UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Chronic construct accessibility in socially phobic and agoraphobic outpatients Capreol, Martha Jean 1991

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

Item Metadata

Download

Media
831-UBC_1991_A8 C36_5.pdf [ 3.7MB ]
Metadata
JSON: 831-1.0302389.json
JSON-LD: 831-1.0302389-ld.json
RDF/XML (Pretty): 831-1.0302389-rdf.xml
RDF/JSON: 831-1.0302389-rdf.json
Turtle: 831-1.0302389-turtle.txt
N-Triples: 831-1.0302389-rdf-ntriples.txt
Original Record: 831-1.0302389-source.json
Full Text
831-1.0302389-fulltext.txt
Citation
831-1.0302389.ris

Full Text

CHRONIC CONSTRUCT ACCESSIBILITY IN SOCIALLY PHOBIC AND AGORAPHOBIC OUTPATIENTS by MARTHA JEAN CAPREOL B.Sc, Queen's University, 1983 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTERS OF ARTS in THE FACULTY OF GRADUATE STUDIES Department of Psychology We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA October 1991 @ Martha Jean Capreol, 1991  In  presenting this  degree  at the  thesis  in  University of  partial  fulfilment  of  of  department  this or  thesis for by  his  or  requirements  British Columbia, I agree that the  freely available for reference and study. I further copying  the  representatives.  an advanced  Library shall make it  agree that permission for extensive  scholarly purposes may be her  for  It  is  granted  by the  understood  that  head of copying  my or  publication of this thesis for financial gain shall not be allowed without my written permission.  Department of  Psychology  The University of British Columbia Vancouver, Canada  D  a  t  e  DE-6 (2/88)  Ortnh^r  1 0  r  1QQ1  ii  Abstract  According to theories of social cognition, individuals screen incoming information from the environment using certain cognitive constructs (Wyer & Srull, 1986). Personally relevant construct systems develop from an individuals's particular history of social interactions (Wyer & Srull, 1986). A concern in social cognition theory is whether there are cognitive processes specific to different complaints (Beck & Emery, 1985). This study investigated whether the content of chronically accessible or salient constructs interpersonal constructs could differentiate individuals with social fears from those with different emotional complaints. The accessibility and salience of social constructs of social phobics, agoraphobics, and normal subjects were examined. No differences were found between the groups on an unstructured measure of construct accessibility. Group differences did emerge on a structured task reflecting salience of specific traits. Individuals with agoraphobia reported that they would be more attentive to the dimensions supportive-critical and enabling-bossy. This is consistent with current conceptualizations of agoraphobics as individuals who do not feel they can cope with the dangers of the outside world, and are compelled to seek help from a 'caregiver' (Beck & Emery, 1985). Social phobics reported that they would not be particularly attentive to any of the traits. This may be a result of socially phobic individuals self-focused attention.  iii Table of Contents Abstract  ii  List of Tables  v  Acknowledgements  vii  Description of Social Phobia  1  Literature Review of Social Phobia and Agoraphobia  2  Introduction  19  Comparison Group  22  Subjects  24  Methods  26  Dependent Measures  26  Supplementary Measures  29  Results  33 Overview  33  Demographical data  33  Social Phobia and Anxiety Inventory  35  Other Clinical Measures  40  Main Analyses: Accessible and Salient Constructs  44  Discussion  54  Appendix 1: Pilot Study  65  Appendix 2: Impression of Others Questionnaire  67  Appendix 3: Others Questionnaire  68  iv Appendix 4: Background Sheet  70  Appendix 5: Statistical Assumptions  72  Bibliography  73  V  List of Tables  Table  1.  Between-group  comparability  of  demographic  characteristics  34  Table 2. The Means and Standard Deviations of the Social Phobia and Anxiety Inventory (SPAI) and it's two subscales, Social Phobia (SPSS) and Agoraphobia (ASS) Table  3.  Between  group  36 comparability  on  Clinical  Background  Factors  38  Table 4. Means and standard deviations on clinical measures for Diagnostic Groups  41  Table 5. Means and standard deviations of the two factors of the Agoraphobic Cognition Questionnaire: ACQ-physical and ACQ-social/behavioural for Diagnostic Groups  43  Table 6. Number of subjects in each diagnostic group who nominated traits under the hypothesized dimensions on the Impression of Others Questionnaire: Part A  45  Table 7.Number of subjects in each diagnostic group who nominated traits under the hypothesized dimensions on the Impression of Others Questionnaire: Part B Social Phobic Situations  46  vi Table 8. Number of subjects in each diagnostic group who nominated traits under the hypothesized dimensions on the Impression of Others Questionnaire: Part B Agoraphobic Situations  . .  47  Table 9. Means and standard deviations of bipolar dimensions on the Others Questionnaire by diagnostic group  49  Table 10. Means and standard deviations by diagnostic group for the mean of a total score for all items on the Others Questionnaire  52  Table 11. Means and standard deviations of traits on the Others Questionnaire by diagnostic group  53  Table 12. The number of subjects from each group who nominated traits under the following dimensions  66  vii Acknowledgements  I gratefully acknowledge the following individuals for their unique contributions. My committee members: Lynn Alden, advisor and committee chair, for her guidance, time, and advice in all stages of the research, and Dan Perlman and Ben Dykman for their clear constructive feedback, and support throughout the project. I am also indebted to Peter McLean and William Koch from the Health Psychology Clinic at the University Hospital (UBC Site), and Jim Quinn from the Burnaby Psychiatric Unit program for agoraphobics, for allowing information about the study to be distributed to participants in their program. I would also like to express my appreciation to my family and friends for their support: Peter Rees, Charlotte Molnar, and Ian Sutherland for offering their time and interest, my sister Kathleen Capreol for becoming chief editor, my parents for their ongoing caring, and in particular George Molnar because with his love and support I feel that anything is possible.  