UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The relationship between interpersonal problems and negative childhood experiences During, Sara May 1992

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

Item Metadata

Download

Media
831-ubc_1992_fall_during_sara_may.pdf [ 9.88MB ]
Metadata
JSON: 831-1.0086554.json
JSON-LD: 831-1.0086554-ld.json
RDF/XML (Pretty): 831-1.0086554-rdf.xml
RDF/JSON: 831-1.0086554-rdf.json
Turtle: 831-1.0086554-turtle.txt
N-Triples: 831-1.0086554-rdf-ntriples.txt
Original Record: 831-1.0086554-source.json
Full Text
831-1.0086554-fulltext.txt
Citation
831-1.0086554.ris

Full Text

T H E R E L A T I O N S H I P B E T W E E N I N T E R P E R S O N A L P R O B L E M S A N D N E G A T I V E C H I L D H O O D E X P E R I E N C E S ty Sara M a y During B.Sc. (Honors), The University of Calgary, 1983 M . A . , The University of British Columbia , 1986 A THESIS SUBMITTED I N P A R T I A L F U L F I L L M E N T O F T H E R E Q U I R E M E N T S FOR T H E D E G R E E OF D O C T O R OF P H I L O S O P H Y i n T H E F A C U L T Y O F G R A D U A T E STUDIES (Psychology) We accept this thesis as conforming to the required standard The University of British Columbia O c t c b e r 1992 ©Sara M a y During, 1992 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 rS'jC'HpLoQ ^ The University of British Columbia Vancouver, Canada Date Och2l ^ 1992. DE-6 (2/88) Abstract The purpose of this investigation was to examine the relationship between adult interpersonal functioning of women, currently i n therapy wherein they were addressing unresolved issues about their chi ldhood maltreatment, and chi ldhood experiences referring to a developmental psychopathology framework. One hundred and twenty women (30 sexual abuse, 30 physical abuse, 30 fami ly disruption, and 30 control) were individual ly presented w i t h a series of audiotapes of three interpersonal situations (conflictual, neutral, dating), and asked to record their self-report of physiological response, self- and other-perceptions and coping responses. Physiological indices (heart rate, systolic and diastolic blood pressure) were also recorded, as wel l as self-report of chi ldhood coping strategies. The data were examined as to whether abuse survivors i n therapy addressing their childhood experiences have more interpersonal problems than non-clinical control subjects or than individuals having other types of traumatic backgrounds and whether any noted interpersonal difficulties are specific to the type of abuse, or general in nature. Results indicated that abusive experiences are related to greater dysfunction i n regards to self- and other-perceptions than exposure to disruptive/chaotic family environments alone. However, the differences appear to be variable-specific and few differences were obtained on physiological and coping variables. Some specific problems i n adult interpersonal functioning were evidenced between the sexual and physical abuse groups. Specifically, in comparison to the physical abuse survivors, sexual abuse survivors reported greater use of different, and perhaps less adaptive, coping strategies in neutral situations. N o other dependent variables significantly differed between the two groups. The results were discussed in terms of theoretical and empirical issues related to abuse-specific outcomes, the specific methodology-employed in this study, and directions for future research. Table of Contents P A G E Abstract i i Table of Contents i v List of Tables v i i i List of Appendices x Acknowledgements xi Historical Overview of Research Trends in Chi ldhood Maltreatment 4 Early Research and Methodological Issues 4 The Investigation of Maltreated Children's Emotional and Psychological Functioning 10 Theoretical Models of C h i l d Maltreatment 16 Attachment Theory: Parent-Infant Interaction 18 The Development of Autonomy 20 Emotional and Cognitive Development 22 Peer Relations 22 Summary of Attachment Theory 24 Developmental Psychopathology 27 Physical Maltreatment 29 Family Chaos and Disruption 32 Sexual Maltreatment 34 Developmental Psychopathology and the Life-Span Perspective 38 Theoretical Relationship Between Chi ldhood Maltreatment, Developmental Psychopathology and Adul t Functioning 40 P A G E Parenting Behavior of A d u l t Survivors of C h i l d Maltreatment 42 The Impact of C h i l d Maltreatment on General Adul t Functioning 49 Chi ldhood physical maltreatment and adult functioning 51 Chi ldhood sexual maltreatment and adult functioning 57 Comparison of Physical and Sexual Maltreatment 63 Pilot Study 65 Purpose of the Present Study 68 Perceptual and Attributional Processes 68 Interpersonal Coping Behaviors 71 Physiological Processes 72 Laboratory Simulation 76 Sample Selection 77 Questions Addressed 80 Flypotheses 83 A d d i t i o n a l Questions 86 M e t h o d 90 Subjects 90 Sexual abuse 90 Physical abuse 90 Family disruption 91 Controls 92 Recruitment and Screening 94 Audiotape Stimulus Materials 100 Audiotape Validation 101 P A G E Dependent Measures 102 Structured self-report 102 Physiological recording 105 A d d i t i o n a l Measures 107 Procedure 111 Results 113 Audiotape Validation Summary 113 Central Analyses: Overview 115 Vic t imizat ion Experience 121 Audiotape One: Neutral Content 122 Self- and other-perception data 122 Open-ended coping data 122 Structured coping data 122 Physiological data 127 Audiotape Two: Confl ictual/Argumentat ive Content 127 Self- and other-perception data 127 Open-ended coping data 127 Structured coping data 127 Physiological data 127 Audiotape Three: Intimate/Dating Content 137 Self- and other-perception data 137 Open-ended coping data 137 Structured coping data 137 Physiological data 137 A d d i t i o n a l Measures: Coping in Chi ldhood 137 Discussion 144 Physiological Measures 149 Structured Coping Measures Social Perception Measures Open-ended Coping Measures Recall of Coping in Chi ldhood A Comparison of Pilot and Central Study Results General Methodological Issues Summary Conclusions and Directions for Future Research References Footnotes Appendices P A G E 154 155 157 160 164 167 172 178 182 209 214 List of Tables P A G E Table 1 Summary of Theoretical Predictions 87 Table 2 Demographic Variable Scores of Experimental and Control Groups 93 Table 3 Mean Therapist-rated Adjustment Scores: Five Categories Reflecting Current Adjustment 95 Table 4 Summary of the Independent and Dependent Variables 110 Table 5 Interrater Agreement Rates: Open-ended Coping Responses 119 Table 6 Mean Self-Perception Scores: Neutral Tape 123 Table 7 Mean Other-Perception Scores: Neutral Tape 124 Table 8 Open-ended Coping Response Frequencies: Neutral Tape 125 Table 9 Mean Coping Scores: Structured Self-report of Use of Coping Strategies in Response to Neutral Tape 126 Table 10 Mean Summary Scores of Self-reported Physiological Response ' 128 Table 11 Mean Physiological Residual Gain Scores 129 Table 12 Raw Mean Physiological Response Scores: Heart Rate 130 Table 13 Raw Mean Physiological Response Scores: Diastolic Blood Pressure 131 Table 14 Raw Mean Physiological Response Scores: Systolic Blood Pressure 132 Table 15 Mean Self-Perception Scores: Conflictual Tape 133 Table 16 Mean Other-Perception Scores: Conflictual Tape 134 P A G E Table 17 Open-ended Coping Response Frequencies: Conflictual Tape 135 Table 18 Mean Coping Scores: Structured Self-report of Use of C o p i n g Strategies in Response to Conflictual Tape 136 Table 19 Mean Self-Perception Scores: Dating Tape 138 Table 20 Mean Other-Perception Scores: Dating Tape 139 Table 21 Open-ended Coping Response Frequencies: Dating Tape 140 Table 22 Mean Coping Scores: Structured Self-report of Use of C o p i n g Strategies in Response to Dating Tape 141 Table 23 Mean W C C L - R Scores: Coping in Chi ldhood 143 Table 24 Summary of Central Results - Investigation 1: 146 Table 25 Summary of Central Results - Investigation 2: 147 Table 26 Summary of Central Results - Investigation 3: 148 List of Appendices P A G E Appendix A Pilot Study 215 Appendix B Therapist Screening Questionnaire ??3 Appendix C Advertisement for Subjects 230 Appendix D Transcripts of Audiotapes 231 Appendix E Social Perception 234 Appendix F Self-Perception 235 Appendix G Coping Strategies 236 Appendix H Structured Coping Response 237 Appendix I Self-report Measure of Physiological Arousal 238 Appendix J Ways of Coping (Revised) 240 Appendix K W C C L - R - Supplementary Items 244 Appendix L Therapist Rating of Client Adjustment 246 Appendix M Consent Form 247 Acknowledgements I w o u l d like to thank the members of my committee, Wolfgang Linden, John Yui l le , and Dan Perlman, and extend special thanks to m y supervisor, L y n n A l d e n , for her enthusiasm, encouragement, guidance, and understanding. I w o u l d also l ike to express my appreciation to agency personnel, patients, and staff at Fraser V a l l e y Menta l Health Services, Chesterfield House, C h i l l i w a c k Social Services, C h i l l i w a c k Transition House, Alberta Children's Hospital , Alberta A d u l t Chi ldren of Alcoholics , and to all others w h o participated. Without their willingness to help and share their experiences, this project w o u l d never have taken place. A l s o , I w o u l d like to thank m y friends and family, particularly m y husband Sam, for their years of help, patience, and faith. T H E R E L A T I O N S H I P B E T W E E N I N T E R P E R S O N A L P R O B L E M S A N D N E G A T I V E C H I L D H O O D EXPERIENCES The abuse of children has a lengthy and well-documented history. Programs to protect victims and scientific investigations of the complex factors involved in abuse are more recent. Early movements to protect chi ldren f rom abuse and exploitation took shape in the 1800s, and addressed such issues as chi ld labor, pr imary education, and health care (Garbarino, Guttmann, & Seeley, 1986). The early reformers were concerned with emotional and spiritual well-being as wel l , and sought to remove children from environments and caregivers that jeopardized the development of self-esteem and the capacity for healthy interpersonal relationships. A s c h i l d abuse became more prominent, there has been the recognition that immediate and/or lasting damage can result, even in the absence of visible physical harm. Awareness of incest, battery, and other forms of abuse has been steadily g r o w i n g since the early 1970s. Increasing numbers of individuals seeking mental health services are identifying themselves as victims of abuse (Bergart, 1986). Current research evidence generally confirms clinical impressions that physical , emotional, and sexual abuse in chi ldhood pose serious risks to mental health into adulthood (e.g., Briere & Runtz, 1988, 1990; Conte, 1985; Egeland, Jacobvitz, & Sroufe, 1988; Finkelhor, 1983, 1990; Finkelhor & Browne, 1985; Gispert, Davis , M a r s h , & Wheeler , 1987; Lamphear, 1986; Steele, 1986; Tsai & Wagner, 1978). Diverse theoretical frameworks (e.g., psychodynamic, attachment theory, learning theory) posit continuity between childhood experiences and adult relationships such that consistent emotional support from caretakers may be a necessary condition for healthy psychological development and for the capacity to form adult relationships (e.g., Alexander, 1992; Bowlby, 1988; Cicchetti , 1989; Flaherty & Richman, 1986; Herzberger & Tennen, 1985). This hypothesis gains support from the observation that adults w h o have experienced childhood abuse, and who have been raised in environments devoid of emotional support and respect, frequently present w i t h a wide range of psychiatric and social disturbances. The aim of the present study was to examine whether adults who experienced different types of childhood/adolescent stressors (i.e., physical abuse, sexual abuse, family chaos) evidence different adaptations (styles of coping wi th these different stressors), and whether these coping styles are reflected in adult adjustment. This issue w i l l be addressed in the literature review that fol lows, beginning w i t h an overview of research trends and related methodological concerns. Theoretical and empirical l i terature i n the areas of early attachment experiences and related childhood development w i l l be summarized in order to highlight dimensions of interpersonal funct ioning hypothesized to reflect continuity between chi ldhood experiences and adult relationships. There is a lack of abundant longitudinal studies that examine the continuity between ch i ldhood experiences and adult functioning. Therefore, an attempt was made to reconstruct information from studies focusing on different age groups of survivors of chi ld maltreatment to examine evidence of continuity. The attachment/ chi ld maltreatment literature also is reviewed to emphasize the process by w h i c h such continuity might occur. The rev iew is organized i n a manner that reflects the major issue under investigation: that is, what is known about differences in the constellation of factors around physically and sexually abusive or disruptive (e.g., chaos, spousal violence, alcoholism) family backgrounds that w o u l d lead one to predict differences in adult outcomes. Historical Overview of Research Trends in Chi ldhood Maltreatment Early Research and Methodological Issues Early research i n the area of chi ld maltreatment was pr imar i ly based on clinical observations and case studies, focusing on the detection, care, and protection of maltreated children (George & M a i n , 1979). While these studies provided some informat ion about the phenomenon, they lacked mthodolog ica l rigor and generalizability. Prior to, and through the early 1970s, a popular research focus in the attempt to explain chi ld maltreatment was on parental characteristics, with reference to the psychiatric model ; that is, the belief that chi ld outcomes are the direct result of parental personality characteristics (Parke & Collmer, 1975). Maltreating parents (particularly maltreating mothers) were typically depicted as impulsive, hostile, and easily angered (e.g.. Brown & Daniels,1968; G i l l , 1971; Spinetta & Rigler, 1972), unhappy, lonely, and neurotic (e.g.. Smith «& Hanson, 1975; Wasserman, 1967), and having a poor self-image and l o w self-esteem (e.g., Blumberg, 1974; Kempe & Kempe, 1978). However, a psychopathological/parental personality deficit model has been criticized in that the results of studies addressing such correlates of abuse have been inconsistent and contradictory (Celles, 1973). Methodological problems such as the use of ready-at-hand populations and the ex-post facto nature of the research have limited the representativeness of the samples (Spinetta & Rigler, 1972). These and other related difficulties have hindered efforts to identify a distinctive set of personality traits discriminating maltreating from non-maltreating parents. Wolfe (1987) suggested that the lack of consensus on parental characteristics Ukely reflects knowledge that child maltreatment is an interactional event that depends to some extent on situational factors. In part as a result of the inadequacies of the psychiatric model , researchers began to focus on the delineation of the behaviors of maltreated children. A number of descriptive studies appeared, h i g h l i g h t i n g the broad range of symptomatology in ch i ld victims. For example, in his research examining the aggressive characteristics of physically abused and neglected children, Reidy (1977) found that the abused children displayed significantly greater aggressive tendencies i n comparison to the control chi ldren in fantasy, free p lay , and the school environment. Neglected children were similar to abused children only in terms of school aggression. S imilar ly , in their study of physical ly abused and neglected children, Hoffman-Plotkin and Twentyman (1984) noted that, when compared to control children, abused children demonstrated the highest levels of aggressive behavior, whereas neglected children engaged in the least number of aggressive interactions w i t h other children. Such results lend some support to the idea that chi ldren f rom various abusive environments experience di f fer ing degrees of dif f iculty w i t h regulation and display of aggression. H o w e v e r , Reidy d i d not indicate ways in which it was ensured that members of the control group were not maltreated, and also failed to control for the potential influence of residence (i.e., at home versus out of home residence). Differences i n self-destructive activity have been observed i n children from different family backgrounds. Green (1978) found a significantly greater incidence of self-destructive behavior (i.e., self-biting, cutting, b u r n i n g , ha i r -pul l ing , head-banging, and suicidal ideation) in abused children in comparison to non-abused and neglected children. The abused children also possessed poor impulse control, minimal frustration tolerance, low self-esteem, and a trend toward the use of motor activity versus verbalization as a preferred mode of expression. However, the fact that Green rel ied u p o n parental report as his outcome measure l imits the conclusions that can be drawn from the data, given the potential biases existent in parental report. While the majority of these early studies indicated that chi ld maltreatment was related to delays and deficits in a number of developmental areas, they often contained numerous methodological limitations. Early findings on the incidence and sequelae of c h i l d maltreatment were compromised, as definitions lacked comparability, reliability, and taxonomic delineation, adequate control groups were frequently not used, varied measures were used to assess developmental outcomes, and results were often imprecisely reported (e.g., Augoust inos , 1987; Conaway & Hansen, 1989; Reidy, 1977). Barahal, Waterman, and M a r t i n (1981) also noted that the majority of early studies were non-specific i n nature (i.e., traits observed in abused children are often those that wou ld describe the overwhelming majority of children seen in clinical settings). There is a relative lack of research focusing on the behavioral characteristics of sexually abused children (many of the studies focusing on sexual maltreatment have used adult female survivors as subjects), and this is particularly evident in the earlier literature. Wolfe and Wolfe (1988) noted that current knowledge of the impact of sexual abuse on chi ld development is p r i m a r i l y derived from clinical observations and assumptions that are not thoroughly tested. These studies lack many of the controls needed to draw f i rm conclusions. For example, numerous researchers have argued that sexual abuse studies often lack representative samples, i n that certain types of cases are more l ikely to be k n o w n to professionals who supply subject pools (Finkelhor & Hotal ing, 1984). Addi t ional ly , small sample sizes are frequently used, inclusion criteria vary widely, and attrition has been poorly addressed (Mash & Wolfe , 1991; Parker & Parker, 1991). V a r y i n g definitions of sexual abuse make it difficult to draw comparisons across studies, and studies often fail to obtain psychological test data (Owens, 1984). M a n y studies restrict the definit ion of sexual abuse to acts by parents, parent substitutes, or other adult caretakers. Al though adults who commit physical a n d / o r emotional assaults on children are almost always parents and/or caretakers, non-family members commit a large percentage of sexual abuse acts (Davis & Leitenberg, 1987). It is also difficult to separate the effects of sexual abuse from other variables w h i c h are highly correlated with developmental trauma, such as low socioeconomic status and a high degree of family disorganization (Henderson, 1983). There is a great need for widely accepted operational definitions to guide research interpretation and comparison across studies. These definitions w i l l need to be comprehensive, clearly delineated, and able to differentiate the heterogeneous groups/sub-groups of maltreatment (e.g., Emery, 1989; Fantuzzo & Twentyman, 1986). It may be insufficient merely to attempt to define abusive acts as discrete variables; rather it has been suggested that researchers might more fruit ful ly investigate the process of child-rearing and consider abuse i n relation to overall quality of care and family climate (e.g.. Mash & Wolfe , 1991; Wolfe, 1987). For example, Starr Jr. (Starr Jr., 1987) found that observers were unable to accurately detect abusive dyads in videotaped p a r e n t / c h i l d interactions. H o w e v e r , they demonstrated improvements when trained to focus on the quality of interaction, rather than on parent and/or child behaviors in isolation. The difficulties in subject definition and selection are closely related to the discrepancy in incidence figures for both physical and sexual maltreatment. A central d i f f i cul ty in the estimation of the incidence and prevalence of chi ld maltreatment is that much abuse escapes detection. Current knowledge is generally based on clinical studies of identifiable cases. For example, Hunter, Ki ls t rom, and Loda (1985) noted that many children's presenting complaints mask the presence of sexual abuse (e.g., abdominal pain, anorexia, drug-use complications, pregnancy). Certain types of cases are more likely to become known to professionals. The age of a ch i ld , the length of the abusive relationship, the socioeconomic status of the family , and the degree of allegiance of the family to the perpetrator may all influence the l ikelihood of detection and reporting. W h e n studies specifically focus on the incidence of physical abuse i n Nor th America, M a s h and Wolfe (1991) report that the second Study of National Incidence and Prevalence of C h i l d Abuse and Neglect (cited in M a s h & Wolfe, 1991) found a rate of 4.94 per 100,000 in 1986. Conversely, Straus and Celles (National Center on C h i l d Abuse and Neglect, 1986) reported that the Nat ional Family Violence Re-Survey found a ratio of 19 cases of physical abuse per 100,000 in 1985. These discrepant findings highlight the need for widely accepted operational definitions and standardized subject selection procedures. Wolfe and Wolfe (1988) stated that estimates of the incidence of child sexual abuse vary w i d e l y according to the method of estimation that is used, such as reference to actual reported cases or projection of incidence based on retrospective studies of non-cl inical populations. For example, Bagley and Ramsay (1985) reported that 21.7 percent of women i n their stratified random Canadian sample reported serious sexual abuse up to age 16. They also noted that the risk of sexual abuse increased significantly after age 8 and that sexual abuse often occurred along w i t h other aspects of family climate, such as marital discord and lack of support. On the other hand, the Committee on Sexual Offences Against Chi ldren and Youths (cited i n Wolfe & Wolfe, 1988) concluded that approximately 50% of females and 30% of males in Canada reported experiencing one or more unwanted sexual acts at some point in their lives. The related issue of outcome measure selection has been problematic i n both sexual abuse and physical abuse investigations. Researchers have often failed to use m u l t i m o d a l assessment approaches or ensure that there is a l ink between the dependent variables used and actual outcomes of c h i l d maltreatment (e.g., Fantuzzo , DePaola, Lambert, M a r t i n o , A n d e r s o n , & Sutton, 1991; Fantuzzo & T w e n t y m a n , 1986; M a s h & Wolfe , 1991; O w i n g s West & P r i n z , 1987). Few investigators have rel ied on standardized outcome measures of cognitive or psychological funct ioning (Wolfe & Wolfe , 1988) or recognized the need for developmentally sensitive and abuse specific measures (Mash & Wolfe, 1991; Wyatt & Newcomb, 1990). The Investigation of Mal t reated Chi ldren ' s E m o t i o n a l and Psychologica l Funct ioning The above-noted methodological l imitat ions lead to uncertainty as to whether and how different forms of child maltreatment are related or responsible for developmental impairment. The recognition that chi ld maltreatment does not occur in a vacuum and that other adverse family and environmental factors could influence chi ld outcome, led to increased investigation of the role of mediating variables and the impact of maltreatment on chi ldren 's emot ional and psychological functioning (Lamphear, 1985; Owings West & Pr inz , 1987). Some of the relevant research w i l l be reviewed below, with particular focus on the findings for children from physically abusive and sexually abusive environments. Smetana, Ke l ly , and Twentyman (1984) noted that prior to 1984, few studies had examined the psychological consequences and socioemotional effects of maltreatment for chi ldren. They examined the impact of maltreatment on children's conceptions of mora l and social convent ional transgressions by investigating physical ly maltreated, neglected, and control children's judgments about moral vignettes. When children were matched on IQ and socioeconomic status, few significant differences were obtained regarding the judgment of moral and social transgressions, except in cases where the event most closely matched the children's o w n maltreatment type. For example, physical ly maltreated children were most l ikely to v iew moral transgressions that caused psychological distress as universe ly w r o n g for others. The researchers suggested that children's interpretation of situations is highly related to behavior, and that interpretation may be related to the nature of maltreatment experienced. In another well-designed study. Dodge, Bates, and Pettit (1990) addressed the effect of physical abuse in early childhood on children's development of aggressive behavior. A demographically diverse sample of 4-year-old children was identified and followed over a 6-month period. Researchers assessed the fo l lowing variables: evidence of physical harm and related variables; ch i ld temperament; children's ability to attend to appropriate and relevant social cues; children's tendencies to attribute hostile intent to others; children's response accessing tendencies; children's abil i ty to generate numerous behavioral responses to social problems, and children's response evaluation tendencies. They used video recorded and cartoon stimuli to assess these aspects of children's patterns of processing social information. The results indicated that physical abuse is a risk factor for later aggressive behavior even when other ecological and biological factors are taken into account (e.g., low socioeconomic status, single parenthood, marital dissolut ion, marital violence, chi ld temperament). Dodge, Bates, and Pettit also noted that the development of aggressive behavior might be mediated by the physically abused children's tendency to acquire deviant patterns of processing social information. Barahal, Waterman, and Mart in (1981) pointed out that many of the studies attending to the psychological consequences of physical maltreatment have been limited by lack of adequate control groups and have been too general in approach (i.e., the traits observed i n maltreated children are those that w o u l d describe the majority of children seen in clinical settings). Therefore they used a group of 6- to 8-year-old children, all associated with the Department of Social Services, to examine the social-cognitive development of abused chi ldren. Results indicated that physically abused children were significantly less confident in their power to impact and shape their experiences, especially unpleasant and frustrating ones. Compared to control chi ldren, physically abused children were less adept in comprehending social role concepts and were consistently more egocentric and insensitive to socioemotional contexts. This study was well-designed, but there are st i l l some methodological problems. The manner i n which maltreament was documented was not clear and some of the physically abused children had been sexually abused. As pointed out previously, it is important to carefully examine the impact of exposure to different and multiple forms of abuse. Researchers also have examined the emotional development of physical ly abused children. For example, K inard (1980) focused on emotional development, carefully matching subjects on potentially confounding variables. K i n a r d found that the phys ica l ly abused children's reports supported the fo l lowing cl inical observations: they were significantly more sad, w i t h d r a w n , and depressed than control ch i ldren ; their sense of personal identi ty was less wel l - formed; they experienced greater difficulty with the task of socialization with the peer group; and they had di f f icul ty w i t h the establishment of trust. A factor that l imits the conclusions of this study is Kinard's failure to utilize multiple sources to confirm the presence or absence of maltreatment. In summary, literature on the psychological functioning of physically abused children suggests that physical abuse has a significant impact on children's social, cognitive, and behavioral development. Turning to sexual abuse, many studies have focused on emotional and psychological concomitants of sexual abuse for adults, although fewer studies have used children as subjects, which severely l imits our knowledge of the impact of sexual abuse for children. A brief review of some of the research on the emotional and psychological concomitants of sexual abuse on children is presented below. Conte and Schuerman (1987) have attempted to identify factors related to an increased impact of chi ld sexual abuse on children's functioning. They used a very large subject pool (369 sexually abused chi ldren and 318 control children) and matched subjects on age, socioeconomic status, and number of children i n the home. Results indicated that children s experience and perception of the sexual abuse were important, and that sexual abuse had greater impact in families that had significant problems i n l iv ing and/or greater indicators of pathology i n family members. Whi le the study was well-designed and relied on multimodal assessment and sophisticated data analyses, interpretation of the results is l imited by the fact that the investigators failed to describe ways that control subjects were screened to ensure that they had not experienced other forms of abuse. In another interesting study of sexually abused children, Wolfe, Gentile, and Wolfe (1989) investigated the presence of post-traumatic stress disorder symptoms i n a study i n c l u d i n g 71 sexually abused ch i ldren and their mothers. The investigators used multiple assessment measures, examined the role of mediating variables, and carefully described the sexually abusive experiences, which are all strengths of the study. Results indicated that sexually abused children were rated by their mothers as displaying high levels of internalizing and externalizing behaviors and high rates of post-traumatic stress disorder symptoms. In general, the children demonstrated global adjustment problems and behaviors specific to sexual abuse, such as sex-associated fears and intrusive thoughts. The investigators noted that decreases in social confidence were related to a self-depreciatory attributional style. While the study is well-designed, it may have been informative to include a control group of differentially traumatized children to help clarify the unique contribution of sexually abusive experiences to post-traumatic stress disorder symptomatology and global adjustment. Addi t ional ly , the study does not state whether the sexually abused subjects had also experienced other forms of maltreament. The behavioral characteristics of child victims of sexual abuse also have been examined by investigators such as Inderbitzen-Pisaruk, Shawchuck, and Hoyer (1992). Results of this study indicated that sexually abused children reported signif icantly greater feelings of depress ion/dysphor ia , and that their parents reported significantly more externalizing and internalizing problems and emotional distress. There are a number of methodological weaknesses in this study. Examiners were not b l ind to the subjects' group status. The investigators failed to measure and control for other disruptive circumstances such as physical abuse and level of family conflict. The subjects came from a number of developmental stages and this was not clearly addressed i n the choice of dependent variables. In summary, research on c h i l d sexual abuse has emphasized the importance of children's experience and perception of the abuse and also has described evidence of global and specific difficulties. Al though the earlier chi ld maltreatment research suffered from numerous methodological limitations, it was still significant for its contribution i n focusing greater attention on the impact of different forms of maltreatment on children (Cicchetti, Toth, & Hennessy, 1989). Recurrent patterns d i d emerge from these studies and have served to guide further research. M o r e research i n the 1980s included the fol lowing improvements: adequately matched control groups; attempts to detail selection criteria more clearly; reports of the psychometric properties of assessment procedures; and an increase i n the sophistication of data analyses. Reseachers also attempted to delineate the consequences of different patterns of maltreatment and emphasized the importance of examining mediat ing variables w h i c h might act to protect maltreated chi ldren from developmental deficits (Augoustinos, 1987; Inderbitzen-Pisaruk et al., 1992). It is essential that such methodological improvements continue if we are to further our knowledge of child maltreatment. Methodological sophistication w i l l help elucidate the noted heterogeneity and lack of un iversa l i ty i n chi ld maltreatment outcome. W h i l e the literature is generally support ive of the suggestion that child maltreatment is detrimental to psychological functioning, no consistent descriptive profile of sexual abuse victims during childhood has emerged (Wolfe & Wolfe, 1988). Differences in outcome appear to be a function, in part, of a number of mediating variables such as age when maltreated, coping abiHties, environmental influences, and attributions, as they have their o w n negative sequelae (e.g., Conaway & Hansen, 1989; Conte & Schuerman, 1987; Fantuzzo, 1990; Wol fe & Wolfe , 1988). It is necessary to consider the role of social and environmental factors and to distinguish carefully between the characteristics of different subtypes of maltreatment as d i f f e r i n g soc io-environmental and maltreatment experiences may result in different patterns of dysfunction across domains of functioning. Theoretical Models of C h i l d Maltreatment The f ie ld of maltreatment has begun to move beyond the delineation of symptoms towards efforts to conceptualize the impact of chi ld maltreatment on i n d i v i d u a l s ' funct ioning. Finkelhor (1990) has emphasized the need for a theoretical framework that could accommodate the specific characteristics of the various forms of child maltreatment and assist in the interpretation and integration of the research results. A lack of comprehensive, integrative conceptual frameworks across investigations has been frequently noted (e.g., Aber , A l l e n , Carlson, & Cicchetti, 1989; Cicchetti, et. al. , 1989; Cole & Putman, 1992; Crittenden & Ainsworth, 1989; Fantuzzo, 1990). M a s h and Wolfe (1991) suggested that current models of child maltreatment have adopted a multifactorial , ecological, transactional approach. In such an approach, the interdependencies and reciprocities among causal variables and outcomes, the dynamic manifestations of abuse, and the numerous different conditions and pathways under which abuse may occur, are central concepts. The impact of specific harmful parenting behaviors on maltreated children might be best understood by use of an organizational/developmental perspective (Cicchetti, 1989; Cicchetti & Braunwald, 1984). C h i l d development is conceived as a series of quali tat ive re-organizations among and w i t h i n behavioral systems. Developmental periods are seen as being characterized by pivotal tasks which must be adequately resolved before the child is able to progress to the next period. Each salient task is thought to contain elements f rom social, emotional, cognitive, and social-cognitive domains, as wel l as their complex inter-relationships (Cicchetti & Braunwald, 1984). Arousal modulation, differentiation of affect, formation of secure attachments, development of individuat ion and autonomy, and enthusiasm for problem-solving are thought to be the most salient developmental tasks in infancy and childhood (Cicchetti, 1989; Cicchetti & Braunwald, 1984). Proponents of an organizat ional /developmenta l perspective v iew mal -treatment as resulting from ongoing transactions among different factors (e.g., child characteristics, caregiver characteristics, environmental factors) and ways in which the factors reciprocally influence each other to lead to disturbances in caregiving. Thus, chi ld maltreatment is understood as a disturbance i n the parent / ch i ld relationship that is first manifest in infancy as insecure, disorganized relationships between caregivers and infants (Crittenden & Ainsworth , 1989; Fantuzzo, 1990). A central facet of organizational/developmental models is that competencies i n ch i ldhood develop across mul t ip le domains and along a cont inuum of progressive stages (Cicchetti, 1989). Therefore, i n order to understand more comprehensively the impact of various forms of maltreatment, a careful considerat ion of deve lopmenta l factors is r e q u i r e d , beg inning w i t h the identification of the qualitative differences i n the competencies and capacities of children at different ages. Researchers need to examine the relationship between attachment, self-systems, and emergent psychopathology i n maltreated children as an adaptation to certain types of child-rearing environments (Aber et al., 1989). For example, children from abusive environments may learn that expression of anger and anxiety threatens their relationship with their caregiver. The development of avoidant behavior may be a way for these children to decrease their arousal levels and cut off, repress, or falsify their expression of anger or anxiety (Cassidy & Kobak, 1988). Cassidy and Kobak (1988) suggest that the defensive component of these avoidant behaviors may remain i n some form in adulthood. For example, such children might be found to hide affection and anger as adults, to be self-reliant, and to have no true interest in interactions. The salient developmental tasks in infancy and chi ldhood w i l l be briefly reviewed to provide a reference point for the later discussion of pathological development in childhood and the continuity of such pathological developmental across the life-span. Attachment Theory: Parent-Infant Interaction The first developmental task or stage that children are thought to face is the formation of a secure attachment relationship with a caregiver. This appears to be particularly salient when children are 6 to 12 months o ld , although proponents of the model suggest that stage-appropriate tasks remain critical to a child's continual adaptation, although decreasing somewhat in salience relative to other newly emerging tasks (Cicchetti, 1989). Bowlby (1988) theorized that the capacity to make bonds w i t h others is a pr incipal component of effective personality functioning and mental health. He hypothesized that the pathway followed by developing individuals and the extent to w h i c h the indiv idual becomes resilient to stressful life events are determined to a significant degree by the pattern of attachment individuals develop dur ing their early years. Bowlby also noted that the patterns that develop have been found to be influenced profoundly by the way caregivers treat their children. A predictable outcome of an infant's attachment behavior is the attainment of proximity to a trusted person (Crittenden & A i n s w o r t h , 1989). A typical pattern of interaction between caregivers and children is described as exploration from a secure base, wherein the caregiver is used as a secure base from which the infant can become acquainted w i t h the world and others in it. Therefore, failure in the response of the caregiver, whether due to physical absence or failure to respond appropriately, is believed to always cause stress, and sometimes to be traumatic (Bowlby, 1988). Cr i t tenden and A i n s w o r t h (1989) address attachment as a qualitative construct, reflecting the degree to which attachment is characterized by feelings of security and insecurity. A number of attachment categories have been proposed as coherent organized strategies infants use for relating to their caregiver i n times of stress, fatigue, and illness, which develop as a function of the infant's interaction w i t h the caregiver (Carlson, Cicchetti , Barnett, & Braunwald, 1989). When the caregiver is sensitive in his or her responsiveness to the infant's signals, and the infant separates from the caregiver w i t h minimal distress and is easily reassured u p o n reunion, then the attachment is described as a secure attachment (Type B). Converse ly , if the caregiver is inaccessible, unresponsive, or inappropriate in responding, and is inconsistent, and the infant is ambivalent when the caregiver is responsive, demonstrates great distress w i t h minimal separation and is hard to soothe, the attachment is described as anxious/ambivalent-resistant (Type C). Finally, when the caregiver is inaccessible, unresponsive and primari ly rejecting, angry and withholding of physical contact, and the infant shows little distress in high stress situations and avoids the caregiver when reunited, the attachment is described as anxious/avoidant (Type A) . (Carlson et a l . , 1989; Crittenden & Ainsworth, 1989). Attachment theory has addressed attachment as a representational construct (Crittenden & Ainsworth , 1989). Chi ldren are seen as developing a representational model as a function of experiences wi th their caregivers. They develop internal working models of self and others that consist of expectations for future events and help them plan their behavior (Bowlby, 1988). Thus, securely attached children view the caregiver as responsive and accessible and themselves as competent of eliciting a response and worthy of it. However, if the caregiver is inconsistent a n d / o r rejecting, the children may feel they cannot trust their caregiver's response and v iew themselves as ineffective in obtaining the caregiver's response. It is hypothesized that the effect of the internal working models that underlie anxious attachments is to change behaviors in ways that exacerbate stress and decrease the likelihood of future secure attachments (Crittenden & A i n s w o r t h , 1989). The Development of Autonomy A second developmental task, salient from approximately 18 to 36 months, is the development of an autonomous self. A caregiver's sensitivity and ability to tolerate a toddler's strivings for autonomy, as w e l l as the capacity to set age-appropriate limits, are integral to the successful resolution of this stage (Cicchetti, 1989). Attachment theorists hypothesize that securely attached children w i l l explore their environment, experience satisfaction at their successes in independently solving problems, and w i l l develop visual self-recognition, an indicator of the emerging self-system and the beginnings of individuat ion. Differentiation of affect and empathetic and prosocial responding also emerge at this time (Cicchetti, 1989; Cicchetti & Braunwald, 1984). C h i l d r e n come to realize that they can have an impact on others, and they develop expression, mediat ion, and control of affective states (Cicchetti & Braunwald, 1984; Erickson & Egeland, 1987). They learn to inhibit inappropriate behavior related to strong affect, self-soothe physiological arousal that strong affect may induce, refocus attention, and organize coordinated action (Gottman & Katz, 1989). Affective and cognitive development are thought to be interactively related. Thus, maltreated children might be hypothesized to develop hypervigilance and ready assimilation of aggressive stimuli , such that they may interpret ambiguous stimuli as threatening and aggressive which may reflect the negative expectations that they have formed about interpersonal relationships. Erickson and colleagues (Erickson, Egeland, & Planta, 1989) speculated that demands at different stages of development may bring out greater problems i n one maltreated group or another. For example, children who are sexually abused in childhood may experience greater problems in the early school years, where there are demands for independence and the establishment of peer relationships. Sexual abuse may decrease the child's chance for increasing social experiences and the establishment of a sense of self and social competence (Cole & Putman, 1992). Conversely, chi ld physical maltreatment that begins in infancy may interfere with the successful resolution of all developmental tasks, and may contribute to the development of h ighly negative internal working models and expectations which may contribute to physically maltreated children behaving in ways that perpetuate their negative expectations. Emotional and Cognit ive Development Researchers also have addressed emotional and social cognitions and perceptions and attributions i n maltreated c h i l d r e n . The development of autonomy, emot ional regulat ion, and cognit ion are closely related to the developmental task of establishment of peer relationships and social cognition, which is part icularly salient dur ing the pre-school and early school-age years (Cicchetti, 1989, Cicchetti & Braunwald, 1984). Social cognition involves the ways in which children develop understanding of other's emotions, attributions for other's behavior, and justification for their own behavior (Smetana & K e l l y , 1989). The development of insecure attachments may place maltreated children at risk for having more generalized poor-quality working models of the self and the self in relation to others (Cicchetti, 1989; Lynch & Cicchetti, 1991). Smetana and Kelly (1989) have suggested that maltreated children's social behavior is related to their interpretation of aggressive and unfair situations, and that researchers therefore need to consider maltreated children's interpretations and structuring of their social w o r l d when examining adjustment. Peer Relationships C h i l d r e n w h o evidence anxious attachments, poor autonomy and modulation of arousal, and biased social cognitions w o u l d appear to be at high risk for the development of poor peer relationships. Fatout (1990) stated that maltreated children's failure to develop a sense of self is a major deterrent in their ability to form relat ionships w i t h others. It is theoretically assumed by attachment researchers that other relationships come to be affected by the infant-parent attachment relationship (Youngblade «& Belsky, 1989). To summarize the literature regarding peer relations of maltreated