intercorrelations between the five NEO

A Proposition for a Parsimonious Taxonomic Convergence of Personality Traits

309

domains have been reported (e.g., Costa, McCrae, & Dye, 1991), with N and C being correlated at -.49 and a correlation between E and O of .43, thus not lending great support to the assumption of orthogonality. If we consider the fact that a correlation of .3 could justify the notion that two scales are measuring a broader super-trait, then this would potentially support the idea that maybe there is ―too much‖ overlap between the traits and thus the variance could be more parsimoniously accounted for by fewer dimensions. Draycott and Kline (1991) appear to agree, arguing, for example, that C may indeed be little more than an aspect of Eysenck‘s Psychoticism. Furthermore, an attempt to examine the theoretical parameters of the FFM using confirmatory factor analysis (CFA) of the NEO-PI-R found a poor fit between the obtained factorial structure and the hypothesised dimensions corresponding to the FFM (Parker, Bagby, & Summerfield, 1992). A comparative analysis of the NEO PI-R as a measure of the FFM and the six-factor personality questionnaire (6FPQ) found that the NEO-PI-R could equally represent the six-factor model of personality (Jackson, Ashton, & Tomes, 1996). In fact, the NEO-PI-R can easily support both a five-factor as well as a three-factor solution, which leads to the conclusion that whilst factor analysis is seen as the ―tool‖ to resolve issues such as five versus three factors, in reality the techniques employed cannot live up to that expectation (Caruso & Cliff, 1997). Since the FFM is rooted in the method of factor analysis, we, as researchers, can only truly evaluate this model with a thorough understanding of this method of analysis. Block (1995) argues that the present and powerful effects of prestructuring are often not sufficiently recognised in relation to the FFM, and that simply the ―sameness‖ of the variable-set may account for the fact that the FFM has been recovered in so many diverse samples. There are others who also advise us to use factor analysis in personality research with caution (Hammond, 1987; Maraun, 1997). For example, Maraun (1997) argues that there are crucial problems in relying upon the Dimensional Interpretation and Simple Structure Process (DISS). For example, most commonly only correlations above .3, or .4 are considered as ―valid‖, thus actively ignoring much of the interrelationships between the variables. He notes that this process is used throughout the FFM literature and yet Maraun comments that the combination of factor analysis and the DISS process does not actually allow the researcher to accurately represent the structure. The argument is that this procedure, by systematically ignoring numerous features of the structure, makes it premature to claim that the structure of traits is five orthogonal dimensions. Considering then that: (a) there seem to be questions surrounding the over-reliance on factor analysis in understanding the structure of personality; (b) the FFM has not always been found to be recoverable through other statistical approaches, and; (c) a three-factor taxonomy can also be found in the Big-5 data, it is surprising that attempts to replicate the FFM using other methodologies are practically nonexistent or at best sparse. Table 1. Mean, standard deviation and Cronbach’s alpha of the NEO-PI-R dimensions (N = 384). M SD alpha N 94.18 26.01 .97 E 111.25 20.60 .75 O 110.05 16.09 .77 A 122.29 16.50 .71 C 109.85 21.72 .83

310

Niko Tiliopoulos

The aim of this chapter was to look, along side the more traditional forms of FA, at the usefulness of an alternative technique, viz. Nonmetric Multidimensional Scaling (NMDS), in order to investigate whether the latter can provide further insight into the underlying structure of personality and the relationship between the FFM and Eysenck‘s Big-3.

METHOD Participants and Procedure A total 384 adults completed the NEO-PI-R and a brief demographic questionnaire. Participants were recruited using several methods, ranging from invitations to existing volunteer participant databases, posters, newspaper articles and adverts. In order to collect a reasonably representative population sample, care was taken to ensure that the participant pool reflected a broad range of ages and educational backgrounds. There were 273 females (71.4%) and 109 males (28.6%), while two people did not reveal their sex. Six age ranges were represented: 17-24 years (23%), 25-34 years (26%), 35-44 years (18%), 45-54 years (14%), 55-64 years (13%) and 65+ years (6%), their mean age being 38.2 years (SD = 15.4 years). Of the sample, 113 (29.4%) individuals were classified as university students, 267 (69.5%) were in various occupations, while 4 (1%) participants were undetermined.

RESULTS Reliability and FA Summaries of trait means, standard deviations and Cronbach‘s alpha reliabilities are presented in Table 1. The dimensions appear highly reliable, especially N and C. A Principle Components Analysis with Varimax rotation was then performed on the 30 facet scales of the NEO-PI-R. The retained first five factors explained 60.6% of the variance and the rotated solution converged in six iterations (Table 2). As it is evident from Table 2, the five factors represent almost unequivocally the five dimensions of the NEO, and the results closely resemble the ones presented in the NEO-PI-R manual (Costa, & McCrae, 1992, p. 44). However, three facets (shown in bold) did not behave as expected. Activity (E4), altruism (A3), and assertiveness (E3) cross-loaded. However, their behaviour did not deviate highly from the one reported in the NEO manual (pp. 44, 45). In the manual, these facets are expected to cross-load in the way they do, albeit not at this magnitude. With the use of Fisher‘s z’ distribution, approximate values for the confidence intervals (CI) of the observed loadings of these facets on other than the dimensions they belong were calculated. The approximate 95% CI for the observed loadings of E4 on C were .36 to .52; for A3 on E were .45 to .59; and finally, for E3 on A were -.36 to -.52. As it can be seen, the expected loadings for these facets tend to fall well within those intervals, with the exception of the loading of E3 on A, which falls .04 points above the lower limit (loading reported in the NEO manual: r = -.32). Overall, and by taking into account my introductory


311

discussion, it seems reasonable to suggest that the NEO-PI-R exhibits quite acceptable coherence and reliability. Table 2. Internal structure of the NEO-PI-R facets after principal components analysis with Varimax rotation

Facets Anxiety (N1) Depression (N3) Self-consciousness (N4) Vulnerability (N6) Angry hostility (N2) Impulsiveness (N5) Self-discipline (C5) Achievement-striving (C4) Competence (C1) Dutifulness (C3) Order (C2) Activity (E4) Deliberation (C6) Straightforwardness (A2) Compliance (A2) Modesty (A5) Tender-mindedness (A6) Assertiveness (E3) Trust (A1) Aesthetics (O2) Ideas (O5) Fantasy (O1) Values (O6) Feelings (O3) Actions (O4) Warmth (E1) Gregariousness (E2) Positive emotion (E6) Altruism (A3) Excitement seeking (E5)

N (14.89%) .84 (.81) .83 (.80) .77 (.73) .75 (.70) .73 (.63) .48 (.49)

C (13.17%)

Domain A (10.93&)

O (10.88%)

E (10.75%)

.78 (.75) .76 (.74) .74 (.64) .73 (.68) .71 (.70) .45 (.42) .40 (.57)

.45 (.54) .69 (.68) .69 (.77) .64 (.59) .62 (.64) -.45 (-.32) .43 (.56)

.36 (.44) .75 (.73) .74 (.75) .71 (.58) .65 (.49) .57 (.50) .55 (.57)

.52 (.55)

.80 (.66) .74 (.66) .64 (.74) .53 (.53) .51 (.58)

Numbers in brackets next to the domains refer to the percentage of the total variance explained by each after rotation. Numbers in brackets next to the loadings show the reported loadings in the NEO manual (p. 44). Bold facets require special attention. Loadings below .40are not shown, except for the bold items.

312

Niko Tiliopoulos

Facet-level NMDS Analysis MDS is a family of statistical modelling techniques that attempt to generate a visual representation of the latent structure of the data in a low-dimensional space – these techniques are also known as perceptual mapping, spatial analysis or smallest space analysis (Borg & Groenen, 1997; Cox & Cox, 2001; Coxon, 1982; Coxon, 2004; Kruskal & Wish, 1978). MDS, like factor analysis, seeks to identify the underlying structure (including, but not privileging the dimensional structure) of a set of data. Unlike, however, factor analysis, NMDS has the ability to generate reasonably ―accurate‖ representations of a set of variables in a low dimensionality (e.g., two or three dimensions), as it can represent less stringent nonlinear or monotonic relationships (Shepard, 1962a, 1962b). Furthermore, MDS and FA are not merely mathematical transformations of each other, not least due to the iterative nature of the solution in MDS and the fact that the sampling distribution of Stress is unknown. Therefore, provided that the distances or similarities between the variables are meaningful, the MDS solutions tend to be more parsimonious and interpretable. Kruskal‘s nonmetric monotone regression algorithm was used (Kruskal, 1964a, b), because it does not require any metric assumptions for the data (such as multidimensional normality or linearity), and it tends to produce a clearer picture of the distances between the variables in comparison to a metric alternative, when the fewest possible dimensions are desirable. Nonmetric MDS (NMDS) was performed on the facet level of the NEO-PI-R using the PROXSCAL approach. In order for these results to be comparable to Maraun‘s solution, I followed precisely the same procedure as he did (see Maraun, 1996, p. 637). Thus, distances were calculated from the correlation matrix of the 30 facets, by using the initial Torgerson‘s (1958, pp. 254-259) classic metric MDS configuration with a primary approach to ties. The use of a metric starting configuration for non-metric scaling is strongly recommended if one wishes to avoid sub-optimal/local minimum solutions (see also, Lingoes & Roskam, 1973). A scree plot of the stress values for each dimensionality suggested the retention of two or three dimensions. A 2-dimensional solution was preferred (Stress1 = 0.1714) as the most parsimonious one, which accounted for 37.4% of the variance and maintained 51.8% of the rank ordering of the dissimilarities (solution converged in 12 iterations). Figure 1 presents a scatterplot of the 2D space. A parallel solution to that provided by PROXSCAL was run using a similar, stable, but two-phase algorithm, viz. the Roskam-Lingoes Smallest Space Analysis program (Lingoes & Roskam, 1973) in the NewMDSX series (Coxon, Brier, & Hawkins, 2004). The transformation was again monotonic and the model was Euclidean distances. The data in the latter case were the original product-moment correlations, but since Pearson correlation coefficients and Euclidean distances are (inversely) monotonically identical, this should make no difference to the solution. The resulting two-dimensional solution was almost exactly identical to the PROXSCAL configuration in Figure 1 (except for a trivial reflection in the vertical axes). The Stress1 value was also identical, and was confirmed by the Spence measure in the MINISSA solution as being an excellent fit (the Spence measure is based upon a large number of simulations of random configurations of p points in r dimensions—Spence, 1979). The expected Spence Random Stress1 for 30 points in 2D is 0.2964, whereas my obtained Stress1 is almost twice as small as that, indicating an excellent fit. Furthermore, the Shepard diagram (Fig. 2), giving the co-plot of data (correlations) and solution distances with the best-fitting monotone-fitting quantities (Kruskal‘s dhat values), shows that the rescaling


313

transformation is virtually linear. This indicates that linearity is satisfied, rather than imposing it as an assumption, as is the case in FA. A final analysis of the data using a linear/metric transformation (MRSCAL in NewMDSX) confirmed once again that my recovered solution and facet structure segmentations are both reliable and stable. Since the axes of the solution in a Euclidean distance MDS model are not dimensionally invariant, but are purely convenience co-ordinates and therefore not an integral part of the solution, nor necessarily relevant to interpretation compared to other structural characteristics, I attempted to describe and understand the space through non-dimensional references. The emerging picture suggests that traits tend to occupy unique regions, and by partitioning the space into five conical segments radiating from a common origin, each with a derived angle of multipliers of 40o (although other configurations may still be plausible—cf. Blas & Forzi, 1998; Hofstee, de Raad & Goldberg, 1992), the regional arrangement of the facets becomes even clearer. This arrangement appears almost identical to the one reported by Maraun (1997, see Fig. 3), with the arbitrary mirroring of the y-axis. Deliberation (C6) seems to be isolated from the rest of the C facets and it appears to be closer to A.

