UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Nonverbal sensitivity as a function of social anxiety Block, Loretta Anna May 1984

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

Item Metadata

Download

Media
831-UBC_1984_A8 B56.pdf [ 3.29MB ]
Metadata
JSON: 831-1.0096222.json
JSON-LD: 831-1.0096222-ld.json
RDF/XML (Pretty): 831-1.0096222-rdf.xml
RDF/JSON: 831-1.0096222-rdf.json
Turtle: 831-1.0096222-turtle.txt
N-Triples: 831-1.0096222-rdf-ntriples.txt
Original Record: 831-1.0096222-source.json
Full Text
831-1.0096222-fulltext.txt
Citation
831-1.0096222.ris

Full Text

NONVERBAL SENSITIVITY AS A FUNCTION OF SOCIAL ANXIETY By LORETTA ANNA MAY BLOCK B .A . , The Univers i ty of B r i t i s h Columbia, 1980 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS in THE FACULTY OF GRADUATE STUDIES (Department of Psychology) We accept t h i s thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA February, 1984 "Q\ Loretta Anna May Block, 1984 In present ing t h i s thes is i n p a r t i a l f u l f i l m e n t of the requirements fo r an advanced degree at the Un ivers i t y of B r i t i s h Columbia, I agree that the L ib ra ry s h a l l make i t f r e e l y a v a i l a b l e fo r reference and study. I fu r ther agree that permission fo r extensive copying of t h i s thes i s fo r s c h o l a r l y purposes may be granted by the head of my department or by h i s or her representat i ves . I t i s understood that copying or p u b l i c a t i o n of t h i s t h e s i s fo r f i n a n c i a l gain s h a l l not be allowed without my wr i t ten permiss ion . Department of The Un ive rs i t y of B r i t i s h Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date /^L/-»V, DE-6 (3/81) Abstract While c l i n i c a l research has led to the development of treatment programs which have e f f e c t i v e l y reduced socia l anxiety in c l i n i c a l and analogue populat ions, one of the major shortcomings in th i s area has been the lack of conceptual c l a r i t y . The present study attempted to e lucidate the nature of the prevalent socia l problem in an invest igat ion of the r e l a t i o n between s e n s i t i v i t y to nonverbal soc ia l cues and social anxiety . Twenty-four s o c i a l l y anxious women and 24 s o c i a l l y nonanxious women were selected from undergraduate classes on the basis of the i r scores on the Social Avoidance and Distress Scale (SAD). Moderately and severely depressed subjects were excluded from the study using the Beck Depression Inventory (BDI). Upon t h e i r a r r i v a l at the laboratory , subjects interacted with a female confederate for 4 minutes. For one ha l f of the i n t e r a c t i o n , the confederate displayed a relaxed demeanor. During the second i n t e r v a l , the confederate a l tered her behaviour to portray an anxious s ta te . Subjects were randomly assigned to relaxed-anxious or anxious-relaxed condi t ions . Following the i n t e r a c t i o n , subjects rated the i r level of comfort and that of the confederate. In add i t ion , subjects were asked to describe how the confederate responded to them during t h e i r conversation. The subsequent essays were coded by two independent judges. Subjects were also asked d i rec t l y whether or not they perceived a change i n the confederate's behaviour. The in teract ions were videotaped and each two minute interval was rated by two independent ( i i ) judges along the fol lowing dimensions: observation, postur ing, gestures, head nods, number of pauses and time spent t a l k i n g . Judges also rated subjects on global measures of social s k i l l and anxiety. Mul t i var ia te analyses of covariance using depression as a covariate were used for data ana lys i s . Based on Croz ier ' s (1979) notion of anxious self -preoccupation i t was predicted that s o c i a l l y anxious subjects would be less sens i t i ve to changes in the nonverbal behaviour of the confederate compared to s o c i a l l y nonanxious subjects . D i f fe rent ia l s e n s i t i v i t y across groups was not demonstrated. S ign i f i cant di f ferences between s o c i a l l y anxious and nonanxious subjects were not obtained on subjects' s e l f - r e p o r t s and only one of six in teract ion measures d i f fe rent ia ted the two groups. Findings were interpretted in l i g h t of the methodological l i m i t a t i o n s of the current study and in the context of the relevant research. The e f fect of s t a t i s t i c a l l y cont ro l l ing for depression was discussed in re la t ion to a n c i l l a r y analyses which excluded t h i s covar iate . It was suggested that future research control for the potential confounding influence of depression when invest igat ing social anxiety . Furthermore, i t was recommended that s e n s i t i v i t y to vocal cues in re la t ion to social anxiety could provide a f r u i t f u l avenue for experimental inqui ry . ( i i i ) Table of Contents Page T i t l e Page i Abstract . . . . . i i Table of Contents iv L i s t of Tables v L i s t of Figures .' v i Acknowl edgements . v i i Introduction 1 Method 26 Results . . . 34 Discussion . . . . 4 8 Appendix A. Social Avoidance and Distress Scale 56 Appendix B. Beck Depression Inventory 59 Appendix C. Results of one-way MANOVA for order e f fects 63 References . . . . 6 9 ( iv ) L i s t of Tables Number Descript ion Page 1 Interrater r e l i a b i l i t i e s 35 2 Means and standard deviations for repeated measures across condit ions and groups (n=12) 67 3 Summary of univar iate F - tes ts for condit ion e f fec t 68 4 Means and standard deviat ions for repeated measures between groups (n=24) 41 5 Means and standard deviations for pos t - in te rac t ion essay across groups (n=24) 45 6 Corrected chi -squared s t a t i s t i c s for subjects' responses to perception of change measures (n=24) . . . . 46 (v) L i s t of Figures Number Descript ion Page 1 Observation and posturing as a function of condit ion 64 2 Gestures and nods as a function of condit ion 65 3 Talking and pauses as a function of condit ion 66 4 Observation, posturing and gestures as a function of soc ia l anxiety and confederate's behaviour 42 5 Nods, ta lk ing and pauses as a function of soc ia l anxiety and confederate's behaviour 43 (vi ) Acknowledgements I would l i k e to acknowledge the contr ibut ions of those ind iv iduals who made possible the completion of th i s t h e s i s . F i r s t l y , I would l i k e to thank Lynn Alden for introducing me to an exc i t ing and rewarding area of research. Furthermore, I would l i k e to thank the members of my committee, Keith Dobson, P h i l i p Smith and Jack Rachman, for the i r time and expert ise in preparing t h i s manuscript. A special note of appreciation goes to Keith without whose patience and support, completion of t h i s thesis would have taken a great deal longer. I would also l i k e to acknowledge the many extra hours of work volunteered by my f r i e n d , Heidi Dopp. In add i t ion , many thanks to my research ass i s tan ts , Linda Lee, Diane Cummins, Pindy Badyal and Cathy Burton. Last ly and most importantly , I would l i k e to thank my parents for the i r unending support over the past 3 years . Their encouragement and love made completion of th i s thesis a r e a l i t y . ( v i i ) 1. The re la t ion between social competence and psychiatr ic disorders i s well es tab l ished. In 1961, Zigler and P h i l l i p s provided data suggesting that premorbid level of social functioning was a good predictor of the adjustment of psychiatr ic patients fol lowing a period of h o s p i t a l i z a t i o n . Since then, a growing number of researchers have supported the notion of social s k i l l d e f i c i t s as a basis for major forms of psychopathology (Argyle and Kendon, 1967; Curran and Wessberg, 1981). Furthermore, social incompetence has been related to numerous other dysfunctional behaviours including alcohol abuse ( K r a f t , 1971), sexual problems (Barlow, 1973), drug addiction (Callner & Ross, 1976), and marital discord ( E i s l e r , M i l l e r , Hersen & A l f o r d , 1974). Research has also indicated that social incompetence represents a s i g n i f i c a n t problem for col lege students. For example, dating problems are frequently accompanied by anxiety , depression and academic f a i l u r e (Hopkins, Malleson and Sarnoff , 1958; Lucas, Kelvin and Ojha, 1966; Arkowitz, Hinton, Perl & Himadi, 1978). Martinson and Zerface (1970) reported that students were more interested in learning how to get along with the opposite sex than with receiving help in choosing a vacation or learning about the i r a b i l i t i e s , i n t e r e s t s , in te l l i gence and p e r s o n a l i t i e s . The prevalence of social problems in col lege populations was investigated by Borkovec, Stone, O'Brien and Kaloupek (1974). They reported that 15.5% of males and 11.5% of females in introductory psychology experienced at least some fear of being with a member of the opposite sex and 32% of males and 38.5% of females experience at least 2 some fear of meeting someone for the f i r s t t ime. More recent surveys substantiate the f ind ing that socia l problems, espec ia l l y in heterosocial s i t u a t i o n s , are a major concern for a large proportion of col lege students (Galassi & G a l a s s i , 1979). Zimbardo (1977) surveyed nearly 5,000 people inc luding both col lege and noncollege populations,. More than 80% of those questioned reported that they were "shy at some point in the i r l i v e s " and 40% of th i s group described themselves as "presently shy." The f indings reviewed thus far underscore the importance of research into the nature and treatment of soc ia l incompetence. Social incompetence i s a widespread problem that can in te r fe re with da i l y functioning and may lead to serious l i f e adjustment problems. Over the course of the l a s t f i f t e e n years , there has been a growing in te res t in the development of treatment programs to increase the soc ia l funct ioning of s o c i a l l y incompetent col lege students and other nonpsychiatric populations (Curran, 1975; Rehm & Marston, 1968; Martinson & Zerface, 1970; Malkiewich and Mer luzz i , 1980). Most of the treatment research may be categorized into one of three major theoret ica l formulations. The c l a s s i c a l condit ioning model states that socal incompetence i s the resu l t of conditioned anxiety (Wolpe, 1973). According to t h i s view, unpleasant social experiences are associated with various cues in soc ia l s i t u a t i o n s . Anxiety i s then aroused when a n t i c i p a t i n g or responding to s i m i l a r s i tuat ions and resu l ts in poor performance, avoidance or escape. I t i s assumed that the problem can be el iminated by deconditioning the anxiety thus al lowing for the expression of more appropriate behaviours. Empirical evidence in support of t h i s conceptual ization comes from studies demonstrating the effect iveness of systematic desensi t izat ion in reducing social anxiety (Bander et a l , 1975; Mitchel l & Orr, 1974; Fishman & Nawas, 1973). The s k i l l d e f i c i t s model states that social incompetence is due to an inadequate or inappropriate behavioural repertoire (Curran, 1977). Limited social s k i l l s do not allow the individual to e f f e c t i v e l y handle the demands of the s i t u a t i o n . The consequence is poor social performance and in some cases, anxiety , avoidance or escape. Social s k i l l s t ra in ings or response - acquis i t ion treatment programs exemplify t h i s model (Twentyman & M c F a l l , 1975; Curran, Gi lbert & L i t t l e , 1976; MacDonald et a l , 1975). Las t l y , the cognit ive - evaluative model defines the source of social incompetence as the faulty appraisal of one's performance and the expectation of negative consequences (Galassi & G a l a s s i , 1979). Because of negative sel f -eval uat ions, excessively hign performance standards, u n r e a l i s t i c expectations, i r r a t i o n a l b e l i e f s , fau l t y perceptions or misinterpretat ion of feedback, s o c i a l l y incompetent ind iv iduals perform poorly in social s i tuat ions or avoid them. Treatment programs based on th is conceptualization have employed strategies such as rat ional restructur ing (Kanter & Goldf r ied , 1979) and self - re inforcement (Rehm & Marston, 1968) to reduce the high level of social anxiety experienced by s o c i a l l y incompetent i n d i v i d u a l s . While these theoret ica l formulations are based on d i f f e r i n g views, treatment programs generated from these three perspectives have demonstrated the i r e f f i cacy in reducing targeted 4 problem areas in s o c i a l l y incompetent populations (Curran, 1977; Malkiewich & Mer luzz i , 1980; Curran & G i lbe r t , 1975). For many years , the majority of research on social incompetence focused on treatment methods. More recent ly , however (Curran, 1977; Hersen & Be l lack , 1977), assessment issues have received an increasing amount of a t tent ion . Bellack (1983) uncovered 35 a r t i c l e s dealing with the assessment of social s k i l l s in adults published in behavioural journals and the Journal of Consulting and C l in i ca l Psychology in 1980 and 1981. The current dilemma facing researchers and c l i n i c i a n s working in th i s area i s summarized n ice ly by Curran and Wessberg (1981): "We have been struggling to deal with issues t h a t , as y e t , have not been defined or c l a r i f i e d " (p. 420). They concur with Hersen and Bel l a c k ' s (1977) observation that most of the simple treatment-outcome studies have been done and that i t i s now time to do the more d i f f i c u l t assessment s tud ies . Curran's (1979) explanation of t h i s issue i s poignantly recounted in an a r t i c l e e n t i t l e d "Pandora's Box Reopened? The Assessment of Social S k i l l s . " The term social incompetence used thus far was chosen a r b i t r a r i l y . Researchers have used a var iety of terms to refer to research in t h i s area including social inadequacy, social dysfunction and social s k i l l s (Curran & Wessberg, 1981; Curran, 1979). While t h i s difference in terminology adds some confusion when reading