children, heightened aggression appears to be present i n maltreated children's, especially physically maltreated children's, peer interactions, and context appears important in the expression of aggressive behavior. A n excessive degree of withdrawal a n d / o r avoidance of interaction with peers has also been noted (e.g., George & M a i n , 1979; Herrenkohl , Herrenkohl , Egolf, & W u , 1991; H o w e s & Eldredge, 1985; Howes & Espinosa, 1985; K a u f m a n & Cicchett i , 1989; R e n k e n , Ege land, M a r v i n n e y , Mangelsdorf, & Sroufe, 1989; Wolfe & Mosk, 1983). Addi t ional ly , different forms of maltreatment appear to have different effects on children's social development. Thus, it appears that a child's social development and peer relationships w i l l be affected both by the particular quality of the parent-child relationship and by the family's interaction wi th the broader social system (e.g., relative isolation). Factors such as social support or its lack may serve to buffer or potentiate the effects of maltreatment (Mueller & Silverman, 1989). Therefore, it is desirable to examine the specific interpersonal problems experienced by i n d i v i d u a l s from a variety of traumatic backgrounds and to attempt to isolate the critical factors that may be related to the establishment and maintenance of such problems. Sun\n\ary of Attachment Theory The review of attachment theory suggests that the pathways fol lowed by developing chi ldren are determined to a significant degree by the pattern of attachment the chi ld develops d u r i n g the early years. By ident i fy ing family experiences that result in different forms of attachment, researchers have been able to identify some of the determinants of children's future development i n areas such as individuat ion, emotional regulation and cognition, and peer relations (Bowlby, 1988). However , the relationship between early attachment and later outcomes is not yet ful ly understood. Research indicates that attachment classifications may change over time, that attachment classifications do not fit al l chi ldren, and that there m a y be other major determinants of children's future resil ience or vulnerabi l i ty (e.g., Belsky & N e z w o r s k i , 1988; Carlson et al , 1989; Crittenden & A i n s w o r t h , 1989). Mueller and Silverman (1989) argue that the central theme of the findings wi th respect to maltreated children's social functioning is that the effect of maltreatment on children's social funct ioning is not necessarily a global phenomenon that manifests itself uniformly across contexts. They suggest that several m e d i a t i n g variables m a y be important i n d e t e r m i n i n g whether maltreatment and attachment patterns w i l l result in maladaptive behavior. Carlson and associates (Carlson et al . , 1989) also emphasized the need to consider compensatory and potentiating mediating variables such as onset, duration and severity of maltreatment, context, and social support in order to understand the diverse pathways toward developmental outcomes. They note that the most widely used method of assessing the quality of infant-parent attachment is the Strange Situation paradigm (Ainsworth , Blehar, Waters, & W a l l , 1978). The Strange Situation was developed with a middle-class sample, but has been used in studies of children, such as maltreated children, who have experienced far greater degrees of caregiver failures than children i n the or iginal sample. Some chi ldren have presented wi th behavior that has been difficult or impossible to classify as secure, anxious/resistant, or anxious/avoidant. Carlson and her colleagues (Carlson et al. , 1989) have reanalysed their data f rom the H a r v a r d C h i l d Maltreatment Project. They found that most of previous coding disagreements came p r i m a r i l y from infants who had disorganized/ disoriented attachments, reflected by a lack of coherent, organized strategies for dealing wi th the stresses of the Strange Situation ( M a i n & Solomon, 1986). Not ing that a significant criticism of the attachment paradigm has focused on the absence of specific hypotheses regarding the outcomes of the different attachment categories, Carlson and colleagues suggest that the use of a disorganized/disoriented category might eliminate misclassified infants, and increase the reliabililty and predictive validity of the attachment system. However, there is a need for more research that utilizes this new category i n order to evaluate its potential usefulness. Crit tenden and A i n s w o r t h (1989) describe another potential attachment pattern observed in children of mothers whose behavior is consistently and severely distorted. Such children have been described as showing high proximity-seeking, high avoidance, and high resistance all dur ing one observation session (Crittenden, 1985a) and have been classified as avoidant/ambivalent. These findings that some children are difficult to accurately classify underscore the value of further research to increase the predictive power of the attachment system (Carlson et al., 1989). This is par t icular ly important in l ight of f indings of exceptions to the theorized relationship between infant attachment and pre-school behavior problems (e.g., Erickson, Sroufe, & Egeland 1985), and the l imited manner in which attachment theory has currently been applied to outcome in sexually abused children (e.g.. Cole & Putman, 1992). The continuity and discontinuity of attachment classifications is another important issue. Pianta, Sroufe, and Egeland (1989) reported low, but significant continuity i n global observations of maternal sensitivity from infancy to pre-school. However, there were cases of discontinuity, which appeared to be closely related to mediating factors such as personal and chi ld stressors. For example, some caregivers may be able to meet the needs and demands of an infant, but for a number of reasons not have the resources to cope w i t h an older child (Erickson et al.,1985). Similarly , Egeland and Sroufe (1981) found that 52% of their maltreatment sample changed attachment classifications from the time their infants were 12 months to when they were 18 months old. Positive changes in attachment appear to be related to changing life events, family support, and out-of-home care. In summary, it is evident that researchers must carefully examine maltreating parents' and their children's behavior over time, differentiating between subtypes of caregiving and context. A s Erickson and associates (Erickson et al . , 1985) emphasized, different subtypes of maltreatment may relate to the manifestation of different kinds of problems, i n different ways, in different contexts. Developmental Psychopathology O r g a n i z a t i o n a l / d e v e l o p m e n t a l theorists are not on ly interested i n understanding the ways in which children resolve developmental tasks, but also hypothesize about the predictive nature of varied resolutions. Cicchetti (1989) stated that although attachment is a stage-salient issue, attachment relations are important across the life span. H e posited that while not all maltreated children experiencing d i f f i c u l t y i n the reso lut ion of stage-salient tasks deve lop subsequent psychopathology, future disturbances in functioning are likely to occur. Proponents of attachment theory have argued that many psychopathological disorders may be brought about by deviations in the development of the attachment system (Carlson et al., 1989). The relation between difficulties in development and specific forms of psychopathology that may emerge, is a domain of developmental psychopathology. Developmental psychopathology is based on the assumption that researchers can use a developmental approach, examining mult iple domains of development, to help understand the variety of factors that play a role in the etiology, course, and sequelae of maladaptive and pathological development. Pathological development is therefore viewed as a lack of integration among socioemotional, cognitive, social -cognitive, and representational competencies, or as an integration of pathological structures (Cicchetti, 1989; Cicchetti & Howes, 1991). Organizat ional /developmental theory implies that the central impact of maltreatment on early development is a distortion i n children's dynamic balance between security-promoting operations a n d competence-promoting operations. Aber and associates (Aber et al., 1989) suggested that this imbalance w i l l prove to be a replicable across-stage feature of the development of maltreated children. They speculated that such a "structuralized" preoccupation w i t h security-promoting operations over competence-promoting operations is a causal/mediating factor that links a history of maltreatment wi th psychopathology in the pre-school years. Sroufe and Rutter (1984) proposed that children's failure to adapt to the environment is a critical component i n the relationship between early experiences and later psychopathology. Maltreated children's methods of adapting to environmental demands may later compromise their ability to form relationships. Maltreated children may lose important opportunities for exposure to social support and buffers against stress when they learn a style of avoidance as a function of poor parental attachment. Maltreated children's difficulties may best be viewed as a product of the transactional relationship between children's emerging personal characteristics, parental treatment, and c i rcumstant ia l /envi ronmenta l factors (Wolfe, 1987). It is possible that different processes and mediators underlie the development of psychopathology for different sub-populations of children-at-risk (Aber et al., 1989). Wolfe (1987) suggested that abused children may suffer the greatest amount of adaptational failure relative to children from distressed families, to the extent that they are deprived of positive adult relationships, exposure to effective problem-solving, and feelings of control or predictability. The extent to which children express developmental impairments may be a function of the person by situation interaction, where child characteristics (e.g., information processing, coping skills, physiological reactions) interact wi th the nature of the stressor (e.g., type, frequency. and duration of maltreatment/family disruption) to determine whether or not the c h i l d w i l l experience trauma and persist in negative chronic stress responses (Cicchetti & Howes, 1991; Wolfe & Jaffe, 1991). A n important issue for investigation is whether different types of childhood (or adolescent) stressors are related to different adaptations (styles of coping with these different stressors). In defining the "stressor", researchers have attempted to delineate the constellation of factors that might dist inguish different types of maltreatment and related outcomes. For example, chi ldren f r o m physical ly maltreating families may not only experience greater violence and vict imization than children from chaotic/disrupted families, but also may be faced w i t h a more extreme fami ly environment that fails to provide appropriate social izat ion opportunities, and is replete with disruptive events that affect development i n both subtle and blatant ways (Wolfe & Jaffe, 1991). It may be more useful to consider the characteristics of different types of dysfunctional family situations, rather than focusing solely on the occurrence or non-occurrence of specific abusive acts. Three types of dysfunctional family situations (i.e., physical maltreatment, chaos and disrupt ion, sexual maltreatment) w i l l be discussed in the fol lowing sections. The literature w i l l be discussed wi th in a developmental psychopathology framework; that is , the relation between difficulties i n development and the emergence of specific forms of psychopathology w i l l be highlighted. Physical Maltreatment Physical maltreatment characterizes one subgroup of maltreating families. Physical ly maltreated children have been found to display a number of problems in various developmental domains. A number of well -designed studies of early attachment relationships have documented that physical ly maltreated children often display anxious attachments, characterized by avoidant, non-compliant, and angry behavior (e.g., Crittenden, 1981; Egeland & Sroufe, 1981; Erickson et a l , 1989; Lyons-Ruth, Connel l , & Z o U , 1989). For example, us ing the Strange Situation, Erickson, Egeland, and Pianta (1989) found that children w h o had experienced a number of forms of maltreatment (e.g., physical abuse, neglect, psychological unavai labi l i ty) d isplayed a relatively high incidence of anxious attachment. Physically maltreated children also have been noted to have difficulty establishing autonomy and secure readiness to learn, over and above effects attributable to family chaos associated w i t h lower class status (e.g., Aber & A l l e n , 1987; Erickson, Sroufe & Egeland, 1985; Erickson et a l , 1989). Addit ional ly , research has pointed to disruptions i n these children's social cognitions and perceptions (e.g.. Dodge, M u r p h y , & Buchsbaum, 1984; Smetana & Kel ly , 1989), as evidenced i n bias towards inferring hostile intent, and a tendency to justify real-l ife aggressive acts as retribution for others' bad actions. Finally, physically maltreated children have been found to have disturbances in peer relations. They have been observed to be more aggressive and/or inhibited in their approaches to peers, and avoidant in response to friendly overtures, w h e n compared to non-physical ly abused children (e.g., Erickson et al . , 1985; George & M a i n , 1979; Herrenkohl et a l . , 1991; Howes & Eldredge, 1985; M a i n & George, 1985). Adolescents who have experienced physical maltreatment also appear to demonstrate developmental impairments . Several studies have begun to document a moderately strong association between c h i l d maltreatment and adjustment problems appearing during adolescence (Wolfe, 1988). Patterns of physical abuse and outcome have been studied in maltreated adolescents, but there is a marked lack of research and theory devoted to this age group. The existing literature suggests that abused adolescents display a number of "patterns" of adjustment, i n c l u d i n g helplessness/dependency, depress ion/ isolat ion, aggres-siveness/hostility, and somatization (Farber & Joseph, 1985; Hjorth & Ostrov, 1982). In their summary of adolescent abuse, Galambos and Dixon (1984) stated that physically abused teenagers suffer from low self-esteem, such that they are more likely to perceive events as contingent on luck, chance, fate and/or powerful others rather than contingent on their own actions. The teens displayed high levels of anxiety and academic problems and possessed poor coping skills, often relying on alcohol and drugs to cope with psychological stress. Final ly, abused adolescents struggle w i t h feelings of anger, violence, and resentment and often display a lack of empathy for peers. If one assumes that empathy underlies social responsibility, then it is not surprising that abused adolescents are more aggressive and delinquent than their non-abused peers. Adolescents face different developmental issues and tasks that may stress high-risk parents who are unable to adaptively meet these new challenges to their parenting. Garbarino (1989) suggested that at-risk adolescents are less socially competent than their peers and experience greater developmental problems. Lewis, Mal louh , and Webb (1989) indicated that whi le most physically abused children do not become violent delinquents (approximately 20% of physically abused children become delinquent), a high percentage of delinquent adolescents have been severely abused. Herrenkohl and colleagues (Herrenkohl et al . , 1991) speculated that abusive pre-school-aged parenting that continues into the school-aged years w i l l result in lowered social competence in adolescence, and that chi ldren whose social competence is low develop into adolescents wi th equally low or even poorer levels of social competence. In summary, children from physical ly maltreating homes have been found to display developmental difficulties throughout chi ldhood and adolescence, characterized by non-compliant, acting out, and aggressive behavior. Family Chaos and Disruption It is theoretically desirable to investigate the development and fami ly characteristics of children from disturbed and chaotic environments that are not physically or sexually abusive. In that child abuse may be viewed as the degree to which a caregiver uses negative, inappropriate control strategies wi th children, it is desirable to understand the impact of abusive environments over and above the effects attributable to family chaos and disruption (Aber et a l , 1989). Researchers have examined the development of children from chaotic and high-risk homes and the results of a number of such studies have been somewhat inconsistent. For example. Dodge and associates (Dodge et a l , 1990) found that children from stressful or chaotic homes that were not physically abusive were not at as great a risk for later aggressive behavior as children from physically abusive homes. Similarly, M a i n and George (1985) noted that non-physically maltreated but stressed children responded with concern, empathy, and sadness to the distress of others, whereas physically maltreated children responded with threats of anger and physical attacks. Conversely Wolfe and Mosk (1983) found that the behavior patterns of the physically abused children in their study resembled the wide range of behavior problems displayed by the children from distressed families. C h i l d witnesses to family violence constitute a group of children from families characterized by chaos and disruption, and a number of investigators have focused on this group as wel l . M a n y of the recent studies i n this area provide support for the existence of a relationship between marital conflict and children's adjustment di f f icul t ies (e.g., social and behavioral incompetence, cognitive impairments) (e.g., Jaffe, Wolfe, Wilson, & Zak, 1986; Fantuzzo et al. , 1991; Wolfe, Zak, Wilson, & Jaffe, 1986; Wolfe, Jaffe, Wilson, & Zak, 1985). However, as Grych and Fincham (1990) indicate, there is still l imited understanding of the mechanisms underlying this relationship. They suggest that the extent and type of adjustment problems associated w i t h f a m i l y conflict m a y di f fer d e p e n d i n g on the developmental task faced by the chi ld, and emphasize the need for future research addressing the cognitive, affective, and behavioral responses of all family members and the role of environmental factors. Research has begun to address such unique characteristics of different types of family environments that may be related to maladaptive outcomes. Cicchetti and Howes (1991) noted that it is important to include the child-rearing context/family climate in maltreating families in a developmental psychopathology framework. They speculated that parenting styles and attitudes of maltreating parents may have the most direct impact on children's indiv idual development. In comparison to adequate and inept mothers, the fol lowing characteristics have been observed in physically abusive mothers: insensitivity to their children's needs; behavior being planned according to the mother's needs; failure to use behavior reciprocally; m i n i m a l enjoyment of parenting characterized by l o w satisfaction w i t h their children and little expressed affection; isolation; failure to encourage their children's autonomy and independence; and possession of high expectations for their children (e.g., Aber et a l , 1989; Aber et al. , 1989; Crittenden, 1981; Trickett, Aber, Carlson, & Cicchetti, 1991). Investigators have noted that much literature suggests that such an environment has deleterious consequences for chi ldren across many domains, including social, emotional, and cognitive development. Sexual Maltreatment In general, the literature in the area of child maltreatment is suggestive of the existence of differences i n the chi ld sequelae associated w i t h different family climates. These differences have been noted to exist over and above the similarities in developmental and behavioral problems that are found to characterize children who have experienced many forms of maltreatment/family environments. Chi ld sexual maltreatment is another form of maltreatment that might be hypothesized to be associated with some unique child sequelae. Wolfe and Jaffe (1991) stated that ch i ld sexual maltreatment, just as other forms of c h i l d maltreatment/ family disrupt ion, can create a vulnerabi l i ty that leads to developmental adjustment problems and/or stress-related disorders. The diversity of interpersonal, affective, and cognitive symptoms exhibited by sexual abuse survivors may theoretically be hypothesized to be mediated by their attachment experiences (Alexander, 1992). Cole and Putman (1992) indicated that a major difficulty i n the area of child sexual abuse is a lack of a developmentally sensitive theoretical organization of outcome, and suggest that self -development is a central o r g a n i z i n g construct for understanding psychopathological outcomes of childhood sexual abuse. The risk of childhood sexual abuse has been shown to increase significantly after approximately age 8, during the prepubertal and pubertal stages (Bagley & Ramsay, 1985; Steele and Alexander, 1981). One could hypothesize that the related stage-salient issue of peer relations might be most affected by chi ldhood sexual abuse. C h i l d sexual abuse may decrease children's chances of exposure to social experiences, and interfere with the establishment of a sense of self-competence in the social w o r l d , and with the integration of positive and negative aspects of the self (Cole & Putman, 1992). Alexander (1992) stated that interpersonal relationships appear to be the area in which many of the long-term conflicts related to sexual abuse are manifest. A d d i t i o n a l l y , children's interpretation of sexually abusive experiences and their perception of their mother's response appear to be related to outcome (Beitchman, Zucker, H o o d , DaCosta, A k m a n , & Cassavia, 1992). It is possible that the impact of c h i l d h o o d sexual abuse compromises earlier developmental accomplishments as w e l l , such as the establishment of self -regulatory functions and trust/sensit ivity i n social relations. The l ikel ihood of such disruptions might be increased by certain contextual factors, such as the child's coping ability and individual differences i n family context (Cole & Putman, 1992). For example, in contrast to the chronicity of the events related to most types of physical ch i ld maltreatment and high risk family environments, some forms of chi ldhood sexual abuse may involve single-incident or short-term events, and may be embedded in a family context that is otherwise stable, particularly if the abuse was extrafamilial . The types of sexual abuse that appear to be the most damaging involve father-figures (i.e., intrafamilial), betrayal of trust, genital contact and force, and longer histories of abuse (e.g., Briere & Runtz, 1988b; Browne & Finkelhor, 1986; Finkelhor, 1990). Finkelhor (1990) pointed out that it has been suggested that as many as one-quarter to one-third of victims of chi ldhood sexual abuse may be without symptoms as measured by clinicians. H e speculated that these children may be the individuals who suffered less severe sexual abuse a n d / o r w h o had adequate psychological and social resources to cope with the sexual abuse. Wolfe and Jaffe (1991) noted that an example of a developmental process that is more or less specific to child sexual abuse is psychosexual development. The onset of puberty and emerging sexuality are stage-salient issues of adolescence, together w i t h the integration of aspects of the self into a coherent whole (Cole & Putman, 1992). These developmental tasks may be uniquely jeopardized by child sexual abuse, given the sexual nature of the abuse and the greater l ike l ihood of sexual abuse, i n contrast to physical abuse, continuing through this stage of development. Whi le sexual abuse does not appear to be associated w i t h any particular diagnostic category, a recurrent f inding is the existence of disturbed psychosexual functioning (e.g., Briere & Runtz , 1990; Browne & Finkelhor, 1986; Finkelhor, 1990). Gelinas (1983) postulated that there are underlying negative effects as w e l l , including: chronic traumatic anxiety (characterized by repetitive nightmares a n d / o r obsessive ideas); panic with the offender or in situations where there are reminders of the offender; denial and dissociations; role-reversal (wherein the child learns to put other's needs before his or her own); under-developed social skills and low esteem; a distorted idea of the balance of obligation and entitlement in relationships; and an increased risk of intergenerational abuse. The empirical literature lends support to proposals such as Gelina's. Conte (1985) studied 369 children who were assessed at or near the point of disclosure of sexual abuse, and a control sample of children. The five most consistently noted difficulties were low self-esteem, fear of abuse-related s t imul i , emotional upset, nightmares and sleep disorders, and withdrawal from usual activities. The abused children were also more likely to be overly anxious to please adults. Wolfe and Jaffe (1991) noted that younger sexually abused chi ldren are reported as having precocious knowledge of sexual behavior and sexual terms, and that older sexually abused children demonstrate unusual and high risk sexual behaviors i n comparison to peers (e.g., excessive masturbation, promiscuity). It may be that sexually v ic t imized chi ldren learn coercive interchanges and related maladaptive and inappropriate sexual behaviors and attitudes. They may develop situation-specific fears, more pervasive fears, a n d / o r a pronounced fear response that may readily generalize to other stimuli (Wolfe & Wolfe, 1988). Some researchers have identif ied post-traumatic stress disorder as an outcome of childhood sexual abuse (e.g., Lindberg & Distad, 1985; Wolfe et al. , 1989) i n an attempt to conceptualize the impact of sexual abuse. Symptoms such as intrusive imagery, feelings of detachment or constricted affect, sleep disturbances, and guilt have been noted in a proportion of sexual abuse survivors. While this formulation has added some insight into the trauma experienced by some victims, researchers also need to consider the family context that mediates the experience of sexual abuse. Family characteristics/climate may be stronger predictors of later adjustment than abuse variables (e.g., duration, force, age of onset) (Alexander, 1992; Cole & Putman, 1992). It is important to account for potential mediators such as family psychopathology and emotional climate (violence, substance abuse, depression, parental coldness), and children's interpretation of their experience, (e.g., attributions, beliefs, cognitions), their perception of significant other's response, and their coping abilities (Bagley & Ramsay, 1985; Crittenden, Fantuzzo, & Lindquist , 1989; Finkelhor, 1990; Inderbitzen-Pisaruk et al . , 1992; Wolfe & Wolfe, 1988). Developmental Psychopathology and the Life-Span Perspective In summary, the literature i n the area of chi ld maltreatment generally supports the conclusion that various forms of child maltreatment and family chaos affect child development in both subtle and blatant ways and that different processes and mediators may underlie outcomes of different subpopulations of children-at -risk, he relation between difficulties in development and specific forms of child psychopathology that may eventuate was discussed f rom the v iewpoint of developmental psychopathology. Given the level of disruption noted in children from maltreating and chaotic families, it is desirable to extend the discussion of the theoret ical re la t ionship between c h i l d maltreatment and deve lopmenta l psychopathology into adul thood. Alexander (1992) emphasized the need to investigate the relative stability of developmental resolution/attachment style, with consideration given to intervening experiences that might be expected to change individual 's internal working models. A s Wolfe and Jaffe (1991) elucidated, relationship formation seems to be impaired i n child victims of maltreatment from early on, setting the stage for later difficulties and maladjustment. The ways i n which different types of chi ldhood stressors and adaptations are reflected in adult adjustment w i l l be discussed in the fo l lowing section. Theoretical Relationship Between C h i l d h o o d Maltreatment, Developmental Psychopathology, and A d u l t Functioning A major research concern is the degree to which maltreated children grow up to be either maltreating adults or to exhibit socially deleterious sequelae of child maltreatment (Starr Jr., M c L e a n , & K e a t i n g , 1991). A n o r g a n i z a t i o n a l / developmental perspective encompasses the tenet that an individual 's degree of vulnerability a n d / o r resilience to stressful life events is strongly influenced by the development and current state of his or her int imate and interpersonal relationships. The capacity to make bonds w i t h others is viewed as a principal feature of effective personality functioning and mental health (Bowlby, 1988; Flaherty & Richman, 1986; Peplau & Perlman, 1982; Young, 1982). Poor parent-child bonds and dysfunctional relations have been observed to impair later interpersonal functioning (e.g., Flaherty & Rickman, 1986; Straus, 1980; White & Straus, 1981). The pattern of interaction that develops between parents and children has been found to be profoundly influenced by the way parents treat their children. Bowlby (1988) argued that such interaction patterns tend to be self-perpetuating, such that there is a tendency to impose earlier patterns onto new relationships. Proponents of organizational models of development conceptualize development as consisting of a number of important age and stage-appropriate tasks, which upon emergence, remain critical to a child's continual adaptation, although decreasing somewhat i n salience relative to other newly emerging tasks. Therefore, even though parent-child attachment is a stage-salient issue, attachment relations are seen to be important across the life span (Cicchetti, 1989). Addit ional ly , Cicchetti suggests that although not all maltreated children who manifest problems in the resolution of stage-salient issues develop subsequent psychopathology, future disturbances in functioning seem likely to occur. C h i l d maltreatment is conceptualized as interfering w i t h the necessary developmental transitions in ways that increase the risk for interpersonal problems and serious psychopathology. Early experiences wi th the caregiver relate to the evolution of the child's expectations of other's availability i n times of stress/need, and a complementary model of the self as worthy or unworthy of care (Egeland et al . , 1988). Individuals may br ing the internal work ing models formed dur ing infancy/chi ldhood forward to adulthood. Therefore, infants and children whose needs have not been adequately or appropriately met may be more likely as adults to have difficulty entering into supportive relationships with others (Parker, Barrett, & Hickie , 1992). Similarly, peer relationships and interaction skills are presumed to be important for later social development and are facilitated by secure attachment relationships with parents and family involvement with social networks (Grych & Fincham, 1990; Howes & Eldredge, 1985; Starr Jr. et al., 1991). Greenberg and Safran (1987) suggested that emotional and interpersonal responding is influenced directly by experiential learning. Maltreating/chaotic family environments can be v iewed as negative learning environments. A child who fails to learn to develop interpersonal trust and effective problem-solving strategies, and who learns to associate maladaptive affect, appraisals, arousal and behaviors w i t h interpersonal situations, has missed important developmental / social ization experiences, and the resulting developmental impairments may interfere w i t h adolescent and adult relationships (Wolfe, 1987; Wolfe & Bourdeau, 1987). The idea that child maltreatment is related to significant adult sequelae has gained both theoretical and empirical support. There have been more empirical studies focusing on survivors of chi ld sexual abuse than on the adult sequelae of physical abuse (Starr Jr. et al., 1991). Although the work with maltreated children is suggestive of di f fer ing sequelae for different types of mal t rea tment / fami ly environments, there is a need for more research extending the chi ld findings to adult outcomes. A major issue is whether different types of ch i ldhood stressors produce different adaptations i n varying domains of functioning, and whether the adaptations are reflected i n areas of adult adjustment. The literature regarding the long-term outcomes of child maltreatment i n the domains of parenting behavior and interpersonal functioning w i l l be discussed, guided by an organizat ional / developmental theoretical perspective. Parenting Behavior of A d u l t Survivors of C h i l d Maltreatment C h i l d maltreatment has been conceptualized as interfering w i t h the necessary developmental transitions in ways that might increase the risk for survivors to have difficulty engaging in supportive relationships with others i n adulthood (e.g., Parker et al . , 1992). The mother-child relationship is a relationship that demands great supportiveness, sensitivity, and responsiveness. Organiza t iona l /deve lop-mental theorists have argued that parents' own childhood experiences and resulting internal models of relationships may influence their response to their children and others (e.g., Crittenden, 1985a, 1988a, b; M a i n & G o l d w y n , 1984). A number of models have been presented as ways to understand the development of severe parent/chi ld conflict. Belsky (1980) has presented an ecological model wherein he states that human behavior should be studied in its o w n context. Thus, as a parent's social context becomes, or is perceived as becoming, increasingly stressful, there is an increased probability of family violence arising as an attempt to gain control over the stress. Alternatively, the social interactional model (e.g., Parke & Collmer, 1975) examines the dynamic interplay between i n d i v i d u a l , family, and social factors in relation to both past and present events. Wolfe (1987) noted that proponents of this model consider parents' learning history, interpersonal experiences, and intrinsic capabilities as predisposing characteristics. To understand parenting behavior, it may be useful to focus on the dynamic relationship between a parent's individual and social characteristics in relation to both past (e.g., exposure to child maltreatment) and current events (e.g., difficult parenting and social situations) (Starr Jr., 1978, 1988; Wolfe & Bourdeau, 1987). Wolfe (1987) hypothesized that low tolerance for stress and dis inhibi t ion of aggression, poor management of acute crises and provocation, and habitual patterns of arousal and aggression with family members, are important factors to investigate. Wolfe presented a transactional model of child maltreatment, focusing on the above processes and others that contribute to hypothesized gradual changes in the parent-child relationship from m i l d to high-risk interactions. Wolfe also suggested that, in comparison to non-maltreating parents, maltreating parents are not as effective or successful in the parenting role, are less flexible or appropriate in their choice of disc ipl inary techniques, and have l imited chi ld management skills in general. These factors, along w i t h maltreating parents' perceptions of adverse family and environmental conditions, are seen to be exacerbated by the parent's failure to develop and use social supports to decrease stress and help w i t h problem-solving (Wolfe, 1987). Thus, the risk for maltreatment is increased when factors that serve to potentiate stress outweigh factors that play a compensatory or buffering role (Cicchetti, 1989; Wolfe, 1987). The results of a number of studies lend support to a transactional view of maltreatment. In comparison to non-maltreating mothers, physically maltreating mothers have been found to: have high rates of physical and verbal aggression (Bousha & T w e n t y m a n , 1984; Lahey, Conger , A t k e n s o n , & Treiber, 1984); demonstrate greater cognitive impuls ivi ty and less motor inhibit ion (Rohrbeck & Twentyman, 1986); discourage autonomy, lack resource access, and fail to enjoy their children (Aber et al . , 1989); and experience increased arousal to parent-child scenes, as well as displaying signs of anticipatory arousal (Wolfe, Fairbank, Kel ly , & Bradlyn, 1983). Parental perception and attributions about c h i l d behavior are important factors that also have been assessed. For example, Re id and associates (Reid, Kavanagh, & Ba ldwin , 1987) observed that, in comparison to non-abusive mothers, physically abusive mothers significantly over-estimated the level of chi ld conduct disorder and aversive behavior of their chi ldren, even though no significant differences were observed in child problem behaviors. The risk associated with increased levels of arousal to chi ld behavior w o u l d appear to be intensified by parental perceptions of the arousal as anger. Such perceptions, together with negatively-biased perceptions of the child and past coping abilities, might diminish physically maltreating parents' ability to use adaptive coping strategies (Lahey et al., 1984). In summary, it appears that maltreating parents differ from non-maltreating parents i n many aspects of their interactions wi th , and perceptions of, their children. M u c h more research has been done i n regards to physically abusive parenting than in regards to sexually abusive parenting. Given that family characteristics may be a stronger predictor of later adjustment than abuse variables per se, the ontogenesis of such m a l a d a p t i v e parent -chi ld interact ion patterns is of major interest. O r g a n i z a t i o n a l / d e v e l o p m e n t a l theorists w o u l d suggest that these parents' experiences wi th in their families-of-origin contribute to the noted difficulties with d is inhib i t ion of aggression, problem-solving and child-management ski l ls , and hostile perceptions and attributions. Vasta and Copitch (1981) argued that adults who experience frustration with a child's behavior may experience autonomic nervous system arousal, which in turn interferes with rational problem-solving and produces higher intensity responding. This pattern of arousal may be a result of an individual's predisposition (e.g., history of chi ld maltreatment, temperamental/physiological arousal, degree of stress) and h i s / h e r learning experiences in regards to relieving arousal. Thus, maltreating parents may begin to respond to cues that have been previously associated with frustration a n d / o r anger (in their o w n chi ldhood and/or with their o w n children), and the result ing behavior toward the chi ld may be intensified by the past experiences. A developmental v iew of parenting suggests that earlier forms of behavior become hierarchically integrated w i t h more complex, recent forms of behavior and then remain potentially active, especially in periods of stress (Wolfe, 1987). Therefore parents, who themselves experienced maltreatment/family chaos and related developmental di f f icul t ies , may fall back on the strategies and competencies learned and developed during childhood. The extent to which maltreated chi ldren become maltreating parents is difficult to state definitively, given the considerable variation i n the reported rates of intergenerational transmission across studies. Kaufman and Zigler (1989) stated that u n q u a l i f i e d acceptance of the hypothesis that abuse is transmitted intergenerationally is unwarranted. They noted that the f indings of different investigations are not readily comparable due to methodological variations, and stated that the best estimate of the rate of intergenerational transmission of abuse is 30% plus or minus 5%. Further research is needed to help delineate the conditions under which intergenerational transmission is likely to occur, w i t h consideration of the compensatory and risk factors, and the relative importance of each of these factors. A factor that has been hypothes ized to relate to intergenerational transmission is that of modell ing. Herzberger and Tennen (1985) speculated that model led aggression, as is the case in maltreating families, may disinhibit children and adults, leading them to believe that aggression is acceptable. The investigators asked college students to rate the abusive treatment of chi ldren portrayed i n vignettes. Results indicated that students who had themselves experienced the chi ld treatment portrayed in the vignettes, viewed the discipl inary methods less harshly and thought they were less severe and more appropriate than d i d students who had not experienced similar treatment as children. The students were less Ukely to believe that discipline w o u l d harm the children's emotional development and thought it was l ikely to decrease additional misbehavior. While Herzberger and Tennen acknowledged that thoughts do not always equal behavior, their results do support the idea that abusive parenting behaviors may be transmitted through the learning of perspectives about discipline (e.g., that aggression is appropriate). Another theoretical position is the organizational/developmental position, wherein it is believed that abusive patterns of behavior are transmitted through mental representations of past relationships (Kaufman & Zigler , 1989). M a i n and G o l d w y n (1984) used observational and interview data to investigate maternal behavior and women's memories of their own interactions wi th their mothers. It was found that, in comparison to women who remembered supportive interactions w i t h their own mothers, women who reported that their mothers were rejecting were more l ikely to reject their own children and to have incoherent recollections of childhood experiences. Similarly, Egeland, Jacobvitz, and Sroufe (1988) examined mothers' chi ldhood relationships and their current child-rearing behaviors i n a study of 267 families. Maltreating mothers abused as children (n=18) reported greater stressful l i fe events, anxiety, and depression i n comparison to non-maltreating mothers w i t h a history of abuse. They were less l ikely than the non-maltreating mothers to have had a relationship w i t h a supportive adult as a child, to have been in treatment, or to have a supportive spouse. The investigators noted that there was no w a y to estimate the accuracy of the mother's recall of childhood experiences, but they argued that one would expect errors and distortions to be more l ikely i n instances where maltreated mothers reported receiving adequate care. Thus, the results suggest that parents' o w n c h i l d h o o d relationships tap an important source of variance in parental behavior. Other researchers have examined the perceptions and attributions of physically abusive parents. Wolfe and LaRose (1985) hypothesized that child abusers (a number of w h o m are survivors of abuse themselves) experience conditioned arousal to stressful events (e.g., chi ld behavior) that resemble previous situations they have encountered. The parents may then attribute such arousal or annoyance to the child's behavior. To empirically test this hypothesis, Wolfe and LaRose showed eight 90-second videotaped scenes s imula t ing stressful chi ld-rearing situations to abusive and non-abusive mothers. Prel iminary results suggested that abusive mothers display differential reactions on dimensions such as sense of control, social perception, and negative affect. S imilarly , Bauer and Twentyman (1985) presented child-related and non-child-related stressful st imuli in the form of 90-second audiotapes to mothers wi th a previous history of chi ld abuse or child neglect or w i t h no k n o w n history of ch i ld maltreatment. Analyses of the attributional data indicated that abusing mothers consistently ascribed more malevolent intentionality to their children than d i d the control mothers. Abusive mothers also displayed a pattern of hyper-responsivity to stressors. Such studies emphasize the need to examine cognitive, attributional, and physiological processes when examining interpersonal behaviors in the area of physical maltreatment. W h i l e these studies demonstrate some s u p p o r t for the not ion of intergenerational transmission of abuse-related mediators , other mediat ing variables related to maltreating parent ing behavior need to be addressed. Addit ional ly , Starr Jr. and colleagues (Starr Jr. et al., 1991) noted that research on the intergenerational transmission of sexual abuse has only recently begun. It may be that parents who were sexually abused as children are more l ike ly than parents without such a history to transmit a distorted perception of sexuality and gender relations. However, it may be difficult to isolate the specific effects of a history of sexual abuse from concomitant family d is rupt ion and other forms of abuse. K a u f m a n and Zigler (1989) emphasize that further research is needed to broaden understanding of the i n d i v i d u a l mechanisms invo lved in the transmission of abuse and their inter-relationships. The Impact of C h i l d Maltreatment on General A d u l t Functioning The literature regarding the parenting behavior of adult survivors of child maltreatment suggests that parents' ch i ldhood relationships tap an important source of variance in parental behavior. Organizat ional/developmental theorists have proposed that the abusive patterns of behavior are transmitted through the internal working models formed dur ing childhood. Egeland, Jacobvitz, and Sroufe (1988) proposed that through early experiences w i t h the caregiver, individuals evolve expectations of the availability of others in times of need, and of the self as worthy or unworthy of care. They suggested that infants whose needs are not appropriately met are more l ike ly as adults to have d i f f i cu l ty entering into supportive relationships with others. Thus, chi ld maltreatment may impact upon m a n y aspects of adult survivor 's funct ioning. For example, Cri t tenden and A i n s w o r t h (1989) speculated that abusing mothers (a number of w h o m are survivors of abuse) have internal models centering around conflict, control, and rejection, accompanied by angry affect. These parents may expect that others w i l l try to dominate them and reject them when they attempt to have their o w n needs met. It may be that maltreated children and children from chaotic homes acquire such maladaptive attitudes and behaviors from their parents, and that these attitudes and behaviors continue to