Figure 1. A 2-dimensional representation of the NEO-PI-R at a facet-level through NMDS. N, O, A, C, occupy conical regions each of an 80o angle, while for E the angle is 40o. Superimposed is the circle of best fit.

314

Niko Tiliopoulos

Figure 2. Co-plot of data (correlations) & solution distances, with the best monotone-fitting quantities (Kruskal‘s dhat values).

Figure 3. Maraun‘s (1997, p. 638) NMDS 2-dimensional representation of the NEO-PI-R facets.


315

In addition, three facets do not plot in the regions defined by their domains. Excitementseeking (E5) appears in the O region, while both trust (A1) and altruism (A3) fall in the region of C (incidentally, my alternative MDS algorithms, discussed above, also identified these ―misplacements‖). This seems to suggest that perhaps theses facets may be structurally and conceptually closer to those domains. The NMDS solution also suggests that only N appears to form a rather distinct personality domain. The boundaries between the rest do not seem to be so clearly defined. In fact, E, O, A, and C appear to form pairs (E with O, and A with C), which may be addressing common, underlying concepts, and perhaps supporting Eysenck‘s convergence toward the Big-3 argument. In order to further explore this facetgrouping, an agglomerative hierarchical cluster analysis was performed on the NMDS facetcoordinates. Various clustering methods were tested. The unweighted group average linkage with Euclidean distances was preferred, because it produced the highest cophenetic correlation coefficient (rc = .75). The use of a cluster configuration distance of d = 1 as a cutoff point (δ1 = .28) suggested the retention of three clusters. Fig. 4 shows cluster membership, with the retained three clusters appearing in bold. This result gives a clearer picture of the formation of the three conceptually distinct groups by the NEO-PI-R facets that arguably resemble Eysenck‘s PEN configuration. Overall, the NMDS solution(s) thus provides additional support for the FFM structure and the internal consistency of the NEO-PI-R. At the same time, however, it also suggests that two dimensions may be adequate to explain the structure of the personality traits of the model.

Figure 4. Hierarchical structure of the NEO-PI-R facet-level NMDS coordinates. The preferred final cluster solution is marked in bold lines.

316

Niko Tiliopoulos

Circular Regression The NMDS map appears to suggest that traits, besides occupying distinct regions, are also arranged in a circular formation on a 2D plane. To test the validity of this observation, a least square circular regression was performed on the 2-dimensional Cartesian coordinates of the facet-scores (Fisher, 1993; Jammalamadaka & SenGupta, 2001). Circular regression is a special type of non-linear regression through which, given the Cartesian or polar coordinates of a set of data, one seeks to generate a circle such that the sum of the squared distances of the data points from the perimeter of that circle is minimised. In the current model, the centre of the circle of best fit was fixed at X, Y = (0, 0). The analysis converged in two iterations and generated a circle with a radius of r = 0.17 (standard error = 0.009, t = 18.01, p < .00001), and an overall RMSEA of .053 (the circle of best fit is superimposed in Fig. 1). The circular regression results suggest that the facets appear to exhibit a strong circular arrangement. This new evidence allows for a polar categorization of the NEO facets, and consequently the FFM, in respect to their angular displacement and their integrative centrality (i.e., their distance from the centre of the regression circle).

DISCUSSION One outcome of this investigation is quite clear: Using traditional methods of factor analytic procedures, previous results have indeed been replicated and the FFM appears to possess a ―sane‖ structure. Nonetheless, as was pointed out previously, this may give a false impression and lead to unnecessarily complex or even wrong interpretations. The underlying structure of personality may well be explained by more parsimonious models, with as few as two factors or dimensions. In that case, viewed from a different perspective (i.e., through NMDS), the discrepancy between the FFM and Eysenck‘s PEN could disappear. The NMDS analysis certainly suggests this to be so. If the NEO and EPQ can both result in a twocomponent outcome then the more parsimonious answer may well lie with Eysenck‘s taxonomy. In light of this, I suggest that the study of personality makes more use of the wide spectrum of multivariate statistics available, to allow for the development and refinement of both empirically and theoretically sound models. MDS is an excellent multivariate technique for identifying and representing latent structures in a more parsimonious way. A potential limitation of traditional factor analytic techniques is that they pre-commit to a linear and dimensional interpretation, while at the same time they tend to project latent structures onto unnecessarily high dimensionality. MDS, on the other hand, generates structural representations at lower dimensionality, which, coupled with its ability to reveal other nondimensional arrangements, such as clustering, allow for greater interpretability. Therefore, whenever possible, MDS may be used in parallel with other structural or factorial models to assist with the deeper understanding of the structures sought, and to prevent data-analytic artefacts being presented as latent ―realities‖.


317

CONCLUSION Probably the most interesting outcome of this study is that personality traits may fall into a circular continuum. This implies that a simpler structural categorisation of personality may involve the interplay between the ―kind‖ of trait (angular displacement) and the ―intensity‖ or ―fundamentality‖ of trait manifestation (integrative centrality). From my recovered arrangement, it appears that the, arguably, more fundamental dimensions of N and C tend to occupy rather distinct regions further away from the centre, while the rest tend to be drawn toward it. In a post-hoc attempt to try to understand that behaviour, I rotated by 90o the 2D picture of Fig. 1 on the x-axis, and thus, in a sense, observed the xz data-configuration. It seems that the facets that appear to move closer to the xy centre are actually moving ―upwards and inwards‖ on the z-axis, while the N and C ones appear to be, more or less, on the same ―plane‖. Interestingly, the 3D arrangement of the facets appears to be conical. In order to get a preliminary statistical impression of this observation, I devised a rather rough least square conical regression algorithm and ran it on the 3D Cartesian coordinates of the facets. The initial results suggest that a conical shape (where the N and C facets make up the circumference of the base and the rest of the facets fall on the sides and toward the apex) appears to have a near excellent fit (RMSEA = 0.035). However, this result and its potentially fascinating implications are beyond the scope of this chapter, and I would simply like to point it out as a focus for future studies. In any case, I make clear my conviction that a constructive way forward for the discipline of traits personology is through the development of a parsimonious taxonomic convergence and the ultimate unification of the models that attempt to explain the structure of personality.

ACKNOWLEDGMENT I would like to thank my co-authors in the original publication of this study for their invaluable contribution and David White for his help with the development and proofing of the circular and conical algorithms.

REFERENCES Bagby, R. M., Costa, P. T., McCrae, R. R., Livesly, J. W., Kennedy, S. H., Levitan., R. D., Levitt, A. J., Joffe, R. T., & Young, L. T. (1999). Replicating the five factor model of personality in a psychiatric sample. Personality and Individual Differences, 17, 11351139. Blas, L. D. & Forxi, M. (1998). The circumplex model for interpersonal trait adjectives in Italian. Personality and Individual Differences, 24, 47-57. Block, J., (1995). A contrarian view of the five factor approach. Psychological Bulletin, 117, 187-215. Boyle, G. J., Matthews, G., & Saklofske, D. H. (Eds.) (2008). Personality theory and assessment. London: Sage.

318

Niko Tiliopoulos

Caruso, J. C., & Cliff, N. (1997). An examination of the five-factor model of normal personality variation with reliable component analysis. Personality and Individual Differences, 23, 317-325. Costa, P. T., & McCrae, R. R. (1992). Revised NEO Personality Inventory and NEO Five Factor Inventory professional manual. Odessa, FL: Psychological Assessment Resources. Costa, P. T., & McCrae, R. R. (1988). Personality in adulthood: A six year longitudinal study of self-reports and spouse ratings on the NEO Personality Inventory. Journal of Personality and Social Psychology, 54, 853-863. Costa, P. T., McCrae, R. R., & Dye, D. A. (1991). Facet scales for agreeableness and conscientiousness: A revision of the NEO Personality Inventory. Personality and Individual Differences, 12, 887-898. Borg, I., & Groenen, P. (1997). Modern multidimensional scaling theory and applications. New York: Springer. Cox, T. F., & Cox, A. A. (2001). Multidimensional scaling (2nd ed.). London: Chapman & Hall. Coxon, A. P. M. (1982). The user’s guide to multidimensional scaling. London: Heinemann Educational Books. Coxon, A. P. M. (2004). Multidimensional scaling. In M. S. Lewis-Beck, A. Bryman, & T. F. Liao (Eds.). The Sage encyclopaedia of social science research methods (Vol. 2, pp. 669672). London: Sage. Coxon, A. P. M., Brier, A., & Hawkins, P. K. (2004, revised ed.). The users’ manual: NewMDSX. Edinburgh: NewMDSX Project. Digman, J. (1990). Personality structure: Emergence of the five-factor model. In M. Rosenzweig & L. Potter (Eds.) Annual Review of Psychology (Vol. 41, pp. 417-440). Palo Alto, CA: Annual Reviews. Draycott, S. G., & Kline, P. (1991). The big three or the big five – the EPQ-R vs. the NEOPI: A research note, replication and elaboration. Personality and Individual Differences, 18, 801-804. Egan, V., Deary, I., & Austin, E. (2000). The NEO-FFI: emerging British norms and an item level analysis suggest N, A and C are more reliable that O and E. Personality and Individual Differences, 29, 907-920. Eysenck, H. (1992). Four ways 5 factors are not basic. Personality and Individual Differences,13, 667-673. Fiske, D. (1949). Consistency of the factorial structures of personality ratings from different sources. Journal of Abnormal and Social Psychology, 44, 329-344. Fisher, N.I. (1993). Statistical analysis of circular data. Cambridge: Cambridge University Press. Goldberg, L. R. (1981). Language and individual differences: The search for universals in personality lexicons. In L. Wheeler (Ed.), Review of Personality and Social Psychology (pp. 141-165). London: Sage. Goldberg, L. R. (1993). The structure of phenotypic personality traits. American Psychologist, 48, 26-34. Guilford, J. P. (1975). Factors and factors of personality. Psychological Bulletin, 82, 802-814. Hammond, S. M. (1987). The item structure of the Eysenck Personality Questionnaire across method and culture. Personality and Individual Differences, 8, 541-549.