1 I. Description of Social Phobia  Social Phobia is defined as "a persistent fear of one or more situations (the social phobic situations) in which a person is exposed to possible scrutiny by others and fears that he or she may do something or act in a way that will be humiliating or embarrassing" (American Psychiatric Association [APA], 1987, p.241). The remaining criteria for a DSMIII-R diagnosis of social phobia include the following: 1) that exposure to the specific phobic stimulus almost invariably provokes an anxiety response; 2) that the situation will be avoided or endured with intense anxiety; 3) that avoidance interferes with the social or occupational, functioning or there is a marked distress over having the fear; and 4) that the person recognizes that the fear is excessive or unreasonable.  Studies have found that social phobia onsets in the mid to late teens (Ost, 1987). The DSM-III-R (APA, 1987) reports that individuals with social phobia may be prone to alcohol, barbiturate or anxiolytic use. In addition, when social or occupational functioning is severely impaired, a depressive disorder may be a complication. Prevalence of this disorder has been reported to be approximately 2% of the general population (Myers et al., 1984). This may be an underestimate since individuals with social anxiety may present for treatment of alcoholism or substance abuse rather than anxiety.  2 II. Literature Review of Social Phobia and Agoraphobia  A. Diagnostic issues  Agoraphobia is defined as the fear of being in situations from which escape might be difficult (or embarrassing) or in which help might not be available in the event of a panic attack (APA, 1987). The diagnostic categories for individuals with agoraphobic symptoms have changed between DSM-III and DSM-III-R. Before the revision, the diagnoses given were agoraphobia with or without panic disorder. Most of the articles reviewed in this proposal use this earlier diagnostic system. DSM-III-R diagnoses are panic disorder with or without agoraphobia and agoraphobia without a history of panic disorder. The latter diagnosis is thought to be very rare. In this discussion, the term agoraphobia will stand for individuals with a history of both panic disorder and agoraphobia unless otherwise stated.  There is a degree of comorbidity between agoraphobic and social phobic symptoms. Amies, Gelder, and Shaw (1983) found that 50% of agoraphobic patients experienced anxiety in social situations, and 25 % of social phobics experienced anxiety in agoraphobic situations, but in both cases at significantly lower intensity than their primary complaint. Earlier studies also reported many individuals warranting both diagnoses using DSM-III criteria. Solyom, Ledwidge, and Solyom (1986) found that 55% of agoraphobics suffered from clinically significant social phobic symptoms as rated by a psychiatrist, and 30% of social phobics suffered from clinically significant agoraphobic symptoms. Unlike previous studies, Cottraux,  3 Mollard and Duinat- Pascal (1988) found that there was little overlap between categories; In their sample, only 12.5% of agoraphobics had secondary diagnosis of social phobia, and 7% of social phobics had a diagnosis of agoraphobia and 6% of panic disorder. Cottraux et al. (1988) followed the recommendations of Liebowitz, Gorman and Fyer (1985) , as well as DSM-III criteria, when making their diagnoses. Liebowitz et al. (1985) outlined that although agoraphobics can develop social fears, a secondary diagnosis of social phobia was only warranted if there were signs of avoidance of social situations linked to the fear of scrutiny, and humiliation, clearly independent of the fear of having a panic attack in front of people. When a patient is socially phobic, he or she should display, besides fear of scrutiny and humiliation, an avoidance of multiple situations linked to the fear of having a situational panic attack, to be qualified for a secondary diagnosis of agoraphobia. Liebowitz et al.'s (1985) recommendations have been incorporated into the DSM-III-R.  B. Interpersonal Issues  The fundamental problem of social phobics is interpersonal in nature, since the main focus of their fear is interacting with other individuals. They are consistently reported to be nonassertive (Cottraux et al., 1988), hypersensitive to criticism and hypervigilant to possible social threats (Liebowitz et al., 1985). Beck and Emery (1985) postulate that the socially anxious individual thinks that (italics are their own) " every action is observed by a crowd of evaluators and appraised as clumsy or skilful, and . . .is judged according to . . .