affect their lives even as adults (e.g., G r y c h & Fincham, 1990; Renken et al., 1989). Therefore, the ramifications of child maltreatment may be far-reaching, with great potential for inf luencing children's adjustment as adults , whether they become parents or not (Hughes, 1986). It is important to discern whether there are specific impacts of various forms of child maltreatment and family chaos on adult functioning, above and beyond any effects that they may have i n common. Terr (1991) suggested that there may be four features of childhood trauma that last for years and are often seen i n adults traumatized as chi ldren. These difficulties include: strongly visualized or otherwise repeatedly perceived memories, repetitive behaviors and physiologic re-enactments, trauma-specific features, and changed attitudes about people, aspects of life, and the future (e.g., profound vulnerability, belief no one can be counted on). Terr speculated that adult survivors of childhood sexual abuse may be more l ikely than survivors of other forms of maltreatment to shrink away from men, or to accost them with friendly overtures. Briere and Runtz (1990) examined the differential adult symptomatology associated w i t h ch i ldhood histories of phys ica l , sexual , and psychological maltreatment. Us ing a group of 277 female university students, substantial and unique relationships were found between different forms of maltreatment and different adult symptomatology. U n i q u e relat ionships were found between retrospective reports of parental psychological maltreatment and subsequent l o w self-esteem, between reports of physical maltreatment and later anger and aggression, and between reports of sexual maltreatment and dysfunctional sexual behavior. W h i l e reports of physical and psychologica l maltreatment often overlapped, subjects who reported experiencing sexual maltreatment typically were not physically maltreated as wel l , and displayed different symptomatology than the physically maltreated subjects. Briere and Runtz suggested that the hypothesis that some adult problems may co-vary with various forms of maltreatment is one that warrants study. The adult functioning of individuals from differing maltreatment backgrounds w i l l be reviewed below. C h i l d h o o d p h y s i c a l maltreatment a n d adul t func t ion ing . A s noted previously, there has been relative inattention to the long-term adult sequelae of physical abuse. However, clinical observations suggest the presence of interpersonal problems long after the abuse has terminated. Physical ly maltreated children's chronic exposure to inappropriate and inadequate vocal, physical , and emotional interaction may form the basis for later social ineptitude (Steele, 1986). It has been suggested that the behavior of some physically abused children (i.e., aggression) is a pattern that endures into adulthood (Lamphear, 1985). In fact, difficulties i n interpersonal functioning constitute some of the most frequently cited problems experienced by abuse victims (e.g., Barahal et al . , 1981; Lamphear, 1986). More specific examination of interpersonal difficulties in childhood suggests that a major factor i n physically abused children's poor peer relations is their heightened aggressiveness towards peers (George & M a i n , 1979; K i n a r d , 1980; Lamphear, 1986; M a i n & George, 1985). Another important factor relates to external locus of control, such that physically abused children have been found to display little confidence i n their power to impact upon and shape their social experiences, especially those that are unpleasant and frustrating (Barahal et al . , 1981). Such beliefs may lead to hopelessness and passivity in relationships. Physically abused chi ldren have been found to experience di f f icul ty comprehending social role concepts, which may render them less able to understand subtle and complex relationships between people (Barahal et al. , 1981). In addition, physically abused children have been found to be more egocentric and insensitive to social-emotional contexts than non-abused children, which places them at greater risk to be unable to effectively take another person's perspective or to accurately evaluate the feelings and motives of others (i.e., empathetic responding) (Barahal et al . , 1981; K i n a r d , 1980; Lamphear, 1985,1986). It appears that interpersonal functioning is a broad construct that is closely tied to psychological development. Healthy psychological functioning appears to be related to social support and the capacity to establish and maintain adult relationships (Flaherty & Richman, 1986; Peplau & Perlman, 1982; Young, 1982). The effects of physical chi ld maltreatment therefore might best be understood in terms of the ways in which maltreatment interferes wi th relationship development. The security of the early relationship formation between parent and child has been l inked to children's emerging mastery of their social environment. Therefore, poor attachment and a related failure to develop a sense of self may predispose a child to increased risk of psychopathology later in life (Wolfe, 1987). Physically maltreated children have been reported to be at risk for development of anxious attachments, such that they either avoid caretakers or are in constant conflict w i t h them (e.g., Crittenden, 1981; Crittenden & A i n s w o r t h , 1989; Erickson et al . , 1989; Egeland & Sroufe, 1981). Cicchetti (1989) emphasized that attachment relationships are important across the life-span. A n adult attachment interview has been developed by George, K a p l a n , and M a i n (1984) to examine attachment relationships in adul thood. It has been suggested that adults who have experienced physical maltreatment in ch i ldhood and developed an anxious-avoidant pattern of attachment evidence related anxiety and dismissiveness of relationships as adults (e.g., Cassidy & Kobak, 1988; Goldberg , 1991). These adults are described as d o w n p l a y i n g the importance of relationships, at times ideal izing their childhood experiences, and struggling to acknowledge the associated impact of their negative experiences. It is possible that adults who failed to develop a sense of trust i n the parent-chi ld relationship continue to struggle to establish interpersonal trust in adulthood. Individuals who fail to learn effective ways to deal w i t h feelings may experience basic mistrust of self and others, helplessness, and low self-worth i n adulthood. M a r t i n and Elmer (1992) attempted to investigate this hypothesis i n a fol low-up study of individuals severely physically maltreated as children. They assessed 19 adults who were physical ly maltreated as chi ldren, but fa i led to include a comparison group, which restricts the conclusions that can be drawn from the study. However , they d i d f ind a high percentage of the subjects were suspicious, felt that others were derogatory, had minimal job skills, and reported that their families-of-origin were controlling, conflict-ridden, and unsupportive. The range of variability in adaptation was extensive, h igh l ight ing the need to consider i n d i v i d u a l s ' perceptions and attributions in regards to their childhood experiences. Briere and Runtz (1988a) also examined the impact of phys i ca l and psychological maltreatment on adults' psychosocial adjustment. W h i l e findings indicated that physical and psychological maltreatment were typically present in the same families, certain types of maltreatment were found to be uniquely associated wi th later psychological difficulties. Paternal physical maltreatment was associated independently w i t h anxiety and somatization, whereas paternal psychological maltreatment was associated with anxiety, depression, interpersonal sensitivity, and dissoc ia t ion . These f indings indicate that different forms of c h i l d h o o d maltreatment may have different effects i n adulthood, as wel l as representing different pathways to a common post-abuse symptom complex. Other tasks thought to be central to development include the emergence, expression, mediation, and control of affective states, conscience development, and impulse control. Disturbances in development i n these domains might relate to angry , aggressive, impuls ive , and unempathetic responding, evidenced i n adolescence and adulthood as delinquent and criminal behavior. Starr Jr. and colleagues (Starr Jr. et al. , 1991) noted that later criminal behavior is one of the most commonly investigated sequelae of chi ld maltreatment, but that studies of this m a l t r e a t m e n t - v i o l e n c e r e l a t i o n s h i p g e n e r a l l y suffer f r o m n u m e r o u s methodological problems. They pointed out that studies often contain mixed samples of abused and neglected children, and that the strength of the relationship varies according to the source of information and definition of maltreatment used. W i d o m (1988) observed that it may be valuable to conceptually and empirically distinguish between forms of maltreatment in samples. She also emphasized the need to obtain val idat ion and substantiation of abuse and criminali ty whenever possible, and to use explicit criteria and multiple measures. These issues must be considered in the evaluation of the empirical findings. Lewis, Mal louh, and Webb (1989) stated that findings of the studies they have conducted suggest that a history of severe childhood physical abuse and witnessing of family violence is significantly associated wi th ongoing violent behaviors in adulthood. They also indicated that evidence suggests that adult criminality is associated w i t h a history of severe chi ld physical abuse. The investigators emphasized that most physically abused children do not become violent offenders, and physical abuse alone is not usually sufficient i n and of itself to cause delinquency. They pointed to the role of neuropsychiatrie vulnerabil it ies in chi ldren, domestic violence, and parental psychiatric illness i n the etiology of aggressive and violent behavior. Similarly, in a study conducted by M c C o r d (1983), parental alcoholism, crime, and aggressiveness were found to increase the probability that child maltreatment w o u l d be damaging to future development and increase the risk of criminality. M c C o r d used a large sample of 253 men and dist inguished between history of neglect, physical abuse, rejection, and adequate care. Whi le group similarities were obtained in regards to lack of affectionate, consistent care, group differences were also found, pointing to the value of distinguishing between forms of maltreatment. M c C o r d reported that 45% of the men who had a history of physical abuse or neglect were convicted for serious crimes, had alcohol problems a n d / o r mental illness, and were more l ikely than rejection and adequate care group members to have had aggressive, alcoholic, and criminal parents. W i d o m (1989) designed a well-constructed cohort study to obtain empirical evidence regarding whether violence begets violence. She inc luded a clear operational definition of abuse and neglect, used a prospective design, separated abuse from neglect, and had a large sample and a well-matched control group. F indings were suggestive that c h i l d maltreatment increases one's risk for delinquency, adult cr iminal behavior, and violent cr iminal behavior. Individuals who had experienced childhood physical abuse had the highest levels of arrest for violent criminal behavior, followed by victims of neglect. M e n had higher rates of del inquency, adult c r imina l i ty , and violent cr iminal behavior than women. Physically abused or neglected women were significantly more l ikely to have an adult arrest than control group women, but the difference between the women's groups for violent crimes was not significant. W i d o m suggested that it is likely that the long-term consequences of physical abuse and neglect for women are manifest in more subtle ways, such as depression and withdrawal . F inal ly , she emphasized the need to consider protective mediating variables, given that a large portion of maltreated children do not grow up to become violent. C h i l d h o o d sexual maltreatment and adult funct ioning. The relatively l imited literature on the adult outcomes of physical chi ld maltreatment indicates that physica l ly maltreated individuals are generally at risk for interpersonal problems characterized by distrust, anger, and aggressiveness. The literature on adult outcomes of childhood sexual abuse is far more extensive, but is l imited by methodological problems. While there is disagreement as to the exact nature of the short- and long-term sequelae of sexual abuse, disturbed interpersonal functioning is frequently cited as a long-term consequence (e.g., Alexander, 1992; Cole & Putman, 1992; Finkelhor, 1990; Gelinas, 1983; Sgroi, Porter, & Blick, 1985; Wolfe & Wolfe, 1988). Wolfe and Wolfe (1988) indicated that three factors appear to have an impact on adult funct ioning: factors related to the abuse itself, cognit ive/attr ibutional factors, and avai labi l i ty of social support. They suggested that adjustment diff icult ies appear to p r i m a r i l y cluster i n the areas of interpersonal and psychological adjustment, sexual satisfaction, sexual dysfunction and deviance, and liability for further victimization. More specifically, the noted long-term effects of chi ldhood sexual abuse include depression, self-destructive behavior, anxiety, feelings of isolation and stigma, poor self-esteem, difficulties trusting others, and substance abuse (Finkelhor, 1990). Finkelhor (1990) noted that methodological and empirical refinements and theoretical developments in the field of childhood sexual abuse have led beyond the study of symptoms to an effort to conceptualize the impact of child sexual abuse on i n d i v i d u a l s ' funct ioning. In keeping w i t h an organizat ional /developmental perspective, Finkelhor hypothesized that chi ldhood sexual abuse has a variety of effects on a number of different mechanisms that may traumatize children by distorting their cognitions and affective capacities. Cole and Putman (1992) also posited that the underlying theme behind the variety of adult difficulties that have been documented is the disruption in the intrapsychic processes of def ining, regulating and integrating aspects of self, and deviations in the ability to experience a sense of trust and confidence in relationships. Thus, difficulties of adults who were sexually abused as children (e.g., poor trust, impaired self-integrity, lack of insight, tendencies toward impulsiveness) may reflect cumulative impairments of self and social functioning, and negative influences in the transition into adulthood and adult roles. O r g a n i z a t i o n a l / d e v e l o p m e n t a l theorists suggest that the divers i ty of interpersonal, affective, and cognitive symptoms exhibi ted by sexual abuse survivors are mediated by the attachment experiences of survivors. Alexander (1992) focused on organizing themes, such as rejection (associated wi th avoidant attachment), role reversal (associated w i t h resistant attachment), a n d fear/unresolved trauma (associated with disorganized attachment). She stated that a disturbance i n attachment is l ikely to be associated w i t h diminished capacity to meet one's needs in appropriate ways, to monitor oneself or others, and to seek help to stop the abuse. Therefore, interpersonal relationships appear to be the area in which many of the long-term conflicts related to child sexual abuse are manifest (Alexander, 1992). A number of studies have focused on psychosocial functioning in adult survivors of ch i ldhood sexual abuse. Parker and Parker (1991) attempted to determine whether childhood sexual abuse was related to poor adult outcome in social functioning. They administered mult iple assessment devices to a large sample of 492 women university students, and found that 135 students reported a history of chi ldhood sexual abuse. Findings indicated that, in comparison to the non-abused women, the abuse group subjects reported more difficulties with social competence. However, the impaired social functioning was reported only when the childhood sexual abuse was associated with poor parental treatment. Harter, Alexander, and Neimeyer (1988) also focused on the social adjustment of adult survivors of childhood sexual abuse. They used a sample of 29 incest survivors and 56 matched controls and a mult imodal assessment technique. It was found that aspects of the abuse per se (onset, duration, nature of abuse) had no significant effect on social adjustment when family and cognitive characteristics were controlled. However, the sexual abuse survivors who had lower ratings of social adjustment perceived themselves as different from significant others, and reported less cohesion and adaptability in their families-of-origin. Al though this study relied on retrospective reports, which are subject to bias, the results suggest that an individual 's personal construction of the sexually abusive experiences and related family environment may be more predictive of negative adult outcome than objective environmental events. Another domain of functioning i n chi ldhood sexual abuse that has been empirically evaluated is that of psychosocial functioning. Bagley and Ramsay (1985) rel ied on a stratified random Canadian sample of 679 adults to examine the psychosocial outcomes of chi ldhood sexual abuse. They found that reports of current depression, suicidal ideas and behavior, and low socioecononnic status were al l more frequent i n the sample w h o reported chi ldhood sexual abuse. The investigators noted that sexual abuse often occurred together w i t h other aspects of family climate, such as parental separation, parental coldness, and lack of support. Given that the effects of childhood sexual abuse were not isolated from the effects of these family factors, the conclusions are l imited to the statement that chi ldhood sexual abuse is but one of a number of disruptive events influencing later mental health. Briere and Runtz (1988b) also examined psychological symptomatology associated wi th childhood sexual abuse. A non-clinical adult sample was used, which included 33 women with a history of childhood sexual abuse, and 191 non-abused controls. Women with a history of childhood sexual abuse reported higher levels of acute and chronic dissociation and somatization and more frequent symptoms of anxiety and depression, i n comparison to controls. Briere and Z a i d i (1989) looked at childhood sexual abuse history and psychosocial sequelae i n female psychiatric emergency room patients. When clinicians were directed to specifically ask patients if they had a history of childhood sexual abuse, 70% of the 50 subjects reported childhood sexual abuse. The results indicated that, i n comparison to non -abused patients, patients with histories of childhood sexual abuse were more likely to have a history of suicidal ideation and attempts, drug abuse, sexual problems, dysfunctional personality traits, and more psychiatric diagnoses in general. While the study relied on patient report, the investigators pointed out that most patients viewed childhood sexual abuse as shameful and stigmatizing, thereby decreasing the l ikelihood of over-reporting a history of childhood sexual abuse. If continuity between the short-term and long-term effects of childhood sexual abuse exists, then it w o u l d fol low that d isrupt ion i n sexual functioning w o u l d be observed into adulthood. While sexual abuse does not appear to be associated with any particular diagnostic category, a recurrent f inding is the existence of disturbed psychosexual functioning (e.g., Briere & Runtz , 1990; Browne & Finkelhor, 1986; Finkelhor & Browne, 1985; McCormack , Janus, & Burgess, 1986; Miselman, 1980; Scott & Stone, 1986). A diversity of specific sexual difficulties have been documented in female victims: feelings of detachment and constricted affect (Briere & Runtz, 1986, 1988a; Lindberg & Distad, 1985); difficulties being emotionally close to males, centering around lack of acceptance, mistrust, denial , and inability to feel the need for affection and dependency (McCormack et al . , 1986; Owens, 1984; Scott & Stone, 1986; Steele, 1986; Tsai & Wagner, 1978); confusion about sex (McCormack et a l , 1986); serious concerns and obsessions about sexuality (Scott & Stone, 1986); distress and guilt related to sexual over-st imulat ion and sexual pleasure (LaBarbera, M a r t i n , & Dozier, 1980) repetition compulsion (Browne & Finkelhor, 1986; Sgroi et al . , 1985; Tsai & Wagner, 1978); and reactions of fear and hostility towards men (Browne & Finkelhor, 1986). For example, Silbert and Pines (1981) found that, among their sample of 200 current and former prostitutes, 60% reported a history of sexual abuse. Of the abused sample, 70% said the sexual exploitation affected their entrance into prostitution, 73% said they were extremely frightened during their abuse, and 53% felt disgusted by sex. It may be that if a c h i l d is forced into a sexual ized relationship and independence is discouraged, she w i l l be more l ike ly to develop a passive dependency which may impair her ability to establish and maintain satisfying sexual relationships. LaBarbera (1984) observed such impairments i n students wi th a history of seductive father-daughter relations. Other problems inc luded poor affective expression and the tendency to attribute danger to male sexuality and female competition. In summary, the empirical literature regarding the consequences of child sexual maltreatment show that such maltreatment may have an impact on adult survivors' interpersonal and psychological adjustment and sexual functioning, and may profoundly distort cognitive and affective capacities. However , the need for research procedures that rule out competing explanations for sjnnptomatology has been emphasized by Briere (1992). H e suggested that relevant factors such as family climate should be included as independent variables or predictors so that one can test directly their interactions wi th childhood sexual abuse. For example, Wyatt and Newcomb (1990) found that the impact of childhood sexual abuse was mediated by important factors such as internal attributions, and that elements of coercion and proximity of abuse (i.e., intrafamilial) were also important variables related to adult outcome. Beitchman and associates (Beitchman et al . , 1992) also emphasized the need to account for potential mediating factors and to control for family variables, as they are thought to have a pivotal impact on individuals ' response to childhood sexual abuse and long-term outcome. Comparison of Physical and Sexual Maltreatment A n important mediating factor i n the adult outcome of chi ld maltreatment may be the form of maltreatment and family disrupt ion that is experienced. W i d o m (1988) stated that examining distinct types of maltreatment is both conceptually and empirically appropriate. For example, Briere and Runtz (1990) obtained significant unique relationships between reports of different forms of child maltreatment and adult outcome. While a number of investigators have examined the differential impact of physical chi ld maltreatment, neglect, a n d / o r psychological maltreatment, very few have i n c l u d e d a comparison group of sexual ch i ld maltreatment. Given that empirical findings have indicated that different forms of maltreatment may be more prevalent at different ages and stages of development, and that infants and children f rom different caretaking environments display different patterns of attachment and related behaviors, one might hypothesize that these differences w i l l be reflected i n differential adult adjustment i n survivors of sexual maltreatment versus physical maltreatment. One could hypothesize that individuals who have been sexually abused are characterized by a cluster of difficulties different than the cluster found i n victims of physical abuse. For example, Finkelhor (1985) suggested that critical differences exist between sexual and physical abuse which may influence treatment, but he focused primari ly on differences in demographic patterns and historical variables. A review of the relevant literature leads one to believe that there may be differences in the interpersonal consequences as w e l l . Finkelhor and Browne (1985) emphasized that researchers should examine the type of trauma experienced w h e n evaluating the effects of abuse. One could hypothesize that victims of sexual abuse wou ld be more l ike ly than vict ims of physical abuse to suffer f rom the effects of traumatic sexualization (i.e., sexual confusion/dysfunction), and to struggle wi th feelings of shame and fear, whereas victims of physical abuse might be seen to have more difficulty centering around anger management, aggression, and cynicism/negativity (e.g., Briere & Runtz , 1988a, 1988b; Browne & Finkelhor, 1986; Galambos & Dixon, 1984; Mart in & Elmer, 1992; M c C o r d , 1983; W i d o m , 1989). There is some recent research which examines the person by situation interaction in either sexual abuse or physical abuse but there is a need for more research w h i c h adequately distinguishes between adult survivors of both forms of abuse w i t h i n the same study. A s noted by Wolfe and McGee (1991) the unique, additive, and/or interactive effects of the various forms of child maltreatment (i.e., physical abuse, sexual abuse, exposure to family violence, psychological maltreatment, neglect) can be determined only by defining and measuring these contructs as distinct entities. If studies were to jointly examine survivors of physical ly and sexually abusive environments, specific and sensitive measures of both historical and outcome variables w o u l d be needed. Whereas numerous measures assess global factors which have been associated wi th child abuse (e.g., locus of control, self-esteem, ttributional style), there is a need for instruments assessing the specific impact of sexually and physically abusive environments (Briere, 1992; Finkelhor, 1986). It is reasonable to assume that there w i l l be variables which w i l l be specific to abuse in general, or to forms of abuse, that w i l l not be identified by global measures. Another shortcoming is that much of the l i terature on maladaptive intrapsychic and interpersonal functioning focuses on one type of deficit in isolation from other deficits. M i l l e r III, M o n t i , Zwick , and Stout (1980) suggested that the nature of skills deficits and other i n d i v i d u a l characteristics might interact with treatment variables, which highlights the need to examine interacting components in greater detail. Pilot Studv In Ught of these methodological concerns, a pi lot study was designed to examine the specific interpersonal problems reported by adult survivors from a variety of traumatic backgrounds, using multimodal assessment including an abuse-specific questionnaire (see Appendix A) . The pilot study was conducted in order to obtain some prel iminary information about which aspects of adult interpersonal funct ioning might best discriminate between survivors of different forms of chi ldhood adversity. A university population was used i n the pilot study for a number of reasons. G i v e n the exploratory nature of the study and the related newness of some of the measures, as well as the fact that access to community resources was not readily available, it was deemed acceptable to use a university population. Other researchers have also utilized university populations (e.g., Briere & Runtz, 1988a, 1988b; Finkelhor, 1980; Fromuth, 1986). The pilot study sought to explore the association of childhood adversity with adult interpersonal functioning through the use of questionnaires asking about interpersonal problems, symptomatic concerns (e.g., shyness, depression), social support , and coping strategies (i.e., strategies for coping with stress). Questionnaire packages were distr ibuted to psychology undergraduate students who agreed to participate. Descriptive infornnation about demographic variables was obtained, as wel l as information about maltreatment history, social support, and coping strategies. A number of measures were included in the questionnaire package to assess subjects' current level of functioning (i.e.. Beck Depression Inventory, Social Avoidance and Distress Scale, Interpersonal Adjective Scales, Inventory of Interpersonal Problems). Details of the pilot study are presented in Appendix A . The results of the pilot work indicated that individuals reporting a history of physical a n d / o r sexual chi ld abuse reported significantly greater interpersonal problems, less social support, and the more frequent use of maladaptive coping strategies i n comparison to individuals without an abusive history. The pilot data support the hypothesis that adult survivors of abuse continue to have problems in interpersonal relationships. The results also indicated that respondents with a history of physical abuse reported significantly greater interpersonal problems on a number of dimensions (e.g., se l f -wor th , over-sensi t ivi ty , dependency, heightened aggressiveness), significantly less social support, and more frequent use of maladaptive coping strategies (e.g., use of physical and verbal aggression), in comparison to non-abused subjects. Survivors of sexual abuse reported significantly less social support than controls. W h e n physically abused subjects were compared to sexually abused subjects, physical ly abused subjects reported significantly greater interpersonal problems on a number of dimensions (e.g., support ing others, over-sensitivity, dependency). The pilot data paint a clearer picture of physically abused individuals than of sexually abused individuals . It seems likely that this was due to the nature of the sexual abuse reported. The pilot sexual abuse sample is non-representative when compared to sexual abuse victims typically used i n clinical research, who report longer durations and higher frequencies of sexual abuse and more frequent family involvement. Another possibility is that survivors of sexual abuse are less w i l l i n g or less able to identify or report problems. Whi le the unrepresentativeness of the sexual abuse subjects in the pilot work does l imit the generalizability of the conclusions, the results are suggestive that dimensions of adult functioning, such as interpersonal relationships (e.g., self-worth, dependency, aggressiveness, distrust), coping strategies, and social support, appear to be salient i n the study of the differential adult outcomes of various forms of child maltreatment. The f indings of the pilot study were used to inform and design the main study i n a number of ways. Given that the pilot study d i d not produce clear results for sexually abused subjects, it was decided to include sexual abuse subjects i n the main study more similar to those more typically used in the research literature, and to attempt to ensure that members of both the physical and sexual abuse groups experienced abuse of similar objectively reported intensity and duration. The results of the pi lot study served to emphasize the importance of i n c l u d i n g mul t id imensional measures of adult funct ioning, particularly focusing on self-perceptions, situational discomfort, and coping strategies. Purpose of the Present Study The central study therefore examined interpersonal funct ioning i n adult women from a variety of traumatic backgrounds. I was pr imari ly interested in the ways in which these individuals processed and responded to social information, given that interpersonal and social functioning appears to be the area i n which many of the long-term consequences of child maltreatment are manifested. Subjects were presented wi th three audiotapes, two of which depicted a troublesome social situation, and one of w h i c h depicted a more neutral social situation. Fol lowing each tape, subjects were asked to imagine themselves in the situation and to answer questions about their response to that situation. Four factors were assessed: (1) subjects' perception of the speaker on each tape; (2) subjects' perception of themselves; (3) subjects' judgments regarding the strategy they w o u l d adopt to cope w i t h the situation; (4) subjects' self-reported and objectively assessed levels of physiological response during the situation. The central question was whether the traumatic interpersonal experiences of abuse survivors resulted in dysfunctional responses to social information, even when that information was presented to them i n a secure, neutral setting (i.e., in the clinic or lab). Perceptual and Attributional Processes A number of researchers and clinicians have suggested that individuals who were abused as children process social stimuli deficiently. For example. Dodge, Bates and Pettit (1990) found that relative to non-abused chi ldren, physical ly abused children displayed deficiencies in the ability to attend to and recall social cues and that the abused children were more l ikely to perceive others as having hostile intentions. Research regarding deficiencies in social information processing in adult surv ivors of abusive environments , p a r t i c u l a r l y p h y s i c a l l y abusive environments, is lacking. It is important to determine whether such deficiencies persist into adulthood. If so, biased information processing might be one mediator between childhood abuse and the dysfunctional interpersonal behavior reported by adult survivors of abuse and by the clinicians who treat them. In the present study, subjects' perceptions of themselves and others were examined to determine if adult survivors displayed negative cognitive biases and perceptual processes relative to those without histories of abuse. The examination of perceptual and attributional processes is theoretically consistent wi th the perspective of the organizat ional/developmental model that individuals develop internal w o r k i n g models, characterized by perceptions and expectations about self, others, and the w o r l d i n general . Researchers are increasingly recognizing the importance of cognitive and perceptual factors i n adequate social behavior. Cognitive processes, such as self-evaluation, attributions about social events, and negative expectations, have been shown to be important mediational components of interpersonal behavior (Buss, 1980; Franzoi & Brewer, 1984; Graziano, Feldesman, & Rahe, 1985). A number of researchers i n chi ld maltreatment have e m p h a s i z e d the importance of assessing i n d i v i d u a l s ' perceptions and beliefs. For example. Dodge and associates (Dodge et al . , 1984) observed that children's perceptions of peers' intentions, not the peers' actual intentions, determined the children's behavioral response to provocation. Skills deficits are often exacerbated by maladaptive perceptual processes (e.g., perceptions of powerlessness, of closeness seen as resulting in pain and vulnerability distorted resentments, and intrusive images), which may serve to negatively bias emotional and behavioral responding. G o l d (1986) explored the relation between chi ldhood sexual vict imization, adult attributional style, and adult interpersonal functioning. Results indicated that sexually vict imized women's adult functioning was related most strongly to their attributional style for bad events. Women who were sexually victimized in childhood and who reported psychological distress (i.e., depression, negative sexual symptoms, dissatisfaction wi th present sexual relations) and l o w self-esteem/social competence were more likely than controls to attribute bad events to internal, stable, global factors, to blame their character and behavior for bad events, and to attribute good events to external factors. Social cognition in maltreated children was examined by Smetana and Kel ly (1989). Chi ldren with histories of different forms of maltreatment were found to vary in moral judgements. The authors speculated that maltreated children's social behavior was related to their interpretation of aggressive and unfair situations. S imilar ly , Wolfe (1987) suggested that individuals ' personal construction of events seems to be more significant than whether the attribution/perceptions are accurate or not. Mueller and Silverman (1989) stated that individuals ' expectations, beliefs, and conceptions (internal working models) are the variables that mediate between a history of maltreatment and subsequent interpersonal behavior. They stressed the value of assessing individuals' concepts of themselves and others. In summary, the empirical literature points to the importance of cognitive and perceptual processes i n adaptive social behavior, and suggests that indiv iduals v^^ith interpersonal problems commonly display deficiencies i n these processes and in social skills. Interpersonal C o p i n g Behaviors Another factor of potential importance to effective interpersonal functioning is the strategy an individual selects for coping with difficult interpersonal situations. A large body of literature illustrates that coping mediates the relationship between stressful events and adaptive outcomes i n general (e.g., A l d w i n & Revenson, 1987; Coyne & Lazarus, 1980; Folkman, Lazarus, Dunkel-Schetter, Delongis, & Gruen, 1986; Kobasa, 1985). Relatively little research has focused on interpersonal coping in general, or on interpersonal coping in abused individuals i n particular. However, a number of researchers have suggested that abused i n d i v i d u a l s may select maladaptive strategies for coping w i t h stressful situations and that this aspect of social judgement may be one mediator of the childhood abuse-adult dysfunction re lat ionship (e.g., W i d o m , 1989). In addi t ion , results of the pi lot study (see Appendix A ) indicated that college students wi th histories of abuse were more likely to select maladaptive strategies for coping with stressful interpersonal events than students without such histories. Organizational/developmental theorists hypothesize that a stage-salient task is the development of autonomy and a sense of self-competency and mastery of the social and physical environment. If children's attempts to be independent and explore and learn from their surroundings are impeded, if they learn that their behavior has little impact on the environment, and if they are exposed to exploitive, aggressive, and/or passive models of problem-solving, it is less l ikely that they w i l l develop adaptive ways to cope with later stressful interpersonal situations. They may cope w i t h such stressful family environments through the use of avoidance and aggressiveness, a n d / o r hypervigi lence and dependency (Crittenden & Ainswor th , 1989). It may be that children's coping competencies and resources assume a central role in determining the adaptive outcome achieved, rather than the type of stress per se (Jaffe, Wi lson, & Wolfe, 1986). For example, Grych and Fincham (1990) speculated that coping behavior is an important mediator between marital conflict, family disruption, and children's adjustment. A number of investigators have obtained evidence of aggressive, withdrawn, résistent, vigilent, a n d / o r dependent interpersonal responses i n maltreated children and adults and suggest that victims of different forms of maltreatment respond differentially (e.g., Bousha & Twentyman, 1984; Briere & Runtz , 1988b, 1990; Crittenden, 1981; Erickson et al., 1989). Therefore, in this study, subjects were asked to describe (in an open-ended format) how they w o u l d cope w i t h the situation on the audiotape. A n open-ended format was selected to al low for maximal individual differences in coping to emerge. As a supplementary measure, a series of structured rating scales concerning coping strategies were also included. Physiological Processes Physiological response to the audiotaped s t imul i was also measured. The value of examining physiological activity in conjunction w i t h traumatic events has been underscored by recent research (e.g., Pennebaker, 1985; Pennebaker & KHhr Beall, 1986). Pennebaker's work revealed the consistently high arousal associated w i t h init ial disclosure of traumatic events. F r o m a broader perspective, the measuren\ent of physiological correlates of social activity is nov^ considered a valuable val idational tool and represents a new subspecialty i n social psychology (i.e., social psychophysiology) (Cacioppo & Tassinary, 1990). The g r o w i n g importance of psychophysiological methods i n clinical assessment is attributable to the complex relationships among phys ica l , cognit ive, behavioral , social , and emotional response systems. Psychophysiologists posit that social, behavioral, and emotional phenomena are often a function of, and reflected i n , physiological processes (Haynes, 1991). Furthermore, there is n o w convincing evidence that cardiovascular reactivity to challenging s t imul i can be reproduced reliably w i t h reliability indices frequently exceeding the critical .80 criterion (Kamarck et al., 1992). Whereas perceptual and coping measures seem more reflective of an indiv idual ' s cognit ive appraisal of the s i tuation (i.e., here is how I see the participants i n the situation, and here is what I w o u l d do to cope w i t h this situation), signs of autonomic response w o u l d be indicative of emotional processes which might serve to disrupt the indiv idual ' s social judgement and behavior. Cicchetti and Braunwald (1984) stated that arousal modulation and differentiation of affect (i.e., emergence, expression, mediation, and control of affective states) are some of the most salient developmental issues in infancy. The child's ability to inhibit inappropriate behavior related to strong positive and negative affect, to self-soothe the physiological arousal that strong affect produces, and to organize coordinated action are important aspects of emotional regulation (Gottman & Katz, 1989). It may be that emotional reactions at the time of abusive events favor development of conditioned emotional responses that may recur in the presence of eliciting stimuli (i.e., situations that are perceived to be dangerous, unpredictable, and uncontrollable) (e.g., Berliner & Wheeler, 1987; Wolfe , 1987). One might speculate that such eliciting stimuli are different for varying forms of maltreatment, such that child sexual abuse is tied to intimacy, unpredictability, and sexually related s t imul i , and that chi ld physical abuse is more closely related to aggressive and dangerous st imuli . Terr (1991) suggested that repetitive behaviors and physiologic re-enactments are features of childhood trauma that appear to last for years and are often seen in adults traumatized as children. Estimations of threat or challenge and related affective arousal are hypothesized to be important mediating variables of children's adjustment to stressful family climates (Grych & Fincham, 1990). If individuals are unable to regulate their emotional arousal, then the arousal may preclude or interfere with further cognitive processing. For example, in their case study of an abusive mother, Koverola , Elliot-Faust, and Wolfe (1984) found that the abusive mother's difficulties controlling her anger and her emotional reactivity to stressful circumstances impeded her ability to use adaptive parenting skil ls . Lahey and colleagues (Lahey et al. , 1984) hypothesized that parents in greater emotional and somatic distress may have a lower threshold for child misbehavior and may react more punitively to it. They found marked differences between physically abusive mothers and matched non-abusive controls on objective measures of parenting behavior and emotional-somatic arousal. Wolfe, Fairbank, Ke l ly , and Bradlyn (1983) speculated that abusive parents may demonstrate increased arousal to problem situations and have greater anticipatory anxiety and arousal to situations they were unable to deal w i t h in the past. U s i n g scenes of ch i ld behaviors, the investigators observed that, in comparison to controls, abusive mothers displayed greater physiological arousal dur ing stressful scenes versus non-stressful scenes. The investigators speculated that the abusive mothers may label the arousal as anger and therefore be less able to use coping strategies. For example, Vasta and Copitch (1981) found that adults faced with child behavior they experienced as frustrating were more l ikely to use higher-intensity responses than when they were not faced with such behavior. Similarly, Gottman and Katz (1989) hypothesized that emotional interactions wi th in families affect children's autonomic functioning and subsequent emotional development. Using objective measures of autonomic nervous system arousal, the investigators found that parenting characterized by coldness, unresponsiveness, and anger may relate to non-compliance and anger in chi ldren, in comparison to the behavior of children of more responsive, nurturing parents. Chi ldren of such cold, unresponsive parents were noted to display greater negative peer interactions, worse health, and h igh vagal tone (tonic functioning of the parasympathetic nervous system vagus nerve). One c o u l d hypothesize that abusive parents' increased arousal and anticipatory anxiety in difficult parenting situations might also be observable in other stressful situations. Given that a percentage of abusive parents have also experienced chi ld maltreatment themselves, it is possible that such elevated emotional arousal and anticipatory anxiety in stressful situations are long-term consequences of c h i l d maltreatment. The e x p l o r a t i o n of the behavioral consequences of arousal and phys io log ica l act ivat ion may p r o v i d e some understanding of mechanisms underlying children's reactions to stressful family events, which might impair adult functioning as wel l . While there is no consensus about the channel of expression - physiological , behavioral , or cognitive - that provides the best source of affective in format ion , there is agreement that muUimodal assessment is needed (e.g.. M a s h & Wolfe , 1991; Wolfe & Bourdeau, 1987). For example, there may be a physiological basis for the development of emotional regulation ability. It seems reasonable that interpersonal s t imul i , such as the audiotaped situations, might evoke signs of conditioned anxiety in individuals w h o were exposed to traumatic interpersonal situations. In the present study both subjective (i.e., self-report) and objective (i.e., psychophysiological) measures of autonomic response were collected. Examining autonomic activity in response to an audiotaped stimulus in a lab setting represents a stringent test of the existence of differential autonomic nervous system activity in abuse survivors as the subject is aware that there is no objective danger in the situation. However , patients with conditions such as phobia or panic disorder have been shown to display strong autonomic responses to "artificial" lab- or clinic-based threats, such as pictures of a feared object or hyperventilation challenge. This indicates that such assessment procedures can provide useful in