Attachment in the Preschool Years Theory, Research, and Intervention
EDITED BY Mark T. Greenberg, Dante Cicchetti, and E. Mark Cummings
The University of Chicago Press Chicago and London
The University of Chicago Press, Chicago 60637 The University of Chicago Press, Ltd., London !""0 #y The University of Chicago All rights reserved. P$#lished !""0 Pa%er#ac& Pa%er#ac& edition !""3 !""3 Prin Printe ted d in in the the Unit United ed 'tat 'tates es of Amer Americ icaa "" "" "( "( "7 "7 ) *3 The University of Chicago Press gratef$lly ac&no+ledges a s$#vention from the ohn -. and Catherine T. T. acArth$r /o$ndation in %artial s$%%ort of the costs of %rod$ction of this vol$me.
Library of Congress Congress Cataoging!in!"ubication Data Attachment in the %reschool years theory, theory, research, and intervention 1 2edited #y ar& T. 4reen#erg, 4reen#erg, -ante Ciccetti, 5. ar& C$mmings, %. cm.The ohn -. and Catherine T. T. acArth$r /o$ndation series on mental health and develo%ment 8ncl$des #i#liogra%hical references. 8'9: 0;<<6;306<";! cloth 8'9: 0;<<6;30630;) %a%er#ac& !. Attachment #ehavior in children. <. Child %sycho%athology. %sycho%athology. 8. 4reen#erg, ar& T. T. 88. Ciccetti, -ante. 888. C$mmings, 5. ar&. 8=. 8=. 'eries. 'er ies. 9/7<3.A7)A(7 !""0 !)).*<>3(;dc<0 (";*"6)7 C8P The %a%er $sed in this %$#lication meets the minim$m re?$irements of the American :ational 'tandard for 8nformation 'ciencesPermanence of Pa%er for Printed Li#rary aterials, A:'8 @3".*(!"(*.
Contents Contri#$tors
vii
Preface The Third Phase of Attachment esearch
iB
Robert N. Emde 8ntrod$ction i istory of of a Colla#oration in in th the 't 't$dy of of Attachment
Biii
Mark T. T. Greenberg, Greenberg, Dante Cicchetti, Cicchetti, and E. Mark Cummings Cummings
"#$T I % T&E'$ETIC#L I(()E( !.
An Drga Drgani niEa Eati tiona onall Pers Pers%ec %ecti tive ve on Atta Attach chme ment nt #eyo #eyond nd 8nfa 8nfanc ncy y 8m%lications for Theory, eas$rement, and esearch
3
Dante Cicchetti, E. Mark Mark Cummings, Mark T. T. Greenberg, Greenberg, and Robert . Marvin <.
A /amily 'ystems /rame+or& for the 't$dy of Attachment
)!
Robert . Marvin and Robert !. te"art 3.
Theo Theoret retic ical al and and et etho hodo dolo logi gica call Cons Consid ider erat atio ions ns in in the the 't$d 't$dy y of Atta Attach chme ment nt and the 'elf in Yo$ng Children (7 #ude Cassidy
*.
Proc Proced ed$r $res es for for 8den 8denti tify fyin ing g 8nfa 8nfant ntss as -iso -isorg rgan aniE iEed ed11-is isor orie ient nted ed d$ring the Ains+orth 'trange 'it$ation
!
Mary Main and #udith o$omon ).
Pare Parent ntsF sF Unres Unresol olved ved Tra$m ra$mat atic ic 5B%e 5B%eri rien ences ces Are Are el elat ated ed to to 8nfa 8nfant nt -isorganiEed Attachment 'tat$s 8s /rightened and1or /rightening Parental Parental 9ehavior 9ehavior the Lin&ing Lin&ing echanismG echanismG
!6!
Mary. Mary. Main and Erik %esse E(E#$C& '* T&E *'$M#L DE+EL'"ME*T "#$T II % $ E(E#$C& #TT#C&ME*T I* "$E(C&''L YE#$(
6.
'
The The -ev -evel elo% o%me ment ntal al eor eorga gani niEa Eati tion on of Atta Attach chme ment nt ela elati tion onsh shi% i%s s 4$idelines for Classification #eyond 8nfancy &aren chneider'Rosen chneider'Rosen
!()
CD:T5:T' 7.
(.
".
!0.
!!.
!<. !<.
!3.
!*.
'ec$ 'ec$ri rity ty of of Tod Toddl dler er;P ;Pare arent nt Atta Attach chme ment nt el elat atio ion n to Chil Childr dren enFFs 'ocio%ersonality /$nctioning d$ring Hindergarten <
"art III % De-eomenta "sychoathoogy of #ttachment Classi Classific ficati ation on of of Attach Attachmen mentt on on a Contin Contin$$m $$m of /elt /elt 'ec$ 'ec$rit rity y 8ll$strations from the 't$dy of Children of -e%ressed Parents E. Mark Cummings To+ard o+ard a Transact ransaction ional al odel odel of ela elatio tions ns #et+ #et+een een Attach Attachmen mentt and -e%ression E. Mark Cummings and Dante Cicchetti Cicchetti "art I+ % Cinica Inter-ention from an #ttachment "ersecti-e -iso -isorde rders rs of Atta Attach chme ment nt and and 'ec$ 'ec$re re 9ase 9ase 9eha 9ehavi vior or in the the 'ec 'econd ond Year of Life Conce%t$al 8ss$es and Clinical 8ntervention ($icia +. +. Lieberman and #eree #eree %. a"$ The Tr Treat eatmen mentt of of Presc Preschool hool Cond$c Cond$ctt Pro# Pro#lem lems s An 8ntegra 8ntegratio tion n of of 9ehavioral and Attachment Conce%ts Matthe" L. *e$t'ec$ri 'ec$rity ty of of Attach Attachmen mentt in Toddler oddlerhoo hood d odi odifyi fying ng Assess Assessmen mentt Proced$res for oint Clinical and esearch P$r%oses #. La"rence (ber and (my #. L. !aker
3!!
33"
37)
3""
*<7
"#$T+ % E"IL'G)E 'ome Considerations egarding Theory and Assessment elevant to Attachments #eyond 8nfancy *63 Mary D. a$ter (ins"orth (ins"orth 8ndeB *("
) J ParentsF Unresolved Tra$matic 5B%eriences Are elated to 8nfant -isorganiEed Attachment Attachment 'tat$s 8s /84T5:5- A:-1D /84T5:8:4 PA5:TAL 95A=8D T5 L8:H8:4 5CA:8' G ary ain and 5ri& esse
T5 'TU-Y D/ h$man attachment can #e encom%assed +ithin t+o maKor #ranches of in?$iry. in?$iry. The first has #een concerned +ith a a descri#ing the normal develo%ment and f$nctioning f$nctioning of an attachment attachment #ehavioral #ehavioral system, a system %res$med %res$med to have evolved evolved to ta&e contin$al contin$al acco$nt of the +herea#o$ts +herea#o$ts of caregivers attachment attachment fig$res, and to %romote %roBimity to those fig$res +henever necessary 9o+l#y 2!"6" !"(<M and b b +ith descri#ing the effects of %otentially tra$matic events, s$ch as loss of attachment attachment fig$res, $%on the f$rther f$nctioning f$nctioning of the individ$al. individ$al. Tra$matic Tra$matic loss of %arents or other attachment fig$res and a#$se #y attachment fig$res are considered li&ely to over+helm the develo%ing attachment #ehavioral system 9o+l#y !"73, !"(0 and to #e contri#$tory factors in de%ression 9ro+n, arris, N 9if$lco !"() and later diffic$lties in %arenting O$inton N $tter !"() . The The seco second nd #ran #ranch ch of in?$ in?$ir iry y has has cent centere ered d $%on $%on indi indivi vid$a d$all diff differe erenc nces es in attachment attachment organiEation. organiEation. -$ring infancy, infancy, these individ$al individ$al differences differences are $s$ally $s$ally identified thro$gh an eBamination of infant #ehavioral res%onses to the %arent in a se%a se%ara rati tion on and and re$n re$nio ion n o#se o#serv rvat atio ion n cond cond$c $cte ted d in the the $nfa $nfami mili liar ar la#o la#ora rato tory ry envi enviro ronm nmen entt Ain Ains+ s+ort orth, h, 9leh 9lehar ar,, Iaters aters,, N Iall all !"7( !"7(. . The thre threee trad tradit itio iona nall categories of infant res%onse to this sit$ation are secure! the secure! the infant sho+s signs of missing the %arent, see&s %roBimity on re$nion, and then ret$rns to %layM avoidant( the infant ignores and avoids the %arent $%on re$nionM and ambiva$entC the the infant is highly distressed and highly foc$sed on the %arent, cannot #e settled #y the %arent, and may see& %roBimity and dis%lay anger in ?$ic& s$ccession. 8n several st$dies, these res%onse categories have #een fo$nd associated +ith %arental #ehavior to+ard the infant, inde%endent inde%endent across %arents a given infant infant may #e sec$re +ith one %arent #$t avoidant +ith the other, and %redictive of the childFs later social;emotional f$nctioning see 9retherton !"() for revie+. This st$dy +o$ld not have #een %ossi#le +itho$t the assistance of Carol 4eorge, $th 4old+yn, and :ancy Ha%lan, +ho +ere critical to the design and cond$ct of the Ad$lt Attachment 8ntervie+s, and Anitra -eoss, +ho coded them. This %a%er has #een thro$gh many drafts and editio editions, ns, and and +e gratef gratef$ll $lly y ac&no+ ac&no+led ledge ge the critic criticism isms, s, ?$erie ?$eries, s, and comme commenta ntarie riess of ary ary Ains+orth, Anne 9e$ter, ohn 9o+l#y, $de Cassidy, 4iovanni Liotti, and $dith 'olomon.
!6!