319

Hofstee, W. K. B., de Raad, B. & Goldberg, L. R. (1992). Integration of the Big Five and circumplex approaches to trait structure. Journal of Personality and Social Psychology, 8, 146-163. Jackson, D. N., Ashton, M. C., & Tomes, J. L (1996). The six factor model of personality: Facets from the big five. Personality and Individual Differences, 21, 391-402. Jammalamadaka, R. S. & SenGupta, A. (2001). Topics in circular statistics. London: World Scientific Publications. John, O. P., Robin, R. W., & Pervin, L. A. (Eds.) (2008). Handbook of personality (3rd ed.). New York: Guilford. Kruskal, J. (1964a). Multidimensional scaling by optimizing goodness of fit to a nonmetric hypothesis. Psychometrika, 29, 1-27. Kruskal, J. (1964b). Nonmetric multidimensonal scaling: A numerical method. Psychometrika, 29, 115-129. Kruskal, J., & Wish, M. (1978). Multidimensional scaling. London: Sage. Lingoes, J. C. & Roskam, E. E. (1973). A mathematical and empirical comparison of two multidimensional scaling algorithms. Psychometrika, 38 (monograph supplement 19). Maraun, M. D. (1997). Appearance and reality: Is the big five the structure of trait descriptors. Personality and Individual Differences, 22, 629-647. Mathews, G., Deary, I. J., & Whiteman, M. C. (2003). Personality traits (2nd ed.). Cambridge: Cambridge University Press. Norman, W. T. (1963). Toward an adequate taxonomy of personality attributes: Replicated factor structure in peer nomination personality ratings. Journal of Abnormal and Social Psychology, 66, 574-583. Parker, J. D. A., Bagby, R. M., & Summerfield, L. J. (1992). Confirmatory factor analysis of the revised NEO Personality Inventory. Personality and Individual Differences, 15, 463466. Shepard, R. N. (1962a). The analysis of proximities: multidimensional scaling with an unknown distance function I. Psychometrika, 27, 125-140. Shepard, R. N. (1962b). The analysis of proximities: multidimensional scaling with an unknown distance function II. Psychometrika, 27, 219-246. Spence, I. (1979). A simple approximation for random ranking STRESS values. Multivariate Behavioral Research, 14, 355-65. Torgerson, W. S. (1958). Theory and methods of scaling. New York: Wiley. Tupes, E. & Christal, R. (1961). Recurrent personality factors based on trait ratings. USAF WADS Technical Note No. 58-61.



Chapter 27

THE ROLE OF CONCEPTUAL ANALYSIS IN PERSONALITY RESEARCH Simon Boag Department of Psychology, Macquarie University, Sydney, NSW 2109, Australia

ABSTRACT The proliferation of personality and individual differences constructs is testament to the vibrancy of research in this field. Furthermore, all such constructs would make some claim to receiving empirical support. However, while additional empirical research will help further determine the merit of these conceptions, this paper proposes that conceptual analysis is another tool available to researchers to help critically evaluate our constructs of interest. Conceptual analysis involves assessing both the clarity and coherency of our constructs as well as examining the relations between them. A useful direction here is scrutinising constructs in terms of intrinsic properties (what something is) and relations (what something does) to help both clarify and evaluate the coherency of these constructs, as well as build bridges between apparently disparate concepts. Given the perceived need for unification in psychology, conceptual analysis provides yet another tool for assessing the relationship of constructs to one another and contributing to a coherent account of ‗persons‘. While conceptual analysis may appear to be too ‗philosophical‘ to some, it is none other than our critical thinking skills in action. Implications for personality psychology are discussed. Keywords: Circular reasoning, Conceptual analysis, Personality, Reification, Relations




322

Simon Boag

INTRODUCTION As early as 1930 the field of personality research was expanding to encompass a wide variety of definitions of personality such that Allport and Vernon (1930) write, ―there seem to be as virtually as many definitions of personality, character and temperament as there are writers on these subjects‖ (p. 681). These authors, however, go on to note that while ―at first glance the problem resembles chaos… the principal disagreements concern the point in a definition where the emphasis falls‖ (p. 681). That is, the various definitions of personality are not mutually exclusive and the differing approaches embrace a number of common personality facets (to varying degrees). Since the 1930s, the number of constructs now found in personality research is even greater, and the International Personality Item Pool (http://ipip.ori.org/ipip/) lists some 200 constructs and scales for assessing various personality components. Given that human beings are complex then it might come as no surprise that such a variety of constructs are available, since these personality facets simply reflect various aspects of persons. Recently, however, some authors have raised concerns about the fragmentary nature of psychology generally and the need for a ‗unified psychology‘ (e.g., Sternberg & Grigorenko, 2001). One specific problem noted involves theoretical and conceptual redundancy, whereby a wide variety of constructs used by different schools appear to be describing the same (or similar) phenomenon. Concurrently, some propose that personality psychology should act as an integrating perspective that reconciles the various sub-disciplines within psychology (e.g., Baumeister & Tice, 1996; Funder, 2001; Mayer, 2005). Here personality can be viewed as the study of the ‗whole person‘ and Funder (2001) writes that the ‗institutional mission‘ of personality psychology ―is to provide an integrative force in an era of scientific specialisation and fragmentation. Personality psychology seeks to bring together the contributions of developmental, social, cognitive and biological psychology into an understanding of whole persons and the dimensions of difference that allow them to be psychologically distinguished from one another‖ (p. 198). However, others have noted that the field of personality itself—as it presently stands— provides a conflicting or confusing array of personality facets that are not easily consolidated into a coherent account of persons (e.g., Baumeister & Tice, 1996; Mayer, 2005). While some have attempted such a synthesis (e.g., Mayer, 2005; McAdams, 1996, 2001), it should be recognised that any such synthesis involves our conceptualisation of persons and thus necessarily involves analysing both the concepts involved and their interrelationships. For instance, Mayer (2005) suggests that personality elements must be relatively distinct from one another and map either onto nervous system or ‗social functions‘ (much like the original tripartite division into motivation, emotion and cognition). However, for this to occur first requires clearly conceptualising what these elements precisely are and then determining their relation to other elements. Consequently some means is required with respect to delineating any particular element or component part. The aim of this chapter is to discuss the role of conceptual analysis in scientific research and specifically in relation to the field of personality. This chapter first discusses the science of conceptual analysis and its relation to empirical research. The nature of conceptual research is then described with particular reference to the distinction between relations and terms standing within relations—and identifying problems that occur when this distinction is not satisfactorily appreciated.

The Role of Conceptual Analysis in Personality Research

323

THE SCIENCE OF CONCEPTUAL RESEARCH While we would all probably agree that empirical research is an essential character of science, the appreciation of the importance of conceptual research in the scientific process is not always explicitly recognised. As Machado and Silva (2007) however write, ―[s]cience has always included the screening of concepts and arguments for clarity and coherence‖ (p. 680), and it is difficult to imagine that as scientists we would shy away from wishing to critically evaluate the concepts we employ in our research. Conceptual and theoretical work involves extricating the logically sound features of theories and rejecting the incoherent aspects, before attempting to execute empirical tests (Michell, 2000). Conceptual research also contributes to the interpretation of empirical findings (Wakefield, 2007), as well as shaping meaningful and valid avenues for future research directions. Consequently, rigorous theoretical and conceptual work provides the underlying backbone of the whole scientific endeavour (Bell, Staines & Michell, 2000; Machado & Silva, 2007). On the other hand, when rigorous critical thinking is neglected then confusion results, particularly with respect to understanding the findings from various empirical fields. For instance, the neuroscientific dream debate has become confused with respect to interpreting the evidence for and against Freud‘s theory of dreams because basic theoretical issues have not been adequately addressed and so leading to contradictory claims (Boag, 2006a, 2006b). Similarly, the empirical investigation of Freudian repression has lead to seemingly paradoxical findings since researchers have failed to sufficiently articulate the theory in question (Boag, 2006c). Furthermore, the subliminal perception literature is also mired in controversy since the basic premises of the research field have not been coherently conceptualised (Boag, 2008a). Accordingly, critical theoretical discussion is not without practical implications, since it both highlights valid avenues of empirical research, as well as saving both time and resources that are otherwise wasted when researchers are ignorant of the theory that is purportedly being tested. Accordingly, science, as an intellectual enterprise, cannot neglect the critical examination of conceptual issues. Of course, conceptual research will sound too philosophical for some and may appear to threaten the emancipation of psychology from its philosophical past. However, given that as scientists we both develop, use, and test constructs, then carefully scrutinizing what we are talking about cannot be ignored. In fact, any scientist conducts conceptual research, more or less implicitly or explicitly, since (for example) formulating scales and tests must be guided by some conceptualisation of what one is looking for. Furthermore, we all attempt to publish our research and communicate our ideas, and so conceptual clarity from conceptual research is a basic element necessary for clear communication (cf. Wakefield, 2007).

CONDUCTING CONCEPTUAL RESEARCH There are many possible approaches to conceptual research (see, for instance, Moors & De Houwer, 2006; Wakefield, 2007), but the approach discussed here has a basis in what can be called a realist position (postulating a real world that we can objectively come to know) and a specifically useful foundation here is the determinist, empiricist position of Scottishborn John Anderson (1962a, 1962b, 1962c, 1962d), and developed by Boag (2005, 2008b),

324

Simon Boag

Hibberd (2005), Mackay (1996), Maze (1983, 1987), Medlow (2008), Michell (1988), and Petocz (1999). This hyper-critical theoretical framework provides a cogent position for rigorously examining the coherency of theories and concepts, in part due to it clarifying the important, but often poorly appreciated, distinction between qualities (or properties), entities (or things) and relations. As Mackie (1962) writes, ―a quality is an intrinsic feature of a thing, it belongs to the thing itself, whereas a relation holds between two or more things‖ (p, 266). More specifically, relations involve at least two or more distinct terms that must have their own intrinsic properties (to constitute what stands in the relation): Anything that can stand …. in any relation at all, must have at least some intrinsic properties. If that were not the case … then we could not understand what it was that was said to have those relationships. A relation can only hold between two or more terms, and a part of what is involved in seeing those terms as related is being able to see them as distinct, that is, as each having its own intrinsic properties, so that we can say what the terms are that are related. This means that each term of the relation must be able in principle to be described without the need to include any reference to its relation to the other (Maze, 1983, p. 24, his italics; cf. Maze, 1954, p. 231; Michell, 1988, p. 234).

By ‗term‘ is not meant some linguistic entity but rather whatever is said to stand in relation to something else, and any relation (e.g., spatial, logical, etc.) can only exist if there are independently existing terms. For instance, if I say that John is different from Mike, then both John and Mike must exist (i.e., have properties) to make such a comparison. Relations, however, should not to be taken as things or properties themselves since a relationship cannot be reduced to either of the terms of the relationship (i.e., relations refer to how the various things stand with respect to one another and are not reducible to either). For example, to say that John is different from Mike is to note a relation between John and Mike and not a property of either (i.e., ‗difference‘ cannot be reduced to either John or Mike). However, again, this is not to say that that John and Mike do not have characteristics that they differ on; only that the relation is between those properties rather than being a property itself. Additionally, there is no suggestion here that relations thereby have a second-rate existence compared to the terms standing in relation or are any less important in terms of the subject matter of scientific research. On the contrary, relations are real features of situations (existing in the same spatio-temporal universe as the things standing within relationships), and so are just as open to scientific enquiry as anything else. In terms of conceptual analysis, then, any personality construct needs to be examined with respect to whether the term refers to a property or quality of something, or to an entity of some description (that consists of its own intrinsic properties), or whether the term refers to specific relations between qualities or entities. Another way of thinking about this is in terms of what something is and what it does. If a quality or property exists then it should have properties that can be identified independently of any relation entered into, which would mean describing the property independently of any particular performance or activity (Maze, 1983). This allows us to examine, clarify, and group our constructs that refer to activities (i.e., what things do rather than what they are). Consider, for instance, constructs such as ‗impulse-control‘, ‗risk-taking‘, ‗sociability‘, ‗adaptability‘, ‗cautiousness‘ and ‗conformity‘. Such constructs suggest activities—controlling impulses, taking risks, adapting to circumstances, being cautious, or conforming to the group—and


325

specifically as relations (e.g., something controls—call this the subject term (S)—such that S controls x, or S conforms to y). Recognising these as relations then allows asking questions concerning exactly what is controlling, taking risks, or conforming, etc. as well as to what the causal antecedents are that lead to these activities. Similarly, ‗identity‘ appears to involve something identifying with something else (such that a subject S identifies with x), and appreciating this relation of identifying also then allows asking questions about S, x, and the specific nature of the relation involved (see, also Boag, in press, with respect to traits as relations). Additionally, the wide variety of ‗self‘-constructs found within the personality literature—self-acceptance, self-monitoring, self-confidence, self-concept, self-esteem, selfefficacy or self-control—can all be analysed in terms of whether these are activities and ask the further question about what precisely the ‗self‘ may be in all such cases (see Boag, 2005 for further discussion here).