format ion about psychological disorders , particularly those in which anxiety plays a central role. Laboratory Simulation A frequent issue raised in discussion of the methodological problems in child maltreatment research is the need for multi-source, mult i -method data collection (e.g., Briere, 1992; Mash, 1991; M a s h & Wolfe , 1991). It is also advantageous to examine child maltreatment in vary ing social contexts in order to broaden the knowledge base, particularly i n l ight of theoretical predictions and empirical findings of differential response across situations and contexts. For example, certain behaviors (e.g., peer distress, rejection) and interactional settings (lack of familiarity, l o w degree of structure, h igh level of demand/threat) seem to be particularly provocative for maltreated individuals (e.g.. M a i n & George, 1985; Muel ler & Silverman, 1989). Laboratory simulations of such provocative behaviors and interactional settings have been used i n studies of ch i ld maltreatment. A number of investigators have used vignettes (audiotaped, v ideotaped, wr i t ten a n d / o r simulated) of social situations to assess dimensions of maltreated individuals ' functioning (e.g.. Dodge et al., 1984; Herzberger & Tennen, 1985; Vasta & Copitch, 1981; Wolfe et al. , 1983). Therefore, audiotapes of varying social situations were used i n the central study to create s t imuli from w h i c h mul t ip le measures of interpersonal functioning could be obtained. The audiotapes also a l lowed for the assessment of adult dysfunction in different contexts. Sample Selection Four groups of adult women were included as subjects: (1) victims of sexual abuse, currently i n treatment wherein they were addressing their chi ldhood experiences; (2) victims of physical abuse currently in treatment wherein they were addressing their childhood experiences; (3) victims of severe family disruption and chaos currently in treatment wherein they were addressing their chi ldhood experiences; (4) control subjects without histories of abuse or family chaos and not i n treatment, who were matched on critical demographic and socio-economic factors. The subjects i n the three clinical groups had identif ied that they had unresolved feelings about ch i ldhood experiences of maltreatment w h i c h they perceived to be negatively influencing their adult relationships in some way. This recognition preceeded their referral to treatment. For example, a number of the subjects were in treatment groups that had been established as groups for survivors of c h i l d h o o d maltreatment. The preceeding l i terature r e v i e w highl ights methodologica l concerns regarding selection bias, inadequate or miss ing comparison groups, and lack of uniform definitions of abuse acts. A goal of this study was to use specific and operationally defined selection criteria and adequate control comparison groups. Criteria were selected to reflect types of abuse that have been shown to be most damaging; those involving betrayal of trust, father figures, and force or coersion (e.g., Briere & Runtz, 1988a; Browne & Finkelhor, 1986; F inkelhor , 1990; Gel inas , 1983). Individuals w i t h histories of chaotic family environments were inc luded as a comparison group w h o shared the features of family dysfunction seen in physically and sexually abusive families, but lacking the specific abusive features (e.g., Briere, 1992; Beitchman et al . , 1992; Emery, 1982; Enos & Handal , 1986; Owings West & Prinz, 1987). The non-clinical group was included to control for the effects of such factors as l iv ing conditions and stressful life events associated with the lower income status that often characterizes many abuse victims. M a n y studies of adult survivors of maltreatment have re l ied upon communi ty a n d / o r college samples. W h i l e these studies p r o v i d e valuable information about the impact of abuse in such populations, the findings may not be generalizable to survivors wi th more severe histories of maltreatment. It has been noted that survivors i n treatment are generally those who have experienced the most severe maltreatment/family chaos (Owings West &c Pr inz , 1987), and are also the most difficult to f ind and recruit (Egeland, 1991). A d u l t survivors of chi ld maltreatment who are in treatment wherein they are addressing their childhood experiences might show different patterns of responding i n comparison to maltreated adults in general. For example, Briere and R u n t z (1990) found differential adult symptomatology in college students associated w i t h three types of ch i ld maltreatment history (i.e., physical, sexual, and psychological maltreatment). H o w e v e r , the investigators noted that such unique re lat ionships between maltreatment history and adult symptomatology may not be found i n clinical samples and emphasized the need for replication with cHnical populations and also w i t h different measures of symptomatology. While focusing on circumscribed subgroups of maltreated individuals limits the generalizability of the findings, the examination of adult survivors who are in treatment wherein they are addressing their childhood experiences may provide specific information about i n d i v i d u a l s who have experienced more severe maltreatment/family chaos. Given that these are the individuals seen by clinicians, there is a need to refine our knowledge about this sub-group i n order to develop effective treatment to meet their needs. It might be argued that treatment interventions derived from the results of studies using college subjects (who may be less likely to be the recipients of these treatment interventions), rather than clinical subjects, are potentially uncomprehensive and, at worst, misguided. Questions Addressed This research addressed three specific questions regarding the nature of the response elicited by the audiotapes. The first question focused on chi ldhood experiences of maltreatment and the related disrupt ion of important domains of development (e.g., attachment, autonomy, emotional regulation and cognition, and peer relations) in a group of already identified dysfunctional individuals being seen in treatment focusing on resolution of their ch i ldhood experiences. The first question concerned the extent to which such c h i l d h o o d abusive experiences correlate with a pattern of adult interpersonal responding that is different than that correlated w i t h exposure to disruptive or chaotic family environments without abuse. Wolfe and M o s k (1983) presented data suggesting that child/adolescent victims of physical abuse d i d not significantly differ from children from chaotic homes in regards to dysfunctional behavior. F r o m an organizat ional/develop-mental perspective, this w o u l d imply that the issue of separating maltreatment from family dysfunction may be more a semantic than a real issue. This suggests that maltreatment and family dysfunction m a y be best conceptualized as the inabil i ty of caregivers to nurture offspring, thereby representing a pervasive, persistent pattern of interaction i n the home environment that fails to foster healthy child development in many ways (Erickson et al. , 1989). We w i l l call this the Family Chaos Position. Alternatively, researchers such as Steele and Alexander (1981) have argued for a two-factor conceptualization of dysfunction. Here, some of the problems of abuse victims are theorized to develop out of the disordered family relationship itself, whereas other problems are theorized to develop more specifically from abusive experiences. We w i l l call this the Abuse Position. This suggests that there is some specificity of outcome of the pathological development related to abusive family environments. This issue was addressed in the current study by comparing subjects from phys ica l ly and sexually abusive environments to subjects wi th histories of family chaos/disruption without physical or sexual abuse. Adopt ing the Family Chaos Position would lead one to expect no differences between the abused and family disrupt ion groups. Alternatively, adopting an Abuse Position w o u l d lead one to expect significant differences between the abused and the family disruption groups i n response to the audiotaped interpersonal situations. The second question concerned the specificity of the dysfunction related to different types of abusive experiences. Given that a person was abused, does the type of the abuse, physical or sexual, and related family environment influence the nature of that individual 's interpersonal difficulties? It is possible that experiences w i t h abusive family environments relate to different patterns of adult dysfunction than experiences w i t h chaotic family backgrounds, but that the extent or nature of that dysfunction does not depend on the exact nature of the abuse. It may be that exposure to abusive environments relates to the development of internal working models characterized by anger, hypervigilence, the belief that one's behavior has little impact on others, and feelings of unworthiness, irrespective of the exact nature of the abuse. We w i l l call this the Generalist Position. O n the other-hand, Steele and Alexander (1981) have suggested that sexual abuse is a particularly devastating experience and that sexual abuse victims might be expected to display greater dysfunction in general or to display specific types of problems or patterns of interpersonal responding (i.e., w i t h sexual or intimate relationships) relative to individuals with other types of traumatic backgrounds. It may be that sexual abuse and the associated family environments are most disruptive to the stage-salient tasks related to social competence and sexuality. We w i l l call this the Specificity Position. In this research, this question was addressed by conducting specific comparisons between the sexual abuse and physical abuse groups. The th ird question was whether women seeking treatment for abusive a n d / o r chaotic family backgrounds differed from appropriately selected non-clinical control subjects in terms of adult interpersonal functioning. The problem of inadequate control groups has been widely documented (Briere, 1992; Dodge et al., 1990; Finkelhor & Hotal ing, 1984; Henderson, 1983; M a s h & Wolfe , 1991). The current literature suggests that women without histories of abuse or family disruption display more adaptive functioning than women w i t h such histories. For example, both Dodge and associates (Dodge et al . , 1990) and W i d o m (1989) found evidence for greater dysfunction i n subjects with histories of abuse than in control subjects matched on demographic and socio-economic characteristics. However, the fact that these studies are among the very few to have included appropriate non-abused controls made it desirable to examine this issue in the present study as well . Therefore, a series of analyses were conducted to compare the three clinical groups (sexual abuse, physical abuse, and family disruption) to a non-clinical control group of women matched on relevant demographic and socio-economic characteristics. Hypotheses Three specific predictions guided the data analyses: 1) Consistent with the Abuse Posit ion, I predicted that women with histories of abusive family environments w o u l d evidence greater dysfunction in their responses to the conflictual and intimate interpersonal situations than would w o m e n w i t h histories of chaotic family backgrounds without abuse. Al though there are some data that question the necessity of distinguishing between forms of maltreatment and family dysfunction, much of the recent research highlights the value of discriminating between forms of maltreatment. Whi le there may not be group differences on all dimensions of development, there is theoretical and empir ica l support for the hypothesis that the demands of different stages of development may bring out greater problems in varying contexts in one group of maltreated individuals or another. Numerous researchers have documented the f inding that maltreated children display different, and potentially more pathological patterns of responding , than c h i l d r e n f rom high-r i sk , d i s r u p t i v e fami ly environments across developmental domains such as attachment, secure readiness to learn, cognition and emotional regulation, and peer interactions (e.g., Aber & A l l e n , 1987; Aber et al., 1989; Crittenden, 1981; Dodge et a l , 1990; Erickson et al., 1989; George & M a i n , 1979; Howes & Espinosa, 1985; Kazdin , Moser, Kolbus, & Bell, 1985; M a i n & George, 1985; Trickett et al . , 1991; Walker , Downey, & Bergman, 1989). A l t h o u g h these d e v e l o p m e n t a l tasks are stage-sal ient , o r g a n i z a t i o n a l / developmental theorists suggest that attachment relations and other stage-appropriate tasks remain critical to individuals ' continual adaptation across the life-span (Cicchetti, 1989; Starr Jr. et al., 1991). I hypothesized that the findings indicating that maltreated children display different, and potentially pathological, patterns of responding i n comparison to children from high-risk disruptive families, w i l l be rephcable i n adult survivors in treatment, such that the chi ldhood patterns of responding w i l l continue into adulthood. 2) Consistent with the Specificity Position, I predicted that women wi th histories of sexual abuse would evidence greater dysfunction i n their responses to the confl ictual and intimate situations than w o u l d w o m e n w i t h histories of physical abuse without sexual abuse. There have been relatively few well-controlled investigations where survivors of sexually and physical ly abusive environments are compared i n the same study. However, Steele and Alexander (1981) theoretically argued that sexual abuse is a uniquely disruptive experience, and that sexual abuse victims might be expected not only to display specific difficulty in sexual situations, but to display greater dysfunction i n general relative to individuals w h o have experienced other forms of maltreatment, such as physical maltreatment. One impact of sexual abuse on development may be the formation of internal work ing models wherein individuals feel trapped, exploited, abandoned, and discounted in interpersonal situations. Steele and Alexander (1981) hypothesized that such working models may lead to pervasive self-doubt, such that individuals are unable to trust their judgment or feelings in sexual and varying other situations. A number of studies have documented that survivors of childhood sexual maltreatment experience difficulties in sexual/ int imate situations (e.g., Briere & Zaid i , 1989; Wolfe et al . , 1989; Wyatt & N e w c o m b , 1990). Briere and Runtz (1990) demonstrated differential adult symptomatology associated wi th histories of sexual versus physica l maltreatment, such that a significant, unique relationship was found between sexual maltreatment and dysfunctional sexual behavior. However , as Steele and Alexander (1981) suggested, sexual maltreatment survivors may display greater dysfunction in general in comparison to physical maltreatment survivors, perhaps reflecting the impact of w o r k i n g models characterized by the belief that one w i l l be trapped, discounted, a n d / o r exploited i n interpersonal situations, and the fact that sexual maltreatment survivors often experience other types of abuse as wel l . Such experiences and work ing models, together w i t h self-doubt and passive/dependent behavior (e.g., Erickson et al . , 1989), could inhibit sexual maltreatment survivors from using adaptive coping strategies i n other difficult interpersonal situations, such as the conflictual situation simulated i n the central study. 3) Consistent with recent research, I predicted that members of the three clinical groups (sexual abuse, physical abuse, and family disruption) would evidence greater dysfunction in their responses to the conflictual and intimate situations than the non-clinical control group. Starr Jr. and colleagues (Starr Jr. et al. , 1991) point out that, although the connection between chi ldhood experiences of maltreatment and later developmental outcomes is the result of numerous, complex interactive factors, the available evidence does support the suggestion that there are significant adult sequelae of childhood maltreatment. It is more l ikely that, i n comparison to individuals from maltreating homes, individuals from lower socio-economic status but relatively stable, home environments wou ld be exposed to caregiving and other environmental factors that w o u l d promote integration among socio-emotional, cognitive, social-cognitive, and representational competencies. Individuals who have experienced maltreatment/ family chaos have consistently been noted to be less adaptive across a number of domains of development, i n comparison to indiv iduals without such family backgrounds (e.g., Kaufman & Cicchetti , 1989; Howes & Eldredge, 1985; Inderbitzen-Pisaruk et al. , 1992; Lyons-Ruth et al. , 1989; Parker & Parker, 1991; Pianta, Egeland, & Hyatt, 1986). A summary of the theoretical predictions is presented i n Table 1. A d d i t i o n a l Questions A supplementary question concerns whether recollections of chi ldhood coping strategies and outcomes differ for women w i t h different chi ldhood experiences. One might speculate that different coping strategies are more l ikely to be used in response to different childhood experiences, and/or meet w i t h greater perceived success. In their conceptual discussion of resilience in child maltreatment victims, Mrazek and Mrazek (1987) hypothesized that coping responses such as information-seeking, risk-taking, cognitive restructuring of painful experiences, and dissociation of affect may be personal characteristics/skills that foster resilience. However , they cautioned that strategies may become maladaptive if over-used a n d / o r if not given up when the stressor no longer exists. It would be interesting to determine if Table 1 Summary of Theoretical Predictions 1. W o m e n wi th histories of abuse S A / P A vs 2. Women w i t h histories of family disruption only F D Abuse Position S A / P A > F D re evidence of dysfunction Family Chaos Position S A / P A = F D re evidence of dysfunction 3. W o m e n with histories of sexual abuse S A vs Women w i t h histories of physical abuse P A Specificity Position SA > P A re evidence of dysfunction Generalist Posit ion SA = P A re evidence of dysfunction 5. 6. Women wi th histories of abuse and family disruption S A / P A / F D vs_ Women with no history of abuse or family disruption C Outcome 1 S A / P A / F D > C re evidence of dysfunction Outcome 2 S A / P A / F D = C re evidence of dysfunction childhood coping strategies used by maltreated children demonstrate any continuity w i t h adult coping strategies. Given that it was not possible to use a prospective design or to obtain independent reports of subjects' chi ldhood coping, the central study relied upon adult survivors' recollections of their childhood coping. The potential value of using adult recall of c h i l d h o o d experiences is highlighted by Beitchman and colleagues (Beitchman et al . , 1992). They proposed that because an adult is able to assess c h i l d h o o d events f rom a different psychological perspective than a chi ld , understanding adults' perspectives of chi ldhood experiences is necessary to unravel the f u l l impact of maltreatment. Addi t iona l ly , the role of cognition and attribution as mediators of maltreatment outcome has been discussed above. Wolfe (1987) noted that individuals ' personal constructions of events seem to be of greater significance than whether the cognition or perspective is accurate. Similarly, Mar t in and Elmer (1992) emphasized the value of examining the ways in which adults v iew their chi ldhood abuse in efforts to understand adult adaptation. Some recent studies have relied upon adult survivors ' recollections of chi ldhood experiences (e.g., Briere and Z a i d i , 1989; Egeland et al., 1988). The Family Chaos and Abuse Positions are applicable to coping strategies as wel l . Proponents of a Family Chaos Position might speculate that adult survivors of abusive experiences w i l l not significantly differ from survivors of chaotic homes in regards to recollections of childhood coping strategies and outcomes. Alternatively, adoption of an Abuse Position w o u l d lead to the prediction that adult survivors of abusive experiences w i l l recall greater chi ldhood use of less adaptive coping strategies, wi th less successful outcomes, when compared to survivors of chaotic homes alone. In accordance w i t h A l d w i n and Revenson's (1987) theory that individuals whose mental health status is poor use different and less effective coping strategies than do individuals w i t h better mental health, therapist ratings of adjustment also were inc luded to ensure that any obtained differences were not due to the confounding effects of mental health status. A number of other potential confounds were examined, such as length of time in treatment, age, marital status, education, occupation, and ethnicity. M e t h o d Subjects Four groups of women were included in the study: Sexual abuse (n=30). The sexual abuse group was comprised of women in treatment wherein they were addressing their chi ldhood maltreatment, who had experienced active sexual abuse during childhood. Sexual abuse was defined as self-report of a l l of the f o l l o w i n g characteristics: (a) sexual body contact (i.e., genital/breast fondl ing , attempted/ completed v a g i n a l / a n a l / o r a l penetration) between any child or adolescent under 18 years of age and a person who was five years older than the victim; (b) use of physical force, threat, or coercion on at least some percentage of occasions; (c) where the offender was someone w i t h i n the victim's family, community or social circle (i.e., abuse characterized by betrayal of trust); (d) where the abuse was of six or more months duration. If the age difference between the vict im and the perpetrator was less than five years, the subject was still included i n the study when the sexual abuse consisted of sexual body contact (as described above) that the respondent d i d not desire a n d / o r involved coercion and manipulat ion. Women were excluded from participation if they had i n d i v i d u a l characteristics (i.e., drug/medicat ion bias; current psychotic a n d / o r depressive, manic-depressive episodes; i l l i teracy/lack of familiarity wi th EngHsh) that w o u l d interfere with their ability to participate. Addit ional ly , women whose only report of sexual abuse was being kissed by a peer or an adult were exluded from this group. Physical abuse (n=30). The physical abuse group was comprised of women in treatment wherein they were addressing their chi ldhood maltreatment, who had experienced active physical abuse during chi ldhood. Physical abuse was defined as self-report of al l of the fol lowing characteristics; (a) physical assault (i.e., being hit with a hand or an object, burned, scalded, bitten, kicked, shaken, choked, slapped, dropped or thrown); (b) characterized by excessive force; (c) resulting in a m i n i m u m of soft tissue damage on at least some percentage of occasions; (d) where the offender was someone wi th in the victims' family, community or social circle; (e) where the abuse was of six or more months duration. Once again, women were excluded from participation if they had individual characteristics as listed under Sexual Abuse that w o u l d interfere w i t h their ability to participate. Addi t ional ly , women whose only report of physical abuse was being spanked dur ing chi ldhood and women who reported experiencing sexual abuse were excluded from this group. Fami ly disrupt ion (n=30). The family disruption group was comprised of women i n treatment wherein they were addressing their chi ldhood experiences, who had experienced chronic family disrupt ion, instability a n d / o r chaos dur ing childhood. For this study, family disruption was defined as self-report of al l of the fol lowing characteristics: (a) high levels of family stress; (b) discord and disruption (e.g., h i g h levels of verbal aggression and host i l i ty between parents; divorce/custody/access disputes; a l c o h o l / d r u g abuse; separation from parent(s); chronic poverty); (c) lack of parental sensitivity of the impact of the disruption on the children, poor communication, and lack of support; (d) where the abuse was of six or more months duration; (e) where there was no reported history of physical a n d / o r sexual abuse. Women were excluded from participation if they had the indiv idual exclusionary characteristics noted under Sexual and Physical Abuse that w ould interfere with their ability to participate. Controls (n=30). The control group was comprised of women from intact families, with no reported family history of: (a) physical a n d / o r sexual abuse; (b) a l c o h o l / d r u g abuse; (c) d i scord and d i s r u p t i o n ; (d) parental psychiatr ic hospitalizations; (e) no personal history of psychiatric treatment. As with the other three groups, w o m e n were excluded from par t i c ipa t ion where i n d i v i d u a l characteristics such as drug/medicat ion bias and i l l i teracy/ lack of familarity with English would interfere with their ability to participate. The groups were compared on various demographic variables (age, marital status, education and occupation, and ethnicity). Descriptive statistics are presented in Table 2. The results of independent samples chi-square analyses and analysis of variance indicated that the groups were not significantly different i n terms of: marital status, X ^ d S , N=120)=22.80, ;p=.10; occupation, X ^ d S , N=119)=22.69, ^=.20; ethnicity, X^ (6, N=93) =5.09, _p=.53; or age, F(3,151)=l.ll, ;p=.35. Analysis of variance indicated that the groups d i d significantly differ i n terms of years of education, F(3,108)=6.83, p.=<.001. Therefore, the years of education variable was initially covaried out of all analyses. However, the covariate d i d not remove a significant degree of variance i n the data. Years of education was therefore not covaried out of subsequent analyses. When the results of the analyses conducted without years of education as a covariate were compared to those conducted with the covariate, no differences were noted at the .01 level. Us ing the .05 level , only three of thirty analyses differed. The three clinical groups were also compared wi th respect to Table 2 Demographic Variable Scores of Experimental and Control Groups Group Demographic Variable Sexual Abuse M S.D. Physical Abuse M S.D. Family Disruption M S.D. Controls M S.D. Age ^ Education ^ 32.03 11.23 (1.91) 33.73 13.50 (3.18) 32.70 12.85 (2.73) 30.63 14.36 (2.09) Marital Status single divorced widowed marr ied longterm relationship separated Occupation welfare c ler ical /manual admin/technical management professional student homemaker N u m b e r 6 2 1 14 1 6 5 7 2 1 1 2 11 Number 2 3 1 9 5 10 5 3 3 1 7 3 8 Number 7 4 0 11 3 5 4 5 2 1 5 4 9 Number 10 1 0 16 2 1 1 14 2 0 7 2 4 Ethnicity: Only eight of 120 subjects were not Caucasian N o r t h American (l=Native; 7=European) ^Age of the subject is stated in years. ° Education refers to subject's years of formal education. current adjustment and treatment exposure. As noted previously , the clinical subjects v^^ ere al l in treatment which was focused on discussion and resolution of their chi ldood experiences w i t h i n their families-of-origin. The cl inical subjects' therapists completed a brief rating of the subjects' adjustment i n intimate and conflictual interpersonal interactions and their treatment exposure (see Table 3). Each item was rated on a 7-point Likert-scale reflecting subjects' ability to respond adaptively (1= not at al l ; 7=consistently). Items were selected to reflect aspects of adjustment noted consistently i n the abuse literature. It is important to note that the groups d i d not significantly differ in terms of therapist-rated adjustment, F(10,148)=1.47, £= .15. Group means were quite similar w i t h low variability, and reflected relatively low levels of adjustment (i.e., means ranging between 3.18 and 5.00). The clinical groups also d i d not differ in terms of treatment exposure, F(4,152)=1.28, £=.28, or treatment improvement, F (2,73) =.53, £= .59 . Recruitment and Screening Women in the three experimental groups (labelled sexual abuse, physical abuse, family disruption) were referred to the research project w i t h their prior consent by therapists from social service agencies, outpatient chi ld abuse divisions of hospitals, and private practices in the Lower Fraser Val ley, British Columbia and in Calgary, Alberta. Subjects were compared across cities on various demographic variables (age, marital status, education, ethnicity). The results of independent samples chi-square analyses indicated that the subjects from the two cities were not significantly different in terms of marital status, X2 (5, N=116)=6.85, p=.23; and Table 3 Mean Therapist-rated Adjustment Scores: Five Categories Reflecting Current A d j u s t m e n t Group A d j u s t m e n t Sexual Physical Family Category Abuse Abuse Disruption M a S.D. M a S.D. M a S.D. Adjustment 1 5.00 (1.33) 4.82 (1.50) 4.96 (1.40) (day-to-day demands) Adjustment 2 4.64 (1.39) 4.52 (1.50) 4.80 (1.38) (general interpersonal coping) Adjustment 3 3.93 (1.46) 3.78 (1.19) 4.48 (1.42) (establishing trust) Adjustment 4 3.61 (1.52) 3.52 (1.50) 4.08 (1.22) (coping in conflictual situations) Adjustment 5 3.18 (1.66) 3.70 (1.61) 4.12 (1.36) (coping i n romantic situations) aMaximum score = 7. Poor adjustment is reflected in low scores, while good adjustment is reflected in high scores. ethnicity, X2(2, N=90)=1.36, £=.51. Analysis of variance and T-test analysis indicated that the subjects from the two cities d id not differ in terms of years of education, F ( l , 108)=2.63, E = - l l , or age, t(114) =.02,2=-99. The w o m e n were referred to the research project by therapists and coordinators at the agencies in which the women were part of the current caseload. The therapists worked i n a number of mental-health, community-based, outpatient facilities. Program directors were contacted, fol lowed by an explanatory meeting between the researcher and agency staff. It was emphasized that therapists were only to refer women who had identified, at the time of treatment referral, that they had experienced childhood maltreatment and whose central reason for referral was a desire to address unresolved feelings about their childhood maltreatment. The therapists were asked to review their caseloads for potential subjects. They then (a) decided if the women met the criteria for inclusion; and (b) could safely participate in the study. The therapists were instructed to exclude from participation women whose i n d i v i d u a l characteristics (i.e., drug/medica t ion bias, current psychotic, a n d / o r depressive/manic episodes, and ilhteracy/lack of familiarity w i t h English) w o u l d interfere with their ability to participate. The therapist w o u l d then describe the study to the patient and al low her to decide whether to participate; a l l participation was voluntary. A n attempt was made to ensure that each referral source referred women from each of the three clinical groups. In general, this pattern of referral was achieved. Exceptions occurred in cases where agency staff were only able to refer a small number (e.g., n=3) of subjects. It was not possible to obtain specific diagnostic information, as the therapists were w o r k i n g i n programs that d i d not util ize psychiatric diagnostic systems. However , it is unl ikely that women who were referred to the current study were i n treatment because of psychiatric conditions and only later disclosed childhood maltreatment, given the nature of the referral sources (i.e., non-psychiatric) and subject criteria. Addi t iona l ly , therapists were clearly directed to only refer women who had identified unresolved feelings about their childhood maltreatment as a central reason for seeking services. The patient was informed that researchers from the Univers i ty of British Columbia , Department of Psychology, were conducting a study focusing on social judgments, looking at how people respond to social situations. They were told that researchers were interested in meeting with women who came from a wide range of family backgrounds in order to learn ways in which childhood experiences might be related to adult response to social situations. Patients were informed that they and their therapist w o u l d be asked to complete questionnaires that dealt wi th relatively specific aspects of their childhood experiences within their families-of-origin, both in regards to actual traumatic experiences and ways that they remembered coping w i t h such experiences. It was emphasized that the informat ion w o u l d be completely anonymous and that their identity w o u l d not appear on any of the materials. The patients were told that they w o u l d be asked to meet w i t h a female graduate student for approximately 40 to 50 minutes at the agency site or at their homes i n order for them to l isten to audiotaped social interactions. It was emphasized that the researchers were extremely flexible in regards to scheduling. and w o u l d be very open to attempting to meet the client's needs in regards to time and location. The audiotaped social interactions were described as two-minute recordings, during w h i c h the patient would hear a man talking to them as if they were engaged in a conversation. They were told they w o u l d be asked to listen to each tape, fo l lowed by the completion of a few brief forms asking about their response to the audiotaped conversation. The physiological recording was described as the patient being asked to wear a blood-pressure cuff, such as used by doctors, that w o u l d take measures of their heart rate and blood pressure while they were resting and while they were listening to the tapes. They were reassured that no risks were reasonably to be expected f rom l istening to the tapes or complet ing the questiormaires, and that they were free to ask any questions they might have and to refuse to participate or terminate participation at any time. Therapists reassured patients that agreeing or refusing to participate wou ld have no impact on the patient i n any way. Finally, they were told that they would be pa id $5.00 for participation, whether they completed the procedures or not. The therapist completed an extensive questionnaire for each client. The questionnaire addressed the client's abusive/disrupt ive experiences as a chi ld , focusing on sexual and physical abuse and family disrupdon. Items were rated on a 7-point Likert - type scale, ranging from "never/not at a l l " (1) to "extreme/ consistently" (7). Other items were responded to i n an open-ended format and responses were later categorized (see Appendix B). The questions were specific and detailed in content and addressed such issues as nature and severity of abuse, offender characteristics, subject perceptions, and contact wi th mental health agencies. The questionnaires were used in order to ensure that subjects had experienced significant abuse of the type that has been shown to be most damaging as referred to i n the Introduction. This screening method proved to be successful, such that overall, subjects i n each group reported extensive disruptive experiences. Whi le there was some with in-group variability, the distributions tended to be negatively-sloped towards greater degrees of severity. Some degree of overlap was noted between the groups. W h i l e only the sexual abuse group was characterized by sexually abusive experiences, 60% of the women in the sexual abuse group reported some physically abusive experiences and 83% also reported some degree of family disruption. A d d i t i o n a l l y , 97% of women i n the physical abuse group reported some degree of fami ly disrupt ion. Such overlap is not surprising given the interrelatedness of different forms of abuse. It is unreasonable to expect that women from severely abusive homes w o u l d not experience a significant degree of family discord and disrupt ion. The differences between the three clinical groups on report of family disrupt ion were significant, X2(2, N=60)=7.50, £=.02. The f inding that the sexual abuse subjects reported family disruption less frequently may reflect the fact that some sexual abuse subjects reported being abused by people outside of their immediate family. Women in the non-clinical control group were recruited through newspaper advertisements and posted notices (see Appendix C). A brief telephone screening interview was administered to determine whether the applicant met the control group criteria. Controls were interviewed to ensure they had no reported history of family disruption or abuse. Control subjects also were informed that they w o u l d be paid five dollars for participation, that participation was voluntary, and that they could withdraw at any time with no consequences. Audiotape Stimulus Materials Three two minute audiotapes were prepared for presentation to the subjects on a stereocassette recorder through headphones. Each subject heard three audio-tapes of a simulated conversation (see Appendix D). One audiotape consisted of an adult male voice engaged in a dating interaction, involving implied physical contact (e.g., arm around shoulder, touching hair) and expression of verbal affection. The second audiotape consisted of an adult male voice engaged in a conflictual interaction where the male was expressing critical and irritated affect. The final audiotape consisted of an adult male voice engaged in a neutral interaction, where the male described shoe stores. Audiotape content was developed to be reflective of situations that w o u l d be theoretically salient for survivors of child matreatment. As wi th the development of the stressors in the Strange Situation, the three audiotaped interactions were chosen because they were thought to be relatively non-traumatic, yet provide graduations of stress that w o u l d ensure the activation of social cognit ion, physiological, and behavior systems of most adults (Carlson et al. , 1989). The male speaker was played by a male actor who was experienced i n theatrical improvisation. A male speaker was chosen rather than a female speaker as the abuse literature suggests that male offenders are more prevalent for both male and female victims. While it could be argued that female perpetrators are common i n cases of fami ly disruption and physical abuse, statistical constraints made it unfeasible to include both male and female speakers. Therefore, a male speaker was chosen to better reflect the experiences of the majority of subjects. Finally, order of presentation was randomized. Order and sequence effects were examined. Audiotape val idat ion . The audiotapes and physiological recording device used i n the central study were piloted using female university graduate students to establish validity and stability. A 15 minute baseline (i.e., the mean of four readings taken at five minute intervals) was established for each of 10 subjects and recording stability was obtained. The subjects were presented w i t h four two-minute audiotapes in randomized order (readings were taken at one and two minutes and a mean score was obtained.) The audiotapes consisted of the three audiotapes described above and a fourth audiotape consisting of an adult male voice engaged i n a conflictual interaction where the male was again expressing angry and irritated affect. A five-minute baseline was obtained between each audiotape (i.e., the mean of readings taken at two and five minutes). Subjects unanimously agreed that one of the two conflictual audiotapes was more reflective of angry and irritated affect; this audiotape was therefore used in all subsequent analyses. It was hypothesized that these pilot subjects (i.e., subjects involved i n audiotape val idation) w o u l d show greater change i n physiological indices in response to the confUctual and dating audiotapes i n comparison to the neutral audiotape. Each experimental audiotape was compared to the neutral audiotape on the three physiological indices. Beginning wi th the dating audiotape, the results of a series of i n d i v i d u a l t-tests indicated that subjects' response to the neutral and dating audiotapes were significantly different (with greater changes associated with the dating audiotape) in terms of mean physiological change in heart rate, t(18)=2.68, £=.05, systolic blood pressure, t(18)=3.43, £= .01 , but not i n diastolic blood pressure, t(18)=2.02, £=.10. Similarly, for the conflictual audiotape, subjects' responses to the neutral and conflictual audiotape were significantly different (with greater changes associated with the conflictual audiotape) i n terms of mean physiological change in : heart rate, t(18)=2.88, £= .01 , systolic blood pressure, t(18)=3.78, £ = . 