[their] confidence and competence" (p. 146). Beck and Emery state that the italicized words represent  4 some of the crucial psychological aspects of evaluation anxiety. Beck and Emery describe the social phobic as encompassing "the notion of a child being subjected to evaluation by adults " (p. 153), and that "other person or persons are involved in paying attention to the 'child'" (p. 154). Some empirical research has looked at the importance of observation and evaluation for social phobics. In a clinical study of severely socially anxious individuals, Nichols (1974) reported that a sense of being watched, a heightened awareness and fear of being evaluated and judged by others, and a fear of situations where the individual was likely to attract attention, were observed. Research has also indicated that socially anxious individuals tend to use a dimension of 'chance of being evaluated' more than those nonanxious (Goldfried, Padawer, & Robins, 1984). This result was not replicated in a follow up study (Robins, 1987). Socially phobic individuals also are concerned with the results of the evaluation, and consistently report fear of negative evaluation (Liebowitz et al., 1985). Beck and Emery (1985) postulate that the attitude of the potential evaluator is very important to the social phobic. Beck and Emery hypothesize than when an individual who is socially phobic enters a social situation, he/she assesses whether the evaluators are accepting and empathetic or rejecting and aloof. Research shows that social phobics are more likely to monitor cues for hostile appraisal, and are more likely to predict, perceive, and recall negative appraisals from others (Halford & Foddy, 1982; Mathews & MacLeod, 1989; Smith, Ingram, & Brehm, 1983). A study by Leary, Kowalski and Campbell (1988) indicated that this may be due to a generalized expectation that others are rejecting and critical. The researchers asked socially anxious and nonanxious individuals to report how they thought an observer would rate them,  5 and another person. They found that the socially anxious individuals not only thought that they would be rated more unfavourably, but that the other person would be as well. Theorists have postulated evolutionary explanations for the interpersonal threat experienced by socially anxious individuals. Ohman (1986) hypothesizes that social fears originate in a dominance/ submissive system. Social fear is one pole of the system, with social dominance at the other end. Social fears are learned in certain individuals, when signs of dominance are paired with an aversive outcome. Trower and Gilbert (1989) also address the evolutionary significance of dominance, but stress the importance of the hostile dominant. They theorize that social anxiety arises from the activation of an evolved mechanism for intra-species threat. In their model, social phobics are seen as on the outlook for social cues that may indicate hostile appraisal of his self presentation. Socially anxious individuals tend to "perceive others as hostile dominants, . . . fear negative evaluation from them and . . . respond, at one level of this disorder, by appeasement and submissive behaviours, and at a more severe level, by primitive actions such as escape or avoidance" (p. 19). Beck and Emery (1985) also theorize that dominance/ submission is an important dimension for those socially anxious. They state that when the evaluator is perceived as more dominant, then submissive behaviours are likely to be mobilized. Related dimensions that Beck and Emery hypothesize that socially anxious individuals use are status, and authority.  Agoraphobia also involves potentially important interpersonal issues: 1. Hypersensitivity to criticism is associated with and may contribute to the development of agoraphobic symptoms in individuals with panic disorder (de Ruiter &  6 Garssen, 1989; Pollard & Cox, 1988). Pollard and Cox (1988) found that agoraphobics scored significantly higher on the Willoughby Personality Schedule, which is a measure of social evaluative anxiety, than panic disorder patients without agoraphobia. Similar results were found using the Interpersonal Sensitivity subscale of the SCL-90-R that measures the extent to which a person worries or is sensitive to being rejected by others (Brown, Munjack & McDowell, 1989; de Ruiter & Garssen, 1989).  2. Dependence issues  Beck and Emery (1985) have proposed that agoraphobics' conflict revolves around issues of dependency, autonomy, and control. They propose that they have had a "lifelong concern about their health or ability to manage their emotions, but have managed to maintain their equilibrium as long as they had available one or more protective figures (parents, siblings, peer group)" (p. 134). Because the individual does not feel able to cope with the dangers of the outside world, they are impelled to seek help from a "caregiver". Beck and Emery state that contact with a caregiver brings relief because it promises access to prompt treatment if help is needed. The caregiver also serves as a reality check and helps evaluate the seriousness of the situation. Beck and Emery describe the disorder as resembling "the child who has been placed in a strange place for the first time", and that other people will "ignore him even to the point of not caring whether something disastrous happens to him" (p. 154). Roth and Argyle (1988) hypothesize that the helpless, dependent manner in which agoraphobics rely on relatives and family members might be an important personality trait  7 that predates the onset of the disorder. This is compatible with Beck and Emery's conceptualization of agoraphobia.  