Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Cognitive processes mediating the effect of expectations on the perception of interpersonal behavior Safran, Jeremy David 1982

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
831-UBC_1982_A1 S34.pdf [ 5.87MB ]
Metadata
JSON: 831-1.0095620.json
JSON-LD: 831-1.0095620-ld.json
RDF/XML (Pretty): 831-1.0095620-rdf.xml
RDF/JSON: 831-1.0095620-rdf.json
Turtle: 831-1.0095620-turtle.txt
N-Triples: 831-1.0095620-rdf-ntriples.txt
Original Record: 831-1.0095620-source.json
Full Text
831-1.0095620-fulltext.txt
Citation
831-1.0095620.ris

Full Text

COGNITIVE PROCESSES MEDIATING THE EFFECT OF EXPECTATIONS ON THE PERCEPTION OF INTERPERSONAL BEHAVIOR by JEREMY DAVID SAFRAN B.A. (Hons.). Simon Fraser University, 1974 M.A., University of B r i t i s h Columbia, 1978 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY i n THE FACULTY OF GRADUATE STUDIES Department oi: Psychology t We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA October, 1982 (g) Jeremy David Safran, 1982 In presenting t h i s thesis i n p a r t i a l f u l f i l m e n t of the requirements for an advanced degree at the University of B r i t i s h Columbia, I agree that the Library s h a l l make i t f r e e l y available for reference and study. I further agree that permission for extensive copying of t h i s thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. I t i s understood that copying or publication of t h i s thesis for f i n a n c i a l gain s h a l l not be allowed without my written permission. Department of (P-S ° S ^ The University of B r i t i s h Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date & c t . T , I ^ 8* ABSTRACT Theorists from diverse psychotherapy t r a d i t i o n s converge i n asserting that maladaptive expectations about int e r a c t i o n s with other people play an important role i n creating and maintaining c l i n i c a l problems. While there i s a consensus that the modification of these dysfunctional expectations i s a v i t a l aspect of e f f e c t i v e therapy, there i s a l s o agreement that these expectations once established, are extremely r e s i s t a n t to change. The present study was conducted to investigate the cognitive mechanisms which mediate the e f f e c t of expectations upon the perception of other people. The objective was to explore the nature of these cognitive processes i n ordinary s o c i a l perception, with the hope of providing c l i n i c i a n s with new in s i g h t s regarding p o t e n t i a l therapeutic interventions. Subjects were given one of two sets of expectations about the interpersonal c h a r a c t e r i s t i c s of a target male actor. They then viewed a videotape of a staged i n t e r a c t i o n between him and a female actor. Subjects were instruc t e d to indicate subjectively s a l i e n t events while observing the videotape, using a modified version of Newtson's (1973) u n i t i z i n g procedure. This was employed as an index of s e l e c t i v e encoding of interpersonal behavior. Following the videotape, subjects were administered a memory recognition test which was designed to d i s t i n g u i s h between s e l e c t i v e memory r e t r i e v a l and s e l e c t i v e memory reconstruction. They then rated the male actor .on a ser i e s of interpersonal adjective scales. - i i -The r e s u l t s confirmed that subjects' impressions of the target person were biased i n a manner which was consistent with th e i r i n i t i a l expectations. Evidence was obtained consistent with the hypothesis that t h i s bias was mediated by the s e l e c t i v e encoding of expectation congruent information. No evidence was obtained f o r the mediating e f f e c t s of eit h e r s e l e c t i v e memory r e t r i e v a l or s e l e c t i v e memory reconstruction. The pot e n t i a l c l i n i c a l implications of these findings are discussed. i i i -TABLE OF CONTENTS Abstract L i s t of tables Acknowledgements Introduction - Dysfunctional expectations and maladaptive behavior Implications f o r psychotherapy Expectations i n perception and cognition - Expectations i n pattern recognition - Selective attention and the encoding of information - Memory - The s e l e c t i v e r e t r i e v a l of information - Selective reconstruction Expectations and the processing of s o c i a l information: Recent empirical investigations Methodological concerns A c r i t i c a l a n a l y s i s of the u n i t i z i n g methodology - What are breakpoints? - At what l e v e l of abstraction does the encoding of s o c i a l information take place? - Perceptual units or salience data? - Tracking subjectively s a l i e n t events The present i n v e s t i g a t i o n - Overview - Developing the expectancy manipulation: T r a i t s as normative conceptual schemata - Pretesting the expectancy manipulation - Developing the videotape vignette - C a l i b r a t i o n of the videotape - Memory recognition test - Interpersonal adjective scales - i v -TABLE OF CONTENTS (Cont'd.) Experimental procedure - Subjects - Practice t r i a l s - Procedures Results - Impression formation - Tracking data - C o r r e l a t i o n a l analyses - Memory data - True memory items - False memory items Discussion - General discussion - Selective encoding or s e l e c t i v e memory? - C l i n i c a l implications - General implications - Tracking subjectively s a l i e n t events: A methodological note References Appendices 1. Expectancy manipulations 2. Videotape transcript 3. False memory items 4. Memory recognition test 5. Interpersonal adjective scales 6. Consent form 7. F i l l e r task 8. Analysis of tracking data: Multiple response format 9. Analysis of ninth agreeable i n t e r v a l 10. Correlations between cumulative i n t e r v a l tracking scores and interpersonal adjective ratings LIST OF TABLES Page Table 1. Group means for expectancy manipulation pretest 45 Table 2. Group means for glo b a l adjective ratings of 49 husband i n vignette t r a n s c r i p t (written version) Table 3. Group means for global adjective ratings of 51 husband i n videotape vignette ( f i n a l version) Table 4. Group means for impression formation data 64 v. Table 5. Group means on tracking task (Dominant i n t e r v a l s ) 67 Table 6. Group means on tracking task (Agreeable i n t e r v a l s ) 68 Table 7. Group means on tracking task (Neutral i n t e r v a l s ) 69 Table 8. Group means on tracking task (Intervals summed) 72 Table 9. Correlations between tracking behaviour and 74 adjective ratings Table 10. Group means for true memory items 77 Table 11. Group means for f a l s e memory items 78 - v i -ACKNOWLEDGEMENTS There are a number of people who made important contributions to th i s project, and whose help I would l i k e to acknowledge. D i m i t r i Papageorgis, Ray Corteen, and Jim Steiger, served as committee members, each contributing hi s own perspective on psychology and p a r t i c u l a r type of expertise. Jerry Wiggins, who chaired the committee, provided encouragement and support throughout the project, and nurtured whatever creative seeds were there. Daniel Kahneman, ea r l y i n the development of my ideas, provided astute c r i t i c i s m which influenced the d i r e c t i o n of my subsequent thinking. Special thanks are due to Debbie Abrami and Rene Weideman, who served as actors i n the videotape vignette which was employed. F i n a l l y , I'd l i k e to acknowledge the influence and support of Park Davidson. Although he did not l i v e to see the f i n a l development of my ideas, his encouragement throughout the f i r s t portion of my graduate career, and during my e a r l i e s t meanderings into the present area, helped set the stage f o r l a t e r developments. - v i i -"What are transferences? They are new additions or f a c s i m i l e s of the impulses and phantasies which are aroused and made conscious during the progress of the analysis; but they have t h i s p e c u l i a r i t y , which i s c h a r a c t e r i s t i c f o r th e i r species, that they replace an e a r l i e r person by the person of the physician. To put i t another way: a whole serie s of psychological experiences are revived, not as belonging to the past, but as applying to the person of the physician at the present moment." (Sigmund Freud, 1905, p. 116) "... These psychotic elaborations of imaginary people and imaginary personal performances are spectacular and seem very strange. But the fa c t i s that i n a great many rela t i o n s h i p s of the most commonplace kind - with neighbours, enemies, acquaintances, and even such s t a t i s t i c a l l y determined people as the c o l l e c t o r and the mailman - variants of such d i s t o r t i o n s often e x i s t . The c h a r a c t e r i s t i c s of a person that would be agreed to by a large number of competent observers may not appear to you to be the c h a r a c t e r i s t i c s of the person toward whom you are making adjustive or maladjustive movements. the r e a l c h a r a c t e r i s t i c s of the other fellow at that time may be of n e g l i g i b l e importance to the interpersonal s i t u a t i o n . This we c a l l parataxic d i s t o r t i o n . " (Harry Stack S u l l i v a n , 1954, p. 26) "My experience i s what I agree to attend to. Only those items which I notice shape my mind - without s e l e c t i v e i n t e r e s t my experience i s utter chaos." (William James, 1950, p. 402) - v i i i -INTRODUCTION Dysfunctional Expectations and Maladaptive  Behavior: Implications f o r Psychotherapy In one of h i s most frequently quoted remarks, George K e l l y (1955) stated over two decades ago that: "A person's processes are psychologically chanellized by the ways i n which he a n t i c i p a t e s events." (p. 46) With t h i s statement, he not only established the fundamental postulate of h i s personal construct theory, but also anticipated the strong emphasis on the role of expectancy factors i n psychological dysfunction and therapeutic change, evidenced by contemporary psychotherapy theorists and researchers ( c f . Bandura, 1974; Beck, 1975; Bootzin, 1979; Carson, 1982; Mahoney, 1974; Wilkins, 1979; Wilson, 1980). While K e l l y has been credited recently f o r the prophetic nature of h i s i n s i g h t s ( c f . L a n f i e l d , 1980; Mischel, 1980) i t i s p a r t i c u l a r l y humbling to note that there i s an important sense i n which the recognition of the i n c r e d i b l e impact of expectations upon interpersonal r e l a t i o n s h i p s , as well as the therapeutic value of i l l u m i n a t i n g and modifying dysfunctional expectations was presaged by Freud. As early as 1895, Freud (Breuer & Freud, 1895) had noted that patients i n analysis often respond to the analyst as i f he were someone e l s e , and hypothesized that i n such cases fee l i n g s which had i n i t i a l l y been experienced towards some other i n d i v i d u a l were being "transferred" on to the - 1 -physician. He christened t h i s processs the transference phenomenon, and described procedures for dealing with i t therapeutically. In so doing, he l a i d the foundation for the i n v e s t i g a t i o n of therapeutic transference; an a c t i v i t y which was subsequently to become p i v o t a l to the e n t i r e enterprise of psychoanalysis. Freud believed that psychoanalytic treatment does not create transferences, but rather illuminates t h e i r existance. He maintained that transference a r i s e s spontaneously i n a l l human re l a t i o n s h i p s , and i s not r e s t r i c t e d to the context of analysis (Lang, 1976). Although Freud i n i t a l l y saw the transference as an impediment to successful a n a l y s i s , he soon began to see i t as a valuable opportunity for gaining in s i g h t into the nature of the patient's intrapsychic c o n f l i c t s , as e a r l i e r unwanted si t u a t i o n s and p a i n f u l experiences were r e l i v e d i n his i n t e r a c t i o n with the analyst (Freud, 1920). By 1940, Freud saw the i n t e r p r e t a t i o n of the transference as being one of the c e n t r a l therapeutic mechanisms i n a n a l y s i s . He believed that through t h i s process the analysis could help the patient gain i n s i g h t i n t o the manner i n which currently dysfunctional behavior was based upon d i s t o r t e d perceptions r e s u l t i n g from past experiences (Freud, 1940). Freud shared an i n t e r e s t with contemporary cognitive therapists i n understanding the manner i n which the q u a l i t y of l i f e can be reduced through d i s t o r t e d perceptions of interpersonal events. His understanding of the mechanisms responsible f o r these d i s t o r t i o n s , however, was based upon his model of psychic energy. Transference was thus seen to r e s u l t from the cathexis of l i b i d i n a l energies on to the person of the analyst (Lang, 1976). - 2 -Sullivan's concept of parataxic d i s t o r t i o n , while borrowed d i r e c t l y from Freud, was formulated i n terms which are much more compatible with the contemporary cognitive metapsychology. Although Freud did acknowledge that transference was not e x c l u s i v e l y a r e s u l t of the a n a l y t i c s i t u a t i o n , S u l l i v a n emphasized more c l e a r l y that parataxic d i s t o r t i o n s are a regular part of everyday s o c i a l perception. They a r i s e out of our early primitive attempts to structure our experiences about ourselves and s i g n i f i c a n t i n d i v i d u a l s i n our environment ( S u l l i v a n , 1940). Experiences are thus structured by forming expectations about what events are l i k e l y to take place i n the future. To paraphrase K e l l y (1955), these expectations subsequently shape a l l new experiences. It i s i n e v i t a b l e that our expectations about the future w i l l be based upon important experiences i n the past. It i s thus not s u r p r i s i n g that our expectations about people and interpersonal r e l a t i o n s h i p s are based upon our experiences with important figures i n the past. Although expectations are a standard part of ordinary s o c i a l perception, interpersonal therapists (e.g., Carson, 1982) hypothesize that neurotic i n d i v i d u a l s develop dysfunctional expectations which lead to d i s t o r t e d perceptions of other people. These d i s t o r t e d perceptions r e s u l t i n maladaptive behaviors. This simple reformulation recasts Freud's basic i n s i g h t from within a cognitive perspective, and c l e a r l y implicates expectations and the structuring of experience as processes warranting further i n v e s t i g a t i o n . If maladaptive expectations are instrumental i n maintaining dysfunctional behavior, i t would seem reasonable that helping c l i e n t s to change the i r expectations might be a useful therapeutic t o o l . And i n f a c t , - 3 -modifying c l i e n t s f a u l t y expectations i s i d e n t i f i e d as a c e n t r a l thrust of diverse therapeutic approaches. Frank (1961) spoke r e l a t i v e l y e a rly on about the importance of coming to understand and subsequently modifying the c l i e n t ' s "assumptive world". Both E l l i s (1977) and Perls (1973) have emphasized the therapeutic value of modifying the c l i e n t ' s "catastrophic expectations" about the future. Beck (1976) argues that depression i s maintained by a c o n s t e l l a t i o n of negative expectations about the s e l f , one's experiences, and the future. Bandura (1977) has proposed that modifying expectations of personal e f f i c a c y i s a common mechanism of a l l e f f e c t i v e therapies. Wilson (1980) has recently r e i t e r a t e d the p o s i t i o n that modifying the c l i e n t ' s expectations i s p i v o t a l i n the therapeutic enterprise. Carson (1982) maintains that dysfunctional expectations about the c h a r a c t e r i s t i c s of other people are key factors i n the perpetuation of neurotic behavior. While there i s a consensus about the importance of modifying dysfunctional expectations there i s also agreement about the f a c t that these expectations are d i f f i c u l t to modify ( c f . Chrzanowski, 1982, Coyne & Segal, 1982; Wachtel, 1982).One reason f o r t h i s i s that people t y p i c a l l y e l i c i t behaviors i n others which confirm t h e i r expectations (Carson, 1982; K i e s l e r , 1982). This i s e s s e n t i a l l y the notion of complementarity of behaviors i n the two person i n t e r a c t i o n which i s so c e n t r a l to the interpersonal perspective of personality (Leary, 1957; S u l l i v a n , 1940). People i n a sense create t h e i r own environments by e l i c i t i n g predictable patterns of behavior from others. The depressed i n d i v i d u a l makes i t d i f f i c u l t f o r others to be f r i e n d l y and intimate with him; the aggressive i n d i v i d u a l e l i c i t s h o s t i l i t y - 4 -i n other people. There i s thus an "unbroken causal loop between s o c i a l perception, behavioural enactment, and environmental reaction" (Carson, 1982). Given the self-perpetuating nature of such recursive loops, i t would seem v i t a l f o r the therapist to have some way of breaking i n t o this v i c i o u s c i r c l e . Wachtel (1982) among others, asserts that i t i s e s s e n t i a l f o r t h i s reason that therapists employ enactive strategies which guide the c l i e n t i n making s p e c i f i c behavioural changes i n the r e a l world. The assumption i s that changes i n the c l i e n t ' s behavior w i l l lead to changes i n the manner that i n d i v i d u a l s t y p i c a l l y react to him, thus disconfirming h i s negative expectations. This, Wachtel believes, provides an important impetus f o r seeking a rapprochement between psychodynamic and behavioural therapies. While the fac t that i n d i v i d u a l s a c t i v e l y influence s o c i a l i n t e r a c t i o n sequences through t h e i r own behavior may contribute to the creation and maintenance of s e l f - f u l f i l l i n g prophecies (Darley & Fazio, 1980), there i s evidence that b e l i e f s about s o c i a l events, once established, tend to r e s i s t disconfirmation even without this environmental reaction aspect of the loop. In a serie s of experiments, Lee Ross and colleagues have demonstrated that subjects w i l l maintain a b e l i e f which i s based upon f a l s e information even a f t e r t h i s information has been d i s c r e d i t e d (Nisbett & Ross, 1980). Ross, Lepper & Hubbard (1975), f o r example, provided subjects with f a l s e feedback regarding t h e i r a b i l i t i e s to d i s t i n g u i s h between authentic and nonauthentic suicide notes. They found that this feedback s i g n i f i c a n t l y influenced subjects' s e l f evaluations of t h e i r own a b i l i t i e s on t h i s task even a f t e r they had been t o l d that the feedback was f a l s e . Moreover, and p a r t i c u l a r l y relevant to the present discussion, subjects who observed the f i r s t set of subjects performing the task, formed impressions about t h e i r - 5 -d i s c r i m i n a t i o n a b i l i t i e s which were even more r e s i s t a n t to change than the impressions of the a c t i v e l y p a r t i c i p a t i n g subjects. More recently, Anderson, Lepper and Ross (1980) demonstrated that subjects who had been induced to develop b e l i e f s about the r e l a t i o n s h i p between r i s k taking and success as a f i r e f i g h t e r , maintained these b e l i e f s even a f t e r the information they had i n i t i a l l y been given was d i s c r e d i t e d . These findings converge with evidence emerging from the areas of c l i n i c a l judgment, p r o b a b i l i t y estimation, and d e c i s i o n making that people are reluctant to abandon inappropriate judgment strategies i n the face of evidence that t h e i r judgments are i n c o r r e c t . Even a f t e r receiving negative feedback, they maintain the i l l u s i o n that t h e i r judgments have been accurate (Einhorn & Hogarth, 1978). Given the importance of expectations to the entire f i e l d of psychotherapy, as well as the apparent resistance of expectations to disconfirmation, i t would seem p o t e n t i a l l y valuable to c l a r i f y our understanding of the cognitive mechanisms through which expectations exert t h e i r influence upon experience. While the statement that expectations shape experience may have some explanatory power, there i s a sense i n which i t i s lacking i n informativeness. It i s c l a r i f y i n g here to draw upon the "structure-process" d i s t i n c t i o n which i s employed i n experimental cognitive psychology (Anderson, 1976, 1978; Estes, 1978). The expectation that an i n d i v i d u a l has at any given point i n time can be thought of as the cognitive structure through which past experience i s represented f o r him. I t i s only through the existence of t h i s cognitive structure that past experience can a f f e c t future behavior. As Anderson (1978) points out, i t i s meaningless to speak about cognitive - 6 -structures without specifying the processes through which these structures are accessed and manipulated. This i s important on both a t h e o r e t i c a l and a p r a c t i c a l l e v e l . On a t h e o r e t i c a l l e v e l , as Anderson suggests, without an understanding of the processes through which these structures are accessed and changed, we have only a p a r t i a l or incomplete understanding of the phenomenon of i n t e r e s t , since d i f f e r e n t processing assumptions w i l l lead to d i f f e r e n t predictions. On a p r a c t i c a l l e v e l , an understanding of the processes which access and manipulate the cognitive structures of i n t e r e s t w i l l p o t e n t i a l l y provide the c l i n i c i a n with new i n s i g h t s into how to go about modifying c l i e n t s ' dysfunctional expectations. Cognitive therapists, while eager to r i d themselves of the t h e o r e t i c a l constraints imposed by a r a d i c a l behaviourist metapsychology, have been somewhat slower to a v a i l themselves of both t h e o r e t i c a l and empirical developments taking place i n the f i e l d of cognitive psychology. While the s i t u a t i o n i s beginning to change ( c f . Arnkoff, 1980; Bower, 1978; Goldfried, 1979) t h i s type of synthesis i s s t i l l i n i t s infancy ( c f . Landau & Goldfried, 1981). A better understanding of the cognitive mechanisms through which expectations normally influence the perception of interpersonal behaviour w i l l p o t e n t i a l l y provide therapists with new i n s i g h t s i n t o how to go about modifying s o c i a l perceptions when they become di s t o r t e d . In the next section some of the relevant l i t e r a t u r e on perception and cognition w i l l be reviewed i n order to provide a t h e o r e t i c a l framework which w i l l then guide an empirical i n v e s t i g a t i o n of the cognitive mechanisms through which expectations influence s o c i a l perception. - 7 -EXPECTATIONS IN PERCEPTION AND COGNITION Expectations i n Pattern Recognition An i n i t i a l clue as to the process through which expectations i n t e r f a c e with the environment can be found i n recent cognitive approaches to the problem of pattern recognition. Models of pattern recognition have been strongly influenced by a r t i f i c i a l i n t e l l i g e n c e technology. One of the e a r l i e r and more i n f l u e n t i a l pattern recognition programs was developed by S e l f r i d g e ( S e l f r i d g e & Neisser, 1960). He was concerned with the problem of how written l e t t e r s are recognized. The model he proposed involved an ac t i v e cognitive search f o r c r i t i c a l features i n the environment to decide between competing hypotheses about the i d e n t i t y of a l e t t e r . More recent models ( c f . Rumelhart, 1977) hold that expectations play a v i t a l r ole i n t h i s search for c r i t i c a l features. Expectations are assigned a formal r o l e i n Rumelhart and S i p i e ' s (1974) model i n the following manner. When a l e t t e r to be recognized i s encountered, i t i s immediately analyzed f o r the presence of c r i t i c a l features. These c r i t i c a l features are combined with any expectations which e x i s t as to what the l e t t e r may be In order to construct a hypothesis as to what the word may be. This hypothesis subsequently guides any further search which i s required f o r c r i t i c a l features. This process can continue i n d e f i n i t e l y u n t i l a s a t i s f a c t o r y i d e n t i f i c a t i o n takes place. Expectations thus function as hypotheses as to what the pattern means, and play a c e n t r a l and indispensable role i n the normal process of pattern - 8 -recognition. In the process of l e t t e r recognition, these expectations are derived from a v a r i e t y of sources such as neighbouring l e t t e r s and neighbouring words. A well known example employed by Neisser (1967) involves the recognition of the symbols " T H E C H T". Most readers w i l l have l i t t l e trouble i n t e r p r e t i n g the second l e t t e r as "H" and the f i f t h l e t t e r as "A", despite the f a c t that the two l e t t e r s possess i d e n t i c a l features. A v a r i e t y of studies have demonstrated that expectations i n l e t t e r and word recognition are activated by both l e t t e r p r e d i c t a b i l i t y and semantic considerations (Estes, 1975; Meyer, Schvanveldt, & Ruddy, 1974; M i l l e r , Bruner, & Postman, 1954; Schvanveldt & Meyer, 1973; Tulving, Mandler, & Baumal, 1964). In other contexts expectations are derived from past experience, present context, and whatever other information sources are a v a i l a b l e (Rumelhart, 1977). It should be f a i r l y c l e a r by t h i s point that i n pattern recognition, expectations, f a r from functioning as an a r t i f a c t , play a regular and c r u c i a l r o l e . Pattern recognition can be conceptualized as an ongoing problem solving a c t i v i t y which i s guided by expectations. In the same sense that appropriate expectations are a prerequisite for accurate perception, inappropriate expectations can d i s t o r t perceptions. A study conducted by Bruner and Potter (1964) i l l u s t r a t e s t h i s point. A s e r i e s of s l i d e s of f a m i l i a r objects were shown to subjects. In one condition, the s l i d e s were i n i t i a l l y out of focus. The focus was gradually sharpened u n t i l the subjects were able to recognize the objects. In the other condition, the s l i d e s were i n focus from the beginning. Subjects i n the f i r s t condition took much longer to recognize the objects than subjects i n the second condition, even a f t e r the s l i d e s were i n focus. The authors interpreted the r e s u l t s to suggest that subjects i n the unfocused condition - 9 -developed expectations as to what the object was before i t was c l e a r l y i n focus. When the expectations were inaccurate ones, correct i d e n t i f i c a t i o n of the objects was impeded even a f t e r they were i d e n t i f i a b l e under normal circumstances. Selective Attention and the Encoding of Information Although the recognition of simple patterns such as l e t t e r s or physical objects, may appear somewhat remote from the type of interpersonal cognition of i n t e r e s t to the c l i n i c i a n , there i s a trend i n modern cognitive theory to eliminate the c l a s s i c a l d i s t i n c t i o n between lower l e v e l perceptual processes and higher l e v e l cognitive processes. There i s an increasing tendency f o r cognitive psychologists to be concerned with the cognitive strategies or mental a c t i v i t i e s that the i n d i v i d u a l a c t i v e l y engages i n while processing information. This emphasis upon con t r o l processes (Atkinson & S h i f f r i n , 1968) appears to be replacing more t r a d i t i o n a l s t r u c t u r a l models of human psychological processes. The r o l e assigned to expectations i n the pattern recognition model described i s that of guiding the search f o r c r i t i c a l features i n the perceived stimulus. This process of perceptual search determines which data are u t i l i z e d f o r further processing and which data are excluded. In more general terms, t h i s process i s usually discussed under the heading of s e l e c t i v e a t t e n t i o n . Early models of s e l e c t i v e a t t e n t i o n were s t r u c t u r a l i n nature (Broadbent, 1958; Deutsch & Deutsch, 1963; Treisman, 1969). A hypothetical cognitive f i l t e r i n g mechanism was postulated to be responsible for admitting some data for higher l e v e l cognitive processing