0 1 , but once again, not in diastoUc blood pressure, t(18)=.94, £>.05. These results support the validity of audiotape content, in that response to the two experimental audiotapes was significantly different when compared to response to the neutral control audiotape for these pilot subjects. The observation that diastolic blood pressure d i d not differ across audiotapes may reflect the fact that diastolic blood pressure does not vary as much as systolic blood pressure (Obrist, 1981). Therefore, one might expect that measures of systolic blood pressure would be more sensitive than measures of diastolic b lood pressure to variations in physiological activity i n response to stimuli such as the audiotapes. Dependent Measures Structured self-report. A number of self-report instruments were used to obtain information across classes of variables. The value of assessing maltreated i n d i v i d u a l s ' social cognitions and attributions was addressed earlier. The specific aspects of social perceptions that have been hypothesized to be important to measure i n maltreated individuals include the social and emotional aspects of self-concept and adequacy, efficacy, estimation of threat or challenge, and conceptions of the basic nature of people as good, bad, or neutral (e.g., Grusec & Walters, 1991; Grych & Fincham, 1990; Milner , 1991; Wolfe & Bourdeau, 1987). It has also been suggested that researchers need to measure cognitions/perceptions in response to specific behaviors/situations rather than measuring general cognitions (Grusec & Walters, 1991). Subjects completed a questionnaire regarding their social perception of the male participant (e.g., "good-bad" 'strong-v\^eak") (see Appendix E), and a question-naire regarding their self-perceptions from the perspective of imagining themselves to be actively involved i n the audiotaped conversations (see A p p e n d i x F). Each questionnaire was comprised of 3 key items measuring global perceptions held by people during social interactions. The items were bipolar-anchored adjective scales, us ing a 7-point Likert scale. These items were chosen from the Semantic Differential (Osgood, Suci, & Tannenbaum, 1957) to represent the three major factors identif ied by Osgood: evaluation, potency, and activity. The Semantic Differential represents a standardized and quantified procedure for measuring the connotations of any given concept for the indiv idual and is a widely used research tool. The Semantic Differential has been noted to possess good item reliability (test-retest r=.85), and moderate factor score reliability (test-retest r=.52 to .79 for females) (Norman, 1969; Osgood et al., 1957). M i l n e r (1991) also has noted that adjective checklists l ike the Semantic Differential have adequate psychometric properties. A d d i t i o n a l l y , adjective checkUsts have been used with both non-pathological and pathological populations, supporting the use with abusive populations. Subjects were asked to describe how they w o u l d cope with the situation using an open-ended format (see Appendix G). Their responses were coded and sorted into 15 coping style scales by coders trained in the use of the scales and b l ind to subject group. These categories were based on content analyses of the data. The scoring procedure is described in detail i n the results section. Milner (1991) has emphasized the value of using two measures of the same construct to help increase the reliability and val idity of the results. Therefore, as a second coping measure, subjects also completed a 6-item self-report measure of their use of specific coping strategies (e.g., "emotionally detach", "get angry and fight back"). Each item was rated on a 7-point Likert scale (l=does not apply a n d / o r not used; 7=used a great deal) (see Appendix H) . Items were chosen to reflect coping styles noted in the pilot study and in the abuse Hterature. A number of researchers in the area of child maltreatment have theoretically a n d / o r empir i ca l ly addressed the value of measuring coping behaviors (e.g.. Dodge et al. , 1990; Grych & Fincham, 1990; M a s h , 1991; M i l n e r , 1991; Wolfe & Bourdeau, 1987). For example. Dodge and colleagues (Dodge et a l . , 1984) relied upon verbal descriptions of behavioral response (e.g., aggression, prosocial). Briere and Runtz (1990) also asked subjects to report on their behavioral response in intimate and conflictual situations. The open-ended format was chosen to al low subjects freedom of response so as to avoid the problem of forcing subjects to fit their response into categories/scales that may not capture their experience. Fo l lowing each audiotape, subjects completed a questionnaire requiring that they rate their level of arousal and physiological reactions to the audiotape on the Body Sensations Questionnaire (BSQ; Chambless, Caputo, Bright, & Gallagher, 1984). This 17-item scale is comprised of items concerning the experience of sensations associated wi th autonomic arousal (e.g., "Feeling short of breath", "Nausea", "Sweating"). Each item is rated on a 7-point scale, ranging from "not at a l l " (1) to "extreme" (7). A total score was derived by summing the i n d i v i d u a l item ratings. The BSQ has been noted to be internally consistent (Cronbach alpha=.87) and moderately reliable (test-retest r=.67). The instrument has been found to fare well on tests of discr iminant and construct val idi ty . (Chambless et al . , 1984) (see A p p e n d i x I). A s i n the measurement of coping behaviors , self-report of physiological arousal was included as a second measure of the same construct (i.e., psychophysiological response). The measurement of subjects' perceptions of their physiological reactions is consistent with organizational/ developmental theory, in that it may be reflective of internal working models of the self as reactive, aroused, and out-of-control i n difficult interpersonal situations. Physiological recording. Subjects' physiological responses (i.e., systolic and diastolic blood pressure, heart rate) were measured us ing an electronic blood pressure monitor w i t h pressure cuff, automatic electrical p u m p , a microprocessor, and digital displays (Model Dinamap 845, Cr i t ikon Corp . , Tampa, Florida). This device was chosen because it was easy to operate and non-invasive, could be carried to different test sites, and could provide ongoing moni tor ing of cardiovascular functions. Linden and Zimmerman (1984) reported generally high intercorrelations between such portable devices and the standard ausculatory method of blood pressure determinat ion (r=.95 (systolic); r=.92 (diastolic); r=.99 (heart rate)). Addi t iona l ly , Manuck and Schaefer (1978) noted that differences i n magnitude of indiv idual responses for physiological responses showed substantial consistency, or reproducibility over a week-long period (r=.68 (systolic); r=.46 (diastolic); r=.69; (heart rate)). Koverola and associates (Koverola et al., 1984) and M i l n e r (Milner, 1991) have emphasized that self-reports of anger and arousal are strengthened when they are accompanied by addit ional objective criteria. A number of researchers have included cardiovascular measures in their studies of maltreatment. For example, Donovan and Leavitt (1989) used measures of heart rate acceleration in their study of maternal patterns of physiological arousal and infant attachment quality. Similarly , Wolfe and associates (Wolfe et al . , 1983) used physiological measures such as heart rate and respirat ion rate in their s tudy of abusive parents' physiological responses to stressful and non-stressful chi ldhood behavior. Gottman and Katz (1989) assessed autonomic nervous system response (e.g., cardiac interbeat interval, pulse transmission time) in both children and adults in their investigation of marital d iscord/ fami ly affect and emotional regulation. They emphasized that the exploration of the behavioral consequences of physiological activation may provide some understanding of the mechanisms underlying individuals ' reactions to stressful situations. Gottman and Katz (1989) noted, however, that the application of physiological techniques to the understanding of interpersonal interactions has largely been u n e x p l o r e d , despite the fact that the a b i l i t y to regulate emot ional experiences/expression has often been described by organizational/developmental theorists as an important developmental milestone in young children. Therefore, the indices of physiological response included in the central study were chosen to explore subjects' ability to regulate emotional experiences. Cardiovascular indices were chosen as they have been used by other researchers in chi ld maltreatment research and have also been shown to be reliable (Kamarck et al., 1992). Addi t ional Measures Subjects were given the Ways of C o p i n g Checkl is t (Revised) ( W C C L - R ; Lazarus & Folkman, 1984) (see Appendix J) and asked to recall childhood coping strategies. The W C C L - R is a 67-item self-report measure of the broad range of cognitive and behavioral strategies people use to manage stressful demands (e.g., "I went over in my m i n d what I w o u l d say or do", "tried to forget the whole thing"). Each item is rated on a 4-point Likert scale (O=does not apply a n d / o r not used; 3=used a great deal). Scores are calculated by summing the ratings across eight scales obtained by factor analysis. The items and factor loadings used were those of A l d w i n and Revenson (1987), and inc lude three problem-focused scales (cautiousness, negotiation, instrumental action), four emotionally-focused scales (escapism, seeking meaning, minimization, self-blame), and one social support scale (support mobihzation). Items were added to the W C C L - R to assess aggressive coping styles (compris ing a n inth scale) and perceptions of outcomes and resolutions (see A p p e n d i x K) . A l d w i n and Revenson's (1987) items and factor loadings were chosen as the coping factors they identif ied were similar to coping strategies identified in the abuse literature. The selection of dependent measures has been frequently addressed in discussions of the methodological issues i n research on c h i l d maltreatment. Investigators have emphasized the need for measures that are developed with reference to maltreatment and l inked to an underlying theoretical model such as the organizational/developmental model (e.g., Briere, 1992; M a s h & Wolfe, 1991). A s discussed earlier, maltreatment researchers have addressed the theoretical and empirical relationship between maltreatment and the development of emotional regulation and cognition, and social/behavioral competence. Investigators such as Mrazek and Mrazek (1987) have discussed the cognitions and behaviors used by maltreated children in attempts to cope with their abusive environments. The role of perceptions and attributions as mediators of maltreatment also has been emphasized (e.g., Wol fe , 1987), and the potential value of examining adult percept ions/recal l of c h i l d h o o d experiences has recently been discussed by researchers in the area of maltreatment (e.g., Beitchman et al . , 1992; Mart in «Se Elmer, 1992). However , there are no consistently used measures of adult recall of c h i l d h o o d coping strategies w h i c h have been developed w i t h reference to maltreatment. Therefore, the W C C L - R was chosen as it is the most popular and best constructed measure of coping strategies (Tennen & Herzberger , 1985), was developed from a wel l - formulated theory of coping, and is supported by an extensive program of research (e.g., Folkman & Lazarus, 1980; Folkman & Lazarus, 1985; Folkman et al. , 1986). The internal consistency has been demonstrated to be good (e.g., 91% rater agreement classifying each item as problem-focused or emotion-focused; Problem-focused scale - Cronbach alpha =.80, emotion-focused scale -Cronbach alpha =.81) and the above cited studies have documented construct and concurrent va l id i ty . A summary of the central independent and dependent variables is presented in Table 4. The clinical subjects' therapists also completed a brief rating of the subjects' adjustment i n intimate and conflictual interpersonal interactions (see A p p e n d i x L). Each item was rated on a 7-point Likert-scale reflecting subjects' ability to respond adaptively (l=not at all ; 7=consistently). Items were selected to reflect aspects of adjustment consistently noted in the abuse literature (e.g., establishment of trust, management of daily demands). This measure was included in an attempt to obtain multisource, multimethod data (e.g.. Mash & Wolfe, 1991; Mi lner , 1991). Table 4 Sumnnary of the Independent and Dependent Variables: Independent Variables: Abuse Group: Audiotape: Dependent Variables: Perceptual: Coping: Physiological: Sexual abuse Physical abuse Family Disrupt ion Controls Neutral content Conflictual content Intimate content Self-Perceptions Other-Perceptions Open-ended coping Structured coping Heart rate indices Diastolic blood pressure indices Systolic blood pressure indices Self-report of physiological response: - Body Sensations Questionnaire (BSQ) Procedure Depending on the source of referral, testing was conducted at the University of British Columbia , the Alberta Children's Hospital (Calgary), the agency site, or at the subject's home. The subjects were escorted individual ly to the testing room by the experimenter. Each subject was asked to sign a consent form containing the general experimental instructions (see A p p e n d i x M ) . The experimenter then reviewed the explanation of the study/measures that was initially presented to the subject when she first met wi th her therapist to discuss potential participation. This explanation was described in the Method section. The subject was shown the physiological recording device and four baseline recordings were obtained over a 15-minute period (Linden, 1985). The subject was seated and fitted with a set of headphones facing a table containing a tape recorder and the self-report questionnaires. Before beginning, the experimenter addressed any questions or concerns raised by the subject. The subject was then instructed that she w o u l d hear three social interactions, and that (a) for each interaction, she was to listen and imagine herself assuming the female's role i n the interaction, and (b) when each interaction was over, she was to stop the tape and complete the set of questionnaires pertaining to that interaction. The instructions also indicated that the subject was free to interrupt or terminate the procedures at any time. The experimenter then activated the stimulus tape and the physiological recording device (two physiological readings were taken over a two minute period at one and two minutes). When the subject had completed the questionnaires for the first interaction, she notified the experimenter and two baseline physiological recordings again were obtained (i.e., readings taken at two and five minutes). The subjects took approximately 5 minutes to complete the questionnaires. Fo l lowing this, the tape recorder was reactivated and the procedure was repeated for the second and third interactions. The subject also completed the W C C L - R , referring to her recollection of childhood coping strategies, and alerted the experimenter when finished. The experimenter paid and thanked the subject for her participation. She explained the rationale of the study again and the subject was encouraged to voice questions or concerns. If any subject displayed distress at any point dur ing the proceedings, the experiment was stopped and the subject's distress was addressed. O n l y two subjects displayed distress. Both these women were sexual abuse victims, and their data were not used. Subjects also were made aware of the appropriate resources available to them, and a referral was made when requested. Results Validity of audiotape content was further supported by the consistent finding of a strong audiotape main effect in the central study. The results of multivariate contrasts reflected the fact that the conflictual audiotape was significantly different in comparison to the neutral audiotape in regards to: physiological response, F(4,115)=36.36, p=<.001; coping responses, F(6,107)=23.82, £=<.001; and self- and other-perceptions, F(6,107)=34.54, p=<.001. Univariate planned contrasts revealed that, in comparison to the neutral audiotape, the conflictual audiotape was seen as more physiologically arousing, F(l,118)=138.95, £=< .001 , gave rise to greater report of use of coping strategies characterized by discussion with friends, F(l,n2)=86.89, £=<.001, gave rise to more attempts to understand the man's behavior, F(l,112)=36.77, £=< .001 , and confrontation, F(I/n2)=57.82, p<.001, and gave rise to greater report of perceiving oneself as weak, F(l,112)=34.44, £< .001, and bad, F(l,112)=85.30, £< .001 , and the man as bad, F(l,112)=128.55, £<.001. In regards to open-ended coping strategies, the conflictual audiotape appeared to elicit greater report of strategies i n v o l v i n g negative emotionality and less frequent report of v iewing the situation as non-problematic or relying on assertive strategies, in comparison to the neutral audiotape. When the conflictual audiotape was compared to the dating audiotape, the results of multivariate contrasts indicated that the audiotapes were significantly different in regards to: physiological data, F(4,115)=9.67, £< .001 ; structured coping data, F(6,107)=12.37, p<.001; and self- and other perceptions, F(6,107)=21.36, £< .001 . Univariate planned contrasts revealed that, in comparison to the dating audiotape, the conflictual audiotape was seen as more physiologically arousing, F(l/118)=36.50, p.<.001, gave rise to greater report of use of coping strategies characterized by discussion w i t h friends, F(l,112)=16.78, p.<.001, gave rise to more attempts to understand the man's behavior, F(l,112)=16.78, g.<.001, and confrontation, F(l,112)=13.65, £< .001 , and gave rise to greater report of perceiving oneself as weak, F(l,112)=11.53, £ = . 0 0 1 , and bad, F(l,112)=61.85, £ = . 0 0 1 , and the man as weak, F(l,112)=9.20, £= .001 , and bad, F(l,112)=85.01, £< .001 . In regards to open-ended coping strategies, the conflictual audiotape appeared to elicit greater report of strategies involv ing negative emotionality and anger, and less frequent report of reliance upon assertive strategies or viewing the situation as non-problematic, in comparison to the dating audiotape. Finally, the dating audiotape was compared to the neutral audiotape. Results of multivariate contrasts revealed that the audiotapes were significantly different in regards to: physiological data, F(4,115)=23.02, £ < . 0 0 1 ; structured coping data, F(6,107)=10.53, £< .001 ; and self- and other-perceptions, F(6,107)=6.92, £<.001. The results of univariate planned contrasts indicated that, i n comparison to the neutral audiotape, the dat ing audiotape was seen as more physio logica l ly arousing, F(l,118)=90.88, £ < . 0 0 1 , gave rise to greater report of use of coping strategies characterized by negotiation, F(l,112)=11.23, £ = . 0 0 1 , discussion w i t h friends, F(l,112)=31.92, £<.001, confrontation, F(l,112)=16.23, £< .001 , and less frequent use of submission/acquiescence, F(l,112)=17.57, £<.001, and gave rise to greater report of viewing the man i n the interaction as weak, F(l,112)=17.91, £< .001 , and passive. F(l,112)=19.62), p<.001. In regards to open-ended coping strategies, in comparison to the neutral audiotape, the dating audiotape appeared to elicit greater report of strategies involving assertiveness, and less frequent report of reliance upon physical withdrawal or viewing the situation as non-problematic. In s u m m a r y , the c o n f l i c t u a l audiotape was v i e w e d as the most physiologically arousing, demanding of coping responses, and evocative of negative self- and other-perceptions. The neutral audiotape was v i e w e d as the least provocative audiotape and the dating audiotape was perceived to fall between the conflictual and neutral audiotapes. Central Analyses: Overview Due to the large numbers of dependent variables i n the analyses, a conservative approach to the data was adopted. The dependent variables were grouped into three sets of variables; self- and other-perception variables , structured coping variables, and physiological variables. Mult ivariate planned orthogonal contrasts were conducted for each of the three audiotapes over each of the three sets of variables. Tabachnick and Fidel l (1989) noted that when there are more than two levels i n a significant multivariate main effect, and when a dependent variable is important to that main effect (as in the central study), it is useful to perform specific comparisons of the dependent variable to pinpoint the source of the significant difference. When the comparisons are planned (a priori) , then they are performed in lieu of omnibus F; that is, the researcher moves straight to comparisons (e.g., M a r a s c u i l o & L e v i n , 1983; Tabachnick & Fidel l , 1989). The central hypotheses addressed group main effects; that is, are there mean differences i n interpersonal functioning - measured by physiological activity, self- and other-perceptions, and coping strategies - associated with differences in group membership? The central study involved a large number of dependent variables together wi th a relatively small sample size. If interaction effects were also addressed, very stringent alpha levels w o u l d be required, reducing the power of the analyses and the chances of f inding any significant interactions. A d d i t i o n a l l y , a greater number of cases per group w o u l d be required to avoid singularity of variance-covariance matrices (Tabachnick & Fidel l , 1989). Tabachnick and Fidel l (1989) also emphasized that in analyses, degrees of freedom are analyzed in i t ia l ly at conventional alpha levels, but that further analyses require very stringent alpha levels, thereby reducing the power of the analyses. They suggest, therefore, that a better strategy than routine repeated measures A N O V A or M A N O V A testing is the use of carefully planned comparisons so that the most interesting comparisons are tested w i t h more powerful alpha levels. Wi th the planned approach, there are fewer comparisons across which the familywise alpha needs to be distributed. Briere (1992) also emphasized the need to use the highest level of measurement possible. Thus, planned comparisons were chosen as the best approach i n that the procedure made conceptual sense w i t h i n the context of the research design (i.e., a priori predictions) and allowed for more powerful analyses. The multivariate planned contrasts procedure assumes homogeneity of variance in each cell, which can be evaluated using the Bartlett-Box procedure. The distribution normality assumption is quite robust and Pillai's Trace test statistic was used as it is the most robust and powerful test criterion (Norusis, 1990). The veracity of the computer-derived multivariate contrasts was established by hand calculations based on formulas found in the SPSS Advanced Statistics Users' Guide (Norusis, 1990). The results were also cross-validated through the use of the B M D P statistical package. A l p h a for the planned contrasts was set at .01 rather than the more liberal .05 in order to make the analyses more stringent and conservative. Significant multivariate analyses were fol lowed by univariate planned contrasts. The first contrast compared the sexual and physical abuse groups together versus the family disruption group. The second contrast compared the sexual abuse group to the physical abuse group. The third contrast was between the sexual abuse, physical abuse, and family disruption groups combined versus the control group. This was done over the three variable groupings noted above. Self- and other-perceptions (i.e., weak-strong; passive-active; bad-good) comprised the first grouping . Structured coping strategies (i.e., emotional detachment, negotiation, discussion wi th friends, attempts to understand the man's behavior, confrontation, and submission/acquiescence) were placed together in the second variable grouping. Physiological responses (e.g., systolic and diastolic blood pressure, and heart rate) and self-report of physiological response (total summary score) comprised the third variable grouping. Residual gain scores were calculated for the physiological responses and these residual gain scores were then treated as raw data. Residual gain scores express post-test scores (e.g., physiological recordings obtained whi le subjects listened to the audiotapes) as the deviation from the post-test - on - pre-test regression line. Therefore, the residual gain procedure partials out the part of the post-test information that is l inearly predictable from the pre-test (e.g., the baseline physiological recordings obtained dur ing rest periods between audiotapes). Essentially, the use of residual gain scores singles out subjects who changed more or less than would be expected from their init ial score. The use of residual gain scores assumes that the correlations between l inked observations (e.g., baseline scores and audiotape scores) w i l l be higher than that between independent observations. The procedure assumes that the mean of random errors over subjects equals zero and that there is equal variance for every condition (e.g., abuse group). Addi t ional ly , intercorrelations of random errors wi th other components is assumed to be equal to zero (Cronbach & Furby, 1970). The remaining class of dependent variables was analysed through the use of Pearson's chi-square analyses. The open-ended coping responses were read independently by two raters. After reading the responses, the raters each devised a list of categories that summarized all the subjects' responses. The agreement rate on the overal l categories (i.e., the extent to w h i c h raters each identif ied the same general category) was 84%. This category system was revised to resolve interrater differences in category nomination. In al l , 15 coping categories were included i n the final classification system. The raters then independently read each answer and assigned it to a category based on detailed descriptions of each category. Interrater agreement rates are presented in Table 5. In general, each subject's response to each audiotape was assigned to one coping category. In several cases, two coping categories (a pr imary and a secondary category) were coded for subjects whose responses combined two coping categories. The most salient strategy was coded first and was used for the data analyses reported below. Interrater Agreement Rates: Open-ended Coping Responses Tape e o p m g Category Conflict^ Tape N e u t r a l ^ Tape Dating^ Tape Cognit ive Reinterpretation 1.00 1.00 1.00 Assertiveness .88 .91 .84 Physical Withdrawal .85 .84 .83 A n g r y Emotional Reaction .86 1.00 1.00 Negative Emotional Reaction .86 .83 .91 Aggression .96 .90 .83 Self-blame .86 1.00 1.00 Compliance - Emotional detachment .82 .90 1.00 Compliance - H i d e feelings 1.00 1.00 1.00 Compliance - Appeasement .85 1.00 1.00 Compliance - Passive acceptance .83 .95 .87 Prosocial - Listen 1.00 .80 1.00 Prosocial -Feel sympathetic 1.00 .88 1.00 Prosocial - Give advice .95 1.00 1.00 N o problem 1.00 .87 .91 ^Maximum score = 1.00 Some categories were infrequently cited; therefore, the data was regrouped into larger units to reflect five basic coping categories: (1) Prosocial or Assertive response; (2) Anger and Aggressive response; (3) Physical Withdrawal response; (4) Responses marked by Compliance or Anxiety/Fear; (5) N o problem cited. Pearson's chi-square analyses (Group by Coping Category) were conducted for each audiotape. In order to investigate order effects, a series of M A N O V A S were conducted for each of the three audiotapes over each of the three sets of variables (i.e., self- and other-perception variables, structured coping variables and physiological variables), using order (i.e., presented in the first, second, or third position) as the grouping variable. The open-ended coping responses were analysed through the use of Pearson's Chi-square analyses. A similar strategy was used to investigate sequence effects, using sequence as the grouping variable. For example, subjects w h o listened to the neutral audiotape first were compared to subjects who heard the neutral audiotape after: the conflictual audiotape; the dating audiotape, and the confl ictual and dat ing audiotapes. The results that f o l l o w w i l l be presented i n discrete sections. The victimization experiences w i l l be reviewed first, fol lowed by an examination of the dependent variables, audiotape by audiotape. Therefore, for the neutral audiotape, the self- and other-perception variables, open-ended coping variables, structured coping variables and the physiological variables, w i l l be addressed in succession. This pattern w i l l then be repeated for the conflictual audiotape and finally for the int imate/dating audiotape. Fol lowing this, the additional measures (i.e., W C C L - R responses) w i l l be discussed. Vict imizat ion Experience The subjects' reported a range of sexually abusive experiences. The age of the vict im at the onset of sexual abuse ranged from 1 to 16 years. The mean age was 6 years (SD=3.00). The sexual abuse lasted from 6 months to 16 years^ with a mean duration of 6.4 years (SD=4.19). In 36.7% of the incidents, the offender was the natural father, in 20% of the incidents was a sibling, and in 30% of the incidents the offender was known to the family (stepfather, relative, family friend). In 56.7% of the cases, fondling was reported as being consistently involved, and in 58.6% of the cases, penetration was reported as being involved more than occasionally. In regards to the physically abusive experiences, the victims' age at the onset of the physical abuse ranged from 1 month to 16 years. The mean age was 5.2 years (SD=5.03). The physical abuse lasted from 2 to 21 years with a mean duration of 10.0 years (SD=5.10). In 51.10% of the incidents, the offender was the natural father, in 28.9% of the incidents the offender was the natural mother, i n 11.1% of the incidents, the offender was a peer, and i n 8.9% of the cases, the offender was a sibl ing. The fol lowing characteristics were reported to be more than occasionally i n v o l v e d by the parenthesized percentage of victims: bruis ing (72.4%); use of belts/sticks (57.2%); cuts (27.9%); kicks from boots/shoes (20.5%); burns/cuts from weapons (10.0%); and, scarring (7.3%). The reported family disruption ranged from m i l d (1) to severe (7) in global severity. The mean severity was slightly above moderate (5) (SD=1.43). The fo l lowing characteristics were reported to be more than occasionally involved by the parenthesized percentage of victims: f ighting/violence and/or verbal aggression between parents (59.2%); emotional abuse (51.7%); parental alcohol abuse (50.1%); neglect/ lack of support (48.3%); poor communication (43.3%); separation from parents (41.7%). Audiotape One: Neutral Content Self- and other-perception data. The mean self- and other-perception scores are presented in Tables 6 and 7. N o planned contrasts were significant (i.e., abuse groups versus family disruption group; sexual abuse versus physical abuse; clinical groups versus controls). Open-ended coping data. The open-ended coping response frequencies for the four groups are presented in Table 8. The Pearson's chi-square analysis for the coping strategies was significant, XHA, N=132)=27.26, p=.007. Post hoc statistical analyses indicated that the significant results were pr imar i ly attributable to differences between the sexual and physical abuse subjects, X^(4, N=72)=12.47, £= .01. The physical abuse subjects reported more frequent use of assertive responses and physical withdrawal, and less frequently perceived the situation to pose no problem, w h e n compared to the sexual abuse subjects. N o other group contrasts were significant. Structured coping data. The mean structured coping scores are presented in Table 9. The results of the planned contrasts indicated that none of the contrasts were significant. Group Self-Perception Sexual Physical Family Controls Category Abuse Abuse Disruption Ma S.D. Ma S.D. M ^ S.D. M ^ S.D. weak-strong 4.66 (1.88) 4.41 (1.27) 4.60 (1.48) 4.72 (1.10) passive-active 4.59 (1.97) 3.86 (1.58) 3.40 (1.48) 3.38 (1.40) bad-good 5.00 (1.77) 4.45 (1.43) 4.20 (1.47) 4.35 (1.08) ^Maximum score = 7 which corresponds to the adjective to the right of each pair. Group Other-Perception Sexual Physical Family Controls Category Abuse Abuse Disruption M a S ^ . M a _S^ . SJD. M a S.D. weak-strong passive-active bad-good 5.00 (2.12) 4.03 (1.50) 5.31 (2.07) 4.69 (1.58) 5.03 (1.52) 4.17 (1.34) 3.83 (1.32) 3.76 (1.60) 4.47 (1.59) 4.83 (1.04) 3.77 (1.04) 4.24 (1.19) aMaximum score = 7 which corresponds to the adjective to the right of each pair. Open-ended C o p i n g Response Frequencies: Neutral Tape* G r o u p Coping Sexual Physical Family Controls Category Abuse Abuse Disruption (n=42) (n=30) (n=32) (n=28) Prosocial/ 16 (38.1) 15 (50.0) 14 (43.8) 18 (64.3) Assertiveness Physical 4 (9.5) 8 (26.7) 8 (25.0) 3 (10.7) withdrawal Anger-Aggression 4 (9.5) 2 (6.7) 5 (15.6) 0 (0.0) Compliance 6 (14.3) 5 (16.7) 3 (9.4) 5 (17.9) Anxiety/Fear N o problem cited 12 (28.6) 0 (0.0) 2 (6.3) 2 (7.1) Numbers in parentheses refer to column percentages * Includes al l subjects run in the larger research project, not just the 30 subjects described i n the selection procedures. Mean Coping Scores: Structured Self-report of Use of Coping Strategies in Response to Neutral Tape Coping Category Sexual Abuse Ma S.D. G r o u p Physical Abuse Ma S.D. Family Disruption M a S.D. Controls M a S.D. Cope 1 (emotional detachment) Cope 2 (negotiation) Cope 3 (discussion with friends) Cope 4 (attempt to understand other's behavior) 3.43 (2.55) 2.36 (2.08) 1.93 (1.80) 2.54 (2.13) Cope 5 1.47 (0.84) (confrontation) Cope 6 (submission/ acquiescence) 3.71 (2.12) 4.45 (2.05) 3.00 (1.80) 2.69 (1.82) 3.59 (1.97) 2.31 (1.82) 3.03 (1.95) 4.63 (1.87) 2.40 (1.61) 2.50 (1.84) 2.90 (1.79) 2.17 (1.56) 3.30 (1.80) 4.40 (2.11) 2.87 (1.91) 1.70 (1.06) 3.27 (2.05) 2.33 (1.90) 3.33 (2.01) aMaximum score = 7 Physiological data. The mean summary scores of self-report of physiological response, physiological residual gain scores, and raw physiological response scores for the four groups are presented in Tables 10-14. The results of the planned contrasts indicated that none of the contrasts were significant. Audiotape Two: Confl ictual/Argumentat ive Content Self- and other-perception data. The mean self- and other-perception scores are presented in Tables 15 and 16. N o planned contrasts were significant (i.e., abuse groups versus family disruption group; sexual abuse versus physical abuse; clinical groups versus controls). Open-ended coping data. The open-ended coping response frequencies for the four groups are presented in Tables 17. In the analyses using the five basic coping strategies (i.e, assertiveness, physica l w i t h d r a w a l , anger-aggression, compliance, and no problem cited), the Pearson's chi-square analysis was significant, X2(9, N=130)=17.63, p.=.04). Post hoc statistical group comparisons yielded no significant results. Structured coping data. The mean structured coping scores are presented in Table 18. The results of the planned contrasts indicated that none of the contrasts were significant. Physiological data. The mean summary scores of self-report of physiological response, physiological residual gain scores, and raw physiological response scores for the four groups were presented in Tables 10-14. The results of the planned contrasts indicated that none of the contrasts were significant. Group Sexual Physical Family Controls Tape Abuse Abuse Disruption Ma S.D. M a S.D. M a S.D. M a S.D. Tape 1 25.63 (13.05) 23.38 (10.84) 26.10 (11.52) 19.67 (4.48) (neutral) Tape 2 51.80 (24.99) 48.63 (18.69) 43.60 (16.02) 43.87 (23.17) (conflictual) Tape 3 45.50 (20.07) 35.20 (14.04) 34.13 (12.49) 34.87 (14.97) (dating) aMaximum score = 119 Group Tape Sexual Physical Family Controls Abuse Abuse Disruption M S.D. M S.D. M S.D. M N e u t r a l heart rate .77 ((2.68) .04 (2.06) -.90 (3.00) -.17 diastolic blood pressure -.75 (6.15) -.97 (4.00) 1.77 (4.93) -.07 systoUc blood pressure .49 (4.77) -.52 (4.06) -.16 ((3.99) .17 ConfHctual heart rate 1.00 (6.27) .27 (5.87) -1.07 (5.37) -.20 diastolic blood pressure .45 (9.26) -.20 (7.65) -.03 (5.93) -.27 systoUc blood pressure .34 (10.17) -.19 (6.96) .60 (7.64) -.75 Dating heart rate .67 (4.07) -.43 (3.94) -1.10 (4.26) .87 diastolic blood pressure -1.47 (8.11) .79 (4.87) -.07 (6.11) .75 systoHc blood pressure -.71 (8.36) .55 (5.18) .62 (7.47) -.47 Group Tape Sexual Physical Family Controls Abuse Abuse Disruption M S.D. M S.D. M S.D. M S.D. Overall Baseline heart rate 78.18 (10.16) 78.43 (11.72) 76.23 (11.09) 73.32 (12.89) Neutral Tape Baseline 77.70 (10.33) 76.64 (9.33) 76.62 (11.17) 74.00 (11.49) Tape Response 77.21 (10.34) 73.55 (16.12) 73.63 (11.07) 72.72 (11.96) Conflictual Tape Baseline 78.25 (10.74) 77.69 (10.61) 75.62 (10.57) 74.47 (11.97) Tape Response 77.22 (13.88) 76.85 (12.70) 73.40 (11.73) 72.20 (10.19) Dating Tape Baseline Tape Response 77.67 (10.64) 78.62 (11.03) 76.27 (10.57) 73.97 (12.34) 76.25 (10.71) 75.48 (11.12) 72.87 (10.84) 72.92 (10.95) Group Tape Sexual Physical Family Controls Abuse Abuse Disruption M S.D. M S.D. M S.D. M S.D. Overal l Baseline Diastolic Blood Pressure 63.98 (9.58) 68.79 (9.34) 66.80 (10.44) 68.38 (8.62) Neutral Tape Baseline 65.65 (10.27) 67.53 (10.26) 68.78 (9.93) 69.35 (6.50) Tape Response 60.83 (10.26) 65.05 (9.65) 67.15 (10.81) 66.35 (8.28) Conflictual Tape Baseline 64.18 (9.01) 71.00 (9.15) 67.60 (10.80) 70.32 (6.87) Tape Response 62.92 (10.61) 66.18 (12.51) 65.75 (10.05) 66.32 (6.35) Dating Tape Baseline 64.12 (7.87) 68.90 (9.63) 68.95 (11.61) 68.68 (9.03) Tape Response 59.75 (11.56) 66.60 (10.33) 64.97 (11.17) 66.80 (7.66) Group Tape Sexual Physical Family Controls Abuse Abuse Disruption M S.D. M S.D. M S.D. M S.D. Overal l Baseline SystoUc Blood Pressure 121.23 (14.97) 117.81 (13.34) 118.38 (11.07) 113.93 (7.33) Neutral Tape Baseline 120.05 (12.67) 116.40(11.49) 118.22 (11.38) 114.22 (7.35; Tape Response 118.93 (12.65) 116.67(12.21) 116.48 (10.79) 113.20 (9.45^ Conflictual Tape Baseline 118.03 (14.74) 119.85(14.36) 116.73 (12.40) 112.92 (9.99) Tape Response 120.18 (16.25) 119.13(13.51) 118.67 (12.75) 113.77 (7.19) Dating Tape Baseline Tape Response 119.28 (14.23) 118.63 (14.70) 118.95 (13.15) 119.70(14.42) 116.67 (10.84) 113.03 (8.69) 118.90 (11.88) 113.97 (8.39) Self-Perception Controls Category Sexual Abuse M a S.D. Group Physical Abuse M a S.D. Family Disruption M a S.D. M a S.D. weak-strong passive-active bad-good 3.28 (2.51) 2.52 (1.48) 3.80 (1.88) 3.83 (2.14) 3.66 (2.45) 3.00 (1.60) 4.07 (2.02) 4.14 (2.13) 2.10 (1.66) 2.95 (1.55) 3.37 (1.61) 2.83 (1.67) aMaximum score = 7 which corresponds to the adjective to the right of each pair. Self-Perception Controls Category Sexual Abuse M a S.D. Group Physical Abuse M a S.D. Family Disruption M a S.D. M a S.D. weak-strong passive-active bad-good 4.86 (2.59) 4.97 (2.29) 3.73 (1.96) 3.62 (2.09) 5.45 (2.06) 5.55 (1.84) 4.67 (1.69) 5.17 (1.28) 1.93 (1.28) 2.41 (1.52) 2.23 (1.10) 2.52 (1.43) aMaximum score = 7 which corresponds to the adjective to the right of each pair. Group Coping Sexual Physical Family Controls Category Abuse Abuse Disruption (n=42) (n=30) (n=31) (n=27) Prosocial/ 11 (26.2) 13 (43.3) 13 (41.9) 16 (59.3) Assertiveness Physical 12 (28.6) 1 (3.3) 4 (12.9) 1 (3.7) wi thdrawal Anger-Aggression 9 (21.4) 8 (26.7) 7 (22.6) 7 (25.9) Compliance 10 (23.8) 8 (26.7) 7 (22.6) 3 (11.1) Anxiety/Fear N o problem cited 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) Numbers i n parentheses refer to column percentages * Includes all subjects run in the larger research project, not just the 30 subjects described in the selection procedures. Mean Coping Scores: Structured Self-report of Use of Coping Strategies in Response to Conflictual Tape Group Coping Sexual Physical Family Category Abuse Abuse Disruption Controls S ^ - A D . Ma S.D. Ma S.D. C o p e l 4.62 (2.44) 5.27 (2.07) 4.90 (2.06) 3.80 (2.22) (emotional detachment) Cope 2 2.90 (1.92) 3.20 (2.17) 3.10 (1.74) 3.53 (2.03) (negotiation) Cope 3 4.24 (2.40) 4.60 (2.24) 4.79 (2.11) 3.90 (1.94) (discussion wi th friends) Cope 4 4.48 (2.25) 4.80 (1.83) 4.03 (1.92) 5.01 (1.82) (attempt to understand other's behavior) Cope 5 4.14 (2.46) 5.53 (2.40) 4.07 (1.93) 4.50 (2.10) (confrontation) Cope 6 3.17 (2.33) 3.27 (2.36) 2.79 (2.21) 2.07 (1.68) (submission/ acquiescence) M a x i m u m score - 7 Audioape Three: Intimate/Dating Content Self- and other-perception data. The mean self- and other-perception scores are presented in Tables 19 and 20. The results of the planned orthogonal contrasts reflected the fact that the contrast between the sexual and physical abuse subjects in comparison to the family disruption subjects was significant, F(6,108)=3.48, £= .003. Univariate planned contrasts revealed that members of the sexual and physical abuse groups were more l ikely than members of the family disrupt ion group to view themselves as weak, F(l,n3)=10.91, p=.001, and bad, F(l,113)=8.10, p=.005. N o other contrasts were significant (i.e., sexual abuse versus physical abuse; clinical groups versus controls). Open-ended coping data. The open-ended coping response frequencies for the four groups are presented in Tables 21. The Pearson's chi-square analysis for the coping strategies was not significant, X^(9, N=132)=17.54, p=.13. Structured coping data. The mean structured coping scores are presented in Table 22. The results of the planned contrasts indicated that none of the contrasts were significant. Physiological data. The mean summary scores of self-report of physiological response, physiological residual gain scores, and raw physiological response scores for the four groups were presented i n Tables 10-14. The results of the planned contrasts indicated that none of the contrasts were significant. Addi t ional Measures: Coping in Chi ldhood The Ways of Coping Checklist - Revised (WCCL-R) responses reflected G r o u p Self-Perception Sexual Physical Family Category Abuse Abuse Disruption Controls SJD. M a S.D. weak-strong 3.10 (1.78) 4.30 (1.64) 4.83 (1.37) 4.61 (1.26) passive-active 3.24 (1.55) 4.33 (1.69) 4.50 (1.36) 4.11 (1.52) bad-good 3.31 (1.91) 4.30 (1.80) 4.90 (1.56) 4.86 (1.55) ^Maximum score = 7 which corresponds to the adjective to the right of each pair. G r o u p Other-Perception Sexual Physical Fannily Category Abuse Abuse Disruption Controls M a S £ . M a _ S £ . S.D. M a S.D. weak-strong 5.62 (1.52) 4.83 (1.64) 4.60 (1.69) 4.75 (1.38) passive-active 5.86 (1.53) 5.63 (1.25) 5.10 (1.45) 5.50 (1.07) bad-good 3.83 (2.21) 4.30 (1.78) 4.10 (1.77) 4.00 (1.56) aMaximum score = 7 which corresponds to the adjective to the right of each pair. Group Coping Sexual Physical Family Controls Category Abuse Abuse Disruption (n=42) (n=30) (n=32) (n=28) Prosocial/ 25 (59.5) 15 (50.0) 24 (75.0) 21 (75.0) Assertiveness Physical 4 (9.5) 5 (16.7) 0 (0.0) 3 (10.7) wi thdrawal Anger-Aggression 3 (7.1) 0 (0.0) 0 (0.0) 1 (3.6) Compl iance Anxiety/Fear 5 (11.9) 7 (23.3) 7 (21.9) 2 (7.1) N o problem cited 5 (11.9) 3 (10.0) 1 (3.1) 1 (3.6) Numbers in parentheses refer to column percentages * Includes all subjects run in the larger research project, not just the 30 subjects described in the selection procedures. Mean C o p i n g Scores: Structured Self-report of Use of Coping Strategies i n Response to Dating Tape G r o u p C o p i n g Sexual Physical Family Category Abuse Abuse Disruption Controls S £ . S £ . M a S.D. M a S.D. C o p e l 4.43 (2.30) 4.33 (2.14) 3.87 (1.91) 3.71 (2.26) (emotional detachment) Cope 2 3.60 (2.31) 3.40 (1.79) 3.27 (1.86) 3.96 (1.86) (negotiation) Cope 3 3.30 (2.41) 4.13 (2.06) 3.20 (2.06) 3.86 (1.96) (discussion w i t h friends) Cope 4 3.53 (2.24) 4.20 (1.79) 3.13 (1.93) 4.04 (2.27) (attempt to understand other's behavior) Cope 5 3.10 (2.34) 3.03 (2.01) 3.10 (1.95) 3.04 (1.90) (confrontation) Cope 6 2.77 (2.08) 2.27 (1.91) 2.40 (1.65) 1.93 (1.70) (submission/ acquiescence) a M a x i m u m score = 7 subjects' retrospective reports of childhood coping, and these data were collected from the subjects as adults (i.e., the data regarding chi ldhood coping were not collected from children). The mean W C C L - R factor scores are presented in Table 23. The results of planned orthogonal contrasts revealed that i n comparison to the subjects in the family disruption group, the abuse subjects d i d not significantly differ in regards to coping strategies and outcome, F(ll,106)=1.58, £=.12. Similarly, sexual abuse subjects d i d not significantly differ from physical abuse subjects, F(ll,106)=.70, £= .73 . Focusing on the clinical subjects in comparison to the control subjects, a significant difference was obtained, F(n,106)=5.71, £< .001 . Specifically, the control subjects reported: more successful childhood outcomes (i.e., positive changes in: the situation, treatment from others, and feelings about oneself), F(l,116)=19.26, £<.001; more successful adult outcomes (i.e., put memories to rest, found meaning in what happened; understands the past; stronger as a result), F(l,116)=11.72, £= .001 ; less rehance on the use of "escapism", F(l,116)=15.87, £ < . 0 0 1 ; and "minimizat ion" , F(l,116)=13.50, £=< .001 ; and greater reliance on the use of "support mobilization", F(l ,116)=20.58,£<.00l2. Mean W C C L - R Scores: Coping i n Chi ldhood Group Factor Sexual Physical Family N a m e Abuse Abuse Disruption Controls M S.D. M S.D. M S.D. M S.D. Historical^ outcome 2.98 (1.32) 2.84 (1.15) 3.49 (1.15) 4.30 (1.47) Current^ outcome 4.97 (0.87) 5.18 (0.93) 5.51 (0.65) 5.80 (0.79) Escapism'^ 1.91 (0.57) 1.87 (0.43) 1.71 (0.49) 1.34 (0.76) Cautiousness^» 1.02 (0.75) 1.35 (0.55) 1.20 (0.58) 1.44 (0.48) Minimizationt» 1.80 (0.67) 1.66 (0.56) 1.54 (0.69) 1.19 (0.52) Support^» M o b i l i z a t i o n 0.83 (0.75) 1.06 (0.70) 0.79 (0.60) 1.56 (0.71) Self-Blameb 1.60 (0.83) 1.63 (0.63) 1.52 (0.54) 1.38 (0.65) Aggressiveness'' 1.30 (0.91) 1.47 (0.88) 1.15 (0.76) 1.12 (0.81) Negotiation^ 1.10 (0.78) 1.22 (0.68) 0.96 (0.60) 1.16 (0.64) Instrumentait» A c t i o n 0.93 (0.60) 1.19 (0.55) 0.96 (0.51) 1.23 (0.50) Seekingt» M e a n i n g 1.18 (0.89) 1.22 (0.81) 1.05 (0.80) 1.14 (0.72) a M a x i m u m Score = 7. Higher scores reflect more positive outcomes. •^Maximum Score = 3. Higher scores reflect more frequent use. Discussion The first question addressed by this study was whether survivors of abusive (i.e., p h y s i c a l / s e x u a l abuse) chi ldhood environments, currently in treatment where in they were addressing unresolved feelings about their c h i l d h o o d experiences, responded to social information i n a less functional manner than women w i t h histories of family chaos without such abuse. The data f rom the coping, physiological , and the social perception measures generally supported the Family Chaos Posit ion; that is, no differences emerged between subjects w i t h chi ldhood histories of abuse and subjects w i t h histories of family disrupt ion in regards to selection of strategies for coping wi th any of the audiotaped interpersonal situations, or i n terms of physiological response to the audiotapes. F r o m an organizat ional/developmental perspective, this w o u l d i m p l y that maltreatment and family dysfunction may be best conceptualized more generally as the inabili lty of caretakers to nurture offspring. Support was found for what I have labelled the Abuse Posit ion on the measures of social perception for the intimate audiotape. Abused subjects were significantly more l ikely than family disruption subjects to perceive themselves negatively on the intimate audiotape. Therefore, it may be that certain diff icult ies (i.e., self-perception i n intimate situations) develop more specifically f rom abusive experiences, beyond those theorized to develop from disordered family relationships. The second question addressed by this study was whether the type of abuse and related family environment (physical or sexual) was related to the nature of subjects' interpersonal difficulties. O n the open-ended coping measure, some support was found for what we have labelled the Specificity Position. Sexually abused subjects were more l ikely to see the neutral audiotape as "no problem", whereas the physically abused subjects were more l ikely to define the situation as problematic and to report that they w o u l d use assertiveness and physica l withdrawal to cope with the situation. The experience of chi ldhood sexual abuse was specifically related to this pattern of interpersonal responding in the neutral situation. Interestingly, no significant differences were found between the sexual abuse and physical abuse groups in terms of coping strategies for the intimate or conflictual situations. These findings, along w i t h the remaining physiological data, structured coping data, and social perception data i n the audiotaped situations supported a Generalist Position; that is, no differences emerged between the two types of abuse. Therefore, the findings appear to support more strongly the theoretical posit ion suggesting that the extent or nature of dysfunction i n abuse survivors does not necessarily depend on the exact nature of the abuse. The th ird question addressed by this study was whether women seeking treatment for traumatic chi ldhood experiences (physical abuse, sexual abuse, or family disruption) differed significantly from non-clinical control subjects matched on socio-economic and demographic variables i n terms of response to social information. The analyses contrasting the three clinical groups to the control group yielded no significant differences. This somewhat surprising f inding highlights the value of including carefully selected non-abused control subjects. A summary of the results are presented i n Tables 24 to 26. Overall , there were few significant differences. It is of note, however, that when one examines the Sumnnary of Central Results - Investigation 1: Is abuse related to greater dysfunction than family disruption alone? Hypotheses Tape Abuse Adds to Family Disruption Abuse Does Not A d d to Family Disruption N e u t r a l Self- and other-perceptions Open-ended coping Structured coping Physiological measures and self-report physiological Conflict Self- and other-perceptions Open-ended coping Structured coping Physiological measures and self-report physiological Dating Self- and other-perceptions Open-ended coping Structured coping Physiological measures and self-report physiological V a V V V Summary of Central Results - Investigation 2: Specificity Issue: Are specific types of dysfunction produced by sexually abusive experiences? Hypotheses Tape Sexual Abuse A d d s to Physical Abuse Sexual Abuse Does Not A d d to Physical Abuse N e u t r a l Self- and other-perceptions Open-ended coping Structured coping Physiological measures and self-report physiological Conflict Self- and other-perceptions V V Open-ended coping V Structured coping V Physiological measures and self-report physiological Dating V Self- and other-perceptions Open-ended coping Structured coping V Physiological measures and self-report physiological Summary of Central Results - Investigation 3: Do women (in treatment) from disruptive backgrounds i n general evidence greater dysfunction than non-clinical controls? Hypotheses Tape Cl inica l Groups Clinical Groups Do Differ from N o n - Not Differ from N o n -Clinical Groups Cl inical Controls N e u t r a l Self- and other-perceptions V Open-ended coping V Structured coping V Physiological measures and self-report physiological V Confl ict Self- and other-perceptions V Open-ended coping V Structured coping V Physiological measures and self-report physiological V Dating Self- and other-perceptions V Open-ended coping V Structured coping V Physiological measures and self-report physiological V obtained means across all of the dependent variables, some trends which lend support to the hypotheses are observable. I w i l l review potential explanations for the f indings in the fol lowing sections and discuss results regarding physiological activity, coping, and social perceptions in greater detail below, w i t h emphasis on theoretical and methodological issues. I w i l l begin w i t h the physiological and structured coping measures where few significant results were obtained. Physiological Measures In regards to the physiological measures, there were no significant differences (i.e., at the conservative .01 level of significance) between the groups on self-report of physiological response and physiological response scores. Trends which followed experimental predictions were often noted in the conflictual audiotape data but failed to reach significance at the conservative significance level that was used. One might argue that a high degree of individual variability eliminated significant group differences. Whi le the within- group variabilities were quite high (see Tables 12-14), they were not unusual in relation to other research using physiological data. Another aspect to consider in regards to the lack of significant physiological findings is that of multiple, interactive influences on cardiovascular functioning. It has been suggested