!6<
A8: A:- 'DLDD:
ecently, +e have develo%ed an intervie+ centered $%on the ad$ltFs descri%tion and eval$ation of his or her attachment;related history 4eorge, Ha%lan, N ain !"(). 't$dy of the ver#atim transcri%ts of this intervie+ %ermits K$dges to classify each ad$lt into one of three ad$lt attachment categories sec$re;a$tonomo$s +ith res%ect to attachment, dismissing of attachment, or %reocc$%ied #y %ast attachments. These ad$lt classificati classifications ons have #een fo$nd to %redict %redict the infantFs infantFs 'trange;'it 'trange;'it$atio $ation n res%onse to the %arent, s$ch that, for eBam%le, sec$re %arents ty%ically have sec$re infants +hile dismissing %arents have avoidant infants ain !"()M ain, Ha%lan, N Cassidy !"()M ain N 4old+yn, in %ressM Ains+orth N 5ich#erg, in %ress. 8n addi additi tion on,, the the %aren %arentF tFss ad$l ad$ltt atta attach chme ment nt clas classi sifi fica cati tion on has has #een #een fo$n fo$nd d dire direct ctly ly %redictive of #ehavior to+ard the offs%ring Cro+ell N /eldman !"((M Iard, Carlson N Hessler !"("#M see also 4rossmann, /remmer;9om#i&, $dol%h, N 4rossmann !"((. 8n this cha%ter, +e disc$ss a significant connecting lin& #et+een these t+o central areas of in?$iry, that is, an association #et+een unreso$ved $oss o/ attachment /igures 0or other attachment're$ated trauma1 as trauma1 as eB%erienced #y the %arent, and the infantFs /ai$ure to /ai$ure to fit to one of the traditional, organiEed 'trange;'it$ation res%onse categories. This is in essence the discovery of a second;generation effect of $nresolved loss of attach attachmen mentt fig$re fig$res, s, +ith +ith the infants infants of %arents %arents +ho are K$dged K$dged $nreso $nresolve lved d +ith +ith res%ect to this %otentially tra$matic eB%erience #eing fo$nd to fit to a ne+ fo$rth infa infant nt atta attachm chmen entt cate catego gory ry no+ no+ term termed ed dis disor organ ganiE iEed ed1d 1dis isori orien ente tedQ dQ ai ain n N 'olomo 'olomon, n, this this vol., vol., and !"(6 !"(6 and %revio %revio$sl $sly y termed termed $nclas $nclassif sifia# ia#leQ leQ ain ain N Ieston !"(!. The The read reader er sho$ sho$ld ld note note that that ! ! the the term term tra tra$m $maQ aQ trad tradit itio iona nall lly y refe refers rs to eB%e eB%eri rien ences ces of inte intens nsee fear fear, terro terrorr, or hel% hel%le lessn ssness ess see see -'; -';888 888; ; 2Ame 2Ameri rica can n Psychiatric Association !"(7, +hich threaten an individ$al +ith %sychological or #ehavioral disorganiEation, altho$gh < +hether any given eB%erience is act$ally tra$matic $ltimately rests +ith the history the individ$al #rings to the eB%erience, and individ$al individ$al %erce%tions. %erce%tions. 5B%eriences 5B%eriences s$ch as %hysical %hysical or seB$al a#$se #y a %arent are almost inevita#ly tra$matic, +hile loss eB%eriences are only %otentially tra$matic, +ith the effects de%endent $%on the individ$al and $%on s$rro$nding conditions. ecent st$dies have demonstrated that not only $nresolved loss of im%ortant fig$res thro$gh death, #$t also $nresolved eB%eriences of %hysical or seB$al a#$se, and even more general recent tra$mas s$ch as a recent close #r$sh +ith death on the %art of the %arent are associated +ith infant D infant D attachment attachment stat$s see #elo+. At this %oint, only a very fe+ of o$r original 9ay Area s$#Kects have descri#ed tra$mas other than the %otential tra$ma im%lied in loss of im%ortant fig$res. ence in this cha%ter o$r attention +ill #e chiefly confined to loss eB%eriences. Ie #egin +ith a revie+ of disorganiEed1disoriented #ehavior %atterns and their se?$elae in recent st$dies. Ie neBt disc$ss the association discovered
8:/A:T -8'D4A:8@5- ATTAC5:T 'TATU'
!63
#et+een %arental lac& of resol$tion of loss eB%eriences and infant disorganiEation ain !"(3, !"()M ain N esse !"(", +hich has led to the develo%ment of a ne+, $nres $nresol olve ved d adu$t attach attachmen mentt catego category ry,, and st$die st$diess re%lica re%licatin ting g and eBtend eBtending ing o$r findings to incl$de $nresolved eB%eriences of %hysical and seB$al a#$se as %redictors of infant D infant D attachment attachment stat$s Ains+orth N 5ich#erg, in %ressM /riedman !"(6M ain, Ha%lan, N esse, $n%$#lished dataM Levine, Iard, N Carlson !"("M Iard, Carlson, Altman, Levine, 4reen#erg, N Hessler !"("a. Ie then eBamine the %atterning of infant disorganiEed1disoriented #ehavior o#served in the Ains+orth 'trange 'it$ation, +hich s$ggests that the infant may at times #e eB%eriencing a fear or distress too intense to #e deactivated thro$gh a shift in attention the Ains+orth ( %attern, ( %attern, yet at least momentarily cannot #e ameliorated thro$gh a%%roach to the attachment fig$re the Ains+orth ! and C %atterns. %atterns. This leads to a disc$ssion of the role that fear may sometimes sometimes %lay in the D infantFs D infantFs eB%eriences +ith the %arents, and to a disc$ssion of the hy%othesis is that the traumati-ed adu$t2s continuing state o/ o/ /ear together "ith its interactiona$behaviora$ concomitants 0/rightened 0/rightened andor /rightening behavior1 is the the mechanism $inking unreso$ved trauma to the in/ant3s dis*$ay o/ disorgani-eddisoriented behavior. behavior. '$ch #ehavior co$ld #e %artic$larly %$EEling or frightening to the infant #eca$se its immediate ca$se +o$ld often lie in the %arentFs res%onse to memories aro$sed #y ongoing events rather than res$lting from those events directly. 'ince the attachment fig$re is normally the sol$tionQ %rovided to the infant for dealing +ith stressf$l or alarming eB%eriences, an infant +ho is frightened #y the attachment fig$re is %resented +ith a %aradoBical %ro#lemnamely, an attachment /igure "ho is at once the source o/ and the so$ution to its a$arm. The cha%ter concl$des +ith a disc$ssion of the (, the (, !, !, and C infant infant attachment classifications as indica indicativ tivee of relati relativel vely y +ell;o +ell;orga rganiE niEed ed #ehavi #ehavior or %atter %atterns ns and infant infant D D res%onse %atterns as indicative of a momentarily irresolva#le conflict. -8'D4A:8@5- 1-8'D85:T5- 8:/A:T ATTAC5:T 'TATU' -5'C8PT8D: A:- '5OU5LA5 8n this section, +e #riefly revie+ the develo%ment of the D the D attachment attachment category and descri#e some of the #ehavior %atterns +hich are regarded as disorganiEed and1or disoriented. /or a more com%lete descri%tion of the D the D system, see ain and 'olomon this vol.. 45nc$assi/iab$e6 Res*onses to the (ins"orth trange ituation The (, The (, !, C classificatory classificatory system +as develo%ed in conK$nction +ith Ains+orthFs st$dy of a sam%le of 9altimore infants and mothers living in rela;
!6*
A8: A:- 'DLDD:
tively sta#le circ$mstances, and has re%eatedly #een s$ccessf$lly a%%lied to the st$dy of infants and %arents in similar lo+;ris&, middle;class sam%les. Iithin o$r o+n $%%e $%%er; r;mi midd ddle le;cl ;class ass 9ay 9ay Area Area sam% sam%le le,, +e had had earl earlie ierr re%ort re%orted ed diff diffic ic$l $lti ties es in classifying classifying the 'trange;'it 'trange;'it$atio $ation n res%onse res%onse of a#o$t !3R of infants infants ain N Ieston !"(!. 't$dies of high;ris&1maltreatment sam%les in +hich children +ere o#served +ith +ith malt maltrea reati ting ng or %syc %sycho hoti ticc %aren %arents ts in cont contra rast st yiel yielde ded d re%or re%orts ts of fre?$ fre?$en entt classi classific ficati ation on diffic diffic$lt $lties ies.. ecogn ecogniti ition on of these these diffic diffic$lt $lties ies often often #egan #egan +ith +ith the o#servation that, $sing the traditional system that classifies every infant as (, !, or !, or C, infants +ere #eing classified as ! sec$re +ith res%ect to a %arent +ho +as in all li&elihood maltreating or frightening see ain N 'olomon, this vol., for a revie+ of these st$dies. Disorgani-edDisoriented Disorgani-edDisorient ed In/ant (ttachment tatus tatus ece ecent ntly ly,, ain ain and and 'olo 'olomo mon n !"(6 !"(6 cond$ cond$ct cted ed an anal analys ysis is of the the 'tran 'trange ge;; 'it$ation #ehavior of fifty;five infants +ho +ere originally considered unc$assi/iab$e +ithin the traditional system ain N Ieston Ieston !"(!, and +ere no+ fo$nd to eBhi#it a diverse arra rray of disorgani-eddisoriented and and some someti time mess seem seemin ingl gly y undirected #ehavi #ehaviors ors.. The more more recent recent instr$ instr$cti ctions ons for identi identifyi fying ng infant infant D D 'trange;'it$ation res%onse +ere, ho+ever, #ased $%on thoro$gh st$dy of <00 infants K$dged to #e disor disorgan ganiEe iEed1d d1diso isorien riented ted,, incl$d incl$ding ing !00 infant infantss from maltre maltreatm atment ent and high;r high;ris& is& sam%les ain N 'olomon, this vol.. 8nfa 8nfants nts are K$dged K$dged to fit fit to the D hereafter, D hereafter, disorganiEed category +hen they sho+ strong or com#ined indices of disorganiEation or disorientation in the %resence of the %arent s$ch as freeEing of all movementM a%%roaching the %arent +ith head avertedM roc&ing on hands and &nees follo+ing an a#ortive a%%roachM moving a+ay from the %arent to the +all +hen a%%arently frightened #y the strangerM screaming for the %arent #y the door $%on se%aration, then moving silently a+ay $%on re$nionM or rising to greet the %arent on re$nion, and then falling %rone to the floor. ost of these #eha #ehavi vior orss a%%e a%%ear ar to #e of a ty%e ty%e etho etholo logi gist stss term term con confl flic ictt #eha #ehavi viors ors,Q ,Q that that is, is, #ehaviors that res$lt from the sim$ltaneo$s activation of incom%ati#le #ehavioral systems see, e.g., inde !"70. Ihile it is not $ncommon for an infant to sho+ some of these #ehaviors at lo+ levels of intensity in stress sit$ations in +hich the %arent is a#sent e.g., freeEing #riefly as the stranger a%%roaches d$ring a se%aration e%isode, +hen #ehaviors of this ty%e are seen at higher levels of intensity in the *arent2s *resence they *resence they a%%ear diffic$lt to eB%lain. /or eBam%le, $%on re$nion, an infant is held on motherFs la% for a long %eriod, d$ring +hich time the infant has a daEed eB%ression and re%eatedly t+ists its hair and raises hands to ears. Dr, an infant is crying lo$dly in the first moments of re$nion and attem%ting to gain the %arentFs la%, then s$ddenly freeEesQ in a given %ost$re for several sec;
