UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Child images of attachment figures and self Head, Tim 1991

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1991_A8 H42.pdf [ 6.34MB ]
Metadata
JSON: 831-1.0054637.json
JSON-LD: 831-1.0054637-ld.json
RDF/XML (Pretty): 831-1.0054637-rdf.xml
RDF/JSON: 831-1.0054637-rdf.json
Turtle: 831-1.0054637-turtle.txt
N-Triples: 831-1.0054637-rdf-ntriples.txt
Original Record: 831-1.0054637-source.json
Full Text
831-1.0054637-fulltext.txt
Citation
831-1.0054637.ris

Full Text

CHILD IMAGES OF ATTACHMENT FIGURES AND SELF by TIM HEAD B.A., The University of B r i t i s h Columbia, 1982 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS i n THE FACULTY OF GRADUATE STUDIES (Department of Mathematics and Science Education) We accept t h i s thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA September 1991 © Tim Head, 1991 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of September 24, 1991 The University of British Columbia Vancouver, Canada Date September 24, 1991 DE-6 (2/88) i i Abstract; Child Attachment-Related Conceptualizations  of Parents. Teacher and Self; Forced Choice Visual  Representations Along a Permitting/Blocking Access Dimension Twenty-three 5 to 7-year-old boys were shown drawings depicting themselves with t h e i r mother, father or main teacher i n one of 8 attachment-related s i t u a t i o n s . Subjects were primarily Caucasian and l i v e d i n a suburban neighbourhood i n Richmond, B.C. They were asked to select from 4 response categories the caregiver "most l i k e " t h e i r mom, dad or main teacher. Response categories were generated from attachment theory. Individual and group inner image p r o f i l e s were developed from the 552 selections. The variable " s i t u a t i o n " , but not the variable "caregiver", was determined by l o g l i n e a r analysis to be s i g n i f i c a n t beyond the .05 l e v e l (p. = .035). The childrens' selections were demonstrated to be relevant and non-haphazard within and across response categories. The analyses provide s i g n i f i c a n t support for the v a l i d i t y of the response categories and the c e n t r a l i t y of the underlying dimension of "permitting/blocking access". In addition, the study supports the notion of viewing main female teachers a f t e r 9 or 10 months with a c h i l d - as an attachment figure to that c h i l d . A f a i r l y generalized meta-structure of i n t e r n a l working models i s suggested by t h i s group p r o f i l e . F i n a l l y , t h i s study gives support to the notion of s i t u a t i o n a l s p e c i f i c i t y of caregiver response i i i under conditions of c h i l d attachment behavioural system ac t i v a t i o n . Table of Contents Abstract 11 L i s t of Tables v i L i s t of Figures v i i Acknowledgements i x Chapter 1: Statement of Problem 1 Rationale and Purpose of Study: General Considerations 3 Phenomenological Considerations 6 Developmental Considerations 8 Inclusion of Teachers as Attachment Figures...11 Background and Theoretical Basis of Problem 16 Chapter 2: Li t e r a t u r e Review: Overview 21 Important Dimensions of Attachment Theory 23 Cognitive and A f f e c t i v e Bias 26 Internal Working Models 35 Chapter 3: Methodology: Overview 48 Rationale for Design 50 The Subjects 54 The Interview Situation 57 The Permitting/Blocking Access Inventory 59 The Friendliness/Unfriendliness Inventory 61 The P i l o t Study 62 Chapter 4: Results: Overview 64 P i l o t Study Results 67 Study Results: Main E f f e c t s and Interactions Between Variables 68 Confidence Intervals 70 Internal Consistency 76 Analyses Based on a Subject's Choice of a Response Category One or More Times 79 Comparison of the Permitting/Blocking Access and Friendliness/Unfriendliness Inventories... 81 Subcategories Within the Ignoring Blocking Access Response Category.. 87 Test for Understanding 88 Chapter 5: Discussion: Interpretation of Cognitive, A f f e c t i v e or Response Bias i n the Results 89 General Discussion 94 Conclusion 103 Future Research Directions 105 References 107 Appendices: Appendix A: Drawings for Permitting/Blocking Access Inventory 120 Appendix B: Drawings for Friendliness/ Unfriendliness Inventory 217 L i s t of Tables Table 1: The Permitting/Blocking Access Inventory: Loglinear analysis 69 Table 2: Confidence l e v e l s based on 23 subjects 70 Table 3: Confidence l e v e l s based on 22 subjects 73 Table 4: Summarized data for a l l subjects i n both inventories 74 Table 5: Anova r e s u l t s within response categories 75 Table 6: P r o b a b i l i t i e s for response categories 79 Table 7: Significance of the d i s t r i b u t i o n across response categories for the f u l l version of the primary measure .82 Table 8: Significance of the d i s t r i b u t i o n across response categories for a reduced version of the primary measure 83 Table 9: Significance of the di s t r i b u t i o n s across response categories for a reduced version and the f u l l version of the secondary measure 84 Table 10: Summary of data from Friendliness/ Unfriendliness Inventory 86 Table 11: Situation by caregiver by response 100 v i i L i s t of Figures Figure 1: Comparison of Permitting/Blocking Inventory and Friendliness/Unfriendliness Inventory for a l l caregivers and others, and f o r mother, father only 81b Acknowledgements I would l i k e to extend my gratitude and appreciation to my advisor, Kip Anastasiou, whose considerable support helped make t h i s research study possible. I am also gra t e f u l to Peggy Koopman and John A l l e n for t h e i r warm support, suggestions and encouragement. Other contributors whom I would l i k e to acknowledge and thank are Stan Kita, Jim Enns, and Walter Boldt. F i n a l l y , while the art work speaks for i t s e l f , I g r a t e f u l l y acknowledge Barbara Clayden for her considerable a r t i s t i c contribution. 1 Chapter 1: Statement of Problem This i s an exploratory study designed to give a p r o f i l e of attachment-related inner images or expectations for a group of 2 3 f i v e to seven year-old boys. The subjects' school and homes are located i n a suburban c i t y near a large metropolitan area i n B.C., Canada. More s p e c i f i c a l l y , t h i s study examines aspects of the childrens' i n t e r n a l working models involving t h e i r expectations of the attachment-related responses (permitting or blocking access) of three adults to various c h i l d s i t u a t i o n s of d i s t r e s s and proximity seeking behaviour. Each c h i l d i s asked to indicate the card which i s "most l i k e " h i s mother, father or main teacher i n a series of forced choice tasks. A s i m i l a r aspect of t h e i r i n t e r n a l working models that i s measured, involves the childrens' expectations of t h e i r own responses to adult attachment figure proximity seeking behaviour. In t h i s case each c h i l d i s asked to indicate which card i s "most l i k e " himself i n a seri e s of forced choice tasks. In a secondary measure children are given a ser i e s of forced choices designed to assess t h e i r tendencies to perceive s i b l i n g s , peers and other s i g n i f i c a n t adults as "unfriendly" or " f r i e n d l y " . This measure i s assessed independently of v i s u a l l y represented c h i l d (or adult) sit u a t i o n s of d i s t r e s s and proximity seeking. I t i s a more 2 general measure, yet s t i l l assumed to measure an aspect of the childrens' working models. The group p r o f i l e i s analyzed i n order to provide empirical support for the v a l i d i t y of the response categories, i n p a r t i c u l a r when u t i l i z e d phenomenologically. The underlying dimension of these response categories (permitting/blocking access) i s determined from theory (Main, Kaplan, and Cassidy, 1985). 3 Rationale and Purpose of Study;  General Considerations This i s an exploratory study of ce r t a i n aspects of childrens' inner working models. I t i s being conducted i n order to determine i f childrens' expectations of s e l f and three caregivers can be represented v i s u a l l y along a ce r t a i n important dimension. As t h i s was an exploratory study, i t i s useful to describe i t s rationale within the context of associated past and recommended future research d i r e c t i o n s . Both the t h e o r e t i c a l work on attachment and in t e r n a l working models by Bowlby (1969, 1973, 1980) and the empirical work done by many others (e.g., Ainsworth, Blehar, Waters, & Wall, 1978; Main & Cassidy, 1988; Sroufe, 1983) strongly suggest a s i g n i f i c a n t amount of s t a b i l i t y i n behavioural patterns from infancy onwards. Bowlby (1973) suggests that some b e l i e f s and expectations a c t u a l l y have a ce r t a i n power to be s e l f - f u l f i l l i n g . Such elements of the in t e r n a l working model are t h e o r e t i c a l l y implicated by d e f i n i t i o n i n the maintenance of important aspects of behavioural s t a b i l i t y . Empirically, i n the attachment research however, there i s a dearth of evidence pointing to any p a r t i c u l a r p o t e n t i a l l y s e l f - f u l f i l l i n g expectations amongst children, except as can be infer r e d from t h e i r behaviour. Crittenden (1988b) states that a c h i l d ' s r e l a t i o n s h i p to h i s mother i s probably best conceptualized 4 i n terms of h i s expectations but notes the d i f f i c u l t i e s i n devising any procedure that d i r e c t l y assesses expectations. This research attempts to show that such i n t e r n a l images or expectations, as aspects of the inner working model, can begin to be systematically and v i s u a l l y represented by demonstrating that t h i s group of subjects shows c e r t a i n pattern(s) of homogeneity i n t h e i r combined selections from the forced choice categories. In addition, t h i s research attempts to show that a l l the response categories generated for the study are relevant to the subjects. These forced choice categories were generated from estimates of some of the important types of responses that a caregiver might make (and vice-versa) to a c h i l d i n some common s t r e s s f u l and proximity seeking s i t u a t i o n s . A rationale for t h i s type of study hinges i n p a r t i c u l a r on the q u a l i t y of the forced-choice categories generated. Main et a l . , (1985) stated that a caregiver can ei t h e r permit or block access to a c h i l d and that access can be permitted consistently or inconsistently. The permitting/blocking access inventory was designed to hinge on the d i s t i n c t i o n inherent i n i t s name. The four response categories include one for "permitting access" (1), two f o r "blocking access" (3, 4), and one roughly midway on the dimension of permitting/ blocking access (2). The dimension of consistency of access i s at best only p a r t i a l l y tapped i n the response categories through the provision of eight d i f f e r e n t s i t u a t i o n s i n which each of the four access 5 p o s s i b i l i t i e s may be selected. I t i s l i k e l y that within c e r t a i n l i m i t s at least, d i f f e r e n t childrens' images or expectations of t h e i r caregivers may be quite d i f f e r e n t even given the same l e v e l of caregiver consistency or the same given d i f f e r e n t l e v e l s of consistency. Caspi and Elder's (1988) 4-generation longitudinal study conducted between 1928 and 1972 c e r t a i n l y supports t h i s point. Using a h i e r a r c h i c a l regression analysis they demonstrated that "early family experience" did nothing to predict "problem adult behaviour" above and beyond "childhood problem behaviour". In other words, "early family experience" i s not necessarily the best predictor of the d i r e c t i o n a person's l i f e w i l l go. Clearly then, there i s necessarily a higher l e v e l of p r o b a b i l i t y that t h i s study accurately measures childrens' images or expectations of t h e i r caregivers than that i t measures how they came to be, or i n p a r t i c u l a r the exact quantity and qu a l i t y of t h e i r attachment-related experiences. Even though t h i s study i s not designed to demonstrate the nature of childrens' experiences, i t may be important to tap a wide v a r i e t y of experiences. Given that a s i g n i f i c a n t proportion of a l l 5 to 7 year-olds have been maltreated (Berger, Knutson, Mehm, & Perkins, 1988; Pelton, 1977; Straus, 1979), i t i s considered useful that the response categories tap some of the images or expectations that a maltreated c h i l d might be expected to develop. The "ignoring blocking access" category taps an issue c l e a r l y 6 central i n cases of neglect (see Crittenden, 1988a). The "angry blocking access" category s i m i l a r l y taps an issue c l e a r l y c e n t ral i n cases of physical abuse (see Crittenden, 1988a), and perhaps also i n cases of neglect (Burgess & Conger, 1978). Sexual abuse, as considered d i s t i n c t l y from physical abuse and neglect, i s not d i r e c t l y tapped by these p a r t i c u l a r response categories. This area aside, "ignoring" and "angry blocking access" categories are considered to be useful i n spanning "normal" and maltreated subjects' i n t e r n a l images or expectations. For that matter, the "permitting" and "midway on access" categories are considered to be equally useful for t h i s purpose. In future studies of t h i s nature, the differences i n degree of perceived "ignoring", "angry", "permitting" or "midway on access" caregiver response for maltreated versus non-maltreated populations i s probably best tapped by differences i n the frequencies of s e l e c t i o n f o r these four access categories. Phenomenological Considerations In a review of 20-30 years of l i t e r a t u r e concerning the e f f e c t s of c h i l d - t r a i n i n g methods on l a t e r personality development Dubin and Dubin (1964) noted "widespread f a i l u r e even to r e a l i s e that i t i s not only parental behaviour to which the c h i l d responds but also h i s perception of parental behaviour" (p. 809). The c h i l d ' s perception of parental 7 behaviour i s hypothesized to be the missing element i n understanding h i s response to parental behaviour. "The one-to-one r e l a t i o n between parental behaviour and c h i l d personality has yet to be demonstrated 11 (p.810). The success of attachment researchers, (e.g., Ainsworth et a l . , 1978; Crittenden, 1988a, 1988b; Main & Stadtman, 1981) i n demonstrating s i g n i f i c a n t correlations between patterns of parental and c h i l d behaviour does not negate the Dubins' point. Furthermore, t h i s i s acknowledged within the theory of the inner working model (Bowlby, 1969, 1973, 1980). We may now have improved pr e d i c t i v e power between patterns of paternal behaviour and patterns of i n f a n t / c h i l d behaviours, between patterns of infant behaviour and c h i l d psychopathology f o r males (Lewis, F e i r i n g , McGuffog, & J a s k i r , 1984), or between "problem c h i l d behaviour" and "problem adult behaviour" (Caspi & Elder, 1988). However, there i s s t i l l a great deal of variance l e f t unexplained with respect to a l l of these rela t i o n s h i p s . As well, Hinde (1982) suggests that what was observed i n the Strange Situation Procedure (Ainsworth & Wittig, 1969) was not the caregiver/child r e l a t i o n s h i p i t s e l f but rather the infant's view of i t , and at a p a r t i c u l a r point i n time. And i n another reference to Bowlby, Main et a l . (1985) noted that the Strange Situation r e f l e c t s a p r i m i t i v e i n t e r n a l working model at a p a r t i c u l a r point i n time. Recently, some attachment researchers have incorporated attempts to d i r e c t l y tap childrens' perspectives into the 8 es s e n t i a l design of t h e i r studies (Cassidy, 1988; Cassidy & Kobak, 1988; Main, Kaplan & Cassidy, 1985). Cassidy (1988) notes "problems associated with the attempt to make observable that which i s in t e r n a l and unseen" (p.123) not the l e a s t being the v a l i d i t y of s e l f - r e p o r t s . Main et a l . (1985) note that i f childrens' representations of attachment are constructed out of c r i t i c a l events (such as separations from an attachment figure) then representations of childrens' responses to such events are of i n t e r e s t . F i n a l l y , Crittenden (1988a) states the need not only to f i n d ways to change the experience of abusing parents and abused children but also to f i n d ways to change t h e i r conceptualizations of t h e i r experience. Indeed, t h i s study i s designed to explore how 5 to 7-year-old childr e n from one suburban neighbourhood conceptualize some aspects of t h e i r experiences with caregivers, r e l a t i v e s and peers. Developmental Considerations The 5 to 7-year-old age group was chosen f o r t h i s study i n part because i t was thought to be the youngest age group generally capable of contributing useful i n t e r n a l information through t h i s format. Indeed, a small but s i g n i f i c a n t minority of the 5-year-olds did not appear to understand or to be capable of following the d i r e c t i o n s s u f f i c i e n t l y . However, at t h i s age i t i s probable that the use of the forced-choice categories i s preferable to more 9 loosely structured interviews because of the r e l a t i v e l y low l e v e l of l i n g u i s t i c and cognitive capacity to give useful, interpretable descriptions of themselves and others. Certainly, categorization of t h e i r responses i s enhanced. At the same time the children themselves are s t i l l being s o l i c i t e d f o r t h e i r own views. Regarding the upper l i m i t of t h i s age range, one of Piaget's best validated findings involves the great d i f f i c u l t y that children under 7 have i n seeing anything from the point of view of anybody else (Piaget, 1924; Piaget & Inhelder, 1948). In the present study t h i s c h a r a c t e r i s t i c , c a l l e d "egocentrism", might be expected to minimize the subjects' tendency to t r y to respond to a perceived expectation of the interviewer. This form of bias then may s i g n i f i c a n t l y increase for groups of children older than 7. Rholes and Ruble (1984) suggest on the basis of two of t h e i r own studies and one by Hel l e r and Berndt (1981) that 5 and 6-year-old children could at best make only f a i r l y l i m i t e d generalizations about others across s i t u a t i o n s . A f t e r the age of 7 or 8 children were increasingly able to perceive broad consistencies i n the behaviour of other child r e n across d i f f e r e n t s i t u a t i o n s . The present study d i f f e r s i n that p a r t i c u l a r caregivers are depicted, and i n attachment-related s i t u a t i o n s . In addition, subjects are not required to make generalizations across s i t u a t i o n s although we may observe generalizations i n t h e i r images or 10 expectations across the d i f f e r e n t s i t u a t i o n s . These other studies do however point to further developmental changes a f t e r about the age of 7. The two Rholes and Ruble studies made an additional relevant point i n that they found evidence against the l i k e l i h o o d that 5 to 7-year-olds' choices r e f l e c t e d a " p o s i t i v i t y bias", that i s , a complete preference f o r p o s i t i v e behaviours. Relative to 9 and 10-year-olds, they did not la b e l the other childrens' behaviour as p o s i t i v e l y , though they were demonstrated to be capable of doing so. To summarize, although i t i s expected that an older group of children, r e l a t i v e to a younger group of children, could incorporate i n t h e i r selections expectations of s e l f and others p o t e n t i a l l y altered or affected by additi o n a l l i f e experience - i t i s not c l e a r l y understood how t h e i r maturing developmental capacities would a f f e c t t h e i r response tendencies. A possible advantage to choosing t h i s age group involves the oft-expressed professional and s c i e n t i f i c opinion that a c h i l d ' s personality i s la r g e l y formed by the time he i s 4 through 6-years-old. Attachment t h e o r i s t s don't tend to make t h i s p a r t i c u l a r