and excluding - 10 -others. The major t h e o r e t i c a l controversy centered around the question of which stage i n the processing of information the f i l t e r i n g mechanism resided. This type of t h e o r e t i c a l debate i m p l i c i t l y assumed that perception and cognition were two d i s t i n c t cognitive processes and that information processing proceeded sequentially from lower l e v e l perceptual processing to higher l e v e l cognitive processing. (Deutsch & Deutsch's (.1963) -model was somewhat of an exception to t h i s rule.) More recent t h e o r e t i c a l developments- challenge the t r a d i t i o n a l l y held d i s t i n c t i o n between perception and cognition and the sequential view of inform-ation processing. Neisser's perceptual cycle model emphasizes the fact that perceptual a c t i v i t y takes place at a number of d i f f e r e n t l e v e l s of abstraction simultaneously and that the mechanisms through which these d i f f e r e n t l e v e l s of perceptual a c t i v i t y operate are s i m i l a r i n nature (Neisser, 1976). In t h i s model, perception i s viewed as a continuous, ongoing c y c l i c a l process through which the perceiver interacts- with the environment. The process of perceptual search i s directed by expectations or cognitive schemata which i n turn are modified on an ongoing basis by feedback from the environment. The modified schemata subsequently c o n s t i t u t e a new s e t of e x p e c t a t i o n s r e g a r d i n g what i s most l i k e l y to occur, and i n t h i s manner continue to guide perceptual exploration. Schemata thus d i r e c t information pickup l i k e continously modified format statements i n computer programs. To quote Neisser (1976), the schema i s : "...that portion of the e n t i r e perceptual cycle which i s i n t e r n a l to the perceiver, modifiable by experience, and somehow s p e c i f i c to what i s being perceived. The schema accepts information as i t becomes a v a i l a b l e at sensory surfaces and i s changed by that information. It d i r e c t s movements and exploratory a c t i v i t i e s that make more information a v a i l a b l e by which i t i s further modified." (p. 54) - 11 -Schemata of varying l e v e l s of g e n e r a l ity and abstraction are embedded within one another i n a h i e r a r c h i c a l manner and operate simultaneously. Thus to return to the i n i t i a l example of word recognition, perception of i n d i v i d u a l l e t t e r s takes place at the same time as perception of whole words. Moreover, these two l e v e l s of perception are completely interdependent and support one another. Palmer (1975) has discussed t h i s phenomemon as i t occurs i n v i s u a l processing and r e f e r s to i t as the "part-whole context". The c e n t r a l thrust of h i s argument i s that the v i s u a l recognition of parts i s impossible without the context of the whole they are embedded i n and by the same token the v i s u a l recognition of the whole i s impossible without the context of the parts which are embedded within i t . S i m i l a r l y , the perception of abstract s o c i a l meaning must always take place within the context of the perception of more concrete s t i m u l i . The mechanism of perception i s the same however, and one type of perceptual cycle must be embedded within the other. As Greene (1976) has noted, a p o t e n t i a l l y u seful perspective from which to view the perception of s o c i a l meaning i s to regard i t as a problem solving task. In contrast to the conventional problem solving task, however, s o c i a l perception requires the s o l u t i o n to an " i l l structured problem" i n the sense that there i s never one optimal and unique s o l u t i o n . There are, however, functional and dysfunctional s o c i a l perceptions. The depressed i n d i v i d u a l who c o n s i s t e n t l y perceives acquaintances as r e j e c t i n g has a maladaptive s t y l e of s o c i a l perception. The s o c i a l l y anxious i n d i v i d u a l who always reads disapproval i n the faces of h i s peers tends to engage i n dysfunctional perceptual a c t i v i t y . - 12 -Although Neisser's perceptual cycle model i s t y p i c a l l y applied to the de s c r i p t i o n of physical object perception, there i s no reason that i t cannot be generalized to the perception of abstract meaning. Greenberg & Safran (1982) provide a simple i l l u s t r a t i o n of the manner i n which the model can be employed to describe the dysfunctional perception of interpersonal behavior c h a r a c t e r i s t i c of the neurotic i n d i v i d u a l . The example they give describes the cognitive-perceptual a c t i v i t y of a s o c i a l l y anxious male c l i e n t who telephones a female acquaintance to ask her out on a date: "... After much pro c r a s t i n a t i o n he f i n a l l y summons up the necessary prerequisite courage and d i a l s her telephone number. At t h i s point i n the i n t e r a c t i o n he already has a c e r t a i n network of expectancies regarding the set of events which are l i k e l y to occur, formed on the basis of past experiences and present context, which serve to automatically d i r e c t h i s perceptual a c t i v i t y from the very beginning. Having completed d i a l i n g he attends to the d u l l sound of the d i a l tone and waits f o r her to pick up the phone at the other end. As soon as he hears her voice, he engages i n p a r t i c u l a r automated perceptual search s t r a t e g i e s . He begins to scan the tone and content of what she i s saying f o r cues of r e j e c t i o n . He d i r e c t s his attention towards the note of surprise i n her voice as she recognizes him. Having sampled t h i s cue from the environment, guided by his schema a n t i c i p a t i n g r e j e c t i o n , he begins to construct a representation of her behavior i n a manner which i s meaningful to him and then seeks further information to f i l l out the construction. He next s e l e c t i v e l y attends to the hesitancy i n her response to h i s i n v i t a t i o n , and encodes t h i s as a r e j e c t i o n rather than as thinking about her schedule. This information i s transformed into confirmation that she doesn't wish to go out with him and i s trying to figure out how to get r i d of him." The preceding example demonstrates the manner i n which dysfunctional expectations may at le a s t i n part r e s i s t change because of a generalized information processing strategy which r e s u l t s i n the search f o r confirming information. Indeed there i s evidence i n the area of abstract problem solving that disconfinning evidence tends to be disregarded as a byproduct - 13 -of t h i s general information processing strategy (Wason, 1968, 1969; Einhorn & Hogarth, 1978). It i s hypothesized here that these same p r i n c i p l e s generalize to the realm of s o c i a l perception. Memory The Selective R e t r i e v a l of Information In the t r a d i t i o n a l s t r u c t u r a l model of memory, information i s seen to be read out of sensory memory in t o short-term memory, and then transferred to long-term memory through rehearsal. The d u r a b i l i t y of the memory trace i s seen to be a function of the memory system i n which i t i s contained. As Neisser (1980) points out, t h i s l i n e a r , passive model of human information processing seems to have l i t t l e scope f o r accounting for a va r i e t y of phenomena of i n t e r e s t to the c l i n i c i a n . In p a r t i c u l a r , i t i s d i f f i c u l t to account f o r the impact of expectations on the perception of interpersonal behavior within t h i s framework. An i n f l u e n t i a l perspective on memory which provides an a l t e r n a t i v e to the t r a d i t i o n a l s t r u c t u r a l model has been a r t i c u l a t e d by Craik and colleagues (Craik & Lockart, 1972; Craik & Tulving, 1975). In th i s "depth of processing" model, the d u r a b i l i t y of the memory trace i s seen to be a byproduct of the manner i n which information i s encoded, rather than the memory system i n which i t i s currently deposited. Consistent with the perceptual cycle model previously discussed, perception i n this approach to memory i s conceptualized as a cognitive process through which s t i m u l i are analyzed at a number of d i f f e r e n t l e v e l s - 14 -simultaneously. These l e v e l s range from analysis of simple physical features at one extreme to complex semantic analyses at the other. An important feature of the depth of processing model i s the recognition of the f a c t that i n the process of perceptual a n a l y s i s , l o g i c a l l y p r i o r stages of analysis do not n e c e s s a r i l y precede l o g i c a l l y subsequent stages of a n a l y s i s . In other words, i n a given perceptual act the relevant stimulus may be analyzed at a r e l a t i v e l y sophisticated semantic l e v e l p r i o r to the analysis of simple physical features. A number of studies (Bobrow & Bower, 1969; Craik & Lockhart, 1972; Craik & Watkins, 1973; Treisman & Tuxworth, 1974) have demonstrated that by modifying the manner i n which s t i m u l i are perceptually processed, the d u r a b i l i t y of the r e s u l t i n g memory trace can be influenced. A representative study by Craik & Tulving (1975) f o r example, found that when exposed to a l i s t of words f o r a l i m i t e d time period, subjects who were asked whether words f i t into s p e c i f i c categories demonstrated better memory on a r e c a l l task than subjects who had been asked questions about typescript. Task demands which induce subjects to process words semantically can thus be seen to produce better word retention than task demands which require processing at the l e v e l of physical features. Craik & Tulving (1975) hypothesize that the memorability of an event i s a function of the elaborateness with which i t i s i n i t i a l l y encoded. In other words, returning to the perceptual problem solving model outlined e a r l i e r , retention w i l l increase i n proportion to the degree of elaborateness and depth of perceptual analysis engaged i n . - 15 -There are at le a s t two ways i n which a s p e c i f i c expectation might re s u l t i n s e l e c t i v e memory f o r interpersonal behavior which i s consistent with that expectation. The f i r s t hypothesis i s that information which i s inconsistent with the expectation i s not as deeply or elaborately encoded as schema consistent information. As Kahneman (1973) has noted, this does not mean that the inconsistent information i s completely neglected, but rather that fewer a t t e n t i o n a l resources are all o c a t e d to i t . In th i s manner unexpected information becomes l e s s consistent with the o v e r a l l organization of information on memory and i s thus l e s s easy to r e t r i e v e . The second, related p o s s i b i l i t y i s that the expectation functions not only as a processing guide at encoding, but also as a r e t r i e v a l schema during memory r e t r i e v a l . It i s thus t h e o r e t i c a l l y possible f o r expectations to induce s e l e c t i v e r e t r i e v a l , independent of any events which take place at encoding. Selective Reconstruction A f i n a l cognitive process through which expectations may a f f e c t the perception of interpersonal behavior i s through the reconstructive aspects of memory. Not only can cognitive schemata s e l e c t i v e l y r e t r i e v e fragments of information from memory, they can also f i l l i n missing values (Minsky, 1975). B a r t l e t t ' s (1932) i n i t i a l work on memory f o r st o r i e s demonstrated that i n d i v i d u a l s w i l l a c t u a l l y change or d i s t o r t s t o r i e s i n a manner which i s consistent with the general state of t h e i r world knowledge. The information i s thus a c t u a l l y assimilated over time to e x i s t i n g schemata. More recently, Mandler (1978) has demonstrated that such changes or d i s t o r t i o n s follow predictable rules f o r story - 16 -schemata. Perhaps one of the more dramatic demonstrations of the reconstructive aspects of memory comes from Piaget (1973). He demonstrated that c h i l d r e n who were asked to reproduce a row of seriated s t i c k s which they had seen some months e a r l i e r , arranged the s t i c k s consistent with t h e i r current concept of s e r i a t i o n rather than with the e a r l i e r reproductions which had been obtained from them. F i n a l l y , i n an important study, Loftus & Palmer (1974) demonstrated that subjects who were asked to describe a car accident they had witnessed were influenced i n t h e i r r e c o l l e c t i o n by the manner i n which the questions were phrased. Subjects who were asked: "How f a s t were the cars going when they smashed in t o one another", remembered the cars as having gone fas t e r then subjects who were asked: "How f a s t were the cars going when they h i t one another". Apparently information fragments which are stored i n memory function as only one source of information employed i n reconstructing a memory. Context and a l l a v a i l a b l e represented knowledge p o t e n t i a l l y relevant to a given event are also employed. In the same manner, general expectations about the s o c i a l environment may guide the reconstruction of interpersonal events. People may remember events, which ' although consistent with t h e i r expectations, did not necessarily transpire. - 17 -EXPECTATIONS AND THE PROCESSING OF SOCIAL INFORMATION: RECENT EMPIRICAL INVESTIGATIONS In the preceeding section three cognitive processes were described which may p o t e n t i a l l y mediate the influence of expectations upon the perception of interpersonal behavior: (1) Selective encoding, (2) Selective r e t r i e v a l , and (3) Selective reconstruction. In recent years there has been some attempt to explore these p o s s i b i l i t i e s e m p i r i c a l l y . In the present section I w i l l describe some of the more relevant i n v e s t i g a t i o n s . In one of the f i r s t of such attempts Zadny & Gerard (1974) demonstrated that actions relevant to the intentions a t t r i b u t e d to an actor were better r e c a l l e d than i r r e l e v a n t actions. Subjects observed a s k i t i n which a student stumbled and dropped a number of items on the ground. P r i o r to viewing the s k i t , subjects were t o l d that the student was a chemistry major, a music major, or a psychology major. On a subsequent memory test, subjects were better able to r e c a l l dropped items which were consistent with the student's d i s c i p l i n e of major (e.g., a s l i d e rule f o r the chemistry major). In a second experiment (Zadny & Gerard, 1974), i t was demonstrated that information regarding the actor's intentions tended to bias memory more when i t was delivered before observation of the s k i t than when i t was delivered a f t e r . These r e s u l t s were interpreted to support the view that s e l e c t i v e memory ar i s e s from s e l e c t i v e encoding rather than s e l e c t i v e r e t r i e v a l . Rothbart, Evans & Fulero (1979) found that subjects had s i g n i f i c a n t l y better r e c a l l f o r the behavior descriptions which were consistent with - 18 -t r a i t s i n i t i a l l y ascribed to groups of men, than f o r both inconsistent and unrelated behavior d e s c r i p t i o n s . They also found a greater e f f e c t when the expectancy was induced p r i o r to presenting the descriptions rather than a f t e r . I t should be noted that this paradigm d i f f e r e d from Zadny & Gerard's (1974) i n two major respects. F i r s t , Rothbart, et a l . (1978) employed a s o c i a l stereotype paradigm, i . e . , they were interes t e d i n s o c i a l perception of groups of men, rather than i n d i v i d u a l s . Second, they assessed memory for behavioural descriptions rather than f o r concrete items. Nevertheless, the i r r e s u l t s e s s e n t i a l l y p a r a l l e l l e d Zadny & Gerard's. Hastie and Kumar (1979) investigated s e l e c t i v e memory f o r behavioural descriptions as w e l l . They had subjects study l i s t s of behaviours which were congruent, neutral, and incongruent with respect to t r a i t s which were i n i t i a l l y ascribed by the experimenter to f i c t i o n a l characters. In contrast to the two previously described studies, r e c a l l was highest f o r behaviours which were incongruent with the i n i t i a l l y ascribed t r a i t . One parameter which may account f o r t h i s contradictory f i n d i n g i s that subjects were warned i n advance that they would be asked to r e c a l l behaviours. Given the task i n s t r u c t i o n s i t seems l i k e l y that they would employ whatever memory strategies were a v a i l a b l e to them during exposure to the stimulus materials. In t h i s context the i n i t i a l l y ascribed t r a i t would be more l i k e l y to function as a mnemonic than as a schema for perceptual search. The task demands would tend to encourage at t e n t i o n to d i s t i n c t i v e n e s s rather than the search f o r confirming evidence. This p o s s i b i l i t y h i g h l i g h t s the importance of employing an experimental paradigm which f a i t h f u l l y captures the e s s e n t i a l c h a r a c t e r i s t i c s of the phenomenon of i n t e r e s t . A memory r e c a l l task may require a very d i f f e r e n t cognitive strategy than an impression formation task. Moreover, forming impressions about f i c t i t i o u s characters on the basis of l i s t s of d i s c r e t e - 19 -behaviours may be d i f f e r e n t i n nature than impression formation i n r e a l l i f e . For this reason, Cohen (1981) attempted to design a more e c o l o g i c a l l y v a l i d study. She had subjects view a videotape of a woman who had previously been described as e i t h e r a waitress or a l i b r a r i a n . The videotape consisted of a staged i n t e r a c t i o n between the woman and her husband, which had been designed to portray features consistent with both waitress and l i b r a r i a n stereotypes. These features were both d i r e c t l y observable (e.g., wears glasses) and i n f e r r a b l e on the basis of the conversation (e.g., t r a v e l l e d i n Europe). Subjects were found to have better recognition f o r stereotype consistent features than f o r inconsistent feature s. While the studies discussed to t h i s point provide some evidence f o r s e l e c t i v e memory f o r expectation consistent information, i t i s unclear whether t h i s memory bias r e s u l t s from s e l e c t i v e r e t r i e v a l or s e l e c t i v e reconstruction. Cantor & Mischel (1977) investigated the reconstructive aspects of memory i n the following manner. They l a b e l l e d f i c t i t i o u s characters as being e i t h e r extraverted or introverted, and then described these characters with a series of other adjectives. Some of these adjectives were conceptually releated to the i n i t i a l l y ascribed t r a i t , while others were unrelated. On a memory recognition task, subjects were biased towards recognizing conceptually related, but nonpresented a d j e c t i v e s . Cantor & Mischel reason that t h i s memory bias r e s u l t s from the use of the a v a i l a b i l i t y h e u r i s t i c (Tversky & Kahneman, 1974). Subjects i n c o r r e c t l y i n f e r that an adjective was presented because i t i s e a s i l y retrieveable i n conjunction with the o v e r a l l schema activated by the i n i t i a l l y ascribed t r a i t . While t h i s study does not c l a r i f y the r e l a t i v e importance of s e l e c t i v e r e t r i e v a l versus s e l e c t i v e reconstruction, i t c l e a r l y demonstrates - 20 -that s e l e c t i v e reconstruction can bias the processing of s o c i a l information. Whether or not t h i s f i n d i n g would generalize to a more e c o l o g i c a l l y v a l i d context, remains unanswered. - 21 -METHODOLOGICAL CONCERNS The studies which have been reviewed provide some evidence that various types of expectations can bias the manner i n which information about other people i s c o g n i t i v e l y processed (cf. Carver &, Scheier, 1981). There are, however; a number of gaps i n the l i t e r a t u r e reviewed. These are as follows: 1. While the studies reviewed do provide some evidence that expectations bias the manner i n which c e r t a i n types of information are processed, none of these studies has d i r e c t l y demonstrated that a concomitant bias i n impression formation r e s u l t s i n such cases. 2. As discussed e a r l i e r , the studies c i t e d f o r the most part, are lac k i n g i n e c o l o g i c a l v a l i d i t y . Processing l i s t s of behavioural descriptions ( c f . Hastie & Kumar, 1979) or adjectives ( c f. Cantor & Mischel, 1977) may require a d i f f e r e n t type of cognitive a c t i v i t y than the type of information processing required by r e a l l i f e s o c i a l perception. Information presented i n written form omits much of the information which i s a v a i l a b l e i n r e a l s o c i a l i n t e r a c t i o n s , such as nonverbal cues and s o c i a l settings (Cohen, 1981). Moreover, any behaviour which takes place i n a r e a l s o c i a l s e t t i n g i s always embedded within the context of those events which take place before and a f t e r . Events unfold themselves through a temporal dimension. For these reasons the s o c i a l perceiver i n r e a l l i f e may have more r e t r i e v a l cues a v a i l a b l e to him when attempting to r e c a l l a s o c i a l i n t e r a c t i o n , thus f a c i l i t a t i n g better memory for s p e c i f i c interpersonal behaviours. A second concern raised - 22 -by Cohen (1981) i s that information provided to subjects i n s o c i a l perception research i s t y p i c a l l y simple and one-dimensional i n nature. Since more information i s a v a i l a b l e to the perceiver i n r e a l l i f e than that which i s t y p i c a l l y provided i n the experiment, i t i s possible that a v a r i e t y of d i f f e r e n t schemata are activated. The e f f e c t of one s p e c i f i c expectation may thus be minimized or cancelled out. 3. A related concern i s that the type of expectancy manipulations t y p i c a l l y employed are not completely germane to the type of s o c i a l perceptual process delineated as a target for concern i n the f i r s t section of t h i s manuscript. In everyday l i f e the expectations which guide the processing of s o c i a l information are t y p i c a l l y generated by the i n d i v i d u a l himself. They are thus self-generated hypotheses which i n theory guide the s o l u t i o n of an i l l - d e f i n e d problem (Greene, 1976). This i s somewhat d i f f e r e n t from the expectation created by t e l l i n g subjects that the target person i s a chemistry major (Zadny & Gerard, 1974) or a waitress (Cohen, 1981). Of course i n a l l fairness to the authors c i t e d , the expectancy manipulations employed were i n some cases completely relevant to the context of the s p e c i f i c i n v e s t i g a t i o n (e.g., Rothbart, et a l . (1978) were interested i n s o c i a l stereotyping and thus described groups of men as e i t h e r f r i e n d l y or i n t e l l i g e n t ) . However, i n order to better simulate the r e a l l i f e s i t u a t i o n i n which the i n d i v i d u a l generates his own expectations about the c h a r a c t e r i s t i c s of the target person, i t would be more appropriate to provide subjects with a c e r t a i n amount of information about the i n d i v i d u a l ' s past behaviour, on the basis of which they could generate t h e i r own hypotheses about future behaviour. - 23 -4. None of the studies reviewed has evaluated whether- or not any memory bias i s the result of'selective r e t r i e v a l , selective reconstruction or some combination of the two. The implication of different cognitive processes may have different c l i n i c a l implications (Kihlstrom & Nasby, 1981). 5. None of the studies reviewed has employed independent indices of encoding a c t i v i t y and memory. Selective attention, i n the s o c i a l cognition l i t e r a t u r e , has t y p i c a l l y been inferred on the basis of selective memory. The strategy of contrasting the performance of a group which has received expectancy information before seeing the stimulus material to that of a group which receives the information afterwards has been useful (cf. Cohen, 1981: Zadny & Gerad, 1974). I t permits the experimenter to rule out the p o s s i b i l i t y that an information processing bias takes place only at the time of r e c a l l . I t does not, however, permit the assessment of the p o s s i b i l i t y that an information processing bias takes place only through selective encoding. I t i s possible that an indiv i d u a l attends s e l e c t i v e l y to certain i n t e r -personal behaviours, and that this selective attention biases impression formation without biasing memory for s p e c i f i c behaviours. Interpersonal judgments are l i k e l y to have a strong affective component, and as Zajonc (1980) has demonstrated, affective judgments are not necessarily represented cognitively. I t i s thus conceivable that an individual might se l e c t i v e l y attend to certain behaviours and form an affe c t i v e impression on this basis without the mediating effect of selective memory for surface information. Without an independent index of encoding a c t i v i t y , however, such a phenomenon would elude detection. 