RELATIONS AND THE ERROR OF REIFICATION The critical importance of recognising the distinction between relations and terms standing within relations (or in terms of what something is and what something does) allows avoiding confusing the two. However, when this distinction is not fully appreciated then confusing the two can lead to the problem of reification—often understood as confusing the abstract with the concrete—but more coherently thought of in terms of mistaking a relation with either an entity or an intrinsic property (Passmore, 1935). Specifically, reification occurs when relationships between things are mistakenly taken to be either properties (or qualities) of things standing within relationships, or taken to be the entities themselves. For instance, to say that John is more ‗sociable‘ than Mike is to note a relationship between John and Mike and, as such, ‗sociability‘ cannot be reduced solely to a property of either. However, to mistakenly consider this difference between John and Mike to be an intrinsic feature of either person would be to reify the relationship (i.e., to mistake a relationship for a property). Unfortunately, any personality term that is based on either descriptions of behaviour (e.g., extraversion, psychoticism) or comparison between individuals may be easily reified, just as the Five-Factor Model has recently been criticised for confusing between-subject with withinsubject variables (Borsboom, Mellenbergh & van Heerden, 2003; Cervone, 1999, 2005; see also Beckmann, Minbashian & Wood, this volume).

DESCRIPTION AND EXPLANATION IN PERSONALITY RESEARCH Another area where conceptual analysis provides clarification is with respect to description, causality and explanation. Description typically involves identification and classification, whereas explanation involves providing an account of origins, development, and causal relationships. For instance, in biology, biologists propose classification systems and taxonomies, and then attempt to explain differences and changes through various mechanisms and processes (see Bechtel, 2005). The same generally holds for personality psychology, which involves both describing personality and providing explanations for both various developments of personality and how personality (however defined) may influence

326

Simon Boag

other aspects of the person. There is, of course, nothing wrong with description (in and of itself), and not knowing the causes of certain relationships does not preclude using description fruitfully for predictive utility. There are, however, problems with confusing description with explanation, particularly with respect to circular explanations. The issue of circular explanation involves saying A causes B when A is equivalent with B. For instance, if we observe someone being sociable (an activity), then name that ‗extraversion‘, and then claim that ‗extraversion‘ causes that sociable behaviour, then we have provided a circular explanation (A causes B when A = B). ‗Trait‘ accounts can easily suffer from circularity when invoked as causes of behaviour and cognition (see Boag, in press) since if traits are behaviourally defined then we would be using behaviour to explain behaviour (cf. Bandura, 1999, p. 203). While one may here retort that prior constellations of behaviour are distinct from present constellations we are though still talking about correlations between past and present behaviour rather than actual causes. None of this is to say, however, that there may not be legitimate trait-explanatory accounts and Jonathon Gerber (this volume) has gone some way to providing solutions to using traits as explanations. Rescuing trait-explanatory accounts, however, requires more than determining distinct causes of trait-behaviour (such as biological factors, etc.) and, instead, what needs to be made explicit is whether certain biological factors (etc.) are the traits in question. Of course, no one is suggesting that various causal factors (e.g., biological processes) do not underlie such trait-behavioural consistencies (see, for instance, Eysenck, 1991, 1997), but if the trait is defined in terms of the behaviour to be explained, then an issue of circularity cannot be avoided. A similar issue of circularity can be raised against supposed functional ‗modules‘ discussed in evolutionary psychology (e.g., Buss & Greiling, 1999; Tooby & Cosmides, 1990), which are generally discussed in terms of activities rather than properties. While not at all doubting the reality of the basic principles of Darwinian evolution, nor claiming that such activities are without causes (or that genetics and brain functioning are not important here), such modules may easily be seen as post- or ad-hoc reifications inferred simply from performance (see also Cervone, 2000). Postulating modules based on observed activities is then not indistinct from the same reasoning that invokes a ‗moral faculty‘ to explain moral reasoning (cf. Passmore, 1935; Boag, in press), or what Maze (1954) refers to as ‗verbal magic‘—―giving a name to a certain kind of event and then using that name as if it accounted for the occurrence of that kind of event‖ (p. 226; cf. McMullen, 1982). Such explanatory tactics only provide pseudo-explanations and add nothing to our understanding of the actual causal factors involved.

AVOIDING CIRCULAR EXPLANATIONS Avoiding circularity is relatively straight-forward, insofar as when we say that A causes B, then both A and B should be distinct from one another (i.e., describable independently of one another) and conceptual research can help here with clarifying testable from untestable hypotheses. Consider the following statement: high mathematical ability causes good mathematics performance. How would one test such claims? In the first case, one would have to be clear about what it meant by ‗high mathematical ability‘. If ‗ability‘ here simply means


327

what one is able to do then the claim is circular (John performs well because he is able to perform well) and it would be a waste of time attempting to research whether the hypothesis is accurate or not (see also Howe, 1990). At the same time, if it was determined that ‗ability‘ in this instance meant particular brain processes then the circularity would be subverted since the postulated cause can be described independently of performance. It is, in fact, most likely, that when the term ‗ability‘ is used it is probably meant as a short-hand for the underlying hidden cause (or causes) of performance and one might possibly then argue that the term is itself therefore harmless. However, by postulating an ‗ability‘ to explain performance suggests a tendency to propose a single cause. It should, though, be borne in mind that there may be multiple causes for any effect (not to mention taking into account the causal field within which cause and effect operate—see Anderson. 1962c, 1962d), or that two sets of distinct causes may lead to the same effect. Consider, for instance, any test performance whereby a variety of factors may be causally implicated (such as personality variables, motivation, fatigue, emotional state, etc.—Revelle, Humphreys, Simon, & Gilliland, 1980; Richardson, 2002). Consequently, if personality psychology is to make substantive explanatory claims it must carefully conceptualise its variables, avoid over-simplifying complexity, and demonstrate that circular reasoning is being avoided. Such conceptual work can only bolster the field of personality‘s claim to being a rigorous, disciplined scientific pursuit.

CONCLUSION In attempting to objectively understand personality, and communicate this understanding to others, we cannot help engage in conceptual research. Since science and precision go hand in hand, precisely understanding what we are talking about (or at least working towards it) should be given the highest priority in scientific research, including our investigations of personality. Conceptual analysis is one tool here for helping achieve this aim. Appreciating the distinction between relations and terms standing in relations provides both an avenue for conceptualising our constructs coherently and avoiding problems of reification and circularity. While some researchers may fear (legitimately!) a tendency to get lost in philosophical word-games, we nevertheless communicate using language, and the meanings that we ascribe to our terms are the lasting legacy of our empirical endeavours. Consequently, given that it is so easy for us to fall into error, there is no excuse for not making use of all the tools available at our disposal for achieving our scientific aims. Personality psychology can only benefit from providing greater focus to such conceptual issues.

REFERENCES Allport, G. W. & Vernon, P. E. (1930). The field of personality. Psychological Bulletin, 27, 677-730. Anderson, J. (1962a). The knower and the known. In Studies in empirical philosophy (pp. 2740). Sydney: Angus & Robertson. (Original work published 1927).

328

Simon Boag

Anderson, J. (1962b). Realism and some of its critics. In Studies in empirical philosophy (pp. 41-59). Sydney: Angus & Robertson. (Original work published 1930). Anderson, J. (1962c). Causality and logic. In Studies in empirical philosophy (pp. 122-125). Sydney: Angus & Robertson. (Original work published 1936). Anderson, J. (1962d). The problem of causality. In Studies in empirical philosophy (pp. 126136). Sydney: Angus & Robertson. (Original work published 1938). Bandura, A. (1999). Social cognitive theory of personality. In D. Cervone & Yuichi Shoda (Eds.), The coherence of personality: Social-cognitive bases of consistency, variability, and organization (pp. 185-241). New York: Guilford Press. Baumeister, R. F. & Tice, D. M. (1996). Rethinking and reclaiming the interdisciplinary role of personality psychology: The science of human nature should be the center of the social sciences and humanities. Journal of Research in Personality, 30, 363-373. Bechtel, W. (2005). The challenge of characterising operations in the mechanisms underlying behavior. Journal of the Experimental Analysis of Behavior, 84, 313-325. Bell, P., Staines, P. & Michell, J. (2000). Logical psych. Sydney: UNSW Press. Boag, S. (2005). Addressing mental plurality: Justification, objections & logical requirements of strongly partitive accounts of mind. Theory and Psychology, 15, 747-767. Boag, S. (2006a). Freudian dream theory, dream bizarreness, and the disguise-censor controversy. Neuro-psychoanalysis, 8, 5-17. Boag, S. (2006b). ―Freudian dream theory, dream bizarreness, and the disguise-censor controversy‖: Response to Commentaries. Neuro-psychoanalysis, 8, 60-69. Boag, S. (2006c). Freudian repression, the common view, and pathological science. Review of General Psychology, 10, 74-86. Boag, S. (2007). ‗Real processes‘ and the explanatory status of repression and inhibition. Philosophical Psychology, 20, 375-392. Boag, S. (2008a). Making sense of subliminal perception. In A. M. Columbus (Ed.), Advances in Psychology Research (pp. 117-139). New York: Nova Science Publishers. Boag, S. (2008b). ‗Mind as feeling‘ or affective relations? A contribution to the School of Andersonian Realism. Theory and Psychology, 18, 505-525. Boag, S. (in press). Explanation in personality psychology: ‗Verbal magic‘ and the FiveFactor Model. Philosophical Psychology. Borsboom, D., Mellenbergh, G. J., & van Heerden, J. (2003). The theoretical status of latent variables. Psychological Review, 110, 203-219. Buss, D. M. & Greiling, H. (1999). Adaptive individual differences. Journal of Personality, 67, 209-243. Cervone, D. (1999). Bottom-up explanation in personality psychology: The case of crosssituational coherence. In D. Cervone & Yuichi Shoda (Eds.), The coherence of personality: Social-cognitive bases of consistency, variability, and organization (pp. 303341). New York: Guilford Press. Cervone, D. (2005). Personality architecture: Within-person structures and processes. Annual Review of Psychology, 56, 423-452. Eysenck, H. J. (1991). Dimensions of personality: 16, 5 or 3?—Criteria for a taxonomic paradigm. Personality and Individual Differences, 12, 773-790. Eysenck, H. J. (1997). Personality and experimental psychology: The unification of psychology and the possibility of a paradigm. Journal of Personality and Social Psychology, 73, 1224-1237.