8:/A:T -8'D4A:8@5- ATTAC5:T 'TATU'
!6)
onds. 8n s$ch cases a a the o#server cannot determine +hat, if anything, is ca$sing the distress, or b b +hy, if the infant +ished to gain the %arentFs la%, it sto%%ed moving in mid;activity. 'ome 'ome of these hese #eha #ehavi vio or %at %attern ternss s$c s$ch h as ster stereo eoty ty%i %ies es are are eB%e eB%ect cted ed in ne$rologically im%aired infants. There is no evidence to date, ho+ever, that D that D reflects any sta#le constit$tional deficiency on the %art of infants in normal sam%les. 8n the larger 9ay Area sam%le !*! infant;mother 'trange 'it$ations analyEed to date, only three o$t of thirty;fo$r infants K$dged D K$dged D +ith +ith one %arent +ere also K$dged D K$dged D +ith +ith the second %arent ain N 'olomon, this vol.. 8nde%endence of $nclassifia#leQ no+, D D atta attach chme ment nt stat stat$s $s +as +as also also fo$n fo$nd d #et+ #et+ee een n moth mother erss and and day;c day;care are careg caregiv iver erss HrentE !"(<. A relation #et+een the D the D category category and %arental #ehavior %atterns +as s$ggested in t+o inde%endent st$dies com%aring maltreating to +ell;matched control fami famili lies es.. 8n #oth #oth st$d st$die ies, s, the the D catego category ry +as fo$nd fo$nd strong strongly ly and s%ecif s%ecifica ically lly associated +ith maltreatment Carlson, Cicchetti, 9arnett, N 9ra$n+ald !"("M Lyons; $th, @oll, Connell, N Ddom !"(7. Ie $ndersc $nderscore ore here here that that the infant infantFFs D D 'trange;'it$ation res%onse in lo+;ris& sam%les s$ch as o$rs is not normally an indication of maltreatment. 8ndeed, +e +ill arg$e here for a ?$ite different, altho$gh related, mechanism. +o$$o"'u* tudies o/ 5nc$assi/iab$e or - In/ants in Lo"'Risk am*$es 8n o$r o$r $%%e $%%err;mid ;middl dlee;cl ;class ass 9ay Area rea sam sam%le, %le, chil childr dren en init nitial ially K$dge $dged d $nclas $nclassif sifia# ia#le le and and later later recode recoded d as disor disorgan ganiEe iEedQ dQ +ith +ith a %artic$ %artic$lar lar %arent %arent in infancy most often sho+ed contro$$ing role;reversing contro$$ing role;reversing res%onses to re$nion +ith that %arent at age 6, #eing generally either %$nitive to+ard the %arent on re$nion or else care caregi givi ving ng.. 8n addi additi tion on,, +hen +hen a chil child d had had #een #een K$dg K$dged ed D $nclassi $nclassifia fia#le #le +ith +ith a %artic$lar %arent in infancy, %arent;child disco$rse for that dyad +as K$dged dysfl$ent five five year yearss late laterr, +hil +hilee disc disco$ o$rs rsee +ith +ith the the seco second nd %are %arent nt refl reflec ecte ted d the the origi origina nall atta attach chme ment nt clas classi sifi fica cati tion on +ith +ith the the seco second nd %are %arent nt #ein #eing, g, for for eBam eBam%l %le, e, fl$e fl$ent nt associated associated +ith gro$% ! infancy classifications or fl$ent;restricted associated +ith gro$% ( gro$% ( infancy infancy classificationsM 'trage N ain !"(). /inally, D D attachment stat$s +ith +ith moth mother er in infan infancy cy %red %redic icte ted d a%%a a%%aren rentt diff diffic ic$l $lti ties es +ith +ith thin thin&i &ing ng regard regardin ing g attachment;related events th$s fearf$l, disorganiEed, and contradictory or irrational; seeming tho$ght %rocesses regarding %arent;child se%arations a%%eared in children +ho had #een classified D classified D five five years %revio$sly Ha%lan !"(7. 'imilarly, in a Charlottesville st$dy, Cassidy fo$nd negative self;descri%tions, and #iEarre or violent descri%tions of interactions #et+een a child and mother doll in attachment attachment;relat ;related ed sit$ations sit$ations +here 6;year;olds 6;year;olds +ere K$dged K$dged contro$$ing in contro$$ing in their D; D; e?$ivalent re$nion res%onse to the mother Cassidy !"(6, !"((. 'till more recently, 4eorge and 'olomon !"(" fo$nd that #oth a motherFs narrative descri%tion of herself as hel%less +ith res%ect to the child,
!66
A8: A:- 'DLDD:
and her %erce%tion of the child as o$t of her control +ere strongly correlated +ith her 6;year;oldFs 6;year;oldFs controlling res%onse to re$nion +ith her in the la#oratory. la#oratory. LACH D/ 5'DLUT8D: D/ DU:8:4 /D ATTAC5:T /84U5' LD'T TDU4 -5AT 5LAT5- TD 8:/A:T -8'D4A:8@5- 1-8'D85:T5ATTAC5:T 'TATU' The The asso associ ciat atio ion n #et+ #et+ee een n %are %arent ntal al lac& lac& of reso resol$ l$ti tion on of mo$r mo$rni ning ng for for lost lost attachment fig$res and infant D infant D attachment attachment stat$s +as discovered in conK$nction +ith a siBth;year follo+;$% follo+;$% st$dy of o$r 9ay Area sam%le ain !"(3, !"()M see ain N Iesto eston n !"(! !"(!,, and and ain ain et al. al. !"() !"() for for a more more com% com%le lete te desc descri ri%t %tio ion n of the the chara charact cteri erist stic icss of this this sam% sam%le le. . 8n this this st$d st$dy y, +e com% com%ar ared ed the the tran transc scri ri%t %tss of a motherFs disc$ssion of her o+n attachment history +ith her childFs 'trange;'it$ation classification d$ring infancy five years earlier. 8n the larger sam%le from +hich the s$#Kects s$#Kects for this st$dy +ere dra+n !(" dyads o#served over a fo$r;year fo$r;year %eriod, !*! infant;mother 'trange 'it$ations analyEed to this date, a maKority of children had #een #een clas classi sifi fied ed as ! sec$ sec$re re +ith +ith moth mother er,, and and only only !6R !6R had had #een #een consi conside dered red $nclas $nclassif sifia# ia#le1 le1dis disorg organi aniEed; Eed;dis disori orient ented ed ain ain N Ieston eston !"(!. !"(!. 8n contra contrast, st, this this re%ort is #ased $%on a selected s$#sam%le of fifty;three mothers and infants t+enty; fo$r girls, t+enty;nine #oys, selected to incl$de as many D infant;mothe infant;motherr dyads as +ere availa#le d$ring the %eriod of the st$dy seventeen, or 3
8:/A:T -8'D4A:8@5- ATTAC5:T 'TATU'
!67
alitiesM and ho+ it may have affected their res%onse to their child. /or the %$r%oses of the c$rrent st$dy, data concerning loss +ere categoriEed as follo+s loss of a close family mem#er older than the s$#Kect %rior to the com%letion of high schoolM later loss of older family mem#ers, and loss of any other im%ortant %ersons at any %oint in timeM no im%ortant loss eB%eriences. As noted earlier, Ad$lt Attachment 8ntervie+s are transcri#ed ver#atim, and each transcri%t is assigned to a single overall classification for the ad$ltFs state of mind +ith res%ect to attachmentQ irres%ective of %artic$lar relationshi%s and eB%eriences. 8n &ee%ing +ith the three;%art Ains+orth infant attachment classification system, o$r original ad$lt attachment classification system assigned each ad$lt to one of three categories, each e?$ivalent to and %redictive of infant categories sec$re1a$tonomo$s +ith res%ect res%ect to attachment attachment associated associated +ith the %arents of ! infants, dismissing of atta attach chme ment nt asso associ ciat ated ed +ith +ith the the %are %arents nts of ( infan infants ts, , or %reoc %reocc$ c$%i %ied ed #y %ast %ast attachment relationshi%s and eB%eriences associated +ith the %arents of C infants. 5arlier st$dies em%loying the Ad$lt Attachment 8ntervie+ also $sed this original, three;%art scoring system e.g., Ho#a& N 'ceery !"((M Cro+ell N /eldman !"((M ain N 4old+yn, in %ress. coring ub:ects /or Lack o/ Reso$ution o/ Mourning o/ Im*ortant +igures Lost through through Death A fo$rth fo$rth,, 5 $nresolvedQ $nresolvedQ ad$lt attachment category has no+ #een develo%ed, +hich is assigned to s$#Kects in conK$nction +ith scores a#ove ) on a nine;%oint scale assessing lac& of $nresolved loss descri#ed #elo+. Li&e infant D infant D attachment attachment stat$s, +hich is assigned in conK$nction +ith a #est;fitting, alternative, or forcedQ (, !, !, or C classificat classification ion so that the infant is act$ally classified classified D(, D!, or DC; or DC; see see ain N 'olo 'olomo mon, n, this this vol. vol., , ad$l ad$ltt $nre $nreso solv lved ed atta attach chme ment nt stat stat$s $s is also also assi assign gned ed in conK$n conK$ncti ction on +ith +ith a #est;f #est;fitt itting ing alterna alternativ tivee ad$lt ad$lt catego category ry.. Th$s, Th$s, an ad$lt ad$lt +ho is K$dged $nresolved +ith res%ect to tra$ma +ill #e classified as, for eBam%le, $nresol $nresolved ved1sec 1sec$re $re or $nreso $nresolved lved1di 1dismi smissi ssing ng of attach attachmen ment. t. 8nsofa 8nsofarr as %ossi# %ossi#le, le, se%arate scores are assigned to the disc$ssion of each individ$al lost. The transcri%t is assigned the highest score given to any loss eB%erience. 9o+l 9o+l#y #y !"( !"(0 0 $ses $ses the the term term mo$ mo$rni rning ngQQ to refe referr to all all the the %syc %sycho holo logi gica call %rocesses, %rocesses, conscio$s and $nconscio$s, $nconscio$s, that are set in train #y loss.Q 8n constr$cting constr$cting a scal cale intended to asse ssess $nres resolved loss oss, +e foc$sed on the conce%t of disorganiEation and disorientation in mental cognitive and affective %rocess as it might #e made evident in s%eech transcri%tions. Ie searched s%ecifically for signs of disor disorgan ganiEa iEatio tion n #oth #oth #eca$se #eca$se ! mental mental %rocess %rocesses es in an attach attached ed indivi individ$a d$all are normally organiEed +ith some contin$ing reference to attachment fig$reshence, loss of an attachment fig$re thro$gh death +ill inevita#ly lead to some disorientation and < #eca$se of several s%ecific