the l i t e r a t u r e , the same questions faces a l l reseachers. What i s social s k i l l ? What are the components of adequate social performance? As Curran (1979) s ta ted , we cannot measure what we cannot define or more importantly as c l i n i c i a n s , 5 we cannot e f f e c t i v e l y t reat that which we cannot measure accurately . To summarize, research on social competence began with a focus on treatment-outcome studies . While treatment methods were found to be e f f e c t i v e , de f in i t i ona l and assessment issues gradually came to the forefront in the l i t e r a t u r e (Curran, 1977). Over the las t seven years , researchers have emphasized the importance of viewing social competence as a "megaconstruct" (Curran, 1979) and the necessity of developing more sophist icated methodologies to assess th i s construct properly (Be l lack , a 1983). Researchers have become keenly aware of the lack of conceptual integration in t h i s area (Schlenker & Leary, 1982) and the urgency of ref in ing our questions and assessment strategies in order to organize and interpret the large body of data that has accumulated (Curran & Wessberg, 1981). Social Anxiety and Social S k i l l Within the realm of assessment issues associated with research in t h i s area, the concepts social s k i l l and social anxiety are frequently referred to without s u f f i c i e n t c l a r i f i c a t i o n . Bellack (1979) reports that "the d i s t i n c t i o n between s k i l l and anxiety and the i r in te r re la t ionsh ip i s not always c lear in the l i t e r a t u r e " (p. 167). While in some studies both var iables are assessed giving them equal weight in treatment-outcome and assessment research (Arkowitz et a l . , 1975; Schwartz & Gottman, 1976), in other cases anxiety i s either not 6 assessed or examined only secondari ly (E i s le r et a l . , 1975; Bellack et a l . , 1976). Due to t h i s inconsistency in the l i t e r a t u r e , statements regarding the re la t ion between social anxiety and social s k i l l must be made with caut ion. Curran (1979) recommends that researchers be wary of the mult ip le e t io log i ca l and maintaining factors of poor social performance in the i r assessment of social s k i l l s . He presents a minimodel of social performance in which social s k i l l s and interference mechanisms (conditioned anxiety or faul ty cogni t ive -eva luat ive processes) are conceptualized as orthogonal constructs . Using a somewhat overs impl i f ied conceptual framework, ind iv idua ls are categorized in one of four ways: adequate social s k i l l s and no interference, adequate social s k i l l s and inter ference, inadequate social s k i l l s and no interference or inadequate s k i l l s and interference. For the purposes of the discussion that fo l lows , i t i s important to underscore the fact that s o c i a l l y unski l led indiv iduals may or may not be s o c i a l l y anxious and s o c i a l l y anxious ind iv idua ls may or may not be s o c i a l l y u n s k i l l e d . Furthermore, i f the s o c i a l l y anxious indiv idual displays s k i l l d e f i c i t s , these d e f i c i t s may or may not be q u a l i t a t i v e l y d i f fe rent than those of the s o c i a l l y unski l led i n d i v i d u a l . For example, the s o c i a l l y anxious indiv idual may know what to say and how to say i t , but may be unable to do so as the result of an interference mechanism while the s o c i a l l y unski l led individual may lack the knowledge to respond appropr iately . The remainder of t h i s introduction w i l l concentrate on the assessment of social anxiety , what researchers have determined thus far and current 7 trends in assessment research to further c l a r i f y t h i s construct . Assessment of Social Anxiety Researchers' e f fo r ts to specify the components of social anxiety have generally focused on behavioural and/or cognitive ind ices . The bulk of t h i s l i t e r a t u r e has placed almost exclusive emphasis on observable response dimensions including speech content, length of conversation, pauses, voice volume, rate of speech, p i t c h , tone, k i n e s i c s , eye contact and fac ia l expression (Bellack & Morrison, 1982). Research f indings indicate that while high and low s o c i a l l y anxious groups have been successful ly d i f fe rent ia ted based on judges' subjective global rat ings of anxiety and s k i l l (Be l lack , 1979; Curran, 1977), s p e c i f i c , objective behavioural measures have yielded inconsistent resu l ts (F ischet t i et al , 1977; Cacioppo et a l . , 1979). In l i g h t of the f a i l u r e of behavioural measures to d i f f e r e n t i a t e contrasted groups of s o c i a l l y anxious i n d i v i d u a l s , researchers have focused a great deal of attention on methodological problems associated with the assessment of social s k i l l s (Be l lack , 1979; Be l lack , 1983; Curran, 1977; Curran, 1979; Wessberg et a l , 1979; Bellack et a l , 1979). Among the issues raised by these and other authors are the r e l i a b i l i t y , v a l i d i t y and u t i l i t y of the predominant assessment strategies which include s e l f - r e p o r t inventor ies , in vivo observation, observation of staged or simulated n a t u r a l i s t i c interact ions and role play t e s t s . On another l e v e l , researchers have become keenly aware of problems regarding the appropriateness of the behavioural units selected for the 8 measurement of social s k i l l s . While some studies have focused on global or molar s k i l l categor ies , others have measured spec i f i c or molecular response components and s t i l l others have assessed both (Hersen 8 Bel lack, 1976; Bel lack, 1979). Given the potential complexity of personal , s i t u a t i o n a l , cu l tural and interact ional var iables involved in the measurement of t h i s construct (Curran & Wessberg, 1981), the importance of a multimethod-multi1evel approach has become a major theme in the assessment l i t e r a t u r e . "Our knowledge of t rad i t iona l psychometric techniques must increase as well as our degree of methodological sophis t icat ion" (Curran, 1979, p .59) . In addit ion to a heightened awareness of experimental design and the use of empir ica l ly val idated assessment s t ra teg ies , the lack of consistent behavioural dif ferences between s o c i a l l y anxious and nonanxious indiv iduals has resulted in the invest igat ion of possible cognit ive dif ferences between these two groups. Smith and Sarason (1975) and O'Banion and Arkowitz (1977) found that s o c i a l l y anxious ind iv iduals recal l more negative information and interpret negative feedback less favorably than do nonanxious ind i v idua ls . Furthermore, they tend to underestimate the i r own performance and expect greater negative evaluations from others, although judges' rat ings of s k i l l may not d i f f e r e n t i a t e high and low s o c i a l l y anxious groups (Clark and Arkowitz, 1975). Goldfried and Sobocinski (1975) and Gormally et al (1971) demonstrated that i r r a t i o n a l be l ie fs were p o s i t i v e l y correlated with s e l f - r e p o r t measures of social anxiety. Consistent with these l a t t e r two s tud ies , Smith et al (1983) provided evidence that s o c i a l l y anxious ind iv idua ls under soc ia l -eva luat i ve threat experience increases in cognit ive a c t i v i t y that r e f l e c t s concern over the evaluation of others . Las t l y , Cacioppo et al (1979) found that s o c i a l l y anxious men generated more negative self -statements while ant ic ipat ing a conversation with an unfamil iar woman than their nonanxious counterparts. While the resul ts of these studies do not provide d e f i n i t i v e evidence regarding the casual role of cognit ive responses in social anxiety , they strongly suggest that social anxiety has an important cognit ive component. Based on th is research Galassi and Galassi (1981) conclude that there i s more evidence of a cognit ive d e f i c i t associated with social anxiety than a behavioural d e f i c i t . A t h i r d outgrowth of the f a i l u r e to specify r e l i a b l e behavioural indices of soc ia l anxiety has been a renewed interest by c l i n i c a l researchers in the dynamic nature of social i n te rac t ion . It has been suggested that simple frequency measures are not adequate to detect behavioural d i f ferences in social s k i l l (Arkowitz et a l . 1975). According to F ischet t i et al (1977), frequency counts ignore the reciprocal nature of a social i n te rac t ion , espec ia l ly the synchronization of responses to a partner's behavioural cues. Given t h i s increased awareness in the interact ional characterst ics of social i n t e r a c t i o n , a growing number of researchers are focusing on the importance of conducting more f ine-grained analyses of response sequences in order to i so la te the fundamental indices of social s k i l l as well as the behavioural markers of social anxiety (Scott & Edel s t e i n , 1981; Glasgow & Arkowitz, 1975). 10 By way of introduction to a discussion of the relevant c l i n i c a l research, a few comments on the nature of social interact ion appear warranted. Whi1e c l i n i c a l researchers have pr imari ly focused upon the intrapersonal d e f i c i t s associated with social anxiety, social psychologists have brought to l i g h t the importance of interpersonal and contextual var iables associated with e f fec t i ve social performance. In a comprehensive review of the experimental work on social performance, Argyle and Kendon (1967) present evidence to suggest that the patterning of action in t ime, para l ingu is t i c features , visual o r ien ta t ion , bodi ly movement, fac ia l expression and looking patterns are a l l important dimensions of a social encounter. These dimensions are said to re f lec t the complex and ever-changing nature of an i n t e r a c t i o n . The terms "working consensus" (Goffman, 1959), " in teract ional synchrony" (Kendon, 1970), " t r a f f i c rules" (Goffman, 1955) and " rhetor ica l s e n s i t i v i t y " (Hart & Burks, 1972) stress the importance of assessing the interact ional var iables associated with e f fec t i ve social performance. The common theme underlying these terms i s that of recognizing and responding to cues emitted by the other person in order to maintain a social i n t e r a c t i o n . According to Argyle and Kendon's (1967) sensorimotor model of social s k i l l , the s o c i a l l y competent indiv idual i s one who continuously monitors the immediate s i tuat ion and modifies his or her performance in l i g h t of continuous external feedback from the other person and internal c r i t e r i a such as his or her desired goals . In order to further understand the nature of social s k i l l and the 11 re la t ion between social s k i l l and social anxiety, c l i n i c a l researchers have broadened the scope of the i r invest igat ions to incorporate the interact ional perspective outl ined above. One of the major proponents of t h i s expansion is Trower (1979, 1980, 1982). According to t h i s author, social incompetence can be accounted for in one of two ways. It may be the resul t of s k i l l component d e f i c i t s which include normative social behaviours (eye contact , gestures, greet ings, part ings , segments of d iscourse) . Social incompetence may also be due to s k i l l process d e f i c i t s . Within th i s category, Trower (1980) includes both cognit ive and perceptual d e f i c i t s which may occur during the course of an i n t e r a c t i o n . Negative or u n r e a l i s t i c expectations, negative a t t r i b u t i o n s , f a i l u r e to monitor one's own behaviours or social cues in the immediate s i tuat ion and f a i l u r e to synchronize one's performance with the other person are a l l c l a s s i f i e d as s k i l l process d e f i c i t s . Morrison and Bellack (1981) agree with Trower's (1980) expansion of the concept of social s k i l l to encompass interact ional or process components. They propose that social perception s k i l l , the a b i l i t y to accurately read the social environment, i s an important dimension of social s k i l l . "Regardless of the magnitude of response s k i l l , the indiv idual cannot perform e f f e c t i v e l y i f he/she does not adequately receive and process the relevant interpersonal s t i m u l i " (p. 70) . Morrison and Bellack specify f i ve components of social perception in social s k i l l : (a) l i s t e n i n g , or attending to the interpersonal partner, (b) getting c l a r i f i c a t i o n , or the a b i l i t y to both perceive confusion and to appropriately seek e luc idat ion from the interpersonal partner, 12 (c) relevance, or insuring that a response is related to the conversation as a whole, as well as to the immediately preceding Communication, (d) t iming , or the performance of responses at appropriate latency, and (e) perception of emotion. While these components are not ident ical to those suggested e a r l i e r , both Morrison and Bellack (1981) and Trower (1980) acknowledge the importance of cognit ive and perceptual processes in the i r conceptualizations of social s k i l l . In both cases, the importance of rece iv ing , interpret ing and responding to social cues emitted by the interpersonal partner i s st ressed. As such, these authors' recommendations that c l i n i c a l invest igat ions into the nature of social s k i l l be expanded to incorporate interact ional var iables represent the ass imi lat ion of a large body of social psychological l i t e r a t u r e which has i l luminated the f luctuat ing nature of social i n te rac t ions . In summary, the current trends in the assessment of social s k i l l and the re la t ion between social s k i l l and social anxiety have developed as resu l t of the f a i l u r e of spec i f i c behavioural measures to r e l i a b l y d i f f e r e n t i a t e contrasted groups. Ef for ts to characterize these constructs have most recently focused on the interact ional or process var iables associated with social i n t e r a c t i o n . Social psychologists have emphasized the importance of monitoring and responding to social cues in order to maintain a social encounter. With th i s in mind, c l i n i c a l researchers are beginning to invest igate the v a l i d i t y of an interact ional approach in the assessment of social s k i l l and social anxiety in which s e n s i t i v i t y to social cues plays a central r o l e . 