generalization, i n part because a c l e a r r e l a t i o n between the inner working model and the various personality constructs has yet to be made. In addition, there i s s t i l l a great deal of debate regarding the r e l a t i v e extent to which various patterns of c h i l d behaviour demonstrate continuity or l a b i l i t y (e.g., Lamb, Thompson, Gardner, Charnov, & Estes, 1984; Thompson, Lamb, & 11 Estes, 1982). Probably the d i s t i n c t i o n between open and closed working models (see Chapter 2) r e f e r r i n g to another continuum by which the inner working model can be described - i s relevant here. I t may be that by a c e r t a i n average age a c h i l d has formed cle a r images or expectations regarding h i s attachment figures and s e l f . However, these images or expectations (and others) may be more or they may be le s s open to mirroring any changes i n l i f e circumstances. This p a r t i c u l a r study does not attempt to measure the degree of openness/closedness of these childrens' working models. However, i t i s possible that the p a r t i c u l a r degree has been established, again more or le s s , by the age of 5 or so; that i s another empirical question. In any event, t h i s study attempts to capture 5 to 7-year-olds' current images or expectations, however l a b i l e they s t i l l might be at t h i s general age and for each p a r t i c u l a r i n d i v i d u a l . Inclusion of Teachers as Attachment Figures Sroufe's (1983) la r g e l y d escriptive study of 40 four-year-old preschoolers goes a long way to e s t a b l i s h empirically that children can and do exhibit attachment behaviours towards t h e i r teachers. Teachers are t e n t a t i v e l y considered to be "attachment figures" to childr e n for the purpose of t h i s study. Not uncommonly another in-home r e l a t i v e may take on s i g n i f i c a n t attachment figure status. However, many children do not have older s i b l i n g s or other 12 in-home r e l a t i v e s who might act as attachment figures. P a r t l y for t h i s reason, each c h i l d ' s most permanent teacher was selected to p o t e n t i a l l y represent a t h i r d attachment figure as depicted i n the two study measures. Generally speaking, each teacher i s considered to p o t e n t i a l l y act as a "secondary" (Ainsworth, 1982) attachment figure and each parent i s considered to be a " p r i n c i p a l " attachment figure. I t i s hoped that t h i s exploration of the inner working models of children w i l l provide useful hints f o r therapists and school personnel. In Carl Rogers' introduction to V i r g i n i a Axline's book (1947), "Play Therapy", he writes that the book i s , "on the surface, an account of the way i n which a teacher has come to function as a therapist, to release the curative forces which e x i s t within each i n d i v i d u a l " ( p . v i i ) . Axline believes that a teacher can make a very s i g n i f i c a n t difference i n a c h i l d ' s l i f e - c o u r s e . She writes, the therapist-teacher i s a l e r t to recognize the feelings the c h i l d i s expressing and r e f l e c t s those feelings back to the c h i l d i n such a manner that the c h i l d gains insight into h i s behaviour. This can be done to a great extent i n any classroom s i t u a t i o n i f the teacher has an understanding of her pupils and an insight into human behaviour. I f the therapeutic r e l a t i o n s h i p has been established between p u p i l and teacher, many children may be helped to gain valuable insight into t h e i r problems before the problems become 13 so unwieldly that they create serious maladjustment, (p.142) Given Caspi and Elder's (1988) evidence that "childhood problem behaviour" mediates the r e l a t i o n between "early family experience" and "negative adult behaviour", i t follows that "childhood problem behaviour" may represent a useful point of intervention i n a p o t e n t i a l l y pathological l i f e - c o u r s e . School probably represents the most s i g n i f i c a n t stage between early negative family experience on the one hand, and trouble with the p o l i c e and negative adult behaviour and parenting on the other. Teachers can become s i g n i f i c a n t attachment figures acting to confirm or disrupt any negative and p o t e n t i a l l y s e l f - f u l f i l l i n g expectations of children. As Crittenden (1988a) and Bowlby (1988) have pointed out, c e r t a i n children need to spend time with someone whose sen s i t i v e responsiveness and communication of p o s i t i v e feelings towards and about them act to allow corresponding p o s i t i v e changes i n t h e i r inner working model. S i m i l a r l y , Summit (1984) claims, "the teacher who re l a t e s e f f e c t i v e l y to a c h i l d may become for that c h i l d an anchor of self-endorsement and a prototype on which to b u i l d other successful r e l a t i o n s h i p s . " (p.33) However, a l t e r n a t i v e l y , "the c h i l d may tend to see the teacher as an extension of the parent. . . and the teacher may f e e l i n the student an inappropriate fear and d i s t r u s t 14 accompanied by t e r r i b l e pressure to perform and abject expectation of f a i l u r e . . ." (p.33). Aber, A l l e n , Carlson and C i c c h e t t i (1989) go so f a r as to say that for older children attachment to primary caregivers should perhaps be subsumed to other stage-s p e c i f i c aspects of the attachment system, for example, t h e i r a b i l i t y to e s t a b l i s h p o s i t i v e r e l a t i o n s with novel adults. I f so, t h i s provides further reason to explore the r e l a t i o n s h i p between children and t h e i r teachers, both i t s a c t u a l i t i e s and i t s p o t e n t i a l i t i e s . Bowlby (1973) also considered t h i s a type of rel a t i o n s h i p worthy of exploration and challenged researchers to e s t a b l i s h empirically that children showed actual attachment behaviours towards teachers. F i n a l l y , Cohn (1990) has conducted what i s probably the most s i g n i f i c a n t exploration of the r e l a t i o n s h i p between child-parent attachment and the r e l a t i o n s h i p of the c h i l d to peers and teachers i n the school s e t t i n g since the Sroufe (1983) study. Cohn found that 6-year-old boys, but not g i r l s , with an insecure attachment to t h e i r mothers are s i g n i f i c a n t l y more l i k e l y to receive lower sociometric ratings and higher ratings on aggressiveness and behaviour problems i n the school. Non-attachment studies s i m i l a r l y provide evidence that "non-maltreated" children growing up under a parenting s t y l e characterized by i r r i t a b i l i t y , explosiveness and threats develop an aggressive s t y l e that i s often generalized from the family to peers and teachers 15 i n the school (Patterson, DeBaryshe, & Ramsey, 1989; Simons, Whitbeck, Conger, & Wu Chyi-In, 1991). Clearly, the c h i l d ' s own perceptions, images or expectations of h i s teacher qua attachment figure i s an area of c h i l d development worthy of study. The p a r t i c u l a r variables i n t h i s study have been chosen to maximize the pot e n t i a l contributions to understanding and in s i g h t into c h i l d behaviour for teachers i n t h e i r capacity as attachment figures and teacher-therapists. Some of the most basic dimensions of attachment theory have been u t i l i z e d for t h i s purpose namely, c h i l d d i s t r e s s states i n t h e i r capacity as activators of the attachment behavioural system, proximity seeking behaviour, attachment figure as "secure base" from which to explore the world, attachment figure behaviour with respect to permitting or blocking access to the c h i l d , and int e r n a l working models as elucidated i n childrens' images or expectations of adult attachment behaviour. 16 Background, Development and Theoretical Basis of the Problem This study follows i n the t r a d i t i o n established by Sroufe (1983) of regarding the teacher as a p o t e n t i a l attachment figure. I t also r e l i e s heavily on a single c l a r i f y i n g statement made by Main et a l . , (1985) that there are b a s i c a l l y only three ways that an attachment figure can respond to c h i l d proximity seeking behaviour, by permitting access consistently, or inconsistently, or by blocking access. The two measures of t h i s study are designed to attempt to tap the i n t e r n a l images or expectations of non-maltreated, p h y s i c a l l y abused and/or neglected children. The maltreated and non-maltreated categories are not considered to be completely d i s t i n c t within the subject group. This idea i s supported i n two quite d i f f e r e n t ways. F i r s t , a c e r t a i n s i g n i f i c a n t proportion of t h i s 5 to 7-year-old group can be expected to have been maltreated. In t h i s case that conclusion i s based on s t a t i s t i c s f o r large populations of children. For example, Straus (1979) i n a study of 1146 families, found that more than 14% of American children, 3 to 17, received abusive violence an average of 10.5 times a year i n the form of "punching, kicking, b i t i n g , h i t t i n g with an object, 'beating up', or 'using a gun'. In addition, the most conservative estimates of the annual incidence of physical abuse of children range from 60,000 to 167,000 i n the U.S. Neglect i s believed to be more than 17 twice as frequent (Pelton, 1977). "Maltreated" can mean maltreatment that has been substantiated by c h i l d protection agencies using one or another c r i t e r i a . However, "maltreated" might also mean maltreatment that would be substantiated by the same c r i t e r i a were a l l the facts known to a c h i l d protection agency. There are many complicated issues associated with the d e f i n i t i o n of maltreatment . Also a c e r t a i n amount of maltreatment by any d e f i n i t i o n goes unreported. The maltreated and non-maltreated categories are not considered to be d i s t i n c t within the subject group from rather a d i f f e r e n t perspective as well. They are not completely d i s t i n c t i n terms of the types of experience (and consequent inner images or expectations) shared to one degree or another by some members of both categories. Bowlby (1973) points out that there i s a vast amount of intermediate experience between those groups of people with extremes of good and bad experience. Each person i n a l l groups grow up with expectations to match his/her early experience. In the intermediate range, provided the rules have been moderate and the sanctions mild and predictable, a person can s t i l l come confidently to believe that support w i l l always be availab l e when needed. But when rules have been s t r i c t and d i f f i c u l t to keep, and when sanctions on breaking them have been severe and es p e c i a l l y when they have included threats to withdraw support, confidence i s 18 l i k e l y to w i l t . (p.209) To i l l u s t r a t e further the commonalities i n experience between groups of children l a b e l l e d "maltreated" or "non-maltreated", George and Main (1979) note s i m i l a r i t i e s between rejected children (without substantiated maltreatment) and t h e i r mothers on the one hand and ph y s i c a l l y abused children and t h e i r mothers on the other. Both groups of children respond with avoidance to the mother. Both groups of mothers engage i n angry, r e j e c t i n g behaviour and aversion to physical contact. Crittenden (1988a) describes abusive caregivers as providing generally needed care accompanied by excessive anger, harshness and/or h o s t i l i t y . Adequately reared children c l e a r l y share the experiences inherent i n "generally needed care". Crittenden uses the same scales for four categories of maltreating mothers and one category of adequate mothers; a l l f i v e categories of caregiver demonstrate one degree or another of the same types of behaviours as c l a s s i f i e d using the shared scales. The s i g n i f i c a n t differences i n degree do di s t i n g u i s h them however. Berger's et a l . ' s (1988) study using the s e l f - r e p o r t s of 4695 college students demonstrated that 12% of them could i d e n t i f y actual physical i n j u r i e s received during childhood. The great majority of them had never been designated as "phy s i c a l l y abused". Furthermore, only 3% of the students considered themselves to have been "physically abused". 19 Cle a r l y , however, t h i s study suggests that a number of these students by the age of 7 had probably been exposed to harsh or angry caregiver behaviour. Simons et a l . ' s (1991) study used parent and adolescent s e l f - r e p o r t s i n an exploration of harsh physical punishment practices. They found that male adolescents were two to three times as l i k e l y to be p h y s i c a l l y punished as females. To be p h y s i c a l l y punished i n the post-childhood years i s considered to be a sign of harsher physical parenting pr a c t i c e s . They do report, however, that although abusive parenting continues to be a s i g n i f i c a n t problem, harsh parenting has declined considerably over the course of the l a s t generation. The point of these examples i s towards the suggestion that the "maltreated" and "non-maltreated" categories may be reasonably conceived as opposite poles on a rough continuum (or set of continuums) of childhood experience. Rather than simply viewing children as maltreated or not i t may well be more useful and less obfuscating to view them i n terms of t h e i r own (phenomenological) range of care-receiving experiences along t h i s continuum (or set of continuums). The use of the early A-B-C attachment c l a s s i f i c a t i o n system allowed many maltreated and semi-dysfunctional infants to be c l a s s i f i e d as "secure" (Spieker & Booth, 1988). This was thought to be a r e s u l t of an i n i t i a l primary focus on normal development (Crittenden, 1988b; Spieker & Booth, 1988). The s i t u a t i o n i s changing for the better i n the strengthening 20 "marriage" of developmental psychology and psychopathology (Belsky & Nezworski, 1988; Sroufe, 1986). However, the determination of the status of "normal" or "abnormal", "maltreated" or "non-maltreated" w i l l always i n e v i t a b l y remain to some extent a " p o l i t i c a l " decision, subject to loc a l e , era and human decision-making bodies. 21 Chapter 2: Literature Review:  Overview This l i t e r a t u r e review i s comprised i n four sections. Some of the underlying dimensions of attachment theory are i n i t i a l l y explained and described including c e r t a i n patterns of behaviour, types of behavioural systems and behavioural strategies. The second section describes the development of a "marriage" between developmental psychology and c l i n i c a l psychology and psychiatry. This "marriage" i s described within a context of the integration of d i s c i p l i n e s . This section i s included to provide further support and rationale for t h i s study's inc l u s i o n of a large amount of material from areas of psychology and psychiatry. The t h i r d section i s a major section and a substantial review of the l i t e r a t u r e on perceptual d i s t o r t i o n or exclusion as i t r e l a t e s to attachment. I t i s included i n large part i n order to help set l i m i t s on the i n t e r p r e t a t i o n of study r e s u l t s . This section also contains a review of the l i m i t e d number of representational studies conducted with the same age group as i n t h i s representational study. The f i n a l section lays out the theory of the inner or representational working model and some of the empirical studies which support and r e f i n e i t . In addition to describing i t s key features and types of structures, examples are given of inner models held by various types of 22 adequate and maltreating populations. Continuity and change i n the inner working model and consequently i n behaviour i s explored. This section concludes with a discussion of applications and future research d i r e c t i o n s . This i s the section most obviously relevant to t h i s study of important aspects of 5 to 7-year-old boys' inner models, that i s , t h e i r inner images or expectations of attachment figures and s e l f . 23 Important Dimensions of Attachment Theory Organized patterns of attachment behaviour may be distinguished from separate, discrete attachment behaviours. Such di s c r e t e behaviours, eg. "smiling", "crying" or "touching", have been traced through early development by frequency counts and generally without regard to t h e i r contextual meaning. I t i s clea r that the same behaviour, for example, "smiling", may lose i t s common (prosocial) meaning i n c e r t a i n contexts such as i n response to the di s t r e s s of another. In other cases large numbers of o r d i n a r i l y d i s s i m i l a r behaviours may be subsumed together according to some meaningful p r i n c i p l e . S i m i l a r l y , i t may be le s s s i g n i f i c a n t ( i n i t i a l l y ) to measure the strength or weakness of d i f f e r e n t relationships than to measure the q u a l i t a t i v e differences i n relat i o n s h i p s ; or i t may be les s pertinent to look at the severity of maladaptation i n development or caregiving than to look at the " s t y l e " or manner of caregiving or of meeting developmental issues (Ainsworth, 1982; Crittenden, 1988b; Sroufe, 1983). There are four types of behavioural systems activated by the Strange Situation Procedure and by the various other assessment too l s of attachment research. The attachment behavioural system i s studied i n interplay with the wariness/fear, a f f i l i a t i o n / s o c i a b i l i t y , and exploration behavioural systems. Single behaviours that could be c l a s s i f i e d i n two systems, e.g., moving toward mother 24 (attachment) while moving away from a stranger (wariness) or the sequential a c t i v a t i o n of more than one system, e.g., o f f e r i n g an object to the stranger ( a f f i l i a t i v e ) then immediately running to mother (attachment) are necessarily considered i n the contexts of ongoing a c t i v i t y and environment (Schneider-Rosen, Braunwalk, Carlson & C i c c h e t t i , 1985). Each behavioural system i s activated by a unique set of conditions. I f one i s activated at low i n t e n s i t y (e.g., attachment) i t i s l i k e l y that another w i l l be activated at high i n t e n s i t y (eg. exploration), and t h i s w i l l determine the behaviour that i s observed. The strange or novel may activate the exploratory system or at a higher i n t e n s i t y (or i n the case of a d i f f e r e n t c h i l d ) , the fear/wariness system, with l i k e l y i n t e n s i f i e d attachment behaviour (Ainsworth, 1982) . Corresponding to the c h i l d ' s attachment behavioural system i s the parental attachment behavioural system. Both systems serve proximity maintenance and protection and nurturance of the c h i l d (Bowlby, 1969). Crittenden (1988b) describes four universal dimensions of attachment theory, i n i t i a l l y set out by Ainsworth (1978). Behaviours that act to bring about proximity or maintain contact correct attachment partners are two f a i r l y straightforward ones. More complex are "avoidance" and "resistance". "Avoidance", involves the avoidance of the attachment figure or of other cues l i k e l y to activate the 25 attachment behavioural system. "Resistance" involves a heightening of responsiveness on the part of the c h i l d to attachment e l i c i t i n g cues and to the attachment figure - to the point of anger and d i f f i c u l t y i n calming down (Carlson, C i c c h e t t i , Barnett, & Braunwald, 1989). These four dimensions or behavioural strategies are thought to underlie the e n t i r e attachment c l a s s i f i c a t i o n system with one possible addition (Crittenden, 1988b, 1989). This i s the " c o n t r o l l i n g " behavioural strategy i d e n t i f i e d i n a population of 6-year-olds by Main and Cassidy (1988). These children, whose behaviour had been very disorganized i n infancy, were observed to control t h e i r caregivers i n e i t h e r a punitive or a caregiving fashion. 