!6(
A8: A:- 'DLDD:
findings from the literat$re concerning res%onses to #ereavement. Th$s, among the disorganiEing and disorienting eB%eriences affecting the #ereaved in the first months s$cc s$ccee eedi ding ng a loss oss are are a necessa necessaril rily y incom% incom%let letee mental mental and #ehavio #ehavioral ral search search %rocesses, b b v$lnera#ility to a dis#elief that the loss has occ$rred or is %ermanent as in flee fleeti ting ng ill$ ill$si sion onss that that the the dead dead %ers %erson on is %rese %resent nt or is a%%ro a%%roac achi hing ng,, c eB%eriences of disorientation in %laces or sit$ations in +hich the dead fig$re had commonly #een fo$nd, and d d $nfo$nded fears of having #een ca$sal in the death itself 9o+l#y !"(0M Par&es !"7<, !"(0M a%hael !"(<. 8f, ho+ever, the o$tcome of mo$rning is favora#le, a reorganiEation of mental %rocesses +ill $ltimately ta&e %lace, leading to rene+ed orientation to the %resent enviro environme nment. nt. 8ndeed, 8ndeed, health healthy y mo$rni mo$rning ng is identi identifie fied d #y 9o+l#y 9o+l#y !"(0 !"(0 as the the s$ccessf$l effort of an individ$al to acce%t #oth that a change has occ$rred in his eBternal +orld and that he is re?$ired to ma&e corres%onding changes in his internal, re%rese re%resenta ntatio tional nal +orld and to reorgani-e, reorgani-e, and %erha %erha%s %s to reorient , his attachment #ehavior accordinglyQ italics o$rs. D$r %resent scale for assessing $nresolved loss from intervie+ transcri%ts ta&es acco acco$n $ntt of stat statem emen ents ts indi indica cati tive ve of cont contin in$a $all ment mental al diso disorrgani ganiEa Eati tion on and and disori disorient entati ation on as sho+n sho+n in ! la%ses la%ses in metaco metacogni gnitiv tivee monito monitorin ring g of reason reasoning ing %rocess %rocesses, es, < la%ses la%ses in metacog metacognit nitive ive monito monitoring ring of disco$ disco$rse rse %rocess %rocesses, es, or 3 re%orts of eBtremely disorganiEed or disoriented #ehavioral res%onses to a death +hen s$ch statements are not accom%anied #y convincing evidence of later s$ccessf$l resol$tion of mo$rning ain, -eoss, N esse !"(". La%ses in metacognitive monitoring of the ty%es to #e descri#ed a%%eared in ad$lts +ho +ere f$nctioning +ell in intellect$ally demanding conteBts, and in a recent analysis +e have s$ggested that these la%ses may res$lt from $neBamined early #eliefs regarding a %artic$lar loss or loss in general 9o+l#y !"(0, and1or from eB%eriences of fear and1or anBiety +hich occ$r d$ring the disc$ssion of the loss eB%erience ain N esse !"(". :ote that +e do not consider dismissingQ disc$ssions of a maKor loss eB%erience 8 K$st considered it a #lessing #y the time it ha%%ened. The neBt year, 8 started high scho school ol.Q .Q as indi indica cati tive ve of diso disorg rgan aniE iEat atio ion n and1 and1or or diso disori rien enta tati tion on.. ath ather er,, s$ch s$ch disc$ssions +o$ld seem to indicate failed mo$rningQthat is, that the %rocess of mo$rning has yet to #egin. 5Bam 5Bam%l %les es of a%%a a%%are rent nt la%s la%ses es in the the metaco metacogni gnitiv tivee monito monitorin ring g o/ reasoni easoning ng *rocesses incl$de *rocesses incl$de ! indications of dis#elief that the %erson is deadfor eBam%le, 8tFs really #etter for him that he is dead, #eca$se no+ he can go on #eing dead and 8 can ta&e care of things li&e he +anted.Q < /eelings of #eing materially ca$sal in a death +here no material ca$se is %resent;for eBam%le, 8 still thin& it might have ha%%ened #eca$se 8 +as so angry +ith her that '$nday that 8 K$st ho%ed she +o$ld die.Q 3 5fforts 5fforts to mani%$late mani%$late the mind so as to ignore the facts or im%licati im%lications ons of a deathfor eBam%le, 8 havenFt had to go #ac& there since he died, so most of the time 8 sort of
8:/A:T -8'D4A:8@5- ATTAC5:T 'TATU'
!6"
%retend heFs still there and 8 can al+ays call him $% if 8 +anted.Q :ote that indications of religio$s #elief in a meta%hysical life after death are not ta&en as la%ses in reasoning %rocesses. oreover, s$ch #eliefs are often stated as ho%es, or as matters of s%ec$lation, as, 8 li&e to #elieve that he is still +atching me from eaven.Q 8ndi 8ndica cati tion onss of a%%ar a%%aren entt $a*ses $a*ses in the metaco metacogni gnitiv tivee monito monitorin ring g o/ discour discourse se *rocesses ta&e ta&e the the form form of alte altere red d disc disco$ o$rse rse d$ri d$ring ng the the disc disc$s $ssi sion on of a deat death, h, s$ggesting that the individ$al has entered into a s%ecial state of mind in +hich orient orientati ation on to the $s$al $s$al conver conversat sation ional al strict strict$res $res are a#sent a#sent.. These These incl$d incl$de e ! Un$s$al attention to detailfor eBam%le, e died forty;t+o years ago last month, on :ovem#er 7, a onday, onday, and right #efore his forty;second #irthday. e +o$ld have #een #een eigh eighty ty;f ;fo$ o$rr, no, no, eigh eighty ty;f ;fiv ive, e, on this his :ove :ovem# m#er er (.Q (.Q < < Poet Poetic ic %hra %hrasi sing ng rhetorical1e$logistic s%eechfor eBam%le, in disc$ssing a death and not else+here +ithin the intervie+, the s$#Kect $ses e$logistic1rhetorical %hrasing, as, 'he +as yo$ng, she +as lovely, she +as dearly #eloved #y all +ho &ne+ her and +ho +itnessed her as she +as torn from $s.Q e%o e%ort rtss of e8treme e8treme disorgani-a disorgani-ation tion or disorientat disorientation ion in behavior behavior follo+ing a death are also ta&en as indices of $nresolved mo$rning, $nless accom%anied #y statements convincingly evidencing resol$tion. These incl$de re%orts of attem%ted s$icide and re%orts of redirected grief, for eBam%le, eBtreme res%onses to the death of a %et or a %$#lic fig$re follo+ing a#sence of res%onse to a lost attachment fig$re. As noted earlier, each s$#Kect is scored on a nine;%oint scale for lac& of resol$tion of mo$rning for each;individ$al lost thro$gh death. 'cores of 3 or less are assigned +hen the s$#KectFs disc$ssion of the loss indicates some contin$ing feeling of regret, and and cont contin in$i $ing ng or reem reemer ergi ging ng affe affect ctio ion n or sorr sorro+ o+,, +ith +itho$ o$tt acco accom% m%an anyi ying ng disorganiEation, for eBam%le, #rief tears in remem#ering a shared love of m$sic y father loved the violinSQ, or regrets eB%ressed for not having had a longer time together. atings of 7 are assigned to individ$als +ho indicate eBcessive fear or g$ilt s$rro$nding the loss, or eBtreme #ereavement reactions, or +hose s%eech #ecomes distinctly disoriented d$ring disc$ssions of the loss. /inally, ratings of " are assigned to indi indivi vid$ d$al alss +hos +hosee tho$ tho$gh ghtt %roc %roces esse sess rega regard rdin ing g the the lost lost fig$ fig$re re are are high highly ly disorganiEed or are in clear violation of the $s$al $nderstanding of %hysical reality and ca$sality. '$#Kects are assigned a rating of " +hen, for eBam%le, they fear having ca$sed the death of an attachment fig$re thro$gh their o+n tho$ght %rocesses d$ring childhood, or indicate s$#tly that the dead %erson is still #elieved to #e alive. :ote that scores indicative of definite $nresolved mo$rning in o$r system +o$ld also often #e indicative of frightening ideation having ca$sed the death of a loved %erson, or #elieving a %erson to #e #oth alive and not;alive, or of frightening or over+helming eB%eriences e.g., s$icide attem%ts.
!70
A8: A:- 'DLDD:
Inter:udge Re$iabi$ity, e8 Di//erences, Di//erences, and !irth 9rder 9rder atings assigned #y one K$dge Anitra -eoss +ere $sed for all fifty;three cases. To estimate interK$dge relia#ility, t+o K$dges esse and -eoss rated thirty cases #lind to one anotherFs ratings and to all information regarding the infant. 8nterrater relia#ility +as high r .(7. Tests for seB differences and #irth order yielded no significant res$lts. :either infant disorganiEation nor motherFs score for lac& of resol$tion of mo$rning +as related to the seB or #irth order of the child. Direct Re$ations Re$ations bet"een Mother3s Mother3s Loss E8*eriences and In/ant Disorgani-ation Disorgani-ation /ifteen mothers in this selected sam%le had lost an older family mem#er +ith +hom they had lived %rior to the com%letion of high school. :ine of these fifteen eB%eriencing early loss 60R had infants K$dged disorganiEed ! +ith them in the 'trange 'trange 'it$ation, 'it$ation, +hile only eight of the thirty;eight thirty;eight remaining mothers
/or o$r follo+;$% st$dies of #oth %arents and children in o$r original 9ay Area sam%le, +e rela relate te o$r o$r findi finding ngss to infa infant ntss +ho +ho +ere +ere both $nclas $nclassif sifia# ia#le le +ithi +ithin n the traditio traditional nal (, !, C classification system and disorganiEed. disorganiEed. !