Although an individual with agoraphobia seems to want a caregiver, seeking support from another person leaves the agoraphobic open to be controlled by that person. According to Beck and Emery, in a typical scenario the individual perceives herself/ himself to be suppressed by the person on whom she/he depends for support. Beck (1983, cited in Beck and Emery, 1985) found that agoraphobics reported a greater investment in mobility and self direction, and sensitivity to being restrained or controlled, than generalized anxiety disordered individuals, depressives, and normals. Goldberg (1986) describes how the need for dependency and autonomy might interact. He discusses two sources for the fear of becoming helpless: "one, being helpless goes against his strong need to be independent and does not fit his idealized self; two, what he really fears is that his need to be nurtured will not be satisfied, and he will simply remain in a helpless state" (p. 145).  Beck and Emery (1985) predict that loss of a caregiver through separation or death could precipitate agoraphobic symptoms. The DSM-III-R (APA, 1987) states, "Separation Anxiety Disorder in childhood and sudden loss or disruption of important interpersonal relationships apparently predispose to the development" of panic disorder with agoraphobia. There have been anecdotal reports supporting this contention, but empirical research has been limited (see Zitrin & Ross, 1988 for discussion). Deltito, Perugi, Mareminni, Mignani,and Cassano (1986) found that 60% of patients diagnosed as agoraphobic reported a history of  8 school phobia in childhood (usually a sequel to separation anxiety), in comparison to no separation anxiety found in panic disorder without agoraphobia. Zitrin and Ross (1988) found significantly more separation anxiety for female agoraphobics, compared to a sample of simple and social phobics. They did not find a significant difference for male agoraphobics. Amies et al. (1983) also found that agoraphobics were significantly more likely to say that the death of a loved one was the cause of their problems than were social phobics. On the other hand, researchers have not found differences for parental losses or severe family disruption during childhood (Thyer, Nesse, Cameron, & Curtis, 1985; Zitrin & Ross, 1988).  Current dependence issues have also been investigated. Agoraphobics are observed to display less acute avoidance behaviour when accompanied by a trusted person (Rachman, 1983). Reich, Noyes and Troughton (1986) found a close association between agoraphobia and dependent personality disorder . Buglass, Clarke, Henderson, Kreitman and Presley (1977), however, found that only 27% reported an awareness of and resentment of dependency. They theorized that this number was not higher because an individual with agoraphobia had a tendency to disguise their dependence on others. On the other hand, the association with the personality trait of dependency has not always been found. King, Bayon, Clark, and Taylor (1988) looked at a population of patients with and without agoraphobia, and did not find any association with dependency. They did find that they were significantly more avoidant, borderline, and neurotic, and significantly less sociable than controls. Because it was a population of panic disordered patients with or without agoraphobia, the association with dependency might have been weakened.  9 Chambless and Mason (1986) found that sex-role inventory measures of masculinity were inversely related to severity for both male and female agoraphobics. Fodor (1974) has argued that sex role stereotyping predisposes women to agoraphobia. Women are allowed to be more fearful and taught to perceive themselves as incompetent and helpless without male assistance. The masculinity subscale measures Instrumentality (e.g., active, superior, but without social undesirable aspects of dominance). This relationship however may be easier explained by lack of assertiveness than dependency needs.  C. Cognitive factors  There are some similarities in the cognitions of various types of phobic individuals (Mizes, Landorf-Fritsche, Grossman & McKee, 1987). As a group, phobics' irrational beliefs centre around approval and disapproval from others and a tendency to avoid problematic situations. To a lesser extent, a tendency to anticipate disastrous consequences was found. Differences have been found in the specific content of the phobic images of agoraphobics and social phobics. Cook, Melamed, Cuthbert, McNeil and Lang (1988) found that social phobics' scripts centres on fears of social performance, and being the centre of attention. In addition, there was evidence in the scripts of hypervigilance to threat, as well as general worry and frustration. Agoraphobics, on the other hand, had scripts that focused on the panic experience as well as the sensation leading up to a panic episode. There were also themes of isolation, entrapment, and loss of control. Hope, Rapee, Heimberg and Dombeck (1990) compared individuals with social phobia  10 to those with panic disorder without agoraphobia using a revised Stroop Colour Naming Task. They found that social phobics experienced more interference when social threat words were used, while panic disordered individuals showed more interference with physical threat words. They concluded that this result supported the theory that social phobics are hypervigilant to social-evaluative threat words. They did not look at individuals with significant agoraphobic symptoms.  