24 It i s not a l l that surprising that independent indices of encoding a c t i v i t y have not t y p i c a l l y been employed since i t i s a cognitive a c t i v i t y which for obvious reasons i s d i f f i c u l t to assess d i r e c t l y . Perceptual a c t i v i t y i s an ongoing, dynamic process (Neisser, 1976). Any I attempt to have the subject report on t h i s a c t i v i t y r e t r o s p e c t i v e l y i s subject to the biasing e f f e c t s of memory, while i t would seem that an attempt to have the subject report on the focus of h i s attention on an ongoing basis would be s u f f i c i e n t l y obtrusive to i n t e r r u p t the n a t u r a l l y occurring perceptual cycle. In l i g h t of these apparent methodological problems, an i n t r i g u i n g p o s s i b i l i t y i s suggested by Massad, Hubbard, and Newtson (1978). They attempted to demonstrate the influence of subject expectations upon the s e l e c t i v e encoding of events using Newtson's u n i t i z i n g methodology to investigate perceptual organization. A l l subjects viewed a short animated f i l m i n which three cartoon characters (a large t r i a n g l e , a small t r i a n g l e , and a c i r c l e ) i n t e r a c t . The f i l m was ambiguous enough to support two contrasting i n t e r p r e t a t i o n s . In one i n t e r p r e t a t i o n , the large t r i a n g l e i s seen as a b u l l y and the small t r i a n g l e and small c i r c l e are seen as victims. In the second i n t e r p r e t a t i o n , the small t r i a n g l e and small c i r c l e are seen as thieves and the large t r i a n g l e i s seen as a v i c t i m . Subjects i n two conditions were i n i t i a l l y provided with two d i f f e r e n t i n t e r p r e t i v e sets supporting these two i n t e r p r e t a t i o n s . They then viewed and u n i t i z e d the f i l m . Following t h i s , they rated the cartoon characters on semantic d i f f e r e n t i a l scales. The r e s u l t s showed subjects with d i f f e r e n t i n t e r p r e t i v e sets both interpreted and u n i t i z e d the cartoon d i f f e r e n t l y . Furthermore, a subsequent reversal i n i n s t r u c t i o n a l set resulted i n a r e v e r s a l i n both o v e r a l l construal of the f i l m and a tendency to u n i t i z e d i f f e r e n t l y . - 25 -While the Massad et a l . (1978) study has some p o t e n t i a l l y i n t e r e s t i n g implications, a number of questions are l e f t unanswered. What i s the precise t h e o r e t i c a l s i g n i f i c a n c e of u n i t i z i n g data? What does a dif f e r e n c e i n u n i t i z i n g mean? How much of a difference i n u n i t i z i n g i s necessary to make a difference? Can we l e g i t i m a t e l y extrapolate from the Massad et a l . (1978) study to the context of everyday perception? In the next section, we w i l l review some of the more important aspects of Darren Newtson's research program on u n i t i z i n g i n order to shed more l i g h t on these issues. - 26 -A CRITICAL ANALYSIS OF THE UNITIZING METHODOLOGY Newtson and colleagues i n a se r i e s of studies have demonstrated that subjects can segment ongoing behavioural sequences i n a manner which i s subj e c t i v e l y meaningful to them and that these s u b j e c t i v e l y p a r t i t i o n e d units play an important and systematic role i n the o v e r a l l i n t e r p r e t a t i o n of the e n t i r e a c t i o n sequence. The procedure through which this subjective . p a r t i t i o n i n g or " u n i t i z i n g " (Newtson, 1973) data are obtained i s a r e l a t i v e l y simple one. A videotape of the behavioural sequence of i n t e r e s t i s played for the subject who i s instruct e d to observe the videotape and to indica t e whenever he f e e l s that "one meaningful behaviour has ended and another one has begun" by pressing a button. The button i s attached to an event recorder which records both frequency and l o c a t i o n i n time of responses. Reasonably high t e s t - r e t e s t r e l i a b i l i t y has been found to e x i s t f o r both frequency of units marked by subjects as we l l as f o r the precise l o c a t i o n of unit markings for subjects (Newtson, 1976; Newtson, Note 1). Newtson maintains that these s u b j e c t i v e l y p a r t i t i o n e d units are accurate r e f l e c t i o n s of the manner i n which i n d i v i d u a l s encode behaviour i n everyday l i f e . A consistent f i n d i n g has been that subjects can vary the frequency of u n i t i z a t i o n at w i l l and that moreover, they s h i f t to using smaller, more frequent units of segmentation when an unpredictable event i s inserted i n the f i l m (Newtson, 1973; Newtson, 1976; Newtson, Note 1). Newtson argues that t h i s s h i f t to smaller units r e f l e c t s the fac t that unpredictable behaviour possesses greater informational value f o r the observer and i s consequently encoded at a f i n e r l e v e l of an a l y s i s . - 27 -Points i n the f i l m which tend to be segmented most frequently by subjects are referred to as "breakpoints" since these are the points at which the sequence i s broken into parts (Newtson & Enquist, 1976). These breakpoints have been demonstrated to possess a number of i n t e r e s t i n g properties. Newtson and Enquist (1976) demonstrated that subjects were more l i k e l y to detect deletions i n the f i l m when these deletions were made at breakpoints rather than at nonbreakpoints. Furthermore, they found that subjects viewing a s e r i e s of breakpoints which had been extracted from the f i l m were able to more accurately formulate descriptions of a c t i o n than subjects who were shown a seri e s of nonbreakpoints. They also rated the sequence as more i n t e l l i g i b l e . In another study, Newtson (1976) demonstrated that subjects were better able to recognize breakpoints extracted from f i l m s they had previously seen than nonbreakpoints extracted from the same f i l m . On the basis of the above r e s u l t s and a seri e s of other studies, Newtson, Rinder, M i l l e r & LaCross, 1978, Newtson, Enquist, & Bois, 197 7; Wilder, 1978), Newtson and colleagues have concluded that breakpoints do possess a psychological r e a l i t y and are meaningfully related to the perception of ongoing behavioural events. What are breakpoints? At what l e v e l of abstraction does the  encoding of s o c i a l i n t e r a c t i o n take place? Newtson assumes that the encoding of s o c i a l i n t e r a c t i o n normally takes place at the l e v e l of concrete behavioural events. While t h i s assumption has characterized h i s work from the beginning, the most clear-cut - 28 -a r t i c u l a t i o n of t h i s p r i n c i p l e can be found i n Newtson, Enquist, and Bois (1977). In th i s study they coded videotapes of simple a c t i o n sequences using the Eshkol-Wachman movement notation (Eshkol, 1973) and proceeded to i d e n t i f y the features which subjects were monitoring at breakpoints. Examples of features i d e n t i f i e d are: (1) right arm, ri g h t forearm, and (2) l e f t upper arm. They concluded that perception of behavioural sequences i n th e i r study consisted of the s e l e c t i o n of successive breakpoints i n the f i l m . Each breakpoint consisted of a simple anatomical feature such as those mentioned above. Perception consisted of the monitoring of change i n l o c a t i o n of those features from breakpoint to breakpoint. This i s consistent with Newtson's o v e r a l l t h e o r e t i c a l conception of the role of breakpoints i n behavioural perception which can be summarized i n the following statement: "In t h i s view, actions are perceived by the disc r i m i n a t i o n and/or s e l e c t i o n of successive "points of d e f i n i t i o n " i n the behavioural stream. These points of d e f i n i t i o n - breakpoints - would also be boundaries of ac t i o n units, because the occurrence of the defining information f o r one a c t i o n provides a basis f o r i t s dis c r i m i n a t i o n from the preceding action. Inspection of breakpoints themselves i s consistent with t h i s i n t e r p r e t a t i o n . A serie s of breakpoints provides an almost comic s t r i p summary of ongoing a c t i o n sequences" (Newtson, Enquist, & Bois, 1977). The view that perception consists of the l i n k i n g of a serie s of perceptual units which are concrete i n nature may be accurate when subjects are assigned the task of monitoring simple behavioural sequences. It should be noted that a l l of Newtson's research has focused upon the u t i l i z a t i o n of such simple a c t i o n sequences, without any true s o c i a l i n t e r a c t i o n or verbal exchange. In a t y p i c a l study, for example, subjects u n i t i z e d sequences i n which a man paced impatiently and i n t e r m i t t e n t l y answered a telephone, a woman set a table with plates and food, or a man systematically removed - 29 -stacks of magazines from a table and reshelved them (Newtson, Enquist, & Bois, 197 7). The assumption that the same un i t s of perception are employed i n everyday s o c i a l perception however, i s more questionable ( c f . T r i e r w e i l l e r , Note 2). Newtson argues f o r a continuity between the type of behavioural perception he investigates and s o c i a l perceptual processes i n the following manner: "Insofar as persons must i n f e r other's t r a i t s , sentiments, or intentions from what others do, the processes of person perception must begin with the perceived actions of other persons." (Newtson, et a l . , 1977). As Craik and Lockhart (19 72) have argued however, perceptual encoding can occur at a v a r i e t y of d i f f e r e n t l e v e l s of a b s t r a c t i o n . R e l a t i v e l y sophisticated analyses of meaning are not n e c e s s a r i l y preceded by the perceptual analysis of the relevant stimulus at a simple concrete l e v e l . Moreover, the l e v e l at which perceptual processing occurs i s influenced by the task i n which the observer i s engaged. Thus, i t i s not s u r p r i s i n g that a subject who i s i n s t r u c t e d to break a simple behavioural sequence into smaller chunks or u n i t s , would attend to r e l a t i v e l y simple concrete features. The requirements of everyday s o c i a l perceptual processing however, would seem to demand a d i f f e r e n t type of perceptual processing. It would presumably be non-adaptive to segment a complex s o c i a l i n t e r a c t i o n into changes i n the p o s i t i o n of physical features and to then i n f e r meaning from these changes, i f the perceiver has the a b i l i t y to encode the information at a deeper l e v e l of meaning i n i t i a l l y . Although the type of stimulus materials employed by Newtson and colleagues i n t h e i r i n v e s t i g a t i o n s may be of t h e o r e t i c a l i n t e r e s t i n and of themselves, they possess a l i m i t e d degree of g e n e r a l i z a b i l i t y to the type of everyday s o c i a l perception i n which the psychotherapy researcher i s i n t e r e s t e d . - 30 -Unfortunately, the a p p l i c a t i o n of Newtson's methodology to the study of more e c o l o g i c a l l y v a l i d stimulus contexts has been v i r t u a l l y non-existent. (One exception i s an unpublished methodological study by T r i e r w e i l e r (Note 2)). Perceptual Units or Salience Data? There i s a subtle inconsistency i n the manner i n which Newtson int e r p r e t s the breakpoint concept. On the one hand, he sees the breakpoint as co n s t i t u t i n g the boundary of demarcation between two perceptual u n i t s . A good example of t h i s conceptualization can be found i n Newtson and Enquist (1976): A series of breakpoints conveys almost comic s t r i p q u a l i t y , i n that they appear to summarize an event very well. Nonbreakpoints on the other hand, appear highly ambiguous, i n that a large number of a l t e r n a t i v e constructions of the event appears to be consistent with them. It i s possible therefore, that greater s e n s i t i v i t y to deletions at breakpoints occurred because the deletions i n t e r f e r e d with unit formation.- If unit formation occurs at breakpoints, then a ser i e s of breakpoints extracted from a f i l m and viewed i n succession should provide a more adequate and understandable summary of the a c t i o n sequence than a comparable s e r i e s of nonbreak- points. (Newtson & Enquist, 1976). The assumption here i s that the breakpoint's informational properties and consequent influence upon o v e r a l l perception, stems from i t s segmentation properties. This emphasis upon segmentation follows n a t u r a l l y from Newtson's e a r l y i n t e r e s t s i n Barker's psychological ecology (Barker, 1963). (There are, however, some important differences between the primary i n t e r e s t s of Barker and of Newtson, which we s h a l l discuss i n a moment.) - 31 -On the other hand, Newtson seems to i n t e r p r e t breakpoints as f o c i i n the stream of events which receive more a t t e n t i o n a l a l l o c a t i o n and more elaborated encoding. Breakpoints are thus those aspects of a behavioural sequence that are most s a l i e n t to observers. This perspective i s well r e f l e c t e d by Newtson (1977), wherein he attempts to ground the breakpoint concept i n Neisser's (1976) perceptual cycle theory. In t h i s perspective, breakpoints are the "points of d e f i n i t i o n " that are monitored for change by the observer. Newtson (197 7) attempts to reconcile these two conceptions of u n i t i z i n g data by asserting that breakpoints are both "the boundaries between a c t i o n units and the points i n the stimulus at which ac t i o n u n i t s are defined". In t h i s conception, however, the notion of perceptual unit appears to be superfluous. It i s i n s t r u c t i v e to remember that Newtson's o r i g i n a l i n t e r e s t i n u n i t i z i n g processes was strongly influenced by Barker's psychological ecology (Barker, 1963). Barker and h i s colleagues were interested i n describing the "natural" units that constituted an i n d i v i d u a l ' s behaviour i n his environment. The b e l i e f was that these units were unambiguously present i n the data and that observers could be trained to recognize them r e l i a b l y . The concern was thus not with the manner i n which the naive subject n a t u r a l l y encodes data but rather with the categorization of information by trained researchers into meaningful units. Consistent with these i n t e r e s t s the coders were not asked to chunk or parse ongoing behaviour i n a s o c i a l context, but instead were given as much time as was required to r e l i a b l y categorize data recorded i n narrative form. It i s thus important to r e a l i z e that Newtson has adapted a methodology which originated i n one context and has applied i t i n a very d i f f e r e n t context. Along with this adaptation have - 32 -come some questionable assumptions about the nature of the task that the subject i s engaging i n when u n i t i z i n g . For one thing, Newtson's subjects are not trained observers. Moreover, Newtson's subjects are being asked to u n i t i z e ongoing s o c i a l behaviour rather than recorded t r a n s c r i p t s . This d i f f e r e n c e may have some important consequences. The subject who i s asked to segment data recorded i n narrative form has as much time as i s required to make a considered de c i s i o n about where one unit ends and another unit begins. This decision can be made i n the context of the e n t i r e n a r r a t i v e . Thus both preceding and subsequent information can influence decisions regarding unit boundaries. This type of task i s conceptually not d i s s i m i l a r from the parsing of narratives (Mandler, 1978). In both of these contexts, i t may be reasonable to assume that the subjects can segment the tr a n s c r i p t i n t o units which n a t u r a l l y cohere together i n a meaningful manner. The i n d i v i d u a l perceiving a t y p i c a l s o c i a l i n t e r a c t i o n however, has less time to think about where the unit boundaries should be i n the ongoing i n t e r a c t i o n . Each p o t e n t i a l unit of data can only be contrasted with the preceding u n i t s . For this reason, the process of segmenting the i n t e r a c t i o n into meaningful units would be considerably more d i f f i c u l t . It may, i n f a c t , be the case that subjects i n the u n i t i z i n g task are marking s a l i e n t events on the videotape rather than d i v i d i n g the sequence i n t o perceptual u n i t s . Recent research on salience e f f e c t s on a t t r i b u t i o n a l processes (Taylor and Fiske, 1978; Smith and M i l l e r , 1979) c e r t a i n l y suggests that complex s o c i a l judgments may be dramatically affected by, or even be ex c l u s i v e l y the product of salience e f f e c t s . - 33 -It i s i n t e r e s t i n g to note that Newtson, despite h i s avowed i n t e r e s t i n perceptual organization and segmentation, provides subjects i n his more recent research with u n i t i z i n g i n s t r u c t i o n s that lend themselves quite e a s i l y to i n t e r p r e t a t i o n as salience i n s t r u c t i o n s (e.g., "Whenever i n your judgment a meaningful a c t i o n occurs, that i s , whenever the person does something, press the botton" (Massad, Hubbard, and Newtson, 1979)). One has only to change the word "meaningful" to "important" (bearing i n mind that the two words can i n some contexts be considered synonomous) to demonstrate how e a s i l y the above phrase can be interpreted as an i n s t r u c t i o n to mark events which are sub j e c t i v e l y s a l i e n t . These i n s t r u c t i o n s contrast quite markedly with the i n s t r u c t i o n s which were employed i n Newtson's e a r l i e r research: I am going to show you a 5-minute videotape of another subject i n a d i f f e r e n t experiment. Now, what I am interested i n here are the units people use to c l a s s i f y other's behaviour. By that I mean that people can vary the l e v e l at which they break up other's behaviour. For example, I could turn around, walk over, push the door closed, turn around, and walk back. Each of these actions could be seen as a d i s c r e t e event, or they might be c l a s s i f i e d into one larger u n i t , such as "closing the door" (Newtson, 1973). Subjects were then t o l d to "press the button when i n your judgment one unit ends and another unit begins". Some of t h i s conceptual confusion appears to have been in h e r i t e d by other researchers. Cohen and Ebbesen (1979) f o r example, while i n s t r u c t i n g subjects to segment an a c t i o n sequence in t o u n i t s , subsequently equate u n i t i z i n g with the a l l o c a t i o n of at t e n t i o n to selected features i n the environment. Subjects were - 34 -instructed to break up the tape i n t o "units or chunks which you f i n d most h e l p f u l i n trying to form an impression of the person". While these i n s t r u c t i o n s apparently request the subject to segment the tape into units, the u n i t i z a t i o n data are interpreted as follows: "... i t seems reasonable that one observer, noting and  presumably u n i t i z i n g ( i t a l i c s mine) a s l i g h t frown on the actor's face, began to regard her as an unhappy person and thus attended to a d d i t i o n a l negative s o c i a l features". (Cohen & Ebbesen, 1979). Apparently " u n i t i z i n g " i s equated with "noting" here. If u n i t i z i n g consists of marking the boundary between two events, how can one u n i t i z e a frown? Of course, subjects may well have interpreted the i n s t r u c t i o n s as a mandate to i n d i c a t e t h e i r focus of a t t e n t i o n during the f i l m , p a r t i c u l a r l y given the context i n which the previously quoted i n s t r u c t i o n s were embedded ( i . e . , "We're interest e d i n how you get out the information that you're trying to remember from these behaviour sequences" and "In t h i s way, you can show us where you get information from the behaviour sequences"). If t h i s i s the case, however, the i n s t r u c t i o n s to break the tape i n t o units are misleading to the reader and p o t e n t i a l l y to the subjects as w e l l . This second p o s s i b i l i t y may introduce unnecessary error variance i n the u n i t i z i n g data. Tracking Subjectively Salient Events The above review suggests that Newtson's u n i t i z i n g methodology may be u s e f u l f o r purposes of i n v e s t i g a t i n g the r e l a t i o n s h i p between expectations - 35 -and the encoding of interpersonal behaviour, provided c e r t a i n conceptual and methodological considerations are kept i n mind. In the previous section, I've discussed two c e n t r a l conceptual problems with Newtson's u n i t i z i n g methodology. The f i r s t issue that was addressed was whether people normally encode s o c i a l i n t e r a c t i o n at the l e v e l of molecular concrete behaviours or the l e v e l of more abstract events. The second issue which was addressed was whether or not subjects normally break up s o c i a l i n t e r a c t i o n into units or attend to selected aspects of the i n t e r a c t i o n . With regard to the f i r s t issue, I argued that the encoding of r e a l s o c i a l i n t e r a c t i o n i n a l l l i k e l i h o o d takes place at the l e v e l of abstract events. With regard to the second issue, I argued that considerable confusion e x i s t s as to whether subjects are marking perceptual units or tracking s u b j e c t i v e l y s a l i e n t events. I concluded that a subjective salience conceptualization of breakpoint data i s more consistent with Newtson's o v e r a l l t h e o r e t i c a l framework, and more l i k e l y from a perspective of cognitive economy and psychological feasibility."'" I t would seem important f o r the task i n s t r u c t i o n s to r e f l e c t e x p l i c i t l y t h i s conceptual c l a r i f i c a t i o n . To the extent that task i n s t r u c t i o n s can be c l a r i f i e d , brought i n l i n e with more na t u r a l l y occurring encoding s t r a t e g i e s , and made easier to implement, r e l i a b i l i t y w i l l be increased. For these reasons, i t would seem desirable to provide subjects with i n s t r u c t i o n s which e x p l i c i t l y ask them to mark s u b j e c t i v e l y s a l i e n t events This conceptualization i s also more consistent with the s e l f - r e p o r t of subjects who p a r t i c i p a t e d i n p i l o t studies which were conducted. - 36 -rather than to segment the vignette into concrete behavioural u n i t s . Given this c l a r i f i c a t i o n i n conceptualization of both task i n s t r u c t i o n s and task behaviour, the term u n i t i z i n g becomes somewhat of a misnomer. For t h i s reason, I s h a l l hereafter use the more appropriate term, tracking to re f e r to the tracking of subjectively s a l i e n t events. - 37 -THE PRESENT INVESTIGATION The present study was designed to investigate / the cognitive processes which mediate the impact of expectations upon the perception of interpersonal behaviour. The i n t e n t i o n was to s p e c i f i c a l l y redress c e r t a i n methodical shortcomings common to a number of the p o t e n t i a l l y relevant studies which have been conducted. To r e i t e r a t e i n b r i e f , these are: 1. f a i l u r e to assess whether or not expectations do indeed bias impression formation. 2. lack of e c o l o g i c a l v a l i d i t y . 3. l a b e l l i n g the target person, rather than inducing subjects to generate t h e i r own expectations. 4. f a i l u r e to d i s t i n g u i s h between s e l e c t i v e r e t r i e v a l and s e l e c t i v e reconstruction. 5. f a i l u r e to provide an independent index of memory and encoding a c t i v i t y . Overview A l l subjects viewed a videotaped husband and wife i n t e r a c t i o n . While they observed the i n t e r a c t i o n , they were asked to track events that were s a l i e n t to them using a modified version of Newtson's u n i t i z i n g procedure. Frequency and l o c a t i o n i n time of button presses were recorded. P r i o r to viewing the vignette, subjects were given one of two expectancy sets regarding the husband's t y p i c a l interpersonal behaviour. Following the f i l m , subjects were asked to rate the husband's behaviour along a ser i e s of adjective scales. In addition, memory f o r events i n the f i l m was assessed. - 38 -It was hypothesized that subjects would attend s e l e c t i v e l y to events i n the vignette which were consistent with the information provided to them p r i o r to the f i l m . This hypothesis was evaluated by comparing the s a l i e n t events marked by subjects i n the two expectancy conditions. It was further hypothesized that t h i s s e l e c t i v e a t t e n t i o n would r e s u l t i n both s e l e c t i v e memory f o r events consistent with subjects' expectations, as w e l l as impression formation which i s consistent with the biasing information. A recognition memory test format was employed which would d i s t i n g u i s h between the e f f e c t s of se l e c t i v e r e t r i e v a l and se l e c t i v e reconstruction. Developing the Expectancy Manipulation:  T r a i t s as Normative Conceptual Schemata An important objective i n developing the expectancy manipulation was to provide subjects with a c e r t a i n amount of information about the target person, on the basis of which they would have the opportunity f o r generating t h e i r own hypotheses. In order to accomplish t h i s i t was necessary to have some t h e o r e t i c a l guide f o r e s t a b l i s h i n g i n advance what s p e c i f i c information would induce the subjects to e s t a b l i s h predictable, s p e c i f i c information processing schemata. Cantor and Mischel (1977) have argued recently that a f r u i t f u l perspective from which to conceptualize the personality t r a i t i s i n terms of i t s role as a cognitive h e u r i s t i c for the observer. In t h i s view, personality t r a i t s function as normative conceptual schemata which guide the cognitive processing of information about other people. When, f o r example, we a t t r i b u t e the t r a i t of extraversion to a person, we expect him to display behaviours consistent with t h i s t r a i t . These expectations d i r e c t both the manner i n which we encode his behaviour as well as the manner i n which we r e c a l l i t . This perspective i s consistent with Alston's (1975) frequency concept of d i s p o s i t i o n s , which maintains that the a t t r i b u t i o n of a given d i s p o s i t i o n to a person i s equivalent to asserting that t h i s person w i l l emit a high frequency of behaviours which are consistent with that d i s p o s i t i o n . In the same vein, Buss and Craik (1980; 1981) have recently argued that interpersonal t r a i t s can be f r u i t f u l l y conceptualized as higher l e v e l categories which allow people to predict c e r t a i n r e g u l a r i t i e s i n interpersonal behaviours which are category members. Their research draws upon the concept of prototypes which was developed i n i t i a l l y by Rosch and colleagues (Rosch, Mervis, Gray, Johnson, and Boyes-Braem, 1976) to c l a r i f y the manner i n which people group together members of a category. The e s s e n t i a l notion i s that cognitive categories consist of "fuzzy sets" which are i l l - d e f i n e d i n that there i s no necessary set of a t t r i b u t e s that defines category membership. This contrasts with the t r a d i t i o n a l , l o g i c a l notion of category membership which asserts that categories can be defined i n terms of necessary and s u f f i c i e n t c r i t e r i a . Rosch maintains that the cognitive process through which people r e a l l y judge whether or not an element i s a member of a category, operates on the basis of perceived resemblance to some ce n t r a l or p r o t o t y p i c a l category member about which people consensually agree. The standard example i s that people consensually agree that the robin i s extremely " b i r d l i k e " or p r o t o t y p i c a l f o r the category of b i r d , whereas other birds such as o s t r i c h e s , are seen as l e s s p r o t o t y p i c a l , or more peripheral to the category. The basic methodology involves i n s t r u c t i n g subjects to rate various p o t e n t i a l category members f o r degree of perceived - 40 -c e n t r a l i t y to that category, and then assessing which category members are r e l i a b l y seen by subjects as p r o t o t y p i c a l . Using this methodology, Buss and Craik (1980; 1981) have been able to generate l i s t s of behaviours which subjects r e l i a b l y perceive as p r o t o t y p i c a l f o r a number of interpersonal t r a i t s . Their research suggested to us a useful methodology f o r a c t i v a t i n g schemata to guide the search f o r information i n the context of the present study. By providing subjects with behavioural information about an actor which i s p r o t o t y p i c a l for an i n d i v i d u a l with a given personality t r a i t , one can presumably activate that t r a i t as a working hypothesis i n the subsequent encoding of information about that actor. This methodology has the d i s t i n c t advantage of providing a c l e a r t h e o r e t i c a l guide for the experimental manipulation of subject expectations. The two interpersonal t r a i t s which were selected to be employed f o r expectancy manipulation purposes were dominance and agreeableness. These two t r a i t s mark the two p r i n c i p a l orthogonal vectors i n the Wiggins (1979) taxonomy of interpersonal behaviour and also correspond to two main dimensions of power and a f f i l i a t i o n which emerge repeatedly i n e m p i r i c a l l y developed taxonomies of interpersonal behaviour (Leary, 1957; Wiggins, 1982). Moreover there i s evidence that these two interpersonal dimensions of power and a f f i l i a t i o n , or agency and communion, are u n i v e r s a l i n man's conceptualization of interpersonal behaviour (White, 1980). There i s thus some assurance that the dimensions that were manipulated were i n a general sense t h e o r e t i c a l l y important. - 41 -In order to gain a concrete appreciation of the p o t e n t i a l relevance of these dimensions to the therapeutic context, imagine the s i t u a t i o n i n which a c l i e n t has the expectation that people w i l l always attempt to dominate her. T y p i c a l l y she finds herself responding to t h i s perceived attempt at domination by v a c i l l a t i n g between two types of behaviour. She alternates between responding with anger, indignation, and an attempt to dominate the other person i n order to protect herself; and responding with overly compliant and submissive behaviour. The l a t t e r type of behaviour i s t y p i c a l l y followed by s e l f - r e c r i m i n a t i o n s , and subsequent withdrawal or h o s t i l i t y . The model of s e l f - f u l f i l l i n g prophecy discussed e a r l i e r suggests that because she i s focused on the interpersonal dimension of power, dominant (or submissive) behaviours w i l l be most s a l i e n t to her. The p o s s i b i l i t y that an i n d i v i d u a l i s engaging i n an a f f i l i t a t i v e response w i l l have l e s s impact upon her, even i f o b j e c t i v e l y he emits an equal number of dominant and agreeable behaviours. The range of p o t e n t i a l l y d i f f e r e n t types of i n t e r a c t i o n s she can have with other people w i l l consequently be r e s t r i c t e d , and