13 Relevant Research F i schet t i et al (1977) examined the timing and placement of social responses in high and low heterosexual -soc ia l ly anxious undergraduate males. Subjects were selected on the basis of behavioural ratings of the i r social anxiety and social s k i l l as well as a sel f - repor t measure of heterosexual-social anxiety. Two s o c i a l l y competent groups and one s o c i a l l y incompetent group, each made up of twelve subjects , were used in the study. Subjects viewed a 10 minute prerecorded tape of a female speaker discussing personal and general aspects of her l i f e and were asked to press a switch whenever they f e l t i t was a good time to respond with a gesture and/or a vocal response. One of the s o c i a l l y competent groups served as the c r i t e r i o n for determining the in terva ls considered to be appropriate places to respond. Findings supported the or ig ina l hypothesis: While the two groups of s o c i a l l y competent indivdual s did not d i f f e r from the group of s o c i a l l y incompetent indiv iduals in the i r frequency of responses, there were s i g n i f i c a n t dif ferences in agreement over the placement or timing of the i r responses. Soc ia l l y competent subjects tended to c luster the i r responses systematical ly whereas s o c i a l l y incompetent subjects tended to respond in a more random fashion. The authors concluded: "The data from th is study point out the importance of expanding invest igat ions into spec i f i c nature of social s k i l l beyond the simple emission of behaviour to accurate placement of those behaviours and the a b i l i t y to discr iminate cues that may guide response synchronization" (p. 193). 14 Peterson et al (1981) used the same methodology to examine the timing and placement of social responses in heterosexual -soc ia l ly anxious women and obtained s i m i l a r , a lbe i t weaker, f indings than those of F ischet t i et al (1977). The attenuated leve ls of s ignif icance were explained in terms of less contrasted groups and a lack of research regarding the social s k i l l s of women which necessitated the use of subject select ion procedures pr imari ly designed for and validated on men. The authors concluded, however, that the i r f ind ings , in combination with those of F ischett i et al (1977), indicate the importance of response synchronization in the assessment of s o c i a l l y imcompetent populations. Trower (1980) investigated the components and processes of s k i l l e d and unski l led social behaviour in s ix ty patients referred for behaviour therapy or social s k i l l s t ra in ing (see previous section for discussion of s k i l l component d e f i c i t s and s k i l l process d e f i c i t s ) . Patients were asked to par t ic ipate in a laboratory-type social interact ion test with one male and one female confederate. The female confederate systematical ly varied her speaking and attending behaviour during the f i r s t and second phases of the i n t e r a c t i o n , each phase l a s t i n g 4 minutes. The male role partner remained s i len t unt i l the th i rd phase at which time he displayed verbal ly aggressive and argumentative behaviours toward the pat ient . Dependent measures included speech, looking at other , s m i l i n g , commmunicative gestures and posture s h i f t s . S k i l l e d patients were found to speak, look, smi le , gesture and change the i r posture more than unski l led pat ients . Furthermore, the s k i l l e d group 15 showed more v a r i a b i l i t y in behaviour in response to s i tuat ional changes. Trower concluded that unski l led patients are def ic ient not only in absolute leve ls of behaviour but in the i r pattern of behaviour as w e l l . Although these f indings must be interpreted in the l i g h t of a se lect psychiatr ic population, Trower addressed two issues in th is study that are of importance to researchers in the area. F i r s t , an attempt was made to empir ica l ly determine some of the essential elements of s o c i a l l y s k i l l e d behaviour. Second, the nature of social incompetence was investigated on a process or interact ional level as well as in terms of s p e c i f i c behavioural components. Unlike the F ischett i et al (1977) and Peterson et al (1981) s tud ies , Trower used a ro le -p lay s i tuat ion to assess his s k i l l e d and unski l led groups, thus employing a measure more representative of social i n t e r a c t i o n . Steffen and Reckman (1978) examined the se lect ive perception and interpretat ion of interpersonal cues in 20 s o c i a l l y anxious and 20 nonanxious male undergraduates. They suggested that s o c i a l l y anxious ind iv iduals may be se lec t i ve l y attending to interpersonal cues and/or d i f f e r e n t i a l l y interpret ing such cues in social s i t u a t i o n s . Subjects were introduced to a female confederate and informed that the i r task was to get to know each other . Their interact ion was video-taped and lasted for 6 minutes. In counterbalanced order across subjects , the confederate behaved pos i t i ve l y for 3 minutes and negatively for 3 minutes. The dependent measures included: (a) a written descr ipt ion by the subject stat ing how his partner responded to him during the i r conversation and (b) a series of 7 - point scales on which the subject 16 rated the frequency with which the confederate smiled, looked at him, asked questions and i n i t i a t e d conversation. Three additional scales asked the subject to rate the extent to which the confederate displayed interest in him, displayed warmth toward him and how generally posi t ive she had responded toward him. Subjects also rated the degree to which they f e l t she would l i k e to work with them in another experiment. D i f fe rent ia l perception of social behaviour was not demonstrated. The only s ign i f i cant f inding was in subjects' response to the item "I bel ieve she would very much l i k e working with me in another experiment." Low anxious subjects agreed with t h i s statement more so than the high anxious subjects . The authors concluded that differences between high and low s o c i a l l y anxious subjects' performances may be more a factor of interpretat ion than the i r perception of social events. Steffen and Reckman (1978) re l i ed solely on se l f - repor t data to assess the d i f f e r e n t i a l perceptions of the i r subjects . The reasons for doing so are p a r t i c u l a r l y unclear given that they had video-taped the subjects' in teract ions with the confererate. Behavioural observations may have revealed di f ferences between groups in the i r responses to changes in the confederate's behaviour. Add i t iona l l y , subjects' level of comfort or discomfort was not assessed. It was not c l e a r l y establ ished that subjects in the high s o c i a l l y anxious groups were more anxious while interact ing with the confederate than low anxious subjects . Another l i m i t a t i o n of Steffen and Reckman's study was the absence of descr ipt ive information regarding the posit ive and negative behaviours of the confederate. As a r e s u l t , the actual nature of the 17 social stimulus remains unknown. In l i g h t of these l i m i t a t i o n s , i t i s d i f f i c u l t to evaluate the s igni f icance of Steffen and Reckman's f ind ings . The studies discussed thus far suggest that interact ional var iables are a potent ia l l y r e l i a b l e index for d iscr iminat ing between s o c i a l l y competent and incompetent groups. When a social s i tuat ion is not held constant ( i . e . the partner's behaviour changes over the course of the i n t e r a c t i o n ) , s o c i a l l y unski l led ind iv iduals can be dist inguished from s o c i a l l y s k i l l e d ind iv iduals on the basis of overt behaviour (Trower, 1980). Furthermore, social l y anxious and nonanxious groups have been shown to d i f f e r in the i r interpersonal sty les of responding to others in a social context; nonanxious ind iv idua ls display greater synchronization in responding to others than do anxious indiv iduals whose responses appear to be random and unrelated to social cues emitted by an interpersonal partner (F ischet t i et a l , 1977; Peterson et a l , 1981). Taken as a whole, these f indings indicate that when presented with a standarized sequence of social cues, s o c i a l l y competent and incompetent groups can be dist inguished on the basis of response synchronization and behavioural v a r i a b i l i t y , measures which re f lec t the dynamic nature of social i n t e r a c t i o n . Nonverbal Communication and Social Anxiety The re la t ion between s e n s i t i v i t y to social cues and social anxiety has only recently been invest igated. As such the robustness of the f indings presented above has not yet been determined. Assessment and 18 procedural techniques are far from sophis t icated . The choice of appropriate dependent measures and the construction of a credible and representative social context in which to assess process or in teract ional var iables such as s e n s i t i v i t y to social cues require further empirical study. F i n a l l y , l i t t l e i s known about the part icu lar social cues to which s o c i a l l y nonanxious indiv iduals attend when interact ing with someone. The research discussed above suggests that s o c i a l l y nonanxious ind iv iduals are more sensit ive to the i r interpersonal partner 's behaviour than anxious i n d i v i d u a l s . However, the basis of the i r s e n s i t i v i t y i s not known. Given the recent impetus of research in t h i s area , the s ign i f icance of nonverbal communication in re la t ion to s e n s i t i v i t y to social cues would appear to be an appropriate s tar t ing point at which to explore some of these issues in greater d e t a i l . Nonverbal behaviour plays an important role in human communication. The s igni f icance of nonverbal communication i s ref lected in the numerous functions i t serves. For example, nonverbal behaviour compliments verbal content by giving the l i s t e n e r some idea of the af fect ive state of the speaker. When presented with messages in which nonverbal and verbal channels are contradictory , l i s teners tend to place more credence on nonverbal behaviours (Shapiro, 1968). Both tone of voice (Mehrabian & Wiener, 1967) and fac ia l expression (Mehrabian & F e r r i s , 1967) have been implicated as of greater importance to observers than verbal content. Nonverbal behaviour i s also a potent re in forcer . Eye contact , hand gestures and head-nods can, when used appropr iately , act as 19 re inforcers during normal interact ions to increase the frequency of selected behaviours emitted by the other partner (Hargie, Saunders & Dickson, 1981). An equally important function of nonverbal behaviour i s the regulation of the flow of communication between speaker and l i s t e n e r . The pattern of gaze i s p a r t i c u l a r l y important in the synchronizing of a social interact ion (Argyle, L a l l j e e & Cook, 1968). F i n a l l y , nonverbal behaviour i s an important source of feedback. In order to sustain communication, the part ic ipants must have a knowledge of how the i r messages are being received. Goffman (c i ted in Argyle & Kendon, 1967) described th is process as the "working concensus" of an encounter. Nonverbal behaviour provides the indiv idual with cues regarding the responses of the other person. Given the s igni f icance of the nonverbal channel in communication, a lack of s e n s i t i v i t y to nonverbal cues could lead to serious impairments in social s i t u a t i o n s . Rosenthal et al (1979) developed the P r o f i l e of Nonverbal Sens i t i v i t y Test (PONS) to assess indiv idual differences in the a b i l i t y to decode cues in a var iety of channels of nonverbal communication. The PONS i s composed of 220 f i lm segments of 20 short scenes portrayed by a young woman. The subject 's task i s to view each segment and c i r c l e the label (one of two al ternat ives provided) that cor rec t l y describes the scene enacted in the segment. Researchers found that subjects scoring higher on the PONS also scored as better adjusted, more interpersonal ly democratic and encouraging, less dogmatic, more extraverted, more popular and more interpersonal l y sensi t ive as judged by acquaintances, c l i e n t s , spouses or supervisors. These f indings are 20 i n t e r e s t i n g , however since social interact ion involves the sending as well as receiving of messages, the PONS provides only l imi ted information regarding d i f f e r e n t i a l s e n s i t i v i t y to nonverbal cues in s i tuat ions where the recipient i s a passive observer. In a study conducted by Christensen et al (1980), s e n s i t i v i t y to nonverbal cues as a function of social competence was examined. Drawing from the l i t e r a t u r e concerning test anxiety and task performance, the authors hypothesized that s o c i a l l y unski l led persons would report greater self -preoccupation when compared to s o c i a l l y s k i l l e d persons. Furthermore, drawing from the same body of research, they hypothesized that anxiety , arousal and sel f - focusing tendencies in social s i tuat ions would resul t in the reduction of an i n d i v i d u a l ' s a b i l i t y to perceive cues displayed by others , in par t icu lar nonverbal cues which are general ly subtle in nature. Before discussing thei r research, an elaboration of the l i t e r a t u r e concerning the ef fects of anxiety on task performance w i l l c l a r i f y the or ig in of the i r hypotheses. Test anxiety studies have supported Easterbrook's a r o u s a l - c u e - u t i l i z a t i o n theory (in Sarason, 1980) which states in par t , that anxiety reduces the range of task- re levant cues u t i l i z e d in