329

Funder, D. C. (2001). Personality. Annual Review of Psychology, 52, 197-221. Hibberd, F. J. (2005). Unfolding social constructionism. New York, US: Springer Science. Howe, M. J. A. (1990). Does intelligence exist? The Psychologist, 3, 490-493. Leahey, T. H. (1997). A history of psychology: Main currents in psychological thought. Upper Saddle River: Prentice Hall. Machado, A. & Silva, F. J. (2007). Toward a richer view of the scientific method: The role of conceptual analysis. American Psychologist, 62, 671-681. Machamer, P., Darden, L. & Craver, C. F. (2000). Thinking about mechanisms. Philosophy of Science, 67, 1-25. Mackay, N. (1996). The place of motivation in psychoanalysis. Modern Psychoanalysis, 21, 3-17. Mackie, J. L. (1962). The philosophy of John Anderson. Australasian Journal of Philosophy, 40, 265-282. Mayer, J. D. (2005). A tale of two visions: Can a new view of personality help integrate psychology? American Psychologist, 60, 294-307. Maze, J. R. (1954). Do intervening variables intervene? Psychological Review, 61, 226-234. Maze, J. R. (1983). The Meaning of Behaviour. London: Allen & Unwin. Maze, J. R. (1987). The composition of the ego in a deterministic psychology. In W. J. Baker, M. E. Hyland, H. Van Rappard & A. W. Staats (Eds.), Current Issues in Theoretical Psychology (pp. 189-199). North Holland: Elsevier Science Publishers. McAdams, D. P. (1996). Alternative futures for the study of human individuality. Journal of Research in Personality, 30, 374-388. McAdams, D. P. (2001). The psychology of life stories. Review of General Psychology, 5, 100-122. McCrae, R. R. & Costa, P. T. Jr. (1995). Trait explanations in personality psychology. European Journal of Personality, 9, 231-252. McMullen, T. (1982). A critique of humanistic psychology. Australian Journal of Psychology, 34, 221-229. Medlow, S. (2008). The ‘paradox’ of mental causation: Solutions from anomalous monism and direct realism. Saarbrücken: VDM Verlag. Michell, J. (1988). Maze‘s direct realism and the character of cognition. Australian Journal of Psychology, 40, 227-249. Michell, J. (2000). Normal science, pathological science and psychometrics. Theory and Psychology, 10, 639-667. Moors, A. & de Houwer, J. (2006). Automaticity: A theoretical and conceptual analysis. Psychological Bulletin, 132, 297-326. Passmore, J. A. (1935). The nature of intelligence. Australasian Journal of Psychology and Philosophy, 13, 279-289. Petocz, A. (1999). Freud, psychoanalysis, and symbolism. Cambridge: Cambridge University Press. Richardson, K. (2002). What IQ tests test. Theory and Psychology, 12, 283-314. Revelle, W., Humphreys, M. S., Simon, L., & Gilliland, K. (1980). The interactive effect of personality, time of day, and caffeine: A test of the arousal model. Journal of Experimental Psychology: General, 109, 1-31. Sternberg, R. J. & Grigorenko, E. L. (2001). Unified psychology. American Psychologist, 56, 1069-1079.

330

Simon Boag

Tooby, J. & Cosmides, L. (1990). On the universality of human nature and the uniqueness of the individual: the role of genetics and adaptation. Journal of Personality, 58, 17-67. Wakefield, J. C. (2007). Why psychology needs conceptual analysts: Wachtel‘s ‗discontents‘ revisited. Applied and Preventative Psychology, 12, 39-43.

INDEX A abuse, 125, 196, 210, 232, 240, 247, 250, 261, 268, 269, 270, 272 academic performance, 102, 112, 292 academic settings, 100 accessibility, 2 accounting, 69, 70, 75, 77, 79, 98, 157, 266, 271 accuracy, 72, 105, 106, 107, 109, 110, 112, 269, 288, 289, 291 achievement test, 105, 106, 107, 110, 135 acid, 194, 198, 199 acquaintance, 68 activity level, 98, 198 adaptability, 13, 324 adaptation, 4, 120, 124, 220, 330 ADHD, 196, 197, 207, 208 adjustment, 80, 152, 278 administrators, 279 adolescents, 80, 114, 196, 207, 208, 209, 210, 215, 216, 217, 219, 220, 221, 222, 305 adulthood, 123, 159, 196, 210, 318 advantages, 6, 100, 258, 283, 284, 285 advocacy, 133 aetiology, 4, 231, 233, 241 affective experience, 152 affective reactions, 19 Africa, 184, 267, 273 agencies, 137 aggregation, 211 aggression, 68, 79, 194, 195, 197 aggressive behavior, 82 agoraphobia, 241, 246 alcohol abuse, 210, 232 alcohol consumption, 260 alcohol problems, 251 alcohol use, 259, 260, 261 alcoholism, 196, 247, 250, 262

alexithymia, 149 algorithm, 207, 312, 317 allele, 198, 200, 204, 205, 208 altruism, 310, 315 ambivalence, 228, 239, 241, 244, 245 American Educational Research Association, 64 American Psychiatric Association, 3, 4, 5, 7, 214, 220, 231, 233, 270, 272 American Psychological Association, 82, 115, 148, 149, 163, 179, 234, 235, 246, 247, 260, 268, 273, 281 anger, 196, 211, 216 anorexia, 225, 227, 232, 233 anorexia nervosa, 233 ANOVA, 155, 227, 243 antecedent variables, 97 anthropologists, 183 antisocial behavior, 194, 261, 272 antisocial personality, 196, 263 antisocial personality disorder, 196, 263 anxiety, 64, 80, 96, 97, 98, 99, 101, 123, 135, 140, 152, 154, 155, 157, 158, 209, 210, 211, 214, 215, 216, 217, 219, 220, 221, 222, 237, 238, 239, 240, 241, 244, 245, 246, 247, 248, 254, 256, 258, 259, 266, 279, 281, 305 anxiety disorder, 123, 214, 215, 220, 221, 222, 237, 238, 240, 241, 244, 245, 246, 247, 248, 305 anxious mood, 210 appetite, 210, 211, 215 aptitude, 27, 32, 107 architecture, 19, 328 Aristotle, 187 arousal, 67, 68, 69, 70, 71, 72, 74, 75, 76, 77, 78, 79, 80, 81, 82, 211, 216, 220, 226, 254, 264, 329 Asia, 189 Asian countries, 135 assertiveness, 142, 310

332

Index

assessment, iv, 1, 4, 5, 6, 21, 32, 34, 41, 56, 71, 83, 108, 109, 112, 113, 139, 140, 143, 148, 189, 210, 211, 215, 219, 220, 221, 222, 227, 241, 259, 260, 261, 263, 273, 275, 291, 304, 317 assessment tools, 6, 210 assimilation, 48, 51, 52, 53, 54, 61, 62, 64 asymmetry, 289, 290, 293 asymptomatic, 224 attachment, 3, 6, 129, 151, 152, 153, 154, 156, 157, 158, 159, 160, 161, 230, 305 attachment theory, 6, 152, 160 Attention Deficit Hyperactivity Disorder, 207, 208 attentional bias, 304 attribution, 80, 157, 238 attribution bias, 157 authoritarianism, 136, 139 authorities, 266 automatic processes, 24 automaticity, 22 autonomy, 13 avoidance, 152, 154, 155, 157, 158, 178, 211, 221, 224, 225, 234, 241, 249, 252, 255, 256, 257, 258, 259, 260, 262, 263

B background noise, 285 barriers, 2 base rate, 269 behaviors, 101, 272 belief systems, 3, 139 benign, 132 bias, 2, 6, 48, 56, 108, 113, 164, 170, 240, 249, 285, 289, 290, 304 Big Five traits, 96 biological markers, 128 biological processes, 326 biological systems, 300 body shape, 224 bombing, 132 bonds, 114, 152, 194 boredom, 230, 277 brain, 6, 28, 29, 31, 34, 128, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 185, 194, 195, 197, 198, 290, 293, 299, 300, 304, 326, 327 brain activity, 172, 299, 300 brain damage, 293 brain functioning, 326 breakdown, 152 breathing, 124, 217, 241 bulimia, 6, 223, 224, 226, 227, 230, 232, 233, 234, 235 bulimia nervosa, 233, 234, 235 bullying, 270

burnout, 47, 281 business environment, 266 buttons, 289

C calibration, 104, 114 campaigns, 136 cancer, 137, 238 candidates, 136, 177, 197, 271 carefulness, 94 caregivers, 224, 301 casting, 269 categorization, 65, 316 category a, 15 catharsis, 124 causal antecedent, 325 causal relationship, 325 causality, 325, 328 causation, 329 central executive, 29 cerebral hemisphere, 290 CFI, 107 childhood, 5, 111, 114, 123, 127, 160, 194, 215, 220, 224 Chile, 134 chimera, 31, 290 China, 133, 134, 136, 171, 172 cholesterol, 135 Christianity, 151 Christians, 153, 157, 158, 159, 161 chromosome, 194, 195, 197 churches, 141, 145 citizenship, 15, 16, 19, 273 clarity, 17, 82, 99, 149, 178, 212, 213, 218, 321, 323 class, 97, 112, 164, 174, 195, 197 classical logic, 164 clients, 240, 245, 248 clinical assessment, 3 clinical diagnosis, 123 clinical disorders, 214, 238, 247 clinical psychology, 5, 6, 240 close relationships, 160, 299 cluster analysis, 315 clustering, 31, 315, 316 cognition, 12, 28, 30, 33, 47, 52, 104, 111, 115, 153, 157, 183, 184, 185, 186, 188, 189, 197, 260, 293, 305, 322, 326, 329 cognitive abilities, 27, 29, 31, 32, 33, 69, 86, 91, 105 cognitive ability, 33, 87, 88, 89, 91, 107 cognitive capacities, 142 cognitive deficit, 184 cognitive deficits, 184 cognitive domains, 104, 108

Index cognitive function, 21, 26, 30, 251 cognitive models, 153 cognitive performance, 25, 30 cognitive process, 25, 26, 27 cognitive psychology, 87 cognitive tasks, 108, 111 cognitive theory, 63, 328 cognitive variables, 26 coherence, 183, 311, 323, 328 collectivism, 138, 181, 235 college students, 233, 261, 263 community, 95, 126, 158, 159, 186, 187, 188, 220, 221, 222, 233 community support, 159 comorbidity, 4, 210 competitive advantage, 97 competitiveness, 133 complexity, 27, 29, 33, 80, 92, 94, 182, 206, 266, 327 compliance, 246, 274 comprehension, 104, 216, 217 computer technology, 13 conceptualization, 84, 255, 258 conditioning, 301 conduct disorder, 6, 194, 196, 205, 206, 207, 208 conductance, 260 conference, 132 configuration, 284, 312, 315, 317 configurations, 312, 313 confinement, 194 conflict, 80, 123, 177, 249, 254, 255, 256, 257, 258, 261, 262, 277, 299 conformity, 268, 324 confounding variables, 72 conscientiousness, 12, 13, 15, 17, 18, 19, 30, 95, 171, 177, 268, 273, 318 conscious awareness, 225 consciousness, 125, 311 consensus, 23, 27, 28, 30, 94 construct validity, 28, 34, 101, 143, 216, 235, 256, 264, 273 consumption, 247, 260 contamination, 214, 257 content analysis, 272 control condition, 55, 56, 57, 59, 62, 72, 75, 76, 77, 78, 232 control group, 22, 24, 77, 126, 199, 200, 202, 203, 223, 226, 230, 231, 237, 241, 242, 243, 244 controlled studies, 245 controlled trials, 244, 245 controversies, 208 convergence, 108, 315, 317 conviction, 128, 267, 317