her interpersonal r e l a t i o n s h i p s w i l l be impaired. In the same manner, one can imagine a c l i e n t who a n t i c i p a t e s that others w i l l r e j e c t him (the warm-cold and agreeable-quarrelsome dimensions are t y p i c a l l y combined i n the Wiggins (1979) system since the u t i l i t y of a f i n e r d i s t i n c t i o n has not been established). As a r e s u l t of h i s expectations he perceives r e j e c t i o n where others might see n e u t r a l i t y or even warmth and agreeableness. He responds to t h i s perceived r e j e c t i o n by withdrawing, thus ensuring that no g r a t i f y i n g interpersonal contact i s made. - 42 -Pretesting the Expectancy Manipulation The dominant and agreeable expectancy manipulations were designed using David Buss' act frequency analysis protocols to provide behavioural exemplars of the relevant t r a i t s . For each of the two expectancy conditions, seven acts were selected from the l i s t of twenty-five most pr o t o t y p i c a l behaviours (Buss & Craik, 1980; 1981) appropriate to that condition. These seven acts were combined with four neutral acts i n order to prevent the character from appearing too implausible. The same four neutral acts were employed i n both conditions. The expectancy manipulations were pretested f o r t h e i r a b i l i t y to act i v a t e the appropriate prototypes or schemata i n the following manner. Ninety psychology undergraduates were employed as a pretest sample. Of these, t h i r t y were males and s i x t y were females. Subjects were t o l d that they would be given information which described a f i c t i t i o u s person. They were randomly assigned to eit h e r the dominant or the agreeable condition and then given a written version of the manipulation appropriate to t h e i r condition. The following task i n s t r u c t i o n s were employed: On the sheet i n front of you there i s a l i s t of 11 acts. Each one of these items describes a behaviour that a f i c t i t i o u s person named David, has engaged i n over the l a s t two weeks. I'd l i k e you to read them over c a r e f u l l y . When you've f i n i s h e d I'd l i k e you to rate the type of person you would expect David to be, on the scales which have been provided. When they had completed reading the information the Wiggins' interpersonal adjective scales (Wiggins, 1979) were administered. - 43 -A 2 X 2 X 2 (expectancy condition X adjective scale X sex) repeated measures ANOVA confirmed the existence of a s i g n i f i c a n t i n t e r a c t i o n between the expectancy conditions and the dominant and agreeable adjective scales, F(l,86) = 393.08, p < .001. Subjects who had been given the dominant manipulation rated the character as more dominant and l e s s agreeable than subjects who had been given the agreeable manipulation (see Table 1). There was no main e f f e c t for sex, and no i n t e r a c t i o n s with the sex f a c t o r , thus confirming that male and female subjects were affected i n a s i m i l a r manner by the expectancy manipulations. The pretest thus confirmed that the expectancy manipulations were functioning i n the desired manner (see Appendix 1 for copies of the manipulations). - 44 -TABLE 1 Group Means f o r Expectancy Manipulation Pretest ADJECTIVE SCALE RATING EXPECTANCY MANIPULATION Dominant Agreeable N X (SD) N X (SD) Dominant 44 53.25 (5.65) 44 31.00 (7.27) Agreeable 46 30.41 (7.25) 46 53.83 (4.52) - 45 -Developing the Videotape Vignette While studies examining the e f f e c t s of expectations upon the processing of s o c i a l information have often employed a l i s t of behaviours or t r a i t s f o r stimulus materials (Cantor & Mischel, 1977; Hastie & Kumar, 1979; Rothbart, Evans, & Fulero, 1979) the present study employed a videotape of a staged s o c i a l i n t e r a c t i o n i n order to increase the e c o l o g i c a l v a l i d i t y of the i n v e s t i g a t i o n . The videotape consisted of a 10-minute scene i n which a husband and wife are shown i n t e r a c t i n g with one another. The scene was developed with the objective of providing equal support f o r dominant and agreeable perceptions of the husband. A s c r i p t f o r the scene was developed i n the following manner. In the f i r s t stage, the twenty-five most p r o t o t y p i c a l acts f o r dominance and the twenty-five most p r o t o t y p i c a l behaviours f o r agreeableness from Buss 1 i n v e s t i g a t i o n s (Buss & Craik, 1980; 1981) were examined f o r behaviours which would be appropriate to a husband and wife i n t e r a c t i o n . Many of the acts had to be excluded f o r various reasons, such as taking place i n a group rather than a dyadic context (e.g., He took the lead i n l i v e n i n g up a d u l l party), or simple inappropriateness to the s p e c i f i c context (e.g., He offered an older person a seat on the bus). A second c r i t e r i o n f o r exclusion was that acts which had been employed i n the expectancy manipulations were not included, i n order to avoid the p o s s i b i l i t y of d i r e c t i n g subjects to look e x p l i c i t l y f o r a given act i n the vignette. On the basis of the acts that were retained an i n i t i a l s c r i p t was written i n which the husband displayed an approximately equal number of dominant and agreeable behaviours. An attempt was also made to intersperse these acts - 46 -with acts presumed to be neither dominant nor agreeable i n nature i n order to decrease dramatic changes back and f o r t h between dominant and agreeable behaviours which would render the character le s s p l a u s i b l e and r e a l i s t i c , as w e l l as exceeding subjects' reaction time c a p a b i l i t i e s when i t came time to track the vignette. This f i r s t version of the s c r i p t was then shown to a panel of psychologists and psychology graduate students f o r an i n i t i a l global appraisal of the extent to which the husband was perceived as being equally dominant and agreeable. On the basis of th e i r reactions, a second version of the s c r i p t was devised which attempted to balance further the perception of dominant and agreeable behaviours. This second version of the s c r i p t was then shown to a group of undergraduate psychology students who were asked to perform two tasks: They were f i r s t i n s t r u c t e d to read the s c r i p t quickly and to then rate the husband's global interpersonal behaviours on two six-point adjective scales: 1) dominant, and 2) agreeable. They were then instruc t e d to go through the s c r i p t a second time and to categorize each of the husband's speech acts as e i t h e r : dominant, agreeable, or neutral. This constituted the f i r s t of a seri e s of i t e r a t i o n s through which the s c r i p t was gradually refined u n t i l i t demonstrated the following c h a r a c t e r i s t i c s : 1) no s i g n i f i c a n t d i f f e r e n c e between global ratings of dominance and agreeableness; and 2) an equal number of perceived dominant versus agreeable speech acts for the husband. For the second c h a r a c t e r i s t i c a minimum of eighty percent intersubject agreement i n item categorization was established as a c r i t e r i o n . - 47 -A f t e r a number of i t e r a t i o n s , the above c h a r a c t e r i s t i c s were f i n a l l y demonstrated. The f i n a l version of the s c r i p t was shown to an independent sample of f i f t y - e i g h t psychology undergraduates. There were twelve males and f o r t y - s i x females i n the group. Global ratings of dominance and agreeableness were analyzed, using a 2 X 2 (sex X adjective rating) between - within ANOVA design with repeated measures on the adjective r a t i n g f a c t o r . There were no s i g n i f i c a n t differences between global ratings of dominance and agreeableness, F(l,56) = .977, p > .327, and there was no s i g n i f i c a n t i n t e r a c t i o n between sex of subject and ratings of dominance and agreeableness, F(l,56) = .136, p >.714 (see Table 2). Although one might a n t i c i p a t e sex differences i n the perception of dominance and agreeableness, the present a n a l y s i s provided no evidence of such differences i n the vignette which had been designed. Subjects categorized eight dominant i n t e r v a l s with greater than eighty percent agreement and eight agreeable i n t e r v a l s with greater than eighty percent agreement. Because there was l e s s agreement about which i n t e r v a l s were neither dominant nor agreeable, the lower l i m i t r e l i a b i l i t y c r i t e r i o n for neutral i n t e r v a l s was established as seventy percent. Eight neutral i n t e r v a l s to be employed i n subsequent research, were selected on t h i s basis. The lowered r e l i a b i l i t y c r i t e r i o n here was not considered to be a problem since i t was not considered e s s e n t i a l that subjects recognize these i n t e r v a l s as neutral, but only that they did not r e l i a b l y categorize these i n t e r v a l s as e i t h e r dominant or agreeable. The twenty-four t r a n s c r i p t i n t e r v a l s of i n t e r e s t are marked i n Appendix 2. - 48 -TABLE 2a Group Means f o r Global Adjective Ratings of Husband i n Vignette Transcript Dominant Agreeable N X (SD) N X (SD) 58 3.95 (.93) 58 4.14 (.95) TABLE 2b Interaction of Subject Sex with Global Adjective Ratings ADJECTIVE SCALE RATING SEX Dominant Agreeable (SD) N X (SD) Male 12 3.67 (1.07) 12 4.00 (.95) Female 46 4.02 (.88) 46 4.17 (.95) - 49 -At t h i s point, the s c r i p t was staged and videotaped. The husband and wife were portrayed by actors. The videotaped scene was shown to an independent sample of twenty psychology undergraduates who were instruc t e d to rate the husband's o v e r a l l behaviour on the Wiggins (1979) interpersonal adjective scales i n order to evaluate whether or not the desired balance had been maintained on the videotape. The subjects saw the husband as s i g n i f i c a n t l y more dominant than agreeable. A second version of the scene was subsequently videotaped i n which the husband's behaviour was modified s l i g h t l y without changing any of the behaviours which had been categorized with over eighty percent r e l i a b i l i t y i n the written s c r i p t . This new videotape was shown to an independent sample of eighteen psychology undergraduates (seven males and eleven females) who once again rated the husband's o v e r a l l behaviour on the adjective scales. The data were analyzed using a 2 X 2 between - within (subject sex X adjective scale) ANOVA. This time subjects saw the husband as equally dominant and agreeable, F(l,16) = .007, p > .935. There was no main e f f e c t f o r subject sex, F(l,16) = .277, p > .606, and no sex X adjective r a t i n g i n t e r a c t i o n , F(l,16) = 2.388, p >.142. Thus, there were no differences between male and female subjects (see Table 3). Since t h i s version of the stimulus scene met the c r i t e r i o n of being balanced f o r dominant and agreeable perceptions of the target person, i t was used i n the subsequent research (see Appendix 2 f o r vignette t r a n s c r i p t ) . - 50 -TABLE 3a Group Means f o r Global Adjective Ratings of Husband i n Videotape Vignette ( f i n a l version) Dominant Agreeable X (SD) N X (SD) 21.61 (4.98) 18 22.00 (4.84) TABLE 3b Interaction of Subject Sex with Global Adjective Ratings ADJECTIVE SCALE RATING Dominant Agreeable X (SD) N X (SD) 22.43 (5.44) 7 20.00 (4.04) 21.09 (4.87) 11 23.27 (5.04) C a l i b r a t i o n of the Videotape The vignette was divided i n t o unequal sized i n t e r v a l s with each i n t e r v a l corresponding to one speech act. These i n t e r v a l s constituted the windows or reference i n t e r v a l s with respect to which subject tracking of subjective salience was evaluated. Although t h i s i s a departure from Newtson's standard methodology i n which the f i l m i s , a r b i t r a r i l y divided into equal sized i n t e r v a l s (Newtson, 1976), i t was presumed to be an improvement f o r the following reason. The c r i t e r i o n f o r deciding the length of a reference i n t e r v a l should maximize the p o s s i b i l i t y that meaningful s i m i l a r i t i e s and differences between i n d i v i d u a l s are not obscured. If a completely a r b i t r a r y reference i n t e r v a l length i s assigned, i t increases the p o s s i b i l i t y that two i n d i v i d u a l s who a c t u a l l y press the button within the same i n t e r v a l of meaning are coded d i f f e r e n t l y . Conversely i t increases the p o s s i b i l i t y that two subjects who, i n r e a l i t y , attend to d i f f e r e n t units of meaning, are coded within the same i n t e r v a l . The speech act i s only one index of the units of meaning within a s o c i a l i n t e r a c t i o n , since within any given speech act there can s t i l l be t r a n s i t i o n s i n meaning as a function of both s h i f t s i n the content of communication as well as s h i f t s i n the metamessage which i s being communicated. The l a t t e r i s strongly influenced by nonverbal behaviours which are not necessarily bounded by speech act units. While i t i s true that the speech act i s only one guide as to the units of meaning, i t has the d i s t i n c t advantage of being e a s i l y and r e l i a b l y i d e n t i f i a b l e , and i n any event i s c e r t a i n l y preferable to s e l e c t i n g reference i n t e r v a l s on an a r b i t r a r y basis. The following procedure was employed for coding the reference i n t e r v a l s . The videotape scene was shown - 52 -to two independent judges, who were also provided with a written t r a n s c r i p t of the vignette. Both judges were advanced psychology undergraduates. One was male and the other was female. The judges were f i r s t shown the videotape scene an i n i t i a l time from beginning to end, without int e r r u p t i o n s . They were then shown the videotape a second time, during which they were asked to follow along with the t r a n s c r i p t . Whenever the husband completed a speech act, the videotape was stopped, and they were asked to categorize the speech act as ei t h e r dominant, agreeable, or neutral. They were permitted to review relevant sections of the videotape as many times as they needed to, i n order to f e e l confident about t h e i r codings. O v e r a l l i n t e r - r a t e r agreement f o r a l l i n t e r v a l s categorized was only f i f t y - n i n e percent. However, on i n t e r v a l s which subjects had previously categorized as either dominant or agreeable with over eighty percent agreement using only the written t r a n s c r i p t , the independent judges achieved one hundred percent agreement, both between themselves, and with the previous ratings. S i m i l a r l y , i n t e r v a l s which had previously exceeded the c r i t e r i o n for c l a s s i f i c a t i o n as neutral i n t e r v a l s were categorized by the independent raters as neutral with one hundred percent r e l i a b i l i t y . Memory Recognition Test As previously discussed, an important objective i n designing the recognition memory test was to devise a format which would enable us to d i s t i n g u i s h between s e l e c t i v e memory due to biased r e t r i e v a l processes and s e l e c t i v e memory due to biased reconstructive processes. This necessitated - 53 -the i n c l u s i o n of dominant, agreeable and neutral items which were a c t u a l l y drawn from the vignette (true items) as well as statements which the husband might p l a u s i b l y have made, but which were not drawn from the vignette ( f a l s e items). With t h i s type of format, biased r e t r i e v a l would manifest i t s e l f as se l e c t i v e recognition accuracy f o r true items consistent with the expectancy condition, whereas biased reconstruction would manifest i t s e l f as a tendency to i n c o r r e c t l y recognize f a l s e items consistent with the expectancy condition. The true memory items were selected simply on the basis of the vignette i n t e r v a l s which had previously exceeded the minimally acceptable r e l i a b i l i t y c r i t e r i a . There were twenty-four such items. There were two important constraints i n developing the f a l s e memory items. F i r s t , they had to appear pl a u s i b l e within the context of the vignette from which they were supposedly drawn. Second, they had to be r e l i a b l y i d e n t i f i a b l e as dominant, agreeable, or n e u t r a l . An i n i t i a l pool of f a l s e memory items was created by generating seventy-two statements which would have been meaningful responses i n context of the vignette but which had never a c t u a l l y been made. Twenty-four of these statements were designed to appear dominant i n nature, twenty-four were designed to appear agreeable i n nature, and twenty-four were designed to be neutral. To i l l u s t r a t e the manner i n which these responses were devised, consider the following example: In one section of the vignette, the wife, Susan, asks her husband, David, i f she can borrow the car. In the vignette, David responds: "Sure, I ' l l take the bus to work." This i n t e r v a l was categorized as agreeable. The following f a l s e a l t e r n a t i v e responses were devised: - 54 -1. f a l s e dominant: "No ... I need the car a l l day tomorrow." 2. f a l s e agreeable: "Sure ... you know I'm always happy to lend you the car i f you need i t . " 3. f a l s e neutral: "Do you need the car i n the morning?" The next step involved the s e l e c t i o n of a subset of those items, which could be r e l i a b l y categorized as dominant, agreeable, or neutral. A new sample of forty-seven psychology undergraduates were instructed to read the vignette. When they had completed t h i s task, they were given a l i s t of the seventy-two f a l s e items presented i n random order, and instructed i n the following manner: "For each of these statements, ask yourself the following question: "If the husband had made this statement during the vignette that I just read, would i t have been: 1) a dominant statement, 2) an agreeable statement, or 3) neutral." On t h i s basis I would l i k e you to categorize each statement as e i t h e r dominant (D), agreeable (A), or neutral (N)." The subjects categorized twelve dominant items and twelve agreeable items with greater than eighty percent i n t e r - r a t e r agreement. They also categorized twelve neutral items with greater than seventy percent 2 i n t e r - r a t e r agreement. This subset of t h i r t y - s i x f a l s e memory items was See Appendix 3 - 55 -then combined with the twenty-four true memory items, to form a memory 3 recognition t e s t with a t o t a l of s i x t y items. Interpersonal Adjective Scales In order to evaluate subjects' perception of the husband's interpersonal behavior i n the vignette, a set of eight adjective scales was administered. These eight scales were selected from amongst the sixteen adjective scales developed by Wiggins (1979) to mark the p r i n c i p a l dimensions i n his circumplex taxonomy of interpersonal t r a i t s . Although only the dominant and agreeable scales were employed i n subsequent analyses i t was considered desirable to embed these two scales within a larger set of adjective scales i n order to render these two dimensions l e s s s a l i e n t , thus reducing the p o s s i b i l i t y that subjects would be responding to any demand c h a r a c t e r i s t i c s which might e x i s t (see Appendix 5). While only these s i x t y items were employed i n subsequent analyses, the - memory test which was a c t u a l l y administered to subjects contained the twenty-four true memory items plus the entire pool of f a l s e memory items which had I n i t i a l l y been generated. The a d d i t i o n a l t h i r t y - s i x items were included i n order to increase the d i f f i c u l t y of the test by increasing the number of memory discriminations which subjects were required to perform. This was considered to be desirable since p i l o t research had found extremely good memory for items drawn from the vignette and there was some concern that such excellent o v e r a l l memory would obscure the presence of any s e l e c t i v e memory e f f e c t . The order of the items i n the memory test was determined on a random basis (see Appendix 4 for memory recognition t e s t ) . - 56 -EXPERIMENTAL PROCEDURE Subjects A l l subjects were psychology undergraduates attending the University of B r i t i s h Columbia. P a r t i c i p a t i o n i n the study was completely voluntary, and there was no form of c r e d i t or f i n a n c i a l remuneration given. Subjects were t o l d i n advance that the experiment was designed to investigate the manner i n which people form impressions of other people, and that i t would involve watching some videotape scenes i n which people were i n t e r a c t i n g , and subsequently answering some questions about the people they had seen. One hundred subjects parti c i p a t e d i n the study. Of these, sixty-nine were female, and thirty-one were male. Subjects were randomly assigned to e i t h e r the dominant or the agreeable expectancy conditions. There were thus f i f t y subjects i n each condition. An attempt was made to balance the experimental conditions with respect to sex. There were t h i r t y - f i v e females and f i f t e e n males i n the dominant condition, and t h i r t y - f o u r females and sixteen males i n the agreeable condition. Practice T r i a l s Subjects were run i n groups ranging i n siz e from one to f i v e . Upon a r r i v a l , a l l subjects were asked to read and sign a consent form (see appendix 6). They then received two prac t i c e t r i a l s on the tracking task p r i o r to the commencement of the experiment proper. It has been argued elsewhere ( T r i e r w e i l l e r , Note 2) that no prac t i c e t r i a l s should be employed - 57 -i n u n i t i z i n g studies since to do so would be to create an a r t i f i c i a l s i t u a t i o n which might not y i e l d a f a i t h f u l p i c t ure of the manner i n which u n i t i z i n g takes place i n r e a l l i f e . I t would seem, however, that tracking s u b j e c t i v e l y s a l i e n t events with a pushbutton i s a somewhat unfamiliar a c t i v i t y to begin with, even i f i n d i v i d u a l s do normally process information s e l e c t i v e l y . The novelty of the tracking task might thus r e s u l t i n a decrement i n r e l i a b i l i t y which conceivably could be reduced through pra c t i c e . Indeed, despite the fa c t that Newtson (Note 1) has demonstrated good r e l i a b i l i t y i n u n i t i z i n g , i t has been demonstrated that good r e l i a b i l i t y with more complex and r e a l i s t i c s o c i a l s t i m u l i i s d i f f i c u l t to achieve i n the absence of p r a c t i c e t r i a l s ( T r i e r w e i l l e r , Note 2). The practice t r i a l s consisted of viewing and tracking two videotape vignettes i n which a husband and wife were seen i n t e r a c t i n g with one another. Each vignette was approximately f i v e minutes i n length, and contained a d i f f e r e n t set of actors. This precaution was employed i n order to reduce the p o s s i b i l i t y that subjects would develop an impression bias with regard to either of the actors i n the experimental vignette on the basis of previous experience. Subjects were asked to track the behaviour of the wife on both pr a c t i c e vignettes i n order to reduce the p o s s i b i l i t y that a bias i n the perception of male characters would be developed. They were introduced to the tracking task i n the following manner: We're interesed i n the impressions you form of d i f f e r e n t people. We're also i n t e r e s t e d i n f i n d i n g out where you get the information which you use i n forming these impressions. In a moment I'm going to show you a short f i l m i n which a husband and wife are i n t e r a c t i n g with one another. I'm interested i n the impression you form of - 58 -the wife. I also want to f i n d out where you get the information which you use to form an impression of her. While you are watching the f i l m , I would l i k e you to press the button whenever she says something which i s h e l p f u l to you i n forming an impression of her. The button i s attached to some equipment which w i l l keep a record of when you press i t . There i s no right way or wrong way to do the task. We just want to know what information i s h e l p f u l to you i n forming your impression. Subjects were also instructed not to look at one another or to communicate amongst themselves i n any manner while the experiment was on, i n order to prevent them from influencing one another. The event recorder was set up i n a d i f f e r e n t room to eliminate the p o s s i b i l i t y that the noise of the pens would function as a cue. Subjects were instruc t e d to hide t h e i r pushbuttons from one another, and the buttons themselves did not emit any noise when pressed. Following the i n s t r u c t i o n s , subjects viewed and tracked the f i r s t p ractice scene. When they were f i n i s h e d they were reassured that they were performing the tracking task c o r r e c t l y . This assurance was designed to increase the p r o b a b i l i t y that subjects would develop a consistent tracking strategy which was subj e c t i v e l y meaningful to them and to reduce random responding which might r e s u l t from a lack of confidence i n t h e i r a b i l i t y to perform the task. At t h i s point the relevant i n s t r u c t i o n s were repeated to prepare subjects f o r the second t r i a l and they were then shown the second vignette. No record was kept of the tracking f o r either of the two t r i a l scenes. Following the completion of the second t r i a l , the subjects were once again reassured, and the experiment proper commenced. - 59 -Procedures A l l subjects were given the following i n s t r u c t i o n s : In a moment I'm going to show you another scene i n which a man and his wife are i n t e r a c t i n g . The husband's name i s David and the wife's name i s Susan. This time I'd l i k e you to watch the f i l m , and to press the button whenever the husband, David, says something which i s h e l p f u l to you i n trying to form an impression of him. Before we begin, however, I'm going to give you some information about David i n advance. I'm going to show you some pictures of him and describe to you a number of d i f f e r e n t thimgs that he has done i n the l a s t two weeks. Subjects were then read the information which was appropriate to t h e i r expectancy condition (Appendix 1). While t h i s information was being read they were shown a s e r i e s of colour s l i d e s which had been taken of the actor who portrayed David i n the f i l m . This was done i n order to increase the impact of the expectancy information by anchoring i t to v i s u a l images, thus making i t more v i v i d and l e s s abstract (Nisbett & Ross, 1980). The same s l i d e s were presented i n the same order, i n both conditions. Following the presentation of the expectancy information, subjects viewed and tracked the stimulus vignette. A record of the tracking was kept, using the event recorder. When they had f i n i s h e d tracking the f i l m a ten-minute delay i n t e r v a l was introduced p r i o r to the administration of the recognition memory t e s t . The purpose of t h i s delay i n t e r v a l was to decrease the p r o b a b i l i t y that subjects would be able to r e t r i e v e information from short-term memory (Kintsch, 1975; Smith & M i l l e r , 1979). During t h i s delay i n t e r v a l subjects were given a f i l l e r task i n order to prevent the employment of any memory rehearsal s t r a t e g i e s . Since i t was important to - 60 -avoid arousing suspicions that a memory test would follow, a f i l l e r task was devised which would appear plausible i n context of the rest of the experiment. Subjects were given a handout upon which were printed a serie s of b r i e f descriptions of i n d i v i d u a l s , and were asked to match these descriptions to ten pictures of i n d i v i d u a l s , which were shown on the s l i d e projector. For each s l i d e , the experimenter waited u n t i l a l l subjects f e l t they had i d e n t i f i e d the appropriate d e s c r i p t i o n on t h e i r answer sheets, before proceeding to the next s l i d e . At the end of the ten minute delay i n t e r v a l the f i l l e r task was ended, regardless of how many s l i d e s had been i d e n t i f i e d (see Appendix 7 for a copy of the f i l l e r task). When the f i l l e r task had been completed, the memory recognition test was administered. The following task i n s t r u c t i o n s were employed: On the handout i n front of you are a serie s of statements. Some of these statements were made by David, i n the l a s t f i l m that you saw, and some of them were not. For each statement, I would l i k e you to put a number ranging from 1 to 6, i n d i c a t i n g how confident you are that David d i d or did not make that exact statement i n the f i l m . Remember, a 1 means you're completely confident that David did not make the statement. A 6 indicates that you're completely confident that David did make the statement. 