task performance. Wine (1971) has elaborated upon t h i s idea by proposing that sel f -centered responses in ter fere with attention to task- re levant st imul i necessary for good performance. The sel f -preoccupied d i rec t ion of attent ion in highly test -anxious persons in comparison to low test -anxious ind iv iduals i s supported by a number of studies (see Sarason, 1980 for a review). Assuming a state of anxiety 21 a rousa l , Christensen et al (1980) predicted that " s o c i a l l y incompetent" indiv iduals would display increased self -preoccupation and a subsequent narrowing of cue u t i l i z a t i o n s imi lar to that of test -anxious subjects . Fifteen high-competence female undergraduates and f i f teen low-competence women were asked to engage in a structured interview with another student (a female confederate). Subjects were given a standardized set of questions to be used during the interview. Ten questions were provided for each of the three topics to be covered. Subjects were instructed that i f the i r partner appeared uncomfortable with a par t icu lar set of questions, they should proceed to the next t o p i c . The experimental manipulation involved the confederate display ing a ser ies of nonverbal cues indicat ing tension during the second set of questions. During the f i r s t and th i rd set of questions, the confederate portrayed a relaxed demeanor. Following the interv iew, subjects completed se l f - repor t scales to measure degree of anxiety , embarrassment, self -consciousness and how much they f e l t the confederate l iked them. Global rat ings by the confederate and experimenter showed that the high and low competence groups did not d i f f e r in terms of the i r behaviour during the interv iew, however the high competence group terminated the second set of questions s i g n i f i c a n t l y e a r l i e r than the low competence group. None of the se l f - repor t measures yielded s i g n i f i c a n t d i f ferences between the two groups. Postexperimental interviews indicated that both groups perceived tension in the confederate at the same point during the second question set . An 22 analys is of the reasons given by the low-com pete nee group for not discontinuing indicated that subjects either f e l t the nature of the questions was not anxiety-provoking or that the tension was at t r ibutab le to the experimental s i tuat ion as opposed to the interview t o p i c . The researchers primary hypothesis was based on the assumption that s o c i a l l y incompetent indiv iduals would experience greater anxiety and sel f - focuss ing tendencies in the experimental s i t u a t i o n . Se l f - repor t measures indicated that the low competence group was not s i g n i f i c a n t l y d i f f e r e n t from the high competence group on either of these two measures. This invest igat ion was not, therefore, an adequate test of the i r hypothesis. As suggested by Christensen et al (1980), the h igh ly -structured format of the interview s i tuat ion may have been the reason for no d i f ferences in anxiety between the two groups. If subjects had been required to take more respons ib i l i t y during the interview with the confederate, group dif ferences may have resu l ted . Summary While c l i n i c a l research has led to the development of treatment programs which have e f f e c t i v e l y reduced social anxiety, one of the major shortcomings in t h i s area has been the lack of conceptual in tegra t ion . Despite the quantity of data that has been generated, terms such as socia l incompetence, social s k i l l and social anxiety have been used interchangeably in the l i t e r a t u r e . Insuf f i c ient attent ion has been given to def in ing these constructs and the nature of the i r r e l a t i o n . Consequently, c l i n i c a l researchers are beset with the ominous task of 23 t reat ing a disorder that has not been adequately def ined . In the current study, the term social incompetence i s used to refer to social problems in general without spec i f ic reference to e t io log ica l or maintaining fac to rs . The term social s k i l l i s used to describe the components of e f f e c t i v e interpersonal behaviour. Social anxiety i s defined as anxiety in social s i tuat ions . It i s suggested that s o c i a l l y anxious and s o c i a l l y unski l led indiv iduals are not equivalent c l i n i c a l populations. The focus of the present study i s on the assessment of social anxiety . More s p e c i f i c a l l y , t h i s study examines the nature of the re la t ion between a part icu lar social s k i l l and social anxiety. For many years social psychologists have acknowledged and studied the re la t ion between social cues and social i n te rac t ion . A handful of c l i n i c a l studies have shown that s o c i a l l y anxious ind iv iduals lack the a b i l i t y to encode and decode social cues. The importance of nonverbal behaviour in human communication suggests that an i n a b i l i t y to accurately monitor and respond to such cues can inter fere with the maintenance of a social i n t e r a c t i o n . The current study invest igates the re la t ion between s e n s i t i v i t y to nonverbal social cues and social anxiety in order to further c l a r i f y the nature of th i s prevalent c l i n i c a l problem. Overview and Experimental Hypothesis The research to be presented examines the re la t ion between nonverbal s e n s i t i v i t y and social anxiety. Borrowing from the Trower (1980) methodology, a social interact ion task involving a confederate 24 displaying changes in the i r behaviour was employed in order to avoid the d i f f i c u l t i e s encountered by Christensen et al (1980). It was expected that a less structured interact ion task would be s u f f i c i e n t l y involv ing so that highly anxious subjects would experience s ign i f i cant discomfort . Unlike the studies reviewed e a r l i e r , the invest igat ion included verbal as well as nonverbal dependent measures. Researchers of nonverbal communication state that verbal and nonverbal components are so c lose ly intertwined that they cannot be studied in i so la t ion of one another. It i s possible that ind iv idua ls may display s e n s i t i v i t y to nonverbal cues emitted by the i r interpersonal partner using verbal and/or nonverbal channels. The study was designed to test the hypothesis that s o c i a l l y anxious ind iv iduals are less sensi t ive to changes in the nonverbal behaviour of the i r interpersonal partner than s o c i a l l y nonanxious i n d i v i d u a l s . Sens i t i v i t y was defined as the a b i l i t y to monitor and respond to changes in the nonverbal behaviour of an in teractant . It was expected that s o c i a l l y anxious indiv iduals would display less v a r i a b i l i t y in the i r behaviour (verbal/nonverbal) across conditions in which the i r partner's behaviour changed than would s o c i a l l y nonanxious ind i v idua ls . This hypothesis was based on data indicat ing that s o c i a l l y anxious ind iv iduals engage in negative se l f -eva luat ions (Cacioppo et a l , 1979) and fear negative evaluations from others (Smith et a l , 1983; Watson & Fr iend, 1969). There i s considerable evidence in the test anxiety l i t e r a t u r e to support the re la t ion between self - focused attention or t a s k - i r r e l e v a n t cognit ions and poor task performance (Wine, 1971). 25 Researchers have integrated th is large body of l i t e r a t u r e within a d i s t r i b u t i o n - o f - a t t e n t i o n model which states that self - focused attention inter feres with attent ion paid to task requirements. The range of cues u t i l i z e d in task performance i s thereby reduced result ing in impaired performance (Sarason, 1980). In the case of social anxiety, ind iv iduals are attending to or evaluating the i r social performance as opposed to focusing on the i r interpersonal partner. Consequently, the i r a b i l i t y to perceive and respond to cues displayed by others i s reduced. Crozier (1979) refers to th is conceptualization of social anxiety as "anxious se l f -preoccupat ion" . In a comprehensive theoret ical paper, Schlenker and Leary (1982) agree with Croz ier 's (1979) proposition that sel f -attention i s an important component of social anxiety. Empirical support for th i s model stems largely from corre lat ional studies which re late task - i r re levant cognitions to social anxiety (Cacioppo et al , 1979) and treatment-outcome studies which indicate that modifying these cognitions can result in the reduction of social anxiety (Kanter & G o l d f r i e d , 1979). While the goal of t h i s study i s neither to prove nor disprove a model of social anxiety , confirmation of the hypothesis could lend support to Croz ier 's (1979) notion of anxious sel f -preoccupation. 26 Method Subjects Twenty-four s o c i a l l y anxious women (mean age 23.96) and twenty-four s o c i a l l y nonanxious women (mean age 23.25) part ic ipated in the study. Subjects were s o l i c i t e d on a voluntary basis from winter , spring and summer undergraduate classes at the University of B r i t i s h Columbia. Subjects were screened for inc lus ion in the study using the Social Avoidance and Distress Scale (SAD; see Appendix A ) , a well-known t r a i t measure of social anxiety developed by Watson & Friend (1969). The SAD i s a 28-item s e l f - r e p o r t questionnaire which measures subjective d i s t ress and avoidance in social s i t u a t i o n s . Social avoidance is defined as avoiding being w i th , ta lk ing to or escaping from others for any reason. Social d is t ress i s defined as the reported experience of a negative emotion, such as being upset, d i s t ressed , tense or anxious in social i n te rac t ions . Item content var ies in terms of the sex of the person with whom the subject i s i n te rac t ing , his/her status re la t i ve to the subject, whether he/she i s a f r iend or stranger and public and pr ivate condi t ions . K-R 20 homogeneity indices of .94 have been obtained and one-month t e s t - r e t e s t r e l i a b i l i t i e s reported in the l i t e r a t u r e are .68 and .79 (Bellack & Hersen, 1979). The mean score for females i s 8.23 and the standard deviation i s 8 . 0 1 . Subjects at leas t one half standard deviat ion above or below the mean (greater than 13 or l ess than 5) were included in the study. The s o c i a l l y anxious group had a mean score of 17.25 and a standard deviation of 3.05 while the s o c i a l l y nonanxious group had a mean score of 2.50 and a standard 27 deviat ion of 1.44. According to the l i t e r a t u r e , there is a general concensus that depressed persons are less s o c i a l l y s k i l l f u l than nondepressed persons (Youngren & Lewinsohn, 1980). Although the bulk of t h i s research has focused on the interpersonal behaviour of depressed psychiatr ic pa t ien ts , there i s some evidence to suggest that social s k i l l d e f i c i t s are also present in depressed col lege students (Gotl ib & Robinson, 1982; Jacobson & Anderson, 1982). In order to control for depression as an a l ternat ive interpretat ion of the data , the Beck Depression Inventory (BDI; see Appendix B) was employed to screen out moderately and severely depressed subjects . The BDI i s a c l i n i c a l l y derived se l f - repor t inventory of state depression which consists of 21 items covering a f f e c t i v e , cogn i t i ve , motivational and physiological areas of depressive symptomatology (Beck, 1967). Bumberry et al (1978) determined that the BDI i s a va l id instrument for the measurement of depression in a un ivers i ty populat ion. Subjects scoring 16 or higher, representing moderately and severely depressed i n d i v i d u a l s , were not used in the study. The mean score and standard deviat ion of the s o c i a l l y anxious group were 6.88 and 3 .92, respect ive ly . The s o c i a l l y nonanxious group had a mean score of 2.38 and a standard deviation of 2 .75. Apparatus and Setting The experimental room consisted of three c h a i r s , a small desk, and a book she l f . A videotape system that included a Sony video camera, a Sony s o l i d - s t a t e video recorder, and a Sony t e l e v i s i o n was used to 28 obtain records of the in te ract ions . The video camera was positioned 12 feet away from the subject and confederate who sat facing one another with chairs turned s l i g h t l y toward the camera. Procedure The SAD and BDI were d is t r ibuted in undergraduate c lasses . Students who expressed interest in par t ic ipat ing were asked to complete the questionnaires and informed that they would be contacted by phone to arrange a convenient time to take part in the study. - Subjects were met ind i v idua l l y by the experimenter (a fourth year female undergraduate) who was blind to the i r condi t ion . Due to the unexpectedly lengthy period of time required to run subjects , the pr inc ipa l invest igator assumed the role of experimenter for 17 subjects co l lected during the summer months. Arrangements were made to ensure that the second experimenter was bl ind to the condition of these subjects . Upon ar r i va l at the laboratory , subjects were asked to take a seat and informed that the second subject (a confederate) had yet to a r r i v e . Several minutes l a t e r , the confederate arr ived and was seated next to the subject . The experimenter then proceeded with the fol lowing in t roduct ion : Social s k i l l s are very important in getting to know people and developing f r iendships . In th is pro ject , we want to assess your social s k i l l s and look at the d i f fe rent styles people use in gett ing to know someone new. There are two parts to t h i s study. F i r s t l y , I w i l l ask you to get to know each other . There i s no r ight or wrong way to act in t h i s s i t u a t i o n . Just do whatever you 29 would normally do when you meet someone for the f i r s t t ime. A video camera w i l l be used during th is part of the study. A l l video tapes w i l l be kept conf ident ia l and w i l l be seen only by ind iv iduals working on the pro ject . Following your conversat ion, I w i l l ask you