26 Cognitive and A f f e c t i v e Bias Disorders of attachment are functions of the child/parent r e l a t i o n s h i p that r e s u l t i n the c h i l d ' s i n a b i l i t y to experience the parent as emotionally a v a i l a b l e and as a r e l i a b l e protector from external danger or in t e r n a l d i s t r e s s (Lieberman & Pawl, 1988). A secure attachment involves the c h i l d ' s f e e l i n g safe and undistressed while with the parent and for short but increasing periods of time while away from the parent. In the meantime, feelings of longing and the desire to restore proximity and contact are part and parcel of a secure attachment (Bowlby, 1969; Ainsworth et a l . , 1978). A secure attachment to mother hinges on the following four factors: mother's s e n s i t i v i t y to signals, her timing of interventions, the c h i l d ' s experience of predictableness with respect to the r e s u l t s of his s o c i a l i n i t i a t i v e s , and the extent to which h i s s o c i a l i n i t i a t i v e s are successful i n helping e s t a b l i s h a r e c i p r o c a l exchange with her (Bowlby, 1969). When a c h i l d i s met by r e j e c t i o n on the part of the attachment figure, a c o n f l i c t i s created for the c h i l d (Main & Stadtman, 1981). The c h i l d i s to some extent torn between attachment behaviour and withdrawal. Bowlby (197 3) i n i t i a l l y discussed t h i s c o n f l i c t i n terms of violence or threats rather than r e j e c t i o n per se on the part of the attachment figure. He noted the l i k e l i n e s s of infants, human and non-human, to c l i n g to the threatening or h o s t i l e attachment 27 figure. Whether the attachment figure's behaviour i s one of physical aggression or a milder form of physical or emotional r e j e c t i o n , the c h i l d w i l l to some degree f i n d himself torn between the opposite responses of approach and withdrawal. There i s pain associated with both physical aggression and r e j e c t i o n . However, when the aggressing or r e j e c t i n g i n d i v i d u a l i s also the attachment figure there i s an additional component of pain i n the form of anger and anxiety associated with the c o n f l i c t between attachment behaviour and withdrawal. Children who avoid the attachment figure and other cues l i k e l y to activate the attachment behavioural system are thought to be deactivating the pa i n f u l conciousness of t h i s c o n f l i c t (Cassidy & Kobak, 1988; Main & Stadtman, 1981). An avoidant c o n f l i c t may be central to the cognitive and behavioural responses of many maltreated children and non-maltreated childre n (Crittenden, 1988a). As early as 1979, George and Main recognized the s i m i l a r i t y between normal rejected toddlers' and p h y s i c a l l y abused toddlers' avoidance of t h e i r caregivers. They shared addit i o n a l c h a r a c t e r i s t i c s as well. When they were not expressing avoidant behaviour they were both s i g n i f i c a n t l y more aggressive and s i g n i f i c a n t l y less empathic to the di s t r e s s of others. These c h a r a c t e r i s t i c s are s i m i l a r to those of avoidant caregivers. That i s , avoidant caregivers were demonstrated to be s i g n i f i c a n t l y more p h y s i c a l l y 28 r e j e c t i n g and i n s e n s i t i v e with infants during the f i r s t year of l i f e when infants are thought to most need physical comforting (Blehar, Ainsworth & Main, unpub. ms.; Marvin, 1977) ; i n addition p h y s i c a l l y r e j e c t i n g mothers were more l i k e l y to handle t h e i r infants roughly during free play (Main & Stadtman, 1981). This "avoidant-aggressive syndrome" (George & Main, 1979) i s p a r t i c u l a r l y i n t e r e s t i n g with regard to physical contact because aversion to physical contact and rough physical handling seem to be such divergent behaviours (Main & Stadtman, 1981; Older, 1981). Cassidy and Kobak (1988) suggest that avoidance i s a s i g n i f i c a n t factor i n the interpretation of research findings as well as i n measurement construction and research design. At the representational l e v e l , as opposed to the behavioural l e v e l , subjects commonly use two defensive strategies, "deactivation of the attachment system" and " i d e a l i z a t i o n " . Avoidance i s central to both strategies. In a study of 52 six-year-old children Cassidy (1988) found that those who were c l a s s i f i e d as avoidant i n infancy were more l i k e l y to receive perfect scores on self-esteem, suggesting less than accurate s e l f - r e p o r t i n g and s p e c i f i c a l l y , " i d e a l i z a t i o n " . I t was, however, not possible to d i s t i n g u i s h between children previously c l a s s i f i e d as secure or avoidant on the basis of perfect scoring alone since about a t h i r d of secure children also received perfect scores. The remaining secure children described themselves p o s i t i v e l y , with some flaws. In a d o l l exercise designed to 29 tap t h e i r views of s e l f i n rel a t i o n s h i p to attachment figures, the 6-year-olds previously c l a s s i f i e d as avoidant were les s l i k e l y to acknowledge the importance of t h e i r r e l a t i o n s h i p to t h e i r mothers, to express a f f e c t i o n or to value and accept themselves. Kaplan and Main (1985) found that 6-year-old children previously c l a s s i f i e d as avoidant, when asked to draw family figures, drew tense or r i g i d figures with l i t t l e i n d i v i d u a l i t y . These figures were unable to reach out to others since they were drawn without arms. F a c i a l expressions were often stereotyped with overemphasized smiles suggesting the masking of negative a f f e c t . Overall, the drawings strongly suggested the deactivation of the attachment behavioural system as expressed at the representational l e v e l . In addition, these same children were shown a series of photographs depicting separations between c h i l d and mother (Hansburg Separation Anxiety Test adapted for age by Klagsbrun and Bowlby, 1976). Those 6-year-olds c l a s s i f i e d as avoidant i n infancy were la r g e l y unable to suggest the use of others as sources of comfort and support or to suggest adaptive coping strategies i n general. They were, however, able to acknowledge d i s t r e s s with respect to the situations depicted. Cassidy and Kobak conclude that the influence of experimental observation on behaviour, s e l f - r e p o r t s and on the various forms of s e l f - r e l a t e d representations - may vary with attachment organization. 30 Based l a r g e l y on c l i n i c a l experience, Bowlby (1973, 1980) claims that i d e a l i z a t i o n of the parent (as opposed to i d e a l i z a t i o n of the sel f ) i s a common phenomenon involving the exclusion from awareness of the model of the bad unloving parent. Some causes include parental r i d i c u l i n g of the c h i l d ' s security-seeking (attachment) behaviors, t h e i r r e i n t e r p r e t i n g r e j e c t i o n as love or motivated by love, and other forms of disavowing or denying the c h i l d ' s anxious, angry or loving feelings towards the parent. Given such circumstances, a model of the bad, unloving parent i s s t i l l thought to ex i s t below conciousness as a "subordinate" model and to continue to influence the i n d i v i d u a l . One mechanism by which c o n f l i c t i n g inner models may occur simultaneously involves the c l a s s i f i c a t i o n of "actual" experience with the attachment figure into many cross-referenced "schemata" (Mandler, 1979) at d i f f e r e n t h i e r a r c h i c a l l e v e l s of memory (Bretherton, Ridgeway, & Cassidy, 1990; Schank, 1982) A f f e c t i v e and cognitive information i s reprocessed, par t i t i o n e d , cross-indexed and summarized i n a v a r i e t y of ways r e s u l t i n g i n the various i n t e r n a l schemata which simulate r e a l i t y . Memories excluded at the episodic l e v e l , that i s , episodic memories excluded from conciousness, may nevertheless be included i n schemata at other l e v e l s thus influencing a person's thinking and behaviour. Bowlby (1973) describes "cognitive development" as consisting of the steps undergone from simple stimulus response to using rules to combine information from 31 perception and memory. By means of these rules the i n d i v i d u a l can predict events i n the world more or less accurately and plan and respond accordingly. An i n d i v i d u a l ' s "cognitive bias", along s i m i l a r l i n e s of reasoning, r e f e r s to the speed, completeness and accuracy of the processing of information. These related concepts describe the functioning of the inner working model. "The storage of images of parents and s e l f i s almost c e r t a i n to be of at l e a s t two d i s t i n c t types" (Bowlby, 1973, p.62). They may be stored in/as "episodic" or "semantic" memory (Tulving, 1972). Episodic memory commonly retains i t s perceptual properties while f i t t i n g i n t e g r a l l y into the rememberer's sense of h i s personal i d e n t i t y or l i f e h i s t o r y . Semantic memory, on the other hand, i s comprised of information stored i n the form of generalizations. These generalizations can vary considerably i n the degree to which they are "constructed" versus "taken i n whole" from others. In a study of 40 adult couples Main et a l . (1985) demonstrated a strong c o r r e l a t i o n between i d e a l i z a t i o n of the adults' parents and r e j e c t i o n by those parents during childhood. Crittenden's studies (1985, 1988a) of parent-child dyads i s strongly suggestive of one possible pathway to i d e a l i z a t i o n of the parent. From as early as 12 months of age some maltreated infants were demonstrated to "cooperate" with t h e i r mothers i n a r i g i d manner while i n h i b i t i n g d i r e c t expressions of negative a f f e c t towards the mother (Crittenden, 1988b). They were described as 32 "compulsively compliant" as opposed to "cooperative" towards the mother (Crittenden & D i L a l l a , 1988). These mothers were found to be s i g n i f i c a n t l y more c o n t r o l l i n g and i n t e r f e r i n g and s i g n i f i c a n t l y less s e n s i t i v e and responsive than other mothers. The i n h i b i t i o n of negative a f f e c t i n those si t u a t i o n s i n which i t would be appropriately expressed i s expected to put these children at a disadvantage i n terms of act u a l l y knowing how they do f e e l . In addition, many ph y s i c a l l y abusive parents (Azar, Robinson, Hekimian, & Twentyman, 1984; Azar & Rohrbeck, 1986; Crittenden, 1988a) and perhaps neglecting parents as well (Azar et a l . , 1984) have an unusually strong need to be viewed as "good" parents. Given patterns of compulsively compliant behaviour, i n h i b i t e d negative a f f e c t and c o n t r o l l i n g parents who perhaps need to be viewed as near-perfect parents, i t i s not d i f f i c u l t to imagine some s i g n i f i c a n t number of these ch i l d r e n "learning" to i d e a l i z e t h e i r parents. Reider and C i c c h e t t i (1989) report d i f f e r e n t forms of information processing d e f i c i t s or, "cognitive control strategies", i n a study of 72 four to nine-year-old children. Maltreated children versus non-maltreated childre n and boys i n general versus g i r l s i n general are more l i k e l y to pay attention to d i s t r a c t i n g s t i m u l i when they are aggressive i n nature ( i . e . , images of guns and knives). When neutral s t i m u l i are i n the foreground and aggressive s t i m u l i are i n the background maltreated childr e n of both genders and boys i n general are s i g n i f i c a n t l y more 33 l i k e l y to attend to i . e . , "the sharpening to" background s t i m u l i . When aggressive and neutral s t i m u l i were reversed with respect to background and foreground, maltreated childre n are able to assimilate the aggressive s t i m u l i more accurately; non-maltreated children of both genders were less l i k e l y to attend to, i e . "the l e v e l l i n g to", the foreground aggressive s t i m u l i . Reider and C i c c h e t t i suggest that the maltreated c h i l d , by turning towards aggressive cues and as a r e s u l t away from non-aggressive ones, may f e e l better prepared to deal with adversity. Given Simons et a l . ' s (1991) finding that adolescent boys are two to three times more l i k e l y than adolescent g i r l s to be p h y s i c a l l y , often harshly, punished by t h e i r parents i t would not seem u n l i k e l y that roughly the same might apply for these 4 to 9-year-old boys and g i r l s . In other words, i n addition to maltreated children of both genders, boys i n general may, by sharpening to aggressive cues, f e e l better prepared to deal with adversity. Furthermore, i f i t could be shown that boys are subject to s i g n i f i c a n t l y greater amounts of physical aggression from persons other than t h e i r parents (e.g., peers, teachers) t h i s explanation would be supported further. Of course, other explanations w i l l no doubt contribute to a f u l l e r understanding of t h i s phenomenon. Reider's and C i c c h e t t i ' s concepts of "sharpening" and " l e v e l l i n g " to s t i m u l i seem to correspond c l o s e l y to Bowlby's concepts of "perceptual v i g i l a n c e " and "perceptual defence" (Bowlby, 1973). One difference perhaps applies to 34 the greater pote n t i a l for a f f e c t i v e n e u t r a l i t y i n the applications of the former p a i r of terms. While "sharpening" and " l e v e l l i n g " apply d e s c r i p t i v e l y with respect to the perceiver's r e l a t i o n to strong a f f e c t i v e s t i m u l i , they don't necessarily apply i n t h i s manner. For example, i f a c h i l d ' s l e v e l l i n g to non-aggressive s t i m u l i occurs because of h i s sharpening to aggressive s t i m u l i , then the a p p l i c a t i o n of the concept of " l e v e l l i n g " to h i s perception appears to be a more neutral one. The ap p l i c a t i o n of the concept of " l e v e l l i n g " to describe the maltreated c h i l d ' s tendency to insulate himself from non-aggressive s t i m u l i i s consistent with r e s u l t s from another study. Aber and A l l e n (1987) demonstrated the maltreated c h i l d r e l a t i v e to the non-maltreated c h i l d to have les s "effectance motivation". By t h i s i t i s meant that le s s i n i t i a t i v e was observed i n the seeking out or mastery of new s i t u a t i o n s . 3 5 Internal Working Models Bowlby, a psychoanalyst, drew heavily from psychoanalytic theories of object r e l a t i o n s (Bretherton et a l . , 1990). His theory of attachment i s narrower, i n that i t doesn't focus on a l l s o c i a l r e l a t i o n s h i p s ; i t i s also more general i n that i t explains both healthy and pathological development. Perhaps i t s greatest break from psychoanalytic theory was i n i t s claim that the attachment system i s a d i s t i n c t motivational system, (Bretherton, 1985). Bowlby was inspired by Craik (1943) as well as by Piaget (1952, 1954) to rework psychoanalytic concepts such as " i n t r o j e c t i o n " , "projection" and "representation" into the metaphor of " i n t e r n a l working models." Most importantly, such a metaphor i s much more amenable to empirical investigation. Empirical evidence suggests that from age 1 infants use working models to forecast the probable future behaviour of the attachment figure (Izard, 1978; Sroufe, 1979). In addition, a very important study by Main and Cassidy (1988) demonstrated strong p r e d i c t a b i l i t y from early security of attachment to aspects of the c h i l d ' s i n t e r n a l working model f i v e years l a t e r . Bowlby (1973) notes four key features of i n t e r n a l working models. The f i r s t involves the i d e n t i t i e s and expected locations of the attachment figures, and how they may be expected to respond. The next feature i s the c h i l d ' s a c c e p t a b i l i t y or lack of a c c e p t a b i l i t y to the attachment 36 figures. The t h i r d i s the use of i n t e r n a l working models i n the perception of events, the forecasting of the future and the construction of plans. F i n a l l y , i t i s most s i g n i f i c a n t whether the c h i l d f e e l s confident or a f r a i d regarding the a v a i l a b i l i t y of attachment figures - occasionally, frequently or most of the time. Such confidence, or lack of, depends on whether the attachment figures are judged to be the type of people who generally respond with support and protection and also whether the s e l f i s judged to be the type of person to whom others, p a r t i c u l a r l y the attachment figures respond with same. Although these two judgements are l o g i c a l l y independent they are i n practice completely intertwined. The models of attachment figures and s e l f are l i k e l y to develop i n complementary and mutually confirming manner (Bowlby, 1973). I f he receives consistently responsive and s e n s i t i v e behaviour with respect to his signals, he forms a representational model of the attachment figure as responsive and accessible and of himself as competent i n e l i c i t i n g her response and worthy of i t . Or, a l t e r n a t i v e l y , he may view himself as i n e f f e c t i v e i n obtaining her cooperation and as unworthy of i t (Crittenden & Ainsworth, 1989). Bowlby (1973) puts i t succinctly, "Whether a c h i l d or adult i s i n a state of security, anxiety or d i s t r e s s i s determined i n large part by the a c c e s s i b i l i t y and responsiveness of his p r i n c i p a l attachment fi g u r e . " (p.23). 