2, 3, 4 and 5 indic a t e varying degrees of confidence between those two extremes. Remember that I want to know i f he made that exact statement i n the f i l m . A copy of the rating scale was provided to the subjects and they were given as much time as they required to complete the memory task. When they were completed, the memory test forms were taken away from them and the adjective scales were administered. Subjects were given copies of the adjective scales and then issued the following i n s t r u c t i o n s : - 61 -I'd l i k e to f i n d out your impression of David's behaviour i n the l a s t f i l m that you saw. For each group of adjectives, please c i r c l e the number which best indicates how accurately those adjectives summarize your impression of David. Please think back to the f i l m while doing t h i s . The rating scale which was employed ranged from 1 (extremely inaccurate) to 8 (extremely accurate) and a copy of t h i s r ating scale was included i n the handout. Following the completion of the adjective scales, subjects were debriefed, thanked f o r t h e i r co-operation, and then dismissed. - 6 2 -RESULTS Impression Formation The hypothesis that the expectancy information would bias subjects' impression of the husband i n an expectation consistent manner was evaluated by comparing the mean ratings of the two experimental conditions on the dominant and agreeable adjective scales. These data were analyzed with a 2 X 2 X 2 (expectancy condition X sex X adjective scales) between-within ANOVA design with repeated measures on the adjective scales. There was a s i g n i f i c a n t i n t e r a c t i o n between the expectancy conditions and the two adjective scales, F(l,96) = .40.842, p < .001. As predicted, subjects given the dominant expectation rated the husband higher on the dominant adjective scale and lower on the agreeable adjective scale. This pattern was reversed for the agreeable subjects (see Table 4). The absence of a main e f f e c t f or the sex f a c t o r , F(l,96) = .001, p > .971, indicates that male and female subjects saw the husband as equally dominant and agreeable. Moreover, there were no s i g n i f i c a n t i n t e r a c t i o n s between: sex X expectancy, sex X adjectives scales, or sex X expectancy X adjective scales, F(l,96) = 2.398, p > .125, F(l,96) = 2.315, p > .131, and F(l,96) = .015, p > .904 re s p e c t i v e l y . These r e s u l t s confirm that the experimental manipulation was equally e f f e c t i v e f o r male and female subjects. There were s i g n i f i c a n t main e f f e c t s f or both expectancy condition, F(l,96) = 7.870, p < .01, and adjective scale r a t i n g , F(l,96) = 15.965, p < .001. These re s u l t s i n d i c a t e that subjects given the dominant expectation rated the husband higher on both adjective scales than did agreeable subjects and that subjects o v e r a l l saw the husband as more agreeable than dominant. TABLE 4 Group Means f o r Impression Formation Data ADJECTIVE SCALE RATING EXPECTANCY MANIPULATION Dominant Agreeable N X (SD) N X (SD) Dominant 50 6.16 (1.42) 50 5.70 (1.58) Agreeable 50 3.86 (1.77) 50 6.74 (1.14) - 64 -Tracking Data The main analysis of the tracking data was conducted using Hotelling's T-Square s t a t i s t i c (Myers, 1979). This was accomplished i n the following fashion. Twenty-four i n t e r v a l s of the f i l m were selected f o r analysis i n the manner previously described. For each i n t e r v a l , the number of subjects who had pressed the button within each expectancy condition was tabulated. In t h i s manner, the proportion of subjects who tracked i n t e r v a l s one through twenty-four was obtained independently f o r a) dominant, and b) agreeable 4 expectancy conditions. After the tracking had been tabulated the i n t e r v a l s were divided into three subsets of eight i n t e r v a l s : 1) dominant, 2) agreeable, and 3) neutral. Separate Hotelling's T-Squares were employed to analyze these three subsets. Sometimes a given subject pressed the button more than once during a given i n t e r v a l . In this case only one button press was counted f o r that subject. A de c i s i o n was made to tabulate the tracking data i n t h i s manner since the t h e o r e t i c a l question of i n t e r e s t had been phrased i n a dichotomous or a l l or nothing manner, i . e . , "Does the subject f i n d the i n t e r v a l h e l p f u l i n forming an impression or not?" In addition i t seemed that i f the data were tabulated i n t h i s manner, the group means would c l e a r l y i n d i c a t e what proportion of the subjects i n each group had designated that i n t e r v a l as important. On the other hand, i t could be argued that tabulating the data i n t h i s manner makes use of only part of the data a v a i l a b l e , and thus p o t e n t i a l l y misrepresents the true state of a f f a i r s . For t h i s reason a separate analysis was conducted i n which a l l button presses f o r every subject were tabulated. The r e s u l t s of t h i s a n alysis are reported i n Appendix 8. The general pattern of r e s u l t s obtained using the two d i f f e r e n t tabulation procedures was i d e n t i c a l . - 65 -For each subset, the eight i n t e r v a l s were considered to be separate dependent v a r i a b l e s . The analysis f o r each subset thus addressed the mult i v a r i a t e question: "Are the mean vectors of the dominant and agreeable expectancy conditions s i g n i f i c a n t l y d i f f e r e n t ? " Since the data entry f o r each subject i n any given reference i n t e r v a l was r e s t r i c t e d to one or zero, the group mean f o r any given i n t e r v a l represents the proportion of subjects i n the group who tracked that i n t e r v a l . On the dominant reference i n t e r v a l s there was an o v e r a l l s i g n i f i c a n t 2 differe n c e between dominant and agreeable subjects, T = 30.017, F(8,91) = 3.484, p < .005. As predicted, dominant subjects tracked dominant i n t e r v a l s to a greater extent than did agreeable subjects (Table 5). While t h i s was true for every one of the dominant i n t e r v a l s , a multiple comparison procedure which set experiment-wise error rate to p < .05 (Myers, 1979) f a i l e d to f i n d s i g n i f i c a n t d ifferences on any of the i n d i v i d u a l mean comparisons. The analysis of agreeable i n t e r v a l s yielded a complementary pattern of r e s u l t s . The mean vectors f o r dominant and agreeable subjects were s i g n i f i c a n t l y d i f f e r e n t , T 2 = 23.112, F(8,91) = 2.683, p < .01, with agreeable subjects tracking agreeable i n t e r v a l s more frequently than dominant subjects. Once again multiple comparisons f a i l e d to y i e l d any s i g n i f i c a n t i n d i v i d u a l comparisons despite the fac t that agreeable subjects c o n s i s t e n t l y scored higher than dominant subjects on every agreeable i n t e r v a l (Table 6). - 66 -TABLE 5 Group Means on Tracking Task (Dominant Intervals) Dominant Agreeable X (SD) X (SD) 1. Susan ... could you c a l l Terry back ... I'm expecting an important c a l l . .66 (.48) .32 (.47) 2. No ... y o u ' l l have to c a l l her back Susan ... I don't want to miss t h i s c a l l . • .62 (.49) .48 (.51) 3. Look ... just make up your mind and give them a c a l l . .86 (.35) .68 (.47) 4. I'm ju s t not i n the mood for Merv G r i f f i n tonight. Common ... l e t ' s have a drink. .52 (.51) .42 (.50) 5. L i s t e n Susan. We owe them a dinner and I r e a l l y want to have them over. .64 (.49) .44 (.50) 6. Don't go storming o f f Susan. S i t down ... That's better. .66 (.78) .46 (.50) 7. No. I just can't afford the extra time i t would take to go to New York. .80 (.40) .48 (.51) 8. Don't pick up that glass with your hands Susan. Use the broom and dustpan. .80 (.40) .74 .44 p <.005 - 67 -TABLE 6 Group Means on Tracking Task (Agreeable Intervals) Dominant Agreeable X (SD) X (SD) 1. Well that sounds good to me too. We can go out f o r Chinese food another time. .64 (.49) .88 (.33) 2. Okay honey. I ' l l t r y to make i t the way you l i k e i t this time. .66 (.48) .70 (.46) 3. Sure. I ' l l take the bus to work. .68 (.47) .72 (.45) 4. That's a good idea. I t ' s a r e a l l y nice dish. .28 (.45) .34 (.48) 5. Okay honey. I'm happy to go along with whatever. .58 (.50) .70 (.46) 6. Sure. I'd be glad to help your s i s t e r move. .64 (.49) .90 (.30) 7. Okay honey. I can always c a l l from the restaurant. .50 (.51) .76 (.43) 8. Okay honey. I ' l l go the the g a l l e r y with you a f t e r dinner. .50 (.51) .76 (.43) p < .01 - 68 -TABLE 7 Group Means on Tracking Task (Neutral Intervals) Dominant Agreeable 1. It was a c t u a l l y quite bad. 2. Oh ... when did he get back into town? 3. Oh ... I see i t . We're almost out, you know. 4. Yeah ... I've got i t . 5. That's encouraging. I hope i t ' s accurate. 6. Not exactly. How about you? 7. What type of a r t does i t have? 8. Okay ... I ' l l be waiting outside. I want to clean the windshield. X (SD) ,08 (.27) ,00 (.00) .04 (.20) .02 (.14) .08 (.27) .04 (.20) .02 (.14) .10 (.30) X (SD) .08 (.27) .04 (.20) .02 (.14) .00 (.00) .14 (.35) .06 (.24) .02 (.14) .16 (.37) - 69 -The T-square analysis on neutral intervals could not be computed since the variance-covariance matrix could not be inverted. An .examination of Table 7 indicates that more than one of the interval means were identical. The variance-covariance matrix could therefore not be inverted due to linear dependencies in the data. In order to compare the tracking of neutral reference intervals to the tracking of dominant and agreeable intervals a cumulative mean score for each of the three subsets of intervals was obtained by summing the individual interval means within each subset. These three cumulative indices were then analyzed as repeated measures in a 2 X 2 X 2 (expectancy condition X sex X interval type) between-within repeated measures ANOVA. There was a significant main effect due to interval type, F(2,192) = 240.577, p < .001, and a significant interaction between expectancy condition and interval type, F(2,192) = 20.406, p < .001. Post hoc comparisons on c e l l means were conducted using Bonferonni t-tests (Myers, 1979). A l l _t-tests were two-tailed. These comparisons demonstrated that overall, subjects tracked both dominant and agreeable intervals significantly more frequently than neutral intervals, t(99) = 22.50, p < .0001 and t(99) = 20.82, p < .0001 respectively. There were no significant differences overall between dominant and agreeable intervals t(99) = 1.14, p > .257 (Table 7). Subjects in the dominant expectancy ;condition tracked dominant intervals more frequently than did subjects in the agreeable condition, t(98) = 4.16, p < .0001, and subjects in the agreeable condition tracked agreeable intervals more frequently than did subjects in the dominant condition, t(98) = 3.01, p <.05 . There were however, no significant differences between - 70 -subjects in the two conditions on neutral i n t e r v a l s , t(98) = .75, p > .226 (Table 8). Subjects i n the dominant expectancy condition tracked dominant intervals: s i g n i f i c a n t l y more frequently than agreeahle i n t e r v a l s , t(49) ^ 2.89, p < .05. Conversely, subjects who received the agreeable manipu-l a t i o n tracked agreeable i n t e r v a l s s i g n i f i c a n t l y more frequently than dominant i n t e r v a l s , t(49) = 5.08, p <.0001 (Table 8). There was no s i g n i f i c a n t main e f f e c t due to subject sex, F(l,96) = 1.142, p > .288, and no s i g n i f i c a n t i n t e r a c t i o n between sex, expectancy condition, and i n t e r v a l type, F(2,192) = 1.719, p > .182. There was a s i g n i f i c a n t i n t e r -action between subject sex and i n t e r v a l type, F(2,192) = 3.83, p < .05. Male subjects- tracked dominant i n t e r v a l s -more frequently, whereas female subjects tracked agreeable i n t e r v a l s more frequently. C o r r e l a t i o n a l Analyses A f i n a l question which was asked with regard to the tracking data was; to what extend could the subjects'- ratings of dominance and agreeableness be pre-di c t e d on the basis of t h e i r tracking behaviour? To answer t h i s question, Pearson r c o r r e l a t i o n s were computed between the cumulative mean scores for each of the three subsets of reference i n t e r v a l s and subjects'- ratings of a) dominance, and b) agreeableness. Collapsing across both expectancy- conditions, the c o r r e l a t i o n between dominant i n t e r v a l s and ratings of dominance was r - «34 ? p <j .0001, while the c o r r e l a t i o n between tracking dominant i n t e r v a l s and ratings of agreeableness was r = -.39, p <• .0001. The c o r r e l a t i o n between tracking agreeable i n t e r v a l s and ratings of dominance was r = -.25, p < ,01, while the - 71 -TABLE 8 Group Means on Tracking Task (Intervals Summed) INTERVAL TYPE EXPECTANCY MANIPULATION Dominant Agreeable Neutral X (SD) X (SD) X (SD) Dominant Subjects 5.56 (1.57) 4.48 (2.35) .38 (.73) Agreeable Subjects 4.02 (2.10) 5.76 (1.87) .52 (1.09) A l l Subjects 4.79 (2.00) 5.12 (2.21) .45 (.93) A l l Subjects (average i n t e r v a l ) 59.50 64.00 .056 - 72 -c o r r e l a t i o n between tracking agreeable i n t e r v a l s and ratings of agreeableness was r = .29, p •< .005. The c o r r e l a t i o n s between tracking neurtal i n t e r v a l s and ratings of dominance and agreeableness were r = -.13, n.s., and r = -.09, n.s., r e s p e c t i v e l y (Table 9). Thus, as expected, subjects who frequently tracked dominant i n t e r v a l s were l i k e l y to rate the husband as dominant and u n l i k e l y to rate the husband as agreeable, whereas subjects who frequently tracked agreeable i n t e r v a l s tended to perceive the husband i n an opposite fashion. Again as expected, there was no s i g n i f i c a n t r e l a t i o n s h i p between tracking n e u r t r a l i n t e r v a l s and r a t i n g the husband as dominant or agreeable. A second analysis was conducted to c o r r e l a t e subjects' tracking behaviour to t h e i r ratings of dominance and agreeableness, within expectancy conditions. In the dominance expectancy condition, there was a s i g n i f i c a n t negative c o r r e l a t i o n between tracking agreeable i n t e r v a l s and r a t i n g the husband as dominant, r = -.31, p <! .05. In the agreeable expectancy condition there was a s i g n i f i c a n t negative c o r r e l a t i o n between tracking dominant i n t e r v a l s and r a t i n g the husband as agreeable, r - -.37, p < .005. These r e s u l t s suggest, that subjects whose tracking behaviour was not biased by the expectancy information, were u n l i k e l y to rate the husband i n an expectation-consistent fashion. Within expectancy conditions, the c o r r e l a t i o n s between tracking expectation-consistent information and r a t i n g the husband i n an expectation-consistent fashion were not s i g n i f i c a n t . The fact that these c o r r e l a t i o n s were not s i g n i -f i c a n t suggests that they may have been spuriously i n f l a t e d by c o l l a p s i n g across expectancy conditions. - 73 -TABLE 9 Correlations Between Tracking Behavior and Adjective Ratings ADJECTIVE RATING TRACKING (INTERVAL TYPE) Dominant Agreeable Dominant ,34 *** -.39 *** Agreeable -.25 * 29 ** Neutral -.13 -.09 * p < .01 ** p < .005 *** p < .0001 - 74 -Memory Data As previously discussed there were two d i s t i n c t subsets of memory items: a) true items, and b) f a l s e items. These two item subsets were analyzed independently of one another. True Memory Items For every subject, o v e r a l l accuracy indices f or dominant, agreeable and neutral items, were obtained by summing confidence ratings f or the eight dominant items, eight agreeable items, and eight neutral items, re s p e c t i v e l y . These three accuracy i n d i c e s were then analyzed as repeated measures i n 2 X 2 X 3 (expectancy condition X subject sex X memory item type) between - within ANOVA. There was a s i g n i f i c a n t main e f f e c t f o r memory item type, F(2,192) = 65.887, p < .001. Bonferonni t - t e s t s indicated that c o l l a p s i n g across conditions, subjects were s i g n i f i c a n t l y more accurate i n both dominant and agreeable items than they were on neutral items, t(99) = 10.39, p <.0001, and t(99) = 7.37, p < .0001 respectively. Subjects were also s i g n i f i c a n t l y more accurate on dominant than on agreeable items, t(99) = 4.24, p < .0001 (Table 10). There were no other s i g n i f i c a n t e f f e c t s i n the a n a l y s i s . Thus, no evidence was found of s e l e c t i v e accuracy f o r memory items which were consistent with expectancy conditions or f o r d i f f e r e n t i a l accuracy between sexes. False Memory Items Ov e r a l l confidence indices f or dominant, agreeable, and neutral items were ca l c u l a t e d by summing the appropriate i n d i v i d u a l items. These data - 75 -were also analyzed with a 2 X 2 X 3 (expectancy condition X subject sex X memory item type) repeated measures ANOVA. There were no s i g n i f i c a n t e f f e c t s i n t h i s analysis (Table 11). Thus no evidence was found f o r expectation consistent memory reconstruction. - 76 -TABLE 10 Group Means f o r True Memory Items MEMORY ITEM TYPE EXPECTANCY CONDITION Dominant Agreeable Neutral X (SD) X (SD) X (SD) Dominant Subjects 43.04 (6.42) 41.54 (5.73) 38.02 (5.51) Agreeable Subjects 44.12 (4.12) 41.90 (4.61) 37.08 (5.09) A l l Subjects 43.58 (5.40) 41.72 (5.18) 37.55 (5.30) - 77 -EXPECTANCY CONDITION Dominant Subjects Agreeable Subjects A l l Subjects TABLE 11 Group Means f o r False Dominant X (SD) 29.76 (11.59) 28.08 (8.47) 28.92 (10.13) Memory Items MEMORY ITEM TYPE Agreeable X (SD) 30.52 (8.20) 31.32 (8.82) 30.92 (8.48) Neutral X (SD) 30.22 (7.76) 30.20 (7.57) 30.21 (7.63) - 78 -DISCUSSION General Discussion The present study c l e a r l y demonstrates that expectations can bias the formation of interpersonal impressions and implicates the s e l e c t i v e encoding of interpersonal behaviour as a mediating mechanism. Subjects tended to track i n t e r v a l s of the f i l m which were consistent with the expectations which had been induced and subsequently formed impressions of the target person which were consistent with both the expectation which had been induced as well as the type of i n t e r v a l s they had tracked. The analysis of between group differences i n the tracking data supported the major hypothesis that subjects would be more l i k e l y to f i n d h e l p f u l and to use information which was consistent with t h e i r i n i t i a l expectations than information which was not. Subjects who had received the dominant expectancy manipulation tracked dominant i n t e r v a l s of the i n t e r a c t i o n s i g n i f i c a n t l y more frequently than subjects who had received the agreeable expectancy manipulation. Conversely subjects who had received the agreeable expectancy manipulation tracked agreeable i n t e r v a l s s i g n i f i c a n t l y more frequently than did subjects who had received the dominant expectancy manipulation. While t h i s pattern emerged co n s i s t e n t l y for a l l i n t e r v a l s which had been categorized as dominant or agreeable, none of the between group differences on i n d i v i d u a l i n t e r v a l s reached s i g n i f i c a n c e when a conservative multiple-comparison procedure was employed. This f a c t , i n addition to simple inspection of the i n t e r v a l means of the two groups (Tables 5 & 6), suggests that the s e l e c t i v e encoding by no means en t a i l e d a - 79 -complete exclusion of information which was not consistent with expectations, but rather e n t a i l e d consistent and s i g n i f i c a n t preference f o r information which was expectation-consistent. This preference ranged i n magnitude from f a i r l y sizeable differences on some i n t e r v a l s (e.g., thirty-two out of f i f t y dominant subjects compared to sixteen out of f i f t y agreeable subjects tracked the f i r s t dominant i n t e r v a l ) , to moderately sized preferences on others (e.g., t h i r t y - t h r e e out of f i f t y dominant subjects compared to twenty-three of the f i f t y agreeable subjects tracked the s i x t h dominant i n t e r v a l ) to n e g l i g i b l e between-group differences on a few of the i n t e r v a l s (e.g., t h i r t y - f i v e out of f i f t y agreeable subjects compared to t h i r t y - t h r e e of the f i f t y dominant subjects tracked the second agreeable i n t e r v a l ) (Tables 5 & 6). It appears then that information which was unanticipated was s t i l l picked up by the subjects, a l b e i t not as frequently as information which was expected. This f i n d i n g i s consistent with Neisser's (1976) p o s i t i o n that " ... perception i s d i r e c t e d by expectations but not c o n t r o l l e d by them." (p. 43). The lack of s i g n i f i c a n t between-group differences on the neutral i n t e r v a l s suggests that encoding differences were s p e c i f i c to relevant information. Since one would expect s e l e c t i v e encoding of relevant information but not of i r r e l e v a n t information, this f i n d i n g provides further evidence that the tracking procedure provided a v a l i d index of encoding a c t i v i t y . The f i n d i n g that subjects i n both conditions tracked s i g n i f i c a n t l y fewer neutral i n t e r v a l s than either dominant or agreeable - 80 -i n t e r v a l s suggests that both schema consistent and inconsistent information were more s a l i e n t to subjects than schema i r r e l e v a n t information ( c f . Hastie & Kumar, 1979). 5 Although the t r a i t s of dominance and agreeableness are t h e o r e t i c a l l y orthogonal (Wiggins, 1979) subjects i n the present study saw the two as s i g n i f i c a n t l y negatively correlated, r = -.47, p < .0001. This departure from expectations may have resulted from the f a c t that i n the present context subjects were ascribing t r a i t s to another i n d i v i d u a l rather than to themselves, whereas Wiggins' (1979) taxonomy of interpersonal t r a i t s i s derived from s e l f - r a t i n g s . The a s c r i p t i o n of t r a i t s to others may follow a d i f f e r e n t set of rules than the a s c r i p t i o n of t r a i t s to the s e l f . It i s also possible that the perceived negative r e l a t i o n s h i p between dominance and agreeableness i n the present study may r e s u l t from the contrived nature of the vignette. In contrast to r e a l l i f e i n t e r a c t i o n s i n which a v a r i e t y of interpersonal behaviours f a l l i n g along various vectors of the circumplex may occur, the vignette i n the present study was designed to h i g h l i g h t dominance and agreeableness almost to the exclusion of other interpersonal behaviours. By narrowing the range of interpersonal behaviours a v a i l a b l e to subjects for comparison purposes, dominant and agreeable behaviours may have appeared to be more negatively correlated than they might i n a more n a t u r a l i s t i c s e t t i n g . A f i n a l p o t e n t i a l factor contributing to the r e l a t i v e lack of salience of neutral i n t e r v a l s i s that i n a d d i t i o n to being i r r e l e v a n t with respect to the t r a i t s of dominance and agreeableness, the neutral acts i n the vignette were quite noninformative with respect to any type of interpersonal judgment. Thus, statements such as: "Yeah, I've got i t , " and "What type of art does i t have?" would not be p a r t i c u l a r l y diagnostic, and were therefore not p a r t i c u l a r l y s a l i e n t to the subjects i n the present study. - 81 -The results of the correlational analyses were consistent with the analysis of between group differences. Collapsing across expectancy conditions, subjects who frequently tracked dominant intervals tended to see the husband as a dominant person and were unlikely to see him as agreeable. Conversely, subjects who fre-quently tracked agreeable intervals tended to see the husband as an agreeable person and were unlikely to see him as dominant. There were no significant relationships between the tracking of neutral intervals and ratings of dominance or agreeableness. Within both expectancy conditions, subjects whose tracking behaviour was not biased by the expectancy information, were unlikely to rate the husband in an expectation-consistent fashion. These findings, together with the analysis of between group differences in tracking data, provide a coherent and theoretically predictable pattern of results. Although the correlational analyses provide evidence regarding the relationship between selective encoding of interpersonal behaviour and impression formation, no causal inferences can be drawn on this basis. Any such inferences would require the direct manipulation of encoding activity i t s e l f . ^ 6 There is evidence in other contexts that the direct manipulation of stimulus salience affects both memory and subsequent attributions about interpersonal behaviour. Taylor and Fiske (1978) , for example, have demons strated in a number of studies that subjects are more lik e l y to attribute causal responsibility to individuals who are more visually salient in the dyadic or group interaction. They argue that simply manipulating visual salience increases the ease with which information about the stimulus person is encoded and thus increases the availability of information about the person in memory. Reyes, Thompson, & Bower (1980), have demonstrated that simply increasing the vividness of written information can increase the l i k e -lihood that information w i l l receive disproportionate attention and w i l l subsequently be weighted more heavily than other information when making legal decisions. - 82 -While the tracking data were consistent with the experimental hypotheses, there was no evidence of e i t h e r s e l e c t i v e memory r e t r i e v a l or s e l e c t i v e memory reconstruction consistent with subjects' expectations. The absence of any such evidence stands i n contrast to a number of studies which have demonstrated a memory bias consistent with i n i t i a l expectations ( c f . Hastie & Kumar, 1979; Rothbart, Evans, & Fulero, 1978; Zadny & Gerard, 1974), and may possibly be accounted for by the increased e c o l o g i c a l v a l i d i t y of the stimulus materials employed i n the present study. Just as i n r e a l l i f e i n t e r a c t i o n s , the information which subjects were subsequently asked to remember, was i n i t i a l l y embedded within the context of a complex s o c i a l i n t e r a c t i o n containing theme, structure, and a complex array of verbal and v i s u a l information. This context may have minimized expectation consistent r e t r i e v a l and reconstruction by providing subjects with s u f f i c i e n t r e t r i e v a l cues to cancel out any such e f f e c t s . ^ While Cohen (1981) di d f i n d an expectancy consistent memory bias even when using more e c o l o g i c a l l y v a l i d stimulus materials than have been common, i t should be noted that there i s an important d i f f e r e n c e between the stimulus materials employed i n her study and those employed i n the present study. Cohen had subjects watch a videotape i n which a woman interacted with her husband and then evaluated memory for various features which were either consistent or inconsistent with the stereotypes of waitress or l i b r a r i a n . These features were ei t h e r subtly slipped into the tape (e.g., artwork was placed on the w a l l ) , or integrated i n t o the conversation (e.g., she mentioned playing the gui t a r , or t r a v e l l i n g to Europe). In the present study the stimulus materials did not, i n the same manner, consist of a v a r i e t y of d i f f e r e n t features, but rather, e x c l u s i v e l y of interpersonal acts. Any given statement that the husband made was meaningfully embedded within a s p e c i f i c i n t e r a c t i o n sequence, and i n turn was meaningful only within the context of that i n t e r a c t i o n - 83 -A second way i n which the increased e c o l o g i c a l v a l i d i t y of the present study may have m i l i t a t e d against the emergence of a memory bias i s that subjects may have been more l i k e l y to encode events at the l e v e l of abstract meaning since the i n t e r a c t i o n they observed i n the present study was more l i f e l i k e and conducive to the consideration of abstract interpersonal issues than i s t y p i c a l l y the case. If the preferred mode of processing was at a f a i r l y abstract l e v e l i t i s possible that any memory s e l e c t i v i t y would be more l i k e l y to emerge at this l e v e l rather than at the l e v e l of surface d e t a i l . Although no expectation-consistent memory bias occurred, i t i s possible that such an e f f e c t would have emerged i f the delay i n t e r v a l between exposure to the stimulus vignette and the memory recognition test had been increased i n length. This e f f e c t would occur i f consistent and inconsistent information decayed at d i f f e r e n t rates. 