333

core personality traits, 100 corporate governance, 266 correlation, 27, 75, 98, 105, 110, 111, 134, 145, 268, 280, 302, 309, 312, 315 correlation coefficient, 75, 110, 111, 312, 315 correlations, 23, 28, 29, 57, 74, 89, 95, 96, 105, 107, 109, 110, 112, 134, 135, 144, 145, 147, 155, 177, 214, 217, 268, 304, 309, 312, 314, 326 corruption, 274 cortex, 24, 33, 35 cost, 29, 145, 171, 172 country of origin, 136 covering, 205, 277 creative process, 87, 90 creative thinking, 92 criminal activity, 139 criminal acts, 133 criminality, 194, 198, 206 criminals, 134, 138, 270 critical thinking, 321, 323 critics, 132, 328 cross-cultural comparison, 135, 139 cross-cultural differences, 135, 139 cues, 17, 80, 124, 152, 225, 226, 227, 230, 231, 232, 234, 235, 255 cultural barriers, 2 cultural differences, 135, 139, 140, 181, 182, 231 cultural studies, 181 culture, 5, 6, 47, 56, 119, 120, 122, 128, 129, 131, 135, 139, 140, 158, 181, 182, 188, 189, 231, 318 current limit, 245 current relationship, 72 curriculum, 96, 106

D danger, 152, 303 Darwinian evolution, 326 data collection, 198, 279, 283, 284, 285, 292 data set, 43, 56, 73, 86, 280 database, 284, 286, 291 decision task, 303 decision-making process, 91 declarative knowledge, 97 defamation, 267 deficit, 196, 207, 208, 250, 293 delinquency, 273 delusions, 152 demographic characteristics, 241 dependent variable, 44, 56, 57, 58, 76, 77, 78, 99, 243, 283 depression, 40, 46, 48, 149, 209, 210, 211, 214, 215, 216, 219, 220, 221, 222, 246, 247, 266, 279, 280, 281

334

Index

desensitization, 123 destruction, 133, 195, 196 detainees, 194, 200, 202 detection, 134, 139, 199, 219, 228, 273, 305 detection system, 305 detention, 198 determinism, 168 devaluation, 211, 216, 219 developmental theories, 112 deviation, 77, 309 diagnosis, 3, 7, 123, 135, 196, 198, 214, 222, 237, 240, 241 Diagnostic and Statistical Manual of Mental Disorders, 3, 231 diagnostic criteria, 123, 214 differential psychology, 26, 84 dimensionality, 312, 316 dimorphism, 81, 291 direct observation, 302 disability, 240, 245 disadvantages, 6, 283, 285 disclosure, 82, 149, 179 discrimination, 177, 199, 211, 225 disorder, 6, 121, 123, 125, 126, 128, 129, 189, 194, 195, 196, 205, 206, 207, 208, 215, 219, 220, 221, 224, 226, 227, 229, 231, 232, 233, 234, 235, 237, 238, 240, 241, 243, 246, 247, 248, 261, 263, 270, 293 displacement, 316, 317 disposition, 132, 170 dissatisfaction, 133, 139 dissociation, 91, 127, 226, 234 distress, 80, 119, 152, 157, 158, 159, 196, 210, 211, 214, 215, 217, 221, 238, 241, 243 disturbances, 210, 211, 219, 227, 232 divergent thinking, 84, 85 diversity, 231, 266, 284 DNA, 194, 195, 199, 200 dogs, 207 dominance, 178, 179, 263 dopamine, 34, 194, 197, 206, 207, 208 dopaminergic, 194, 196, 197, 207 dosage, 22, 237 double helix, 194 downward comparison, 37, 39, 40, 41, 43, 44, 45, 46 draft, 3, 63, 216 drawing, 93 DRD4 gene, 194, 197, 198, 199, 200, 205 dream, 323, 328 drug abuse, 269, 270 drug consumption, 247 dualism, 163, 166, 167, 168, 169, 170

E eating disorders, 210, 224, 226, 227, 231, 232, 233, 234, 235, 251, 253, 263 eating disturbances, 227, 232 educational background, 267, 310 educational objective, 100 educational programs, 30 educational psychology, 23 educational research, 20, 93 educational settings, 96, 97 effort level, 13, 14 electric current, 195 electrophoresis, 195, 199 elementary school, 115, 126 emotion, 37, 38, 45, 46, 47, 65, 69, 80, 120, 142, 143, 144, 146, 151, 159, 171, 183, 188, 189, 197, 210, 215, 220, 234, 259, 311, 322 emotion regulation, 159 emotional conflict, 254 emotional disorder, 194 emotional distress, 119, 211 emotional exhaustion, 281 emotional experience, 38, 222 emotional hardiness, 142 emotional information, 69, 79 emotional intelligence, 5, 6, 80, 81, 82, 141, 142, 143, 145, 146, 147, 148, 149, 171, 173, 174, 178, 279 emotional reactions, 48, 69 emotional responses, 42 emotional stability, 95, 154 emotional state, 209, 210, 214, 215, 216, 217, 219, 222, 327 emotional valence, 254 emotionality, 238, 262 empathy, 80, 142, 148, 172, 240, 270 empirical studies, 38, 133, 139 employability, 173 employment, 85, 177, 266, 267, 268, 269, 272, 274 enemies, 133, 152 engineering, 171, 172 England, 141, 264 English language proficiency, 54 entrepreneurs, 276 environmental control, 285 environmental factors, 196 environmental influences, 153, 208, 301 ERA, 64 ethanol, 199 etiology, 196, 208, 234 excitation, 68, 71, 80, 81, 86 exclusion, 52, 64, 65, 164, 300

Index executive function, 26, 28, 29, 33, 286, 287 executive functioning, 33 executive functions, 26, 287 executive processes, 24, 26, 28, 29, 104 exercise, 5, 67, 68, 71, 72, 78, 93, 94, 95, 98, 100, 101, 127, 164, 216 exertion, 68, 72, 216, 217 experiences, 3, 37, 38, 39, 40, 47, 70, 104, 120, 122, 124, 127, 152, 153, 158, 159, 224, 257, 276, 301, 302, 304 experimental condition, 68 experimental design, 226, 250 expertise, 24, 25, 30, 34 exploration, 159 exposure, 52, 225, 226, 241, 257, 269 externalizing disorders, 195 extinction, 259 extraversion, 12, 16, 262, 298, 325, 326 eye movement, 303

F facial expression, 144, 146 factor analysis, 74, 89, 133, 143, 183, 220, 280, 309, 312, 319 faith, 3, 123, 159, 165, 168 false negative, 271 false positive, 136, 271 family conflict, 261 family members, 231 feedback, 72, 291 feelings, 11, 12, 14, 39, 42, 45, 46, 67, 68, 69, 78, 79, 104, 123, 124, 132, 142, 174, 183, 224, 230, 231 fidelity, 240 field trials, 207 financial crisis, 6 first degree relative, 196 fitness, 99, 100, 142 five-factor model, 11, 12, 16, 101, 102, 183, 189, 247, 318 flexibility, 19, 142 fluid, 5, 21, 26, 33, 34, 35, 85, 88, 92 fluid intelligence, 5, 21, 33, 34, 35, 88, 92 foundations, 91, 120, 121, 138 fragments, 195, 199 fraud, 6, 265, 266, 267, 269, 270, 271, 272, 273, 274 free recall, 113 freedom, 75, 132 frequencies, 194, 199, 200, 201, 202, 203, 204, 205, 207 friendship, 223, 225, 227, 230, 231, 232 functional fixedness, 85, 87

335

G gambling, 250, 261, 262, 271, 289 gel, 195, 199, 200 gender differences, 64, 134 general intelligence, 33, 34 general knowledge, 53, 54, 56, 57, 58, 59, 62 generalized anxiety disorder, 220, 246 Generalized Anxiety Disorder, 221, 222 genes, 194, 195, 196, 197, 206, 207, 301 genetic factors, 196 genetic information, 194 genetic linkage, 196 genetic marker, 4, 206 genetics, 4, 207, 326, 330 genotype, 194, 195, 200, 201, 205 geography, 189 Germany, 114, 178 Gestalt, 84 goal setting, 99 goal-setting, 94, 101 God, 3, 8, 121, 122, 124, 134, 136, 138, 139, 147, 152, 154, 157, 159, 163, 164, 165, 166, 167, 168, 169, 170 good deed, 121 governance, 266, 274 grades, 41, 46, 58, 61, 96, 114 Greece, 234 Greeks, 184, 187 grounding, 86 group identity, 158 group interactions, 243 group membership, 240 grouping, 198, 205, 242, 315 Guatemala, 134 guidance, 124, 158, 239

H hallucinations, 153, 157 handedness, 177, 293 haplotypes, 200, 203, 204, 206, 207 happiness, ix, 42, 121, 125, 127, 132, 142 health problems, 120, 126, 127, 210 health services, 194, 195, 276 heart rate, 71, 75, 260 hedonism, 147 hemisphere, 172, 173, 179, 290 hemispheric asymmetry, 289 hemisphericity, 171, 172, 173, 174, 175, 176, 177, 178 heterogeneity, 195, 247 high school, 46, 263

336

Index

high school grades, 46 holism, 184, 188 honesty, 267, 268, 269, 273, 285 hopelessness, 211, 216, 219 hostility, 299, 311 human behavior, 305 human brain, 128, 178, 197, 198 human condition, 121 human development, 1, 97, 297 human nature, 7, 132, 133, 247, 328, 330 humanistic psychology, 329 hydrogen, 194 hydrogen bonds, 194 hyperactivity, 194, 196, 207, 208, 293 hyperarousal, 211, 217 hypnosis, 119, 120, 123, 124, 128, 129 hypnotherapy, 128 hypothesis, 44, 45, 46, 57, 61, 70, 71, 75, 77, 90, 96, 97, 168, 169, 173, 176, 186, 187, 215, 216, 219, 224, 226, 235, 239, 309, 319, 327

I ideal, 189, 245 idiosyncratic, 52 image, 289, 290 imagery, 64, 123 images, 92, 164, 166, 174 imagination, 129 immigration, 31 impacts, 1, 274, 302 implicit memory, 86 impulses, 324 impulsive, 95, 251, 256, 259, 260, 262, 263 impulsiveness, 251, 259, 261 impulsivity, 194, 196, 197, 198, 207, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 263 inattention, 198 incarceration, 194, 200, 205 INCOM, 42, 43, 44 incompatibility, 163, 164, 165, 167, 168, 169 incumbents, 275, 276, 277 independence, 14, 112, 142 independent variable, 44, 243 India, 120, 171, 172 individual differences, iv, 1, 2, 3, 4, 5, 6, 12, 13, 16, 28, 37, 38, 39, 40, 47, 48, 63, 69, 87, 91, 97, 101, 103, 105, 128, 198, 232, 237, 238, 247, 251, 262, 268, 281, 300, 301, 302, 308, 318, 321, 328 individualism, 138, 181, 183 individuality, 329 Indonesia, ix, 119, 120, 128, 129, 181, 182, 183 inertia, 21, 32, 211, 216, 219 inferences, 13, 256

inferiority, 39, 45, 48 information processing, 152, 159, 237, 238, 259 informed consent, 54 inhibition, 28, 29, 86, 198, 251, 252, 253, 255, 257, 259, 260, 262, 263, 286, 328 inhibitor, 252, 262 insecurity, 152, 154 integration, 12, 33, 128, 161, 207 intelligence, 5, 6, 21, 22, 26, 27, 29, 30, 31, 32, 33, 34, 35, 54, 55, 80, 81, 82, 83, 85, 88, 89, 90, 92, 114, 115, 141, 142, 143, 145, 146, 147, 148, 149, 171, 173, 174, 178, 183, 252, 263, 279, 287, 292, 302, 303, 305, 308, 329 intelligence scores, 174 intelligence tests, 88, 90 interaction effect, 57, 59, 62 interaction effects, 57, 59 interdependence, 15 interference, 29, 252, 253, 286, 287, 289, 293 intermediaries, 195 internal consistency, 43, 56, 69, 105, 141, 143, 145, 147, 196, 241, 278, 293, 315 internal controls, 271, 273 Internal structure, 311 Internet, 292, 293 interpersonal conflict, 80 interpersonal interactions, 96 interpersonal relations, 1, 95, 142, 160, 238, 241, 243, 277 interpersonal relationships, 95, 142, 160, 238, 243, 277 intervention, 17, 23, 24, 119, 128, 135, 136, 159, 238, 239, 240, 241, 245, 246, 247, 265, 281 intimacy, 152, 154, 158, 230 introns, 195 introversion, 305 inversion, 199 investors, 266 IP address, 286 IQ scores, 32, 171 irritability, 210, 211, 216 Islam, 137, 138 isolation, 148, 230, 235 Israel, 263