8:/A:T -8'D4A:8@5- ATTAC5:T 'TATU'
!7!
them in the 'trange 'it$ation five years %revio$sly. The t+elfth infant had #een originally assigned to the eBtremely avoidant ( avoidant (! ! category, #$t +as %laced in the D' the D' e?$ivalent e?$ivalent contro$$ing category contro$$ing category +hen o#served again at 6 years. 8t is interesting to note that this mother descri#ed herself as having #een rethin&ing her early losses only recently. More Recent tudies Re$ating Mother2s Traumatic Traumatic E8*eriences to In/ant (ttachment tatus Ains+orth and 5ich#erg in %ress st$died an $nselected Charlottesville sam%le of forty;five forty;five infants and mothers, com%aring com%aring infant (, !, C , and D and D 'trange;'it$ation attachment classifications to motherFs Ad$lt Attachment 8ntervie+ res%onses a fe+ months later. The Ad$lt Attachment 8ntervie+ transcri%ts +ere scored for lac& of resol$tion of mo$rning #y ary Ains+orth +ho +as #lind to infant 'trange;'it$ation attachment stat$s. As in the 9ay Area st$dy, individ$als assessing infant 'trange; 'it$ation attachment stat$s +ere #lind to all other information regarding the dyad. As o%%osed to the 9ay Area st$dy, ho+ever, the Charlottesville sam%le +as $nselected for infant;mother attachment categories. !. 8n o$r o$r o+n o+n st$d st$dy y, +e fo$n fo$nd d moth motheerFs rFs lac& ac& of reso resoll$ti $tion of mo$ mo$rni rning significant significantly ly associated associated +ith the infantFs infantFs #est;fitti #est;fitting ng gro$% C insec$re;am#ivalent attachment attachment category category ain N esse !"(". Th$s, the mother +as mother +as significantly more li&ely li&ely to sho+ sho+ $nreso $nresolve lved d mo$rni mo$rning ng for lost lost attach attachmen mentt fig$res fig$res +hen the in/ant2s #est;fitting attachment category +as C than than +hen the #est;fitting category +as ( +as ( or or ! $sin $sing g a tradi traditi tion onal al (, !, C infant attachment category analysis. Ains+orth and 5ich#e 5ich#erg rg fo$nd fo$nd motherF motherFss $nreso $nresolve lved d mo$rni mo$rning ng signif significa icantl ntly y associ associate ated d +ith +ith her %lacem %lacement ent in the infant infant C ;e?$ivalent ;e?$ivalent %reocc$%ied %reocc$%ied ad$lt attachment attachment category category.. This s$ggests that ad$lts +hose state of mind +ith res%ect to attachment is %reocc$%ied may #e es%ecially v$lnera#le to loss and other tra$matic eB%eriences, and ma&es sens sensee of the the fact fact that that +e fo$n fo$nd d a high higher er %ro% %ro%or orti tion on of C th than ( or ! infants disorganiEed. <. /e+ Charlottesville mothers had eB%erienced early loss of a %arent none of these five had D had D infants. infants. Ihen loss +as more incl$sively defined to incl$de loss of im%o im%ort rtan antt %erso %ersons ns at any any %oin %oint, t, an assoc associa iati tion on +ith +ith infa infant nt D attachment attachment stat$s stat$s a%%eared as a trend #$t failed to reach statistical significance. T+elve o$t of thirty mothers mothers eB%erien eB%eriencin cing g loss loss had D D infants *0R, +hile three o$t of fifteen <0R mothers +ith no loss eB%erience had D had D infants infants P .<0. 3. /ifteen of the forty;five infants in the Charlottesville st$dy +ere assigned to D attach attachmen mentt stat$s. stat$s. ($$ ($$ eight mothers K$dged $nresolved +ith res%ect to loss of attachment fig$res $sing an earlier version of o$r scale had disorganiEed infants, +hile %ost;hoc %ost;hoc analyses indicated indicated that t+o more mothers of D infants D infants +o$ld have #een assigned to $nresolved stat$s $sing a more com%letely develo%ed scale ain et al. !"(". These findings s$%%ort o$r 9ay
!7<
A8: A:- 'DLDD:
Area st$dy ain !"(3, !"() in concl$ding that it is $nresolved mo$rning for loss of an attachment fig$re that is es%ecially e s%ecially strongly associated +ith infant D infant D attachment attachment stat$s and not the death as s$ch. *. /inally, /inally, for the mothers of five D five D infants infants in the Charlottesville st$dy, st$dy, there +ere no definite indications indications of $nresolved $nresolved mo$rning. mo$rning. 8n t+o of these five cases, ho+ever, ho+ever, infant D infant D attachment attachment stat$s stat$s +as %redicted %redicted in advance advance from other tra$mas tra$mas +hich +ere considered $nresolved. 8n one case, the mother had #een a#$sed #y her %arents. 8n a second second es%eci es%eciall ally y instr$ instr$cti ctive ve case, case, the mother mother had almost almost died died of an eBtrem eBtremely ely dangero$s disease +ith s$dden onset K$st %rior to the 'trange 'it$ation. /rom the motherFs disc$ssion of this close #r$sh +ith death, the K$dge Ains+orth considered the mother still $nresolved and again correctly %redicted infant D infant D attachment attachment stat$s. ece ecent ntly ly,, com% com%ari ariso son n #et+ #et+ee een n ad$l ad$ltt and and infan infantt atta attach chme ment nt stat stat$s $s has has #een #een eBtended to t+o %overty sam%les of #lac& and is%anic teenaged mothers living in inner;city :e+ Yor&, some of +hom eB%erienced %hysical or seB$al a#$se. 9oth st$die st$diess $sed $sed the ne+ infant infant and ad$lt ad$lt attach attachmen mentt classi classific ficati ation on system systems, s, that that is, incl$ding infant D infant D and and ad$lt $nresolved categories. 8n one st$dy, infant attachment stat$s +as fo$nd highly concordant (7.)R fo$r;category match +ith conc$rrent assessments of the teenage motherFs attachment organiEation Levine et al. !"("M in anot anothe herr, the the Ad$l Ad$ltt Atta Attach chme ment nt 8nte 8nterv rvie ie+ + +as +as admi admini nist stere ered d %rena %renata tall lly y, and and the the motherFs motherFs attachment attachment category category +as again fo$nd %redictive %redictive of the infantFs infantFs 6"R fo$r; category matchM Iard Iard et al. !"(". 8n #oth of these sam%les, as in the 9ay Area and Charlottesville sam%les, the moth motherF erFss $nres $nresol olve ved d stat stat$s $s +as +as fo$nd fo$nd s%ec s%ecif ific ical ally ly %redi %redict ctiv ivee of the the infa infant ntFFs disor disorgan ganiEe iEed1d d1diso isorien riented ted attach attachmen mentt stat$s stat$s.. These These adoles adolescen centt mothers mothers +ere +ere often often fo$nd $nresolved +ith res%ect to maltreatment eB%eriences. A history of %hysical or seB$al a#$se in %arents of some D infants D infants had also #een noted in a small n n !) st$dy of #lind and %artially sighted infants /riedman !"(6. :ot only $nresolved loss of attachment fig$res, #$t also $nresolved a#$se and other recent $nresolved tra$matic eB%eriences a%%ear to #e lin&ed to D to D attachment attachment stat$s in o$r o$r in;% in;%ro rogre gress ss anal analys ysis is of a ne+ ne+ 9ay 9ay Area Area sam% sam%le le ai ain, n, Ha%l Ha%lan an,, N esse esse,, $n%$#lished data. A:ALY'8' D/ T5 -8=5'5 AAY D/ -8'D4A:8@5- 1 -8'D85:T5- 8:/A:T 95A=8D' 8' /5A A:-1D 8:898T8D: D/ ATTAC5:T 95A=8D 8:=DL=5-G Consideration of the #ehavior of infants %laced in the D category D category leads nat$rally to s%ec$lation regarding #oth the immediate ca$se of the #ehavior o#served +ithin the 'trange 'it$ation e.g., +hy the infant falls %rone at a %artic$lar moment and the more general, ontogenetic ca$se e.g., eB%eriences in interaction +ith the %arent +hich have led the infant to develo% this
8:/A:T -8'D4A:8@5- ATTAC5:T 'TATU'
!73
%artic$lar %attern of res%onse. As the reader is already a+are, +e +ill shortly s$ggest that frightened and1or frightening %arental #ehavior may %rovide the lin& #et+een the %arentFs eB%erience of $nresolved tra$ma and the infantFs disorganiEed1disoriented #ehavior as eBhi#ited in the 'trange 'it$ation. ere +e eBamine some of the diverse forms of D of D #ehaviors #ehaviors o#served. Dne salient theme that r$ns thro$gh the disorganiEe disorganiEed1dis d1disorient oriented ed #ehavior #ehavior of D infa infant ntss is that of contradiction or inhibition o/ action as it is being undertaken undertaken indeed, an $ndermining of action +hich occ$rs almost as soon as action is initiated. 8n many cases, this a%%ears to #e an inhi#ition or contradiction of action #eg$n s%eci s%ecifi fica call lly y +ith +ith res%e res%ect ct to atta attach chme ment nt.. Th$s Th$s,, imme immedi diat atel ely y $%on $%on the the %aren %arentF tFss a%%earance in the door+ay, the infant orients, then %laces hand to mo$thM or rises, then falls %roneM or cries, calls, and eagerly a%%roaches the %arent, then t$rns a#o$t and avoids the %arent +ith a daEed eB%ression. eB%ression. Later, Later, in the same e%isode, the infant may a%%roach the %arent o#li?$ely or +ith head avertedM cling to the %arent +hile averting gaEeM cry, +hile moving a+ay from the %arentM ma&e hesitant, sto%;start movements on a%%roach to the %arentM or $nderta&e movements of a%%roach +hich have a slo+, lim%, $nder+aterQ ?$ality as tho$gh #eing sim$ltaneo$sly inhi#ited or contradicted. 'everal of these #ehavior %atterns s$ggest that the %resence of at least a limited fear may act to inhi#it or contradict movements of a%%roach or %roBimity see&ing. /ear may also %lay a role in freeEing,Q a fre?$ently a%%earing disor disorgan ganiEe iEed1d d1diso isorien riented ted #ehavi #ehavior or also also eBhi#it eBhi#ited ed #y some some animal animal s%ecie s%eciess +hen +hen frightened. The infant +ho stills or freeEes in moments of distress im%resses the o#server o#server as having no alternative alternative sol$tion sol$tion neither a %erson, %erson, nor a location, to +hich flight may #e ta&en. Ihen the %arent is %resent, this seems an anomalo$s %attern. There are also more direct mar&ers of fear among the #ehaviors +hich +e have la#eled disorganiEed.Q A%%rehensive movements and eB%ressions are mar&ers for the D categor category y, as are fearf$l fearf$l eB%ress eB%ression ions, s, eBtrem eBtremely ely tentat tentative ive a%%roac a%%roaches hes,, moving moving s+iftly a+ay immediately $%on or K$st follo+ing a%%roach to the %arent, and tense, vigilant vigilant #ody %ost$res. D$r im%ression im%ression is, then, that - behavior o/ten invo$ves the start , /o$$o"ed by the inhibition o/ an attachment se