to complete several questionnaires independently. After completing the forms, I w i l l answer any questions you may have about the study. The subject and confederate were then given consent forms which they were asked to read and s ign . Following t h i s , the experimenter turned on the video camera, instructed the subject to begin the conversation and l e f t the room. The experimenter reentered the lab at the end of four minutes and escorted the confederate to another room where she was supposedly completing the remaining questionnaires. After f i l l i n g out the quest ionnaires, the subject was informally interviewed to check for any suspicions regarding the experimental procedures and then f u l l y debr iefed. Confederate's Behaviour Two female undergraduates were recruited to play the role of confederate. Each student interacted with 24 subjects and remained bl ind to the condit ion of each subject and the spec i f i c nature of the experimental hypothesis. The confederate role was comprised of two phases. In the relaxed phase, the confederate's behaviour included head nodding, appropriate eye contact , expressive hand gestures, smil ing and a comfortable body posture. In the anxious phase, the confederate's behaviour consisted of nonsignaling hand gestures (f ingers tw i tch ing , c l ing ing hands, hands grabbing chair or braced unnatura l ly ) , poor eye 30 contact ( infrequent and of short durat ion) , movements of the mouth (b i t ing l i p s , mouth closed t ight l y ) and posture s h i f t s . These nonverbal cues were chosen on the basis of work done by Waxer (1979) who i d e n t i f i e d a d i s t i n c t nonverbal cue c luster associated with anxiety . During both phases the confederate was minimally responsive to subjects' questions and did not i n i t i a t e conversation. Each phase was two minutes in durat ion. The experimenter, who was outside the room during the i n t e r a c t i o n , signaled the confederate at the end of two minutes by coughing b r i e f l y . Subjects were randomly assigned to relaxed-anxious and anxious-relaxed condi t ions . A p i l o t study was conducted in order to ensure that the two phases of the confederate's role were accurately discriminated by naive observers. Throughout the data c o l l e c t i o n per iod, weekly t ra in ing sessions were held to minimize var ia t ions in the confederate's performance and maintain a c lear d i s t i n c t i o n between phases. Upon completion of the data c o l l e c t i o n , observer ratings were co l lected by two undergraduate judges using a 7-point scale to determine i f the selected nonverbal cues successful ly conveyed relaxat ion and anxiety (1 = anxious, tense; 7 = re laxed, at ease) . Measures Videotapes of the subjects' in teract ions with the confederate were coded independently by two trained undergraduate judges who were b l ind to the spec i f i c nature of the study. The fol lowing measures were obtained: 31 Nonverbal-nonvocal behaviours, (a) observation - the f ract ion of each two minute interval when the subject looked at the confederate, whether or not i t could be ascertained that the subject was making d i rec t eye contact with the confederate; (b) posturing - the number of gross body s h i f t s in each interval ( i . e . leaning forward or backward against the c h a i r , arms c ross ing , legs c ross ing , sh i f t ing posit ion in chair in some way); (c) communicative gestures - the number of expressive hand gestures (hand movements which emphasize verbal content) ; (d) head nods - due to the d i f f i c u l t y of defining head nods in terms of d iscreet behavioural u n i t s , they were scored on the basis of occurrence/nonoccurrence over 5-second i n t e r v a l s . Nonverbal-vocal behaviours, (a) number of pauses in each two minute interval (periods of s i lence las t ing 5 seconds or more); (b) time spent t a l k i n g . Verbal-vocal behaviours, (a) s i tuat ional statements - statements about the experimental s i t u a t i o n ; (b) s e l f - d i s c l o s i v e statements -statements about s e l f ; (c) questions about the confederate's comfort/discomfort; (d) questions (excluding those regarding the confederate's comfort/discomfort). Global ra t ings . Judges rated subjects on global measures of s k i l l 32 (1 = not at a l l s k i l l f u l , handles s i tuat ion poorly; 7 = very s k i l l f u l , handles s i tuat ion e f fec t i ve ly ) and anxiety (1 = very anxious, tense; 7 = very comfortable, relaxed) based on the ent i re 4 minute i n t e r a c t i o n . S e l f - r e p o r t measures. After interact ing with the confederate, subjects rated the i r own comfort/discomfort and that of the confederate using 7-point sca les . Subjects were then asked to describe in the i r own words how the i r partner responded to them during thei r conversation. The resu l t ing essays were read by two undergraduate judges (bl ind to the subject 's condit ion and the spec i f i c nature of the experimental hypothesis) who bracketted of f a l l references to spec i f i c overt behaviours, quotations of the confederate's statements, general descr ipt ions of the confederate's interpersonal s t y l e , perceptions of change in the confederate's behaviour and statements about s e l f . Subjects' scores were calculated as the total number of statements that f e l l under each category with the exception of perception of change in the confederate's behaviour which was coded as e i ther yes (the subject perceived a change) or no (the subject did not perceive a change). This measure was adopted from the work of Steffen and Reckman (1978) to assess subjects' attent ion to the confederate's behaviour during thei r i n t e r a c t i o n . The p i l o t study revealed that in the majority of cases subjects were reporting the i r responses to the confederate instead of describing the confederate's behaviour. Consequently, a fourth measure was added 33 which asked subjects whether or not they observed a change in the i r partner 's behaviour and i f so, to describe what they observed. Judges coded subjects' responses as e i ther yes (the subject perceived a change) or no (the subject did not perceive a change). 1 ^Before data c o l l e c t i o n was completed, two videotape casettes were s t o l e n . As a r e s u l t , three dependent measures could not be coded. These were s i tuat ional statements, s e l f - d i s c l o s i v e statements and quest ions. Since th is was not a r e p l i c a t i o n study and these measures were not c r i t i c a l in test ing the central hypothesis of the study, i t was decided to proceed with data analysis without th i s information. 34 Results After a l l dependent measures were coded, i t was observed that two var iables demonstrated almost no var iat ion across subjects . Quotations by the confederate, a measure used to categorize essay responses, and subjects' questions regarding the confederate's comfort during the i r in teract ion occured in only 3 and 9 cases, respect ive ly . Given the marked absence of information in these measures, they were not included in the subsequent data analyses and d iscuss ion . In te r - ra ter R e l i a b i l i t e s Table 1 l i s t s the inter rater r e l i a b i l i t e s for the essay categories and videotape rat ings . Pearson Product-Moment correlat ions were used to calculate agreement between judges. The cor re la t ion coef f i c ients for posturing and descr ipt ive statements about the confederate low indicated only moderate r e l i a b i l i t y although a l l ot other r e l i a b i l i t y c o e f f i c i e n t s are high. Although not indicated by the overal l r e l i a b i l i t y index, judges' level of agreement increased substant ia l l y over the course of coding statements about the confederate. In the case of posturing however, the judges reported some d i f f i c u l t y in d is t inguish ing gross body sh i f t s from f iner body movements. Given these moderate c o r r e l a t i o n s , s i g n i f i c a n t relat ionships based on posturing and statements about the confederate are considered suggestive and require further empirical v e r i f i c a t i o n . 35 Table 1. Interrater R e l i a b i l i t i e s Essay Categories Correlations (r) n=48 References to Speci f ic Overt Behaviours .95 Descript ive Statements about Confederate .70 Statements about Self .90 Perception of Change .91 Videotape Ratings Correlat ions (r) n=48 Global s k i l l .83 Global anxiety .81 Observation .91 Posturing .69 Gestures .96 Nods .91 Time spent ta ik ing .96 Pauses (_> seconds) .89 A l l cor re lat ions have j) < » u u l 36 Confederates' Behaviour In te r - ra te r r e l i a b i l i t y for judges' ratings of the two confederates across relaxed and anxious in te rva ls was .85 using the Pearson Product-Moment co r re la t ion c o e f f i c i e n t . The mean rat ing of re laxat ion for both confederates was 6 . 7 1 . In the anxious i n t e r v a l , the confederates' mean rat ings were 2.33 and 1.46. A 2(Intervals) X 2(Confederates) analys is of variance (ANOVA) y ie lded a s i g n i f i c a n t main e f f e c t for i n t e r v a l s , Hi,92) = 2203.11, jp_ < . 0 1 , for confederates, Hi ,92) = 18.21, £ < .01 and for t h e i r i n t e r a c t i o n , Hi ,92) = 18.21, £ < . 0 1 . The main e f f e c t for confederates and the interva l by confederate in teract ion were due to the s i g n i f i c a n t difference between confederates in the anxious i n t e r v a l . A 2(Confederates) X 2(SAD) ANOVA on subjects' perceptions of the confederates' degree of comfort or discomfort during the i r in teract ion was car r ied out in order to determine whether or not t h i s s t a t i s t i c a l d i f ference between confederates was perceived by subjects . There was no main e f f e c t for SAD, f_(l,44) = 2 .47 , ns, for confederates Hi ,44) = 0.07 ns, or t h e i r in teract ion H i ,44) = 0 .07 , ns. The di f ference between the confederates i n the anxious interval does not appear to have been a c r i t i c a l factor in subjects' perceptions of the i r i n te rac t ions . Based on t h i s a n a l y s i s , the di f ference between confederates i n the anxious in terva l as perceived by judges' ratings was not incorporated into the data analys is presented below. 37 Subjects' Global S k i l l and Anxiety Judges' mean global rat ings of social s k i l l were 4.08 for the s o c i a l l y anxious group and 5.00 for the soc ia l l y nonanxious group. A two- ta i led t - t e s t indicated that th i s was not a s ign i f i cant d i f ference (£ > .05) . The s o c i a l l y anxious group had a mean global anxiety rat ing of 4.71 compared to 5.63 for the nonanxious group. This di f ference was s t a t i s t i c a l l y s i g n i f i c a n t , _t(46) = - 2 . 0 6 , j) < .05 . A t - t e s t was performed on subjects' s e l f - r a t i n g s of comfort/discomfort following the i r interact ion with the confederate. Soc ia l l y anxious subjects had a mean rating of 3.75 and nonanxious subjects had a mean rating of 5.00. This dif ference was also s t a t i s t i c a l l y s ign i f i cant _t(46) = - 3 . 4 1 , £ < . 0 5 . Las t l y , two - ta i led t - t e s t s revealed a s ign i f i cant dif ference between s e l f - r a t i n g s of comfort/discomfort and judges' global ratings of anxiety for the s o c i a l l y anxious group, t^46) = - 2 . 1 1 , £ < .05 , but not for the nonanxious group, _t(46) = - 1 . 7 5 , _p_ > .05 . In summary, these f indings indicate that while objective judges did not observe a di f ference in the level of social s k i l l displayed by s o c i a l l y anxious and nonanxious subjects , they did observe a di f ference on the i r leve ls of anxiety . The l a t t e r result i s in agreement with s o c i a l l y anxious subjects' s e l f - r e p o r t s that they experienced greater anxiety during the i r interact ions with the confederate than s o c i a l l y nonanxious subjects . However, anxious subjects perceived themselves as more uncomfortable than did objective judges. There was no such discrepancy between s e l f - r a t i n g s and objective global ratings of anxiety for the s o c i a l l y nonanxious group. 38 Depression Scores As reported e a r l i e r , the mean BDI scores for s o c i a l l y anxious and nonanxious groups were 6.88 and 2 .38, respect ive ly . A 2(SAD) X 2(Condition) ANOVA revealed a s i g n i f i c a n t main e f fec t for SAD, £(1,44) = 20.70, £ < . 0 1 . The F ra t ios for the main e f f e c t for condit ion and the SAD X condit ion in teract ion were not s i g n i f i c a n t . Based on th is f i n d i n g , BDI scores were incorporated into the main data analys is as a covar iate . Order Ef fects A one-way mul t i var ia te analys is of variance (MANOVA) with repeated measures was conducted to determine whether or not the order in which the relaxed and anxious in te rva l s were portrayed to subjects by the confederate had a s i g n i f i c a n t e f f e c t , i r respect ive of subject group or i n t e r v a l . The MANOVA revealed a nonsignifcant main e f fec t for cond i t ion , 1F(6,41) = 1 .35, ns (see Appendix C for summary table and f i g u r e s ) . Addit ional analyses were car r ied out to invest igate the r e l a t i o n between SAD, condit ion and i n t e r v a l . A 2(SAD) X 2(Condition) X 2( Interval ) mul t i var ia te analys is of covariance (MANCOVA) with depression as the Covariate showed a nonsigni f icant e f fec t for the SAD 1 H o t e l l i n g s method for c a l c u l a t i o n of mul t ivar iate F's i s used throughout the Results sect ion (Winer, 1971). 39 X condit ion X interval i n t e r a c t i o n , £(6,17) = . 8 2 , ns. The SAD X condit ion interact ion was also not s i g n i f i c a n t , £(6,17) = 2 .15 , ns. L a s t l y , the mult ivar iate F's for the SAD X interval in teract ion and the main e f f e c t for SAD were nons ign i f i cant , although some univar iate F's were s i g n i f i c a n t . These l a t t e r two resu l ts w i l l be discussed further in the next sect ion . .Given the absence of a s i g n i f i c a n t re la t ion between condit ion and SAD, and given that condit ion was counterbalanced across groups, order ef fects were not considered in subsequent data ana lys i s . Interact ion Measures A 2(SAD) X 2(Interval) MANCOVA with BDI scores as the covariate for repeated measures was employed to analyze nonverbal-nonvocal and nonverbal-vocal dependent measures (see Table 4 ) . The main e f fec t for SAD was nons ign i f icant , f_(6,40) = 1.58, ns, however univar iate F - tests indicated a s i g n i f i c a n t e f fec t for observation between groups, £(1,45) = 7 .45, £ < . 