J Java, 120 job performance, 13, 15, 18, 26, 97, 252, 262, 269 job satisfaction, 15, 16, 19, 281 jurisdiction, 269 justification, 131, 133, 134, 139, 166

Index

K knowledge acquisition, 24, 27, 30, 96 Korea, 133

L labour force, 172 labour market, 271 landscape, 97 language proficiency, 54 languages, 99, 214 laptop, 70, 71 laterality, 290 Latin America, 135 leadership, 141, 142, 145, 148, 178 learning, 5, 18, 20, 25, 26, 27, 29, 30, 33, 35, 91, 93, 94, 95, 96, 97, 98, 99, 100, 101, 103, 110, 111, 112, 113, 114, 123, 188, 194, 207, 238, 240, 246, 249, 255, 260, 262, 263, 288, 289, 291, 292, 293, 297, 301, 302, 304 learning environment, 113 learning outcomes, 93, 96, 97, 99, 100 learning process, 93, 94, 96, 97, 98, 99, 100 learning styles, 291 learning task, 255, 289 left-hemisphere, 172 legislation, 266 limbic system, 197 Limitations, 46, 63, 226, 244 linear model, 13, 14, 15, 19, 20 linearity, 312, 313 literacy, 142 local government, 267 locus, 64, 81, 194, 195, 200 logical reasoning, 290 loneliness, 224, 230, 233, 234, 235 longitudinal study, 101, 318 long-term memory, 25lying, 154, 155, 156

M Machiavellianism, 273 machinery, 196 major depression, 220 majority, 2, 16, 23, 75, 107, 186, 217, 227, 271, 278 Malaysia, 134, 136 management, 24, 54, 99, 274, 275, 276, 280, 281, 284 manipulation, 41, 46, 77, 86, 270 mapping, 307, 312 markers, 4, 68, 72, 87, 107, 128, 173, 179, 196, 199, 205, 308 Marlowe-Crowne Social Desirability Scale, 227

337

marriage, 7, 189 mathematics, 56, 57, 62, 326 matrix, 22, 75, 88, 312 media, 136, 240 mediation, 159, 160 medication, 123, 243 membership, 240, 315 memory, 21, 22, 25, 26, 33, 34, 35, 86, 91, 120, 123, 124, 128, 129, 197, 207 memory capacity, 34, 35 memory performance, 207 menopause, 70, 80 mental ability, 34, 88, 94 mental development, 122 mental disorder, 4, 7, 119, 121, 123, 129, 189, 196, 207, 220, 221, 272 mental health, 3, 4, 120, 126, 127, 129, 152, 194, 195, 196, 198, 210, 220, 222, 230, 233, 276, 281 mental illness, 4, 119, 120, 126, 129, 238 mental image, 64 mental imagery, 64 mental processes, 3 messages, 144, 146, 148 messengers, 197 meta-analysis, 18, 95, 96, 97, 98, 101, 102, 208, 238, 239, 247, 268, 292 metabolism, 195 metabolizing, 197 metacognition, 5, 103, 104, 110, 114 metacognitive skills, 110, 112 methodological implications, 100, 250 methodology, 13, 14, 79, 86, 225, 240, 250, 258, 285 military, 97, 101 misconceptions, 136 modelling, 13, 14, 15, 26, 97, 312 moderates, 69, 79, 178 modification, 125, 305 modules, 283, 284, 286, 291, 292, 326 molecular weight, 199 molecules, 195 monitoring, 69, 104, 110, 112, 113, 325 mood change, 69, 144 mood disorder, 210, 215 mood states, 221 moral reasoning, 326 motivation, 20, 30, 33, 96, 97, 101, 111, 134, 142, 164, 189, 194, 197, 238, 239, 244, 245, 246, 248, 252, 258, 260, 284, 322, 327, 329 motor activity, 197 motor control, 194, 253, 261 MRI, 293 multidimensional, 3, 82, 233, 247, 312, 318, 319 multiple regression, 44, 100

338

Index

multiple-choice questions, 104 multivariate statistics, 48, 316

N narcissism, 46, 47, 270 narcissistic personality disorder, 270 nationality, 56 negative affectivity, 40, 45, 46, 262 negative consequences, 254 negative emotions, 39, 40, 45 negative outcomes, 250 negative reinforcement, 80 neglect, 31, 164, 323 nervous system, 120, 322 Netherlands, 37, 42, 43 neural network, 86 neural networks, 86 neurons, 86, 195 neuropsychological tests, 261 neuropsychology, 305 neuroscience, 28, 33 neurotransmitter, 194, 197 New South Wales, 11, 21, 65, 151 New Zealand, 101, 246, 266, 272, 281 Nietzsche, 38, 39, 48 noise, 277, 285 nonconscious, 5, 85, 87, 88, 90 normal distribution, 56 North America, 2, 6, 189, 231 Norway, 193 nuclear family, 124 nucleic acid, 194 nucleotides, 194 nucleus, 194 nurses, 199, 275, 276

O obesity, 112 objectivity, 269 obstacles, 136, 144, 146 occupational groups, 277 OCD, 243, 247 offenders, 198, 258, 262, 265, 266, 267, 269, 270, 271, 272, 273 omission, 147, 217 openness, 12, 17, 30, 143, 171, 177, 187 openness to experience, 12, 17, 30, 143 opportunities, 30, 32, 113, 177, 208, 271, 272 Oppositional Defiant Disorder, 195 optimism, 78, 80, 108, 142, 148 organism, 195 organizational stress, 281

orthogonality, 308 overlap, 84, 86, 157, 176, 197, 227, 241, 250, 309

P Pacific, 273 pain, 46, 48, 222 panic disorder, 246 parallel, 22, 88, 152, 172, 186, 290, 312, 316 paranoia, 151, 153, 157, 159 Parisians, 31 password, 286 path analysis, 101 path model, 97 pathogenesis, 197 pathology, 235 pathophysiology, 194 pathways, 194, 262 pattern recognition, 172 PCR, 195, 199 Pearson correlations, 155 peer review, 268 percentile, 155 perceptual learning, 91 perceptual processing, 52 performers, 288 permission, 119, 163, 307 perpetrators, 269, 270, 272 personal communication, 42, 240 personal goals, 20, 98, 102, 123 personal identity, 52 personal problems, 144, 146 personal relations, 123, 157 personal relationship, 123, 157 personal responsibility, 132 personality characteristics, 12, 40, 173, 252 personality constructs, 99, 161, 237, 238, 268 personality differences, 173 personality dimensions, 17, 18, 96 personality disorder, 4, 8, 153, 161, 196, 236, 263, 270 personality factors, 3, 5, 12, 30, 93, 94, 97, 98, 105, 174, 177, 279, 319 personality measures, 97 personality research, 12, 51, 52, 93, 254, 258, 262, 309, 322 personality scales, 96 personality traits, 4, 16, 18, 20, 68, 70, 75, 94, 95, 96, 97, 98, 99, 100, 114, 129, 134, 135, 136, 152, 160, 161, 173, 181, 182, 188, 189, 248, 251, 254, 258, 265, 301, 303, 308, 315, 317, 318 personhood, 187, 188 persuasion, 153 pessimism, 80, 148

Index phobia, 48, 243 photographs, 290 physical activity, 93, 94, 95, 98, 100, 101, 102 physical exercise, 5, 68, 71, 78, 216 physical fitness, 99, 100 physical health, 80, 98, 277 physiological arousal, 68, 70, 71, 77, 78, 211, 216 pilot study, 240, 244, 245, 248 PISA, 135 plasticity, 23, 26, 27, 33, 128 platform, 199, 284 pleasure, 37, 38, 39, 40, 42, 45, 46, 48, 210 political parties, 137, 138 polymorphism, 195, 198, 200, 204, 208 polymorphisms, 194, 198, 199, 200, 202, 203, 204, 205, 207 poor performance, 147, 217 positive correlation, 105, 107, 110 positive emotions, 124 positive mood, 144, 146 positive relationship, 112, 176 postmodernism, 164, 165 posttraumatic stress, 126, 128, 129, 247 potassium, 197 prayer, 152, 155, 157 predicate, 167, 169 predictive validities, 96 predictive validity, 98, 196 predictor variables, 44 prefrontal cortex, 24, 33 prejudice, 163, 164, 165, 170 prejudices, 170 prevention, 119, 120, 196, 210, 219, 271, 273 primary school, 103, 111, 219 priming, 53, 63, 64, 159, 230 prior knowledge, 94 prisoners, 143, 146 proactive interference, 29 probability, 104, 259, 288, 289 probe, 200 problem drinking, 253, 261 problem solving, 35, 84, 86, 87, 142, 241, 287 problem-solving, 24, 27, 84, 85 problem-solving task, 84, 85 procedural knowledge, 96 processing biases, 159 productivity, 177 professional development, 143, 147 programming, 171, 172 project, 32, 123, 141, 185, 286, 316 proliferation, 321 promax rotation, 74 property crimes, 198

339

proposition, 86 proteins, 195, 197 psychiatric diagnosis, 7 psychiatric disorders, 210 psychiatrist, 122 psychoanalysis, 123, 328, 329 psychobiological systems, 120 psychological assessments, 198 psychological development, 120 psychological distress, 80, 210 psychological functions, 198 psychological health, 121 psychological problems, 121, 235, 245, 277 psychological processes, 13, 18, 123, 189, 303 psychologist, 164 psychometric properties, 69, 139, 143, 147, 214, 219, 250, 253 psychopathology, 4, 7, 135, 153, 225, 227, 234, 238, 247, 254, 260, 262, 263, 264 psychopaths, 262, 263, 270, 273 psychopathy, 207, 270, 273 psychosis, 129, 151, 152, 159 psychosocial factors, 80 psychosocial functioning, 128 psychotherapy, 7, 99, 129, 237, 238, 245, 247 psychotic symptoms, 161, 240 psychoticism, 4, 325 PTSD, 119, 125, 126, 129, 243, 247 public schools, 198 pumps, 288, 289 punishment, 255, 257, 259, 260, 261, 262