7 (Cont'd.) sequence. Thus, f o r example, when the husband says: "I'm just not i n the mood f o r Merv G r i f f i n tonight ... come on ... l e t ' s have a drink", i t only makes sense i n context of the i n t e r a c t i o n the husband and wife are having about whether or not they should watch T.V. In turn, when one r e c a l l s the statement, one automatically r e c a l l s the relevant i n t e r a c t i o n sequence. In contrast, the type of stimulus features employed by Cohen (1981) are not n e c e s s a r i l y embedded within a s p e c i f i c i n t e r a c t i o n context. Thus one can r e c a l l that there was a picture on the wall, or r e c a l l that the woman plays a g u i t a r , without necess a r i l y r e c a l l i n g an interpersonal context i n which these features were embedded i n the f i l m . In short, there appears to be a greater degree of embeddedness or interrelatedness between stimulus features and context, i n the present study than i n Cohen's (1981), and t h i s greater embeddedness may have provided more r e t r i e v a l cues for subjects. - 84 -Although t h i s p o s s i b i l i t y was not ruled out i n the present study, i t i s noteworthy that the predicted impression formation bias did emerge a f t e r the ten minute delay i n t e r v a l without the apparent mediation of any memory bias. It would appear then that although biases i n memory r e t r i e v a l and reconstruction may r e s u l t from the schematic processing of s o c i a l information, they are not necessar i l y the mediating mechanisms through which expectations influence the perception of interpersonal behaviour. Furthermore, there i s some evidence that lengthening the delay i n t e r v a l does not n e c e s s a r i l y increase the degree of expectation-consistent -memory bias r e s u l t i n g from schematic; processing (Cohen, 1981). Although there was no evidence of expectation consistent r e t r i e v a l or reconstruction i n the present study, subjects i n both expectancy conditions were more accurate i n r e t r i e v i n g both dominant and agreeable items than they were i n r e t r i e v i n g neutral items and more accurate i n r e t r i e v i n g dominant items than they were i n r e t r i e v i n g agreeable items. The superior recognition f o r both dominant and agreeable items r e l a t i v e to neutral items p a r a l l e l s the findings that both dominant and agreeable subjects tracked s i g n i f i c a n t l y fewer neutral i n t e r v a l s than e i t h e r dominant or agreeable i n t e r v a l s . An examination of group means on the tracking task (Tables 5, 6 & 7) ind i c a t e s that indeed the neutral i n t e r v a l s received much less a t t e n t i o n from the subjects than either dominant or agreeable i n t e r v a l s . Inspection of the l a s t row of the data presented i n Table 8, indicates that the average neutral i n t e r v a l was tracked by less than 1 of the 100 subjects (.056 to be p r e c i s e ) . This compares with 59.5 of the one hundred subjects who tracked the average dominant i n t e r v a l , and 64 of the 100 subjects who tracked the average agreeable i n t e r v a l . This v i r t u a l absence of any - 85 -a t t e n t i o n whatsoever to neutral acts may account f o r the r e l a t i v e l y poor r e t r i e v a l of neutral i n t e r v a l s . The fi n d i n g that subjects showed more accurate recognition memory f o r dominant than f o r agreeable memory items was not p a r a l l e l e d by the tracking data. There i s thus no evidence to support the hypothesis that dominant acts were more s a l i e n t to subjects than agreeable acts. A possible explanation f o r the superior memorability of dominant items i s that they tended to be more thematically heterogeneous and thus more d i s t i n c t i v e than the agreeable items. The dominant acts tended to be quite varied i n nature i n the sense that some involved g i v i n g advice (e.g., Look ... just make up your mind and give them a c a l l ) , some involved forbidding the wife to do something (e.g., Don't go storming o f f , Susan ... s i t down), some involved dominating a d e c i s i o n about a mutual a c t i v i t y (e.g., I don't r e a l l y f e e l l i k e watching t.v. ... l e t ' s have a drink before dinner instead), and so on. In contrast, the agreeable items were r e l a t i v e l y homogeneous i n nature i n the sense that they a l l involved simply going along or accomodating (e.g., Okay honey, I'm happy to go along with whatever; or, Okay honey ... I ' l l go to the g a l l e r y with you a f t e r dinner). Also note that four of the eight agreeable items a c t u a l l y s t a r t with exactly the same words ( i . e . , Okay honey). Because of this lack of d i s t i n c t i v e n e s s , these items may have been less e a s i l y retrieveable from memory since they may have had fewer d i s t i n c t i v e r e t r i e v a l cues associated with them. Whether dominant interpersonal acts are inherently more memorable than agreeable acts or were only so because of the p a r t i c u l a r acts employed i n the present study, i s an empirical question. - 8 6 -It i s noteworthy that while subjects i n both expectancy conditions showed more accurate recognition f o r dominant than agreeable memory items, they did not see the husband i n the vignette as more dominant than agreeable. Quite the contrary, i n f a c t , they saw the husband as s i g n i f i c a n t l y more agreeable than dominant. This discrepancy h i g h l i g h t s the fact that i n the present study no consistent r e l a t i o n s h i p was found between the impressions that subjects formed of the husband, and either memory r e t r i e v a l or memory reconstruction. Selective Encoding or Selective Memory? The above pattern of r e s u l t s suggests that when complex, e c o l o g i c a l l y v a l i d stimulus materials are employed, expectations may influence the perception of interpersonal behaviour more through a process of s e l e c t i v e encoding than through biased memory f o r events. How i s i t possible f o r se l e c t i v e encoding to bias impression formation without the mediating mechanism of s e l e c t i v e memory f o r events? One p o s s i b i l i t y i s that subjects discount expectation inconsistent information rather than forgetting or d i s t o r t i n g i t . This discounting may take the form of simply deciding that a given piece of information i s i r r e l e v a n t to the task of impression formation, or i t may take a more sophisticated form i n which the subject a t t r i b u t e s expectation inconsistent behaviours to suspect motives. Subjects when queried about th e i r thought process during the debriefing session would often make comments such as: "I thought that the husband was b a s i c a l l y a dominant person, who did agreeable things i n order to get his own way". Future research should investigate the p o s s i b i l i t y of t h i s type of discounting process. - 87 -A related p o s s i b i l i t y i s that although there were no expectation consistent biases i n memory f o r the dialogue of the vignette, as previously discussed, subjects may have encoded and subsequently r e c a l l e d events at the l e v e l of abstract meaning. The statement: "Don't pick up that glass with your hands, Susan", may simply have been encoded as: "He's r e a l l y being bossy now", while the statement: "Sure, I'd be glad to help your s i s t e r move" may have been encoded as: "He's being r e a l l y accommodating and agreeable". As Neisser (1976) asse r t s : "In reading or i n l i s t e n i n g , one perceives the meaning of words and sentences, the d r i f t of an argument, or the undertone of f e e l i n g that may be represented. These perceptions often seem very d i r e c t , i n the sense that we become aware of the meanings without seeming to notice the physical d e t a i l s that provide evidence f o r them." (P. 71) With both of the above p o s s i b i l i t i e s , tracking biases i n the study may have represented a s e l e c t i v e weighting of expectation consistent information rather than a complete i n a t t e n t i o n to inconsistent information. If t h i s was the case, i t might be argued that the tracking index r e f l e c t s a higher l e v e l cognitive process rather than s e l e c t i v e encoding. As repeatedly emphasized i n the the introductory section of t h i s study, however, the d i s t i n c t i o n between perceptual and cognitive processes may be more apparent than r e a l . As Smith & M i l l e r (1979) have argued, a t t r i b u t i o n a l processing can take place "at the time of encoding and storage of a stimulus as well as at the time of i t s r e t r i e v a l " . This i n t e r p r e t a t i o n i s not at a l l incompatible with contemporary theories of a t t e n t i o n which conceptualize s e l e c t i v e a t t e n t i o n i n terms of degree of a t t e n t i o n a l a l l o c a t i o n (Kahneman, 1974; Norman & Bobrow, 1975) and depth of processing (Craik & Tulving, 1975) rather than s e l e c t i v e f i l t e r i n g of information. - 88 -A f i n a l p o s s i b i l i t y i s that when a subject tracked a given i n t e r v a l , he registered an a f f e c t i v e impression associated with that i n t e r v a l . Zajonc (1980) has recently reviewed evidence supporting the p o s i t i o n that cognition and a f f e c t are under the co n t r o l of two p a r t i a l l y independent systems and that a f f e c t i v e reactions are not necessarily represented c o g n i t i v e l y . A f f e c t i v e judgments can be made independently of the types of cognitive processes t y p i c a l l y presumed to precede a f f e c t i v e reactions, yet at the same time can influence judgments i n a v a r i e t y of contexts. It seems l i k e l y that s o c i a l judgments are p a r t i c u l a r l y l i k e l y to have an important a f f e c t i v e component, and i t i s possible, f o r example, that judgments about the husband i n the present study were made along a simple dimension of good - bad, and then translated i n t o more complex adjective dimensions. C l i n i c a l Implications Regardless of which hypothesis or combination of the above hypotheses most accurately accounts f o r the data, the f a c t that subjects exhibited no expectation consistent bias i n eit h e r memory r e t r i e v a l or memory reconstruction may have some i n t e r e s t i n g c l i n i c a l i m plications. It suggests that the c l i e n t who has formed a dysfunctional impression of a s o c i a l i n t e r a c t i o n , may s t i l l have a v a i l a b l e i n memory the surface information upon which the i n i t i a l d i s t o r t i o n was based. In such cases i t may be us e f u l f o r the therapist to help the c l i e n t r e t r i e v e the appropriate information from memory and to then assess for any dysfunctional discounting of information or m i s i n t e r p r e t a t i o n of i t s meaning or s i g n i f i c a n c e . * A second p o t e n t i a l c l i n i c a l i m p l i c a t i o n of the above pattern of r e s u l t s i s that i t may be therapeutically u s e f u l for the c l i n i c i a n to help the c l i e n t to become aware of the manner i n which he s e l e c t i v e l y encodes expectation consistent information and to then begin to modify this encoding a c t i v i t y . This process can presumably be accomplished i n a number of ways. One p o s s i b i l i t y consists of prescribing exercises i n vivo to help the c l i e n t develop greater awareness of his encoding a c t i v i t i e s . The therapist can i n s t r u c t the c l i e n t to ask himself what his expectations are i n any given interpersonal s i t u a t i o n , and to then pay attention to the events which appear to confirm those expectations. Once the c l i e n t has become adept at t h i s process the therapist can i n s t r u c t him to change the procedure and to begin to a c t i v e l y seek events which disconfirm h i s expectations. In this manner the c l i e n t can compare the two encoding s t y l e s and the s o c i a l perceptions which r e s u l t from them and gain a sense of the manner i n which he a c t i v e l y influences h i s own experience. Another p o s s i b i l i t y consists of using the therapeutic r e l a t i o n s h i p as a mini laboratory i n which the therapist can stop the c l i e n t at s t r a t e g i c points to i n q u i r e as to which therapist behaviours are being s e l e c t i v e l y weighted. Group therapy s i t u a t i o n s would be i d e a l for t h i s type of exploration, since the therapist could p o t e n t i a l l y help the c l i e n t to become aware of the manner i n which he s e l e c t i v e l y encodes the interpersonal behaviour of a v a r i e t y of d i f f e r e n t target people. Group exercises i n which partners take turns pinpointing the interpersonal acts which they f i n d most h e l p f u l i n forming an impression of the other person could be used for t h i s purpose. - 90 -While the type of interventions described here tend to focus on what interpersonal information i s encoded by the c l i e n t , there i s no reason f o r them not to focus on how the c l i e n t encodes information, as well. It may be u s e f u l to stop c l i e n t s at s t r a t e g i c points during s o c i a l i n t e r a c t i o n s to explore the abstract meaning which they extract from a s p e c i f i c event. One p o s s i b i l i t y , f o r example, might involve videotaping therapy sessions with dysfunctional m a r i t a l couples, and subsequently reviewing the tape with the couple. Using this procedure, the therapist could stop the tape at various points to assess which events the two spouses deem s i g n i f i c a n t , as well as the meaning abstracted by both partners from those events. While the pattern of r e s u l t s emerging i n the present study leads to some i n t r i g u i n g speculation about p o t e n t i a l therapeutic interventions, i t must be remembered that the extent to which these findings would generalize to a c l i n i c a l population remains unanswered. It i s possible that expectations which are based upon more global b e l i e f systems or more enduring assumptions about the world would have a q u a l i t a t i v e l y d i f f e r e n t impact upon the processing of s o c i a l information, than the expectations which were manipulated experimentally i n the present i n v e s t i g a t i o n . It also seems l i k e l y that a f f e c t i v e and motivational considerations have a greater impact upon s o c i a l cognition i n a c l i n i c a l context than i n the present study and might p o t e n t i a l l y r e s u l t i n information processing biases which did not emerge i n the present study. F i n a l l y , i t i s probable that the precise nature of the information processing bias i n s o c i a l perception i s dependent upon the p a r t i c u l a r type of psychological disorder. Magaro (1980), for example, argues that - 91 -paranoids and paranoid schizophrenics employ "top - down" processing strategies which bend a l l incoming information to confirm e x i s t i n g schemata. It seems l i k e l y that the extent to which schemata w i l l accommodate i n the face of inconsistent information varies as a function of the s e v e r i t y of the disorder. There i s thus an important sense i n which the present i n v e s t i g a t i o n , while providing evidence about the cognitive mechanisms which mediate the e f f e c t of expectations upon s o c i a l perception i n general, must be considered only a preliminary step i n the d e l i n e a t i o n of cognitive processes involved i n c l i n i c a l l y dysfunctional s o c i a l perception. General Implications Although the present study was conducted s p e c i f i c a l l y with an eye towards c l i n i c a l implications, there are a v a r i e t y of other contexts i n which a refinement of our understanding of the cognitive mechanisms through which expectations influence the perception of interpersonal behaviour may be important. The impact of expectations upon s o c i a l stereotyping has long been a topic of i n t e r e s t to s o c i a l psychologists. Another relevant phenomenon i s the influence of teachers' expectations upon student behaviour i n classroom s i t u a t i o n s . S i m i l a r l y , courtroom settings and personnel decisions are contexts i n which expectancy-biased information processing may plan an important r o l e . F i n a l l y , returning to the context of c l i n i c a l psychology, neither the diagnostician, nor the therapist are exempt from the e f f e c t s of expectancy-biased information processing. The impression that a diagnostician forms of a patient may be biased by p r i o r information provided by diagnostic l a b e l s . Expectations based upon previous experiences - 92 -i n the therapist's personal l i f e may bias h i s perception of his c l i e n t s . This b r i e f l i s t provides some i n d i c a t i o n of the ubiquity of expectancy e f f e c t s i n important interpersonal s i t u a t i o n s and underscores the value of c l a r i f y i n g the nature of the relevant cognitive mechanisms. Since the f a i l u r e of the present study to confirm the mediating role of biased memory f o r surface information, stands i n contrast with other research as previously described, r e p l i c a t i o n of t h i s r e s u l t i s e s s e n t i a l . As Hogarth (1981) has recently noted, however, judgment research which focuses upon d i s c r e t e incidents may y i e l d a f a u l t y picture of the accuracy of judgmental processes which normally operate upon continuous information i n the r e a l world. Analogously, s o c i a l perception research using d i s c r e t e stimulus materials such as l i s t s of behaviours may y i e l d an inaccurate r e f l e c t i o n of the cognitive processes involved i n everyday s o c i a l perseption. For this reason, future studies i n v e s t i g a t i n g the cognitive processes mediating expectancy e f f e c t s would be well advised to employ e c o l o g i c a l l y v a l i d stimulus materials which more c l o s e l y r e f l e c t the complexity of r e a l l i f e s o c i a l i n t e r a c t i o n s than has t y p i c a l l y been the case. The present study was an i n i t i a l step i n this d i r e c t i o n . Tracking Subjectively Salient Events:  A Methodological Note The r e s u l t s of the present study suggest that the tracking methodology employed can be a us e f u l procedure f o r purposes of exploring the perception - 93 -of interpersonal behaviour. The network of re l a t i o n s h i p s which emerged between the tracking data and other variables was consistent and t h e o r e t i c a l l y g meaningful, thus providing a form of construct v a l i d a t i o n of the procedure. This i s important since the u n i t i z i n g procedure upon which the methodology was based has t y p i c a l l y been employed i n the context of very simple, a r t i f i c i a l stimulus materials. The fact that the predicted pattern of r e s u l t s emerged i n the tracking data despite the fact that t h i s was not p a r a l l e l e d by the memory data, demonstrates the importance of employing independent indices of encoding and memory i n s o c i a l perception studies, and points to the tracking procedure as a promising candidate for t h i s purpose. The fact that subjects' adjective ratings could be predicted on the basis of t h e i r tracking behaviour suggests that the tracking procedures may p o t e n t i a l l y be useful i n c o r r e l a t i o n a l studies, designed to investigate differences i n encod-ing behaviour between c l i n i c a l and normal groups. Another p o t e n t i a l a p p l i c a t i o n of the tracking methodology may be as a c l i n i c a l diagnostic t o o l which can be employed to assess and pinpoint c h a r a c t e r i s t i c a l l y dysfunctional encoding a c t i v i t i e s i n c l i e n t s . A procedure could for example, be devised i n which a standard set of videotaped vignettes dealing with common interpersonal themes such as domination, submission, expression of approval, a f f e c t i o n , r e j e c t i o n , etc. 8 See Appendix 10 for further discussion of the r e l a t i o n s h i p between tracking and other v a r i a b l e s . - 94 -are developed. The c l i e n t would then be asked to i d e n t i f y with one of the i n d i v i d u a l s i n the vignettes and to track events which are h e l p f u l to him i n forming an impression of the other i n d i v i d u a l . The c l i n i c i a n would then be able to use the tracking data to i s o l a t e the type of interpersonal behaviours to which the c l i e n t i s p a r t i c u l a r l y prone to attend. In t h i s manner; the procedure could be employed .as a form of cognitive TAT (Meichenbaum, 1977) which y i e l d s information about cognitive s t y l e rather than more t r a d i t i o n a l personality or motivational v a r i a b l e s . A f i n a l issue warranting consideration, is- the p o s s i b i l i t y that the tracking data r e s u l t s i n the present study, represent the operation of demand character-i s t i c s , rather than subjects' true encoding patterns. An a l t e r n a t i v e hypothesis which might account for the r e s u l t s obtained, i s that subjects were attempting to respond to perceived task demands by tracking i n t e r v a l s which were consistent with the information presented i n the expectancy manipulation. Although t h i s p o s s i b i l i t y cannot be ruled out completely, i t would seem u n l i k e l y , since at no point i n the experiment was i t made e x p l i c i t that consistent information would be more appropriate than inconsistent information. I t would nevertheless be advisable for future studies to con t r o l f o r t h i s p o s s i b i l i t y more c a r e f u l l y . - 95 -REFERENCE NOTES 1. Newtson, D. Task and observer s k i l l f actors i n accuracy of assessment of performance. Unpublished Manuscript, U n i v e r s i t y of V i r g i n i a , 1976. 2. T r i e r w e i l e r , S.T., The unit of perception i n the processing of s o c i a l information: An evaluation of current theory and methodology. Unpublished Master's Thesis, University of I l l i n o i s at Urbana -Champaign, 1980. - 96 -REFERENCES Alston, W.P. T r a i t consistency and conceptual a l t e r n a t i v e s f o r personality theory. Journal For the Theory of S o c i a l Behaviour, 1975, 5, 17-48. Anderson, C.A., Lepper, M.R. & Ross, L. Perseverence of s o c i a l theories: The r o l e of explanation i n the persistence of d i s c r e d i t e d information. Journal of Personality and So c i a l Psychology, 1980, 39, 1037-1049. Anderson, J.R. Arguments concerning representations f o r mental imagery. Psychological Review, 1978, 85, 249-277. Anderson, J.R. Language, memory and thought. H i l l s d a l e , New Jersey: Erlbaum, 1976. Arnkoff, D.B. Psychotherapy from the perspective of cognitive theory. In M.J. Mahoney (Ed.), Psychotherapy process: Current issues and future  d i r e c t i o n s . New York: Plenum Press, 1980. Atkinson, R.C. & S h i f f r i n , R.M. Human memory: A proposed system and i t s con t r o l process. In K.W. Spence & J.T. Spence (Eds.), The psychology of learning and motivation: Advances i n research and theory (Vol. 2). New York: Academic Press, 1968. Bandura, A. S e l f - e f f i c a c y : Towards a unifying theory of behavioral change. Psychological Review, 1977, 84, 191-215. Bandura, A., Adams, N.E., Hardy, A.B. & Howells, G.N. Tests of the gene r a l i t y of s e l f - e f f i c a c y theory. Cognitive Therapy and Research, 1980, 4, 39-66. Barker, R.G. The stream of behavior. New York: Appleton-Century- Crofts, 1963. B a r t l e t t , F.C. Remembering. Cambridge, England: Cambridge Uni v e r s i t y Press, 1932. Beck, A.T. Cognitive therapy and the emotional disorders. New York: International U n i v e r s i t i e s Press, 1976. Bobrow, S.A. & Bower, G.H. Comprehension and r e c a l l of sentences. Journal  of Experimental Psychology, 1969, 80, 55-61. Bootzin, R.R. & Lick, J.R. Expectancies i n therapy research: Interpretive a r t i f a c t or mediating mechanism. Journal of Consulting and C l i n i c a l  Psychology, 1979, 47, 852-855. Bower, G.H. Contacts of cognitive psychology with s o c i a l learning theory. Cognitive Therapy and Research, 1978, 2, 123-146. Breur, J . & Freud, S. Studies on h y s t e r i a . Standard E d i t i o n 2. London: Hogarth, 1895. - 97 -Broadbent, D.E. Perception and communication. London: Pergamon, 1958. Bruner, J.S. & Potter, M. Inference i n v i s u a l recognition. Science, 1964, 144, 424-425. Buss, D.M. & Craik, K.H. The act frequency analysis of personal d i s p o s i t i o n s : Aloofness, gregariousness, dominance and submissiveness. Journal of Personality-, 1981. Buss, D.M. & Craik, K.H. The frequency concept of d i s p o s i t i o n ! Dominance and pr o t o t y p i c a l l y dominant acts. Journal of Personality, 1980, 43, 379-392. Cantor, N. & Mischel, W. T r a i t s as prototypes: E f f e c t s on recognition memory-. Journal of Personality and S o c i a l Psychology, 1977, 35, 38^48. Carson, R.C. S e l f - f u l f i l l i n g prophecy, maladaptive behaviour and psychotherapy. In J.C. Anchin & D.J. K i e s l e r (Eds.), Handbook of interpersonal psychotherapy, New York: Pergamon Press, 1982. Carver, C.S. & Scheirer, M.F. Attention and s e l f - r e g u l a t i o n : A control-theory approach to human behavior. ' New York: Springer-Verlag,' 1981. ~ ~ ~~ Chrzanowski, G. Interpersonal formulations of psychotherapy: A contemporary model. In J.C. Anchin & D.J. K i e s l e r (Eds.), Handbook of interpersonal psychotherapy, New York: Pergamon Press, 1982. Cohen, C E . Person categories and s o c i a l perception: Testing some boundaries of the processing e f f e c t s of p r i o r knowledge. Journal of Personality and  Soc i a l Psychology, 1981, 40, 441-452. Cohen, C E . & Ebbeson, E.B. Observational goals and schema a c t i v a t i o n : A t h e o r e t i c a l framework for behavior perception. Journal of Experimental S o c i a l Psychology, 1979, 15, 305-329. ' " Coyne, J.C. & Segal, L. A b r i e f , s t r a t e g i c , i n t e r a c t i o n a l approach to Psycho-therapy. In J . C Anchin & D.J. K i e s l e r (Eds.)', Handbook of interpersonal psychotherapy-, New-York: Pergamon Press-, 1982. —' Craik, F.l.M. & Lockhart, R.S. Levels of processing; A framework for memory research. Journal of Verbal Learning and Verbal Behavior, 1972, 11, 671-684. Craik, F.l.M. & Tulving, E. Depth of processing and the retention of words- i n episodic-memory-. Journal ..of Experimental Psychology: General, 1975, 104, 268^294. Craik, F.l.M. and Watkins, 'M.J. The ro l e of rehersal i n short term -memory. Journal, of V e r b a l Learning and 'Verbal Behavior, 1973, 12, 599-607. Darley-, J.M. and Fazio, R.H. Expectancy confirmation processes a r i s i n g i n the s o c i a l i n t e r a c t i o n sequence. American Psychologist, 1980, 35, 867-881. Deutsch, J.A. & Deutsch, D. Attention: Some t h e o r e t i c a l considerations. Psychological Review, 1963, 70, 80-90. - 98 -Einhorn, H.J. & Hogarth, R.M. Confidence i n judgment: Persistence of the i l l u s i o n of v a l i d i t y . Psychological Review, 1978, 85, 395-416 E l l i s , A. Reason and emotion i n psychotherapy. Seacaucus, New Jersey: Lyle Stuart, 1977. Eshkol, N. Moving writing reading. T e l Aviv: Movement notation society, 1973. Estes, W.K. The information-processing approach to cognition: A confluence of metaphors and methods. In W.K. Estes (Ed.), Handbook of learning and  cognitive processes (Vol. 5). H i l l s d a l e , New Jersey: Erlbaum, 1978. Estes, W.K. The focus of i n f e r e n t i a l and perceptual processes i n l e t t e r i d e n t i f i c a t i o n . Journal of Experimental Psychology: General, 1975, 1, 122-145. Fodor, J.A. & Bever, T.G. The psychological r e a l i t y of l i n g u i s t i c segments. Journal of Verbal Learning and Verbal Behavior, 1965, 4, 414-420. Frank, J.D. Persuasion and healing. Baltimore: John Hopkins Un i v e r s i t y Press, 1961. Freud, S. Beyond the pleasure p r i n c i p a l . Standard E d i t i o n 18. London: Hogarth, 1920. Freud, S. Fragment of an analysis of a case of h y s t e r i a . Standard E d i t i o n 7. London: Hogarth, 1905. Freud, S. The technique of psychoanalysis. From: An o u t l i n e of  psychoanalysis. Standard E d i t i o n 23. London: Hogarth, 1940. Go l d f r i e d , M.R. Anxiety reduction through cognitive-behavioral intervention. In P.C. Kendall & S.D. Hollon (Eds.), Cognitive-Behavioral  Interventions. New York: Academic Press, 1979. Go l d f r i e d , M.R. & D ' Z u r i l l a , T.J. A behavioral-analytic model f o r assessing competence. In Spielberger, CD. (Ed.), Current topics i n c l i n i c a l and  community psychology, Vol. 1. New York: Academic Press, 1969. Gottman, J. M a r i t a l i n t e r a c t i o n : Experimental i n v e s t i g a t i o n s . New York: Academic Press, 1979. Gottman, J . & Markman, H. Experimental designs i n psychotherapy research. In S. G a r f i e l d and A. Bergin (Eds.), Handbook of psychotherapy and  behavior change, second e d i t i o n . New York: John Wiley & Sons, 1978. Greenberg, L.S. & Safran, J.D. Encoding and cognitive therapy: Changing what c l i e n t s attend to. Psychotherapy: Theory, Research and Practice, 1982, In Press. Greene, D. S o c i a l perception as problem solving. In J.S. C a r r o l l & J.W. Payne (Eds.), Cognition and s o c i a l behavior. H i l l s d a l e , New Jersey: Lawrence Erlbaum, 1976. - 99 -Hastie, R. & Kumar, P.A. Person memory: Personality t r a i t s as organizing p r i n c i p l e s i n memory f o r behaviors. Journal of Personality and S o c i a l  Psychology, 1979, 37, 25-38. Hogarth, R.M. Beyond discr e t e biases: Functional and dysfunctional aspects of judgmental h e u r i s t i c s . Psychological B u l l e t i n , 1981, 90, 197-217. James, W. The p r i n c i p l e s of psychology, Volume 1. New York: Dover, 1950. Jones, R.A. S e l f - f u l f i l l i n g prophecies: S o c i a l , psychological and p h y s i o l o g i c a l e f f e c t s , of expectancies. H i l l s d a l e , New Jersey: Lawrence Erlbaum, 1977. Kahneman, D. Attention and e f f o r t . Englewood C l i f f s , New Jersey: P r e n t i c e - H a l l , 1973. K e l l y , G.A. The psychology of personal constrcuts. New York: Norton, 1955. K i e s l e r , D.J. Confronting the c l i e n t - t h e r a p i s t r e l a t i o n s h i p i n psychotherapy. In J.C. Anchin & D.J. K i e s l e r (Eds.) Handbook of  interperonal psychotherapy. New York: Pergamon Press, 1982. K i e s l e r , D.J. Some myths of psychotherapy research and the search for a paradigm. Psychological B u l l e t i n , 1966, 65, 110-136. Kihlstrom, J.F. & Nasby, W. Cognitive tasks i n c l i n i c a l assessment: An exercise i n applied psychology. In P.C. Kendall & S.D. Hollon (Eds.), Assessment strategies f o r cognitive-behavioral interventions. New York: Academic Press, 1981. Kintsch, W. Memory representations of text. In R.L. Solso (Ed.), Information processing and cognition. H i l l s d a l e , New Jersey: Erlbaum, 1975. Landau, R.J. & G o l d f r i e d , M.R. The assessment of schemata: A unifying framework f o r cognitive, behavioral and t r a d i t i o n a l assessment. In P.C. Kendall & S.D. Hollon (Eds.), Assessment strategies f o r  cognitive-behavioral interventions. New York: Academic Press, 1981. Landfield, A.W. Personal Construct Psychology: A theory to be elaborated. In M.J. Mahoney (Ed.), Psychotherapy process: Current issues and future  d i r e c t i o n s . New York: Plenum Press, 1980. Lang, R. The therapeutic i n t e r a c t i o n . Volume 2. New York: Jason Aronson, Inc., 1976. Leary, T. Interpersonal diagnosis of personality. New York: Ronald Press, 1957. Loftus, E.F. & Palmer J.C. Reconstruction of automobile destruction: An example of the i n t e r a c t i o n between language and memory. Journal of Verbal  Learning and Verbal Behavior, 1974, 11, 585-589. - 100 -Magaro, P.A. Cognition i n schizophrenia and paranoia. H i l l s d a l e , New Jersey: Erlbaum, 1980. Mahoney, M.J. Cognition and behavior modification. Cambridge, Mass.: B a l l i n g e r , 1974. Mandler, J.M. A code i n the node: The use of a story schema i n r e t r i e v a l . Discourse processes, 1978, 1, 14-35. Massad, CM., Hubbard, M. & Newtson, D. Selective perception of events. Journal of experimental s o c i a l psychology, 1979, 15, 515-532. Meichenbaum, D. Cognitive behavior modification: An i n t e g r a t i v e approach. New York: Plenum, 1977. Meyer, D.E., Schvaneveldt, R.W., & Ruddy, M.G. Functions of phonemic and graphic codes i n v i s u a l word recognition. Memory and cognition, 1974, 2, 309-321. M i l l e r , G.A., Bruner, J.S., & Postman, L. F a m i l i a r i t y of l e t t e r sequences and tachistoscopic i d e n t i f i c a t i o n . Journal Of Genetic Psychology, 1954, 50, 129-139. Minsky, M.A. A framework for representing knowledge. In P.H. Winston (Ed.), The psychology of computer v i s i o n . New York: McGraw-Hill, 1975. Mischel, W. George K e l l y ' s a n t i c i p a t i o n of psychology: A personal t r i b u t e . In M.J. Mahoney (Ed.), Psychotherapy process: Current issues  and future d i r e c t i o n s . New York: Plenum Press, 1980. Myers, J.L. Fundamentals of experimental design (Third E d i t i o n ) . Boston: A l l y n & Bacon: 1979. Neisser, U. Cognitive psychology. New York: Appleton-Century- Crofts, 1967. Neisser, U. Cognition and r e a l i t y : P r i n c i p l e s and implications of  cognitive psychology. San Francisco: Freeman, 1976. Neisser, U. Three cognitive psychologies and t h e i r i m p l i c a t i o n s . In M.J. Mahoney (Ed.), Psychotherapy process: Current issues and future  d i r e c t i o n s . New York: Plenum Press, 1980. Newtson, D. A t t r i b u t i o n and the unit of perception of ongoing behavior. Journal of Personality and S o c i a l Psychology, 1973, 28, 28-38. Newtson, D. Foundations of a t t r i b u t i o n s : The perception of ongoing behavior. In J . Harver Ickes & R. Kidd (Eds.), New d i r e c t i o n s i n  a t t r i b u t i o n research. H i l l s d a l e , New Jersey: Erlbaum, 1976. Newtson, D. & Enquist, G. The perceptual organization of ongoing behavior. Journal of Experimental S o c i a l Psychology, 1976, 12, 436-450. Newtson, D., Enquist, G., & Bois J. The objective basis of behavior u n i t s . Journal of Personality and S o c i a l Psychology, 1977, 35, 847-862. - 101 -Newtson, D., Rinder, R., M i l l e r , R., & LaCross, K. E f f e c t s of a v a i l a b i l i t y of feature changes on behavior segmentation. Journal of Experimental  S o c i a l Psychology, 1978, 14, 379-388. Nisbett, R. & Ross, L. Human inference: Strategies and shortcomings of  s o c i a l judgment. New Jersey: P r e n t i c e - H a l l , 1980. Norman, D.A. and Bobrow, D.G. On data l i m i t e d and resource l i m i t e d processes. Cognitive Psychology, 1975, 7, 44-64. Palmer, S.E. The e f f e c t s of contextual scenes on the i d e n t i f i c a t i o n of objects. Memory and Cognition, 1975, 3, 519-526. Perl s , F.S. The ge s t a l t approach and eye witness to therapy. New York: Bantam, 1973. Piaget, J. Memory and i n t e l l i g e n c e . New York: Basic Books, 1973. Piaget, J. The ori g i n s of i n t e l l i g e n c e i n chil d r e n . New York: International U n i v e r s i t i e s Press, 1952. Reyes, R.M., Thompson, W.C., & Bower, G.H. Judgemental biases r e s u l t i n g from d i f f e r i n g a v a i l a b i l i t i e s of arguments. Journal of Personality and S o c i a l  Psychology, 1980, 39, 2-12. Rosch, E., Mervis, C.B., Gray, W.D., Johnson, D.M., & Boyes-Braem, P. Basic objects i n natural categories. Cognitive Psychology, 1976, 8, 382-439. Ross, L., Lepper, M.R., & Hubbard, M. Perseverance i n s e l f perception and s o c i a l perception: Biased a t t r i b u t i o n a l processes i n the deb r i e f i n g paradigm. Journal of Personality and So c i a l Psychology, 1975, 32, 880-892. Rothbart, M., Evans, M., & Fulero, S. Recall f o r confirming events: Memory processes and the maintenance of s o c i a l stereotypes. Journal of  Experimental S o c i a l Psychology, 1979, 15, 343-355. Rumelhart, D.E. Introduction to human information processing. New York: Wiley, 197 7. Rumelhart, D.E., & Si p l e , P. The process of t a c h i s t o s o p h i c a l l y recognizing words. Psychological Review, 1974, 81, 91-118. Schank, R., & Abelson, R.P. S c r i p t s , plans, goals, and understanding: An  inquiry i n t o human knowledge structures. H i l l s d a l e , New Jersey: Lawrence Erlbaum, 1977. Schvanveldt, R.W. & Meyer, D.E. R e t r i e v a l and comparison processes i n semantic memory. In S. Kornblum (Ed.), Attention and performance IV. New York: Academic Press, 1973. Se l f r i d g e , O.G., & Neisser, U. Pattern recognition by machine. S c i e n t i f i c  American, 1960, 203, 60-68. - 102 -Smith, E.R. & M i l l e r , F.D. Salience and the cognitive mediation of a t t r i b u t i o n . Journal of Personality and S o c i a l Psychology, 1979, 37, 2240-2252. S u l l i v a n , H.S. Conceptions of modern psychiatry. New York: Norton, 1940. S u l l i v a n , H.S. The p s y c h i a t r i c interview. New York: Norton, 1954. Taylor, S.E. & Fiske, S.T. Salience, Attention and a t t r i b u t i o n : Top of the head phenomena. In L. Berkowitz (Ed.), Advances i n experimental s o c i a l  psychology (Vol. 11). New York: Academic Press, 1978. Treisman, A.M. Strategies and models of s e l e c t i v e a t t e n t i o n . Psychological  Review, 1969, 76, 282-299. Treisman, A. & Tuxworth, J. Immediate and delayed r e c a l l of sentences a f t e r perceptual processing at d i f f e r e n t l e v e l s . Journal of Verbal Learning and  Verbal Behaviour, 1974, 13, 38-44. Tulving, E., Mandler, G., & Baumal, R. Interaction of two sources of information i n tachistoscopic word recognition. Canadian Journal of  Psychology, 1964, 18, 67-71. Tversky, A., & Kahneman, D. Judgment under uncertainty: H e u r i s t i c s and biases. Science, 1974, 185, 1124-1131. Wachtel, P.L. Interpersonal therapy and active intervention. In J.C. Anchin & D.J. K i e s l e r (Eds.), Handbook of interpersonal psychotherapy. New York: Pergamon Press, 1982. Wason, P.C. Reasoning about a r u l e . Quarterly Journal of Experimental  Psychology, 1968, 20, 273-281. Wason, P.C. Regression i n reasoning? B r i t i s h Journal of Psychology, 1969, 60, 471-480. White, G.M. Conceptual universals i n interpersonal language. American  Anthropologist, 1980, 82, 759-781. Wiggins, J.S. A psychological taxamony of t r a i t d e s c r i p t i v e terms: The interpersonal domain. Journal of Personality and S o c i a l Psychology, 1979, 37, 395-412. Wiggins, J.S. Circumplex models of interpersonal behaviour i n c l i n i c a l psychology. In P.C. Kendall & J.N. Butcher (Eds.), Handbook of research  methods i n c l i n i c a l psychology. New York: Wiley Interscience, 1982. Wilder, D. P r e d i c t a b i l i t y of behaviour, goals, and unit of perception. Personality and S o c i a l Psychology B u l l e t i n . 1978, 4, 604-607. Wilkins, W. Expectancies i n therapy research: Discriminating between heterogeneous nonspecifics. Journal of Consulting and C l i n i c a l  Psychology, 1979, 47, 837-845. - 103 -Wilson, G.T. Toward specifying the nonspecific factors i n behaviour therapy: A s o c i a l - l e a r n i n g analysis. In M.J. Mahoney (Ed.), Psychotherapy process: Current issues and future d i r e c t i o n s . New York: Plenum Press, 1980. Zadny, J. and Gerard, H.B. A t t r i b u t e d intentions and informational s e l e c t i v i t y . Journal of Experimental S o c i a l Psychology, 1974, 10, 34-52. Zajonc, R.B. Feeling and thinking: Preferences need no inferences. American Psychologist, 1980, 35, 151-175. - 104 -A P P E N D I C E S - 105 -APPENDIX 1 Expectancy Manipulations A. Dominant 1. He persuaded the others to accept h i s opinion on the issue. 2. He t o l d h i s co-^worker how to do h i s job. 3. On the auto t r i p , he decided which d i r e c t i o n s to take when they got l o s t . 4. He bought himself a new summer jacket. (N)* 5. He monopolized the conversation. 6. He ordered a cheeseburger f o r lunch. (N) 7. He stubbed h i s toe on the kitchen table. (N) 8. He took a stand on the issue without waiting to f i n d out what others thought. 9. He played a game of tennis with a f r i e n d . (N) 10. He issued orders to get the group organized. 11. He took charge of things at the committee meeting. B. Agreeable 1. He remained patient when the car ran out of gas. 2. He asked others what they would l i k e to do at the Sunday p i c n i c . 3. He r e a d i l y did the dishes a f t e r dinner. 4. He bought himself a new summer jacket. (N) 5. He t r i e d a new food at the restaurant because h i s f r i e n d suggested i t . 6. He ordered a cheeseburger f o r lunch. (N) 7. He stubbed h i s toe on the kitchen table. (N) 8. He danced when asked even though he d i s l i k e d the muskc. 9. He played a game of tennis with a f r i e n d . (N) 10. He offered an older person h i s seat on the bus. 11. He stopped at the post o f f i c e for a f r i e n d even though he was i n a rush. *Neutral items designated by "N". - 106 -APPENDIX 2 Vignette Transcript* H = Husband W = Wife H 1: (walks into house and says to wife) Hi Susan ... W - (gives husband a hug) Hi David ... How was your day? H 2: (returning hug) Oh ... i t was ikay ... I suppose ... the a i r conditioner broke i n the o f f i c e ... and i t made i t r e a l l y d i f f i c u l t to work t h i s afternoon. W - I can imagine ... H 3: Anyways ... they're supposed to be f i x i n g i t for tomorrow ( s i t s down and begins to look at the newspaper). ** \] - David ... that makes me r e a l l y angry when you s t a r t reading the newspaper the second you come home. H - (putting newspaper away) I'm sorry honey ... that wasn't very nice of me. I guess I'm f e e l i n g kind of stressed. W - What are you f e e l i n g stressed about? H - Well ... i t was a d i f f i c u l t day at work today. (telephone rings ... wife answers i t ) W — Oh ... Hi Terry ... how are you ... (pause) ... not bad ... p_ H 4: ( i n t e r r u p t s ) Susan ... could you c a l l Terry back ... I'm expecting an important phone c a l l . W - (to f r i e n d on phone) Could you hold no a minute please Terry? (to husband) We'll just be a few minutes, David. * D = dominant items with > 80% intersubject agreement i n categorization A = agreeable items with > 80% i n t e r s u b j e c t agreement i n categorization N = neutral items with >80% i n t e r s u b j e c t agreement i n categorization ** This segment of the f i l m was not included i n the main analyses. For further discussion see Appendix 9. - 107 -APPENDIX 2 Vignette Transcript (Cont'd.) p_ H 5: No ... y o u ' l l have to c a l l her back Susan ... I don't want to miss t h i s c a l l . W - (to f r i e n d on phone) Terry ... I ' l l have to c a l l you back ... David's expecting an important phone c a l l ... okay ... I ' l l t a l k to you l a t e r ... goodbye. H 6: Thanks. W - Who are you expecting a c a l l from? H 7: Oh ... I t o l d an out of town c l i e n t that he could reach me at home a f t e r work. W - Is he a new c l i e n t ? H 8: Yeah ... He just opened up an account with the firm l a s t week ... (pause) ... L i s t e n ... While I'm waiting f o r the c a l l , we can decide where we want to go f o r dinner. W - Where would you l i k e to go? H 9: Well ... I'm kind of p a r t i a l to Chinese food, myself. W - I'd rather go out f o r Greek food ... we haven't done that i n a long time ... A H 10: Well ... that sounds good to me too ... we can go out f o r Chinese food another time. Where would you l i k e to go? W - How about that Greek restaurant that we went to with your brother that time? H 11: Sure ... that sounds fime. Do you want to go f o r a l a t e or an e a r l y dinner? W - How about 7:30? H 12: (glancing at watch) Okay ... so we can relax f or a while f i r s t . W - Um hm ... How was the t r a f f i c on the way home? N H 13: It was a c t u a l l y quite bad ... W - I thought i t might be ... I heard on the news that there had been a bad t r a f f i c accident on the bridge. H 14: Yeah ... i t was a 5-car pile-up i n the middle of rush hour t r a f f i c . Commuting to work i n t h i s c i t y i s r e a l l y getting to be a drag ... By the way ... Have you decided whether or not you're going to take that part-time job? - 108 -APPENDIX 2 Vignette Transcript (Cont'd.) W - No ... not yet. H 15: You should take i t ... i t would be good f o r you ... W - I know ... but I can't make up my mind. I) H 16: The longer you think about i t ... the harder i t w i l l get ... look ... just make up your mind ... give them a c a l l ... and y o u ' l l f e e l 100 times better afterwards. W - You know something ... you're righ t ... okay ... I ' l l c a l l them tomorrow. H 17: Good ... (pause) ... were there any c a l l s f o r me today? W - Your brother c a l l e d . N_ H 18: Oh . .. when did he get back in t o town? W - Just l a s t night. H 19: Did he have a good time? W - He says he had a great time ... He also said he'd i n v i t e us over to h i s place when the s l i d e s are developed. H 20: You know ... i t seems l i k e he just l e f t on h i s t r i p a few days ago ... W - I know ... t h i s l a s t month has passed r e a l l y quickly ... H 21: Well ... I'm looking forward to seeing him again ... W - So am I ... (pause) David ... what time i s i t ? H 22: (glancing at his watch) 7:00. W - I'd l i k e to watch Merv G r i f f i n on T.V. H 23: I don't r e a l l y f e e l l i k e watching T.V. Let's have a drink before dinner instead. W - But I'd l i k e to watch T.V. D_ H 24: I'm just not i n the mood for Merv G r i f f i n tonight ... come on ... l e t ' s have a drink. W - Well ... okay. H 25: Good ... I ' l l make them (husband gets up and makes drinks while they continue talking) ... I'm r e a l l y f e e l i n g kind of stresses tonight . .. maybe a drink w i l l relax me a b i t ... How are you feeling? - 109 -APPENDIX 2 Vignette Transcript (Cont'd.) W - Well I'm a c t u a l l y f e e l i n g kind of t i r e d . I didn't sleep very w e l l l a s t night. H 26: Anything wrong? W - No ... you know I just get insomnia from time to time ... you're lucky ... you never seem to have that problem. H 27: That's true ... I haven't f o r quite a while. W - I don't know what causes i t ... maybe i t ' s because I haven't been getting any exercise ... H 28: That could be i t . W - David ... don't make the drinks too strong t h i s time ... you always make them too strong (spoken i n a chiding i r r i t a t e d fashion) . A H 29: (warmly) Okay honey ... I ' l l t r y to make i t the way you l i k e i t th i s time. I always forget and assume you l i k e i t the same way as I do. Where's the tonic? W - I t ' s i n the back l e f t hand corner of the f r i d g e . H 30: Oh ... I see i t ... we're almost out, you know. W - No we're not ... there's another b o t t l e i n the cupboard. H 31: Oh ... W - Can you f i n d i t ? N H 32: Yeah ... I've got i t . W - Is there a b o t t l e of 7-up there as well? H 3 3: Um hm . .. W - Good ... Oh, by the way, David ... I need the car tomorrow during the day ... okay? A H 34: Sure ... I ' l l take the bus to work (brings the drinks and s i t s down beside her) Here's your drink. W - Thanks (tastes i t ) ummm ... that's good ... H 35: I put a drop of lemon j u i c e i n i t . W - So that's the magic ingredient. H 36: Um hm. W - Where did you le a r n that? - 110 -APPENDIX 2 Vignette Transcript (Cont'd.) H 37: I know a guy who drinks i t that way. W - Well ... i t ' s a nice taste. H 38: (takes a s i p of drink) ... You know ... we should have P h i l l i p and Kathy over f o r dinner next week. W - Oh David ... I don't want to ... I f i n d them so boring. H 39: I know ... but they're my friends and we owe them a dinner. W - Oh ... p_ H 40: L i s t e n Susan ... we owe them a dinner and I r e a l l y want to have them over ... i t won't be that bad. W - Well okay ... when do you want to have them over? H 41: Friday. W - Okay H 42: Good ... I ' l l give P h i l l i p a c a l l . W - Well, I suppose i t won't be so bad having them over ... we haven't seen them i n a while. H 43: We don't have to make a long evening of i t e i t h e r . W - That's true ... I think I ' l l cook that chicken dis h that I make ... you know ... the one with the wine and the bacon? A H 44: That's a good idea ... i t ' s a r e a l l y nice dish. W - But you know ... come to think of i t ... I don't think that P h i l l i p l i k e s chicken. H 45: Oh, don't worry ... I'm sure h e ' l l eat i t . W - Well ... I don't want to take the chance. A H 46: Okay honey ... I ' l l be happy to go along with whatever. W - Maybe I ' l l make f i s h . Are there any good places to buy salmon t h i s time of year? H 47: They usually have pretty good f i s h at that l i t t l e market by the beach, don't they? W - That's true. Well ... I ' l l see i f they have any good f r e s h salmon and i f they do ... we can barbeque i t . H 48: Okay ... (pause) ... (takes a sip and says) So how did you spend your day today? - I l l -APPENDIX 2 Vignette Transcript (Cont'd.) W - I f i n i s h e d reading that novel. H 49: You mean, "Riders i n the Chariot"? W - Um hm. H 50: What did you think of i t ? W - . Well ... I thought i t was well written ... I enjoyed i t ... what di d you think? H 51: (hes i t a n t l y ) Well ... to be honest ... I didn't l i k e i t . W - ( t e s t i l y ) What was wrong with i t ? H 52: Well ... I found i t kind of pretentious. W - How was i t pretentious? (becoming more i r a t e ) H 53: Well ... I just thought the author was trying too hard to be sch o l a r l y . W - (angry ... getting up to leave) Well that's your opinion! I'm going to take a shower! p_ H 54: Don't go storming o f f , Susan ... s i t down. W - ( s i t s down) H 55: That's better. Now what are you so upset about? W - Oh ... I don't know ... I guess I'm j u s t i n an i r r i t a b l e mood. H 56: Just because we have d i f f e r e n t opinions about something, i t doesn't mean that one of us i s wrong and the other one r i g h t , you know ... and i t doesn't meand that I l i k e you any l e s s . W - (calmer now) Yeah ... I know ... you're r i g h t ... I'm sorry i f I overreacted. H 57: Do you want me to make you another drink? W - No ... that's okay ... thanks ... I'm s t i l l working on t h i s one. H 58: You okay now? W - (si n c e r e l y ) Yeah ... I f e e l better ... I guess I could use a vacation. H 59: Yeah ... I think I could use one myself ... It seems that by th i s time of the year I r e a l l y s t a r t to f e e l t i r e d . (pause) Remember how good we both f e l t a f t e r our t r i p to C a l i f o r n i a l a s t year? - 112 -APPENDIX 2 Vignette Transcript (Cont'd.) W - Yeah ... that was nice ... H 60: I was thinking about i t just the other day. It was fun camping ... wasn't i t . , W - Um hm ... remember that b e a u t i f u l lake that we swam in? H 61: Yeah ... the water was c r y s t a l c l e a r ... and a perfect temperature. W - A l l that swimming was great f o r the appetite. H 62: Yeah ... there's also something about cooking outdoors which makes food taste better (pause) ... You know ... we have to decide where we're going on holidays t h i s year. We've got to f i n a l i z e our t r a v e l plans within the next two weeks. W - (agreeably) Okay. H 63: Let's spend the f i r s t four days i n Toronto, and then spend two days browsing around Montreal. W - Well ... I'd rather go to New York than Montreal. p_ H 64: No ... I just can't a f f o r d the extra time i t would take us to go to New York ... we'll have to do that some other time. W - Well . . . okay ... do you want to make the reservations or should I? H 65: I ' l l do i t ... I'm looking forward to t h i s vacation ... I hope the weather's nice. W - So do I ... I don't think Montreal w i l l be as much fun i n the r a i n . I think they're predicting good weather though. H 66: Where did you hear that? W - I heard a long range forecast on CBC radio t h i s morning. N_ H 67: That's encouraging ... I hope i t ' s accurate. W - Well ... they've been pretty accurate so f a r t h i s year. H 68: That's true ... (pause) W - By the way David ... have you got any plans f o r Saturday? H 69: No ... not r e a l l y ... W - My s i s t e r c a l l e d up to ask i f you would be able to help her move i n t o her new apartment. - 113 -APPENDIX 2 Vignette Transcript (Cont'd.) A H 70: Sure ... I'd be glad to help your s i s t e r move. W - That's r e a l l y nice of you. H 71: No problem ... I'd forgotten she was moving t h i s week. W - Well . .. she was o r i g i n a l l y supposed to move a couple of weeks l a t e r ... but the previous tenant vacated e a r l y . H 72: I see ... W - You should see i t ... i t ' s r e a l l y a large place (wife gestures with her hands i n the process, a c c i d e n t a l l y knocks over her glass and breaks i t ) Oops (she begins to pick up the broken glass with her hands) p_ H 73: Don't pick up that glass with your hands, Susan - Use the broom and dustpan. (wife does so) H 74: That's better (pause while wife completes task) W - I think we've broken almost every glass i n that set now. H 75: I know ... i t ' s a shame too ... because they're r e a l l y nice glasses. W - Oh well ... we can always get some more ... what time i s i t ? H 76: (looks at watch) About 7:15. W - I'd l i k e to be going ... I'm hungry. H 77: What about my c a l l ? W - I don't know about that ... but I'm starved. A H 78: (hugs wife) Okay honey ... I suppose i f you r e a l l y want to go now ... I can always try to c a l l them from the restaurant. W - Do you remember how to get to the restaurant? N_ H 79: Not exactly ... how about you? W - Sort of. H 80: Well I'm sure between the two of us we can f i n d i t . W - You know what I'd r e a l l y l i k e to do afterwards? H 81: What? - 114 -APPENDIX 2 Vignette Transcript (Cont'd.) W - I'd l i k e to see that new art g a l l e r y that opened up downtown. I know you don't r e a l l y l i k e browsing around g a l l e r i e s , but how about indulging me tonight? A H 82: Well ... okay honey ... I ' l l go to the g a l l e r y with you a f t e r dinner. W - Thanks David ... I appreciate i t ... I hear that i t ' s supposed to be r e a l l y i n t e r e s t i n g . Terry was there the other day and r e a l l y enjoyed i t . N H 83: What type of art does i t have? W - Well i t ' s supposed to be quite varied ... but a l o t of i t i s apparently new a r t which i s being produced i n Quebec ... Anyways I ' l l get my coat ... N_ H 84: Okay ... I ' l l be waiting outside ... I want to clean the windshield. - 115 -APPENDIX 3 False Memory Items Dominant Items ( > 80% agreement) 1. I don't want to go to New York, Susan. 2. Just cook the chicken and stop worrying about i t . 3. Don't make trouble ... Susan. 4. Susan ... I need to use the phone ... now. 5. You spend too much time procr a s t i n a t i n g . You've got to make up your mind now. 6. I have no i n t e n t i o n of watching T.V. tonight. 7. I'm not going to have f i s h again. 8. No ... I need the car a l l day tomorrow. 9. Susan ... I want to have them over ... and that's a l l there i s to i t . 10. Susan ... t h i s i s the second time I'm t e l l i n g you. I have to use the phone. 11. I'm not going to f i n i s h t h i s conversation because you're getting upset. 12. I'd l i k e you to c a l l them up and i n v i t e them over. Agreeable Items (>, 80% agreement) 1. I know i t ' s a d i f f i c u l t decision. I'm sorry i f I pressured you. 2. Sure ... you know I'm always happy to lend you the car i f you need i t , 3. If that's r e a l l y what you want ... I ' l l go to the g a l l e r y with you. 4. I can see your point about the book. 5. I ' l l watch Merv G r i f f i n with you i f you want. 6. I'm kind of hungry, too. I can c a l l them l a t e r . 7. I'd be happy to drive down to the market and see. 8. Maybe we can both give your s i s t e r a hand. 9. If you'd rather make something else ... that's f i n e with me too. 10. I'm sorry i f I hurt your f e e l i n g s ... Susan. 11. Maybe we can work out some sort of compromise. 