!7*
A8: A:- 'DLDD:
may #e more indicative of conf$sion than of fear, since the infant may #e res%onding eith either er to o#vi o#vio$ o$sl sly y confl conflic icti ting ng sign signal als, s, or to sign signals als or event eventss it %erce %erceiv ives es as conflicting or conf$sing see HrentE !"(
8:/A:T -8'D4A:8@5- ATTAC5:T 'TATU'
!7)
fined as any of the follo+ing eB%eriences occ$rring %rior to the com%letion of high school loss of a close family mem#er thro$gh accident, loss of #oth %arents, seB$al a#$se #y a %arent, %hysical a#$se #y a %arent. Dnly nine of o$r fifty;three mothers re%orted events of this ty%e, +ith at least one mother eB%eriencing at least one of these tra$mas. These eB%eriences did a%%ear to #e associated +ith $nresolved mo$rning. /ive of the t+elve mothers *!R +ere K$dged to eB%erience definitely $nresolved mo$rning, #$t only fo$r among the remaining forty;one +omen !0R had eB%erienced tra$mas of this &ind. D$r o+n informal and non#lind o#servations of the #ehaivor of the %arents of D disorganiEed infantsas o%%osed to the %arents of organiEed infantsd$ring the 'trange 'it$ation tend to %rovide s$%%ort for o$r hy%othesis that the %arent of the D infant may #e frightening or frightened. 8n these informal o#servations, +e have noted the follo+ing %atterns, each of +hich seems to $s li&ely to frighten an infant, either #y #eing directly threatening or #y indicating fright on the %art of the %arent. ost ost of the more more fright frighteni ening ng %attern %atternss listed listed here have have not yet #een noted in o$r st$dies of lo+;ris& sam%les and are instead ta&en from videota%es of maltreatment and high;ris& dyads. 5nusua$ =oca$ atterns These incl$de ! sim$ltaneo$s voicing and de;voicing intonation es%ecially d$ri d$ring ng greet greetin ing g or +hen +hen init initia iati ting ng %hys %hysic ical al conta contact ct lead leadin ing g to an omin omino$ o$s, s, or ha$nted,Q tone or effect. Th$s, the %arent may greet the infant +ith a sim$ltaneo$sly voiced and de;voiced i.Q This is a #reathy, eBtended, falling intonation +hich can #e recreated #y saying iQ +hile %$lling in on the dia%hragm. < ParentFs voice has s$dden mar&ed dro% in intonation intonation to dee% or lo+ %itch. Ihen mar&ed, mar&ed, s$ch changes are startling, es%ecially +hen the s%ea&er is a +oman +hose %itch and intonation s$ddenly seem to #elong to a male s%ea&er. 5nusua$ Movement atterns These incl$de ! %arent s$ddenly moves o#Kect or o+n face very close to infantFs face loomingQ. < ParentFs movements or %ost$res are %art of a %$rs$it se?$ence. 3 Parent %resents conflicting signals #y, for eBam%le, calling infant +hile standing +ith hands on hi%s and nec& and chin K$tted for+ard in a threatening %ost$re. * Un%redicta#le invasions of the infantFs %ersonal s%ace, as the %arentFs hands s$ddenly sliding from #ehind or across the infantFs face or throat. ) ParentFs handling of the infant s$ggests s$ggest s eBtreme timidity. timidity. 6 Parent is eBtremely res%onsive to any indications of reKection on the %art of the infant, as, for eBam%le, sl$m%s and folds %revio$sly eBtended hands on la% and +aits, foc$sed on infant +ith a %leading loo&, +hen infant ma&es an im%atient gest$re.
!76
A8: A:- 'DLDD:
5nusua$ *eech Content Thes Thesee incl incl$de $de ! ! %aren %arentt im%l im%lie iess that that infa infant ntFFs acti action onss co$l co$ld d have have harm harmf$ f$ll conse?$encesa conse?$encesa Yo$Fll &ill that little st$ffed #ear if yo$ do thatSQ b U$$ohhS /rightened inta&e of #reath as infant moves toy car across #are floor. 4onna have an accide accidentS ntS 5very#o 5very#odyF dyFss gonna gonna get &illed &illedSQ SQ < '$dden '$dden initia initiatio tion n of games games +ith +ith a frightening s%eech content, if accom%anied #y an $n$s$al, frightening %attern of movement and intonation8Fm gonna get yo$SQ 3 -irect indications of fear of the infant, as, for eBam%le, #ac&ing a+ay from the infant +hile directing the infant not to follo+ in a stammering, a%%rehensive voice-onFt follo+, d;donFt.Q 8t is self;evident that a maltreating %arent +ill #ehave in +ays that are frightening to an infant. Ihy, on the other hand, a %arent +ho dis%lays only frightened rather than threatening #ehavior +o$ld %rod$ce fear in an infant is less o#vio$s and is a to%ic that demands closer eBamination. As stated a#ove, it seems li&ely that s$ch ?$al ?$alit itie iess of %aren %arenta tall #eha #ehavi vior or are are often often rela relate ted d to the the %are %arent ntsF sF o+n o+n tra$ tra$ma mati ticc eB%eriences rather than to some as%ect of the ongoing %arent;infant interaction. 8f this is the case, these %arental #ehaviors are no do$#t all the less com%rehensi#le to the infant and most li&ely seem not only $n%redicta#le as %atterns of #ehavior #$t also ineB%lica#le in origin. +urther Consideration o/ Dynamics 9ccurring )hen )hen the arent2s +rightened !ehavior Resu$ts /rom /rom ast Traumas Traumas Ie %res %res$m $mee that that the the infa infant nt is e?$i e?$i%% %%ed ed +ith +ith #eha #ehavi vior oral al syst system emss +hic +hich h contin$o$sly monitor the environment as +ell as the accessi#ility of the attachment fig$re. fig$re. -e%ending -e%ending on the circ$mstances, circ$mstances, c$es to danger +ithin +ithin the environment environment +ill then trigger inhi#ition of action a#o$t to #e ta&en, flight, and1or attachment #ehavior. The %arentFs o#served eB%ressions of fear +ill %rovide one s$ch c$e, and, in the $s$al case, case, these these eB%ress eB%ression ionss +ill +ill a alert the infant to some o#serva#le danger in the immedi immediate ate enviro environme nment, nt, for eBam%l eBam%le, e, an a%%roac a%%roachin hing g animal animal,, or an a%%aren a%%arently tly dangero$s edge or s$rfaceM b b lead to changes in the infantFs #ehavior +hich in some +ay +ay effec effecti tive vely ly red$ red$ces ces its its dang danger erMM and and c #e ra%i ra%idl dly y s$cc s$ccee eede ded d #y %rote %rotect ctiv ivee %arental action, th$s red$cing the infantFs infantFs state of alarm. Under these more $s$al circ$mstances, the so$rce of the danger to the infant is in fact eBternal to #oth infant and %arent. Ihen, ho+ever, the so$rce of the dangerQ leading to %arental eB%ressions of fear or anBiety is internal to the still;tra$matiEed %arent, a a the %arent may not #e oriented to any o#vio$sly frightening as%ect of the environment, or b b may #e oriented to an as%ect of the environment +hich has #ecome associated +ith her history and internal state #$t is not intrinsically alarming, and c c is at any rate $nli&ely to ta&e satisfactory %rotection action. 8n this case incom*rehensib$e #eca$se, !. The %arentFs frightened #ehavior +ill #e incom*rehensib$e #eca$se,
8:/A:T -8'D4A:8@5- ATTAC5:T 'TATU'
!77
stemming from internal factors, its so$rce +ill #e either a indiscerni#leM or b b if discerni#le d$e to the %arentFs orientation it +ill #e something that the %arent has associated associated +ith her o+n tho$ght tho$ght %rocessesfor %rocessesfor eBam%le, eBam%le, a %hotogra%h %hotogra%h +ith remote associations to her earlier tra$ma. <. D+ing to the fact that +hat is %rod$cing fear in the %arent is attachment; related, the infant may occasionally #ecome conf$sed in the %arentFs mind either +ith the %arent herself or +ith those attachment fig$res +hose death or #ehavior %rod$ced the original tra$ma. 8n these cases+hen, for eBam%le, the %arent fears having #een ca$sal in the death of an attachment fig$rethe %arent may indicate to the in/ant that the in/ant itse$/ is the source o/ the a$arm . 3. Adding to these %otential com%lications is a nat$ral tendency to+ard flight in individ$als eB%eriencing fear. ence, the %arent a a may indicate a desire desire to get a"ay /rom the immediate environment andor the in/ant +hich in/ant +hich +o$ld greatly intensify the infantFs state of fear, or instead b b may even subt$y indicate indicate a *ro*ensity *ro*ensity to /$ee to the in/ant as a haven o/ sa/ety. sa/ety. 8t a%%ears, therefore, that many of the li&ely se?$elae to a %arent #eing alarmed #y an internal so$rce +o$ld lead to conf$sion and fear on the %art of the infant, and hence to disorganiEation and disorientation #elo+. -8'D4A:8@AT8D:, -8'D85:TAT8D: A:- T5 ATTAC5:T 95A=8DAL 'Y'T5 To this %oint, +e have sho+n that ! an intervie+;#ased method of assessing $nresolved tra$ma has #een develo%ed +hich relies chiefly $%on a%%arent la%ses in the metacognitive monitoring of reasoning or disco$rse d$ring disc$ssions of loss or other attachment attachment;relat ;related ed tra$ma tra$ma< < < to %red %redic ictt infa infant nt D attachment attachment stat$s. igh scores for $nresolved tra$ma on the %art of the %arent have #een fo$nd %redictive of infant D infant D attachment attachment stat$s in t+o +hite middle;class sam%leso$r original 9ay Area sam%le ain N esse !"(", and the Charlottesville sam%le st$died #y Ains+orth and 5ich#erg in %ress. As this cha%ter goes to %ress, the relation #et+een infant D attachment stat$s and $nresolved ad$lt attachment stat$s 3 has #een re%licated in t+o ado;
2
Until very recently, researchers +ere advised to $se the indices of disorganiEation and disorientation in tho$ght %rocesses d$ring disc$ssions of a loss in order to identify $nresolved tra$ma of other &inds. Ie have no+ com%leted a draft of a se%arate scale for assessing $nresolved eB%eriences eB%eriences of %hysical a#$se, and +e are a%%lying it to a ne+ 9ay Area sam%le of infants and mothers. 3
8n #oth the 9ay Area and Charlottesville sam%les, almost all infants +hose mothers +ere K$dged $nresolved $nresolved had infants K$dged disorganiEed disorganiEed,, #$t the D attachment D attachment stat$s of a minority of infants in each sam%le remained $neB%lained. Ihile some instances co$ld conceiva#ly #e acco$nted for #y tra$mas not descri#ed #y the %arent, +e $nderscore that in all %ro#a#ility other factors are o%erating in a s$#stantial minority of D of D cases cases for eBam%le, ne$rological diffic$lties, conf$sion regarding %arental signaling or some other as%ect of the eB%erimental %roced$re, and t em%orary stresses of other &inds.