D. Demographic Factors  Amies et al. (1983) found that social phobics reported an earlier age of onset (19 years in comparison to 24). In the study, both groups had approximately one decade of ongoing symptoms before seeking treatment. Many studies have confirmed this general finding (for example, Solyom et al., 1986; Cameron, Thyer, Nesse & Curtis, 1986; Ost, 1987). Amies et al. (1983) also found that social phobics were younger at referral (mean of 30.7 compared to 37.2), and had a higher proportion of males in the sample (males:females60:40 compared to 14:84). In addition, social phobics were found to have come from a higher social class on average. They were also less likely to have been married (5% as compared to 32%). Other researchers (Persson & Nordlund, 1985; Solyom, Ledwidge & Solyom, 1986) have confirmed these results. Social phobics have also been found to have a higher educational and occupational status (Persson & Nordlund, 1985; Solyom et al., 1986), and higher scores on verbal IQ, and higher SES of parents (Persson & Nordlund, 1985). Persson and Nordlund (1985) found that agoraphobia was associated with mother's working  11 outside the home during patient's childhood, but this is probably due to a confound with lower SES.  E. Symptom Profile  Somatic Symptoms Amies et al. (1983) found that individuals with agoraphobia and social phobia reported different somatic symptoms. Social phobics reported more blushing, and a tendency to report more muscle twitching, while agoraphobics reported weakness in the limbs, difficulty breathing and dizziness, and faintness, as well as a tendency to report fainting and ringing in the ears. Cameron et al. (1986) gave the Anxiety Symptom Differential Questionnaire to a variety of anxiety disordered patients and found that agoraphobics were high on severity of reported anxiety symptoms while social phobics were intermediate.  Panic Attacks Cameron et al. (1986) found that when comparing acute panic symptoms from DSMIII anxiety disorders, agoraphobics reported more panic symptoms than social phobics. In a comparative clinical and psychometric study, Cottraux, Mollard and Duinat-Pascal (1988) also found a higher panic frequency in agoraphobics, and found that psychometrically this was the principal factor differentiating agoraphobic from social phobics.  12 Other symptoms Amies et al. (1983) found that the groups did not differ on anxiety, or obsessional symptoms. Solyom et al. (1986) also found no difference in the general level of anxiety symptoms measured by the IP AT Anxiety Scale Questionnaire (Cattell & Sheier, 1963). Cottraux et al. (1988) also found comparable levels of generalized anxiety and avoidance. In contrast, Cameron et al. (1986) found that agoraphobics were more anxious, using the StateTrait Anxiety Inventory. Depersonalization and derealization was more frequent and more severe for agoraphobics (Amies et al., 1983). Persson and Nordlund (1985) also found that agoraphobics reported more depersonalization. In addition, simple phobias were found more often with agoraphobia (Cottraux et al. 1988).  Specific Questionnaires  Fear Survey Schedule Amies et al. (1983) found that social phobia and agoraphobia could be differentiated by their main complaint. Yet, agoraphobics did report anxiety in social phobic situations and vice versa (50% and 25% respectively). However, subjects showed significantly less anxiety in situations "belonging" to the other group than their main complaint. Interestingly, the rank order of social phobia situations by agoraphobics is the same as social phobic patients, but not the reverse. One interpretation of this is that the social anxiety felt by agoraphobics is of the same nature as social phobics, but only differing in degree. Cerny, Himadi, and Barlow  13 (1984) looked at the number of fears reported and found that agoraphobics reported a greater number of fears than social phobics, who reported more fears than simple phobics.  SCL-90R Munjack, Brown, and McDowell (1987) compared social anxiety in patients with social phobia and panic disorder without agoraphobia and found differential response patterns on the SCL-90R. Individuals with social phobia reported more interpersonal sensitivity, while panic patients reported more somatization. The authors concluded that the findings strengthen the belief, shared by Liebowitz, that there are fundamental differences between the social anxiety experienced by primary social phobics and panic disordered patients who report social anxiety. However, they did not compare social phobic to panic disordered with agoraphobia individuals. Since panic disordered individuals with agoraphobia report more interpersonal sensitivity than those without, how they compare to social phobics is still unclear.  Alcohol Use There are conflicting reports in the literature about the comparative degree of drug use in social phobics and agoraphobics. Amies et al., 1983 found that excessive alcohol use was more common for social phobics (20%-7%). On the other hand, Thyer, Parrish, Himle, Cameron, Curtis and Nesse (1986) compared anxiety disorders and found alcohol abuse to be greatest in agoraphobics, although the level was not significantly higher than for social phobics.  14 Depression and Suicide Depressive symptomatology is associated with both social phobia and agoraphobia (Amies et al., 1983). There is not a clear trend for depression to be differentially associated with one condition. For example, Amies et al. (1983) did not find differences in reported depressive symptoms, but did find that social phobics reported a higher rate of parasuicidal acts (14% compared to 2%). Solyom et al. (1986) also found no differences in the level of depression measured by self and psychiatric ratings. In contrast, Heimberg and Barlow (1988) reported that agoraphobics appeared more depressed than social phobics.  Impairment Social phobics have been found to report more occupational and educational difficulties, but fewer domestic problems than agoraphobics (Solyom et al., 1986). This may be due to a confound with sex distribution.  