0 1 . The wi th in -subjects factor ( in terva l ) produced a highly s i g n i f i c a n t main e f f e c t , £(6,41) = 7 .37 , £ < .001. According to the subsequent univar iate analyses, the fo l lowing measures produced s i g n i f i c a n t e f fec ts across i n t e r v a l s : gestures, £(1,46) = 18.56, £ < .001, nods £(1,46) = 12.76, £ = .001, time t a l k i n g , £(1,46) = 26.29, £ < .001, and pauses, £(1,46) = 10.71, £ < . 0 1 . The mul t i var ia te test for *Hotel l ings method for ca l cu la t ion of mul t i var ia te F's i s used throughout the Results section (Winer, 1971). 40 the SAD X interval in teract ion e f fect was nons igni f icant , FJ6.41) = 1.27, ns. Univariate F - tests did reveal a s i g n i f i c a n t difference between groups in the relaxed interva l for gestures F_(l,46) = 5 .70, £ < . 0 5 . Figures 4 and 5 summarize in graphic form the s i g n i f i c a n t f indings of t h i s a n a l y s i s . The s o c i a l l y nonaxious group spent a s i g n i f i c a n t l y greater amount of time observing the confederate across in terva ls compared to the anxious group. Furthermore, a change in the confederate's behaviour from relaxed to anxious resulted in s i g n i f i c a n t decreases in the frequency of nods and gestures and time spent ta lk ing across groups. Soc ia l l y anxious and non-anxious subjects also displayed a s i g n i f i c a n t increase in number of pauses when the confederat's behaviour changed. L a s t l y , the analysis revealed that the nonanxious group gestured more frequently than the anxious group when the confederate was relaxed and displayed a s i g n i f i c a n t decrease in gesturing when the confederate was anxious. 41 Table 4. Means and Standard Deviations for Repeated Measures between Groups3 Interval Relaxed Anxious Repeated Measures Stati s t i c HSAb LSAC HAS LSA Observation M 81.96 97.71 78.29 99.25 (seconds) SD 31.53 25.29 21.10 25.59 Posturing M 1.79 1.92 1.71 1.75 (frequency) SD 2.37 1.84 2.24 2.23 Gestures M 6.29 10.08 4.33 3.25 (frequency) SD 8.74 8.59 5.42 3.37 Nods M 4.58 6.67 3.42 4.79 (frequency) SD 3.78 3.78 1.99 3.08 Ta 1 k i ng M 43.46 54.08 35.33 40.79 (seconds) SD 17.01 17.63 14.79 13.54 Pauses M 1.71 1.00 2.63 1.67 (frequency) SD 1.52 .98 1.56 1.81 a n = 24 bHSA = High Social Anxiety CLSA = Low Social Anxiety 42 100-1 90-OBSERVATION (seconds) gO-High Social Anxiety Low Social Anxiety 1.80-POSTURING (frequency) uo-1.60-GESTURES (frequency) 8 -6 -4 -Relaxed Anxious Confederate's Behaviour Figure 4. Observation, posturing and gestures as a function of social anxiety and confederate's behaviour 43 High Social Anxiety Low Social Anxiety NODS (frequency) 6 5-4-1 55-TALKING (seconds) 45. 35H PAUSES (frequency) 2.50 2.00-1.50-Relaxed Anxious Confederate's Behaviour Figure 5. Nods, talking and pauses as a function of social anxiety and confederate's behaviour 44 Post - Interact ion Measures A one-way MANCOVA across SAD using BDI as the covariate was conducted to analyze subjects' written responses col lected fol lowing the i r interact ion with the confederate. Table 5 presents a summary of the means and standard deviations for the post - interact ion essay. The main ef fect for SAD was not s i g n i f i c a n t , £(3,43) = .454, ns . The resu l ts of the subsequent univariate F-tests were as fo l lows : references to confederate's behaviour, £(1,46) = 4 .82, ns , descr ip t ive statements about the confederate, £(1,46) = .337, ns , and statements about s e l f , £(1,46) = .203 , ns . The two se l f - repor t measures used to assess whether or not subjects able to perceive a change in the confederate's behaviour were analyzed the chi-squared s t a t i s t i c . Table 6 summarizes the resul ts of t h i s a n a l y s i s . There were no s i g n i f i c a n t differences between s o c i a l l y anxious and nonanxious subjects on e i ther measure. The z s t a t i s t i c (Johnson, 1976) was employed to examine within group differences across the p r o b a b i l i t i e s of these two measures. The proportion of subjects who perceived a change in the confederate's behaviour was s i g n i f i c a n t l y greater for the s o c i a l l y anxious group, ^z(0.0S) = - 3 . 7 5 , and the s o c i a l l y nonanxious group, £(0.05) = - 3 . 4 2 , when asked d i r e c t l y i f they observed a change. 45 Table 5. Means and Standard Deviations for Post - Interact ion Essay across Groups (n=24) Stat i s t i c Social Anxiety Coded Categories High Low References to Confederate's Behaviours M 1.58 2.75 SD 1.89 2.17 Descriptive Statements • about Confederate M 5.29 5.13 SD 2.49 1.99 Statements about Self M 2.63 2.75 SD 3.23 2.40 46 Table 6. Corrected Chi-squared S t a t i s t i c s for Subjects' Responses to Perception of Change Measures Measure Group(n=24) Perception of Change Yes No 2 P Essay-Perception HSA a 6 18 of change LSA b 11 13 1.45 .277 Perception of change HSA LSA 19 5 22 2 .66 .413 a High Social Anxiety D Low Social Anxiety 47 In summary, analyses of the post - in teract ion measures revealed no s i g n i f i c a n t dif ferences between the s o c i a l l y anxious and nonanxious groups in terms of the frequency of references to spec i f i c overt behaviours emitted by the confederate, descr ipt ive statements regarding the confederate's interpersonal s t y l e , or statements about themselves while interact ing with the confederate. Furthermore, no di f ferences were observed when subjects' post - in te ract ion essays were coded according to the absence or presence of the perception of a change in the confederate's behaviour. F i n a l l y , when s p e c i f i c a l l y asked whether or not they observed a change in the behaviour of the confederate, both s o c i a l l y anxious and nonanxious groups showed a s i g n i f i c a n t increase in the number of subjects who answered yes. However, there was no s i g n i f i c a n t group ef fect on th is measure. 48 Discussion The resul ts of t h i s study do not support the hypothesis that s o c i a l l y anxious ind iv iduals are less sens i t ive to changes in the nonverbal behaviour of thei r interpersonal partner than s o c i a l l y nonanxious ind iv iduals when level of depression is cont ro l l ed . Both groups reported observing a change in the confederate's behaviour according to thei r post - in teract ion s e l f - r e p o r t s . Furthermore, verbal and nonverbal interact ion measures as rated by objective judges revealed a s i g n i f i c a n t group X confederate's behaviour interact ion on only one of s ix va r iab les . The nonanxious group displayed a s i g n i f i c a n t l y greater decrease in the frequency of the i r communicative gestures when the confederate portrayed an anxious state compared to the anxious group. However, both s o c i a l l y anxious and nonanxious subjects displayed decreases in time spent t a l k i n g , nodding, and gesturing as well as a s i g n i f i c a n t increase in the number of pauses when the confederate appeared anxious. In terms of absolute leve ls of behaviour, the nonanxious group spent more time observing the confederate than did the anxious group. S ign i f i cant group differences were not obtained on measures of postur ing, gestures, nodding, time spent ta lk ing or number of pauses. The f a i l u r e of f i ve of the six spec i f i c behavioural measures to d i f f e r e n t i a t e between s o c i a l l y anxious and nonanxious indiv iduals in the current study i s consistent with the notable absence of r e l i a b l e behavioural indices in past research. These f indings lend support to the recommendation made by F ischet t i et al (1977) that researchers 49 expand the i r enpi r ica l horizons beyond the measurement of simple frequencies in order to more f u l l y understand the nature of social anx iety . Judges' global ratings of subjects' s k i l l and anxiety revealed no d i f ference between groups in terms of social s k i l l ; however, a s i g n i f i c a n t dif ference was observed in the i r leve ls of anxiety . Anxious subjects displayed greater levels of anxiety than the i r nonanxious counterparts. In add i t ion , anxious subjects perceived themselves as more uncomfortable than did objective judges rating the i r degree of anxiety . This discrepancy between judges' ratings and subjects' s e l f -reports was not observed for the nonanxious group. These results are consistent with previous f indings in the l i t e r a t u r e (Clark & Arkowitz, 1975) and support the notion that behavioural indices are not adequate measures of social anxiety (Galassi & G a l a s s i , 1981). The discrepancy between anxious subjects' se l f - repor ts and judges' global ratings suggests the presence of an ' interference mechanism' which is cognit ive in nature (Curran, 1979). One possible explanation i s that s o c i a l l y anxious ind iv idua ls fear that they may be evaluated negatively by others (Watson & Fr iend, 1969; Smith et a l , 1983) and as a r e s u l t , experience greater anxiety . There was a s ign i f i cant order ef fect across groups on the in teract ion measures. The pattern of f indings for the order ef fects was var iable and therefore not readi ly in te rp re tab le . An attempt to explain these resu l ts in re lat ion to social anxiety would be conceptually confusing given th is lack of consistency in the data . It i s recommended 50 that the influence of a relaxed, then anxious interpersonal partner versus an anxious, then relaxed partner be examined independently of social anxiety before attempting to investigate the re la t ion between social anxiety and systematic changes in the behaviour of an interpersonal partner. Methodological Considerations While the present study attempted to provide an experimentally sound test of the or ig ina l hypothesis, there were a number of methodological shortcomings that should be considered when evaluating the f ind ings . F i r s t l y , the duration of the social interact ion task was 4 minutes. Trower (1980) employed a s imi lar experimental manipulation and obtained posi t ive r e s u l t s ; however, the interact ion between the subjects and confederates lasted 12 minutes. Steffen and Reckman (1978) also used a n a t u r a l i s t i c social interact ion task but found no support for the i r hypothesis. The duration of the interact ion task in thei r invest igat ion was 6 minutes. It is possible that the shorter time durations used in Steffen and Reckman's study as well as the present study were only s u f f i c i e n t to demonstrate an expected decrease in the frequency of behaviours when ind iv iduals interact with an unresponsive confederate. On the other hand, a longer interact ion period may have revealed a s i g n i f i c a n t dif ference between s o c i a l l y anxious and nonanxious groups. A second possible shortcoming of t h i s study involves the choice of dependent va r iab les . The interact ion measures employed (postur ing, 51 gestures, nods, observation, pauses, time talk ing) were l imi ted to assessing changes in gross behaviours. More refined measures such as f a c i a l movements may have provided for the p o s s i b i l i t y of d iscr iminat ing subtle behavioural responses. In terms of the global ratings of anxiety and social s k i l l recal l that judges' rated each subject based on her overal l performance during the interact ion task. It remains unknown as to whether or not global ratings on these two dimensions fol lowing each 2 minute interval would have demonstrated differences between s o c i a l l y anxious and nonanxious subjects in thei r responses to the confederate's behaviour. F i n a l l y , a few comments are in order regarding the two post-in teract ion measures of the subjects' perceptions of change in the confederate's behaviour. Due to the retrospective nature of these measures, i t was not possible to determine at what point subjects observed a change. Once again, had such data been obtained, i t may have discr iminated s o c i a l l y anxious and nonanxious groups on the basis of the immediacy or latency of the i r perceptions of the change in the confederate's behaviour. Before leaving th is discussion of methodological concerns associated with the dependent measures, i t i s important to take note of the d i f ference in data obtained from the two perception of change measures. In the f i r s t instance, subjects were asked to describe in the i r own words how the i r partner responded to them during the i r conversat ion. The second measure was more d i rec t and asked subjects the fol lowing question: "Did you notice a change in your partner 's level of comfort or discomfort over the course of the conversation?" When asked 52 i n d i r e c t l y , 25% of the s o c i a l l y anxious group and 46% of the nonanxious group described a change in the confederate's behaviour. When asked d i r e c t l y , 79% of the s o c i a l l y anxious group and 92% of the nonanxious stated that they observed a change. While the differences between the two groups on either measure were not s i g n i f i c a n t , the di f ference between the two measures was s i g n i f i c a n t . It appears that the more d i rec t measure acted as a cue which aided subjects in reca l l i ng the confederate's behaviour change during the in te rac t ion . Furthermore, t h i s f inding suggests that the ind i rect measure i s not va l id as an absolute measure of subjects' perceptions. In fu ture , researchers in th is area who chose to employ s imi la r retrospective measures of subjects' perceptions must determine whether or not they wish to assess a l l perceptions made by the i r subjects or only those perceptions which are most immediately recal led by subjects . One last methodological l i m i t a t i o n of the present study which may have contributed to the negative findings was the role of the confederate. The s h i