that it is wrong to attempt to define the influence of any specific behavior on a single aspect of cardiovascular function in the absence of any consideration of the complex interrelationships among al l aspects of cardiovascular functioning. One model of cardiovascular functioning posits that individuals engage in cognitive appraisal of st imuli and respond either w i t h an increase in cardiovascular/motor act ivity c o i n c i d i n g w i t h fear/anxiety and thought, or w i t h a decrease in cardiovascular/motor activity, coinciding w i t h an alert, attentive, non-affective mode (WilUams, 1981). Therefore, the way that a situation is appraised becomes central to cardiovascular response. A number of researchers i n the area of maltreatment have emphasized the role of appraisal and attributions in emotional arousal (e.g., Gottman & Katz , 1989; Grych & Fincham, 1990, Wolfe et al . , 1983). Undoubtedly there are i n d i v i d u a l differences in such cognitive appraisals and cardiovascular reactivity. Addit ional ly , a person may be a reactor or a nonreactor on different physiological indices (e.g., heart rate, systolic blood pressure). Wolfe and Jaffe (1991) suggest that the extent to which maltreated subjects develop difficulties is a function of the person by situation interaction, where i n d i v i d u a l characteristics interact wi th the nature of the stressor to determine outcome. If one assumed a high degree of ind iv idua l variabil i ty in cognitive appraisal a n d / o r physiological reactivity to different indices in the present study, then subtle group differences might be obscured. Another possible explanation for the weak physiological findings may relate to treatment issues. A l l members of the clinical groups were engaged in treatment, focusing on their traumatic childhood experiences, at the time of participation. One could assume that such treatment w o u l d encompass varying methods of addressing the subjects' traumatic chi ldhood experiences (e.g., imaginai , verbal, written, role-play). It may be that this exposure to abuse-salient issues/emotions contributed to a gradual habituation of subjects' physiological responsivity to traumatic st imuli . Such a hypothesis is consistent wi th the conclusion of Pennebaker (1985), who proposed that conf id ing , w r i t i n g d o w n , or otherwise translating negative experiences reduces autonomic activity. For example, i n one study, Pennebaker asked subjects (N=15) to rate traumatic experiences and the degree to which they had confided the event to others, and to complete a health inventory assessing experience of serious health problems (e.g., ulcers, high blood pressure) and number of physician visits in the past year. Subjects who had an upsetting trauma that they had not confided to others reported more symptoms, diseases, and visits to physicians than subjects w h o had confided to others about their traumatic experience(s) or who had not experienced trauma. Abused subjects in the present study l ikely engaged i n such activities as confiding to others in the course of therapy. This may have reduced between-group differences in autonomic reactivity. It may be that confiding i n treatment acts as a compensatory mediating factor between early and later maladaptation. Conte and Schuerman (1987) stress the value of identifying risk factors related to maltreahnent that therapy can alter. It may be that autonomic reactivity is such a risk factor. Habituation of physiological reactivity fo l lowing conf iding to a therapist might help explain the lack of physiological results in the central study i n contrast to the significant physiological results obtained in the audiotape validation study. The audiotape val idat ion study demonstrated that the tapes are potent s t imul i . Therefore, the fact that the clinical subjects in the central study d i d not demonstrate differential physiological response to the audiotapes suggests that something about the clinical subjects themselves leads to the lack of physiological results. The audiotape validation study subjects were control women who were not i n therapy, while the clinical subjects were in therapy. Pennebaker's suggestion that confiding negative experiences i n therapy reduces autonomic reactivity offers a possible explanation of the obtained results. Addi t iona l ly , one might argue that the use of a more ambiguous audiotape, or of a s l ightly different stimulus, such as a videotape, w o u l d help shift the observations from trends to significant results. Mi lner (1991) emphasized that when using laboratory simulations the challenge is to develop situations that are as natural as possible and that contain opportunities to observe the characteristics of interest. He pointed out that structured laboratory situations often provide access to only a l imited sample of behavior and may be so structured that the appropriate response is obvious. It is possible that the audiotapes used in the present study were not ambiguous a n d / o r natural enough to provide the opportunity to observe group differences in autonomic reactivity. It also is important to note that even if group differences on the physiological measures were found, questions of ecological val idi ty might remain. However , ethical considerations l imit the use of less structured, more naturalistic interpersonal interactions. Finally, it may be that subjects require idiosyncratic, very specific triggers/cues to elicit physiological reactivity. If this was the case, the traumatic physiological conditioning may not be general enough to be detected in a laboratory situation. Studies in the area of maltreatment where physiological reactivity has been noted have generally used very specific child behaviors as triggers (e.g., Gottman & Katz , 1989; Vasta & C o p i t c h , 1981; Wolfe et al . , 1983). One could speculate that an i n d i v i d u a l w o u l d need to have experienced abuse and resulting physiological conditioning of immense severity to respond to general triggers, such as those in the audiotapes. Al ternat ive ly , it may be that subjects s i m p l y d i d not experience physiological conditioning during their abuse, or that the abusive experiences had significantly greater impact on other aspects of emotional expression or on cognitions/social information processing. Mash (1991) emphasized the need to expand the measurement procedures used to assess emotional expression. He pointed to indicators of emotion such as voice quality, nonverbal facial displays, visual responsiveness, body orientation, and aggregate measures of affect. There is no consensus i n the maltreatment literature, however , as to the channel of expression that might provide the richest source of affective information. Addit ional ly , it may be that the long-term impacts of maltreatment are more clearly expressed in cognitive processing. There is theoretical support for this suggestion that chi ld maltreatment may have its greatest impact on the development of internal work ing models of relationships (Alexander, 1992; Bowlby, 1988; Cole & Putman, 1992; Crittenden «& Ainsworth, 1989). Rieder and Cicchetti (1989) suggested that the disrupted internal models of the self and relationships that maltreated children develop predispose them to manifest disrupt ions i n cognitive control func t ion ing (i.e., mechanisms that determine ways individuals organize, interpret and use information) and cognitive-affective balance. They proposed that such disrupted cognitive control functioning, along wi th information-processing deficits also noted i n maltreated children (e.g., Camras, Grow & Riboray, 1983; Dodge et al., 1984; D u r i n g & M c M a h o n , 1991), may relate to the negative expectations maltreated ch i ldren have formed about relationships. Investigations u t i l i l z ing m u l t i m o d a l assessment techniques are needed to increase the empirical support for such theoretical assertions. Structured Coping Measures Turning to structured self-report of coping strategies, no significant (i.e., at the conservative .01 level of significance) differences were obtained on these single-item measures. It is possible that the items lacked reliability, or d i d not ful ly capture the exact nature of the coping strategies subjects w o u l d be most l ikely to report using in the audiotaped situations. Lazarus and Folkman (1984) have argued that the best measures of coping are precise and individual ized. Therefore, it is possible that the structured coping items used in this study failed to yield significant results given their generalized nature. The lack of significant findings i n regards to self-report of coping strategies also may reflect treatment effects. It is possible that, irrespective of group membership, the clinical subjects i n the present study addressed ways to respond to, and cope w i t h , difficult interpersonal situations as part of their treatment. If this was the case, potential group differences could be obscured. A s noted i n the d iscuss ion of the phys io log ica l measures, treatment may func t ion as an environmental mediating factor between childhood experiences and adult outcome. For example, Egeland and associates (Egeland et al., 1988) found that when compared to abused mothers who were abusing their own children, abused mothers who were able to break the cycle of abuse w i t h their own children were more l ikely to have participated in therapy at some point i n their lives. Alternatively, it is possible that group differences were lost in the use of a self-report format. One might speculate that subjects' reported response might not coincide with actual performance. Such a confounding factor operating equally across groups w o u l d c loud subtle group differences. Whi le self-report measures have been noted to have a number of strengths (e.g., high face val idity, ease of administration and scoring, access to private events), they must be carefully interpreted in regards to the degree they may represent biased, inaccurate, or seriously distorted views of family relationships (e.g.. Mash, 1991; Briere, 1992). Social Perception Measures In regards to the social perception variables, some evidence was obtained that at least some of the abused subjects currently in treatment differed from the family disruption subjects. Dodge, Bates and Pettit (1990) found that, in comparison to non-abused children, abused children developed different information processing styles; they were less attentive to relevant social cues and more biased toward attributing hostile intent. In this study, in comparison to adults not physically or sexually abused as children, adult survivors of abuse were more l ikely to attribute negative characteristics to themselves in dating situations. Thus, where significant results were obtained in the central study, it appeared that the subjects from physically and sexually abusive environments tended to process social information differently, and perhaps less adaptively, i n intimate situations i n comparison to subjects f rom disrupted/chaot ic but non-abusive environments. These perceptions and social information processing strategies may negatively influence individuals ' perceptions of others, events, and their o w n efficacy, potentially resulting in less adaptive responding. This highlights the need to address perceptual and cognitive factors i n research and treatment, i n that investigators are increasingly recognizing the importance of such factors in adequate social behavior. A d d i t i o n a l l y , the examination of perceptual and attributional processes is theoretically consistent wi th the perspective of the organizational/ developmental model that individuals develop internal models, characterized by perceptions and attributions about self, others, and the w o r l d i n general. Furthermore, there is growing empirical evidence that maltreated women's adult functioning is strongly related to their attributions and perceptions (e.g.. G o l d , 1986; Wyatt & Newcomb, 1991). The lack of more significant findings in regards to social perception measures may reflect treatment effects as suggested for structured coping measures. If subjects focused on their cognitions and perceptions in treatment, w i t h the goal of challenging and altering maladaptive cognitions/perceptions, then potential group differences could have been lost. It is also important to note that social perceptions were assessed through a self-report format. Issues in the use of self-report mothers were discussed earlier. Whi le subjects may have learned more adaptive perceptions in treatment and reported the same, it is possible that actual behaviors might not coincide with subjects' self-report. Addit ional ly , it may be that other methods of assessing social perceptions are more sensitive to potential group differences. For example, researchers have examined perceptual and attributional processes in maltreated indiv iduals using self-report measures of locus of control , moral judgements, family history, and response to hypothetical vignettes. F inal ly , it is possible the obtained results are a true reflection of this interpersonal domain in survivors of childhood abuse and family disruption who are currently in treatment focusing on their childhood experiences. Open-ended Coping Measures Finally, in regards to the remaining class of dependent variables, open-ended coping measures, some group differences were noted between sexual abuse and physical abuse subjects currently in treatment. Sexual abuse subjects appeared to be unique in their report of specific, open-ended coping strategies. The fact that results were obtained using an open-ended format may be a reflection that such a format allows subjects to report the use of very individual ized strategies and also allows them to express strategies in any way desired. In comparison to physical abuse subjects on the neutral tape, sexual abuse subjects were more l ikely to view the situation as non-problematic and therefore to choose to talk and prolong the interaction. Terr (1991) suggested that adults who were sexually maltreated as children may either shrink away from men, or accost them w i t h friendly overtures. The open-ended coping question was designed to address subjects' social judgement and planning. The results suggest that the sexual abuse subjects either d i d not possess a planned coping strategy and/or choose to attempt to handle situations in ways that may be quite different, and potentially less adaptive, than how others who have not experienced sexual abuse w o u l d choose to respond. It may be that female survivors of ch i ldhood sexual abuse are part icular ly l i k e l y to exhibit poor judgement w h e n interacting w i t h men w h o are not o b v i o u s l y abusive or untrustworthy, but who are more subtly dysfunctional. Similarly, Dodge, Bates and Pettit (1990) found that in comparison to non-abused children, abused children were less l ikely to demonstrate good judgement and generate competent solutions to interpersonal problems. Alternatively, the sexual abuse subjects' report of "no problem" in the neutral interaction may reflect the use of an actual coping strategy, characterized by emotionally distancing or dissociating oneself from the interaction. It is interesting to note that sexual abuse subjects were more l ikely than members of the other three groups (although not signficant at the .01 level) to report "no problem" in response to the dating tape. These findings that some abused individuals f ind situations acceptable that others f ind unacceptable, are similar to those of Herzberger and Tenner (1985). They found that college students who had experienced moderate and severe disc ip l inary encounters as chi ldren, were more l ike ly to rate vignettes depicting s imilar encounters as acceptable and appropriate i n comparison to students without such childhood encounters. It may be particularly valuable to examine aspects of social behavior under ambiguous conditions such as that depicted on the neutral audiotape. Maltreated indiv iduals may be more l ikey to display negative attributional biases under ambiguous conditions. When threat/conflict exists in interpersonal situations, it may obscure differences in the responses of survivors of different forms of chi ldhood maltreatment. The conflict might prove to be so overt that most i n d i v i d u a l s w o u l d respond s imi lar ly (e.g., w i t h self-protective behaviors) . However , i n ambiguous situations where there is no overt threat or conflict, aggressively- or negatively-biased views and behaviors might be more l ikely to appear. For example. Dodge and associates (1984) found that, in comparison to popular and average children, socially deviant children could accurately identify hostile intentions but were biased toward infering hostile intent when the intention was actually prosocial or accidental. Thus, group differences appeared in situations that d i d not contain overt conflict. The investigators suggested that such a negative attr ibutional bias might increase the l i k e l i h o o d of aggressive responding. Attr ibt ional style has been found to be strongly related to sexually vict imized women's adult functioning (e.g. G o l d , 1986; Wyatt & N e w c o m b , 1990). This highlights the need to address social judgement and planning in both research and treatment, and to examine attributional styles of survivors of other forms of childhood maltreatment. There also is theoretical support for an emphasis on social judgement and coping behaviors. Organizational/developmental theorists propose that development of a sense of self-competency and mastery of the social and physical environment is a salient developmental task, and suggest that coping behaviors are important mediators between c h i l d h o o d adversity and later adjustment. The lack of more significant results for open-ended coping i n the present study may be a reflection of treatment effects, as detailed for the previous dependent variables. The explanations offered for the social perception data i n regards to variable sensitivity and nature of the interpersonal domain are also applicable for the open-ended coping data. A d d i t i o n a l l y , it may be that women evidence disruptions in more subtle aspects of coping behaviors. A number of researchers have noted that maltreatment is not as strongly related to aggressiveness i n females as in males, and suggest that females may evidence more subtle disruptions in social judgement and coping strategies (e.g., Renken et al. , 1989; Rieder & Cicchetti, 1989; W i d o m , 1989). Recall of Coping in Chi ldhood A d d i t i o n a l analyses examined women's recollections of chi ldhood coping strategies and outcomes. The obtained findings indicate that women who grew up i n abusive, disruptive or relatively stable families may differ substantially in the types of retrospective perceptions of coping responses used dur ing childhood, and in the perceived chi ldhood and adulthood outcomes of these coping strategies. In comparison to women from more stable control families, women from abusive and disruptive families recalled greater reliance on emotionally-focused strategies, such as escapism and minimizat ion, and less reliance on problem-focused and/or social supportive strategies, such as cautiousness and support mobil izat ion. They were more l ikely to perceive that their coping efforts were ineffective when considering both short-term and long-term impact. Such results are distressing i n their implicat ion that children from abusive and disruptive families f ind themselves faced w i t h environmental demands that call for coping strategies beyond the emotional and physica l /developmental resources available to them. Chandler (1982) suggested that adults have many resources to d r a w upon i n coping with stress, whereas options available to children are much more l imited. They are therefore left to rely on coping strategies that appear to be ineffective and even potentially further damaging (e.g., drug/a lcohol and self-abuse as a means of temporary escape). While the whole family environment i n abusive and disruptive homes appears to be generally harmful in regards to recollection of childhood coping, the results of the present study suggest that when adequate comparison groups are used to control for the shared effects of various forms of maltreatment/disruption, some specific differences i n recollection of childhood coping are found to be associated w i t h particular forms of maltreatment. Subjects experiencing sexual a n d / o r physical abuse in chi ldhood reported perceptions of less positive childhood and adulthood coping outcomes and greater reliance on the use of escapism and minimizat ion when compared to subjects from disrupt ive and control families (referring to group means). Further, while subjects i n each of the clinical groups reported recollection of significantly poorer coping and outcomes in comparison to subjects i n the control group, subjects experiencing sexual abuse i n particular recalled the poorest adulthood coping outcomes, the greatest use of escapism and minimization, and the least use of cautiousness. It w o u l d appear that while women who experienced sexual abuse in childhood report recollection of some of the same childhood coping strategies and outcomes as women who experienced other forms of maltreatment, sexual abuse victims display the greatest degree of disruption in childhood coping, as reflected in adult recall. Newberger and De Vos (1988) stated that knowing more about the meaning a c h i l d makes of her maltreatment and fami ly experiences is cr i t ical for understanding how the experiences affect the chi ld . It may be unreasonable to expect children from abusive and disruptive home environments to use adaptive coping strategies if they do not perceive that a problem exists or that they have access to adaptive strategies or faith in their ability to implement such strategies effectively. Examination of the subjective comments that subjects wrote in response to the W C C L - R suggest that the option of using more adaptive coping strategies may not be available to children in dysfunctional homes. Subjects reported that: they often received unhelpful and blaming responses from adults they turned to for help: they could not console themselves by thinking that "things could be worse" because they d i d not believe that things could be worse; attempts to express their feelings were met w i t h further abuse; they could not "maintain their pride" because they d i d not perceive that they had any pride; they could not accept sympathy and support from others, because there was no-one to offer it; they perceived their situation as so overwhelming and punishing that they could not rely on seeking meaning i n the situation or on prayer; and they could not remove themselves from the situation because, as children, there was nowhere safe they could go. Such subjective findings highlight the need to adopt a developmental and environmental perspective when studying the impact of childhood maltreatment (Compas, 1987). It is important to note that the W C C L - R was given out of interest only, in an attempt to generate ideas for future studies, particularly those i n the area of potentiating and compensatory mediating factors. Aber and colleagues (Aber et al., 1989) referred to organizational/ developmental principles to describe chi ldhood coping responses in maltreated children as adaptations to certain types of chi ld-rearing environments. For example, Rieder and Cicchetti (1989) proposed that the hypervigilance and ready assimilation of aggressive st imuli noted i n maltreated children may have initially developed as a coping strategy, alerting the child to signs of danger and keeping affect from increasing to a level that might interfere with coping attempts. They pointed out, however, that such strategies may eventually become engrained and maladaptive in more average interactions. The use of prospect ive designs w o u l d contr ibute va lua b le i n f o r m a t i o n as to the continuity/discontinuity of the use of such coping strategies throughout the life-span. There are a number of problems inherent wi th the use of retrospective, self-report data. Subjects were asked to focus on coping strategies they used i n childhood. For some of the subjects many years had passed since their chi ldhood and they may have experienced difficulty accurately recalling how they responded as children. Whi le there was no way to estimate the accuracy of the women's recall, one might expect that distortions in recall w o u l d be more l ikely where the women reported adaptive coping strategies. Given that coping strategies such as drug use, suicide attempts, and promiscuity are generally seen as stigmatizing and shameful, over-reporting of such strategies seems unlikely (e.g.., Briere & Za id i , 1989; Egeland et a l . , 1988). Alternatively, subjects' current beliefs about how they coped as children and the outcomes of such coping, may have been biased by what they learned in therapy. Such a finding wou ld still be of interest, as a common goal of therapy is to alter survivors' cognitions and attributions to help patients v i e w their experiences differently and/or to reach a level of acceptance. For example, Carlson and colleagues (Carlson et a l , 1989) noted that adult survivors of malteatment who d i d not repeat the cycle with their o w n children seemed to have better resolved their negative feelings about their maltreatment (e.g, mourned the loss of adequate parenting as a child). It is also important to consider that different individuals and different groups of individuals (e.g., sexual abuse survivors, family disruption survivors) may have referred to quite different chi ldhood situations when reporting coping strategies. One could assume that subjects in different groups were coping w i t h qualitatively different types of situations as wel l . This situational variability makes it difficult to determine whether differences (or lack of differences) between groups are due to the coping strategies the subjects used or to the circumstances wi th which subjects had to cope. Further research in this area would be helpful in clarifying these questions. It is desirable to delineate whether adult recall truly reflects that different forms of maltreatment increase the l ike l ihood of an individual ' s use of specific coping strategies, or whether recall is more reflective of individual characteristics that serve to bias which strategies are recalled. It may be that an indiv idual ' s personal construction of their maltreatment and their attempts to cope are more predictive of maladaptive outcomes than objective environmental events (Harter et al . , 1988). It is necessary to use longitudinal designs to investigate this issue more directly, and also to empirically demonstrate the reliability and validity of the use of the W C C L - R wi th maltreated individuals . A Comparison of Pilot and Central Study Results The pilot study that was prepared to guide the central study also addressed the association of childhood adversity wi th adult interpersonal functioning. The pilot study focused on individuals from sexually, physically and emotionally maltreating environments and a control population. In contrast to findings in the central study. pilot study results were more strongly suggestive that exposure to maltreating environments is related to self-report of more adult interpersonal problems than that reported by control subjects. Pi lot study subjects who had experienced c h i l d h o o d maltreatment reported more problems than controls w i t h self-perceptions, dependency and aggressiveness and more frequent use of maladapative coping strategies, based on self-report of real life experiences not i n a neutral laboratory setting. It is important to note that the pilot study used a university student population while the central study used a clinical population. It may be that the significant differences obtained using a university population reflect the fact that students may represent a broader cross-section of the populat ion than clinical subjects and therefore provide an indication of survivor funct ioning across a broader range of psychological adjustment. M a n y of the studies that have found significant differences between the reports of maltreated individuals and controls have relied on university populations (e.g., Briere & Runtz , 1988a, 1988b, 1990; Harter et al. , 1988; Parker & Parker, 1991). Briere and Runtz (1990) have cautioned that the unique relationships observed between types of maltreatment and adult symptomatology may not be found in clinical samples, and emphasized the need for replication with clinical populations. It is beneficial to examine chi ld maltreatment i n different social contexts (e.g., university students, c l inical groups, random representative community samples), as findings that are stable across mult iple studies using multiple contexts have greater generalizability to the total population of survivors of maltreatment (Briere, 1992). Briere emphasized, however, that one needs to be extremely cautious i n making inferences regarding individuals with similar demographic/social status and abuse histories. Alternatively, the pilot study ut i l i zed more self-report measures w h i c h subjects completed anonymously. Perhaps abuse survivors are more w i l l i n g to acknowledge poor interpersonal functioning anonymously than when they are face-to-face wi th an experimenter. Finally, the central study relied upon control group subjects, selected to be demographically similar to women in the cl inical groups. The resulting control group subjects may have therefore shared some of the historical and daily struggles and difficulties characteristic of the clinical group subjects (e.g., poverty, frequent moves, stressful working conditions), thereby obscuring potential group differences. In contrast, one might suggest that the pilot study control group subjects were more distinct in their historical and dai ly situations and stresses i n comparison to the abuse groups subjects in the pilot study. It w o u l d be informative to empirically investigate such hypotheses, particularly i n l ight of theoretical speculations that there may be generally harmful family environments and that general family variables may have a pivotal impact on children's response to maltreatment and on long-term outcome (e.g., Beitchman et al. , 1992; Fantuzzo, 1990; Jaffe et al . , 1986). Therefore, researchers need to take into account children's total experience of victimization and stress from a variety of sources. When the sexual and physical abuse survivors are compared, no significant differences i n reported coping strategies were obtained in the pilot study whereas the two groups significantly differed in the central study in response to the neutral tape. Futhermore, whereas the sexual abuse subjects displayed greater overall dysfunction than the physical abuse subjects in the central study when one examines the tables containing group means, the opposite result was obtained in the pi lot study. One might hypothesize that these different findings across studies is a reflection of the difference between the members of the sexual abuse subjects in the two studies. W h i l e many of the sexual abuse pilot subjects reported experiencing one-time incidents of abuse, with only 27% of the offenders being within the subject's family circle, the sexual abuse subjects i n the central study reported sexually abusive experiences lasting from six months to 16 years w i t h approximately 65% of the offenders being within the subject's family circle. These differences i n duration, frequency and degree of family involvement between the two studies may have resulted in the obtained differences. There is theoretical and empirical support for this suggestion. If chi ld maltreatment is hypothesized to negatively impact the development of internal models of the self and others, it follows that maltreatment of greater severity w i l l have greater impact. M a n y researchers have found that the severity of women's sexual abuse in childhood directly influenced outcome (e.g., Briere & Runtz, 1988a; Browne & Finkelhor, 1986; Gelinas, 1983; Wyatt & Newcomb, 1990). General Methodological Issues The difficulties encountered i n the construction of methodologically sound studies are repeatedly highlighted i n literature in the area of ch i ld abuse. It is essential for the accurate interpretation of research to control potential confounding variables that might account for obtained results. In order to ensure that differential group response to the experimental manipulation was not merely a reflection of demographic differences between the groups, the four groups i n the present study were balanced across a number of demographic variables. Subjects d i d not significantly differ in marital status, age, occupation or ethnicity. One might also hypothesize that differential response reflects differential treatment exposure such that subjects w h o have more experience in treatment are more able to adaptively respond to the audiotaped social interactions. This potential confound was controlled i n the present study, such that there was no significant difference i n treatment exposure between groups. Members of the three clinical groups also d i d not differ in therapist-rated treatment improvement. Final ly, it could be argued that subjects who are more adjusted in general w i l l display more adaptive coping. Therapists i n the present study rated their clients on general measures of adjustment and no significant group differences were obtained. Addi t ional ly , the vast majority of subjects, independent of clinical group membership, were rated as being relatively poorly adjusted. F ina l ly , it is possible that the subjects i n each group were not equally representative of the total population they were drawn from. If the base rates for adjustment differed across groups, then the subjects referred to the study for each group may have been differentially representative of the total population for their group. For example, given that the subjects were noted to be relatively poorly adjusted, a referral bias would exist if one of the group populations contained a number of more adequately adjusted individuals who were therefore not adequately represented i n the subject pool. I attempted to obtain information from therapists about the number and nature of potential subjects who did not participate. However , therapist compliance was poor for this task, and therapsits were unable or unwi l l ing to provide this data. It is recommended that attempts be made to obtain such information in future research in this area. W h i l e every attempt was made to compare the experimental groups on relevant demographic and treatment variables, as Briere (1992) recommended, it is not always clear on which variables subjects should be matched. It is possible that some unmonitored, but relevant, variables other than maltreatment experiences still discriminate the groups. There is literature to suggest that it is diff icult to isolate different forms of abuse given the high degree of overlap. Henderson (1983) pointed out that it also is difficult to separate the effects of specific forms of abuse (e.g., sexual and physical abuse) from those of other variables that are highly correlated w i t h developmenta l t rauma, such as h i g h degree of f a m i l y disorganization. Wolfe and Mosk (1983) hypothesized that disturbances i n abused children's development are more a function of family events and interaction patterns shared by a wide range of distressed families than of isolated abusive episodes. However , other researchers (e.g.. Dodge et al . , 1990) note that when samples are representative and family ecological factors (e.g., poverty, marital violence, family instability) and child biological variables (e.g., early health problems and temperament) are adequately addressed in the subject groups, physical child abuse is a risk factor for the development of aggressive behavior. These results highlight the need for meaningful control groups and detailed descriptions of subject characteristics. While the subjects in the sexual and physical abuse groups i n the present study d i d share characteristics of disrupted and chaotic family relationships, they also possessed the additive experiences of sexual and physical abuse. It is unlikely to expect that children who experience sexual and physical abuse w i l l not experience other aspects of family disruption, particularly given the inter-relatedness of these aspects of developmental traumas (Briere, 1992; Henderson, 1983). Rather, it may be more accurate to speak of degrees of disruptive experiences during childhood. The subjects in the current study might be conceptualized as having experienced the f o l l o w i n g degrees of disrupt ion: no d i s rupt ion ; disordered, chaotic family relationships; disordered family relationships and physical abuse; and disordered, chaotic family relationships, aspects of physical abuse, and sexual abuse. One can argue that this is a more clinically val id expression of the experiences of survivors of chi ld abuse. The question that follows is whether differences between groups are quantitatively or qualitatively different (e.g., Dietrich, Starr, Jr., & Weisfeld, 1983). This question is not directly addressed in the present study as it w o u l d be extremely diff icult to determine whether increasing degrees of disruption have cumulative effects or whether the whole is greater than the sum of its parts. It may be that different disruptive factors combine to create an abusive experience that is uniquely different from the experience of any of the contributing factors indiv idual ly . Thus, chi ld (and adult) adjustment problems may not be compounded by the presence of any one famil ial stressor in isolation, but by interactions of an accumulation of famil ia l stressors such as poverty, interparental violence, substance abuse, illness and isolation, in combination with minimal compensatory factors (e.g., Fantuzzo et al . , 1991; M a s h & Wolfe , 1991). A d o p t i o n of such a theoretical model of developmental psychopathology underscores the value of prospective designs using multisource, mult imethod measurement of potentiating familial stressors, quality of parental care, compensatory factors, and outcome. In any scientific research, it is important to demonstrate the val idi ty and strength of experimental manipulations. The central manipulation in the present study i n v o l v e d the exposure to audiotaped social interactions which varied in content and hypothesized impact. Across numerous dependent variables, a strong audiotape effect was noted. The neutral tape was specifically designed to be less threatening a n d / o r arousing than the two clinical tapes. This intended differential impact was supported, thereby helping to establish the validity of the manipulation. Subjects consistently reported less arousal and perceived threat i n response to the neutral/control interaction than to the conflictual and dating interactions. It also is important to establish that differential impact is not a reflection of tape presentation. Therefore, the presentation was randomized and each tape fell equally into the first, second, and third placements. Addit ional ly , there was no evidence of tape order or sequence effects. Final ly , as noted i n the methods section, a male speaker was chosen to be more representative of the majority of the womens' experiences; for example, Wolfe and Wolfe (1988) suggested that approximately 97 to 98% of child sexual abuse offenders are male. The conflictual tape d i d not prove to be a highly discriminatory stimulus. In general, the tape appeared to be perceived as too powerful and not ambiguous enough. It may be that one sees more clear response biases when decisions have to be made about ambiguous situations versus clear-cut situations. If the conflictual tape was seen as clearly negative and assertiveness-evoking by the vast majority of subjects, this w o u l d over-shadow subtle group differences. S u m m a r y The central question investigated in the present study was whether different types of childhood stressors produce different adaptations (styles of coping with the different stressors), and whether coping styles are reflected in adult adjustment, in a group of already identif ied dysfunctional ind iv idua ls w h o were being seen in treatment wherein they were addressing unresolved feelings about their childhood experiences. The results indicate that abusive experiences are related to greater dysfunct ion i n regards to social information processing, as reflected i n self-perceptions i n response to da t ing interact ions, than does exposure to disruptive/chaotic family environments alone. However , the differences generally appear to be variable-specific and few differences were obtained on physiological and coping variables. Turning to the predictions fo l lowing from a Specificity Position, the results indicate that some specific problems in adult interpersonal functioning are evidenced between the sexual and physical abuse groups in neutral situations. Specifically, in comparison to physical abuse survivors in treatment, sexual abuse survivors in treatment reported significantly greater use of different, and perhaps less adaptive, coping strategies in neutral situations. Overa l l , however, few significant group differences were obtained which lends support to a Generalist Position; the extent and nature of adult dysfunction observed in abuse survivors does not depend on the exact nature of the abuse. Alexander (1992) suggested that it may be more product ive to classify maltreated i n d i v i d u a l s according to the important attachment relationships concurrent w i t h the maltreatment, given the overlap of symptoms between forms of maltreatment and the lack of sets of symptoms. Organizat ional /developmental theorists speculate that cessation of abuse does not automatically alter the internal working models of relationships that individuals develop as children. M a i n and G o l d w y n (cited i n Mash, 1991) discussed the development of an interview for the assessment of the continuity of attachment patterns into a d u l t h o o d . W h i l e such measures of adul t perceptions of family/attachment experiences may have some predictive val idity, they should be used cautiously as they may not be accurate representations of chi ldhood experiences (Mash, 1991). Addi t iona l ly , it may be that different forms of child maltreatment a n d / o r attachment relationships represent different pathways to a common post-abuse adult symptom complex (Briere & Runtz, 1988b). It is important to note that there was an overlap of physical and sexual abuse in the sexual abuse group used in this study. While such overlap is not surprising given the interrelatedness of different forms of abuse, it may have obscured group differences. Comparison within the sexual abuse group of those subjects with sexual and physical abuse versus those wi th sexual abuse only might reveal important within-group differences. Given the relatively small number of women in the sexual abuse group who d i d not experience concurrent physical abuse, the feasibility of such comparisons is somewhat restricted. A d d i t i o n a l l y , in order to draw any conclusions from such a comparison, it w o u l d be necessary to ascertain that the level of sexual abuse was similar for the subjects in the sexual-abuse-only and sexual-with-physical-abuse groups. However, future research involving such comparisons is desirable, particularly in light of findings such as those of Dietrich and associates (Dietrich et a l . , 1983) that children w h o have experienced mult iple forms of maltreatment are qualitatively different from those who have experienced only one form of maltreatment. F inal ly , i n regards to whether women seeking treatment for sexual abuse, physical abuse and family disruption differ significantly from non-clinical controls, no significant differences were obtained on any of the dependent measures discussed above. This is an interesting f inding, which may be an indication that when appropriately selected, demographical ly and socio-economically s imilar controls are used, few significant differences are found between w o m e n wi th abusive and disruptive family histories and women without such histories. Starr Jr. and associates (Starr Jr. et a l , 1991) emphasized that research focusing on the impact of maltreatment needs to differentiate the specific effects of vary ing forms of maltreatment from potentially confounding associated factors, such as poverty, unemployment, chaotic neighborhoods, frequent moves, health concerns, and lack of resources. For example, after accounting for family stress variables, Wolfe and Mosk (1983) found no significant differences in child behavior problems of children from abusive versus distressed families. Similarly, Trickett and associates (Trickett et a l . , 1991) noted that for the lowest socioeconomic status families i n their investigation, the differences between the abusive and non-abusive families were basically nondiscernable. It is possible, therefore, that children's general experiences in these two types of families are quite similar, leading to similar developmental sequelae. This hypothesis could best be tested in prospective longitudinal studies, focusing on familial climate and stressors. Al ternat ively , it may be that the treatment experiences of women from abus ive/disrupt ive families bring them to a level of interpersonal funct ioning resembling that of women from demographically and socio-economically similar family backgrounds. A s discussed prev ious ly , treatment may function as a compensatory mediating factor. If treatment exposure serves to alter maltreated i n d i v i d u a l s ' internal w o r k i n g models, then one might anticipate changes in developmental trajectories so as to get explainable discont inui ty (Belsy & Nezworski , 1988). In order to empirically evaluate the role of treatment exposure, it w o u l d be informative to include a control group of women from s imi lar ly abusive/disruptive families who have never received treatment. However , there may be some logistical restraints to the inclusion of such a group, as women who have not sought treatment for their chi ldhood experiences also may be reluctant to participate i n research which would entail the discussion of experiences they have tried to deny or forget. Overal l , few significant (i.e., .01 level of significance) group differences were obtained i n the current study. When one examines the obtained means across the dependent variables, some trends w h i c h are in hypothesized directions are observable. It may be that variables and stimulus materials other than those used in this study may reflect group differences. For example, the present study employed a male speaker on the audiotapes. Subtle differences may have been observed if both a male and a female speaker were used. Addit ional ly , the study was conducted i n a controlled laboratory setting; subjects interacting in real-life