!7(
A8: A:- 'DLDD:
lescen lescentt %overt %overty y sam%le sam%less and infant infant disorg disorgani aniEed1 Eed1dis disori orient ented ed 'trang 'trange;'i e;'it$a t$atio tion n res%onse has #een %redicted %rior to the #irth of the child Levine et al. !"("M Iard et al. !"(". A#ov A#ove, e, +e s$gg s$gges este ted d a %oss %ossi# i#le le rela relati tion on #et+ #et+ee een n $nre $nreso solv lved ed tra$m ra$maa and and frightened frightened and1or frightening frightening #ehavior on the %art of the %arent. Ie Ie concl$de +ith a disc$ssion of ho+ s$ch #ehaviors may #e related to infant D infant D attachment attachment stat$s. 9o+l#y 9o+l#y 2!"6" 2!"6" !"(< !"(< has descri descri#ed #ed the attach attachmen mentt #ehavi #ehaviora orall system system as a develo%ing in all gro$nd;living %rimates raised in any #$t highly a#normal rearing conditions, b b leading to a foc$s $%on maintaining %roBimity to one or a fe+ selected caregivers, and c c %roviding the infant +ith the chief #ehavioral means #y +hich it can ass$re its o+n s$rvival. The system is %res$med to ta&e contin$al acco$nt of thes thesee envir environ onme ment ntal al and and inte interna rnall cond condit itio ions ns that that +o$l +o$ld d norm normal ally ly lead lead to the the acti activa vati tion on or termi termina nati tion on of dis% dis%la lays ys of atta attach chme ment nt #ehav #ehavio ior ran and d henc hencee to #e contin$ally conte8t'sensitive. 8n conte8t'sensitive. 8n general, the system is organiEed to increase infant; caregiver %roBimity in sit$ations of threat or alarm, #$t to %ermit activation of eB%loratory #ehavior +hen the sit$ation is not threatening and %roBimity is ass$red. Among the conditions %rovo&ing alarm are ! $nfamiliar environments and < threatened or act$al se%arations from the attachment fig$re. 8n considering considering individ$al individ$al differences differences in %atterns %atterns of attachment attachment as o#served o#served in the 'trange 'it$ation, it is +orth+hile to note that the f$nctioning of the attachment #ehavioral system as descri#ed a#ove is most readily o#serva#le in infants K$dged secure +ith +ithin in the the 'tran 'trange ge 'it$ 'it$at atio ion. n. Th$s Th$s,, a rela relati tive vely ly smoo smooth th and and orga organi niEe Eed d transition #et+een eB%loratory #ehavior and attachment #ehavior is o#served in the res%onse of these and only these infants, +ho eB%lore the environment so long as %roBimity %roBimity to the %arent is ass$redM ass$redM t$rn their attention attention to attachment attachment +hen %roBimity %roBimity is threatenedM threatenedM and, follo+ing follo+ing a dis%lay dis%lay of attachment attachment #ehavior #ehavior directly directly $%on re$nion re$nion +ith the %arent, ret$rn once more to eB%loration. The ( The ( and and C infant infant 'trange;'it$ation res%onses also a%%ear relatively +ell;organiEed, altho$gh more com%leB and conceiva#ly more v$lnera#le to disorganiEation than the sec$re attachment %atterns. At a theoretical level, +e $nderstand the avoidant infant to #e minimi-ing the minimi-ing the dis%lay of attachment #ehavior relative to the level of activation of the atta attach chm ment ent #eha #ehavi vior oral al syst system em,, #y reducin educing g reactivi eactivity ty to /ear' /ear'e$i e$icit citing ing 0attachment'e$iciting1 cues cues 9o+l#y !"(0M ain, in %ress, and !"(!. 9ehavioral organiEation is then maintained thro$gh a shift in attention a+ay from the %arent +hich is largely accom%lished thro$gh eB%lorationa strategy that may also serve to maintain +hat;
8:/A:T -8'D4A:8@5- ATTAC5:T 'TATU'
!7"
ever %roBimity to the %arent seems %ossi#le ain !"(!M ain N Ieston !"(<. * The insec$re;am#ivalent infant also a%%ears organiEed in its %ersistent foc$s $%on the careg caregiv iver er.. 5B%r 5B%res essi sing ng alarm alarm over over even even rela relati tive vely ly $nal $nalarm armin ing g even events tsh hen ence ce,, ma8imi-ing the dis%la dis%lay y of attach attachmen mentt #ehavi #ehavior orthi thiss seemin seemingly gly hy%erv hy%ervigi igilan lantQ tQ infa infant nt dra+s dra+s or atte attem% m%ts ts to dra+ dra+ the the atte attent ntio ion n of an $n%r $n%red edic icta ta#l #ly y res% res%on onsi sive ve caregiver. The factors leading to the %rod$ction of the D the D %attern %attern of #ehavior are inherently com%leB. com%leB. A com%rehens com%rehensive ive disc$ssion +o$ld re?$ire ! a more detailed detailed foc$s $%on the role of conflict in infant res%onses to the 'trange 'it$ation, and relatedly < consideration of the n$m#er and nat$re of the %arameters +hich serve to control res%onses for infants assigned to differing attachment classifications. The 'trange; 'it$ation #ehavior of the sec$re infant may, for eBam%le, #e $nder the control of caregiver location only and therefore involve little or no conflict. The #ehavior of the avoidant avoidant and am#ivalent am#ivalent infants infants may #e additional additionally ly controlled #y the %ast #ehavior #ehavior of the attachment fig$re, so that for these infants li&ely caregiver res%onse as +ell as caregiver location m$st contin$ally #e ta&en into acco$nt ain, in %ress. 8n that their caregivers have ty%ically failed to %rovide negative terminating feed#ac& to attach attachmen mentt #ehavi #ehavior or in the home home sit$at sit$ation ion,, #eing #eing either either reKect reKecting ing of attach attachmen mentt #eha #ehavi vior or Ain Ains+ s+ort orth h et al. al. !"7( !"7(MM re%l re%lic icat ated ed in t+o t+o f$rt f$rthe herr sam% sam%le less #y ain ain N 'tadtman !"(! or $n%redicta#ly res%onsive Ains+orth et al. !"7(, the #ehavior of these infants in the more stressf$l 'trange 'it$ation may #e in %art the %rod$ct of conflicta conflict that can nonetheless %rod$ce relatively organiEed o$tcomes as the avoidant infantFs eB%loration, +hich may #e a form of dis%lacement #ehavior. ) /inally, the more dramatically conflicted #ehaviors a%%earing in disorganiEed infants $n%redicta#le #oth in timing and formmay seem chaotic %recisely #eca$se of a history of frightening eB%eriences +hich #elo+ %rovide miBed %ositive and negative feed#ac& to the attachment #ehavioral system. A com%lete disc$ssion of the origin and significance of %attern D #ehavior D #ehavior +o$ld involv involvee sim$lt sim$ltane aneo$s o$s eBamin eBaminati ation on of the #ehavi #ehavior or of (, !, !, and C infants and is #eyond the sco%e of this cha%ter. cha%ter. Previo$sly, ho+ever, ho+ever, +e had 4
The avoidant %attern may #e v$lnera#le to #rea&do+n $nder increased stress, ho+ever, at +hich time the %revio$sly avoidant infant may a%%roach the %arent in a disorganiEed disorganiEed manner as see Ains+orth Ains+orth et al. !"7( descri%tio descri%tion n of diffic$lty diffic$lty;to;cl ;to;classi assify fy %roBimity %roBimity see&ing in %attern %attern ( infants s$#Kected to a second 'trange 'it$ation. 5
To many %eo%le acc$stomed to o#serving infant 'trange;'it$ation #ehavior $sing the three; %art category system, the am#ivalent %attern seems m$ch the least +ell organiEed. This may #e in %art #eca$se a maKority of C infants infants are in fact disorganiEed ain N esse !"(" #$t may also res$lt res$lt from from o#serv o#servati ations ons of the infant infant +ho first first see&s, see&s, then then angril angrily y reKec reKects, ts, the mother motheran an a$ternation in a$ternation in #ehavior %atterns +hich is one %ossi#le res$lt of conflict. 8f, 8f, ho+e ho+eve verr, +e cons consid ider er the the C nondi nondisor sorga ganiE niEed ed infant infant as ma8imi-ing of attachment attachment #ehavior, this alternation may a%%ear less disorganiEed. 9o+l#y !"73 s$ggests that, inside of some relationshi%s, angry #ehavior is $ltimately %roBimity %romoting.