37 Although models of the p r i n c i p a l attachment figures, generally the mother and then the father, are believed to be at f i r s t independent they eventually through some unknown process become more or less loosely integrated into a generalized model of attachment figures (Crittenden & Ainsworth, 1989). I t i s i n terms of t h i s generalized model, i n t r i c a t e l y intertwined with h i s model of s e l f , that he perceives h i s entire s o c i a l world. To varying degrees t h i s continues throughout the l i f e cycle. Towards an i l l u s t r a t i o n of t h i s point, one study by Crittenden and D i L a l l a (1988) made c l i n i c a l observations of elementary school c h i l d r e n involving unusually compliant and v i g i l a n t behaviours directed towards a l l adults, i n addition to t h e i r abusing parents. They claimed that, at the age of 3 6 months, the children had adopted such behaviour as an adaptive strategy based on a f l e x i b l e open model of r e l a t i o n s h i p s . Apparently between 36 months and 5 or 6 years the pattern had become r i g i d and unresponsive. In addition, Crittenden (1985), i n a study of 121 maltreating and adequate caregivers found that p h y s i c a l l y abusing caregivers appear to have inner models related to ideas of "power" and " s c a r c i t y of needed resources" (both psychological and p h y s i c a l ) ; t h e i r inner models involve them i n coercive and non-reciprocal interactions to meet t h e i r own needs. For neglecting caregivers the idea of " s c a r c i t y " i s combined with a sense of helplessness and despair. Only adequate caregivers are able to e s t a b l i s h r e c i p r o c a l 38 re l a t i o n s h i p s or perceive others as w i l l i n g and able to meet t h e i r needs and themselves as able to e n l i s t t h e i r support. In another study, Pastor (1981) described 62 securely and insecurely attached toddlers i n play si t u a t i o n s with the mother and a peer. Secure toddlers were more sociable and p o s i t i v e l y oriented towards both mother and peer, as were t h e i r mothers more supportive and appropriately d i r e c t i v e . Park and Waters (1989) tested 33 four-year-olds with t h e i r best friends i n a 1-hour free play session. They found that secure-secure pairs were happier, more responsive and harmonious and less c o n t r o l l i n g than secure-insecure p a i r s . In another empirical study, Slade (1987) describes the difference between "secure" and "insecure" childr e n not as a difference i n cognitive competencies per se but rather i n the way cognitive competencies inter a c t with s o c i a l competence. Secure children are better able to e n l i s t the support and " s c a f f o l d i n g " they need from the s o c i a l environment. When children f e e l secure about the a v a i l a b i l i t y of attachment figures when needed - competence, s e l f - r e l i a n c e and knowledge about the world i s fostered (Crittenden & Ainsworth, 1989). They are a c t u a l l y "freed to attend to other aspects of t h e i r l i v e s " (p.445). Attachment behaviour i s most intensely activated under s t r e s s f u l conditions evoking alarm or anxiety (Bowlby, 1973). I f an attachment figure i s sought without success or i f there i s doubt as to the a v a i l a b i l i t y of an attachment figure should one be needed the c h i l d w i l l f e e l anxiety. This fear 39 response to i n a c c e s s a b i l i t y to mother can u s e f u l l y be regarded as a basic adaptive response i n terms of species s u r v i v a l . In fact, when the c h i l d ' s attachment system i s active i t e l i c i t s caregiving behaviour from the parent (Shaver, Hazan, & Bradshaw, 1988). S i m i l a r l y an active parental caregiving system brings about behaviours that ease the c h i l d ' s d i s t r e s s . However, i t i s not only short-term anxiety but often anxious attachment that i s brought about by the kinds of experiences that shake a c h i l d ' s confidence i n the attachment figure's a v a i l a b i l i t y when needed (Bowlby, 1973) . The concepts of "continuity" and "change" are continually i n tension and interplay i n the empirically supported development of the t h e o r e t i c a l notion of the i n t e r n a l working model. Both continuity and change must be acknowledged i f the concept of the inner working model i s to be a tenable one. Although the majority of attachment studies demonstrate continuity, more or less successfully, some studies s i m i l a r l y focus upon change i n inner models and security of attachment (e.g., Gaensbauer & Harmon, 1982; Lamb et a l . , 1984; Thompson et a l . , 1982). At l e a s t two of these studies (Egeland & Farber, 1984; Vaughn, Egeland, & Sroufe, 1979) focussed upon the systematic nature of change i n inner models and security of attachment. Cassidy (1988), i n the following, describes aspects of the "continuity" notion: Although some of the pressure for continuity comes from 40 the environment, the working models organize and help mold that environment; by seeking p a r t i c u l a r kinds of people and by e l i c i t i n g p a r t i c u l a r behaviours from them, the in d i v i d u a l p a r t i c i p a t e s i n the creation of h i s or her own environment. A d d i t i o n a l l y , Bowlby suggests that working models also t r i g g e r defensive processes that act to s e l e c t i v e l y exclude c e r t a i n information from being processed and hence also contribute to continuity, (p.133) Sroufe and Fleeson (1986) focus on "expectations" as c a r r i e r s of continuity: Expectations are the c a r r i e r s of r e l a t i o n s h i p s . Carrying forward a l l of the s p e c i f i c behaviours and response chains from previous interactions would be an overwhelming task, but a l i m i t e d set of expectations can generate countless behavioural reactions, f l e x i b l y employed i n a v a r i e t y of s i t u a t i o n s . One's ori e n t a t i o n concerning others, one's expectations concerning t h e i r a v a i l a b i l i t y and l i k e l y responses, and what, i n general terms, one can do (or cannot do) to increase the l i k e l i h o o d of f a m i l i a r responses are strongly shaped by e a r l i e r relationships, (p.68) Several home observation studies (Ainsworth et a l . , 1978; Belsky, Rovine, & Taylor, 1984; Grossmann, Grossmann, Spangler, Suess, & Unzer, 1985; Maslin & Bates, 1983) 41 support a connection between infants' (inferred) expectations of r e j e c t i o n or responsiveness - and t h e i r secure versus insecure behaviours. In at l e a s t one study (Blehar, Lieberman, & Ainsworth, 1977) other simultaneously taken infant c h a r a c t e r i s t i c s f a i l e d to predict l a t e r s ecurity or ins e c u r i t y . Sroufe (1983) t i e d security i n infancy to self-esteem at the age of 4. Krazier, Fryer, & M i l l e r (1988) gave evidence supporting a l i n k between s e l f -esteem and the a b i l i t y to learn and use abuse-prevention s k i l l s . Bowlby (1973) and Epstein (1980) suggest the connection e x i s t s between self-esteem and expectations of s e l f and others. However, to date one study (Cassidy, 1988) has t r i e d but f a i l e d to support the claim. Additional support for the idea that expectations are the c a r r i e r s of relationships may be found i n various non-attachment studies. For example, Dodge and Richards (1985) showed aggressive children to be biased towards in t e r p r e t i n g agemates' behaviour as h o s t i l e or malicious. Waas (1988) compared two groups of low and high aggressive and rejected 3rd and 5th-grade boys to a group of non-rejected boys. When given no other s o c i a l information, the aggressive groups made more h o s t i l e a t t r i b u t i o n s of hypothetical peers depicted i n drawings, and suggested more h o s t i l e responses. Parke and Slaby (1983) demonstrated that aggressive childr e n were l i k e l y to come from homes i n which d i s c i p l i n e i s harsh and punitive. Hart, Ladd, & Burleson (1990) found that 5 to 7-year-old children whose mothers were more power-assertive 42 i n t h e i r d i s c i p l i n a r y s t y l e s received lower peer ratings on being l i k e d and were more l i k e l y to expect successful outcomes fo r unfriendly/assertive methods of resolving peer c o n f l i c t s . Ricks (1985) adapting Epstein's (1973, 1979, 1980) theory proposed that change i n inner working models may occur through change within the same early r e l a t i o n s h i p s across time. A l t e r n a t i v e l y , i t may occur through repeated experiences i n other relationships or through e s p e c i a l l y strong emotional experience within a single r e l a t i o n s h i p that disconfirms e a r l i e r postulates or models. Bretherton et a l . (1990) point out that despite constraints a c h i l d ' s inner model must adapt as the attachment r e l a t i o n s h i p develops. Needs that change with development must be r e f l e c t e d i n revisions to the inner model. They also mention external factors or changes a f f e c t i n g the attachment r e l a t i o n s h i p such as a developing chronic i l l n e s s or loss of the parent's job which may require further revisions to the c h i l d ' s model. In addition, they note that cognitive development can be presumed to a f f e c t the developing complexity of the c h i l d ' s inner model. Crittenden (1988a) proposed: the goal of intervention with both compliant and acting out abused children must be to engage them i n i n t e r a c t i o n with s e n s i t i v e l y responsive adults soon enough and long enough that t h e i r patterns of 43 i n t e r a c t i o n (and resultant i n t e r n a l representational models) are not l i m i t e d to those derived from i n t e r a c t i o n with t h e i r parents, (pp. 183-184) F i n a l l y , Bowlby (1988) d i r e c t l y refers to one commonality between the manner i n which models are formed and yet may also be changed: There are, i n fact, no more important communications between one human being and another than those expressed emotionally, and no information more v i t a l f o r constructing and reconstructing working models of s e l f and others than information about how each f e e l s toward the other, (pp. 156-157) Crittenden and Ainsworth (1989) point out the importance of having inner models not only accurately based on experience, but also open to new input and consequent adjustment. I t i s h e l p f u l to define some additional concepts i n order to elaborate upon t h i s point. Crittenden (1989) describes two additional dimensions of the inner working model, or inner "representational" model, delineated by Bowlby and l a r g e l y neglected or misunderstood i n the l i t e r a t u r e to-date. The extremes of these dimensions are "open" and "closed", "working" and non-working". Open models are open to new interpretations and 44 predictions. Closed models interpret a l l behaviour i n terms of the e x i s t i n g model. . . Working models allow cognitive manipulation of possible responses. Non-working models do not allow cognitive exploration of behavioural a l t e r n a t i v e s , (p.11) Critteden gives arguments based on her research f o r categorizing the models of abusive parents as "closed but working", neglecting parents as "closed and non-working", marginally-maltreating parents as "open but non-working" and adequate parents as "open and working". Given that persons generally develop several attachment rel a t i o n s h i p s , i t i s h e l p f u l to delineate the meta-structure of inner working models. The simplest one, which Crittenden suggests i s associated with abusing mothers, i s the "single i n t e r n a l representational model". Piaget's (1952) concept of "assimilation" i s relevant here such that a l l r e l a t i o n s h i p s are interpreted i n terms of the one model. More complex and consistent with r e a l i t y are "multiple, unrelated i n t e r n a l representational models". Since there i s one model for each relat i o n s h i p , the i n d i v i d u a l i t y of each re l a t i o n s h i p but not the coherency between them i s recognized. Marginally-maltreating mothers are associated with t h i s meta-structure. "Accommodation" of new information i s the associated pattern described by Piaget. Most complex and consistent with r e a l i t y i s the "generalized model with d i f f e r e n t i a t e d r e l a t i o n s h i p - s p e c i f i c submodels". Adequate mothers are associated with t h i s meta-structure, 45 displaying both assi m i l a t i o n of information about rel a t i o n s h i p s to the generalized model and accommodation of information to the s p e c i f i c submodels of re l a t i o n s h i p s . F i n a l l y , i t w i l l be useful to delineate the three best understood memory systems (Crittenden, 1989; Tulving 1979, 1985) . "Procedural" memory i s the f i r s t to develop and i s captured i n observations of infant behaviour i n the Strange Situ a t i o n . I t consists of f a m i l i a r behavioural routines c a r r i e d out unconsciously and extensively throughout the l i f e cycle. The "episodic" memory system i s believed to develop mostly a f t e r the age of f i v e (Fivush and Slackman, 1986) , and consists of s p e c i f i c personal memories encoded v i s u a l l y or l i n g u i s t i c a l l y . They are more e a s i l y open to conscious review and revisement than procedural memory. F i n a l l y , "semantic" memory i s conscious, impersonal, generalized and undated (Crittenden, 1989; Tulving, 1989). I t contains, among other things, generalizations about s p e c i f i c r e lationships and relationships i n general. I t consists of a person's own conclusions based on d i r e c t experience and on information received from others. Semantic memory i s not evident u n t i l a f t e r the development of representation (Bretherton, 1984; Crittenden, 1989). The representation of s e l f and others forms the context by which l a t e r experience i s interpreted (Sroufe, 1986). Representation would appear to be a form of generalization. The episodic and semantic memory systems (at least) contain both a f f e c t i v e and factual knowledge. Such 46 knowledge may vary between the memory systems f o r each i n d i v i d u a l . Since eit h e r form of knowledge may be dist o r t e d , the associated memory system may be di s t o r t e d accordingly. Each individual's inner working model i s as dis t o r t e d or accurate as i t s associated memory systems. In addition, i t i s probable that each memory system has i t s own inner working model. In any event, " i n d i v i d u a l s appear to d i f f e r i n the extent to which they can co-ordinate d i f f e r e n t memory systems" and "individuals may have concious access to a l l or only some of these memory systems" (Crittenden, 1989, pp. 8-9). The above delineations associated with inner working models are conceptual tools which should prove instrumental fo r two types of research. "Normal" developmental change may be explored using assessments which c l e a r l y specify the le v e l s of memory, and the dimensions and meta-structures of inner models being assessed. In addition, the rela t i o n s h i p s and the behavioural systems most relevant at the age l e v e l s being considered could be stated. To-date not many assessments i n t h i s general area have been directed p r i m a r i l y for the exploration of developmental change (Crittenden, 1989). These conceptual too l s may help to change that. The other relevant form of research i s that directed towards intervention. Behavioural, cognitive and psychodynamic therapies may be viewed as working with procedural, semantic and episodic models primarily, i n that 47 order. (Psychodynamic therapies may be viewed as directed towards uncovering episodic memories i n order to revise semantic models.) The two topics which Crittenden describes as most i n need of research and elaboration are "(a) the r e l a t i o n of the development of in t e r n a l representational models to treatment of children of d i f f e r e n t ages and, (b) determination of the model(s) ( i e . procedural, episodic, semantic or combination) with which to intervene" (p.27). 48 Chapter 3; Methodology  Overview This research uses an exploratory design to map out a group p r o f i l e i n terms of subjects' inner images or conceptualizations of attachment figures i n r e l a t i o n to s e l f . The three or four response categories used i n the primary measure are generated from theory (Bowlby, 1969, 1973, 1980; Bretherton et a l . , 1990) as opposed to s t a t i s t i c a l analyses such as c l u s t e r analysis. This study begins to gather empirical support for the v a l i d i t y of four response categories by demonstrating consistent patterns across categories, within categories and between inventories or measures. The response categories constitute the dependent var i a b l e . The independent variables are "caregiver" and " s i t u a t i o n " of which there are three and eight respectively i n the main measure. Non-parametric s t a t i s t i c s (loglinear) are used to determine the s i g n i f i c a n c e or i n s i g n i f i c a n c e of the independent variables across the response categories. Chi square s t a t i s t i c s are used to i l l u s t r a t e a dditional patterns across response categories. Parametric s t a t i s t i c s (anova) are used to demonstrate in t e r n a l consistency within each of the i n d i v i d u a l response categories for the primary measure. Descriptive s t a t i s t i c s are used to compare the two inventories. They are also used to show additional patterns within and across the response categories for the primary 49 measure. F i n a l l y , a short t e s t of understanding of some of the basic v i s u a l and verbal concepts required f o r adequate p a r t i c i p a t i o n i s described. This study does not attempt to l a b e l i n d i v i d u a l subjects i n the study. Rather, the subjects are viewed c o l l e c t i v e l y i n order to determine i f a consistent group p r o f i l e e x i s t s . The p a r t i c u l a r patterns inherent i n the group p r o f i l e are used to investigate the v a l i d i t y of the response categories i n the primary measure. Deviations from homogeneity within the group p r o f i l e are based on a normal curve confidence i n t e r v a l c r i t e r i o n . I t i s not suggested that o u t l i e r s that are defined as a r e s u l t of the use t h i s c r i t e r i o n have any diagnostic s i g n i f i c a n c e or v a l i d i t y . The c r i t e r i o n and the r e s u l t i n g o u t l i e r s serve only to describe the group p r o f i l e i n terms of the r e l a t i v e proportions of homogeneity and non-homogeneity with respect to the dependent vari a b l e . 50 Rationale for Design The establishment of the v a l i d i t y of the response categories generated by attachment t h e o r i s t s i s a f i r s t step; i t should precede t h e i r u t i l i z a t i o n i n comparative experimental studies, preventive work with children, etc. The exploratory design of t h i s study i s organized around an in v e s t i g a t i o n of the v a l i d i t y of i t s two measures. The primary measure i s the Permitting/Blocking Access Inventory. The secondary measure i s the Friendliness/Unfriendliness Inventory. The main advantage of non-parametric over parametric s t a t i s t i c s with regard to relationships across response categories i n t h i s case involves the extremely skewed nature of the d i s t r i b u t i o n . The normal d i s t r i b u t i o n requirement for the use of parametric s t a t i s t i c s could not be met across response categories (although i t could to a much greater degree within response categories). On the other hand, the repeated design manova s t a t i s t i c s had the advantage over l o g l i n e a r s t a t i s t i c s i n terms of the capacity to consider repeated measures. When repeated design manovas were attempted s i g n i f i c a n t "within subjects e f f e c t s " were found. These are most l i k e l y associated with the f a i l u r e to meet the normal d i s t r i b u t i o n requirements. Loglinear models are robust i n the face of repeated measures however, and are frequently used i n such cases. 51 The advantages of l o g l i n e a r over ordinary c h i square methods are several: the l a t t e r , according to Fienberg (1980): (a) confuses the marginal r e l a t i o n s h i p between a p a i r of categorical variables with the r e l a t i o n s h i p when other variables are present (b) does not allow f o r the simultaneous examination of these pairwise relationships, (c) ignores the p o s s i b i l i t y of three-factor and higher order interactions among the variables, (p.l) In addition, l o g l i n e a r models have the same modelling f l e x i b i l i t y as analysis of variance and regression, and they are also interpretable (Christiansen, 1990). Confidence i n t e r v a l s are determined for two purposes. In the f i r s t case they are used to demonstrate the degree to which the d i s t r i b u t i o n s within categories of the dependent var i a b l e are normal. This can provide the ra t i o n a l e f o r using anovas within response categories. That, i n turn, allows the two halves of the group to be compared as a check for i n t e r n a l consistency. Confidence i n t e r v a l s have the additional purpose i n t h i s study of providing a reasonable c r i t e r i o n by which subjects can be sorted with regard to homogeneity or non-homogeneity of response. Without such a c r i t e r i o n the degree of homogeneity associated with t h i s group p r o f i l e could not be described. This c r i t e r i o n may of course have to be adjusted i n the l i g h t of any future studies which contribute to establishing the v a l i d i t y of 52 these response categories or which describe other group p r o f i l e s . A simple but important means to describe the group p r o f i l e and support the v a l i d i t y of the constructs involves determining the percentage of subjects who chose each response category one or more times. This i s done fo r each of the four categories and for the two "blocking access" categories combined. These categories are further described by determining the p r o b a b i l i t y that a category w i l l be chosen one or more times when another category i s known to have been chosen one or more times. This provides evidence of the degree of dependence or independence of the response categories to each other. This i n turn, allows further speculation as to the interdependent nature of the categories. The secondary measure, the Friendliness/Unfriendliness Inventory, has the advantage of allowing descriptions of greater numbers of individuals and groups of i n d i v i d u a l s from the subject's point of view. There i s , however, no context or s i t u a t i o n described i n t h i s measure. The depicted persons are merely i l l u s t r a t e d i n poses. This allows i n e f f e c t a comparison across inventories of the r e l a t i v e d e s c r i p t i v e value of varying context on the one hand and lack of context or s i t u a t i o n on the other. At the same time, the subject's mother and father are depicted i n both inventories, allowing for more precise comparisons of subject response. 