F. Personality Measurements  Extraversion Amies et al. (1983) found that social phobics had significantly lower extraversion scores on the Eysenck Personality Inventory than agoraphobics, whose scores were similar to normals. Oei, Gross and Evans (1989) found the same results using the extraversion dimension of the Maudsley Personality Inventory.  15 Assertion Cottraux et al. (1988) found that social phobics had very low assertion scores on the Rathus Assertion Schedule. Lack of assertion was found to be the principal factor that psychometrically characterized social phobics in comparison to agoraphobics.  Minnesota Multiphasic Personality Inventory (MMPI) Cottraux et al. (1988) found that social phobics had a more severe neurotic profile, with significantly higher elevations on the psychasthenia, and anxiety scales.  Willoughby Personality Schedule (WPS) Social phobics were found to have significantly higher scores on this measure of social anxiety than other diagnostic groups (simple phobia, sex disorders, obsessive compulsives, and agoraphobia) (Turner, Meles & DiTomasso, 1983). However, 68% of agoraphobics and obsessive compulsives did have clinically significant WPS scores.  Dutch Battery (van Zuuren, 1987) Agoraphobic women were characterized by a strong field dependence, neurosomatism, self-sufficiency, rigidity, and to a lesser extant low intelligence, defensive attitude, inadequacy and femininity. Agoraphobic men were only mildly characterized by low self esteem, low intelligence, and by feeling externally controlled. Socially phobic women were characterized by social anxiety, social inadequacy, low self esteem and high intelligence. Socially phobic men shared these as well as being characterized by introversion.  16 G. Background Factors  Perceived parenting styles for agoraphobics and social phobics have been studied among outpatients (Arrindell, Emmelkamp & Monsma, 1983) and inpatients (Arrindell, Kwee, Methorst, Van der Ende, Pol, & Moritz, 1989). The pattern of results was the same, only the magnitude of correlations differed. In comparison to nonanxious controls, socially phobic patients rated both parents as more rejective, less emotional warm, and as having been more over protective. Agoraphobics rated both parents as having been less emotionally warm than nonanxious controls, but only there mothers as rejective. When comparing social phobics to agoraphobics, there were significant differences with the social phobics assigning more negative ratings. These included rating both parents as more rejective, rating both parents lower in affection ratings. Similarly, Amies et al. (1983) found that social phobics reported that their fathers were more dominant, and that they had unsatisfactory relationships with them. It is important to note that these are perceived patterns, and might be a result of memory distortions.  Persson and Nordlund (1985) found that there was a trend for social phobics to perceive conflicts with their parents as a cause of their disorder more often than agoraphobics did. Agoraphobics, on the other hand, perceived the experience of someone's death as a cause of their disorder significantly more than social phobics.  17 H. Family History  Reich and Yates (1988) compared the family history of emotional disorders in the relatives of panic disordered, nonanxious, and socially phobic individuals. Social phobics had significantly more relatives with social phobia than panic disordered individuals, and a trend more than normals. Relatives of panic disordered individuals had significantly more diagnoses of panic disorder, alcohol abuse and generalized anxiety disorder. Incidence of major depressive disorder in relatives is complicated by the high degree of depressive symptoms in the probands.  I. Biological Challenge Studies and Drug Response  Differences have been found in biological challenge studies (Liebowitz et al., 1984). These studies involve exposing individuals to different chemicals, and observing if the agents precipitate a panic in the subjects. For example, social phobics have been found to panic less in response to sodium lactate (Liebowitz et al., 1984). Differential response to drug treatment has also been reported (see Liebowitz, M.R., Campeas, R., Levin, A., Sandberg, D. & Papp, L., 1987 for discussion). This data must be taken cautiously due to many methodological problems found in the research. Clinical studies show that agoraphobics respond well to tricyclics and MAO inhibitors, but not to betablockers, while the pattern for social phobics is not as clear. Recent research points to MAO inhibitors being more effective for social phobics than beta-blockers, and tricyclics not being  18 effective at all (Liebowitz et al., 1987).  J. Summary of Comparative Studies Until the 1980 DSM-III, social phobia was not considered a discrete disorder and many were diagnosed as agoraphobics (Liebowitz et al., 1985). Amies et al. (1983) were the first to empirically compare individuals who met the criteria for social phobia or agoraphobia. Since then, other comparative studies have been published. This empirical research supports the distinction, in clinical subjects, between social phobia and agoraphobia. For example, social phobics and agoraphobics have been found to differ on demographic factors, somatic and panic symptoms, personality measures of extraversion and assertion, family history and background, biological challenge reactions and drug response. Less consistent differences have been found in alcohol use and level of depression. Interpersonal issues have not been directly compared.  