interpersonal situations might display differential response i n regard to coping strategies and/or physiological indices. The ethical constraints on the use of more stressful real-life situations were previously discussed. It is also possible that a referral bias existed. If therapists referred only their better adjusted clients, then the lack of group differences may be understandable. However, as discussed previously, the vast majority of subjects were rated by therapists as being relatively poorly adjusted. Alternatively, it may be that the results are a true reflection of the measured interpersonal domains i n adult survivors of abuse currently in treatment. It may be that there are other important aspects of interpersonal functioning and/or different mediating variables. C h i l d maltreatment is an area where research developments and increasing knowledge are burgeoning. Thus, the dependent variables chosen at the time of inception of the present study may not remain those that are preferred in the research climate of the 1990's. For example, the role of compensatory mediating factors has received much attention in more recent literature. Cicchetti and Braunwald (1984) emphasized that there are many factors that may mediate between early and later adaptation and maladaptation that may al low alternative outcomes to occur. There is consensus that the impact of child maltreatment is not universal, and that some children present without symptoms, or show signs of recovery (e.g., Finkelhor, 1990; Wolfe et al . , 1989). Numerous researchers have pointed to the role that victims' support systems play in decreasing the impact of maltreatment (e.g.. Conte & Schuerman, 1987; Egeland et al. , 1988; Mrazek & Mrazek, 1987; Mueller & Silverman, 1989; Wolfe et al. , 1985; Youngblade & Belsky, 1989). It may be important to measure the nature of support systems not only in chi ldhood, but also i n adulthood. Research is beginning to demonstrate that the presence of emotionally supportive friendships in adulthood is related to more adaptive functioning i n adult survivors of maltreatment (e.g., Egeland et al. , 1988; M a r t i n & Elmer, 1992; Harter et al . , 1988; Wolfe, 1987). The quaUty of adult relationships appears important to consider. It w o u l d be beneficial to obtain informat ion regarding maltreated adults' dai ly interpersonal interactions to determine if patterns are discernable (e.g., stable/warm, avoidant/distant, conflict-ridden). For example, it is theoretically consistent w i t h organizational models (which propose that maltreated individuals develop a v iew of others and the world as unsafe) to suggest that maltreated adults may rely on avoidance and withdrawal to attempt to alleviate distress during social interactions perceived to be difficult or threatening (e.g., Emery, 1989; Wolfe, 1988). Examination of both past and current family stressors may be a key area in the study of mediating variables. The importance of family climate has increasingly been addressed in recent research. Factors such as economic disadvantage, d i s rupt ions i n f a m i l y e q u i l i b r i u m , employment stressors, environmenta l disorganization, health and nutrit ion deficits, and social isolation have al l been noted to have their o w n negative sequelae (Fantuzzo, 1990; Lyons-Ruth et al. , 1989; Kaufman & Cicchetti, 1989; Owings West & Prinz, 1987; Parker & Parker, 1991; Pianta et al. , 1986; Strean, 1988; Wolfe & Jaffe, 1991). Thus, maltreatment researchers might profitably focus on more general family background characteristics and on other events in childhood as they affect current interpersonal experiences and self-esteem (Parker & Parker, 1991). Conclusions and Directions for Future Research Researchers i n the area of maltreatment have begun to emphasize that there is a need for theoretical models recognizing that the connections between childhood family experiences and later developmental outcomes are the result of mult iple interactive factors. The lack of comprehensive, integrative conceptual frameworks across investigations has been frequently noted (e.g., Aber et al. , 1989; Cicchetti et al., 1989; Crit tenden & Ainsworth , 1989; Starr Jr. et al . , 1991). The organizat ional/ deve lopmenta l perspective encompasses such a mul t i fac tor ia l , ecologica l , transactional approach to maltreatment. Maltreatment is seen as resulting from ongoing transactions among different factors and ways in which the factors reciprocally influence each other to lead to disturbances in caregiving (e.g., Bowlby, 1988; Crittenden & Ainsworth, 1989; Cicchetti, 1989). Wolfe and McGee (1991) noted that this emphasis on mul t i fac tor ia l and interact ive processes calls for methodologies aimed at detecting developmental differences appearing among maltreated i n d i v i d u a l s . Prospective l o n g i t u d i n a l studies w h i c h emphasize deve lopment /organizat ion throughout the l i fe-span and which use develop-mentally sensitive outcome measures are necessary to further our understanding of indiv idual development in relation to child maltreatment (Starr Jr. et al. , 1991). A number of methodological considerations w i l l need to be addressed in such prospective studies. There is a great need for better operationalization of the constructs and groups of interest, standardization of the measures of constructs, and use of nneasures that are based on more focused theoretical models of behavior (Mash, 1991). There also is a need to adopt multisource-multimethod approaches, where numerous aspects of maltreatment and indiv idua l functioning are assessed across varying contexts. A s Briere (1992) emphasized, the development of such conceptual ly and methodologica l ly sophist icated studies w i l l make great contributions to attempts to disentangle the antecedents, correlates, and impacts of child maltreatment. The vast majority of the women in the present study spoke of verbal abuse and condemnation that was extensive both i n content and duration. Women's subjective comments often reflected their perceptions that this emotional abuse (i.e., being verbally denegrated, berated a n d / o r ignored and isolated) was the most destructive aspect of their abusive and chaotic families-of-origin. Many stated that while the physical and/or sexual abuse was periodic in nature, the emotional abuse was constant. Garbarino and Gi l l iam (1980) proposed that psychological abuse and neglect are at the heart of the overall maltreatment problem, in that victims' w o r l d views become dominated by negative feelings and self-defeating styles of relating w i t h people. They further suggested that such a w o r l d v iew impairs the development of communicat ion ski l ls , patience, moderate goal-setting, and empathy. There is a great need for empirical research documenting the relationship between emotional abuse and developmental damage, given the suggestion that emotional abuse is not only more prevalent, but also potentially more destructive than other forms of abuse (Garbarino et al . , 1986; Hart & Brassard, 1987; McGee & Wolfe, 1991). Future research concerning temperamental traits, treatment experience, life experience and contact wi th positive "parent surrogates" may help further our understanding of the mult idimensional factors involved i n long-term outcomes. Researchers need to consider the role of perinatal problems and developmental/ learning problems, as wel l as the fit between a child's temperamental style and capabilities and the environmental expectations and demands. A s Chess and Thomas (1984) pointed out, studies should examine not on ly i n d i v i d u a l development, but the individual wi thin a specific family. As discussed earlier, it is also important to address whether victims were able to develop good and lasting relationships as children, whether inside or outside the nuclear family. This, in part, relates to the factor of social isolation from support systems wherein family members are cut off from feedback and support. Parker and Barnett (1988) found a consistent l ink over time between reports of perceived deficiencies in both maternal care during childhood and in the availability of close supportive relationships during adulthood. Other researchers have also noted that the supportive relationships available to developing children exert a continuing influence on psychological health over time, and ultimately influence adjustment in adulthood (e.g., Benson & Heller , 1987; Egeland et al., 1988; Holahan & Moos, 1987). Some support for these findings were obtained i n our pilot study as well . Newberger and De Vos (1988) recommended that researchers i n the area of invulnerabil i ty adopt a transactional model which acknowledges that influences evolve over time and that changes in one domain may affect changes in all other domains. In light of the present findings and those noted above concerning individual differences in outcome, future research should attempt to clearly define and measure central mediating variables, focusing attention on protective processes a n d / o r mechanisms operating at key turning points i n people's lives (e.g., Carlson et a l . , 1989; Cicchetti & Braunwald, 1984; Rutter, 1987). Research on invulnerable children within abusive and disruptive homes should emphasize the identification of protective factors that account for children's development of coping competence. O u r understanding would also be furthered by the use of prospective longitudinal studies of abused children which examine such protective factors i n adulthood (e.g.. Black, 1991; Briere, 1992; Mart in , 1982). Such a focus on developmental sequelae across the life-span and their relation to mediator variables is an important next step for research in the area of child abuse (Starr Jr. et al., 1991). A s our understanding of the developmental, family background, and stress variables related to adult outcome increases, intervention and treatment resources can be focused on vulnerable children and on the particular conditions that render them vulnerable. Treatment for adult survivors can also be refined i n order to meet the mult idimensional and developmental needs of i n d i v i d u a l clients. As Chess and Thomas (1984) emphasized, preventative and therapeutic interventions can make a difference at all age periods. Aber, J.L., & A l l e n , J.P. (1987). Effects of maltreatment in young children's socioemotional development: A n attachment theory perspective. Developmental Psychology. 23(3), 406-414. Aber, J.L., & A l l e n , J.P., Carlson, V. , & Cicchetti, D. (1989). The effects of maltreatment on development during early chi ldhood: Recent studies and their theoretical, clinical, and policy implications. In D. Cicchetti and V . Carlson (Eds.), C h i l d maltreatment: Theory and research on the causes and consequences of child abuse and neglect (pp. 579-619). Cambridge: Cambridge University Press. Ainsworth, M.D.S . , Blehar, M . , Waters, E., & Wall,S. (1978). Patterns of attachment: A psychological study of the Strange Situation. Hi l lsdale , N.J . : Erlbavmi. A l d w i n , C M . , & Revenson, T .A. (1987). Does coping help? A reexamination of the relation between coping and mental health. Tournai of Personality and Social Psychology, 53, 337-348. Alexander, P.C. (1992). Application of attachment theory to the study of sexual abuse. Tournai of Consulting and Clinical Psychology, 60(2). 185-195. Augoustinos, M . (1987). Developmental effects of chi ld abuse: Recent findings. C h i l d Abuse and Neglect. U, 15-27. Bagley, C , & Ramsay, R. (1985). Chi ld abuse in childhood: Psychosocial outcomes and implications for social work practice. Journal of Social Work - H u m a n Sexuality. 46, 33-43. Barahal, R . M . , Waterman, J., & Mart in , H . P . (1981). The social cognitive development of abused children. Journal of Consult ing and Clinical Psychology, 49,508-516. Bauer, W.D. , & Twentyman, C.T. (1985). Abusing, neglectful, and comparison mothers' responses to child-related and non-child-related stressors. Journal of Consulting and Clinical Psychology, 53, 335-343. Beck, A . (1978). Beck Inventory. Philadelphia, P .A. : Center for Cognitive Therapy. Beitchman, J .H. , Zucker, K.J. , H o o d , J.E., DaCosta, G . A . , A k m a n , D. , & Cassavia, E. (1992). A review of the long-term effects of child sexual abuse. C h i l d Abuse and Neglect, 16,101-118. Belsky, J. (1980). C h i l d maltreatment: A n ecological integration. American Psychologist, 35, 320-325. Belsky, J., & Nezworsk i , T. (1988). Clinical implications of attachment. In J. Belsky and T. Nezworski (Eds.), Clinical implications of attachment (pp. 3-17). Hillsdale: Lawrence Erlbaum Associates, Inc. Benson, C.S. & Heller, K . (1987). Factors in the current adjustment of young adult daughters of alcoholic and problem drinking fathers. Journal of Abnormal Psychology, 96, 305-312. Bergart, A . (1986). Isolation to intimacy: Incest survivors i n group therapy. Social Casework: The Journal of Contemporary Social Work , M a y , 266-275. Berlinger, L. , & Wheeler, J.R. (1988). Treating effects of sexual abuse on children. Journal of Interpersonal Violence, 2(4), 415-434. Black, M . (1991). Longitudinal studies in child maltreatment: Methodological considerations. In R . H . Starr Jr. & D . A . Wolfe (Eds.), The effects of child abuse and neglect: Issues and research (pp. 129-143). N e w York: The Gui l ford Press. Blumberg, M . L . (1974). Psychopathology of the abusing parent. American Journal of Psychotherapy, 28, 21-29. Bousha, D . M . , & Twentyman, C.T. (1984). Mother-child interactional style in abuse and control groups: Naturalistic observations in the home. Journal of Abnormal Psychology, 93(1), 106-114. Bowlby, J. (1988). Developmental psychology comes of age. American Journal of Psychiatry, 145(1), 1-10. Briere, J. (1992). Methodological issues in the study of sexual abuse effects. Journal of Consulting and Clinical Psvchologv, 60(2), 196-203. Briere, J., & Runtz, M . (1986). Suicidal thoughts and behaviors in former sexual abuse victims. Canadian Journal of Behavioral Science, 18, 413-423. Briere, J., & Runtz, M . (1988a). Multivariate correlates of childhood psychological and physical maltreatment among university women. C h i l d Abuse and Neglect, 12,331-341. Briere, J., & Runtz, M . (1988b). Symptomatology associated with childhood sexual victimization in a nonclinical adult sample. C h i l d Abuse and Neglect, 12, 51-59. Briere, J., & Runtz, M . (1990). Differential adult symptomatology associated with three types of child abuse histories. C h i l d Abuse and Neglect, 14, 357-364. Briere, J., & Z a i d i , L .Y. (1989). Sexual abuse histories and sequelae in female psychiatric emergency room patients. American Journal of Psychiatry, 146, 1602-1606. Brown, J .A., & Daniels, R. (1968). Some observations on abusive parents. C h i l d Welfare. 4Z, 89-94. Browne, A . , & Finkelhor, D. (1986). Impact of child sexual abuse: A review of the research. Psychological Bulletin, 99, 66-77. Buss, A . H . (1980). Self-consciousness and social anxiety. San Francisco: W . H . Freeman and Company. Cacioppo, J.C., & Tassinary, L . G . (1990). Inferring psychological significance from physiological signals. American Psychologist, 45, 16-28. Camras, L . , Grow, J.G., & Ribordy, S. (1983). Recognition of emotional expression by abused children. Journal of Clinical C h i l d Psychology, 12, 325-328. Carlson, V . , Cicchetti, D. , Barnett, D. , & Braunwald, K . G . (1989). Finding order in disorganization: Lessons from research on maltreated infants' attachment to their caregivers. In D. Cicchetti and V . Carlson (Eds.), C h i l d maltreatment: Theory and research on the causes and consequences of child abuse and neglect (pp. 494-528). Cambridge: Cambridge University Press. Cassidy, J., & Kobak, R.R. (1988). Avoidance and its relation to other defensive processes. In J. Belsky and T. Nezworski (Eds.), Cl inical Implications of Attachment (pp. 300-323). Hillsdale: Erlbaum. Chambless, D.L . , Caputo, G .C . , Bright, P., & Gallagher, R. (1984). Assessment of fear in agoraphobics: The body sensations questionnaire and the agoraphobia cognitions questionnaire. Journal of Consulting and Cl inical Psychology, 52, 1090 1097. Chandler, L . A . (1982). Chi ldren under stress: Understanding emotional adjustment reactions. Springfield: Charles C. Thomas, Publisher. Chess, S. & Thomas, A . (1984). Origins and evolution of behavior disorders from infancy to early adult life. N e w York: Brunner/Mazel , Inc. Cicchetti, D. (1989). H o w research on child maltreatment has informed the study of child development: Perspectives from developmental psychopathology. In D. Cicchetti and V. Carlson (Eds.), C h i l d maltreatment: Theory and research on the causes and consequences of child abuse and neglect (pp. 377-431). Cambridge: Cambridge University Press. Cicchetti, D. , & Braunwald, K . G . (1984). A n organizational approach to the study of emotional development in maltreated infants. Infant Mental Health Journal, 5(3), 172-183. Cicchetti, D. , & Howes, P .W. (1991). Developmental psychopathology in the context of the family: Illustrations from the study of child maltreatment. Canadian Journal of Behavioral Science, 23(3), 257-281. Cicchetti, D. , Toth, S.L., & Hennessy, K. (1989). Research on the consequences of child maltreatment and its application to educational settings. Topics in Early Chi ldhood Special Education, 9(2), 33-55. Cole, P . M . , & Putman, F.W. (1992). Effect of incest on self and social functioning: A developmental psychopathology perspective. Tournai of Consult ing and Clinical Psychology. 60(2), 174-184. Compas, B.E. (1987). Coping with stress during childhood and adolescence. In S. Chess and A . Thomas (Eds.), Annual progress in child psychiatry and child development. N e w York: Brunner /Mazel , Inc. Conaway, L.P. & Hansen, D.J. (1989). Social behavior of physically abused and neglected children: A critical review. Clinical Psychology Review. 9, 627-652. Conte, J.R. (1985). The impact of sexual abuse on children. Paper presented at the National Conference on C h i l d Abuse and Neglect, Chicago. Conte, J.R., & Schuerman, J.R. (1987). Factors associated with an increased impact of child sexual abuse. C h i l d Abuse and Neglect, 11, 201-211. Coyne, J.C., & Lazarus, R.S. (1980). Cognitive style, stress perception, and coping. In I.L. Kutash & L.B., Schlesinger (Eds.), Handbook on stress and anxiety. San Francisco: Jossey Bass Publishers. Crittenden, P . M . (1981). Abusing, neglecting, problematic, and adequate dyads: Differentiating by patterns of interaction. Merril l-Palmer Quarterly, 9(2), 201-218. Crittenden, P . M . (1985a). Maltreated infants: Vulnerability and resilience. Tournai of C h i l d Psvchologv and Psychiatry, 26, 85-96. Crittenden, P . M . (1985b). Social networks, quality of child-rearing, and child development. C h i l d Development, 56, 1299-1313. Crittenden, P . M . (1988a). Fannily and dyadic patterns of functioning in maltreating families. In K. Browne, C. Davies, and P. Stratton (Eds.), Early prediction and prevention of child abuse (pp. 161-189). N e w York: Wiley. Crittenden, P . M . (1988b). Relationships at risk. In J. Belsky and T. Nezworski (Eds.), Clinical implications of attachment (pp. 136-174). Hil lsdale , N J : Erlbaum. Crittenden, P . M . , & Ainsworth , M.D.S . (1989). C h i l d maltreatment and attachment theory. In D. Cicchetti and V. Carlson (Eds.), C h i l d maltreatment: Theory and research on the causes and consequences of chi ld abuse and neglect (pp. 432-463). Cambridge: Cambridge University Press. Cronbach, L.J., & Furby, L. (1970). H o w should we measure "change" - or should we? Psychological Bulletin. 74, 68-80. Davis, G.E. , & Leitenberg, H . (1987). Adolescent Sex Offenders. Psychological Bulletin. 101, 417-427. de Jong-Gierveld, J. (1987). Developing and testing a model of loneliness. Tournai of Personality and Social Psychology. 53,119-128. Dietrich, K . N . , Starr Jr., R . H . , & Weisfeld, G.E. (1983). Infant maltreatment: Caretaker infant interaction and developmental consequences at different levels of parenting failure. Pediatrics. 72(4), 532-540. Dodge, K . A . , M u r p h y , R.R., & Buchsbaum, K. (1984). The assessment of intention-cue detection skills in children: Implications for developmental psychology. C h i l d Development. 55, 163-173. Dodge, K . A . , Bates, J.E., & Pettit, G.S. (1990). Mechanisms in the cycle of violence. Science. 250.1678-1683. Donovan, W . L . & Leavitt, L . A . (1989). Maternal self-efficacy and infant attachment: Integrating physiology, perceptions and behavior. C h i l d Development, 60, 460-472. During, S .M. , & M c M a h o n , R.J. (1991). Recognition of emotional facial expressions by abusive mothers and their children. Tournai of Cl inica l C h i l d Psychology, 20(2), 132-139. Egeland, B. (1991a). A longitudinal study of high-risk families: Issues and findings. In R . H . Starr Jr. and D . A . Wolfe (Eds.), The effects of chi ld abuse and neglect: Issues and research (pp. 33-56). N e w York: The Gui l ford Press. Egeland, B. (1991b). From data to definition. Development and Psychopathology, 3, 37-43. Egeland, B., & Sroufe, L . A . (1981). Attachment and early maltreatment. C h i l d Development, 52, 44-52. Egeland, B., Jacobvitz, D . , & Sroufe, L . A . (1988). Breaking the cycle of abuse. Chi ld Development, 59, 1080-1088. Emery, R.E. (1989). Family violence. American Psychologist, 44(2), 321-328. Enos, D . M . , & Handal , P.J. (1986). The relation of parental marital status and perceived family conflict to adjustment in white adolescents. Journal of Consulting and CUnical Psvchology, 54, 820-824. Erickson, M.F . , & Egeland, B. (1987). A developmental view of the psychological consequences of maltreatment. School Psychology Review, 16(2), 156-168. Erickson, M.F . , Egeland, B., & Pianta, R. (1989). The effects of maltreatment on the development of young children. In D. Cicchetti and V. Carlson (Eds.), C h i l d maltreatment: Theory and research on the causes and consequences of child abuse and neglect (pp. 647-684). Cambridge: Cambridge University Press. Erickson, M . F . , Sroufe, L . A . , & Egeland B. (1985). The relationship between quality of attachment and behavior problems in preschool in a high-risk sample. Monographs of the Society for Research in C h i l d Development, 50, 147-166. Fantuzzo, J.W. (1990). Behavioral treatment of victims of child abuse and neglect. Behavior Modificat ion. 14(3), 316-339. Fantuzzo, J.W., DePaola, L . M . , Lambert, L . , Mart ino, T., Anderson, G . , & Sutton, S. (1991). Effects of inter parental violence on the psychological adjustment and competencies of young children. Tournai of Consulting and Cl inica l Psychology, 59(2), 258-265. Fantuzzo, J.W., & Lindquist, C . U . (1989). The effects of conjugal violence on children: A review and analysis of research methodology. Tournai of Family Violence. 4(1), 77-94. Fantuzzo, J.W., & Twentyman, C.T. (1986). C h i l d abuse and psychotherapy research: Merging social concerns and empirical investigations. Professional Psychology Research and Practice. 17(5), 375-380. Farber, E.D. , & Joseph, J.A. (1985). The maltreated adolescent: Patterns of physical abuse. C h i l d Abuse and Neglect. 9. 201-206. Fatout, M.F . (1990). Consequences of abuse on the relationships of children. Families in Society: The Journal of Contemporary H u m a n Services, 71, 76-81. Finkelhor, D . (1980). Risk factors in the sexual victimization of children. C h i l d Abuse and Neglect, 4, 265-273. Finkelhor, D. (1983). Common features of family abuse. In D. Finkelhor, R.J., Celles, G.T. Hotaling, & M . A . Straus (Eds.), The dark side of families: Current family violence research. Beverly H i l l s : Sage Publications, Inc. Finkelhor, D. (1985). Sexual abuse and physical abuse: Some critical differences. In E . H . Newberger & R. Bourne (Eds.), Unhappy families. Littleton: PSG Publishing Company, Inc. Finkelhor, D. (1986). Designing studies on the impact and treatment of child sexual abuse. In D . Finkelhor (Ed.), A sourcebook on child sexual abuse. Beverly Hi l ls : Sage Publications. Finkelhor, D . (1990). Early and long term effects of child sexual abuse: A n update. Professional Psvchologv Research and Practice, 21(5), 325-330. Finkelhor, D. , & Browne, A . (1985). The traumatic impact of child sexual abuse: A conceptualization. American Tournai of Orthopsychiatry, 55, 530-541. Finkelhor, D. , & Hotaling, G.T. (1984). Sexual abuse in the National Incidence Study of C h i l d Abuse and Neglect: A n appraisal. C h i l d Abuse and Neglect. 8, 23-33. Flaherty, J .A., & Richman, J.A. (1986). Effects of childhood relationships on the adult's capacity to form social supports. American Tournai of Psychiatry, 143, 851-855. Folkman, S., & Lazarus, R.S. (1980). A n analysis of coping in a middle-aged community sample. Tournai of Health and Social Behavior, 21, 219-239. Folkman, S., & Lazarus, R.S. (1985). If it changes it must be a process: Study of emotion and coping during three stages of a college examination. Journal of Personality and Social Psychology, 48,150-170. Folkman, S., Lazarus, R.S., Dunkel-Schetter, C , DeLongis, A . , & Gruen, R.J. (1986). Dynamics of a stressful encounter: Cognitive appraisal, coping, and encounter outcomes. Journal of Personality and Social Psychology, 50, 992-1003. Franzoi, S.L., & Brewer, L . C . (1984). The experience of self-awareness and its relation to level of self-consciousness: A n experimental sampling study. Journal of Research in Personality, 18, 522-540. Fromuth, M . E . (1986). The relationship of childhood sexual abuse wi th later psychological and sexual adjustment in a sample of college women. C h i l d Abuse and Neglect, 10, 5-15. Galambos, N . L . , & Dixon, R . A . (1984). Adolescent abuse and the development of personal sense of control. C h i l d Abuse and Neglect, 8, 285-293. Garbarino, J. (1989). Troubled youth, troubled families: The dynamics of adolescent maltreatment. In D. Cicchetti and V. Carlson (Eds.), C h i l d maltreatment: Theory and research on the causes and consequences of child abuse and neglect (pp. 685 -706). Cambridge: Cambridge University Press. Garbarino, J. & Gi l l iam, G . (1980). Understanding abusive families. Lexington: D . C Heath & Company. Garbarino, J., Guttmann, E., & Seeley, J.W. (1986). The psychologically battered child. San Francisco: Jossey-Bass Publishers. Gelinas, D.J. (1983). The persisting negative effects of incest. Psychiatry. 46, 312-332. Celles, R.J. (1973). Chi ld abuse as psychopathology: A sociological critique and reformulation. American Journal of Orthopsychiatry, 43, 611-621. Celles, R.J. (1978). Violence toward children in the United States. American Journal of Orthopsychiatry, 42, 580-592. George, C , & M a i n , M . (1979). Social interactions of young abused children: Approach, avoidance and aggression. C h i l d Development, 50, 306-318. G i l , D . G . (1970). Violence against children: Physical child abuse i n the United States. Cambridge, Mass.: Harvard Univesity Press. Gispert, M . , Davis, M . , Marsh, L . , & Wheeler, K. (1987). Predictive factors i n repeated suicide attempts by adolescents. Hospital and Community Psychiatry, 38, 390-393. G o l d , E.R. (1986). Long-term effects of sexual victimization in childhood: A n attributional approach. Journal of Consulting and Clinical Psychology, 54, 471-475. Goldberg, S. (1991). Recent developments in attachment theory and research. Canadian Journal of Psychiatry, 36, 393-400. Gottman, J .M. , & Katz, L.F. (1989). Effects of marital discord on young children's peer interaction and health. Developmental Psychology, 25(3), 373-381. Graziano, W . G . , Feldesman, A . , &c Rahe, D.F. (1985). Extraversion, social cognition and the salience of aversiveness in social encounters. Journal of Personality and Social Psychology. 49, 971-980. Green, A . H . (1978). Self-destructive behavior in battered children. American Journal of Psychiatry, 135, 579-582. Greenberg, L.S., & Safran, J.D. (1987). Emotion in psychotherapy. N e w York: The Gui l ford Press. Grusec, J.E., & Walters, G .C . (1991). Psychological abuse and childrearing belief systems. In R . H . Starr Jr. and D . A . Wolfe (Eds.), The effects of child abuse and neglect: Issues and research (pp. 186-202). N e w York: The Gui l ford Press. Grych, J .H. , & Fincham, F.D. (1990). Marital conflict and children's adjustment: A cognitive-contextual framework. Psychological Bulletin, 108(2), 267-290. Hart, S .N. , «&: Brassard, M.R. (1987). A major threat to children's mental health: Psychological maltreatment. American Psychologist, 42, 160-165. Harter, S., Alexander, P .C. , & Neimeyer, R . A . (1988). Long-term effects of incestuous child abuse in college women: Social adjustment, social cognition, and family characteristics. Journal of Consulting and Clinical Psychology, 56(1), 5-8. Haynes, S .N. (1991). Clinical applications of psychophysiological assessment: A n introduction and overview. Journal of Consult ing and Clinical Psychology, 3(3), 307-308. Henderson, J. (1983). Is incest harmful? Canadian Journal of Psychiatry, 28, 34-40. Herman, J., Russell, D. , & Trocki, K. (1986). Long-term effects of incestuous abuse in childhood. American Journal of Psychiatry, 143(10), 1293-1296. Herrenkohl, R .C. , Herrenkohl, E.C.,Egolf, B.P., & W u , P. (1991). The developmental consequences of child abuse: The Lehigh longitudinal study. In R . H . Starr Jr. and D . A . Wolfe (Eds.), The effects of child abuse and neglect: Issues and research (pp. 57-81). N e w York: The Guilford Press. Herzberger, S.D., & Tennen, H . (1985). The effects of self-relevance on judgements of moderate and severe disciplinary encounters. Journal of Marriage and the Familv, 47, 311-318. Hjorth, C.W., & Ostrov, E. (1982). The self-image of physically abused adolescents. Tournai of Youth and Adolescence, 11, 71-76. Hoffman-Plotkin, D. , & Twentyman, C.T. (1984). A mult imodal assessment of behavioral and cognitive deficits in abused and neglected preschoolers. C h i l d Development, 55, 794-802. Holahan, C.J. & Moos , R . H . (1987). Risk, resistance and psychological distress: A longitudinal analysis with adults and children. Journal of Abnormal Psychology, 96,3-13. Horowitz , L . M . (1979). O n the cognitive structure of interpersonal problems treated in psychotherapy. Journal of Consulting and Cl inica l Psychology, 47, 5-15. Howes, C , & Eldredge, R. (1985). Responses of abused, neglected and non-maltreated children to the behavior of their peers. Tournai of A p p l i e d Developmental Psychology, 6, 261-270. Howes, C , & Espinosa, M . P . (1985). The consequences of child abuse for the formation of relationships with peers. C h i l d Abuse and Neglect, 9, 397-404. Hughes, H . M . (1986). Research with children in shelters: Implications for cUnical services. Chi ldren Today. 15(2). 21-25. Hunter, R.S., Ki l s t rom, N . , & Loda, F. (1985). Sexually abused children: Identifying masked presentations in a medical setting. C h i l d Abuse and Neglect, 9, 17-25. Inderbitzen-Pisaruk, H . , Shawchuk, C.R., & Hoier, T.S. (1992). Behavioral characteristics of child victims of sexual abuse: A comparison study. Tournai of Clinical C h i l d Psychology, 21(1), 14-19. Jaffe, P., Wolfe, D . A . , Wilson, S., & Zak, L . (1986). Similarities in behavioral and social maladjustment among chi ld victims and witnesses to family violence. American Journal of Orthopsychiatry, 56(1), 142-146. Jaffe, P., Wilson, S., & Wolfe, D . A . (1986). Promoting changes in attitudes and understanding of conflict resolution among child witnesses of family violence. Canadian Journal of Behavioral Science, 18, 356-366. Kamarck, T.W., Jennings, J.R., Debski, T.T., Glickman-Weiss, E., Johnson, P.S., Eddy, M.J . , & Manuck, S.B. (1992). Reliable measures of behaviorally-evoked cardiovascular reactivity from a PC based test battery: Results from student and community samples. Psychophysiology, 29, 17-28. Kaufman, J., & Cicchetti, D. (1989). Effects of maltreatment on school-age children's socioemotional development: Assessments in day-camp settings. Developmental Psychology, 25(4), 516-524. Kaufman, J. & Zigler, E. (1989). The intergenerational transmission of child abuse. In D. Cicchetti and V. Carlson (Eds.), C h i l d maltreatment. Theory and research on the causes and consequences of child abuse and neglect (pp. 129-150). Cambridge: Cambridge University Press. K a z d i n , A . E . , Moser, J., Colbus, D. , & Bell, R. (1985). Depressive symptoms among physically abused and psychiatrically disturbed children. Journal of Abnormal Psychology, 94(3), 298-307. Kempe, R.S., & Ken\pe, C H . (1978). C h i l d abuse. Cambridge: Harvard University Press. K inard , E . M . (1980). Emotional development in physically abused children. American Journal of Orthopsychiatry, 50, 686-696. Kobasa, S . C (1985). Stressful life events, personality, and health: A n inquiry into hardiness. In A . Monat & R.S. Lazarus (Eds.), Stress and coping: A n anthology. Second edition (pp. 174-188). N e w York: Columbia University Press. Koverola, C , Elliot-Faust, D. , & Wolfe, D . A . (1984). Clinical issues in the behavioral treatment of a child abusive mother experiencing multiple life stress. Journal of Clinical C h i l d Psvchology. 13(2), 187-191. LaBarbera, J.D. (1984). Seductive father-daughter relationships and sex roles in women. Sex Roles, 11, 941-951. LaBarbera, J.D., Martin, J.E., & Dozier, J.E. (1980). C h i l d psychiatrists' v iew of father-daughter incest. Chi ld Abuse and Neglect. 4,147-151. Lahey, B.B., Conger, R.D. , Atkeson, B . M . , & Treiber, F .A. (1984). Parenting behavior and emotional status of physically abusive mothers. Journal of Consult ing and Clinical Psychology. 52(6), 1062-1071. Lamphear, V.S. (1985). The impact of maltreatment on children's psychosocial adjustments: A review of the research. C h i l d Abuse and Neglect, 9, 251-263. Lamphear, V . (1986). The psychosocial adjustment of maltreated children: Methodological limitations and guidelines for future research. C h i l d Abuse and Neglect, 10, 63-69. Lazarus, R.S., & Folkman, S. (1984). Stress, appraisal and coping. N e w York: Springer Publishing Company. Lewis, D.O. , Mal louh, C , & Webb, V. (1989). C h i l d abuse, delinquency, and violent criminality. In D. Cicchetti and V. Carlson (Eds.), C h i l d maltreatment: Theory and research on the causes and consequences of child abuse and neglect (pp. 707-721). Cambridge: Cambridge University Press. Lindberg, F . H . , & Distad, L.J. (1985). Post-traumatic stress disorders in women who experienced childhood incest. C h i l d Abuse and Neglect, 9, 329-334. Linden, W. (1985). Cardiovascular response as a function of predisposition, coping behavior and stimulus type. Journal of Psychosomatic Research, 29, 611-620. Linden, W. , & Zimmermann, B. (1984). Comparative accuracy of two new electronic devices for the noninvasive determination of blood pressure. Biofeedback and Self-Regulation, 9, 229-239. Lyons-Ruth, K. , Connell , D.B., & Z o l l , D. (1989). Patterns of maternal behavior among infants at risk for abuse: Relations with infant attachment behavior and infant development at 12 months of age. In D. Cicchetti and V . Carlson (Eds.), C h i l d maltreatment: Theory and research on the causes and consequences of child abuse and neglect (pp. 464-493). Cambridge: Cambridge University Press. M a i n , M . , & George, C . (1985). Responses of abused and disadvantaged toddlers to distress in agemates: A study in the day care setting. Developmental Psychology, 21(3), 407-412. Marascuilo, L . A . , & Levin,J.R. (1983). Multivariate Statistics in the Social Sciences: A Researcher's Guide. Monterey: Brooks/Cle Publishing Company. Lynch, M . , & Cicchetti, D . (1991). Patterns of relatedness i n maltreated and non-maltreated children: Connections among multiple representational models. Development and Psychopathology, 3, 207-226. M a i n , M . , & G o l d w y n , R. (1984). Predicting rejection of her infant from mothers representation of her o w n experiences: A preliminary report. International Tournai of C h i l d Abuse and Neglect, 8, 203-207. M a i n , M . , & Solomon, J. (1986). Discovery of a disorganized disoriented attachment pattern. In T.B. Brazelton & M . W . Yogman (Eds.), Affective development in infancy (pp. 95-124). N o r w o o d , N.J. : Ablex. Manuck, S.B., & Schaefer, D .C . (1978). Stability of individual differences in cardiovascular reactivity. Physiology and Behavior. 21, 675-678. Mart in , H.P . (1982). Abused children - What happens eventually? In K. Gates (Ed.), C h i l d abuse, a community concern (pp.154-169). Scarborough, Ont.: Butterworth and Co. Ltd . , Mart in , J.A., & Elmer, E. (1992). Battered children grown up. A follow-up study of individuals severely maltreated as children. Chi ld Abuse and Neglect, 16, 75-87. Mash, E.J., «& Wolfe, D . A . (1991). Methodological issues i n research on physical child abuse. Cr iminal Tustice and Behavior, 18(1). 8-29. Mash, E.J. (1991). Measurements of parent-child interaction i n studies of child maltreatment. In R . H . Starr Jr. and D . A . Wolfe (Eds.), The effects of child abuse and neglect: Issues and research (pp. 203-256). N e w York: The Gui l ford Press. M c C o r d , J. (1983). A forty year perspective on effects of chi ld abuse and neglect. C h i l d Abuse and Neglect. 7_, 265-270. McCormack, A . , Janus, M . D . , & Burgess, A . (1986). Runaway youths and sexual victimization: Gender differences in an adolescent runaway population. C h i l d Abuse and Neglect, 10, 387-395. McGee, R . A . , & Wolfe, D . A . (1991). Psychological maltreatment: Towards an operational definition. Development and Psychopathology, 3, 3-18. Meisselman, K . C . (1980). Personality characteristics of incest history psychotherapy patients: A research note. Archives of Sexual Behavior, 9, 195-196. Mil ler III, I.W., M o n t i , P . M . , Zwick, W.R., & Stout, R.L. (1980). The socially inadequate patient: Incidence rate, demographic and clinical features, and hospital and posthospital functioning. Journal of Consulting and Clinical Psychology, 48,375-382. Milner , J.S. (1991). Measuring parental personality characteristics and psychopathology in child maltreatment research. In R . H . Starr Jr. and D . A . Wolfe (Eds.), The effects of child abuse and neglect: Issues and research (pp. 164-185). N e w York: The Guil ford Press. Mueller, E., & Silverman, N . (1989). Peer relations in maltreated children. In D . Cicchetti and V . Carlson (Eds.), C h i l d maltreatment: Theory and research on the causes and consequences of child abuse and neglect, (pp. 529-578). Cambridge: Cambridge University Press. Mrazek, P.J., & Mrazek, D . A . (1987). Resihence in child maltreatment victims: A conceptual exploration. C h i l d Abuse and Neglect, H , 357-366. Newberger, C M . & DeVos, C. (1988). Abuse and victimization: A lifespan develop-mental perspective. American Journal of Orthopsychiatry, 58 (4), 505-511. Norman, W.T. (1969). Stability characteristics of the Semantic Differential. In J.G. Snider & C . E . Osgood (Eds.), Semantic Differential Technique: A sourcebook. Chicago: A l d i n e Publishing Company. Norusis, M.J. (1990). SPSS advanced statistics: Users' guide. Release 4. Chicago: SPSS, Inc. Obrist, P .A . (1981). Cardiovascular psychophysiology: A perspective. N e w York: Plenum Press. Osgood, C.E. , Suci, G.J., & Tannenbaum, P . H . (1957). The measurement of meaning. Urbana, III.: University of Illinois Press. Owens, T . H . (1984). Personality traits of female psychotherapy patients with a history of incest: A research note. Journal of Personality Assessment, 48, 606-608. Owings-West, M . , & Prinz, R.J. (1987). Parental alcoholism and childhood psychopathology. Psychological Bulletin, 102(2), 204-218. Parke, R.D. , & Collmer, C .W. (1975). Chi ld abuse: A n interdisciplinary analysis. In E . M . Hetherington (Ed.), Review of child development research: Volume 5. Chicago: University of Chicago Press. Parker, G . , & Barnett, B. (1988). Perceptions of parenting i n childhood and social support in adulthood. American Journal of Psychiatry, 145, 479-482. Parker, G.B., Barnett, E .A. , & Hickie, I.B. (1992). From nature to network: Examiiùng links between perceptions of parenting received in childhood and social bonds in adulthood. American Journal of Psychiatry, 149, 877-885. Parker, S., & Parker, H . (1991). Female victims of child sexual abuse: A d u l t adjustment. Journal of Family Violence. 6(2). 183-197. Pennebaker, J.W. (1985). Traumatic experience and psychosomatic disease: Exploring the roles of behavioural inhibition, obsession, and confiding. Canadian Psychology, 26, 82-95. Pennebaker, J.W., & K l i h r Beall, S. (1986). Confronting a traumatic event: Toward an understanding of inhibition and disease. Tournai of Abnormal Psychology, 95, 274-281. Peplau, L . A . , & Perlman, D. (1982). Perspectives on loneliness. In L . A . Peplau & D. Perlman (Eds.), Loneliness: A sourcebook of current theory, research and therapy (pp. 1-20). N e w York: John Wiley and Sons, Inc. Pianta, R.C., Egeland, B., & Hyatt, A . (1986). Maternal relationship history as an indication of developmental risk. American Tournai of Orthopsychiatry, 56(3), 385-397. Pianta, R.C., Sroufe, L . A . , & Egeland, B. (1989). Continuity and discontinuity in maternal sensitivity at 6, 24, and 42 months in a high-risk sample. C h i l d Development, 60, 481-487. Reid, J.B., Kavanagh, K . , & Baldwin, D.V. (1987). Abusive parents' perception of child problem behavior: A n example of parental bias. Journal of Abnormal C h i l d Psychology, 15(3), 457-466. Reidy, T.J. (1977). The aggressive characteristics of abused and neglected children. Tournai of Cl inical Psychology. 33, 1140-1145. Renken, B., Egeland, B., Marvinney, D. , Mangelsdorf, S., & Sroufe, L . A . (1989). Early childhood antecedents of aggression and passive-withdrawal in early elementary school. Tournai of Personality. 57(2). 257-281. Rieder, C , & Cicchetti, D . (1989). Organizational perspective on cognitive control functioning and cognitive-affective balance i n maltreated children. Developmental Psvchology, 25(3), 382-393. Rohrbeck, C . A . , & Twentyman, C.J. (1986). Mul t imodal assessment of impulsiveness in abusing, neglecting and nonmaltreating mothers and their preschool children. Journal of Consulting and Clinical Psychology, 54(2), 231-236. Rutter, M . (1987). Psychosocial resilience and protective mechanisms. American Tournai of Orthopsvchology, 5Z, 316-331. Scott, R.L. , & Stone, D . A . (1986). M M P I profile constellations in incest families. Tournai of Consult ing and Clilnical Psychology, 54, 364-368. Sgroi, S .M. , Porter, F., & Blick, L. (1985). A conceptual framework for child sexual abuse. In S . M . Sgroi (Eds.), Handbook of clinical intervention in child sexual abuse (pp.9-37). Lexington: D.C. Heath and Company. Silbert, M . H . , & Pines, A . M . (1981). Sexual child abuse as an antecedent to prostitution. C h i l d Abuse and Neglect, 5, 407-411. Smetana, J.G., Ke l ly , M . , & Twentyman, C.T. (1984). Abused, neglected, and nonmaltreated children's conceptions of moral and social-conventional transgressions. C h i l d Development. 55, 277-287. Smetana, J.G., & Kel ly , M . (1989). Social cognition in maltreated children. In D. Cicchetti and V. Carlson (Eds.), Chi ld maltreatment: Theory and research on the causes and consequences of child abuse and neglect (pp. 620-646). Cambridge: Cambridge University Press. Smith, S .M. , & Hansen, R. (1975). Interpersonal relationships and child rearing practices in 214 parents of battered children. British Journal of Psychiatry, 127, 611-621. Spinetta, J.J., & Rigler, D. (1972). The child-abusing parent: A psychological review. Psychological Bulletin. ZZ/ 296-304. Sroufe, L . A . , & Rutter, M . (1984). The domain of developmental psychopathology. C h i l d Development, 55, 17-29. Starr, R . H . , Jr. (1978). The controlled study of the ecology of child abuse and drug abuse. C h i l d Abuse and Neglect. 2.19-28. Starr, R . H . , Jr. (1987). Clinical judgement of abuse-proneness based on parent-child interactions. Chi ld Abuse and Neglect. 11, 87-92. Starr, R . H . , Jr. (1988). Physical abuse of children. In V.B. Van Hasselt, A.S. Bellack, R.L. Morrisson, and M . Hersen (Eds.), Handbook of family violence (pp. 119-155). N e w York: Plenum. Starr, R . H . , Jr., MacLean, D.J., & Keating, D.P. (1991). Life-span developmental outcomes of child maltreatment. In R . H . Starr Jr. and D . A . Wolfe (Eds.), The effects of child abuse and neglect: Issues and research (pp. 1-32). N e w York: The Gui l ford Press. Steele, B.F. (1986). Notes on the lasting effects of early child abuse throughout the life cycle. C h i l d Abuse and Neglect, 10, 383-391. Steele, B.F., & Alexander, H . (1981). Long-term effects of sexual abuse i n childhood. In P. Beezley Mrazek & C H . Kempe (Eds.), Sexually abused children and their families. N e w York: Pergamon Press. Straus, M . A . (1980). Stress and child abuse. In C H . Kempe & R.E. Heifer (Eds.), The battered child 3rd edition (pp.86-103). Chicago: University of Chicago Press. Straus, M . A . , & Celles, R.J. (1986). Societal changes and change in family violence from 1975-1985. Journal of Marriage and the Family, 48, 465-479. Strean, H.S. (1988). Effects of childhood sexual abuse on the psychosocial functioning of adults. Social Work, 33, 465-467. Tabachnick, B .C. , & Fidell , L.S. (1989). Us ing multivariate statistics 2nd E d . N e w York: Harper Collins Publishers, Inc. Tennen, H . , & Herzberger, S. (1985). Ways of Coping scale. In D.J. Keyser & R.C. Sweetland (Eds.), Test critiques: Volume 3. Kansas City, M O : Test Corporation of America . Terr, L . C (1991). Chi ldhood traumas: A n outline and overview. American Journal of Psychiatry, 148,10-20. Trickett, P .K. , Aber, J.L., Carlson, V . , & Cicchetti, D. (1991). Relationship of socio-economic status to the etiology and developmental sequelae of physical child abuse. Developmental Psychology, 27(1), 148-158. Tsai, M . , & Wagner, N . N . (1978). Therapy groups for women sexually molested as children. Archives of Sexual Behavior, 7, 417-427. Vasta, R., & Copitch, P. (1981). Simulating conditions of child abuse i n the laboratory. C h i l d Development, 52. 164-170. Vondra , J.I., Barnett, D. , & Cicchetti, D . (1990). Self-concept, motivation and competence among preschoolers from maltreating and comparison families. C h i l d Abuse and Neglect. 14, 525-540. Walker, E., Downey, G . , & Bergman, A . (1989). The effects of parental psychopathology and maltreatment on child behavior: A test of the diathesis-stress model. C h i l d Development, 60, 15-24. Wasserman, S. (1967). The abused parent of the abused child. Chi ldren . 14, 175-179. Watson, D. , & Friend, R. (1969). Measurement of social-evaluative anxiety. Journal of Consulting and Clinical Psvchology. 33, 448-457. White, S.D., & Straus, M . A . (1981). The impHcations of family violence for rehabilitation strategies. In S. Mart in, L . Sechrest & R. Redner (Eds.), N e w directions in the rehabilitation of criminal offenders. Washington, D . C : National Academy of Sciences. W i d o m , C.S. (1988). Sampling biases and implications for child abuse research. American Journal of Orthopsvchiatry. 58(2), 260-270. W i d o m , C.S. (1989). The cycle of violence. Science, 244. 