53 Chi square s t a t i s t i c s are used to demonstrate the s i g n i f i c a n c e l e v e l s of skewness across response categories i n both inventories. Histograms provide v i s u a l evidence of the s i m i l a r i t i e s and differences of the group d i s t r i b u t i o n s for the two inventories. Both l o g l i n e a r and manova models were inadequate for t h i s job due primarily to the i n s u f f i c i e n t number of responses c a l l e d for on the secondary measure. Seventeen of the 23 subjects were given a short t e s t of understanding using submeasures of the secondary inventory. They were f i r s t asked to point to the drawing which i s "the most f r i e n d l y and nice". Their s e l e c t i o n was then covered and they were asked to point to the drawing which i s "the most angry and unfriendly". From t h i s t e s t i t i s decided whether or not subjects appear to have an adequate understanding of the verbal and v i s u a l concepts required to p a r t i c i p a t e i n the study. 54 Subjects The subjects comprised one group of 23 f i v e to seven-year old boys i n kindergarten or grade one. At the time of the study there were eleven 5-year-olds, eight 6-year-olds and four 7-year-olds. The age range was 5 years, two months - to 7 years, 4 months with an average of 74 months. Most of the subjects are Caucasian. The 5 or 6 that are not Caucasian represent a number of e t h n i c i t i e s . A l l were attending school and l i v i n g i n a suburban neighbourhood i n Richmond, B.C. for at l e a s t nine months p r i o r to the study. Seventeen were l i v i n g at the time with both parents; 6 were l i v i n g only with t h e i r mothers and were v i s i t i n g regularly with t h e i r fathers. A l l boys had one of s i x d i f f e r e n t female teachers as h i s main teacher for that school year. Letters were sent to a l l 80 sets of parents with 5 to 7-year-old boys i n kindergarten and grade one. I t was offered to parents that any information gained about t h e i r c h i l d r e n would be shared with them. Belsky and Nezworski (1988) found t h i s basic courtesy to be a major motivator behind parental cooperation. T h i r t y sets of parents responded, 27 of them a f f i r m a t i v e l y . Of the 27, 26 requested to be informed about t h e i r sons' r e s u l t s a f t e r the study. Of the 27 boys, 1 of the 5-year-olds appeared extremely uncomfortable and stated c l e a r l y that he did not want to 55 leave the classroom to go with the t e s t e r . None of the other boys objected i n any noticeable way; i n f a c t , most of them were c l e a r l y eager to go. Some data were taken f o r each of them. Three other five-year-olds c l e a r l y were not able or w i l l i n g to cooperate adequately on the exercise and t h e i r data were not analyzed. The remaining 23 boys completed the e n t i r e exercise and t h e i r data were analyzed. School personnel were not aware of o f f i c i a l substantiation of abuse i n the case of any of the 23 subjects. However, one of the boys (#22) was believed by the school to have been sexually abused by someone whose i d e n t i t y was not known to the school. This was reported to the school by the parents. Another boy (#23) was suspected by the school to have been eithe r p h y s i c a l l y or sexually abused based on aspects of h i s behaviour. I t was decided for two reasons to include these subjects i n the analysis. The f i r s t i s simply that the abuse was not substantiated. The second and more important reason i s that t h i s group was not systematically selected to represent a completely non-maltreated population. A l l of the kindergarten and Grade 1 teachers at the school were female. Boys were selected over g i r l s for no c a t e g o r i c a l l y imperative reason. The school was chosen because the d i s t r i c t administrators, school p r i n c i p a l and teachers were w i l l i n g to p a r t i c i p a t e . A d d i t i o n a l l y , the school and i t s environment were reasonably representative of a middle-income suburban neighbourhood. The school i s situated on a quiet t r a f f i c - r e g u l a t e d street with l o t s of green playing space and the p r i n c i p a l and teachers were s i g n i f i c a n t l y f r i e n d l y and relaxed. Such a physical and s o c i a l environment could be expected to have a p o s i t i v e e f f e c t on the children. 57 The Interview Situation A l l subjects were introduced to the interviewer by the p r i n c i p a l or v i c e - p r i n c i p a l i n groups of 2 to 7 depending on the number of subjects i n any p a r t i c u l a r classroom. The introduction was b r i e f and informal with an opportunity provided for children to ask the interviewer questions and shake hands with him. Subjects were informed that they would be shown drawings and asked questions about them that they would be able to answer. They were t o l d as well that there would be an opportunity to colour, do puzzles and/or play a game of cards, as they wished. Within a few days of introduction each c h i l d was i n d i v i d u a l l y picked up at h i s classroom and escorted by the interviewer to the v i c e -p r i n c i p a l ' s o f f i c e . (Neither the v i c e - p r i n c i p a l nor her o f f i c e was associated with a d i s c i p l i n a r y role.) The o f f i c e i s about 10 by 12 and contains a desk and a c h i l d - s i z e table and set of 3 chairs. Stuffed animals were placed on the table within reach of any c h i l d who might desire them fo r a greater sense of security. Only 2 or 3 children a c t u a l l y held them. The games mentioned above were placed s l i g h t l y out of reach of the c h i l d u n t i l a break was required or the exercise was finished. Seventeen subjects f i n i s h e d the exercise without a break i n 15 to 20 minutes. The rest asked f o r a break eit h e r with or without prompts from the interviewer. A l l subjects spent 15 to 25 minutes i n relaxing a c t i v i t i e s before returning to t h e i r classrooms. 58 P r i o r to t h i s some subjects appeared s l i g h t l y fatigued and a few mentioned being t i r e d . No subject at any time appeared or mentioned being distressed. There were no reports from teachers or the p r i n c i p a l that the interview had disturbed any of the children. Seventeen of the children returned b r i e f l y to the interviewing o f f i c e within 10 days of i n i t i a l interviewing for less than 5 minutes each. Children were given a short t e s t to determine t h e i r degree of understanding of some of the e a r l i e r interviewing directions and also of t h e i r understanding of the f a c i a l expressions and body language inherent i n the drawings. This t e s t was not given p r i o r to the study i n order to avoid possible bias due to the formation of perceptual or cognitive sets formed as a r e s u l t of t e s t i n g . 59 The r e m i t t i n g / B l o c k i n g Access Inventory There are 8 submeasures of t h i s task, one for each s i t u a t i o n depicted. Each s i t u a t i o n i s depicted i n a separate series of 4 drawings for each of "mother", "father" and "main teacher". Therefore each submeasure contains 12 drawings. The c h i l d i s depicted i n every drawing with one of the three caregivers. However, i n 1 of the 8 submeasures the roles of " i n i t i a t o r " and "responder" are reversed between the c h i l d and each caregiver. In each of the three series of each submeasure an adult (or i n one case the child) i s e i t h e r permitting access (roughly means "responding s e n s i t i v e l y " ) , blocking access, or i s midway on the dimension of permitting/blocking access. In each case where the adult i s blocking access, he/she i s doing so i n e i t h e r an "ignoring" manner or i n a more e x p l i c i t l y "unfriendly/aggressive" manner. Each task i s presented to the c h i l d i n the following manner. "OK, B i l l y , imagine that i n each picture t h i s i s you. I t looks l i k e you are s i c k i n bed ( a l t e r n a t i v e l y "cut knee", e t c . ) . I also want you to imagine, B i l l y , that t h i s could be your dad (or mom or teacher) i n each one of these pictures. Which one of the four i s most-like your dad?" In the two or three instances when a c h i l d commented that the picture did not look l i k e h i s mom or dad ( s p e c i f i c a l l y "no moustache" or "hair i s too short") he was b a s i c a l l y asked to be a "good pretender" and "pretend r e a l hard". This 60 strategy seemed to take care of the problem at l e a s t insofar as no c h i l d persisted i n mentioning the problem again and a l l c h i l d r e n were then able to begin and complete the interview. S i m i l a r l y , i n the case of the reversed submeasure (" s e l f " ) , the task i s presented i n the following manner. "OK, B i l l y , imagine that i n each picture t h i s i s your mom (or dad or teacher). I t looks l i k e she wants to give you a hug. I also want you to imagine, B i l l y , that t h i s could be you i n each of these pictures. Which one of these four pictures i s most-like you?" For 11 of the subjects the 4 drawings i n each series were presented i n order from most "permitting" to l e a s t "permitting" with the "angry blocking access" card coming l a s t . For 12 of the subjects the most "permitting" card was switched with the "ignoring blocking access" card i n a d i f f e r e n t 2 of the 3 series i n each submeasure (counterbalancing). Each presentation unit consisted of a s e r i e s of 4 drawings, each on 8 1/2 by 11 pages and attached together to a s i n g l e piece of b r i s t o l board. 6 1 The Friendliness/Unfriendliness Inventory In t h i s inventory, persons, groups of persons and buildings are depicted. Neither r e l a t i o n s h i p s i t u a t i o n s nor the c h i l d himself are depicted. There are s i x + n submeasures of t h i s task, where n = the number of older s i b l i n g s plus the number of non-parental adults i n the home. The submeasures shared by a l l subjects are "mother", "father", "teachers", "peers", "school" and "home". Each submeasure involves the presentation of three drawings ranging from most f r i e n d l y on the l e f t to l e a s t f r i e n d l y on the r i g h t . This inventory was presented i n parts p r i o r to, i n the course of, and a f t e r the presentation of the Permitting/Blocking Access Inventory. The "mother", "father" and "peers" submeasures were presented f i r s t because they constituted simple tasks to get the children started. The submeasures for " s i b l i n g s or other in-home r e l a t i v e s " , "school" and "home" were considered to be les s important and so were placed at the very end when c h i l d fatigue as a poten t i a l cause of bias might be greatest. The submeasure for "teachers" was presented between the middle and the end, a f t e r seven of the eight Permitting/Blocking Access submeasures. 62 The P i l o t Study The p i l o t study was conducted with 4 subjects. Five-year-olds were not included i n order to create an opportunity to practice and re f i n e the procedures with more mature subjects. The p i l o t was approached with a readiness to a l t e r procedures as necessary or appropriate (e.g., counterbalance aspects of the presentation of tasks to t e s t for apparent forms of response bias, shorten and spread out the t e s t i n g sessions). The p i l o t had the following purposes: (a) To determine the optimal number of card tasks per session and the number of sessions required, (b) To t e s t the appropriateness of the age range for the card tasks, (c) To reveal any problems i n the procedures used to implement the card tasks, (d) To reveal any problems which the childre n might have i n understanding and responding to the d i r e c t i o n s , (e) To provide an opportunity to practice presenting and interviewing s k i l l s and to detect any problems i n the schedule and format, (f) To reveal any d i f f i c u l t i e s which the interview s i t u a t i o n might create with respect to the childrens' subjective feelings of stress or d i s t r e s s , (g) To reveal any obvious or poten t i a l forms of response bias 63 (h) To o t h e r w i s e d e t e r m i n e t h e a p p r o p r i a t e n e s s and s e n s i t i v i t y o f t h e i n s t r u m e n t f o r t h e o b j e c t i v e s o f t h e s t u d y . 64 Chapter 4: Results  Overview The Permitting/Blocking Access Inventory: Loglinear analysis determined a s i g n i f i c a n t r e l a t i o n s h i p between the variable " s i t u a t i o n " and the variable "response" beyond the .05 l e v e l . There was no s i g n i f i c a n t three-way in t e r a c t i o n between "response", " s i t u a t i o n " , and "caregiver". Loglinear analyses do not d i s t i n g u i s h between dependent and independent variables so the s i g n i f i c a n t i n t e r a c t i o n of "response" and " s i t u a t i o n " i s equivalent to a s i g n i f i c a n t main e f f e c t for " s i t u a t i o n " . S i m i l a r l y , there i s no s i g n i f i c a n t main e f f e c t for "caregiver" and no in t e r a c t i o n e f f e c t between " s i t u a t i o n " and "caregiver". Confidence l e v e l s were determined for each of the four response categories as well as for the combined "blocking access" categories. Two of 23 subjects were outside of the 95% confidence i n t e r v a l for a normal d i s t r i b u t i o n on the "permitting access" response category (1). At the 99% confidence l e v e l only one subject was excluded. These general r e s u l t s are i d e n t i c a l to the general r e s u l t s f o r the "midway on access" response category (2). In addition, the subject excluded from the 99% confidence i n t e r v a l f o r categories 1 and 2 was the only subject excluded from the 95% and 99% confidence i n t e r v a l s for the "ignoring" (3) and 65 the combined "blocking access" (3-4) categories. The "angry blocking access" (4) category had three completely d i f f e r e n t subjects excluded from the 95% confidence i n t e r v a l , but none for the 99% confidence i n t e r v a l . When the subject who was excluded from the 99% confidence i n t e r v a l for categories 1,2,3 and 3-4 - was l e f t out of the analysis, one new subject and one previously mentioned subject became excluded at the 99% confidence i n t e r v a l s for three response categories. Thirteen percent (3 of 23) of subjects were c l a s s i f i e d as o u t l i e r s . Anovas were applied to the i n d i v i d u a l collapsed response categories (1,2 and 3-4). In each case the response category constituted the dependent va r i a b l e . The independent variables were "caregiver", and " s i t u a t i o n " . The two groups ("halves") of subjects used i n the anova had received counterbalanced presentations of the inventory. No s i g n i f i c a n t differences were found between groups on each of the three analyses. One hundred percent of subjects selected response category 1 one or more times. In addition, 57%, 22%, 22% and 30% selected response categories 2,3,4 and 3-4, respectively, one or more times. Seventy percent of subjects selected categories 2,3 or 4 one or more times. The knowledge that a subject chose category 1 one or more times does not provide additional information about the p r o b a b i l i t y of a subject choosing any other response category one or more times. In t h i s sense category 1 i s 66 independent from the other categories. In the same sense categories 2 and 3 are p o s i t i v e l y dependent. Categories 2 and 4 are independent one way and negatively dependent the other. Categories 3 and 4 are p o s i t i v e l y dependent. Highly s i g n i f i c a n t c h i squares were found f o r the d i s t r i b u t i o n s of responses across response categories. The Friendliness/Unfriendliness Response Inventory Highly s i g n i f i c a n t c h i squares were found f o r the d i s t r i b u t i o n of responses across the three categories. They were skewed i n the same d i r e c t i o n as i n the primary inventory. A l o g l i n e a r analysis of the re l a t i o n s h i p between "caregiver" (mother and father only) and the two response variables from the two d i f f e r e n t inventories produced a nea r - s i g n i f i c a n t Pearson c h i square s t a t i s t i c (p_ = .06). Again the r e l a t i v e lack of data produced from t h i s inventory probably affected the re s u l t s of the analysis. 67 P i l o t Study Results I t was determined i n the p i l o t study that most of the procedures were unfolding as predicted. The children appeared to understand and to be able to follow the dir e c t i o n s appropriately. I t was decided to make two changes during the course of the p i l o t study. The "home" and "school" submeasures of the secondary inventory were determined to be inappropriate because the children tended to think that the bars were "blinds" or "curtains". A c t u a l l y t e l l i n g them "these are supposed to be bars" would of course defeat the purpose of the submeasure. In addition, i t was decided that the order of the presentation of response categories should be counterbalanced among subjects. This was concluded on the basis of a heavy s e l e c t i o n by the 4 subjects of the response category i n the f i r s t p o s i t i o n . A l l data from the 4 p i l o t subjects were treated i d e n t i c a l l y to the data from the other 19 subjects. The procedures followed for p i l o t and non-pilot subjects were i n every way i d e n t i c a l . 68 Study Results: Main E f f e c t s and Interactions Between Variables There i s a s i g n i f i c a n t r e l a t i o n s h i p between the var i a b l e " s i t u a t i o n " and the variable "response" beyond the .05 l e v e l . There i s no s i g n i f i c a n t three-way i n t e r a c t i o n between "response", " s i t u a t i o n " , and "caregiver". There i s no s i g n i f i c a n t i n t e r a c t i o n between "response" and "caregiver". Loglinear analyses do not d i s t i n g u i s h between dependent and independent variables so the s i g n i f i c a n t i n t e r a c t i o n of "response" and " s i t u a t i o n " i s equivalent to a s i g n i f i c a n t main e f f e c t for " s i t u a t i o n " . S i m i l a r l y there i s no s i g n i f i c a n t main e f f e c t for "caregiver" and no i n t e r a c t i o n e f f e c t between " s i t u a t i o n " and "caregiver" beyond the .05 l e v e l . No recoding was c a r r i e d out i n t h i s analysis. The Pearson chi square s t a t i s t i c for the f u l l 4 response categories i s not s i g n i f i c a n t beyond the .05 l e v e l (p_ = .092). There i s some doubt then that " s i t u a t i o n " i s s i g n i f i c a n t for the f u l l (1,2,3,4), though not the collapsed (1,2,3-4) model. This i s the case because for small-sample d i s t r i b u t i o n s the Pearson c h i square s t a t i s t i c i s more, accurate than the Likelihood Ratio c h i square s t a t i s t i c (Fienberg, 1980). This i s a l l the more true f o r samples such as the one i n t h i s study with many observed counts of "0" and "1". 