f t from an anxious to relaxed state (or v ice versa) was dramatic and therefore an un l ike ly representation of the t y p i c a l l y subtle nonverbal cues that character ize social i n te rac t ions . Had the confederate's change in behaviour been less obvious, the s o c i a l l y anxious group may not have perceived the change as e a s i l y as the nonanxious group. Conclusions In order to interpret the f indings of the present study, they must 53 be examined not only in l i g h t of the i r methodological soundness, but in re la t ion to the relevant research as w e l l . The present study s t a t i s t i c a l l y control led for the influence of depression as an a l ternat ive interpretat ion of the r e s u l t s . Depression was not contro l led for in any of the relevant studies reviewed e a r l i e r . Anc i l la ry analyses excluding depression as a covariate revealed that while d i f f e r e n t i a l s e n s i t i v i t y across groups was not demonstrated, the number of s i g n i f i c a n t group dif ferences on the interact ion measures was substant ia l l y d i f f e r e n t . Soc ia l l y nonanxious subjects spent s i g n i f i c a n t l y more time observing the confederate, nodding, and t a l k i n g , and paused less frequently than did the s o c i a l l y anxious subjects . Furthermore, subjects' essay responses revealed that the nonanxious group recorded a greater number of references to spec i f i c overt behaviours emitted by the confederate than did the anxious group. These resu l ts underscore the importance of cont ro l l ing for the impact of depression in studies invest igat ing the nature of social anxiety. While the outcome of the present study remains unchanged when depression is s t a t i s t i c a l l y c o n t r o l l e d , i t remains an empirical question as to whether or not the f indings of past research are strong enough to remain when the test of sigfnicance i s more s t r ingent . Two of the three studies reviewed ear l i e r which employed a l i v e social interact ion test to investigate the re lat ion between s e n s i t i v i t y to social cues and social anxiety did not obtain d i f f e r e n t i a l s e n s i t i v i t y across groups (Steffen & Reckman, 1978; Christensen et al , 1980). Furthermore, these two studies and the present study selected 54 col lege students as subjects . In contrast to t h i s , the th i rd invest igat ion which u t i l i z e d a l i v e social interact ion test (Trower, 1980), selected subjects from a pool of psychiatr ic patients diagnosed as neurot ic , phobic, and having personality d isorders . Although Trower's f indings supported the hypothesis that s o c i a l l y unski l led ind iv iduals are less responsive to s i tuat ional cues than s k i l l e d i n d i v i d u a l s , the heterogenous nature of the subject sample and the use of verbal and nonverbal cues on the part of the confederates do not allow one to speculate on the re la t ion between s e n s i t i v i t y to nonverbal social cues and social anxiety. Based on the findings of the current study and the relevant research, i t appears that s o c i a l l y anxious ind iv iduals are no less sensit ive to nonverbal social cues than s o c i a l l y nonanxious i n d i v i d u a l s . Recommendations While current f indings do not support Croz ier 's (1979) proposition that s e l f - a t t e n t i o n i s an important component of social anxiety, the theory of anxious self -preoccupation should not be abandoned prematurely. The experimental evidence suggests that social anxiety i s not related to a decrease in s e n s i t i v i t y to nonverbal social cues. However, there i s evidence to suggest that s o c i a l l y anxious indiv iduals may be less sens i t i ve to vocal cues when compared to nonanxious i n d i v i d u a l s . F ischet t i et al (1977) and Peterson et al (1981) employed a prerecorded tape of a female voice to demonstrate that heterosexual-s o c i a l l y anxious 55 col lege students lack the a b i l i t y to discr iminate cues that guide the timing and placement of social responses. One possible d i rect ion for future research would be to test out the g e n e r a l i z a b i l i t y of these f indings based on vocal dependent measures in the context of a l i v e socia l i n t e r a c t i o n . A second avenue for experimental inquiry stems from the negative resu l ts of the present study as well as those of Steffen and Reckman (1978) and Christensen et al (1980). In both structured and unstructured social s i t u a t i o n s , s o c i a l l y anxious indiv iduals appear to have the a b i l i t y to perceive nonverbal cues. However, i t may be the case that s o c i a l l y anxious indiv iduals interpret these cues d i f f e r e n t l y than the i r nonanxious counterparts. Recall that in the Christensen et al (1980) study, s o c i a l l y competent subjects responded more immediately to cues of tension displayed by the confederate even though competent and incompetent groups became aware of the tension at the same point . Post-experimental interviews revealed that s o c i a l l y incompetent subjects tended to a t t r ibute the confederate's tension to s i tuat ional factors versus the confederate's behaviour. Additional experimentation is needed to c l a r i f y the re la t ion between social anxiety and perception, in terpretat ion and responsiv i ty . Furthermore, the present study did not address the issue of sex d i f fe rences . There i s evidence to suggest that females are , in general , better decoders of nonverbal cues than males ( H a l l , 1978). Further research i s needed to better understand the re la t ion between sex and social anxiety . APPENDIX A Social Avoidance and Distress Scale 57 Social Avoidance and Distress Scale The fol lowing questions are concerned with your b e l i e f s , fee l ings , and a c t i o n s . Decide whether each statement i s more true or fa lse as applies to you personally and c i r c l e either true (T) or fa lse (F) after each. Work qu ick l y , giving your f i r s t reaction to each statement. If a statement i s sometimes true and sometimes f a l s e , decide which i s more typical of you personal ly . 1. I feel relaxed even in unfamil iar social s i tua t ions . T F 2. I t r y to avoid s i tuat ions which force me to be very soc iab le . T F 3. It i s easy for me to relax when I am with strangers. T F 4. I have no par t i cu lar desire to avoid people. T F 5. I often f ind social occasions upsett ing. T F 6. I usual ly feel calm and comfortable at social occasions. T F 7. I am usually at ease when ta lk ing to someone of the opposite sex. T F 8. I t r y to avoid ta lk ing to people unless I know them w e l l . T F 9. If the chance comes to meet new people, I often take i t . T F 10. I often feel nervous or tense in casual get-togethers in which both sexes are present. T F 11. I am usually nervous with people unless I know them w e l l . T F 12. I usual ly feel relaxed when I am with a group of people. T F 13. I often want to get away from people. T F 14. I usual ly feel uncomfortable when I am in a group of people I don't know. T F 15. I usual ly feel relaxed when I meet someone for the f i r s t t ime. T F 16. Being introduced to people makes me tense and nervous. T F 17. Even though a room i s f u l l of strangers, I may enter i t anyway. T F 58 18. I would avoid walking up and jo in ing a large group of people. T F 19. When my superiors want to talk with me, I talk w i l l i n g l y . T F 20. I often feel on edge when I am with a group of people. T F 21. I tend to withdraw from people. T F 22. I don't mind ta lk ing to people at part ies or social gatherings. T F 23. I am seldom at ease in a large group of people. T F 24. I often think up excuses in order to avoid social engagements. T F 25. I sometimes take the respons ib i l i t y for introducing people to each other . T F 26. I t ry to avoid formal social occasions. T F 27. I usual ly go to whatever social engagements I have. T F 28. I f ind i t easy to relax with other people. T F APPENDIX B Beck Depression Inventory Beck Inventory 60 On th is questionnaire are groups of statements. Please read each group of statements c a r e f u l l y . Then pick out the one statement in each group which best describes the way you have been feel ing the PAST WEEK, INCLUDING TODAY! C i r c l e the number beside the statement you picked^ If several statements in the group seem to apply equally w e l l , c i r c l e each one. Be sure to read a l l the statements in eeach. group before making your choice. 1 0 I do not feel sad. 1 I feel sad. 2 I am sad a l l the time and I can ' t snap out of i t . 3 I am so sad or unhappy that I can't stand i t . 2 0 I am not p a r t i c u l a r l y discouraged about the future . 1 I feel discouraged about the future . 2 I feel I have nothing to look forward t o . 3 I feel that the future i s hopeless and that things cannot improve. 3 0 1 do not feel l i k e a f a i l u r e . 1 I feel I have fa i l ed more than the average person. 2 As I look back on my l i f e , a l l I can see i s a lo t of f a i l u r e s . 3 I feel I am a complete f a i l u r e as a person. 4 0 1 get as much s a t i s f a c t i o n out of things as I used t o . 1 I don't enjoy things the way I used t o . 2 I don't get real sa t i s fac t ion out of anything anymore. 3 I am d i s s a t i s f i e d or bored with everything. 5 0 1 don't feel p a r t i c u l a r l y g u i l t y . 1 I feel g u i l t y a good part of the t ime. 2 I feel quite g u i l t y most of the t ime. 3 I feel g u i l t y a l l of the t ime. 6 0 1 don't feel I am being punished. 1 I feel I may be punished. 2 I expect to be punished. 3 I feel I am being punished. 7 0 1 don't feel disappointed in myself . 1 I am disappointed in myself . 2 I am disgusted with myself . 3 I hate mysel f . 8 0 I don't feel I am any worse than anybody e l s e . 1 I am c r i t i c a l of myself for my weaknesses or mistakes. 2 I blame myself a l l the time for my f a u l t s . 3 I blame myself for everything bad that happens. 10 11 12 13 14 15 16 17 0 1 2 3 0 1 2 3 0 1 2 3 0 1 2 3 0 1 2 3 0 1 2 0 1 2 3 0 1 2 0 1 2 3 61 don't have any thoughts of k i l l i n g myself . have thoughts of k i l l i n g myself, but I would not carry them out. would l i k e to k i l l myself. would k i l l myself i f I had the chance. don't cry anymore than usual , cry more now than I used t o . cry a l l the time now. used to be able to c ry , but now I can' t cry even though I want to . am no more i r r i t a t e d now than I ever am. get annoyed or i r r i t a t e d more eas i l y than I used t o . feel i r r i t a t e d a l l the time now. don't get i r r i t a t e d at a l l by the things that used to i r r i t a t e me. have not lost interest in other people, am less interested in other people than I used to be. have los t most of my interest in other people, have lost a l l of my interest in other people. make decisions about as well as I ever could . put off making decisions more than I used t o . have greater d i f f i c u l t y in making decisions than before. can ' t make decisions at a l l anymore. don't feel I look any worse than I used t o . am worried that I am looking old or unat t ract ive , feel that there are permanent changes in my appearance ook unat t ract i ve , believe that I look ugly. that make me can work about as well as before, t takes me extra e f fo r t to get started at doing something, have to push myself very hard to do anything, can ' t do any work at al 1 . can sleep as well as usual , don't sleep as well as I used t o . wake up 1-2 hours e a r l i e r than usual and find i t hard to get back to s leep. wake up several hours e a r l i e r than I used to and cannot get back to s leep. don't get more t i red than usual , get t i red more eas i l y than I used t o . get t i red from doing almost anything, am too t i red to do anything. 62 18 0 My appetite is no worse than usual . 1 My appetite i s not as good as i t used to be. 2 My appetite i s much worse now. 3 I have no appetite at a l l anymore. 19 0 I haven't lost much weight, i f any l a t e l y . 1 I have lost more than 5 pounds. I am purposely t ry ing 2 I have los t more than 10 pounds. to lose weight by eat -3 I have lost more than 15 pounds. ing less Yes No 20 0 I am no more worried about my health than usual . 1 I am worried about physical problems such as aches and pains; or upset stomach; or const ipat ion . 2 I am verry worried about physical problems and i t ' s hard to think of much el se. 3 I am so worried about my physical problems, that I cannot think about anything el se. 21 0 I have not noticed any recent change in my interest in sex. 1 I am less interested in sex than I used to be. 2 I am much less interested in sex now. 3 I have lost interest in sex completely. APPENDIX C Results of one-way MANOVA for Order E f fects High Social Anxiety Low Social Anxiety — i 1 — Relaxed Anxious CONDITION 1 + Relaxed Anxious CONDITION 2 I 1 1 1 — 1 — • Relaxed Anxious Relaxed Anxious CONDITION 1 CONDITION 2 Figure 1. Observation and posturing as a function of condition High Social Anxiety Low Social Anxiety 65 CD c r CD CO U J CC § LU CD 10-8 6 -— r i — Relaxed Anxious CONDITION 1 -t- — — i — — Relaxed Anxious CONDITION 2 cr co cr cu CO Q O 8-6-4-2-— I H 1 -+- 1 — • Relaxed Anxious Relaxed Anxious CONDITION 1 CONDITION 2 Figure 2. Gestures and nods as a function of condition 66 _ _ High Social Anxiety Low Social Anxiety ^ so CO TD C § 40 CD i 30 20 H - g 2.50 CD §•2.00 % 1-50 1.00 I 1 1 l I Relaxed Anxious Relaxed Anxious CONDITION 1 • / CONDITION 2 / / / / / / / l / M f w S 1 / 4 • • 1 1 Relaxed Anxious CONDITION 1 Relaxed Anxious CONDITION 2 Figure 3. Talking and pauses as a function of condition 67 Table 2. Means and Standard Deviations for Repeated Measures across Conditions and Groups3 Condition 1 Condition 2 Repeated Relaxed Anxious Anxious Relaxed Measures S t a t i s t i c HSAb LSAC HSA LSA HSA LSA HSA LSA Observation M 93.25 99.08 75.75 89.58 80.83 108.92 70.67 96.33 (seconds) SD 29.83 24.43 22.04 32.18 20.77 11.39 30.17 27.14 Posturing M 1.25 1.33 .83 1.17 2.58 2.33 2.33 2.50 (frequency) SD 1.29 .98 1.19 .83 2.71 2.99 3.08 2.32 Gestures M 7.83 10.25 5.42 4.00 3.25 2.50 4.75 9.92 (frequency) SD 10.46 9.22 7.22 4.18 2.59 2.24 6.70 8.33 Nods M 5.42 7.17 3.00 4.08 3.83 5.50 3.75 6.17 (frequency) SD 4.66 3.90 2.00 2.39 1.99 3.61 2.56 3.74 Tai king M 47.17 52.33 33.4Z 39.17 37.25 42.42 39.75 55.83 (seconds) SD 17.43 13.75 15.63 14.73 14.33 12.67 16.46 21.31 Pauses M .92 1.08 2.92 2.08 2.33 1.25 2.50 .92 (frequency) SD 1.08 .99 1.31 1.31 1.78 2.18 1.51 .99 a n = 12 DHSA = High Social Anxiety CLSA = Low Social Anxiety Table 3 . Summary of Univariate F - tes ts for Condition E f fect 68 Dependent Var iable df F S ign i f icance Observation 1,46 .00 ns Posturing 1,46 6.10 .02 Gestures 1,46 1.07 ns Nods 1,46 .01 ns Talking 1,46 .03 ns Pauses 1,46 .00 ns Mu l t i va r ia te F 6,41 1.35 ns 69 References Argyle, M. & Kendon, A. The experimental analysis of social performance. In L. Berkowitz (Ed . ) , Advances in Experimental Social Psychology (Vo l . 