19 III. Introduction Current conceptualizations of social phobia propose that it is, at least in part, cognitive mediated (Butler, 1990). If these theories are correct, one would expect to be able to find cognitive phenomena that uniquely characterize social phobics. This study was designed to compare the social cognitions of social phobics, agoraphobics and normals. Of particular interest were the interpersonal constructs that these groups use to process social information. One question was whether the concepts suggested in clinical descriptions of social phobia and agoraphobia would emerge as chronically accessible constructs (Higgins, King, & Mavin, 1982) in processing social information. According to theories of social cognition (Wyer & Srull, 1986),, individuals screen incoming information from the environment using certain cognitive constructs. "Construct systems can be considered as a kind of scanning pattern which a person continually projects upon his world. As he sweeps back and forth across his perceptual field he picks up blips of meaning (Kelly, 1955, p. 145). " People typically use personally relevant dimensions when processing information. The dimensions that are habitually used are defined to be chronic constructs. Since an individual's construct system develops from their particular history of social interactions, personally relevant dimensions may differ. Individual differences have been found in the particular kinds of constructs that are actually present in the memory (Markus, Smith & Moreland, 1985). This is defined as construct availability. It is also possible for individuals to possess many of the same constructs but differ in the readiness with which each construct is used in information processing (Higgins et al., 1982). This is defined as construct accessibility.  20 Researchers have found individual differences in the accessibility of the social constructs individuals use to evaluate others (Bargh, Bond, Lombardi & Tota, 1986; Higgins et al., 1982; Markus et al., 1985). There is also evidence there is considerable stability over time in an individual's chronic constructs (Dornbusch, Hastorf, Richardson, Muzzy & Vreeland, 1965). Researchers have grouped individuals according to shared accessible constructs, and demonstrated that the particular type of construct influenced information processing (Bargh et al., 1986; Higgins et al., 1982; Markus et al., 1985). For example, individuals who had an accessible construct of shyness were more likely to see evidence of shy behaviour when given ambiguous vignettes to rate (Bargh et al., 1986). Socially anxious individuals may process information about others and perceive their social environment in a manner that contributes to or maintains their anxieties. Cognitive theory predicts that the use of different interpersonal constructs will affect the manner in which social information is processed. The accessible chronic constructs of socially phobic, panic disordered with agoraphobia and nonanxious subjects were compared. Thematic similarities in the chronic interpersonal constructs of these preselected groups were contrasted. There has been little research investigating the interpersonal factors involved in social phobia. Leary et al. (1988) did find that socially anxious individuals perceived that others would rate them, as well as another person, more negatively than nonanxious subjects. This could be explained by a generalized perception that others are critical, and negatively evaluative. If individuals with social phobia anticipate others to be more critical, the inappropriate fear shown would be a logical result of that prediction. Goldfried, Padawer, and Robins (1984) also found that evaluation was salient for those socially anxious, and found  21 that they tended to use a dimension of 'chance of being evaluated' more than those nonanxious. This result, however, was not replicated in a follow up study (Robins, 1987). Other interpersonal factors have been proposed, but have not yet been empirically tested. Trower and Gilbert (1989) hypothesize that socially anxious individual tend to construe relationships as dominance ranked, fear negative evaluation from hostile dominants, tend to perceive others as hostile dominants, and tend to respond self defensively with submissive behaviour (Trower & Gilbert, 1989).  Beck and Emery (1985) have also  speculated that dominance and relative status are important dimensions for social phobics. A pilot study suggests that other dimensions might be important for those higher in social anxiety (see Appendix 1). The dimension that socially anxious subjects used in common when describing others was trustworthy-untrustworthy. This included two subdimensions of trustworthiness: sincerity (honesty, genuineness) and responsibility (reliability). For this study, it was hypothesized that there will be differences between anxious and nonanxious individuals in the thematic content of their chronic constructs. It was also anticipated that the thematic content of the social phobics' chronic constructs would be different from agoraphobics. From the literature, it was postulated that constructs that involve dimensions of criticism, dominance, and evaluation would be more accessible for socially anxious individuals. It was also hypothesized the dimensions suggested by the pilot study, sincerity and responsibility, would be more accessible.  