!(0
A8: A:- 'DLDD:
dra+n attention to the conflict eB%erienced #y the infant +ho is %hysically reKected #y the attachment fig$re ain !"(!M ain N Ieston !"(<. Ie s$ggested that this insec$re;avoidant infant is theoretically ca%t$red +ithin a %ositive feed#ac& loo% in +hich reKection aro$ses attachment #ehavior +hich, reKected again #y the attachment fig$re, leads to still f$rther aro$sal of the attachment #ehavioral system. Ie Ie descri#ed this sit$ation as inherently disorganiEing, #$t +e s$ggested that the reKected infant esca% esca%es es #eha #ehavi viora orall diso disorg rgani aniEa Eati tion on thro thro$gh $gh a shif shiftt in atte attent ntio ion n a+ay a+ay from from the the attachment fig$re and from attachment;eliciting changes in the environment. This shif shiftt in atte attent ntio ion n can can %ro#a %ro#a#l #ly y only only #e acco accom% m%li lish shed ed,, ho+e ho+eve verr, +hen +hen a the eB%erience of %arental reKection has #een relatively consistent and nonthreatening and b +hen +hen the the cond condit itio ions ns in the the s$rro s$rro$n $ndi ding ng envi environ ronme ment nt are only only mode moderat ratel ely y stressf$l. 8f +e are correct regarding the $n%redicta#ly frightening #ehavior %atterns sho+n on occa occasi sion on #y stil still; l;tr tra$ a$ma mati tiEe Eed d %aren %arents ts,, then then +e can can see see that that thei theirr infa infant ntss are are confronted +ith an inherently %er%leBing set of circ$mstances. 8n contrast to #oth avoidant and am#ivalent infants;+ho may #e frightened #y diffic$lties in o#taining caregiver caregiver res%onsiveness res%onsiveness in stressf$l stressf$l sit$ationsth sit$ationsthee fear the D D infant eB%eriences stems from the %arent as its so$rce. Placed in the 'trange 'it$ation in the com%any of a %arent +hose #ehavior has #een $n%redicta#ly frightening, then, the D the D infant infant may at moments moments $navoida#ly $navoida#ly eB%erience eB%erience an alarm s$fficiently s$fficiently intense that the activation activation of atta attach chme ment nt #eha #ehavi vior or cann cannot ot #e syst systema emati tica call lly y cont control rolle led. d. Ihen Ihen the the effec effects ts of atta attach chme ment nt;el ;elic icit itin ing g c$es c$es cann cannot ot #e mini minimi miEe Eed d as they they a%%a a%%aren rentl tly y can can for for the the avoidant avoidant infant, increasing %roBimity %roBimity to the attachment attachment fig$re is the $s$al sol$tion. sol$tion. /or /or the the infan infantt +ho +ho has #een #een alar alarme med d $n%r $n%red edic icta ta#l #ly y #y the the atta attach chme ment nt fig$ fig$re, re, ho+ever, a%%roach to the %arent may #e occasionally interr$%ted #y dis%lays of disorganiEed1disoriented #ehavior %recisely #eca$se memories involving fear of the %arent have momentarily #ecome aro$sed. 8n concl$sion, it seems a%%arent that frightening #ehavior on the %art of the still; tra$matiEed %arent sho$ld lead to disorganiEed1disoriented infant #ehavior, since the infant is %resented +ith an irresolva#le %aradoB +herein the haven of safety is at once the so$rce of the alarm. oreover, the conflict #et+een o%%osing tendencies to a%%roach and to flee from the attachment fig$re stems from a single eBternal signal thr threa eate teni ning ng or fear fearf$ f$ll %are %arent ntal al #eha #ehavi vior orM M is inte intern rnal al to the infa infant ntMM is self self;; %er%et$atingM and is eBacer#ated #y %lacement in a stressf$l sit$ation. 5/55:C5' Ains+orth, . -. '.M 9lehar, . C.M Iaters, 5.M N Iall, '. !"7(. atterns !"7(. atterns o/ attachment7 ( *sycho$ogica$ study o/ the strange situation. illsdale, situation. illsdale, :.. 5rl#a$m. Ains+orth, . -5. '., N 5ich#erg, C. 4. in %ress. 5ffects on infant;mother attachment of
8:/A:T -8'D4A:8@5- ATTAC5:T 'TATU'
!(!
motherFs $nresolved loss of an atta chment fig$re or ot her tra$matic eB%erience. 8n P. arris, . 'tevenson;inde, N C. Par&es, Attachment across the life cycle. :e+ Yor& Yor& o$tledge. American Psychiatric Association Committee on :omenclat$re !"(7. Diagnostic and statistica$ manua$ o/ o/ menta$ disorders, III'R. Iashington, III'R. Iashington, -.C. American Psychiatric Association. 9o+l#y, 9o+l#y, . 2!"6" !"(<. (ttachment !"(<. (ttachment and $oss7 =o$. =o$. >. (ttachment . :e+ Yor& 9asic 9oo&s. 9o+l#y, 9o+l#y, . !"73. (ttachment !"73. (ttachment and Loss7 =o$. =o$. ?. e*aration. :e+ e*aration. :e+ Yor& 9asic 9oo&s. 9o+l#y, 9o+l#y, . !"(0. (ttachment !"(0. (ttachment and $oss7 =o$. =o$. @. Loss, sadness sadness and de*ression. de*ression. :e+ Yor& 9asic 9oo&s. 9retherton, 8. !"(). Attachment theory etros%ect and %ros%ect. 8n 8. 9retherton N 5. Iaters eds.. Gro"ing *oints o/ attachment theory and research. research. Monogra*hs o/ the ociety /or Research in Chi$d Deve$o*ment )0!;<, )0!;<, 'erial :o. <0", 3;3). 9ro+n, 4.M arris, T.M N 9if$lco, A. A. !"(). Long;term effects of early loss of %arent. 8n . $tter, C. 8Eard, N P. ead eds.. De*ress eds.. De*ression ion in chi$dhood7 Deve$o*menta$ *ers*ectives. *ers*ectives. :e+ :e+ Yor& 4$ildford. Carlson, =.M =.M Cicchetti, -.M 9arnett, -.M N 9ra$n+ald, H. !"(". -isorganiEed1disoriented attachment relationshi%s in maltreated infants. Deve$o*menta$ infants. Deve$o*menta$ sycho$ogy <), sycho$ogy <), no. *, )<);)3!. Cassidy, Cassidy, . !"(), A%ril. Attachment and the self at siB. Pa%er %resented at the #iennial meeting of the 'ociety for esearch in Child -evelo%ment, Toronto. Toronto. Cassidy, Cassidy, . !"(6. Attachment and the self at siB. Un%$#lished doctoral dissertation, University of =irginia. =irginia. Cassidy, Cassidy, . !"((. Child;mother attachment and the self in siB;year;olds. Chi$d Deve$o*ment AB, AB, !
!(<
A8: A:- 'DLDD:
ain, . !"(!. Avoidance Avoidance in the service of attachment A +or&ing %a%er. %a%er. 8n H. 8mmelmann, 4. 9arlo+, 9arlo+, L. Petrinovich, N . ain eds., !ehaviora$ deve$o*ment7 The !ie$e/e$d interdisci*$inary *ro:ect %%. %%. 6)!;6"3. :e+ Yor& Yor& Cam#ridge University Press. ain, . !"(3. :e+ %ros%ects in the st$dy of attachment. Pa%er %resented at the ann$al meeting of the American Academy of Child Psychiatry, Psychiatry, 'an /rancisco. ain, . !"(), A%ril. An ad$lt attachment classification system 8ts relation to infant;%arent attachment. Pa%er %resented at the #iennial meeting of the 'ociety for esearch in Child -evelo%ment, Toronto. Toronto. ain, . in %ress. Cross;c$lt$ral st$dies of attachment organiEation ecent st$dies, changing methodologies, and the conce%t of conditional strategies. %uman Deve$o*ment. ain, ., N Cassidy, . !"((. Categories of res%onse to re$nion +ith the %arent at age siB Predicted from infant attachment classifications and sta#le over a one;month %eriod. Deve$o*menta$ %eriod. Deve$o*menta$ sycho$ogy <*, sycho$ogy <*, no. 3, *!);*<6. ain, .M -eoss, A.M N esse, 5. !"(". A system for assessing lac& of resol$tion of mo$rning from intervie+ transcri%ts. Un%$#lished man$scri%t, -e%artment of Psychology, Psychology, University of California, 9er&eley. ain, ., N 4old+yn, . in %ress. 8ntervie+;#ased ad$lt attachment classifications elated to infant; mother and infant;father attachment. Deve$o*menta$ attachment. Deve$o*menta$ sycho$ogy. sycho$ogy. ain, ., N esse, 5. !"(". 8ntervie+;#ased assessments of a %arentFs $nresolved $nresolved tra$ma are related to infant -Q attachment stat$s Lin&ing %arental states of mind to infant #ehavior o#served in a stressf$l sit$ation. '$#mitted man$scri%t, +or&ing title. ain, .M Ha%lan, :.M N Cassidy, Cassidy, . !"(). 'ec$rity in infancy, childhood and ad$lthood A move to the level of re%resentation. 8n 8. 9retherton N 5. Iaters Iaters eds., Gro"ing *oints o/ attachment theory and research. research. Monogra*hs o/ the ociety /or Research in Chi$d Deve$o*ment )0! )0! ;<,'erial :o. <0", 66; !0*. ain, ., N 'olomon, . !"(6. -iscovery of a ne+, insec$re;disorganiEed1disoriented insec$re;disorganiEed1disoriented attachment %attern. 8n T. 9. 9raEelton N . Yogman eds., (//ective deve$o*ment in in/ancy %%. in/ancy %%. ");!<*. :or+ood A#leB. ain, . N 'tadtman, . !"(!. 8nfant res%onse to reKection of %hysical contact #y the mother Aggression, avoidance and conflict. #ourna$ conflict. #ourna$ o/ the (merican (cademy o/ Chi$d sychiatry <0, sychiatry <0, <"<;307. ain, ., N Ieston, -. !"(!. The ?$ality of the toddlerFs relationshi% to mother and father. father. Chi$d Deve$o*ment )<, )<, "3<;"*0. ain, ., N Ieston, -. !"(<. Avoidance Avoidance of the attachment fig$re in infancy descri%tions and inter%retations. 8n C. . Par&es N . 'tevenson;inde eds., The *$ace o/ attachment in human behavior %%. behavior %%. 3!;)". London Tavistoc&. Par&es, C. . !"7<. !ereavement7 tudies o/ grie/ in adu$t $i/e. London Tavistoc&. Par&es, C. . !"(0. 9ereavement co$nseling -oes it +or&G !ritish Medica$ #ourna$ !, #ourna$ !, 7*0;7*3. O$inton, -., N $tter, . !"(). Parenting #ehavior of mothers raised in care. 8: A. . :ichol ed., Longitudina$ studies in chi$d chi$d *sycho$ogy and *sychiatry7 *sychiatry7 *ractica$ $essons /rom /rom research research e8*erience %%. e8*erience %%. !)7;<6!. Chichester and :e+ Yor& Iiley. ad&e;Yarro+ ad&e;Yarro+,, .M C$mmings, 5. .M H$cEyns&i, L.M N Cha%man, . !"(). Patterns of attachment in t+o; and three;year;olds in normal families and families +ith %arental de%ression. Chi$d Deve$o*ment )6, )6, ((*;("3. a%hael, 9. !"(<. The anatomy o/ bereavement . :e+ Yor& 9asic 9oo&s. 'trage, A., N ain, . !"(), A%ril. Parent;child disco$rse %atterns at 6 years %redicted from the organiEation of infant attachment relationshi%s. Pa%er given at the #i ennial meeting of the 'ociety for esearch in Child -evelo%ment, Toronto. Toronto. =ol&mar, /. /. .M oder, 5. L.M N 'iegel, A. 5. !"(0. -iscre%ant social comm$nications. Deve$o*menta$ comm$nications. Deve$o*menta$ sycho$ogy !6, sycho$ogy !6, no. ), *");)0). =ol&mar, /. ., N 'iegel, A. 5. !"7". Yo$ng childrenFs res%onses to discre%ant social comm$nications. #ourna$ o/ Chi$d sycho$ogy and sychiatry <0, !3";!*". Iard, .M Carlson, 9.M Altman, '. C.M Levine, L.M 4reen#erg, . .M N Hessler, -. 9. !"("a. Predicting infant;mother attachment from adolescentFs +or&ing models of relationshi%s. '$#mitted man$scri%t. Iard, .M Carlson, 9.M N Hessler, -. 9. !"("#. Adolescent Adolescent mother;infant attachment 8nteractions, relationshi%s and adolescent +or&ing models. '$#mitted man$scri%t.