3 ) . New York: Academic Press, 1967. Argy le , M., L a l l j e e , M. & Cook. M. The ef fects of v i s i b i l i t y on interact ion in a dyad. Human Relat ions , 1968, 2_1_, 3 - 1 7 . Arkowitz, H., L i ch tenste in , E., McGovern, K., & Hines, P. The behavioral assessment of social competence in males. Behavior Therapy, 1975, 6s 3 -13 . Arkowitz, H., Hinton, R., P e r l , J . & Himadi, W. Treatment strategies for dating anxiety in col lege men based on r e a l - l i f e p r a c t i s e . The  Counseling Psychologist , 1978, 7_, 41-46. Bander, K.W., Steinke, G.V., A l l e n , G . J . & Mosher, D.L. Evaluation of three d a t i n g - s p e c i f i c treatment approaches for heterosexual dating anxiety . Journal of Consulting and C l i n i c a l Psychology, 1975, 4 3 , 259-265. Barlow, D.H. Increasing heterosexual responsiveness in the treatment of sexual dev ia t ion : A review of the c l i n i c a l and experimental evidence. Behavior Therapy, 1973, £ ,655-671. Beck, A .T . , Depression: Causes and Treatment. Ph i lade lph ia : Univers i ty of Pennsylvania Press , 1967. Be l lack , A . S . , Hersen, M. , & Turner, S.M. General ization ef fects of soc ia l s k i l l s t ra in ing in chronic schizophrencics: An experimental a n a l y s i s . Behaviour Reserach and Therapy, 1976, 14, 391-398. 70 Be l lack , A . S . , Hersen, M. & Lamparski, D. Role-play tests for assessing social s k i l l s : Are they va l id? Are they useful? Journal of  Consulting and C l i n i c a l Psychology, 1979, 47_, 335-342. Bel lack, A . S . & Hersen, M. Research and Pract ice in Social S k i l l s  T ra in ing . New York: Plenum Press , 1979. Bel lack , A .S . A c r i t i c a l appraisal of strategies for assessing socia l s k i l l . Journal of Behavioral Assessment, 1979, 1_, 2^ , 157-176. Bel lack, A . S . & Morrison, R.L. Interpersonal dysfunct ion. In A .S . Be l lack , M. Hersen & A .E . Kazdin (Eds . ) , International Handbook of  Behavior Modif icat ion and Therapy. New York: Plenum Press, 1982. B e l l a c k , A .S. Recurrent problems in the behavioral assessment of social s k i l l . Behaviour Research and Therapy, 1983, 21_, U 29-41. Borkovec, T .D. , Stone, N.M., O 'Br ien, G.T. and Kaloupek, D.G. Evaluation of a c l i n i c a l l y relevant target behavior for analog outcome research. Behavior Therapy, 1974, £ , 503-513. Bumberry, W., O l i ver , J . M . & McClure, J . N . Val idat ion of the Beck Depression Inventory in a univers i ty population using psychiatr ic estimate as the c r i t e r i o n . Journal of Consulting and C l i n i c a l  Psychology, 1978, 46, J^, 150-155. Cacioppo, J . T . , Glass, C R . & Mer luzz i , T.V. Self -statements and s e l f -evaluat ions: A cognitive-response analysis of heterosocial anxiety. Cognit ive Therapy and Research, 1979, 3_, 3, 249-262. C a l l n e r , D.A. & Ross, S.M. The r e l i a b i l i t y and v a l i d i t y of three measures assert ion in a drug addict populat ion. Behavior Therapy, 1976, 7, 659-667. Christensen, A. & Arkowitz, H. Prel iminary report on pract ice dating and feedback as treatment for col lege dating problems. Journal of  Consulting and C l i n i c a l Psychology, 1974 , 2£, 92-95. Christensen, D., Far ina, A. & Boudreau, L. Sens i t i v i t y to nonverbal cues as a function of social competence. Journal of Nonverbal Behavior, 1980, 4 , 3, 146-156. C la rk , J .V . & Arkowitz, H. Social anxiety and se l f -eva luat ion of i n t e r -personal performance. Psychological Reports, 1 9 7 5 , 3 6 , 211-221. C roz ie r , R. Shyness as anxious sel f -preoccupat ion. Psychological  Reports, 1979, 44, 959-962. Curran, J . P . Social s k i l l s t ra in ing and systematic desens i t i zat ion in reducing dating anxiety . Behaviour Research and Therapy, 1975, 13, 65-68. Curran, J . P . & G i l b e r t , F.S. A test of the r e l a t i v e effect iveness of a systematic desens i t i zat ion program and an interpersonal s k i l l s t ra in ing program with date anxious subjects. Behavior Therapy, 1975, 6, 510-521. Curran, J . P . & G i l b e r t , F.S. & L i t t l e , L.M. A comparison between behavioral r e p l i c a t i o n t ra in ing and s e n s i t i v i t y t ra in ing approaches to heterosexual dating anxiety. Journal of Counseling Psychology, 1976, 2 3 , 3 , 190-196. Curran, J . P . S k i l l s t ra in ing as an approach to the treatment of heterosexual soc ia l - a n x i e t y . Psychological B u l l e t i n , 1977, 84, 140-157. Curran, J . P . Pandora's box reopened? The assessment of social s k i l l s . Journal of Behavioral Assessment, 1979, 1_, 1_, 55-71. Curran, J . P . & Wessberg, H.W. Assessment of social inadequacy. In D. Barlow (Ed . ) , Behavioral Assessment of Adult Disorders. New York: The Guildford Press , 1981. E i s l e r , R.M., M i l l e r , P . M . , Hersen, M. & A l f o r d , H. Ef fects of assert ive t ra in ing on marital i n t e r a c t i o n . Archives of General Psychiatry , 1974, 30, 643-649. E i s l e r , R.M. , Hersen, M. , M i l l e r , P .M. , & Blanchard, E.B. S i tuat ional determinants of assert ive behaviors. Journal of Consulting and C l i n i c a l Psychology, 1975, 43, 330-340. F i s c h e t t i , M., Curran, J . P . & Wessberg, H.W. Sense of t iming: A s k i l l d e f i c i t in he tero sexual-soc i al l y anxious males. Behavior Modi f icat ion , 1977, 1_, 2, 179-194. Fishman, S. & Nawas, M. Treatment of polysomatic or global problems by systematic desens i t i za t ion . In R. Rubin (Ed . ) , Advances in Behavior Therapy. New York: Academic Press , 1973. Ga lass i , J . P . & G a l a s s i , M.D. Modif icat ion of heterosocial s k i l l s d e f i c i t s . In A .S . Bel lack and M. Hersen (Eds.) Research and Pract ise in Social S k i l l s T ra in ing . New York: Plenum Press, 1979. Glasgow, R.E. & Arkowitz , H. The behavioral assessment of male and female social competence in dyadic heterosexual i n te rac t ions . Behavior Therapy, 1975, 6\, 488-498. Goffman, E. On face-work: An analysis of r i t u a l elements in socia l i n t e r a c t i o n . Psychiat ry , 1955, 18, 213-221. 73 Goffman, E. Presentation of Se l f in Everday L i f e . New York: Doubleday, 1959. Go ld f r ied , M.R. & Sobocinski , D. E f fec t of i r r a t i o n a l be l ie fs on emotional arousal . Journal of Consulting and C l i n i c a l Psychology, 1975, 43, 504-510. Gormally, J . , Sipps, G. , Raphael, R., Edwin, D. & Varv i l -Weld , D. The re la t ionsh ip between maladaptive cognitions and social anxiety . Journal of Consulting and C l i n i c a l Psychology, 1981, 49 , 2, 300-301. G o t l i b , I.H. & Robinson, L.A. Responses to depressed ind i v idua ls : Discrepancies between s e l f - r e p o r t and observer-related behavior. Journal of Abnormal Psychology, 1982, 91_, 4 , 231-240. H a l l , J . Gender e f fec ts in decoding nonverbal cues. Psychological B u l l e t i n , 1978 , 85 , 845-857. Hargie, A . , Saunders, C. & Dickson, D. Social S k i l l s in Interpersonal Communication. London: Croom Helm L t d . , 1981. Hart, R.P. & Burks, D.M. Rhetorical s e n s i t i v i t y and social i n t e r a c t i o n . Speech Monographs, 1972, 39, 2, 75-91. Hersen, M. & Bel lack, A .S . Social s k i l l s t ra in ing for chronic psychiat r ic pat ients : Rat ionale , research f indings and future d i r e c t i o n s . Comprehensive Psychiatry , 1976, 17, 559-580. Hersen, M. & Be l lack , A.S. Assessment of social s k i l l s . In A.R. Ciminero, K.S. Calhoun, & H.E. Adams (Eds . ) , Handbook of Behavioral Assessment. New York: John Wiley & Sons, 1977. Hopkins, J . , Malleson, N. & Sarnoff , I. Non- inte l lectual correlates of success and f a i l u r e among univers i ty students. B r i t i s h Journal of Educational Psychology, 1958, 28, 23-35. Jacobson, N.S. & Anderson, E.A. Interpersonal s k i l l and depression in co l lege students: An analysis of the timing of sel f - d i s c i o s u r e s . Behavior Therapy, 1982, 13, 271-282. Johnson, R. Elementary S t a t i s t i c s . (Second Edit ion) North Sc i tuate , Mass.: Duxbury Press, 1976. Kanter, N.J. & G o l d f r i e d , M.R. Relat ive effect iveness of rat ional restructur ing and se l f - cont ro l desensi t i zat ion in the reduction of interpersonal anxiety . Behavior Therapy, 1979, 10_, 472-490. Kendon, A. Movement coordination in social i n t e r a c t i o n . Acta Psychologica, 1970, 32_, 100-125. K r a f t , T. Social anxiety model of a lcohol ism. Perceptual and Motor S k i l l s , 1971, 33, 797-798. Lucas, C . J . , K e l v i n , R.P. & Ojha, A.B. Mental health and student wastage. B r i t i s h Journal of Psychiatry , 1966, 112, 277-284. MacDonald, M.L. , L indquis t , C.V. , Kramer, J . A . , McGrath, R.A. & Rhyne, L.D. Social s k i l l s t r a i n i n g : Behaviour rehearsal in groups and dating s k i l l s . Journal of Counseling Psychology, 1975, 22_, 2» 2 2 4 -230. Malkiewich, L.E. & M e r l u z z i , T.V. Rational restructur ing versus desens i t i zat ion with c l i e n t s of diverse conceptual l e v e l s : A test of a cl ient-treatment matching model. Journal of Counseling Psychology, 1980, 27, 453-461. 75 Martinson, W.D. & Zerface, J.P. Comparison of individual counseling and a social program with nondaters. Journal of Counseling  Psychology, 1970, 17_, 36-40. Mehrabian, A. & Ferris, S.R. Inference of attitudes from nonverbal communication in two channels. Journal of Consulting Psychology, Mehrabian, A. & Wiener, M. Decoding of inconsistent communications. Journal of Personality and Social Psychology, 1967, 36, 1, 109-114. Mitchell, K.R. & Orr, T.E. Note on treatment of heterosexual anxiety using short-term massed desensitization. Psychological Reports, 1974, 35, 1093-1094. Morrison, R.L. & Bellack, A.S. The role of social perception in social s k i l l . Behavior Therapy, 1981, 12, 69-79. 0'Banion, K. & Arkowitz, H. Social anxiety and selective memory for affective information about the self. Social Behavior and Personality, 1977, 5_, 2, 321-328. Rehm, L.P. & Marston, A.R. Reduction of social anxiety through modification of self-reinforcement: An instigation therapy technique. Journal of Consulting and Clinical Psychology, 1968, 32, 065-574. Rosenthal R., Hall, J.A. Archer, D., DiMatteo, M.R. & Rogers, P.L. The Pons Test: Measuring sensitivity to nonverbal cues. In S. Weitz (Ed.), Nonverbal Communication. New York: Oxford University Press, 1979. 1967, 31, 3, 248-252. 76 Sarason, 1.6. Test Anxiety: Theory, Research and Applications. New Jersey: Lawrence Erlbaun Associates, 1980. Schlenker, B.R. &Leary,_M.R. Social anxiety and sel f-presentation: A conceptualization and model. Psychological Bulletin, 1982, 92, 1_, 641-669. Schwartz, R.M. & Gottman, J.M. Toward a task analysis of assertive behavior. Journal of Consulting and Clinical Psychology, 1976, 44, 910-920. Scott, W.O.N. & Edelstein, B.A. The social competence of two interaction strategies: An analog evaluation. Behavior Therapy, 1981, 12_, 482-492. Shapiro, J .6. Responsivity to facial and linguistic cues. Journal of  Communication, 1968, 18, 11-17. Smith, R.E. & Sarason, 1.6. Social anxiety and the evaluation of o negative interpersonal feedback. Journal of Consulting and Clinical  Psychology, 1975, 43, 3, 429. Smith, T.W., Ingram, R.E. & Brehm, S.S. Social anxiety, anxious self-preoccupation and recall of self-relevant information. Journal of  Personality and Social Psychology, 1983, 44, 6_, 1276-1283. Trower, P. Fundamentals of interpersonal behaviour: A social-psychological perspective. In A.S. Bellack & M. Hersen (Eds.), Research and Practice in Social Ski l i s Training. New York: PIenum Press, 1979. Trower, P. Situational analysis of the components and processes of behavior of socially skilled and unskilled patients. Journal of  Consulting and Clinical Psychology, 1980 , 48 , 3, 327-339. 77 Trower, P. A generative model of social skills. In J.P. Curran & P.M. Monti (Eds.), Social Skills Training: A Practical Handbook for Assessment and Treatment. New York: Guilford Press, 1982. Twentyman, C.T. & McFall, R.M. Behavioral training of social skills in shy males. Journal of Consulting and Clinical Psychology, 1975, 43, 384-395. Watson, D. & Friend, R. Measurement of social-evaluative anxiety. Journal of Consulting and Clinical Psychology, 1969, 33, 448-457. Waxer, P.H. Nonverbal cues for anxiety: An examination of emotional leakage. Journal of Abnormal Psychology, 1977, 86, 4, 306-314. Wessberg, H.W., Mariotto, M.J., Conger, A.J., Farrell, A.D. & Conger, J.C. Ecological validity of role plays for assessing heterosocial anxiety and skill of male college students. Journal of Consulting and Clinical Psychology, 1979, 47_, 3, 525-535. Wine, J.D. Test anxiety and direction of attention. Psychological Bulletin, 1971, 76, 92-104. Winer, B.J. Statistical Principles in Experimental Design. (Second Edition) New York: McGraw-Hill Book Company, 1971. Wolpe, J. The Practise in Behavior Therapy. New York: Pergamon Press, 1973. Youngren, M.A. & Lewinsohn, P.M. The functional relation between depression and problematic interpersonal behavior. Journal of Abnormal Psychology, 1980, 89, 3_, 333-341. Zigler, E. & Phillips, L. Social competence and the process reactive distinction in psychopathology. Journal of Abnormal and Social Psychology, 1961, 65, 215-222. Zimbardo, P.G. Shyness. New York: Jove Publ ications, Inc., 1977. ..' j » e s 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0096222/manifest

Comment

Related Items