160-166. Wiggins, J.S. (1979). A psychological taxonomy of trait-descriptive terms: The interpersonal domain. Journal of Personality and Social Psychology, 37, 395-412. Wil l iams, R.B. (1981). Behavior factors in cardiovascular disease: A n update. In J.W. Hurst (Eds.), Update V: The Heart (pp. 219-230). N e w York: M c G r a w - H i l l . Wolfe, D . A . (1985). Child-abusive parents: A n empirical review and analysis. Psychological Bulletin. 97(3), 462-482. Wolfe, D . A . (1987). C h i l d abuse: Implications for child development and psychopathology. Newbury Park: Sage Publications. Wolfe, D . A . (1988). Chi ld abuse and neglect. In E.J. Mash and L . G . Terdal (Eds.), Behavioral assessment of childhood disorders 2nd Edition (pp. 627-669). N e w York: The Gui l ford Press. Wolfe, D . A . , & Bourdeau, P . A . (1987). Current issues in the assessment of abusive and neglectful parent-child relationships. Behavioral Assessment, 9, 271-290. Wolfe, D . A . , Fairbank, J.A., Kel ly , J .A., & Bradlyn, A.S. (1983). C h i l d abusive parents' psychological responses to stressful and non-stressful behavior in children. Behavioral Assessment, 5, 363-371. Wolfe, D . A . , & Jaffe, P. (1991). C h i l d abuse and family violence as determinants of child psychopathology. Canadian Journal of Behavioral Science, 23(3), 282-299. Wolfe, D . A . , Jaffe, P., Wilson, S., «& Zak, L. (1985). Chi ldren of battered women: The relationship of child behavior to family violence and maternal stress. Journal of Consulting and Clinical Psvchologv, 53(5), 657-665. Wolfe, D . A . , & LaRose, L. (1985). A video-simulated stress procedure for abusive parents: Mult i -modal assessment of parental tolerance, arousal and coping. In J. Lutzker (Chair), Chi ld abuse and neglect: Treatment, research and theory. Symposium conducted at the meeting of the Association for Advancement of Behavior Therapy, Houston. Wolfe, D . A . , & McGee, R. (1991). Assessment of emotional status among maltreated children. In R . H . Starr Jr. and D . A . Wolfe (Eds.), The effects of child abuse and neglect: Issues and research (pp. 257-277). N e w York: The Gui l ford Press. Wolfe, D . A . , & Mosk, M . D . (1983). Behavioral comparisons of children from abusive and distressed families. Tournai of Consulting and Cl inica l Psychology, 5 L 702-708. Wolfe, D . A . , Zak, L . , Wilson, S., & Jaffe, P. (1986). C h i l d witnesses to violence between parents: Critical issues in behavioral and social adjustment. Journal of Abnormal C h i l d Psvchologv, U{1), 95-104. Wolfe, V . V . , Gentile, C , & Wolfe, D . A . (1989). The impact of sexual abuse on children: A PTSD formulation. Behavior Therapy, 20, 215-228. Wolfe, V . V . , & Wolfe, D . A . (1988). The sexually abused child. In E.J. Mash and L . G . Terdal (Eds.), Behavioral assessment of childhood disorders 2nd Edition (pp. 670 -714). N e w York: The Gui l ford Press. Wyatt, G.E. & Newcomb, M . (1990). Internal and external mediators of woman's sexual abuse in childhood. Journal of Consulting and Clinical Psychology, 58(6), 758-767. Young, J.E. (1982). Loneliness, depression and cognitive therapy: Theory and application. In L . A . Peplau and D. Perlman (Eds.), LoneUness: A sourcebook of current theory, research and therapy (pp. 379-406). N e w York: John Wiley and Sons, Inc. Youngblade, L . M . , & Belsky, J. (1989). C h i l d maltreatment, infant-parent attachment security, and dysfunctional peer relationships in toddlerhood. Topics i n Early Chi ldhood Education, 9(2), 1-15. ^One of the criteria for inclusion in the sexual abuse group was that the abuse was of six or more months duration. One subject reported experiencing abuse of approximately one- and three-months duration. This subject was included as there was some discrepancy regarding reported duration(s) of abuse and the abuse was severe in nature (ie., multiple perpetrators, h igh frequency of occurrence, family member as perpetrator and perception of the abuse as severe by the victim). -Supplementary Analyses Secondary analyses were conducted to compare each of the clinical groups alone with the control group. These were the comparisons of interest for the self-and other-perceptions, open-ended coping, structured coping, and physiological data for all three of the audiotapes. Tape One: Neutral Content - Supplementary Analyses Self-and other-perception data. The mean self- and other-perception scores were presented in Tables 6 and 7. The sexual abuse subjects significantly differed from the control subjects, F(6,108)=3.22, £= .006 . Univariate planned contrasts revealed that members of the sexual abuse group were more likely than members of the control group to view themselves as being active, F(l,113)=8.05, £=.005, and to view the man i n the interaction as being strong, F(l,113)=8.13), £=.005. Thus, the sexual abuse subjects' judgements about the neutral social situation appear to be quite different than those of the control subjects, and are perhaps maladaptive. The remaining two contrasts between the physical abuse group versus the control group. and the family disruption group versus the control group were not significant. Open-ended coping data. The open-ended coping response frequencies for the four groups were presented in Tables 8. The results of the planned contrasts indicated that none of the group contrasts were significant. Structured coping data. The mean structured coping scores were presented in Table 9. The results of the planned contrasts indicated that none of the group contrasts were significant. Physiological data. The mean summary scores of self-report of physiological response, physiological residual gain scores, and raw physiological response scores for the four groups were presented in Tables 10-14. The results of planned contrasts indicated that none of the group contrasts were significant. Tape Two: Confl ictual/Argumentat ive Content - Supplementary Analyses Self-and other-perception data. The mean self- and other-perception scores were presented i n Tables 15 and 16. None of the group contrasts revealed differences at conventional significance levels. Open-ended coping data. The open-ended coping response frequencies for the four groups were presented in Tables 17. In the analyses using the five basic coping strategies (i.e., assertiveness, physical withdrawal, anger-aggression, compliance, and no problem cited), post hoc statistical group comparisons revealed that only the comparison between the sexual abuse group and the controls reached significance, X ^ O , N=69)=11.54, £= .01 . Visual inspection of the data indicated that, in comparison to the control subjects, sexual abuse subjects reported more frequent use of physical withdrawal and negative emotions and less frequent use of assertive responses. Structured coping data. The mean structured coping scores were presented in Table 18. The results of planned contrasts reflected the fact that none of the group contrasts were significant at the .01 level of significance. Physiological data. The mean summary scores of self-report of physiological response, physiological residual gain scores, and raw physiological response scores for the four groups were presented in Tables 10-14. The results of planned contrasts reflected the fact that none of the group contrasts were significant. Tape Three: Intimate/Dating Content - Supplementary Analyses Self- and other-perception data. The mean self- and other-perception scores were presented i n Tables 19 and 20. The sexual abuse subjects themselves significantly differed from the control subjects, F(6,108)=4.01, p=<.001. Univariate planned contrasts revealed that members of the sexual abuse group were more l i k e l y than members of the control group to v i e w themselves as: weak, F(6,113)=13.80, p.=<.001; and bad, F(l,113)=11.58, p.=<.001. The remaining two contrasts between the physical abuse and control groups, and the family disruption and control groups, were not significant. Open-ended coping data. The open-ended coping response frequencies for the four groups were presented in Tables 21. The results of post hoc statistical group comparisons for the coping strategies indicated that none of the comparisons were significant. Structured coping data. The mean structured coping scores were presented in Table 22. None of the group contrasts revealed significant differences. Physiological data. The mean summary scores of self-report of physiological response, physiological residual gain scores, and raw physiological response scores for the four groups were presented in Tables 10-14. None of the contrasts were significant (e.g., clinical groups versus controls) at the .01 level of significance. W C C L - R -Supplementary Analyses We then looked directly at the differences between the clinical subjects in comparison to the control subjects. In comparison to subjects in the control group, subjects in the sexual abuse group differed significantly in reported coping strategies and outcomes, F(ll,106)=5.69, p.<.001. Specifically, sexual abuse subjects reported: less successful chi ldhood outcomes, F(l,116)=15.98, £< .001 ; less successful adult outcomes, F(l,116)=15.44, g.<.001; greater reliance on the use of "escapism", F(l,116)=14.48, £< .001, and "minimization", F(l/n6)=14.78, £<.001; and less reliance on the use of "cautiousness", F(l,116)=7.52, p.=.01, and "support mobi l iza t ion" , F(1416)=16.61,£<.001. The differences between the physical abuse versus the control subjects were also significant, F(ll,106)=4.07, £< .001. The physical abuse subjects reported: less successful c h i l d h o o d outcomes, F(l,116)=18.49, g.<.001; less successful adult outcomes, F(l,116)=9.52, p_=.003; greater reliance on the use of "escapism", F(l,116)=12.08, p=.001, and "minimization, F(l,116)=8.77, £=.004; and less rehance on the use of "support mobilization", F(l,116)=7.12, £= .01 . F i n a l l y , the control subjects also signif icantly differed from the family disruption subjects, F(ll,106)=2.81, £=.003. The difference was primarily attributable to the f inding that the family disruption subjects reported significantly less reliance on the use of "support mobil ization", F(l,116)=18.68, p<.001. The mean W C C L - R factor scores were presented in Table 23. A P P E N D I C E S Appendix A Pilot Study M e t h o d For the purpose of the study, sexual abuse was defined as sexual body contact (i.e., genital/breast fondling, attempted/completed vag ina l /ana l /ora l penetration) prior to age 18 by someone 5 years older than the victim. If the age difference was less than 5 years, the subject was included in the study if the sexual abuse involved sexual b o d y contact that the respondent d i d not desire or i n v o l v e d coercion/manipulation. Physical abuse was defined as physical assualt (i.e., being hit with a hand or an object, slapped, burned, scalded, bitten, kicked, shaken, choked, dropped or thrown). Emotional abuse was defined as verbal harassment (i.e., constant berat ing, f a u l t - f i n d i n g , t e r ror iz ing , rejecting, c r i t i c i s m , r i d i c u l e , belittlement, humiliat ion). P a r t i c i p a n t s . Two hundred and seventy-three male and female under-graduate psychology students attending the Univers i ty of Br i t ish Columbia participated in the study. The students ranged in age from 17 to 38, with a mean age of 19 (SD = 2.02). The study was explained to the classes as a whole. Students were free to participate or not, and were given research credit for their participation. Procedure. A brief description of the study's rationale, clinical application, and participant requirements was presented to the classes. Questionnaire packages were distributed to those students who wished to participate. They were advised that they could terminate their participation at any time w i t h no effect on course standing, and were assured of the confidentiality and anonymity of their responses. A debriefing form and contact numbers were inc luded in each package. The participants took the forms home and returned them in an unmarked envelope at the fol lowing class period. Measures. Descriptive information about the participants' gender, age, and marital status was obtained on the cover sheets of two abuse-related questionnaires. The first of these questionnaires addressed abuse history, social support, and coping strategies. Respondents were asked to rate their abusive experiences, family dynamics , social support, and coping strategies on 7-point scales w h i c h were anchored w i t h the descriptive phrases "never"(l) and "consistently/always"(7). O p e n - e n d e d probes were also i n c l u d e d where appropriate . The second questionnaire measured situational discomfort/avoidance in interpersonal settings. These abuse-specific measures were designed for the present study. The fol lowing measures were included in the questionnaire package to assess the subjects' present level of functioning: 1) The Beck Depression Inventory (BDI; Beck, 1978), a clinically derived 21-item self-report scale, was used to assess depression. 2) The Social Avoidance and Distress Scale (SADS; Watson & Friend, 1969) was used to assess social/interpersonal interaction style. 3) The Interpersonal Adjective Scales - Revised ( lAS-R; Wiggins , 1979), provided a self-report of self-perception. Respondents are presented w i t h a list of 64 words describing people's personal characteristics and asked to indicate on a 7-point scale how accurately each word described them. 4) The Inventory of Interpersonal Problems (IIP; Horowitz , 1979). The IIP is a 127-item self-report measure of the broad range of difficulties people encounter in interpersonal relationships. Each item is rated on a 5-point Likert scale (0 = not at all ; 4 = extremely). Scores are calculated across major clusters of problem behavior, obtained by cluster analysis. The Vict imization Experiences A number of statistical procedures were appl ied to the data i n order to investigate interpersonal functioning. Eighty-eight of the participants reported experiencing some degree of physical abuse, 33 reported some degree of sexual abuse; and 160 reported some degree (i.e., 2 or greater on a 1 to 7 Likert scale, with 1 representing "never" and 7 representing "consistently/always") of emotional abuse (overlap existed across reports of the varying forms of abuse). For the physical abuse group, the age of victims at the time of onset of the abuse ranged from 1-18 years (X=7, SD=3.63). In 33% of the reports, the abuse lasted for 3 years, and 36% of the cases occurred weekly, 8% daily. In 69% of the reports, the offender was the victim's natural father and in 42% of the reports, the offender was the victim's natural mother. For the sexual abuse group, the age of the vict im at the time of onset ranged from 2-18 years (X=10, SD=5.10). In 44% of the reports, the abuse lasted one day, and in 47% of the reports it occurred only once. In 32% of the reports, the offender was a family friend, in 21% of the reports the offender was a stranger, in 17% of the reports the offender was a relative, and in 10% of the reports the offender was the natural For the emotional abuse group, the age of the v ic t im at the time of onset ranged from 1-18 years (X=9, SD=3.93). In 12% of the reports, the abuse lasted for 10 years, in 42% of the reports occurred weekly, 37% daily. In 54% of the reports the offender was the natural father, and in 41% of the reports, the offender was the natural mother. Analyses The following analyses focus on the physical and sexual abuse groups only, as these were the groups included in the central study. Multivariate analyses of variance were conducted to compare: (a) the abused subjects as a whole wi th the control subjects; and (b) the physical abuse versus sexual abuse versus control subjects, on self-report of interpersonal problems. Univariate F tests were then used to examine the nature of obtained differences. Multivariate analyses of variance also were conducted to compare: (a) the abused subjects as a whole wi th the control subjects; and (b) the physical abuse versus sexual abuse versus control subjects, on reported level of social support and use of coping strategies. One-way analyses of variance and Student-Newman-Keuls post hoc analyses were used to examine the nature of obtained differences. One-way analyses of variance were conducted to investigate whether differences existed between the physical abuse, sexual abuse and control groups on i n d i v i d u a l IIP items endorsed. Addi t iona l ly , Pearson's Correlat ion Coefficients were calculated between items contained in the abuse questionnaire, to investigate Appendix A cont'd, relationships between variables within groups. Results and Discussion The differences between respondents who reported experiencing physical , sexual a n d / o r emotional abuse in childhood in comparison to respondents with no reported history of abuse in regards to report of experience of interpersonal problems (as measured by the IIP) approached significance, F (1,256)=3.31, p=.07. The abuse group as a whole reported significantly less social support and more frequent use of maladaptive coping strategies, when compared to the control group, F(l,221)=4.77, £= .03 . Specifically, abused respondents reported that, as children they less often felt that there were others who cared about them and upon w h o m they could rely for support. Addi t iona l ly , they were more l ike ly than the controls to cope wi th stress by keeping the problems inside themselves, and by becoming verbally aggressive/assaultive. They were less likely to use talking with others or work ing matters through with others as coping strategies. Significant differences were also found among the physical abuse, sexual abuse, and control groups, F(3,749)=1.59, £ = . 0 2 . The results indicated that respondents w i t h a history of physical abuse reported experiencing significantly greater interpersonal problems than controls on the IIP scales reflecting self-worth, and di f f i cul t ies being support ive , be ing too g i v i n g to others, heightened aggressiveness, over-sensitivity, excessive eagerness to please, and dependency on others. In comparison to individuals with a history of sexual abuse, physical abuse victims reported significantly more problems being supportive of others, and perceived themselves as being too giving to others, overly sensitive, too eager to please, and excessively dependent on others. In comparison to controls, physical abuse victims reported significantly less social support and the more frequent use of maladaptive coping strategies. Specifically, physical abuse victims were less l ikely than controls to report feeling that there were adults or children who cared about them during childhood (and were significantly less likely than victims of sexual abuse to report feeling that there were children who cared about them during childhood). They also were less likely than controls to report that they currently believed that there were individuals who cared about them and whom they could turn to for support. Addit ional ly , physical abuse victims reported using significantly more physical and verbal aggression to cope with stress than d i d controls. Victims of sexual abuse were significantly less l ikely than controls to report feeling that there were adults who cared about them i n childhood. N o significant differences were noted in reported coping strategies between sexual abuse victims and physical abuse victims, or between sexual abuse victims and controls. One might hypothesize that this lack of significant findings is a reflection of the non-representativeness of the members of the sexual abuse group when compared to sexual abuse victims typically used in research, who report longer durations and higher frequencies of abuse, and more family involvement. L o o k i n g at endorsement of indiv idual items on the IIP, victims of physical abuse reported significantly more difficulty than victims of sexual abuse on the fol lowing items: "It is hard for me to do what another person wants me to do"; "It is hard for me to be supportive of another person's goals in life"; "I avoid other people too much"; and "I blame myself too much for causing other people's problems". It also should be noted that victims of physical abuse were found to significantly differ from the controls on 23 of the 127 items. For example, victims of physical abuse reported experiencing diff iculty getting along wi th others and accepting another's authority, and felt that they tried to control people too much, were too afraid of others, were too suspicious of others, were too envious and jealous, and wanted to get revenge against people too much. Finally, within the physical abuse group, significant correlations were noted between items on the abuse questionnaire. A report of physical abuse was positively correlated w i t h report of feeling responsible for the onset and continuance of the abuse; partially attributing the abuse to family stress and offender instability; and feeling that other family members (siblings i n particular) also had experienced physical, sexual and/or emotional abuse. W i t h i n the sexual abuse group, a report of sexual abuse was positively correlated with the report of: genital contact; intercourse; physical force and verbal threat; and, feeling that other family members (siblings i n particular) also had experienced sexual abuse, or feeling uncertain as to whether other family members also were abused. A report of sexual abuse was negatively correlated with the report of feeling responsible for the onset of the abuse. It also is interesting to note that across groups, a report of feeling responsible for the onset and/or continuance of the abuse was positively correlated w i t h a report of the use of physical aggression, verbal aggression, and anger as coping responses. Appendix B T H E R A P I S T S C R E E N I N G Q U E S T I O N N A I R E Please read the fo l lowing items, and indicate by circl ing the appropriate category, to what extent you feel each item reflects the current adjustment of this client. 1 ) Able to manage day-to-day demands in occupational, emotional, and social functioning: 1 2 3 4 5 6 7 not at al l consistently 2) A b l e to generate in te rpersona l c o p i n g strategies that are pe rsona l l y and soc ia l ly acceptable/satisfactory across a range of interpersonal situations: 1 not at al l consistently 3) Ab le to establish and maintain trust in interpersonal relationships: 1 2 3 4 5 6 not at al l consistently 4) A b l e to generate pe rsona l l y adap t i ve and cons t ruc t i ve c o p i n g strategies d u r i n g conf l ictual/aggressive interpersonal interactions: 1 not at al l consistently 5) Able to respond in personally adaptive and constructive ways in romant ic/sexual relationships: 1 2 3 4 5 6 7 not at al l consistently This questionnaire addresses chi ldhood experiences of your client. We appreciate that this is complex information and that you may not know the answers to some questions. If so, that is fine - just leave the question blank. 1 ) H o w long has this client been in treatment? (please be specific) approximate # of sessions. months 2 ) In your opinion, how much has this client improved over the course of treatment? 1 2 3 4 5 6 7 not at al l somewhat great ly 3) What is the client's ethnic background? P H Y S I C A L A B U S E 1 ) Has the client ever experienced physical abuse? 1 2 3 4 never occasional ly 2 ) D id the abuse ever result in bruising? 1 2 3 4 never occasional ly 3) D id the abuse every result in cuts/bleeding? 1 2 3 4 never occasional ly 4) D id the abuse result in scarring/permanent damage? 1 2 3 4 never occasional ly 5) Were belts or sticks ever used? 1 2 3 4 never occasional ly 6) Was the client ever kicked with boots/shoes? 1 2 3 4 never occasional ly consistently consistently 6 7 consistently 6 7 consistently consistently consistently 7) Was the client ever burned/cut wi th weapons? (e.g., hangers, knives, cigarettes): 1 2 3 4 5 6 7 never occasional ly consistently 8) H o w old was the client when the abuse first occurred? (Please give age in years) 1 2 3 4 5 6 7 never occasional ly consistently 9 ) For how long d id the abuse occur? (Please give specific # of months & / o r years) 10) O n average, how often d id the abuse occur? 1 2 3 4 5 6 7 month ly week ly d a i l y 11) H o w wel l d id the client know the abuser? 1 2 3 4 5 6 7 not at al l somewhat very much 12) To what extent d id the client trust a n d / o r care about the abuser? 1 2 3 4 5 6 7 not at al l somewhat very much 13) H o w fearful of the abuser was the client? 1 2 3 4 5 6 7 not afraid somewhat very afraid 14) Was the abuser male or female? 15) Approximately how old was the abuser when the abuse began? 16) Who was the abuser? (e.g., mother, stepfather....) 17) How was the abuse disclosed? 18) Were social service agencies ever involved? 1 2 3 4 5 6 7 not at al l somewhat very much In general, how helpful was your client's contact with these agencies? 1 2 3 4 5 not at al l somewhat 6 7 very helpful 20) In general, how supportive and responsive were family members? 1 2 3 4 5 6 7 not at al l somewhat very much 21) To what extent d id the client feel responsible for the start/onset of the abuse? 1 2 3 4 5 6 7 not at al l somewhat very much 22) To what extent d id the client feel responsible for the abuse continuing? 1 2 3 4 5 6 7 not at a l l somewhat very much 23) D id the client feel that there were other things (outside of herself) that contributed to the abuse starting or continuing: 1 2 3 4 5 6 7 not at al l somewhat very much 24) A l l things considered, how severe would you say the abuse was? 1 2 3 4 5 6 7 m i l d moderate severe S E X U A L A B U S E 1 ) Has the client ever experienced sexual abuse? 1 2 3 4 5 6 7 never occasional ly consistently 2) Were verbal threats involved? (e.g., threats of hurt ing others/ te l l ing others/using physical force) 1 2 3 4 5 6 7 never occasional ly consistently 3) Was kissisng or touching of clothed parts of the body ever involved? 1 2 3 4 5 6 7 never occasional ly consistently Was manual touching of unclothed breasts/genitals ever involved? 1 2 3 4 5 never occasional ly 7 consistently 5) Was oral touching of unclothed breasts or genitals ever involved? 1 2 3 4 5 never occasional ly 6) Wou ld the offender be touching the client? 1 2 3 4 5 never occasional ly 7) Wou ld the client be touching the offender? 1 2 3 4 5 never occasional ly 8) Was penetration/intercourse ever involved? 1 2 3 4 5 never occasional ly consistently consistently consistently consistently 9 ) H o w old was the client when the abuse first occurred? (Please give age in years). 10) For how long d id the abuse occur? (Please give specific # of months & / o r years) _ 11 ) O n average, how often d id the abuse occur? 1 2 3 4 5 6 month ly week ly 12) H o w well d id the client know the abuser? 1 2 3 4 5 6 not at al l somewhat 13) To what extent d id the client trust and /o r care about the abuser? 1 2 3 4 5 6 not at al l somewhat 14) H o w fearful of the abuser was the client? 1 2 3 4 5 6 not afraid somewhat 7 daily very wel l very much very afraid 15) Was the abuser male or female? 16) Approximately how old was the abuser when the abuse began?. 17) Who was the abuser? (e.g., natural father, teacher) 18) How the abuse disclosed? 19) Were social service agencies ever involved? 1 2 3 4 5 6 7 not at al l somewhat very much 20) In general, how helpful was your client's contact with these agencies? 1 2 3 4 5 6 7 not at al l somewhat very much 21) In general, how supportive and responsive were family members? 1 2 3 4 5 6 7 not at al l somewhat very much 22) To what extent d id the client feel responsible for the start/onset of the abuse? 1 2 3 4 5 6 7 not at al l somewhat very much 23) To what extent d id the client feel responsible for the abuse continuing? 1 2 3 4 5 6 7 not at al l somewhat very much 24) D i d the client feel that there were other things (outside of herself) that contributed to the abuse starting or continuing? 1 2 3 4 5 6 7 not at al l somewhat 25) A l l things considered, how severe would you say the abuse was? 1 2 3 4 5 m i l d moderate very much 7 severe F A M I L Y D I S R U P T I O N 1) Has the client ever experienced family disruption? 1 2 3 4 5 6 7 never occasional ly consistently 2) A l l things considered, how severe would you say the family disrupt ion was? 1 2 3 4 5 6 m i l d moderate 7 severe 3) Was f ight ing/violence and /o r verbal aggression between the parents involved? 1 2 3 4 5 6 7 never occasional ly consistently 4) Was parental alcohol abuse involved? 1 2 3 4 5 never occasional ly 6 7 consistently 5 ) Were other disruptive factors involved? fl'lease list: 1 2 3 4 never occasional ly consistently Feel free to add any comments or information that you feel wou ld be useful. We are grateful for the time and effort you have given to this project. Appendix C A D V E R T I S E M E N T F O R SUBJECTS Social Judgment Study-W o m e n , aged 19 to 40, are required for a research project associated w i t h the Department of Psychology of the University of British Columbia . This research is looking at social judgments, that is, how people respond to social situations. It w i l l involve approximately 40 to 50 minutes, during which y o u w i l l be asked to make judgments about taped interactions. Y o u w i l l be paid $5 for our participation. If interested or for addi t ional information, contact the Social Judgment Study, University of British Columbia , 228-5581 or 228-2198. Appendix D T R A N S C R I P T S O F A U D I O T A P E S Dating Audiotape: So, what d i d y o u think of the movie? .... Yeah, I thought so too. I really l iked it Yeah, Robert DeNero is one of my favourite actors. Yeah, he's good, l ike he puts on weight, he loses weight for parts, he's amazing, he really takes his craft seriously. So what should we do now? .... What, we don't have to do anything but it's still early. It's really nice out, I mean it's so brisk. Do y o u want to go for a walk? Great, where shall we go? H o w about down by the beach? It's Uke a really short walk from here It's gonna be a little windier d o w n by the water though. A r e you sure you're gonna be warm enough? .... Your hands feel so cold. Do you want m y jacket? .... Okay W e l l , let me put m y arm around y o u , that'll keep us both warm .... Boy, it sure is nice tonight. Look at all the stars .... W e l l , this is a good place to sit, why don't we sit here for a while and just look at the lights Boy, it sure gets cool when the w i n d kicks up , doesn't it? See, if we move closer toether we can share body heat.... I had a geat time tonight. Look, it's getting too cold out. M y place is really nearby, just a short walk. W h y don't we go up and w a r m up. I can put on coffee or a dr ink if y o u want .... Y o u k n o w , you're really pretty .... A drink, definitely a good idea. I love your hair, beautiful hair. Y o u are so gorgeous I'd like to spend moe time with you. Neutral Audiotape: Yeah, well I don't know. I f ind that you have to look around for shoes in this town. That's all there is to do. I tell you, a city like Montreal , boy are there ever shoes there. When y o u walk d o w n the street; what's the main street? St. Catherines, that's right, St. Catherines, and you walk d o w n - actually there's a street right off St. Catherines. It's a street you get off like on the metro .... I can't remember any of the metro stops. L ike , I was only there twice, but I remember the street. I think it's south of St. Catherine's, but you walk around on the street and there are clothing stores and these wonderful places to buy shoes, not cheap, not l ike the shoes you get i n Vancouver, just excellent shoes, and the difference is they want to sell there .... I mean, that's the real difference. The salesmen come out and they're a little pushy at first; it takes a little while to get used to them, but once you see their enthusiasm on their face, I mean they really want to sell and you know what, they know how to do it. I mean, of course, they're all on commission so you're, you know, the sale that you get is really important, but boy when they come up w i t h a big smile on their face and they're immediately starting to match shoes to socks to suits. You walk i n for a pair of socks and f ind yourself wi th a whole new wardrobe just because of this guy's manner .... I mean, they are so good, but you k n o w here, I mean it's really difficult to f ind shoes. It really is .... When I was walking d o w n , there's a place on Robson Street, again I can never remember the name, but they, y o u know they have nice shoes and they look nice and they look expensive. You know, it's that place where they d i d that thing on Cr ime Stoppers, do you remember where all the shoes were stolen because the guys had left them out back; it's that place, it's an Italian place .... Nice looking shoes, a little expensive, but looks are deceiving because you know I bought a pair, in fact these that I'm wearing now, you can see that immediately they started to get discoloured, a little rain on them and I know what you're thinking, .... I d i d put leather protector on them, but still a little rain and they're already turning brown and they're beginning to crack at the toes, and look at that, what whole thread has come off the back and I'm going to have to re-thread these shoes or something H o w embarassing because I d i d pay about $160 and it is a great leather and there is nothing like the smell of leather for the first little whi le when you get it home ... But, what shoddy workmanship. I mean, it says "made i n Italy", but I think, I don't k n o w , it might be, but nothing l ike the shoes y o u buy in Montreal , nothing like those shoes. What do y o u think you're doing wi th that .... Y o u always do that and it drives nne crazy .... I can't believe you still can't figure it out. Do I have to sit you d o w n and explain it to you one more time? Do I have to take y o u and sit you d o w n and tell y o u once again why what you do drives me crazy? Every single day I have to sit y o u d o w n and do this. Do you realize that? I explain it to y o u over and over again. It's l ike I'm talking to a brick wal l Is that it. A r e you a brick wall? A r e you a child? Do I have to treat you like a child? I think I probably do. You're a child. .... I'm tired of having to deal wi th this day i n and day out. I'm sick of it What do y o u mean you're tired? .... O h , I've heard that before y o u know. I've heard "I'm trying". Y o u say that every day. "I'm trying". W e l l , obviusly you're not trying hard enough .... What's wrong with you? Get your act together because no-one else acts this way Y o u know, it's like you're doing it on purpose to make m y life hell .... Y o u f ind that amuzing, do you? You think it's funny. W e l l , I don't happen to f ind it very funny, so take that smile off your face or I'll wipe it off for y o u .... You k n o w , sometimes I could just hit you. Y o u k n o w that? You're absolutely impossible to get along with and sometimes I could just hit you. Appendix E S O C I A L P E R C E P T I O N In\agine yourself interacting w i t h this person. What is your impression of this man? Circle the appropriate number. This man is: C O L D 1 2 3 4 5 6 7 W A R M C O N C E R N E D 1 2 3 4 5 6 7 I N D I F F E R E N T P O W E R L E S S 1 2 3 4 5 6 7 P O W E R F U L I N C O N T R O L 1 2 3 4 5 6 7 H E L P L E S S A N X I O U S 1 2 3 4 5 6 7 C A L M N O N - A N G R Y 1 2 3 4 5 6 7 A N G R Y O U T G O I N G 1 2 3 4 5 6 7 W I T H D R A W N D E T A C H E D 1 2 3 4 5 6 7 I N V O L V E D S A D 1 2 3 4 5 6 7 H A P P Y S T R O N G 1 2 3 4 5 6 7 W E A K P A S S I V E 1 2 3 4 5 6 7 A C T I V E G O O D 1 2 3 4 5 6 7 B A D Appendix F SELF P E R C E P T I O N Imagine yourself interacting wi th this person. H o w does this person make you feel about yourself? Circle the appropriate number. This person makes me feel:: C O L D 1 2 3 4 5 6 7 W A R M C O N C E R N E D 1 2 3 4 5 6 7 I N D I F F E R E N T P O W E R L E S S 1 2 3 4 5 6 7 P O W E R F U L I N C O N T R O L 1 2 3 4 5 6 7 H E L P L E S S A N X I O U S 1 2 3 4 5 6 7 C A L M N O N - A N G R Y 1 2 3 4 5 6 7 A N G R Y O U T G O I N G 1 2 3 4 5 6 7 W I T H D R A W N D E T A C H E D 1 2 3 4 5 6 7 I N V O L V E D S A D 1 2 3 4 5 6 7 H A P P Y S T R O N G 1 2 3 4 5 6 7 W E A K P A S S I V E 1 2 3 4 5 6 7 A C T I V E G O O D 1 2 3 4 5 6 7 B A D Appendix G C O P I N G STRATEGIES What w o u l d y o u do to cope wi th this man? That i s / h o w w o u l d y o u handle this situation? Appendix H S T R U C T U R E D C O P I N G R E S P O N S E To what extent w o u l d you (circle the appropriate number) 1) Emotionally detach yourself from the situation; emotionally withdraw 1 2 3 4 5 6 7 does not apply used a great deal & / o r not used 2) Negotiate and reason w i t h h im 1 2 3 4 5 6 7 does not apply used a great deal & / o r not used 3) Look for friends to discuss this situation wi th 1 2 3 4 5 6 7 does not apply used a great deal & / o r not used 4) Try to understand why he was behaving this way. Try to f ind meaning or make sense of his behavior. 1 2 3 4 5 6 7 does not apply used a great deal & / o r not used 5) Get angry and fight back; stand up and confront h im 1 2 3 4 5 6 7 does not apply used a great deal & / o r not used 6) Go along wi th whatever he said 1 2 3 4 5 6 7 does not apply used a great deal & / o r not used S E L F - R E P O R T M E A S U R E O F P H Y S I O L O G I C A L A R O U S A L Imagine yourself interacting wi th this person. Read the fo l lowing items, and indicate by circ l ing the appropriate category, to what extent you wou ld be experiencing each of the bodi ly sensations. Heart palp i tat ions: 1 2 not at al l Pressure in chest: 1 2 not at al l Numbness in arms or legs: 1 2 not at al l T ingl ing in fingertips: 1 2 not at al l Numbness in another part of your body: 1 2 not at a l l Feel ing short of breath: 1 2 not at al l Dizz iness: 1 2 not at al l Blurred or distorted vision: 1 2 not at al l Nausea : 1 2 not at al l Butterflies in stomach: 1 2 not at al l 7 extreme 7 extreme 7 extreme 7 extreme 7 extreme 7 extreme 7 extreme 7 extreme 7 extreme 7 Knot in stonnach: 1 2 3 not at al l L u m p in throat: 1 2 3 not at al l Wobbly or rubber legs: 1 2 3 not at al l Sweat ing: 1 2 3 not at al l D r y throat: 1 2 3 not at al l Feeling disoriented and confused: 1 2 3 not at al l Feeling disconnected from your body: 1 2 3 not at al l 1 cont'd 4 5 6 7 extreme 4 5 6 7 extreme 4 5 6 7 extreme 4 5 6 7 extreme 4 5 6 7 extreme 4 5 6 7 extreme 4 5 6 7 extreme W A Y S O F C O P I N G (Revised) Please read each item below and indicate, by circl ing the appropriate category, to what extent you used it to cope wi th the stressful events in your chi ldhood/adolescence. Used Used Used N o t some- quite a great used w h a t a bit deal 1. Just concentrated on what I had to do next -the next step. 2. I tried to analyze the problem in order to understand it better. 3. Turned to work or substitute activity to take my mind off things. 4. I felt that time would make a difference -the only thing to do was to wait. 5. Bargained or compromised to get some-thing positive from the situation. 6. I d id something which I didn't think would work, but at least I was doing something. 7. Tried to get the person responsible to change his or her mind. 8. Talked to someone to find out more about the situation. 9. Cr i t ic ized or lectured myself. 10. Tried not to bum my bridges, but leave things open somewhat. 11. Hoped a miracle wou ld happen. 12. Went along wi th fate; sometimes I just have bad luck. 13. Went on as if nothing had happened. 14. I tried to keep my feelings to myself 15. Looked for the silver l in ing, so to speak; tried to look on the bright side of things. 16. Slept more than usual. 0 0 0 0 0 0 0 0 0 2 2 2 2 2 2 2 2 2 3 3 3 3 3 3 3 3 3 N o t used Used some-w h a t Used quite a bit Used a great deal 17. I expressed anger to the person(s) who caused the problem. 0 1 2 3 18. Accepted sympathy and understanding from someone. 0 2 3 19. 1 told myself things that helped me to feel better. 0 2 3 20. I was inspired to do something creative. 0 2 3 21. Tried to forget the whole thing. 0 2 3 22. 1 got professional help. 0 2 3 23. Changed or grew as a person in a good way. 0 2 3 24. I waited to see what wou ld happen before doing anything. 0 1 2 3 25. 1 apologized or d id something to make up. 0 2 3 26. I made a plan of action and fol lowed it. 0 2 3 27. 1 accepted the next best thing to what 1 wanted. 0 2 3 28. 1 let my feelings out somehow. 0 1 2 3 29. Realized I brought the problem on myself. 0 2 3 30. I came out of the experience better than when I went in. 0 1 2 3 31. Talked to someone who could do something concrete about the problem. 0 1 2 3 32. Got away from it for a whi le; tried to rest or take a vacation. 0 1 2 3 33. Tried to make myself feel better by eating, dr inking, smoking, using drugs or medication, etc. 0 1 2 3 34. Took a big chance or d id something very risky. 0 2 3 35. I tried not to act too hastily or fo l low my first hunch. 0 1 2 3 36. Found new faith. 0 1 2 3 37. Maintained my pride and kept a stiff upper l ip. 0 1 2 3 Used Used Used N o t used some-w h a t quite a bit a great deal 38. Rediscovered what is important in life. 0 2 3 39. Changed something so things wou ld turn out alright. 0 2 3 40. Avo ided being wi th people in general. 0 2 3 41. Didn't let it get to me; refused to think too much about it. 0 2 3 42. I asked a relative or friend I respected for advice. 0 2 3 43. Kept others from knowing how bad things were. 0 2 3 44. Made light of the situation; refused to get too serious about it. 0 2 3 45. Talked to someone about how I was feeling. 0 1 2 3 46. Stood my ground and fought for what I wanted. 0 2 3 47. Took it out on other people. 0 2 3 48. Drew on my past experiences; I was in a similar situation before. 0 1 2 3 49. I knew what had to be done, so I doubled my efforts to make things work. 0 2 3 50. Refused to believe that it had happened. 0 2 3 51. I made a promise to myself that things would be different next time. 0 1 2 3 52. Came up with a couple of different solutions to the problem. 0 2 3 53. Accepted it, since nothing could be done. 0 2 3 54. I tried to keep my feelings from interfering wi th other things too much. 0 1 2 3 55. Wished that I could change what had happened or how I felt. 0 2 3 56. I changed something about myself. 0 1 2 3 N o t used Used some-w h a t Used quite a bit Used a great deal 57. I daydreamed or imagined a better time or place than the one I was in. 0 1 2 3 58. Wished that the situation would go away or somehow be over with. 0 1 2 3 59. Had fantasies or wishes about how things might turn out. 0 1 2 3 60. I prayed. 0 2 3 61. I prepared myself for the worst. 0 2 3 62. I went over in my mind what 1 would say or do. 0 1 2 3 63. I thought about how a person I admire would handle this situation and used that as a model. 0 1 2 3 64. I tried to see things from the other person's point of v iew. 0 1 2 3 65. I reminded myself how much worse things could be. 0 1 2 3 66. I jogged or exercised. 0 2 3 67. I tried something entirely different from any of the above. (Please describe) 0 2 3 Appendix K W C C L - R S U P P L E M E N T A R Y ITEMS Aggression Subscale Items: I struck out at the person. I became verbally aggressive and challenging. I felt l ike exploding at the person. I h i t /k i cked some object like a table or a wal l . I said mean/sarcasatic things. Outcome - Historical : What was the outcome of your attempts to cope with these events while they were happening? Circle the appropriate number. 1 2 3 4 5 6 7 made the made the situation worse situation better 1 2 3 4 5 6 7 made others made others treat me worse treat me better 1 I felt worse about myself Outcome - Current: I felt better about myself What is the current outcome of your efforts to cope wi th these events over the years? Circle the appropriate number. 1. I've put the memories of these experiences to rest. I've come to peace wi th these events. 1 2 3 4 5 6 7 not descriptive very descriptive of me at all of me 2. These events continue to interfere wi th m y relationships with others. 1 2 3 4 5 6 7 not descriptive very descriptive of me at all of me 3. I've found meaning in what happened to me. 1 2 3 4 5 6 7 not descriptive very descriptive of me at all of me 4. I continue to wonder about and think about these events. I don't understand them. 1 2 3 4 5 6 7 not descriptive very descriptive of me at all of me 5. I feel I am stronger as a result of what happened to me. 1 2 3 4 5 6 7 not descriptive very descriptive of me at all of me 6. I've decided that my life has been ruined as a result of these events. 1 2 3 4 5 6 7 not descriptive very descriptive of me at all of me T H E R A P I S T R A T I N G O F C L I E N T A D J U S T M E N T Please read the fo l lowing items, and indicate by circling the appropriate category, to what extent y o u feel each item reflects the current adjustment of this client. 1. Able to manage day-to-day demands in occupational, emotional, and social funct ioning. 1 2 3 4 5 6 7 not at all consistently 2. Able to generate interpersonal coping strategies that are personally and socially acceptable/satisfactory across a range of interpersonal situations. 1 2 3 4 5 6 7 not at all consistently 3. Able to establish and maintain trust in interpersonal relationships. 1 2 3 4 5 6 7 not at all consistently 4. Able to generate personally adaptive and constructive coping strategies during conflictual/aggressive interpersonal interactions. 1 2 3 4 5 6 7 not at all consistently 5. Able to respond i n personally adaptive and constructive ways in romantic/sexual relationships. 1 2 3 4 5 6 7 not at all consistently C O N S E N T F O R M I, (first name), consent to participate in the project titled "The relationship between interpersonal funct ioning and chi ldhood experiences" being conducted by Sara Dur ing and Dr. L y n n A l d e n at the Universi ty of British C o l u m b i a Psychology Department (ph. 228-2198). I have been informed of the requirements of m y participation in this study w h i c h involves one session lasting 40 to 50 minutes, and understand I w i l l be paid $5 for my participation whether I complete all of the procedures or not. The purpose of the study is to complete a series of questionnaires regarding my response to three audiotaped social interaction sequences. I w i l l also provide physiological data (heart rate and blood pressure), and w i l l complete a historical coping questionnaire. I am aware that m y identity and information w i l l be kept confidential through the use of numerical coding. There are no risks reasonably to be expected by myself as a result of participation in this project. I understand that I am free to refuse to participate or my withdraw my consent at any time, and that such actions w i l l not affect me in any manner or any present or future service I may receive. I know that I am entitled to make any inquiries concerning the procedures to ensure that I understand them fully. I have received a copy of this consent form, have read and understood it, and agree to participate in this study. Witness Research Participant (first name only) 

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:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0086554/manifest

Comment

Related Items