12. I f e e l the same way as you do about Montreal i n the r a i n . - 116 -APPENDIX 3 False Memory Items (Cont'd.) Neutral Items (> 70% agreement) 1. I broke a glass just the other day. 2. Which dish do you mean? 3. What w i l l you make for them? 4. What's on T.V.? 5. I didn't know your s i s t e r was moving. 6. It was r e a l l y busy at the o f f i c e today. 7. Yes,... I remember the lake. 8. Has Terry gone to the g a l l e r y yet? 9. I'm going to use a new recipe t h i s time. 10. Susan ... i s that Terry on the phone? 11. It's always hard deciding where to go. 12. Of course I remember how to get to the restaurant. - 117 -APPENDIX 4 Memory Recognition Test Below are a s e r i e s of statements. Some of these statements were made by David i n the l a s t f i l m you saw, while others were not. On the answer sheet provided, please rate each statement according to how confident you are that t h i s exact statement was made by David, on a scale of 1 to 6. Remember ... I want to know i f he made that exact statement i n the f i l m . 1 2 3 4 5 6 completely completely confident he confident he didn't say i t s a i d i t 1. I broke a glass the other day. 2. I know i t ' s a d i f f i c u l t d e c i s i o n . I'm sorry I pressured you. 3. P h i l l i p doesn't l i k e f i s h . I think you should make something else . 4. Sure ... you know I'm always happy to lend you the car i f you need i t . 5. Not exactly. How about you? 6. If that's r e a l l y what you want I ' l l go to the g a l l e r y with you. 7. I'm glad they're pr e d i c t i n g good weather. 8. I can see your point about the book. 9. I ' l l watch Merv G r i f f i n with you i f you want. 10. I'm r e a l l y glad you l i k e i t . 11. Which d i s h do you mean? 12. I'm kind of hungry too. I can c a l l them l a t e r . 13. What w i l l you make f o r them? 14. No ... y o u ' l l have to c a l l her back, Susan. 15. I don't want to go to New York ... Susan. . 16. Don't get angry Susan ... i t ' s not going to help the s i t u a t i o n . 17. Well ... that sounds good to me too. We can go out f o r Chinese food another time. 18. Just cook the chicken ... and stop worring about i t . 19. Oh ... when d i d he get back in t o town? 20. What type of art does i t have? 21. Don't make trouble ... Susan. 22. What's on T.V.? - 118 -APPENDIX 4 Memory Recognition Test (Cont'd.) 2 3 . Sure ... I ' l l take the bus to work. 2 4 . You've had a r e a l l y good appetite l a t e l y . 2 5 . I'd be happy to drive down to the market and see. 2 6 . Okay ... I ' l l f i n i s h reading the paper. 2 7 . Do you need the car i n the morning. 2 8 . I didn't know your s i s t e r was moving. 2 9 . Okay honey ... I suppose i f you r e a l l y want to go now ... I can always try to c a l l them from the restaurant. 3 0 . You're a great cook ... Susan. 3 1 . I don't f e e l l i k e going out for Greek food. 3 2 . It was r e a l l y busy at the o f f i c e , today. 3 3 . Susan ... I need to use the phone ... now. 3 4 . Would you l i k e me to put some extra tonic i n your drink? 3 5 . I t was a c t u a l l y quite bad. 3 6 . Yeah ... I r e a l l y enjoyed being with you on that holiday. 3 7 . You spend too much time procrastinating. You've got to make up your mind now. 3 8 . I know. I don't think we've eaten out much i n general l a t e l y . 3 9 . Are you okay, Susan? I hope you didn't hurt y o u r s e l f . 4 0 . Maybe we can both give your s i s t e r a hand. 4 1 . No ... I don't f e e l l i k e going to the g a l l e r y . 4 2 . I have no i n t e n t i o n of watching T.V. tonight. 4 3 . Yes ... I remember the lake. 4 4 . Well okay ... i f i t ' s only a few minutes. 4 5 . I'm sorry Susan ... I don't mean to make i t too strong. 4 6 . Oh ... I see i t . We're almost out you know. 4 7 . Don't go storming o f f Susan ... s i t down. 4 8 . Has Terry gone to the g a l l e r y yet? 4 9 . Why don't you want to have them over? 5 0 . I invented i t . 5 1 . I'm not going to have f i s h again. 5 2 . He t r i e d too hard to imitate D.H. Lawrence. - 119 -APPENDIX 4 Memory Recognition Test (Cont'd.) 53. No ... I need the car a l l day tomorrow. 54. Susan ... could you c a l l Terry back. I'm expecting an important phone c a l l . 55. If you'd rather make something else ... that's f i n e with me too. 56. Yeah ... I've got i t . 57. I don't think I make the drinks too strong. 58. I'm sorry i f I hurt your f e e l i n g s , Susan. 59. I'm too busy to help your s i s t e r move t h i s weekend. 60. I'm going to use a new recipe t h i s time. 61. Please Susan ... I'd r e a l l y l i k e to have them over. 62. Yes ... I think i t ' s by the new bank downtown. 63. That's a good idea. It's r e a l l y a nice dish. 64. I'm just not i n the mood f o r Merv G r i f f i n tonight ... comeon, l e t ' s have a drink. 65. Okay ... I ' l l be waiting outside. I want to clean the windshield. 66. Well ... i t ' s about time he c a l l e d . 67. I r e a l l y didn't l i k e the lake. 68. Susan ... I want to have them over ... and that's a l l there i s to i t . 69. Okay honey ... I ' l l try to make i t the way you l i k e i t t h i s time. 70. L i s t e n Susan ... we owe them a dinner and I r e a l l y want to have them over. 71. Maybe we can work out some sort of compromise. 72. Susan ... i s that Terry on the phone? 73. No ... P h i l l i p doesn't l i k e chicken. 74. Sure ... I'd be glad to help your s i s t e r move. 75. Oh ... what did he have to say f o r himself? 76. Susan ... t h i s i s the second time I'm t e l l i n g you ... I have to use the phone. 77. I t ' s always hard deciding where to go. 78. Of course I remember how to get to the restaurant. 79. I f e e l the same way as you do about Montreal i n the r a i n . - 120 -APPENDIX 4 Memory Recognition Test (Cont'd.) 80. Don't pick up that glass with your hands, Susan. Use the broom and dustpan. 81. I'm not going to f i n i s h t h i s conversation because you're getting upset. 82. Is there something about the job which i s putting you off ? 83. I'd l i k e you to c a l l them up and i n v i t e them over. 84. I'm sorry i f I hurt your f e e l i n g s ... Susan. 85. The longer you think about i t ... the harder i t w i l l get. Make up your mind and give them a c a l l . 86. I can't think of any places offhand ... but I think salmon i s i n season t h i s time of year. 87. I'm r e a l l y looking forward to seeing by brother. 88. Can I make you a drink while you're on the phone? 89. Yes ... I remember the lake. 90. Okay honey ... I ' l l be happy to go along with whatever. 91. That's encouraging. I hope i t ' s accurate. 92. Not exactly ... how about you? 93. Well ... okay honey ... I ' l l go to the g a l l e r y with you a f t e r dinner. 94. You'd better pick up that glass before we go. 95. Please don't be angry with me Susan. 96. No ... I just can't a f f o r d the extra time i t would take up to go to New York. - 121 -APPENDIX 5 Wiggins Interpersonal Adjective Scales Below are 8 r a t i n g scales f o r summarizing your impression of David's behaviour i n the l a s t f i l m you saw. Each scale i s l a b e l l e d by 4 adjectives. For each scale, please consider the 4 adjectives together. Your task i s to i n d i c a t e how accurately each scale describes your impression of David, from 1 (extremely inaccurately) to 8 (extremely a c c u r a t e l y ) . Please think back to the f i l m while doing t h i s . 1 2 3 4 5 6 7 8 extremely very quite s l i g h t l y s l i g h t l y quite very extremely inaccurate inaccurate inaccurate inaccurate accurate accurate accurate accurate extraverted, outgoing e n t h u s i a s t i c , cheerful 4 5 submissive, timid meek, self-doubting 4 5 quarrelsome, impolite discourteous, d i s r e s p e c t f u l 4 5 6 agreeable, courteous co-operative, accommodating 4 5 6 c a l c u l a t i n g , s l y t r i c k y , c r a f t y 4 5 aloof, a n t i s o c i a l unneighbourly, d i s t a n t 4 5 dominant, assertive f o r c e f u l , domineering 4 5 l a z y , unproductive disorganized, i m p r a c t i c a l 4 5 6 - 122 -APPENDIX 6 Consent Form This research i s designed to investigate the fashion i n which people make s o c i a l judgments. We are interested i n fi n d i n g out how people obtain the information to make these judgments. You w i l l be shown some videotape scenes i n which two people are i n t e r a c t i n g . We w i l l be asking you to push a button whenever something takes place which helps you to form an impression of these people. Following t h i s you w i l l be asked to match some descriptions of people to t h e i r p i c t u r e s . The e n t i r e study w i l l take about 60 minutes. You are free to withdraw from the study at any time or to refuse to answer any questions you do not wish to answer. Neither of these actions w i l l be held against you i n any way. I have read and understand the above statement and f r e e l y agree to p a r t i c i p a t e i n the study. Name: Date: - 123 -APPENDIX 7 F i l l e r Task Please match the following descriptions to the pictures which are shown to you on the screen. A. Simon P. i s a graduate student i n anthropology. He i s married and has one small c h i l d . His favourite sports are s k i i n g and tennis. He a c t i v e l y p a r t i c i p a t e d i n track and f i e l d sports during high school, but i s now no longer a c t i v e i n those areas. He was born i n Philadelphia and moved to Vancouver to attend graduate school. B. Susan T. i s a community health nurse. She i s s i n g l e . She enjoys playing chess and other board games. While she i s not p a r t i c u l a r l y a t h l e t i c , she does swim occasionally. One of her favourite pastimes i s singing i n the church choir. She has been doing this since childhood and enjoys i t thoroughly. C. Andrea B. i s a dental hygienist. She i s currently engaged to be married. Her fiance i s a lawyer. Her favourite pastimes are s a i l i n g and s k i i n g . She was born i n Vancouver and has l i v e d here a l l her l i f e . She and her finace are intending to move to another c i t y when they are married, but they have not yet decided where. D. P h y l l i s S. i s a teacher. She i s separated from her husband. Although she received her teacher t r a i n i n g some time ago, she has not been able to f i n d permanent work i n the c i t y , and has been working as a substitute teacher f o r the l a s t two years. For t h i s reason, she i s considering moving to a smaller town, where there i s a greater demand f o r teachers. Her favourite pastimes are reading and hiking. E. Shaun K. i s an undergraduate student who i s majoring i n biology. He i s thinking of applying to medical school when he i s f i n i s h e d h i s degree, but may also apply to graduate school i n biology. His favourite sports are hiking, mountain climbing and s k i i n g . He very r a r e l y does any l e i s u r e time reading, but when he does i t i s usually science f i c t i o n . - 124 -APPENDIX 7 F i l l e r Task (Cont'd.) F. Tom R. Is a stock broker. He and h i s wife move to Vancouver recently from Montreal. One of his favourite pastimes i s gourmet cooking. He a l s o enjoys golf and f i s h i n g . He enjoys reading, and i s p a r t i c u l a r l y interested i n h i s t o r i c a l f i c t i o n . He has not done much t r a v e l l i n g , but he and h i s wife are planning to go to Europe f o r a vacation next year. G. Paul T. i s a lawyer. He has three c h i l d r e n . When he was younger, he always wanted to be a j o u r n a l i s t , but decided to go i n t o law for reasons of greater f i n a n c i a l s e c u r i t y . Now he i s f e e l i n g bored with his work, and i s considering a career change. His favourite pastimes are golf and tennis. H. Anne K. i s a potter. She i s married and has no chi l d r e n . She i n i t i a l l y began making pottery as a hobby, but over time became more serious about i t . She i s now quite successful at her trade, and makes pottery for a number of stores i n the c i t y . For l e i s u r e a c t i v i t i e s she l i k e s to weave and read poetry. I. Kathy D. i s a s o c i a l worker. Her area of s p e c i a l t y i s family counselling. She, h e r s e l f , comes from a large family i n a small town. She moved to Vancouver to attend u n i v e r s i t y and has l i v e d here since that time. She l i k e s to s a i l and to cook gourmet meals. J. Donna P. i s married and has two grown c h i l d r e n . Although she was trained as a biochemist, she cu r r e n t l y does not work. This i s her second marriage. Her f i r s t husband died a few years ago. Her current husband i s r e t i r e d and i s reasonably w e l l o f f . They lead a f a i r l y a c t i v e l i f e , enjoying a v a r i e t y of d i f f e r e n t types of entertainment including theatre, movies, and music. They also t r a v e l extensively. - 125 -APPENDIX 7 F i l l e r Task (Cont'd.) K. Janet L. c u r r e n t l y works as a c l e r k at family court. She majored i n h i s t o r y at u n i v e r s i t y . She has recently separated from her boyfriend of f i v e years and i s now unattached. Her major pastimes are drawing and painting. Her artwork i s quite important to her, and i n fa c t at times she has t r i e d to go profe s s i o n a l . She has not been able to make a l i v i n g at i t , however, and has had to f i n d other jobs to support h e r s e l f . L. Diane H. i s a graduate student i n English l i t e r a t u r e . She has l i v e d with her boyfiend f o r 3 years. She i s very a c t i v e i n the women's movement, and i s cu r r e n t l y e d i t i n g a book on women's poetry. She takes yoga and modern jaz dance f o r exercise. M. Ben M. i s married and has three c h i l d r e n . He owns a sporting goods store and i s an a c t i v e sports enthusiast himself. He enjoys f i s h i n g , g o l f , and camping. He and h i s wife have a small cabin on one of the Gulf i s l a n d s , where they l i k e to spend much of t h e i r spare time. Another l e i s u r e a c t i v i t y which he enjoys i s l i s t e n i n g to opera. N. Doug L. i s a graduate student i n r e l i g i o u s studies. This i s h i s f i r s t year i n the program and he i s enjoying i t thoroughly. He i s very pleased about t h i s , since he had some d i f f i c u l t y deciding what he wanted to do a f t e r he f i n i s h e d h i s B.A. The other major options he was considering were e i t h e r psychology graduate school or medical school. He drive a t a x i on weekends to support himself and to pay t u i t i o n . 0. Margaret A. i s married and has three c h i l d r e n . Two of them are married, and the t h i r d i s s t i l l l i v i n g at home. She i s employed as a s p e c i a l education teacher i n the school system. She and her husband have recently moved out of a large house and bought a smaller condominium. The move was somewhat d i f f i c u l t f o r them at f i r s t , but they are now adjusting to i t quite w e l l . - 126 -APPENDIX 7 F i l l e r Task (Cont'd.) P. Deborah H. might be described as going through a period of t r a n s i t i o n . She was recently divorced from her husband and she also l e f t the job where she was employed for the l a s t f i v e years. She has been thinking of returning to l i v e i n England where she was born. For t h i s reason, she has decided to go back there f o r a holiday to see i f she enjoys i t . Her major pastimes are reading and playing the recorder. She has recently joined a chamber group and finds t h i s very rewarding. Q. Ken T. i s an undergraduate who i s studying theatre and f i n e a r t s . His ambition i s to become a professional actor. He recently t r i e d out f o r a small role i n a u n i v e r s i t y play. Although he did f a i r l y w e l l he d i d not get the part. He i s f e e l i n g quite discouraged about t h i s , but h i s i n s t r u c t o r has t o l d him that i t i s rare f o r f i r s t year students to get parts. His major pastime i s photography and he recently placed f i r s t i n a p r o v i n c i a l competition. - 127 -APPENDIX 8 Analysis of Tracking Data: Multiple Response Format In the r e s u l t s section, a d e c i s i o n was made to analyze subjects' tracking responses as binary data. In other words, i t was assumed that a given subject e i t h e r did or d i d not track a given i n t e r v a l . Multiple responses for a subject, within a given i n t e r v a l were ignored i n t h i s a n a l y s i s . Although as previously discussed, there are reasons f o r considering the data i n t h i s fashion, i t can also be argued that t h i s procedure ignores important information. A subject who tracks a given i n t e r v a l more than once may be giving some indication, of the degree of subjective salience of that p a r t i c u l a r i n t e r v a l . Since at this point i n time there are no accepted conventions regarding the analysis of tracking data, this p o s s i b i l i t y cannot be ruled out, and f o r this reason the r e s u l t s obtained when multiple responses were scored w i l l be reported i n the present section. The T-square analysis indicated that, o v e r a l l , subjects i n the dominant condition tracked dominant i n t e r v a l s s i g n i f i c a n t l y more frequently than d i d subjects i n the agreeable condition, T 2 = 25.711, F(8,91) = 2.984, p < .005. There was a consistent bias i n t h i s d i r e c t i o n on every i n t e r v a l , but multiple comparisons yielded no s i g n i f i c a n t differences on any i n d i v i d u a l mean comparisons. Subjects i n the agreeable condition tracked the subset of agreeable i n t e r v a l s s i g n i f i c a n t l y more frequently than subjects i n the dominant condition, T 2 = 24.125, F(8,91) = 2.80, p < .01. Again there was a consistent bias In this d i r e c t i o n on every i n t e r v a l , but no s i g n i f i c a n t differences on i n d i v i d u a l i n t e r v a l s . On the repeated measures ANOVA there was a s i g n i f i c a n t main e f f e c t due to i n t e r v a l type, F(2,192) = 220.211, p < .001, and a s i g n i f i c a n t i n t e r a c t i o n between expectancy condition and i n t e r v a l type, F(2,192) 22.469, p < .001. There were no other s i g n i f i c a n t e f f e c t s . Post hoc comparisons were again conducted with Bonferonni t - t e s t s . A l l tests were two-tailed. - 128 -APPENDIX 8 Analysis of Tracking Data: Multiple Response Format (Cont'd.) Collapsing across expectancy conditions, subjects tracked both dominant and agreeable i n t e r v a l s s i g n i f i c a n t l y more frequently than neutral i n t e r v a l s , t(99) = 20.08, p < .001, and t(99) = 20.66, p < .001 respectively. There were no s i g n i f i c a n t differences o v e r a l l between dominant and agreeable i n t e r v a l s , t(99) = 1.15, p > .759. Subjects tracked dominant and agreeable i n t e r v a l s equally frequently. The post hoc comparisons r e p l i c a t e d the r e s u l t s of the T-square i n that subjects i n the dominant condition tracked dominant i n t e r v a l s more frequently than did subjects i n the agreeable condition, t(98) = 4.02, p <. .001, and subjects i n the agreeable con d i t i o n tracked agreeable i n t e r v a l s more frequently than did subjects i n the dominant condition, t(98) = 2.83, p < .05. There were no s i g n i f i c a n t d ifferences between subjects i n the two conditions on neutral i n t e r v a l s , t(98) = .82, p > .125. The present r e s u l t s c l o s e l y p a r a l l e l those obtained when multiple responses were not scored, and suggest that the two procedures may be used interchangeably i n future research. - 129 -APPENDIX 9 Analysis of Ninth Agreeable Interval Inspection of the f i l m t r a n s c r i p t indicates the presence early on i n the i n t e r a c t i o n of an i n t e r v a l during which the husband's behaviour (although not categorized as such) appears to be quite agreeable i n nature ( i . e . , "I'm sorry, that wasn't very nice of me. I guess I'm f e e l i n g kind of stressed"). Despite the apparent agreeableness of this act, however, i t was not included i n the main analyses. This i s because t h i s section was added on a f t e r subjects rated the husband's behaviour i n the i n i t i a l videotape as s i g n i f i c a n t l y more dominant than agreeable, i n order to balance the ratings (see p. 50). For t h i s reason, agreement i n categorization of this i n t e r v a l had not been em p i r i c a l l y established with a large sample, and i t was thus excluded from the main a n a l y s i s . A second analysis of the tracking data, was however conducted, i n which t h i s n i n t h i n t e r v a l was included. This i n t e r v a l , i n f a c t , turned out to be p a r t i c u l a r l y s e n s i t i v e to expectancy e f f e c t s , as evidenced by the fact that the T-square value rose to T 2 = 46.75, F(9,90) = 4.47, p < .0001, when i t was included (binary data). The group tracking means f o r the i n t e r v a l were: 1. binary data - dominant subjects: .52, agreeable subjects: .92 2. multiple response format - dominant subjects: .54, agreeable subjects: 1.00. The s e n s i t i v i t y of t h i s i n t e r v a l to expectancy e f f e c t s may stem from the f a c t that because i t was so close to the beginning of the i n t e r a c t i o n , agreeable subjects were p a r t i c u l a r l y well primed to attend to i t since t h e i r agreeable expectations had not yet been modified by exposure to any dominant acts whereas dominant subjects were p a r t i c u l a r l y l i k e l y to disregard i t since i t was only the f i r s t instance of an expectancy disconfirming act. This s e n s i t i v i t y to expectancy e f f e c t s was also true of the f i r s t dominant - 130 -APPENDIX 9 Analysis of Ninth Agreeable Interval (Cont'd.) i n t e r v a l (see Table 5). One factor p o t e n t i a l l y contributing to expectancy confirmation processes i s that i n d i v i d u a l s may terminate the i n t e r a c t i o n f a i r l y e a rly on i n the i n t e r a c t i o n sequence before many disconfirming events take place (Darley & Fazio, 1980). This early termination may combine with the type of i n t e r v a l primacy e f f e c t discussed here, i n order to further maintain dysfunctional expectations. - 131 -APPENDIX 10 Correlations Between Cumulative Interval  Tracking Scores and Interpersonal Adjective Ratings Another possible means of evaluating the construct v a l i d i t y of the tracking procedure consists of assessing the pattern of r e l a t i o n s h i p s which emerge between tracking a c t i v i t y and the eight interpersonal dimensions which were sampled from Wiggins (1979) circumplex taxonomy. Is i t possible to predict subjects' ratings of the husband along these eight dimensions on the basis of t h e i r tracking a c t i v i t y ? We have already seen that subjects who frequently tracked dominant acts were l i k e l y to rate the husband high on the dominant scale, whereas subjects who frequently tracked agreeable i n t e r v a l s were l i k e l y to perceive the husband as an agreeable i n d i v i d u a l . The circumplex model also enables us to postulate a hypothetical ordering of r e l a t i o n s h i p s between tracking a c t i v i t y and the perception of other interpersonal dimensions and to then evaluate the actual degree of f i t with t h i s hypothesized pattern. It i s t h e o r e t i c a l l y predictable, for example, that subjects who frequently tracked dominant i n t e r v a l s should tend to give the husband t h e i r highest ratings on the dominant adjective scale and t h e i r lowest ratings on the submissive adjective scale since the hypothetical c o r r e l a t i o n between dominance and dominance i s .10 while the hypothetical c o r r e l a t i o n between dominance and submissiveness i s -.10. The ordering of ratings on the other adjective scales should be predictable on the basis of the proximity of any given dimension to the dominant dimension on the circumplex. One would predict then that the positive r e l a t i o n s h i p between tracking dominant i n t e r v a l s and ratings on the c a l c u l a t i n g adjective scale would be weaker than the p o s i t i v e r e l a t i o n s h i p between tracking dominant i n t e r v a l s and ratings on the dominant scale, but stronger than the p o s i t i v e r e l a t i o n s h i p between tracking dominant i n t e r v a l s and ratings on the lazy scale (the t h e o r e t i c a l c o r r e l a t i o n between dominance and the c a l c u l a t i n g and lazy dimensions are .50 and -.75 respectively. - 132 -APPENDIX 10 Correlations Between Cumulative Interval  Tracking Scores and Interpersonal Adjective Ratings (Cont'd.) In the following tables, the adjective scales are presented i n order of t h e i r t h e o r e t i c a l l y predicted c o r r e l a t i o n s with the tracking a c t i v i t y from highest to lowest (tracking values were calculated by summing each subject's responses f o r the eight relevant i n t e r v a l s ) . The hypothetical c o r r e l a t i o n s between t r a i t p a i r s i n the circumplex are compared with the obtained c o r r e l a t i o n s between t r a i t s and tracking behaviours. It was predicted that the order of c o r r e l a t i o n s would be perserved but not that the exact c o r r e l a t i o n values would be r e p l i c a t e d . Correlations Between Dominant I n t e r v a l  Tracking Score and Adjective Scales Obtained multiple adjective scale hypothetical binary data responses 1. dominant 1.00 .34 .34 2. c a l c u l a t i n g .50 .19 .15 3. extraverted .50 .04 * .11 4. quarrelsome .00 .37 * .29 * 5. agreeable .00 -.41 * -.39 * 6. aloof -.25 .10 .04 7. lazy -.75 -.05 -.11 8. submissive -1.00 -.24 -.25 Correlations Between Agreeable Interval Tracking Score and Adjective Scales Obtained multiple adjective scale hypothetical binary data responses 1. agreeable 1.00 .27 .29 2. extraverted .50 .02 .02 3. submissive .00 .11 * .11 * 4. dominant .00 -.27 * -.25 * 5. lazy -.25 -.23 -.23 6. c a l c u l a t i n g -.50 -.32 -.32 7. aloof -.75 -.32 -.32 8. quarrelsome -1.00 -.39 -.37 * These c o r r e l a t i o n s did not conform to the predicted ordering. - 133 -APPENDIX 10 Correlations Between Cumulative Interval  Tracking Scores and Interpersonal Adjective Ratings (Cont'd.) Inspection of the above tables indicates a reasonably good f i t between the hypothesized and obtained c o r r e l a t i o n orderings, with a few exceptions to t h i s r u l e . The most notable of these are the i n f l a t e d c o r r e l a t i o n s between the dominant i n t e r v a l tracking score and the quarrelsome and agreeable adjective ratings i n the f i r s t table, and the i n f l a t e d c o r r e l a t i o n s between the agreeable i n t e r v a l tracking score and the submissive and dominant adjective scales i n the second table. These departures from the hypothesized ordering can be accounted f o r by the f a c t that subjects tended to see the dimensions of dominance and agreeableness as negatively c o r r e l a t e d i n the context of the present study (see discussion s e c t i o n ) . The quarrelsome - agreeable dimension thus tended to merge or become interchangeable with the dominant - submissive dimension. As previously discussed t h i s c o l l a p s i n g of dimensions may have resulted from the f a c t that the a s c r i p t i o n of t r a i t s to others follows a d i f f e r e n t set of rules than the a s c r i p t i o n of t r a i t s to the s e l f and/or the fac t that dominant and agreeable acts were set i n contrast to one another i n the context of the stimulus vignette i n a fashion which may have made them appear to be polar opposites. Bearing t h i s i n mind, i t appears that the f i t between the hypothetical and obtained c o r r e l a t i o n orderings was a c t u a l l y quite good. These data suggest that there i s a t h e o r e t i c a l l y predictable l i n k between encoding a c t i v i t y as measured by the tracking procedure and the perception of a v a r i e t y of d i f f e r e n t dimensions of interpersonal behaviour. It i s thus consistent with the hypothesis that the s e l e c t i v e encoding of events has an important mediating e f f e c t upon the perception of interpersonal behaviour and provides a form of construct v a l i d a t i o n of the tracking measure. F i n a l l y , these data h i g h l i g h t the u t i l i t y of Wiggins (1979) circumplex taxonomy of interpersonal t r a i t s f o r making t h e o r e t i c a l predictions to f a c i l i t a t e t h i s type of person perception research. - 134 -

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}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            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.831.1-0095620/manifest

Comment

Related Items