69 Table 1 The Permitting/Blocking Access Inventory: Loglinear Analysis response categories variables df Likelihood Pearson p_ Ratio c h i c h i square square 1,2,3,4 response caregiver s i t u a t i o n 83 83 81.0 82.0 541 509 1,2,3-4 response caregiver s i t u a t i o n 60 60 67.2 .242 64.0 .337 1,2,3,4 response s i t u a t i o n 21 21 33.4 .042* 29.9 .092 1,2,3-4 response s i t u a t i o n 14 14 27.4 .017* 24.9 .035* 1,2,3,4 response caregiver 6 6 11.2 .080 11.4 .077 1,2,3-4 response caregiver 4 4 6.7 6.5 149 161 p_ < .05, Confidence Levels Confidence i n t e r v a l s are determined for each of the four response categories as well as for the combined "blocking access" categories. Two of 23 subjects are outside the 95% confidence i n t e r v a l for a normal d i s t r i b u t i o n on the "permitting access" response category. At the 99% confidence l e v e l only 1 subject i s excluded. These general r e s u l t s are i d e n t i c a l to the general r e s u l t s f o r the "midway on access" response category. In addition, the subject excluded from the 99% confidence i n t e r v a l f o r categories 1 and 2 i s the only subject excluded from the 95 and 99% confidence i n t e r v a l s for the "ignoring" and the combined "blocking access" categories. The "angry blocking access" category has 3 completely d i f f e r e n t subjects excluded from the 95% confidence i n t e r v a l , but none for the 99% confidence i n t e r v a l . Table 2 Confidence l e v e l s based on 23 subjects Subscale Permitting Midway Ignoring Angry Collapsed (1) (2) (3) (4) (3-4) Subjects 23 23 23 23 23 Total Responses 484 48 12 8 20 Mean 21 2.08 .5 .35 .9 71 Table 2 (continued) Subscale Permitting Midway Ignoring Angry Collapsed (1) (2) (3) (4) (3-4) Sum of squares 381 218 41.75 11.22 48.22 Standard deviation 4.07 3.08 1.35 .70 1.45 68% C.I. 17 to 24 0 to 5 0 to 1.9 0 to 1 0 to 2.4 68% of N 16 16 16 16 16 No. i n 68% C.I. 19 20 21 20 20 95% C.I. 15 to 24 0 to 8 0 to 3 0 to 1.8 0 to 3.7 95% of N 22 22 22 22 22 No. i n 95% C.I. 21 21 22 20 22 Identity of Ss. out #15,#21 #15,#21 #21 #8,#13,#21 #21 99% C.I. 12.5 to 24 0 to 10 0 to 4 0 to 2 0 to 4.6 99% O f N 23 23 23 23 23 No. i n 99% C.I. 22 22 22 23 22 Identity of Ss. out #21 #21 #21 - #21 When subject #21 (see Table 3) i s excluded from the analysis because of h i s obvious " o u t l i e r " status, subjects 72 #10 and #15 are excluded from the 99% confidence i n t e r v a l i n one and i n two response categories, respectively. Three subjects then, may reasonably be c l a s s i f i e d as o u t l i e r s -#10, #15, and #21 - on the basis of being excluded at one or the other 99% C.I. The use of a cut-off point approximately two standard deviations from the mean i s not unusual i n psychological studies. The importance of i t s a r b i t r a r i n e s s i s discussed i n Chapter 5. Three other subjects can perhaps be considered to be "borderline o u t l i e r s " - #8, #23, and #13 - on the basis of being excluded from one or more 95% C.I. when subject #21 i s l e f t out of the analysis. Thirteen percent (3 of 23) of subjects are c l a s s i f i e d as o u t l i e r s . I f the 3 borderline o u t l i e r s are included then 26% of subjects are c l a s s i f i e d as o u t l i e r s . 73 Table 3 Confidence l e v e l s based on 22 subjects Subscale Permitting Midway Ignoring Angry Collapsed (1) (2) (3) (4) (3-4) Subjects 22 22 22 22 22 Total Resp. 477 37 6 8 14 Mean 22.00 1.68 .27 .36 .64 Sum of squares 185 137 10.38 11.09 24.40 95% C.I. 16.2-24 0-6.6 0-1.6 0-1.8 0-2.7 95% of N 21 21 21 21 21 No. i n 95% C.I. 19 21 21 19 20 Identity of Ss. out #10,#13,#15 #15 #10 #8,#13,#23 #10,#13 99% C.I. 14.3-24 0-9 0-2.1 0-2.1 0-3.3 99% of N 22 22 22 22 22 No. i n 99% C.I. 21 21 21 22 22 Identity of Ss. out #15 #15 #10 Table 4 Summarized data for a l l subjects i n both inventories Inventories F./U. I. P./B. I. Subscales Subjects A B C 1 2 3 4 1 2 1 1 20 04 0 0 2 5 1 0 23 01 0 0 3 4 2 0 22 02 0 0 4 4 0 0 24 00 0 0 5 2 1 0 24 00 0 0 6 3 1 0 24 00 0 0 7 3 0 1 24 00 0 0 8 3 1 0 22 00 0 2 9 5 0 0 23 01 0 0 10 4 1 0 16 05 3 0 11 1 2 2 20 02 1 1 12 5 1 0 23 01 0 0 13 3 1 0 15 06 1 2 14 3 1 1 23 01 0 0 15 5 1 0 14 10 0 0 16 4 1 0 22 02 0 0 75 Table 4 (continued) Inventories F./TJ. I. P./B. I. Subscales Subjects A B C 1 2 3 4 17 6 0 0 24 00 0 0 18 3 1 0 22 00 1 1 19 2 2 0 24 00 0 0 20 5 0 0 24 00 0 0 21 3 1 1 07 11 6 0 22 4 0 0 22 02 0 0 23 2 3 0 22 00 0 2 Note. F.U./I. - Friendliness/Unfriendliness Inventory P.B./I. - Permitting/Blocking Access Inventory Internal Consistency The r e s u l t s of anovas.are given i n Table 5 for the collapsed response categories, "permitting access" (1), "midway on access" (2) "blocking access" (3-4). In no instance was the variable "halves" s i g n i f i c a n t e i t h e r as a main e f f e c t or as part of an i n t e r a c t i o n . This v a r i a b l e corresponds to the two groups of 11 and 12 subjects who were presented the drawings i n counterbalanced order. Presentation order was varied i n t h i s manner i n order to check f o r one form of response bias. The group of 23 subjects could have been s p l i t up i n numerous other ways as well i n order to demonstrate in t e r n a l consistency. The main e f f e c t s and the in t e r a c t i o n of " s i t u a t i o n " and "caregiver" are s i g n i f i c a n t for the "midway on access" response category. These r e s u l t s should be subsumed to the l o g l i n e a r r e s u l t s for the c o l l e c t i v e l y analyzed response categories. 77 Table 5 Anova r e s u l t s within response categories Response category Main e f f e c t Z Sig. or i n t e r a c t i o n of F Permitting (1) halves .02 .909 caregiver .47 .717 s i t u a t i o n 1.02 .644 halves by caregiver . 22 .834 halves by s i t u a t i o n 1. 00 .650 caregiver by s i t u a t i o n .62 .774 halves by caregiver by s i t u a t i o n .23 .942 Midway (2) halves 1.49 .231 s i t u a t i o n 1.59 .173 halves by s i t u a t i o n .96 .478 halves 1.58 .216 caregiver 2.34 .108 halves by caregiver . 10 .905 s i t u a t i o n 2.79 .028* caregiver 4.16 .028* s i t u a t i o n by caregiver 2.59 .019* Table 5 (continued) Response category Main e f f e c t or i n t e r a c t i o n Sig. of F Combined Ignoring and Angry (3-4) halves .02 .909 caregiver .29 .795 s i t u a t i o n .33 .873 halves by caregiver .03 .970 halves by s i t u a t i o n .23 .927 caregiver by s i t u a t i o n .18 .968 halves by caregiver by s i t u a t i o n .15 .977 Note. No F given f o r response by halves by caregiver by s i t u a t i o n - for "midway on access" response category p_ < .05, 79 Analyses Based on a Subject's Choice of a Response  Category One or More Times One hundred percent of subjects selected the "permitting access" (1) response category one or more times. Fifty-seven percent (13 of 23) selected the "midway on access" (2) category one or more times. Twenty-two percent (5 of 23) selected each of "ignoring" (3) and "angry (4) blocking access" response categories one or more times. Clearly, the "22%" for each of categories 3 and 4 represents two d i f f e r e n t but overlapping subgroups since 30% (7 of 23) of subjects were found to choose the collapsed "blocking (3-4) access" category one or more times. Seventy percent (16 of 23) of subjects chose categories 2, 3 and/or 4 one or more times. Table 6 P r o b a b i l i t i e s for response categories or more Is 100% or more 2s 57% or more 3s 22% or more 4s 22% or more 3s or 4s 30% The prob. of 1 or more 3s given 1 or more 2s 4/9=44% The prob. of 1 or more 4s given 1 or more 2s 2/11=18% The prob. of 1 or more 3s or 4s given 1 or more 2s 4/9=44% The prob. of 1 The prob. of 1 The prob. of 1 The prob. of 1 The prob. of 1 80 Table 6 (continued) The prob. of 1 or more 2s given 1 or more 3s 4/5=80% The prob. of 1 or more 3s given 1 or more 4s 3/5=60 % The prob. of 1 or more 4s given 1 or more 3s 3/5=60 % The prob. of 1 or more 2s given 1 or more 4s 2/5=40 % The knowledge that a subject chose response category 1 one or more times does not provide additional information about the p r o b a b i l i t y of a subject choosing any other response category one or more times. In t h i s sense categories 2 and 3 are c l e a r l y p a r t i a l l y dependent since the knowledge that a subject has chosen one of them 1 or more times increases the p r o b a b i l i t y that the subject has chosen the other 1 or more times from 22% to 44% or from 57% to 80%. The p r o b a b i l i t y of one or more 3s given one or more 2s i s the same (44%) as the p r o b a b i l i t y of one or more 3s or 4s given one or more 2s. However, the knowledge that there are one or more 4s could a c t u a l l y lower the p r o b a b i l i t y of one or more 2s from 57% to 40%. From t h i s data 2 and 4 are independent one way and negatively dependent the other. On the other hand, categories 3 and 4 are c l e a r l y p o s i t i v e l y dependent since the knowledge that there are one or more of either 3 or 4 increases the p r o b a b i l i t y of there being one or more of the other from 22% to 60%. 81 Comparison of the Permitting/Blocking Access and  Friendliness/Unfriendliness Inventories A l o g l i n e a r analysis of the re l a t i o n s h i p between "caregiver" (mother and father only) and the two response variables from the two inventories produced a near-s i g n i f i c a n t Pearson c h i square s t a t i s t i c (p_ = .06). The Likelihood-Ratio chi square had a much higher l e v e l of p r o b a b i l i t y , but the Pearson i s more accurate for small samples such as t h i s one (Fienberg, 1980) . Again the r e l a t i v e lack of data produced from t h i s inventory affected the r e s u l t s of the analysis. This was due to the fact that l o g l i n e a r analysis requires that a l l data f o r a l l variables be consecutively numbered with each numerical category represented. Substantial recoding was required r e s u l t i n g i n the loss of information. The r e l a t i o n s h i p between d i s t r i b u t i o n s across inventories i s perhaps best expressed v i s u a l l y (see Figure 1). The f i r s t (1), second (2) and t h i r d (3-4) response categories of the collapsed Permitting/Blocking Access Inventory are compared to the f i r s t (A), second (B) and t h i r d (C) response categories of the Friendliness/Unfriendliness Inventory. The comparison i s made fo r both f u l l and p a r t i a l versions of both inventories. The f u l l version compares the data for a l l "caregivers" i n the primary inventory and a l l "others" i n the secondary inventory. The p a r t i a l version compares "mother" and "father" only across the two inventories. 8 1 b FIGURE 1: Comparison of Permit t ing /Blocking Inventory and F r i e n d l i n e s s / Unfr iendl iness Inventory for A l l Caregivers and Others , and for Mother, Father Only 100-• 90 80 70 60 -• 50 -• 40 30 20 . . 10 10G-90 . . d) 80 X in c 0 a in Q) L U-• 70 60-50 -0) D) (D •P C a) u aj 40 + 30 • 20 10 Tst" 2 n d 3 r d R e s p o n s e C a t e g o r i e s F u l l Vers ion 1 s t 2 n d 3 r d Mother, Father only Permit t ing /Blocking Inventory 88:9:3 ( f u l l ) F r i e n d l i n e s s / U n f r i e n d l i n e s s Inventory 75:20:5 ( f u l l ) Both inventories 88:8:4 ( p a r t i a l ) 65:31:4 ( p a r t i a l ) Skewness towards "most f r i e n d l y " i n the case of the secondary measure and towards "most permitting" i n the case of the primary measure i s obvious; t h i s i s r e f l e c t e d i n the highly s i g n i f i c a n t (p < .001) c h i squares for the f u l l and p a r t i a l versions of each measure i n Tables 7, 8 and 9. Table 7 Significance of the d i s t r i b u t i o n across response categories  for the f u l l version of the primary measure Inventory Permitting/Blocking Access No. of subjects 23 Variables caregiver, s i t u a t i o n Categories 3 8 Responses/subject 24 Total responses 552 Response category Permitting Midway Ignoring Angry (1) (2) (3) (4) Total responses 484 48 12 8 Proportion 88% 9% 2% 1% chi square = 1176.8 c r i t i c a l value for df=3 i s 16.3 1176.8 > 16.3 * p_ < .001 83 Table 8 Significance of the d i s t r i b u t i o n across response categories  for a reduced version of the primary measure Inventory Permitting/Blocking Access No. of subjects 23 Variables caregiver s i t u a t i o n Categories 2 (mother, father) 7 ("self" excluded) Responses/subject 14 Total responses 322 Response category Permitting Midway Blocking (1) (2) (3-4) Total responses 284 25 13 Proportion 88% 8% 4% chi square = 605.2 c r i t i c a l value for df=2 i s 13.8 605.2 > 13.8 p_ < .001 84 Table 9 Significance of the d i s t r i b u t i o n s across response categories  fo r a reduced version and the f u l l version of the secondary  measure (A) (B) Inventory Friendliness/ F riendliness/ Unfriendliness Unfriendliness No. of subjects 23 23 Variable other other Categories mother, father mother, father, peers, teachers, r e l a t i v e s No. of categories 2 5 Responses/subj ect 2 4-6 Total responses 46 110 (A) Response category Most f r i e n d l y Midway on Least f r i e n d l y f r i e n d l i n e s s Total responses 30 14 2 Proportion 65% 31% 4% ch i square = 29.84 c r i t i c a l value f o r df=2 i s 13.8 29.84 > 13.8 85 Table 9 (continued) (B) Response category Most f r i e n d l y Midway on Least f r i e n d l y f r i e n d l i n e s s Total responses 82 22 6 Proportion 75% 20% 5% chi square = 57.2 c r i t i c a l value for df=2 i s 13.8 57.2 > 13.8 E < .001 86 One s l i g h t advantage of the secondary inventory i s that i t explores inner images for a wider v a r i e t y of objects (see Table 10). Unfortunately, the submeasures of "home" and "school" are of no use since many of the subjects assumed the bars on the windows to be "blinds" or "curtains". The submeasures of "mother" and "father" were the most useful since they could be d i r e c t l y compared on the primary measure. The r e l a t i o n of the d i s t r i b u t i o n to primary measure d i s t r i b u t i o n s i s observable i n Figure 1. Table 10 Summary of data from Friendliness/Unfriendliness Inventory Response category (A) (B) (C) Variable "others" mother 14 9 0 father 16 5 2 teachers 18 4 1 r e l a t i v e s 13 3 2 peers 21 1 1 87 Subcategories Within the Ignoring Blocking Access  Response Category A p o s t e r i o r i , i t was noted that the "ignoring blocking access" response category was depicted i n 2 rather d i f f e r e n t ways. In 4 of the situations ("bully", "cut knee", "smiling c h i l d reaching", "distressed c h i l d reaching") a rear p r o f i l e of the caregiver i s given. In 2 of the si t u a t i o n s ("self", "puzzle-schoolwork") the p r o f i l e of the face and body i s shown. The other 2 situations ("sick", " s p i l t milk-late") depict hybrid "ignoring" versions for each caregiver or across caregivers. Six of the 12 selections of the "ignoring" category for the group occurred i n the f i r s t set of 2 si t u a t i o n s distinguished above. Only 1 of the 12 occurred i n the set of 4 situations distinguished above. I f the hybrid versions are included such that the "ignoring" response category i s divided into a set that shows f a c i a l features and a set that does not, then i t i s found that 11 of 12 "ignoring" selections depicted f a c i a l features. Furthermore, there were 13 p o s s i b i l i t i e s without f a c i a l features to choose from compared to 11 with f a c i a l features. The 2 sets - with and without f a c i a l features - are d i s t r i b u t e d almost equally across caregivers. 88 Test for Understanding Seventeen of the 23 subjects whose data were used i n t h i s study and 1 subject whose data were not used were given a short t e s t of understanding using submeasures of the secondary inventory. Subjects were shown the submeasures for "teachers", "peers", "mother", "father", " s i s t e r " , and "brother" one at a time. They were f i r s t asked to point to the drawing which i s "the most f r i e n d l y and nice". Their s e l e c t i o n was then covered and they were asked to point to the drawing which i s "the most angry and unfriendly". One c h i l d got 2 of 12 responses correct. This c h i l d was already strongly suspected of not being able to respond on the basis of adequate understanding of the measures. His data were not used at a l l i n t h i s study. The other 17 subjects each received a minimum of 10 correct scores. There were a t o t a l of 6 errors from f i v e subjects out of a possible 204. A l l errors involved confusions between the "most f r i e n d l y " and "midway on f r i e n d l i n e s s " categories. Three subjects who made single errors were retested on the submeasures. Two of the 3 corrected t h e i r error and no new errors were made. I t i s concluded that subjects appear to have an adequate understanding of the verbal and v i s u a l concepts required to p a r t i c i p a t e i n the study. 89 Chapter 5: Discussion:  Interpretation of Cognitive. A f f e c t i v e or Response Bias i n the Results With respect to the study goal of representing childrens' i n t e r n a l images or conceptualizations, i t i s not necessary to demand that such images be free of the e f f e c t s of cognitive or a f f e c t i v e bias. These images, as aspects of the childrens' inner working models, are subject to the same forms of bias as inner models i n general. This study attempts to access the inner working model pr i m a r i l y at the l e v e l of semantic memory (P. Crittenden, personal communication, June, 1990). I t i s possible that s a l i e n t episodic memories also enter d i r e c t l y into the r e s u l t s for the more concrete s i t u a t i o n - s p e c i f i c measure. However, i t i s thought f o r the most part that episodic memory i s only being accessed i n d i r e c t l y v i a semantic memory. Both semantic and episodic memory are subject to cognitive and a f f e c t i v e d i s t o r t i o n . One form of bias associated with the semantic l e v e l of memory i s the i d e a l i z a t i o n of the parent. The highly skewed r e s u l t s towards the "permitting access" response category and the subsequent c e i l i n g e f f e c t suggest the p o s s i b i l i t y that t h i s form of c o g n i t i v e / a f f e c t i v e bias i s present. On the other hand, Bowlby (1973, 1980) and Main et a l . (1985) discuss i d e a l i z a t i o n of the parent as t y p i c a l l y associated with r e j e c t i o n and low self-esteem. 