Q quality of life, 238, 243 questioning, 2, 4

R random configuration, 312 rash, 256, 259, 261 rating scale, 104, 105 rationality, 114, 163, 165, 184 reaction time, 255, 284, 285, 286 reactions, 19, 33, 48, 68, 69, 97, 277, 293, 301, 302, 303, 305 reading, 3, 56, 57, 62, 288 realism, 91, 113, 114, 170, 329 reality, 64, 124, 127, 142, 163, 165, 167, 168, 286, 309, 319, 326 reasoning, 22, 24, 26, 27, 28, 29, 30, 33, 35, 165, 170, 171, 174, 175, 176, 186, 290, 292, 297, 299, 321, 326, 327 reasoning skills, 174

340

Index

recall, 14, 113, 224, 228 receptors, 195, 197 recognition, 69, 164, 172, 173, 271, 277 recommendations, iv, 234 redundancy, 322 reflexes, 299, 304 regression, 44, 100, 123, 156, 177, 246, 307, 312, 316, 317 regression model, 44, 156 reinforcement, 80, 263, 302 reinforcers, 259 rejection, 168, 299, 300, 301, 302, 303, 304, 305 relatives, 196 relativism, 170 relaxation, 99, 241 relevance, 1, 3, 30, 54, 55, 124, 197, 264 reliability, 43, 56, 70, 81, 95, 105, 106, 108, 112, 141, 143, 145, 147, 154, 158, 196, 207, 241, 267, 273, 308, 311 religion, 3, 5, 7, 8, 119, 120, 148, 152, 154, 159, 160, 165, 166 religiosity, 134, 136, 139, 161 religious beliefs, 129, 131, 158, 159, 160 religiousness, 161 repair, 82, 149, 178 replication, 68, 79, 107, 153, 194, 318 resentment, 39, 48, 131 residuals, 14 resilience, 93, 96, 101, 102, 278 resistance, 240, 259 resolution, 84 resource allocation, 263 resource management, 54 resources, 23, 25, 159, 172, 266, 275, 277, 278, 279, 280, 323 response format, 108 response time, 285 responsiveness, 17 restructuring, 241 reversal learning, 255, 260, 289 rewards, 208, 255, 274 rhythm, 288 right hemisphere, 173 rights, iv, 194, 196 risk factors, 136 risk-taking, 252, 263, 289, 292, 324 RMSEA, 107, 316, 317 RNA, 194, 195 Royal Society, 81 Russia, 193, 198

S sampling distribution, 312

Sarbanes-Oxley Act, 274 scaling, 307, 312, 318, 319 Scandinavia, 220 schema, 153, 231, 233, 234, 236 schizophrenia, 161 schizotypal personality disorder, 153, 161 schizotypy, 6, 151, 153, 154, 155, 156, 157, 158, 159, 160 school activities, 109 schooling, 26, 122 science education, 114 scientific knowledge, 165 scientific method, 85, 329 sclerosis, 207 SCO, 37, 39, 40, 41, 42, 44, 45, 46 screening, 71, 135, 161, 240, 245, 265, 266, 267, 268, 269, 271, 274, 323 SEC, 274 second language, 58, 60, 62, 223, 228, 229, 230, 231, 232 secondary students, 252 secularism, 6, 163, 170 self esteem, 305 self representation, 52 self-awareness, 142, 234, 250 self-concept, 5, 45, 51, 52, 63, 65, 101, 108, 109, 114, 325 self-confidence, 5, 33, 103, 104, 105, 108, 110, 111, 112, 113, 114, 325 self-control, 80, 94, 325 self-discipline, 94, 98 self-doubt, 135, 140 self-efficacy, 20, 94, 97, 241, 325 self-esteem, 40, 45, 46, 97, 270, 325 self-evaluations, 61, 62, 96 self-interest, 134 self-monitoring, 104, 112, 325 self-perceptions, 112 self-presentation, 227 self-reflection, 112 self-regard, 142 self-regulation, 17, 114, 142, 262 self-reports, 63, 100, 109, 318 self-worth, 114 sensation, 108, 252, 254, 255, 257, 263, 264 sensation seeking, 252, 254, 255, 257, 263, 264 sensations, 224, 257 sensitivity, 48, 71, 139, 238, 259, 261, 262, 263, 297, Serbia, 131, 134 setting goals, 241 sex, 64, 145, 146, 174, 194, 257, 264, 310 sex chromosome, 194 sex differences, 64, 146

Index sexual activities, 261 sexual dimorphism, 81 sexual orientation, 72, 119, 123 shape, 3, 55, 56, 81, 120, 166, 223, 224, 317 shortness of breath, 68, 72 short-term memory, 25 Siberia, 184 SIGMA, 56, 64 signals, 14, 195, 200 significance level, 57, 243 signs, 20, 157 simple linear regression, 44 Singapore, 113, 129, 131 skill acquisition, 97 skills training, 99 skin, 260 slavery, 31 sleep disturbance, 211 smoking, 257 SNP, 195, 204 soccer, 53 sociability, 324, 325 social anxiety, 238 social attitudes, 134, 135, 136, 139 social behaviour, 53, 64, 261, 270 social cognition, 305 social comparison, 37, 38, 39, 40, 42, 44, 45, 46, 47, 48, 51, 52, 53, 54, 57, 58, 59, 60, 61, 62, 63, 64, 65 social comparison theory, 37, 38, 39, 40, 47 social competence, 142 social construct, 165, 170, 329 social context, 52 social contract, 186 social costs, 280 social desirability, 152, 232, 234, 293 social environment, 5, 38 social group, 300 social identity, 65 social institutions, 31 social justice, 132 social life, 138 social network, 230, 232 social norms, 134 social phobia, 48 social programs, 32 social psychology, 47, 49, 65, 181, 182, 261 social responsibility, 132, 142 social rules, 195 social sciences, 328 social situations, 214, 299 social skills, 42, 142 social structure, 184

341

social support, 160, 281 social withdrawal, 153 socialization, 261, 263 software, 13, 25, 27, 71 South Africa, 81, 267, 273 South Korea, 231, 233 spatial ability, 64 spatial location, 22 speech, 71, 132, 166, 239 spelling, 286 spirituality, 3, 5, 122, 161 stakeholders, 266 standard deviation, 72, 73, 77, 89, 309, 310 standard error, 14, 74, 202, 205, 316 stars, 142, 185 State-Trait Anxiety Inventory for Children, 210, 222 statistics, 43, 48, 125, 154, 155, 156, 175, 233, 269, 316, 319 stereotypes, 21, 32 stimulus, 22, 52, 71, 78, 223, 224, 225, 227 storage, 22, 28, 128 stressors, 123, 278, 279, 281 structural changes, 25 structural characteristics, 313 structuring, 307, 308, 309 subgroups, 235 subjectivity, 267 substance abuse, 196, 240, 247, 250, 261, 262, 268 substance use, 251, 252, 254, 258, 260, 261, 262, 263 substitution, 199, 300, 302 substitutions, 198 substrates, 300 success rate, 125, 291 suicidal ideation, 266 suicide, 136, 210 suicide bombers, 136 supernatural, 121, 163, 164, 165, 166, 167, 169, 170 supervision, 240 supervisor, 240, 269 supervisors, 291 suppression, 129, 160, 259 survey, 139, 143, 147, 173, 186, 228, 266, 276, 280, 291 Sweden, 193, 206, 228 syllogisms, 184, 185, 189 symbolism, 329 sympathy, 38, 47, 48 symptomology, 226 symptoms, 4, 80, 123, 126, 161, 196, 209, 210, 211, 214, 215, 216, 217, 219, 220, 221, 222, 233, 238, 240, 247, 266, 281 syndrome, 211, 214, 215

342

Index

synergistic effect, 244 synthesis, 194, 322 systematic desensitization, 123 systemic sclerosis, 207

T tandem repeats, 195 task conditions, 22 task demands, 292 task difficulty, 15 task performance, 15, 27, 34 task-specific strategies, 24 taxonomy, 12, 100, 104, 272, 309, 316, 319 temperament, 157, 292, 322 tenants, 122 tension, 209, 210, 211, 214, 215, 216, 217, 219, 222 terrorism, 132, 135, 137 terrorist acts, 133 terrorist groups, 135, 137, 139 terrorists, 132, 135, 138, 139 tertiary education, 96, 99 test anxiety, 96, 99, 101, 135 test items, 88, 89 test scores, 89 testing, 6, 14, 31, 35, 83, 85, 96, 142, 168, 171, 211, 216, 238, 268, 273, 301 testosterone, 291 test-retest reliability, 56, 196 theft, 195, 196, 266, 268, 269, 271 theoretical psychology, 164 therapeutic approaches, 159 therapeutic intervention, 159 therapeutic practice, 123 therapy, 5, 6, 99, 119, 126, 127, 129, 224, 226, 233, 234, 235, 236, 237, 238, 239, 241, 243, 244, 245, 246, 247, 248 think critically, 170 thoughts, 11, 12, 14, 124, 132, 183, 224, 239, 256, 262, 270 threats, 225, 232, 234 thymine, 194 time periods, 75 time variables, 62 time-frame, 96 traditions, 2, 105 traffic control, 275, 276 training, 5, 21, 22, 23, 24, 25, 26, 27, 28, 30, 31, 32, 33, 34, 97, 98, 99, 100, 101, 122, 198, 240 training programs, 21, 32 trait anxiety, 101 trajectory, 23, 237, 238, 239, 241, 242, 243, 244 transactions, 167, 266 transcendence, 3

transformation, 43, 56, 229, 290, 292, 312 transformations, 312 translation, 221 transmission, 129 transport, 132 trauma, 125, 126, 128, 129 trial, 22, 89, 240, 244, 246, 248 triggers, 120 trustworthiness, 272, 273 Turkey, 279 turnover, 197 typology, 233

U UK, 19, 33, 34, 65, 91, 114, 276, 280, 281 underlying mechanisms, 249, 250, 256, 257, 258, 289 uniform, 250, 254 United Kingdom, 132, 141 universality, 2, 330 universe, 121, 183, 185, 324 universities, 31 unusual perceptions, 157 unusual perceptual experiences, 153 use of force, 136

V valence, 80, 97, 254 validation, 82, 97, 135, 149, 158, 221, 233, 260, 268, 269 variations, 4, 15, 68, 184 video, 67, 71, 72, 73 video-recording, 71 violence, 132, 133, 136, 139, 198 violent behaviour, 195 violent crime, 266 visions, 151, 329 visual processing, 90, 225, 234 visual stimuli, 22, 81, 225 vocabulary, 114, 234 vulnerability, 4, 120

W wages, 172 Wales, 11, 21, 65, 151, 283 Western countries, 172 white-collar, 6, 272 William James, 3, 152 withdrawal, 34, 153 witnesses, 45, 113 work environment, 18, 171, 269, 277, 278, 279 work roles, 278

Index workers, 104, 123, 177, 280, 281, 291 working memory, 21, 22, 26, 33, 34, 35, 207 workplace, 5, 11, 12, 94, 97, 142, 173, 177, 178, 268, 271, 273, 276, 277, 279, 280, 281 work-related stress, 280 world order, 132 World War I, 132

343

World Wide Web, 172 worldview, 3, 164, 165, 169 worry, 97, 210, 220, 222, 241

Y young women, 226

[Simon Boag, Niko Tiliopoulos] Personality and In

Recommend Documents