90 There i s no evidence of low self-esteem suggested i n the childrens' images of t h e i r own responses to caregivers. Furthermore, there i s no s i g n i f i c a n t difference i n the way the group views the three caregivers. I f i d e a l i z a t i o n of parents i s a factor i n t h i s group p r o f i l e then i t s e f f e c t s , or the e f f e c t s which i t represents, have generalized to include the three categories of caregivers. In addition, to the extent that the concrete s i t u a t i o n - s p e c i f i c nature of the primary measure can access episodic memory more or less d i r e c t l y , any " i d e a l i z a t i o n of parent" e f f e c t could be reduced. This, of course, assumes that the i d e a l i z a t i o n of the parent occurs v i a the semantic l e v e l of the inner model and that t h i s l e v e l might be bypassed. Cassidy and Kobak (1988) state that "avoidance" i s central to i d e a l i z a t i o n of the s e l f and deactivation of the attachment system at the representational l e v e l . They suggest that "avoidance" i n p a r t i c u l a r and attachment organization i n general can influence the e f f e c t s of experimental observation on s e l f - r e l a t e d representations. Of course, the attachment organization of these subjects i s not known. S u p e r f i c i a l examination of the data could indicate that there i s l i t t l e evidence of "avoidance" as a form of c o g n i t i v e / a f f e c t i v e bias. This i s based on the r e l a t i v e l y large proportion of "permitting access" responses and on the i n t e r p r e t a t i o n that t h i s indicates a c t i v a t i o n rather than deactivation of the attachment behavioural system. Such an interpretation i s not j u s t i f i e d f o r at 91 l e a s t two reasons. A r e l a t i v e l y small proportion of "non-permitting access" responses may be i n d i c a t i v e of cognitive and a f f e c t i v e bias. Secondly, images of responsive caregivers do not necessarily indicate a lack of deactivation of the attachment behavioural system amongst the chil d r e n i n general. That s t i l l requires that a substantial inference be made - although i t i s tempting to do so. We do know four relevant things about childr e n of t h i s age who have avoidant attachment organizations. (a) They tend to represent themselves i d e a l l y . (b) They tend to represent others as distant (Cassidy, 1988). (c) They are able to acknowledge d i s t r e s s i n themselves (Kaplan & Main, 1985). (d) They don't tend to reach out to others (Cassidy, 1988). The t h i r d fact i s relevant because seven of the eight s i t u a t i o n s depict a c h i l d i n d i s t r e s s . The f i r s t point could be relevant i f i t could be assumed that the childrens' representations of themselves as "permitting of access" are i n d i c a t i v e of ideal self-representation. One s t i l l has to interpret the exceptions, however. The tendency to represent others as distant i s most relevant, but again the p o s s i b i l i t y must be faced that a r e l a t i v e l y small proportion of "non-permitting access" responses may be i n d i c a t i v e of "avoidance" or avoidant attachment organization. The fourth point can be assessed most e a s i l y and d i r e c t l y i n the childrens' expectations of t h e i r own responses to the caregivers. Six children selected the "midway on access" or "ignoring" response categories. None 92 selected the "angry" response category. One c h i l d selected each of these two categories once (#13). One selected the "ignoring" category twice and the "midway on access" category once (#21). Interestingly, these children achieved o u t l i e r or borderline o u t l i e r status. The other four childr e n each selected the "midway on access" category once. The drawings presented to the children can also be viewed as s t i m u l i i n themselves. Several of the sit u a t i o n s as well as the "angry" caregiver responses, i f not the "ignoring" caregiver responses, could act as s t i m u l i s u f f i c i e n t to activate the attachment behavioural system. I t i s possible that i n one or more cases the subject was i n e f f e c t withdrawing from a noxious s t i m u l i and approaching a p o s i t i v e s t i m u l i when he chose the "permitting access" response category. I t would be i n t e r e s t i n g to present the drawings to a group of children with the most noxious stimulus, the "angry" caregiver response category removed. There i s the p o s s i b i l i t y of at le a s t two other general forms of response bias i n e f f e c t i n t h i s study. F i r s t , the childrens' responses could be influenced by a desire to respond as they imagine the interviewer wants them to respond (observer e f f e c t ) . Piaget (1924) and Piaget and Inhelder (1948) demonstrated that children under the age of 7 have great d i f f i c u l t y seeing from another's point of view ("egocentrism"). This fact reduces but does not eliminate the p o s s i b i l i t i e s of t h i s form of response bias. There i s i n addition the p o s s i b i l i t y of a general p o s i t i v i t y bias, 93 that i s , a tendency to select only " p o s i t i v e " behaviours. Again the notable c e i l i n g e f f e c t for the "permitting" response category i s suggestive. Two sets of studies are relevant although the point remains unsettled. Reider and C i c c h e t t i (1989) demonstrated that 4 to 9-year-old boys but not g i r l s "sharpen to" aggressive cues. One can conclude from t h i s that the male subjects i n t h i s study were at lea s t l i k e l y to accurately note the "angry blocking" response category even though i t was the lea s t selected. In addition, a p a i r of studies by Rholes and Ruble (1984) showed that 5 to 7-year-old children did not l a b e l the behaviours of other children as p o s i t i v e l y as d i d 9 to 10-year-olds even though they were demonstrated to be capable of doing so. I t can not be ruled out that the r e s u l t s of t h i s study are subject to mood or other short-term within subject factors. However, the o v e r a l l homogeneity and general patterns of responses for the group suggest the minimization of the e f f e c t s of such factors. This i s important i n terms of the r e l i a b i l i t y of the study. In addition, a l l subjects appeared to be fresh and healthy versus t i r e d or i l l . One subject (#21) seemed p a r t i c u l a r l y sombre throughout the interview. Although very p o l i t e , he never came close to cracking a smile. This was noted at the time as unusual. However, i t i s not known to what extent t h i s was t y p i c a l of the boy or merely a short term c h a r a c t e r i s t i c . 94 General Discussion P a r t i c u l a r l y with an exploratory design such as t h i s one a f a i l u r e to prove some of one's hypotheses i s not overly consequential. They are mentioned only i n passing, t h e i r usefulness having been expended. One such hypothesis was that about 75% to 80% of subjects would show a primarily "permitting" p r o f i l e and 20% to 25% - a p r i m a r i l y "blocking" p r o f i l e . A p rimarily "permitting" p r o f i l e f o r an i n d i v i d u a l would consist of about 75% to 80% "permitting" selections (1) and 20% to 25% "blocking" selections (3-4). The "midway on access" category was not i n i t i a l l y conceived of as a v a l i d category i n i t s e l f . I t was planned that i f a subject chose response category 2 he would be asked some questions l i k e , "what i s he/she doing? what i s he/she thinking? what i s he/she feeling?" I t was hypothesized that the c h i l d ' s answers would allow c l a s s i f i c a t i o n i n categories 1,3 or 4, or perhaps i t would provide some unexpected information. I t was i n part a method to "open up" the closed forced-choice categories. I t was a d d i t i o n a l l y a way to " f i l l out" the dimension or continuum from "permitting" to "blocking access". I t was also hypothesized that the variable "caregiver" would be shown to be moderately s i g n i f i c a n t and at l e a s t as s i g n i f i c a n t as the variable " s i t u a t i o n " . The variable "caregiver" was demonstrated not to be s i g n i f i c a n t . The variable " s i t u a t i o n " was demonstrated to be s i g n i f i c a n t , at least for the collapsed response 95 categories (1,2,3-4). The second response category indeed served one of i t s functions, namely the " f i l l i n g out" of the "permitting" to "blocking access" dimension. I t was selected 8% of the time, twice as often as the combined "blocking access" categories. I t was also chosen by nearly twice as many children (57% vs. 30%). Furthermore, once a c h i l d chose t h i s response category he gave the interviewer no reason whatsoever to r e c l a s s i f y i t i n one of the other categories. Even the one subject (#21) most deserving of the term " o u t l i e r " did not match the description of the primarily "blocking" p r o f i l e f or an i n d i v i d u a l . His p r o f i l e i s about 25% "permitting", 25% "blocking" and 50% "midway on access". 7 in d i v i d u a l s show a 100% "permitting" p r o f i l e . Ten others show a 92% to 96% "permitting" p r o f i l e . C learly, the "permitting" and "midway on access" response categories were selected much more often than was expected and the "blocking access" categories much less often than expected. As a r e s u l t , the c r i t e r i o n used to di s t i n g u i s h o u t l i e r s from a r e l a t i v e l y homogeneous group, i f there i s to be any c r i t e r i o n at a l l , appears more stringent i n i t s e f f e c t than would have been expected on the basis of e a r l i e r hypotheses. In other words, using normal d i s t r i b u t i o n confidence i n t e r v a l s as the c r i t e r i o n to define an o u t l i e r , a 67% "permitting access", 21% "midway on access", and 12% "blocking access" p r o f i l e (#10) i s s u f f i c i e n t i n one case to define an o u t l i e r . Using more stringent confidence i n t e r v a l 96 c r i t e r i a to define "borderline o u t l i e r s " causes two subjects (#8, #23) with 92% "permitting" and 8% "angry blocking" access p r o f i l e s to be so defined. This defines a group p r o f i l e described by having 13% (3 of 23) o u t l i e r s , or 26% (6 of 23) o u t l i e r s or borderline o u t l i e r s . These c l a s s i f i c a t i o n s do not have diagnostic v a l i d i t y or si g n i f i c a n c e . The confidence i n t e r v a l c r i t e r i a are somewhat ar b i t r a r y and future studies may develop more useful, more empirically based and/or less a r b i t r a r y c r i t e r i a . They are used here to describe the group i n terms of i t s s i m i l a r i t i e s and differences along a rather l i m i t e d number of dimensions or variables ( i e . "caregiver", " s i t u a t i o n " , "response"). O u t l i e r s and borderline o u t l i e r s have an operational or s t a t i s t i c a l d e f i n i t i o n as used i n t h i s study. However, i t i s premature to give them a t h e o r e t i c a l d e f i n i t i o n . Subjects who meet t h i s operational d e f i n i t i o n , and subjects who do not meet i t , need to be compared on other psychological measures whose v a l i d i t y and r e l i a b i l i t y are established. There are several other patterns that can be distinguished across response categories and inventories. These patterns provide support for the relevance, the non-haphazardness, and the in t e r n a l consistency of the subjects 7 selections across or within the response categories. By t h i s means, the patterns provide - and are submitted as -support for the v a l i d i t y of the response categories and t h e i r associated construct, "inner images" or 97 "conceptualizations". Stronger evidence of v a l i d i t y w i l l require future studies with s i m i l a r and d i f f e r e n t groups of subjects. The most obvious pattern across categories i n both inventories i s the highly s i g n i f i c a n t skewed nature of the d i s t r i b u t i o n s . They are s i m i l a r d i s t r i b u t i o n s but not to a s t a t i s t i c a l l y s i g n i f i c a n t degree. I t i s the case, however, that i f the f i r s t two categories of the Friendliness/Unfriendliness measure are collapsed the category r a t i o changes from 15:7:1 to 22:1. The collapsed (1,2,3-4) Permitting/Blocking measure response category r a t i o i s 22:2:1. A case could be made for c o l l a p s i n g (in the "Friendliness/Unfriendliness ratio) the f i r s t two categories on the basis that the second category r e a l l y i s a l o t more l i k e the f i r s t category than the t h i r d . I t was suspected that there would be a higher proportion of "permitting access" responses than "most f r i e n d l y " responses pertaining to these parents. There are two related reasons for t h i s . F i r s t l y , the parents of t h i s subject group were c l e a r l y not selected i n order to form any kind of high-risk p r o f i l e group. Secondly, the s i t u a t i o n s depicted were i n t e n t i o n a l l y constructed around issues of c h i l d stress and d i s t r e s s . In general, these parents would be expected to be more s e n s i t i v e l y responsive to t h e i r children i n such s i t u a t i o n s . However, i f a high-risk group of parents or a maltreating group of parents had been systematically selected t h i s might not be the case. 98 I t i s not possible, however, to confidently i n t e r p r e t the r e l a t i o n of the r e s u l t s from the two inventories. The secondary inventory i s perhaps most useful i n i t s capacity to h i g h l i g h t the s u p e r i o r i t y of a measure which depicts multiple s i t u a t i o n s over one depicting no s i t u a t i o n s whatsoever. The demonstration that there i s i n t e r n a l consistency within response categories i s of course not strong evidence of v a l i d i t y . On the other hand, without i n t e r n a l consistency, v a l i d i t y i s seriously threatened. The fact that at l e a s t 22% of subjects selected each response category one or more times combined with the fact that i n t e r n a l consistency i s demonstrated within each response category provides good empirical support f o r the relevance and the v a l i d i t y of the response categories. Further exploration of the i d e a l number and forms of response categories for the dimension of "permitting/blocking access" i s suggested however. The "ignoring blocking access" category appears to be mutually dependent with both the "midway on access" and the "angry blocking access" categories. The "angry blocking access" and the "midway on access" categories appear to be e i t h e r independent or negatively dependent. Conclusions must be tentative but i t appears that the "ignoring blocking access" response category i s between the other two on a "permitting/blocking access" continuum. 99 On the basis that the "ignoring" and "angry" categories are both "blocking access" categories, i t was thought l i k e l y that an association would be found between the two. However, the very d i f f e r e n t kinds of association found between the "ignoring" and "midway on access" categories on the one hand, and the "ignoring" and "angry blocking access" categories on the other, was not foreseen. I t i s a d d i t i o n a l l y noteworthy that subjects who selected the "ignoring blocking access" category almost always selected the side p r o f i l e depiction showing part of the face over the depiction of the caregivers' backs. This was the case despite the fact that there was at l e a s t an equal opportunity to s e l e c t the l a t t e r version of the "ignoring" response category. This has implications for the design of future studies. 100 Table 11 Sit u a t i o n by caregiver by response Response category (1) (2) (3) (4 Si t u a t i o n Caregiver Distressed c h i l d mother 20 1 1 1 reaching father 20 3 0 0 teacher 16 7 0 0 t o t a l 56 11 1 1 Cut knee mother 22 1 0 0 father 19 3 0 1 teacher 22 1 0 0 t o t a l 63 5 0 1 Sick mother 18 4 1 0 father 18 4 0 1 teacher 20 1 0 0 t o t a l 55 10 2 1 Bully mother 22 1 0 0 father 21 2 0 0 teacher 21 2 0 0 t o t a l 64 5 0 0 Puzzle-Schoolwork mother 20 1 2 0 father 20 2 0 1 teacher 20 0 2 1 t o t a l 59 3 4 2 Table 11 (continued) Response Category (1) (2) (3) (4 S i t u a t i o n Caregiver S p i l t milk-Late mother 20 1 1 1 father 19 1 1 2 teacher 21 2 0 0 t o t a l 60 4 2 3 Smiling c h i l d mother 23 0 0 0 reaching father 22 1 0 0 teacher 20 3 0 0 t o t a l 65 4 0 0 Se l f mother 20 1 2 0 father 22 0 1 0 teacher 18 5 0 0 t o t a l 60 6 3 0 102 The s i g n i f i c a n c e of the variable " s i t u a t i o n " and the i n s i g n i f i c a n c e of i t s i n t e r a c t i o n with, or the main e f f e c t of, the variable "caregiver" i s c e r t a i n l y i n t e r e s t i n g . The greatest difference i n the childrens' responses appears to be between the depiction of the c h i l d reaching f o r a hug with a distressed look on h i s face and the depiction of the c h i l d reaching for a hug with a smile on h i s face (see Table 11). Category 1 received more; a l l other categories received less responses when the c h i l d was depicted smiling. Children also appear to view t h e i r caregivers as responding more s e n s i t i v e l y when they have cut t h e i r knee as opposed to when they are sick. A l l caregivers are seen to respond p r o t e c t i v e l y for the most part to the s i t u a t i o n of the c h i l d being b u l l i e d . The "distressed c h i l d reaching for a hug" s i t u a t i o n received the most "midway on access" responses (11 of 48). The " c h i l d frustrated with puzzle or schoolwork" s i t u a t i o n received the most "ignoring blocking access" responses (4 of 12). The "distressed c h i l d has s p i l t h i s milk" s i t u a t i o n received the most "angry blocking access" responses (3 of 8). Conclusion This study provides some support for the v a l i d i t y of the four response categories. In doing so, i t supports the notion of the c e n t r a l i t y of the underlying dimension of "permitting to blocking" access suggested by attachment t h e o r i s t s , Main, Kaplan and Cassidy (1985). The "midway on access" response category, to the extent that i t s v a l i d i t y i s here supported, i s an additional response category on the dimension of "permitting to blocking access". Although the underlying dimension was described by Main, Kaplan and Cassidy, t h i s response category was not. This study also supports the contention that childrens' inner images or conceptualizations of s e l f and others can be represented v i s u a l l y . Methodological rigourousness requires that conclusions presented i n the remaining section be considered tentative. The v a l i d i t y of the primary measure requires further support, i n part because of the small s i z e of the sample. In addition, t h i s study i s i n essence an in t e r n a l v a l i d i t y study. One should not use the presence of patterns i n the data to support the v a l i d i t y of the response categories -and then use the si g n i f i c a n c e l e v e l s of variables determined i n r e l a t i o n to the response categories - to make further conclusions. On the other hand, to the extent that the response categories are validated on a t h e o r e t i c a l basis and 104 by various related attachment studies, the following conclusions are supported. The three caregivers that were depicted i n t h i s study were considered to be primary attachment figures i n the case of the parents and t e n t a t i v e l y , secondary attachment figures i n the case of the main teachers. A l l of the s i t u a t i o n s that were depicted i n the study were considered to be attachment-related situations, although some more than others. The minimum c r i t e r i o n used, i f proximity seeking was not depicted, was that the c h i l d be depicted i n a state of d i s t r e s s . The only other variable, "response category", was also generated from attachment theory. Each of the 552 subject selections involved the c h i l d and one caregiver i n one s i t u a t i o n . The variable " s i t u a t i o n " i s s i g n i f i c a n t beyond the .05 l e v e l with p_ = .035 (using the more appropriate Pearson s t a t i s t i c ) for the collapsed response categories. Neither the variable "caregiver" nor i t s i n t e r a c t i o n with the variable " s i t u a t i o n " i s s i g n i f i c a n t . A f a i r l y generalized meta-structure of i n t e r n a l working models i s suggested by t h i s group p r o f i l e of 5 to 7-year-old boys. Even the o u t l i e r s responded s i m i l a r l y across caregivers. "Multiple unrelated inner working models" i n which the i n d i v i d u a l i t y of each r e l a t i o n s h i p but not the coherency between them i s recognized - i s not suggested for these subjects. In addition, the data on teachers r e l a t i v e to parents can be interpreted as evidence that main female 105 teachers a f t e r 9 or 10 months of teaching a c h i l d may be considered to be an attachment figure to that c h i l d . F i n a l l y , several studies (Bauer and Twentyman, 1985; Kadushin and Martin, 1981; Thomson, Paget, Bates, Mesch, & Putman, 1971; Zussman, 1980) have noted s i t u a t i o n a l s p e c i f i c i t y of caregiver response under general conditions. This study gives support to the notion of s i t u a t i o n a l s p e c i f i c i t y of caregiver response under conditions of c h i l d attachment behavioural system a c t i v a t i o n . Future Research Directions I f inner images can be successfully explored for t h i s group of chi l d r e n i t may be done for other groups -including groups of g i r l s , older children, abused children, peer-rejected children, etc. Comparisons can then be made amongst d i f f e r e n t groups and t h e i r p r o f i l e s . I t may eventually be possible to use t h i s kind of research d i r e c t l y to f a c i l i t a t e normal development or intervention and/or i n d i r e c t l y through the development of research theory i n c h i l d development. This p a r t i c u l a r l i n e of research may i n the future explore one or more of the following d i r e c t i o n s : (a) the development of new forced choice categories, e.g., "malicious laughter" as a caregiver response; or "ignoring blocking access" responses s p e c i f i c a l l y with and/or without f a c i a l features; (b) the development of new s i t u a t i o n s , e.g., " c h i l d asking caregiver to j o i n him i n play" or v i c e -versa; or " c h i l d has damaged a material possession with 106 caregiver present"; or " c h i l d awakes from a nightmare with parent present"; (c) the comparison of members of one group i n terms of t h e i r inner images across time; (d) the comparison of d i f f e r e n t types of groups i n terms of t h e i r inner images; (e) the comparison of inner image p r o f i l e s for one group to attachment c l a s s i f i c a t i o n s or other behavioural/cognitive measures taken at various points i n time. A c h i l d ' s cognitive capacities at d i f f e r e n t ages are an important developmental consideration i n the current and i n future study designs. The development of cognitive capacities over time could be one factor that s e r i o u s l y hampers attempts to develop i n the t h i r d d i r e c t i o n mentioned. On the other hand, i f developmental change can be i s o l a t e d from actual differences i n the content of the inner images of the i n d i v i d u a l s ' i n t e r n a l working models there may be no problems; however, there have been few assessments directed primarily towards exploring developmental change i n t h i s general area (Crittenden, 1989) . Research and development i n the f i r s t two d i r e c t i o n s above can ultimately contribute to the o v e r a l l v a l i d i t y of the inventory i n i t s f i n a l form. The f i n a l two d i r e c t i o n s mentioned involve an e s s e n t i a l extension of the study i n terms of i t s g e n e r a l i z a b i l i t y and external v a l i d i t y . References Aber, J.L. & Al l e n , J.P. (1987). E f f e c t s of maltreatment on young children's socioemotional development: An attachment theory perspective. Developmental psychology. 23(3), 406-414. Aber, J.L., A l l e n , J.P., Carlson, V., & C i c c h e t t i , D. (1989). The e f f e c t s of maltreatment on development during early childhood: recent studies and t h e i r t h e o r e t i c a l , c l i n i c a l , and p o l i c y implications. In V. Carlson and D. C i c c h e t t i (Eds.), Child maltreatment (pp. 579-616). New York: Cambridge University Press. Ainsworth, M.D.S. (1982). Attachment: Retrospect and prospect. In J . Stevenson-Hinde and C. Murray Parkes (Eds.), The place of attachment i n human infancy (pp. 3-30). New York: Basic Books. Ainsworth, M.D.S., Blehar, M.C., Waters, E., & Wall, S. (1978). Patterns of attachment: A psychological study of the strange s i t u a t i o n . H i l l s d a l e , NJ: Erlbaum. Ainsworth, M.D.S. & Wittig, B.A. (1969). Attachment and the exploratory behavior of one-year-olds i n a Strange Situ a t i o n . In B.M. Foss (Ed.), Determinants of infant  behavior v o l . 4 (pp.113-116). London: Methuen. Axline, V. (1947). Play therapy. New York: Ballantine. Azar, S.T., Robinson, D.R., Hekimian, E. & Twentyman, C T . (1984). U n r e a l i s t i c expectations and problem-solving 108 a b i l i t y i n maltreating and comparison mothers. Journal of consulting and c l i n i c a l psychology. 52(4), 687-691. Azar, S.T. & Rohrbeck, CA. (1986) . Child abuse and u n r e a l i s t i c expectations: Further v a l i d a t i o n of the Parent Opinion Questionnaire. Journal of consulting and  C l i n i c a l Psychology. 54(6), 867-868. Bauer, W.D. & Twentyman, C T . (1985). Abusing, n e g l e c t f u l , and comparison mothers' responses to c h i l d - r e l a t e d and non-child-related stressors. Journal of consulting and  c l i n i c a l psychology. 53(3), 335-343. Belsky, J . & Nezworski, T. (1988). C l i n i c a l implications of attachment. In J. Belsky & T. Nezworski (Eds.), C l i n i c a l  implications of attachment (pp. 3-17). H i l l s d a l e , NJ: Lawrence Erlbaum Associates, Inc. Belsky, J . , Rovine, M., & Taylor, D. (1984). The or i g i n s of i n d i v i d u a l differences i n infant-mother attachment: Maternal and infant contributions. Child development. 55, 718-728. Berger, A.M., Knutson, J.F., Mehm, J.G., & Perkins, K.A. (1988). The se l f - r e p o r t of punitive childhood experiences of young adults and adolescents. Child abuse and neglect. 12, 251-262. Blehar, M.C, Ainsworth, M.D.S., & Main, M. (unpub.). Mother-infant i n t e r a c t i o n relevant to close bodily contact: A longitudinal study. Blehar, M.C, Lieberman, A.F., & Ainsworth, M.D.S. (1977). Early face-to-face i n t e r a c t i o n and i t s r e l a t i o n to l a t e r 109 infant-mother attachment. Child development, 48, 182-194. Bowlby, J . (1969). Attachment and loss, vol.1: Attachment. New York: Basic Books. Bowlby, J . (1973). Attachment and loss, vol.2: Separation. New York: Basic Books. Bowlby, J . (1980). Attachment and loss, vol.3: Loss. sadness, and depression. New York: Basic Books. Bowlby, J . (1988). A secure base. New York: Basic Books. Bretherton, I. (1984). Symbolic play: The development of s o c i a l understanding;. New York: Academic Press. Bretherton, I. (1985). Attachment theory: Retrospect and prospect. Monographs of the Society for Research i n Child Development. 50(1-2, S e r i a l No. 209). Bretherton, I., Ridgeway, D., & Cassidy, J . (1990). The r o l e of i n t e r n a l working models i n the attachment re l a t i o n s h i p : Can i t be studied i n three-year-olds?. In M. Greenberg, D. C i c c h e t t i , and E. Cummings (Eds.), Attachment i n the preschool years: theory, research and intervention (pp. 273-308). Chicago: University of Chicago Press. Burgess, R.L. & Conger, R.D. (1978). Family i n t e r a c t i o n i n abusive, n e g l e c t f u l , and normal families, C h i l d  development. 49, 1163-1173. Carlson, V., C i c c h e t t i , D., Barnett, D., & Braunwald, K.G. (1989). Finding order i n disorganization: Lessons from research on maltreated infants' attachments to t h e i r caregivers. In V. Carlson & D. C i c c h e t t i (Eds.), Child 110 maltreatment (pp. 525-531). New York: Cambridge University Press. Caspi, A. & Elder, G.H., J r . (1988). Emergent family patterns: The intergenerational construction of problem behaviour and rela t i o n s h i p s . In Hinde and Stevenson-Hinde (Eds.), The relationships within f a m i l i e s : Mutual  influences (pp. 218-240). Oxford: Clarendon Press. Cassidy, J . (1988). Child-mother attachment and the s e l f i n six-year-olds. Child development. 59, 121-134. Cassidy, J . , & Kobak, R.R. (1988). Avoidance and i t s r e l a t i o n to other defensive processes. In J . Belsky and T. Nezworski (Eds.), C l i n i c a l implications of attachment (pp. 300-323). H i l l s d a l e , NJ: Lawrence Erlbau Associates, Inc. Christensen, R. (1990). Log-linear models. New York: Springer-Verlac Inc. Cohn, D.A. (1990). Child-mother attachment of six-year-olds and s o c i a l competence at school. Child development. 61, 151-162. Craik, K. (1943). The nature of explanation. Cambridge: Cambridge University Press. Crittenden, P. (1983). The e f f e c t of mandatory protective daycare on mutual attachment i n maltreating mother-infant dyads. Child abuse and neglect, 7, 297-300. Crittenden, P. (1985). Social networks, q u a l i t y of c h i l d rearing, and c h i l d development. Child development. 56, 1299-1313. I l l Crittenden, P. (1988a). Family and dyadic patterns of functioning i n maltreating families. In K. Browne, C. Davies, and P. Stratton (Eds.), Early p r e d i c t i o n and  prevention of c h i l d abuse (pp. 161-189). New York: Wiley. Crittenden, P. (1988b). Relationships at r i s k . In J . Belsky and T. Nezworski (Eds.), C l i n i c a l implications of  attachment (pp. 136-174). H i l l s d a l e , NJ: Lawrence Erlbaum Associates, Inc. Crittenden, P. (1988c). Distorted patterns of r e l a t i o n s h i p i n maltreating f a m i l i e s : The r o l e of i n t e r n a l representation models. Journal of reproductive and infant  psychology. 6, 183-199. Crittenden, P. (1989). Internal representational models of attachment rela t i o n s h i p s . Paper presented at the World Association of Infant Psychiatry and A l l i e d D i s c i p l i n e s , Lugano, Switzerland. Crittenden, P. & Ainsworth, M.D. (1989). C h i l d maltreatment and attachment theory. In V. Carlson and D. C i c c h e t t i (Eds.), Child maltreatment (pp. 432-463). New York: Cambridge University Press. Crittenden, P. & D i L a l l a , D.L. (1988). Compulsive compliance: The development of an i n h i b i t o r y coping strategy i n infancy. Journal of abnormal c h i l d  psychology. 16(5), 585-599. Dodge, K., & Richard, B. (1985). Peer perceptions, aggression, and the development of peer r e l a t i o n s . In J . Pryor and J . Day (Eds.), The development of s o c i a l 112 cognition (pp. 35-58). New York: Springer-Verlag. Dubin, R. & Dubin, E.R. (1964). Children's s o c i a l perceptions: A review of research. C h i l d development, Egeland, B. & Farber, E.A. (1984). Infant-mother attachment: Factors related to i t s development and changes over time. Ch i l d development. 55. 753-771. Epstein, S. (1973). The self-concept r e v i s i t e d or a theory of a theory. American psychologist. 28, 404-416. Epstein, S. (1979). Natural healing processes of the mind: 1. Acute schizophrenic disorganization. Schizophrenia b u l l e t i n . 5, 315-321. Epstein, S. (1980). The self-concept: A review and the proposal of an integrated theory of personality. In E. Staub (Ed.), Self-esteem and psychotherapy (pp. 82-131). New York: Wiley. E v e r i t t , B.S. (1977). The analysis of contingency tables. New York: Halsted Press. Fienberg, S.E. (1980). The analysis of c r o s s - c l a s s i f i e d categorical data. Cambridge, MA: M.I.T. Press. Fivush, R., & Slackman, E. (1986). The a c q u i s i t i o n and development of s c r i p t s . In K. Nelson (Ed.), Event knowledge: Structure and function i n development (pp. 71-96). H i l l s d a l e , NJ: Erlbaum. French, D. C. (1988). Heterogeneity of peer-rejected boys: Aggressive and nonaggressive subtypes. C h i l d development. 59, 976-985. Gaensbauer, T.J. & Harmon, R.J. (1982). Attachment behavior 113 i n abused/neglected and premature infants. In Emde and R. Harmon (Eds.). Development of attachment and a f f i l i a t i v e  Systems (pp. 263-279). New York: Plenum Press. George, C. & Main, M. (1979). Social interactions of young abused children: Approach, avoidance, and aggression. Ch i l d development. 50. 306-318. Grossmann, K.E., Grossmann, K., Spangler, G., Suess, G., & Unzer, L. (1985). Maternal s e n s i t i v i t y i n northern Germany. Monographs of the Society for Research i n Child  Development, 50(1-2, S e r i a l No. 209). Hart, CH., Ladd, G.W., & Burleson, B.R. (1990). Children's expectations of the outcomes of s o c i a l s t r a tegies: Relations with sociometric status and maternal d i s c i p l i n a r y s t y l e s . Child development. 61, 127-137. Helle r , K.A., & Berndt, T.J. (1981). Developmental changes i n the formation and organization of personality a t t r i b u t i o n s . Child development, 52. 683-691. Hinde, R.A. (1982). Attachment: Some conceptual and b i o l o g i c a l issues. In C. Parkes and J . Stevenson-Hinde (Eds.), The place of attachment i n human behavior (pp. 60-76). New York: Basic Books. Izard, C E . (1978). Emotions as motivations: An evolutionary-developmental perspective. In R.A. Dienstbier (Ed.), Nebraska Symposium on Motivation. 163-200. Lincoln, NE: University of Nebraska Press. Kadushin, A. & Martin, J . (1981). Child Abuse: An Interactional Event. New York: Columbia University Press. 114 Kaplan, N., & Main, M. (1985). Internal representations of attachment at six-years as indicated by family drawings and verbal responses to imagined separations. In M. Main (Chair), Attachment: A move to the l e v e l of  representation. Symposium at Society f o r Research i n  Ch i l d Development, Toronto. Krazier, S.K., Fryer, G.E., & M i l l e r , M. (1988). Programming for preventing sexual abuse and abduction: What does i t mean when i t works? Child Welfare League of America. 67(1), 69-78. Lamb, M.E., Thompson, R.A., Gardner, W.P., Charnov, E.L., & Estes, D. (1984). Security of i n f a n t i l e attachment as assessed i n the "strange s i t u a t i o n " : I t s study and b i o l o g i c a l interpretation. The behavioral and brain  sciences. 7, 127-171. Lewis, M., F e i r i n g , C., McGuffog, C., & J a s k i r , J . (1984). Predicting psychopathology i n six-year-olds from early s o c i a l r e l a t i o n s . Child development, 55, 123-136. Lieberman, A.F. & Pawl, J.H. (1988). C l i n i c a l applications of attachment theory. In J . Belsky and T. Nezworski (Eds.), C l i n i c a l implications of attachment (pp. 327 -351). H i l l s d a l e , NJ: Lawrence Erlbaum Associates, Inc. Main, M. & Cassidy, J . (1988). Categories of response to reunion with the parent at age 6: Predictable from infant attachment c l a s s i f i c a t i o n s and stable over a 1-month period. Developmental psychology. 24 (3). 415-426. Main, M., Kaplan, N., & Cassidy, J . (1985). Security i n 115 infancy, childhood, and adulthood: A move to the l e v e l of representation. Monographs of the Society for Research i n Chil d Development, 50(1-2, S e r i a l No. 209). Main, M. & Stadtman, J . (1981). Infant response to r e j e c t i o n of physical contact by the mother. American academy of  c h i l d psychiatry. 292-307. Mandler, J.H. (1979). Categorical and schematic organization i n memory. In C.R. Puff (Ed.), Memorv organization and  structure (pp. 259-299). New York: Academic Press. Marvin, R.S. (1977). An ethological-cognitive model f o r the attenuation of mother-child attachment behavior. In T.M. Alloway, L. Krames, and P. P l i n e r (Eds.), Advances i n the  study of communication and a f f e c t . Vol. 3. Attachment  behavior (pp. 25-60). New York: Plenum Press. Maslin, C., & Bates, E. (1983). Precursors of anxious and  secure attachment: A multivariate model at age s i x  months. Symposium conducted at the bie n n i a l meeting of the Society for Research i n Child Development, Detroit. Matas, L., Arend, R.A., & Sroufe, L.A. (1978). Continuity of adaptation i n the second year: The re l a t i o n s h i p between qu a l i t y of attachment and l a t e r competence. Ch i l d  development. 49, 547-556. Older, J . (1981). A restoring touch for abusing f a m i l i e s . C h i l d abuse and neglect. 5, 487-489. Park, K.A. & Waters, E. (1989). Security of attachment and preschool friendships. Child development. 60, 1076-1081. Parke, R., & Slaby, R. (1983). The development of 116 aggression. In M. E. Hetherington (Ed.), Handbook of  c h i l d psychology: S o c i a l i z a t i o n , personality, and s o c i a l  development (pp. 547-641). 4, New York: Wiley. Pastor, D.L. (1981). The q u a l i t y of mother-infant attachment and i t s r e l a t i o n s h i p to toddlers' i n i t i a l s o c i a b i l i t y with peers. Developmental psychology. 17(3). 326-335. Patterson, G.R., DeBaryshe, B.D., & Ramsey, E. (1989). A developmental perspective on a n t i s o c i a l behavior. American psychologist. 44, 329-335. Pelton, L. (1977). Child abuse and neglect and protective  intervention i n Mercer County. New Jersey. Bureau of Research, New Jersey D i v i s i o n of Youth and Family Services. Piaget, J . (1924). The language and thought of the c h i l d . New York: Harcourt, Brace. Piaget, J . (1952). The o r i g i n of i n t e l l i g e n c e i n children. New York: International U n i v e r s i t i e s Press. Piaget, J . (1954). The construction of r e a l i t y i n the c h i l d . New York: Basic Books. Piaget, J . & Inhelder, B. (1948). The c h i l d ' s conception of space. New York: Humanities Press. Reider, C. & C i c c h e t t i , D. (1989). Organizational perspective on cognitive control functioning and c o g n i t i v e - a f f e c t i v e balance i n maltreated children. Developmental psychology r 25(3). 382-393. Rholes, W.S. & Ruble, D.N. (1984). Children's understanding of d i s p o s i t i o n a l c h a r a c t e r i s t i c s of others. C h i l d 117 development, 55, 550-560. Ricks, M.H. (1985). The s o c i a l transmission of parental behavior: Attachment across generations. Monographs of the Society for Research i n Child Development. 50(1-2, S e r i a l No. 209). Schank, R.C. (1982). Dynamic memory: A theory of reminding and learning i n computers and people. Cambridge: Cambridge University Press. Schneider-Rosen, K., Braunwalk, K.G., Carlson, V., & C i c c h e t t i , D. (1985). Current perspectives i n attachment theory: I l l u s t r a t i o n from the study of maltreated infants. Monographs of the Society For Research i n Ch i l d Development. 50(1-2, S e r i a l No. 209). Shaver, P., Hazan, C., & Bradshaw, D. (1988). Love as attachment: The integration of three behavioral systems. In M. Barnes (Ed.), The psychology of love (pp. 68-99). New Haven: Yale University Press. Simons, R.L., Whitbeck, L.B., Conger, R.D., & Chyi-In, W. (1991). Intergenerational transmission of harsh parenting. Developmental psychology. 27(1). 159-171. Slade, A. (1987). Quality of attachment and early symbolic play. Developmental psychology. 23(1). 78-85. Spieker, S.J. & Booth, C.L. (1988). Maternal antecedents of attachment qua l i t y . In J . Belsky and T. Nezworski (Eds.), C l i n i c a l implications of attachment (pp. 95-135). H i l l s d a l e , NJ: Lawrence Erlbaum Associates, Inc. Sroufe, L.A. (1979). Socioemotional development i n infancy. 118 In J . Osofsky (Ed.)/ Handbook of infant development (pp. 462-515). Sroufe, A. (1983). Infant-caregiver attachment and patterns of adaptation i n preschool: The roots of maladaptation and competence. In M. Perlmutter (Ed.), Minnesota  Symposia i n Ch i l d Psychology. Vol. 16 (pp. 41-81). H i l l s d a l e , NJ: Lawrence Erlbaum Associates, Inc. Sroufe, A. (1986). Appraisal: Bowlby's contribution to psychoanalytic theory and developmental psychology; Attachment: Separation: Loss. Journal of c h i l d psychology  and psychiatry. 27(6). 841-849. Sroufe, A. (1988). The role of infant-caregiver attachment i n development. In J . Belsky and T. Nezworski (Eds.), C l i n i c a l implications of attachment (pp. 18-38). H i l l s d a l e , NJ: Lawrence Erlbaum Associates Inc. Sroufe, L.A., & Fleeson, J . (1986). Attachment and the construction of relat i o n s h i p s . In W. Hartup and Z. Rubin (Eds.), Relationships and development (pp. 51-71). New York: Cambridge University Press. Straus, M. (1979). Family patterns and c h i l d abuse i n a na t i o n a l l y representative American sample. Ch i l d abuse  and neglect, 2, 213-225. Summit, R.C. (1984). The impact of c h i l d abuse: Psycho-s o c i a l and educational implications. In S. Leung (Ed.), Mental health and the school (pp.19-40). Thompson, R.A., Lamb, M.E., & Estes, D. (1982). S t a b i l i t y of infant-mother attachment and i t s r e l a t i o n s h i p to changing 119 l i f e circumstances i n an unselected middle-class sample. Chi l d development. 53, 144-148. Thomson, E.M., Paget, N.W., Bates, D.W., Mesch, M., & Putman, T.I. (1971). Child abuse; A community challenge. New York: Henry Stewart and Childrens Aid Society. Tulving, E. (1972). Episodic and semantic memory. In E. Tulving and W. Donaldson (Eds.), Organization of memory (pp. 382-403). New York: Academic Press. Tulving, E. (1979). Memory research: What kind of progress? In L.G. Nilsson (Ed.), Perspectives on memory research: Essavs i n honour of Uppsala University's 500th anniversary (pp. 19-34). H i l l s d a l e , NJ: Lawrence Earlbaum Associates. Tulving, E. (1985). How many memory systems are there? American psychologist. 4f)(4) , 385-398. Tulving, E. (1989). Remembering and knowing the past. American s c i e n t i s t . July-August. 361-367. Vaughn, B., Egeland, B., & Sroufe, A. (1979). Individual differences i n infant-mother attachment at twelve and eighteen months: S t a b i l i t y and change i n families under stress. C h i l d development. 50, 971-975. Waas, G. (1988). Social a t t r i b u t i o n a l biases of peer-rejected and aggressive children. Child development. 59, 969-975. Zussman, J.U. (1980). S i t u a t i o n a l determinants of parental behavior: E f f e c t s of competing cognitive a c t i v i t y . C h i l d  development. 51 792-800. Appendix A: Drawings f o r P e r m i t t i n g / B l o c k i n g Access Inventory 121 122 123 125 126 127 1S8 129 130 131 132 133 134 135 136 138 139 140 141 142 143 144 145 147 148 149 150 151 152 153 154 155 156 157 1 5 8 159 160 161 162 1 6 3 164 165 168 169 170 171 172 173 174 175 176 177 178 179 180 182 185 186 1 8 8 183 / 191 192 194 195 1 9 6 197 1 9 8 199 200 201 202 203 204 205 206 207 208 209 2 1 0 211 212 213 214 215 Appendix B: Drawings f o r F r i e n d l l n e s s / U n f r i e n d l i n e s s Inventory 217 218 219 220 221 222 223 224 225 2 2 6 227 228 229 231 232 1 233 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0054637/manifest

Comment

Related Items