1  22  IV. Comparison Group  Agoraphobics make an interesting comparison group for social phobics. Agoraphobia and social phobia have many fundamental issues in common since they are both phobic disorders. There is also current interest in the interpersonal sensitivity of agoraphobics. On the other hand, there is debate in the literature surrounding dependency issues for agoraphobics, a concern that is not shared by social phobics. Past dependency issues include an association with separation anxiety in children and adolescents (Zitrin & Ross, 1988). Current dependency issues include reliance on certain support givers, and an association with the personality trait of dependence (Reich et al., 1986). Beck and Emery (1985) also describe how the dimension of dependence is interrelated with control, and autonomy. The agoraphobic is seen as reluctant to get too close to a caregiver, lest they be dominated, or too far, lest they encounter a situation where they need help. The phobic images of agoraphobics have scripts that have themes of isolation, entrapment, and loss of control (Cook et al., 1988). The pilot data suggested similar, but slightly different in essence, content to the accessible constructs (see Appendix A). The traits that individuals higher in agoraphobic fears tended to use the dimension egotistic-nonegotistic when describing other people. This included two subdimensions, selfish-considerate and conceited-modest. This suggests a theme of being aware of and focusing on the extent to which an individual thinks of others or only thinks of themselves. This could be related to Beck and Emery's conceptualization of an agoraphobic as an individual who is afraid that they will be ignored if they need help. The  23 thematic content of the chronic constructs of agoraphobics was hypothesized to more likely include dimensions of bossiness, control, abandonment, and safety. From the pilot study, it was also postulated that dimensions of selfish-considerate and conceited-appreciative would be more accessible. 3  24 V. Subjects  Subjects for the two clinical groups were recruited from two psychiatric outpatient programs: Health Psychology Clinic at University Hospital (UBC Site), and the Burnaby Psychiatric Unit outpatient program (Se-Cure) for agoraphobics. Individuals were considered for the study if they had received a DSM-III-R diagnosis for social phobia or panic disorder with agoraphobia from their referring psychiatrist. Twelve individuals were recruited from the Health Psychology Clinic: Nine social phobics (6 men, 3 women) and three agoraphobics (2 men, 1 women). All socially phobic individuals reported a generalized pattern of social fears. Six individuals with agoraphobia (6 women) were recruited from Se-Cure. Demographic characteristics of the subjects can be seen in Table 1. Control subjects were recruited from university classes. To be included, control subjects had to meet 4 criteria: a) no history of psychiatric inpatient treatment, psychotropic medication, or outpatient treatment for social phobic or agoraphobia, b) received scores on the revised Anxiety Disorder Interview Schedule (ADIS-R) (DiNardo, O'Brien, Barlow, Waddell and Blanchard (1983) that confirmed they did not meet diagnostic criteria for agoraphobia or social phobia, c) no history of panic attacks, and d) matched clinical subjects on demographic qualities (age, sex, marital status, and years in school). The ADIS-R is a structured interview that is designed to facilitate differential diagnosis among DSM-III-R anxiety disorders, while ruling out psychosis, substance abuse, and major affective disorders. The inter-rater reliability ( K coefficient) for social phobia is .77 and Agoraphobia with panic attacks is .86 (DiNardo et al., 1983).  25 As a check on diagnostic status, the Social Phobia and Anxiety Inventory (SPAI) was administered to all subjects (Turner, Beidel, Dancu & Stanley, 1989). Mean scores for each group can be seen in Table 2. This instrument is designed to discriminate between social phobia and agoraphobia. This 45 item inventory has two subscales. The social phobia subscale has 32 items on the cognitive, somatic, and behavioural dimensions of social fear. The agoraphobia subscale includes 13 items based on DSM-III criteria for Agoraphobia. The total score for the SPAI is derived by calculating the difference between the two subscale scores. The Agoraphobia subscale serves as a suppressor variable to control for complaints of social anxiety that are only part of the larger clinical picture for agoraphobia. A cutoff score of 80 was recommended to minimized false positives and negatives. Using a Pearson 1  correlation, test-retest reliability for the SPAI done at a two week interval was .86. The instrument demonstrated good internal consistency for the social phobia and agoraphobia subscales (Cronbach's alpha = .96 and .85, respectively).  Three individuals recruited from Se-Cure with psychiatric diagnoses of agoraphobia had SPAI scores over 80. Ideally these individuals would not have been included. The ADIS-R, however, was administered to these subjects and confirmed the diagnosis of agoraphobia and not social phobia. They were included in the agoraphobic sample to insure adequate n's for statistical analysis.  26 VI. Methods  A. Dependent Measures  Unstructured measurements of Construct Accessibility Social perception was assessed by an unstructured task (Impression of Others Questionnaire) consisting of two parts. In part A, chronic construct accessibility was measured using the free response measure derived by Higgins, King and Mavin (1982) (see appendix 2). This procedure involved asking subjects to list up to 10 traits that best describe the following: (a) a type of person they sought out, (b) a type of person they avoided, (c) a type of person they liked, (d) a type of person they disliked, and (e) a type of person they frequently encountered. Each question was presented on a separate sheet of paper, and the ordering of the five questions was counterbalanced across respondents. In accordance with Higgins et al. 's (1982) operationalization of construct accessibility in terms of output primacy, a given respondent's chronically accessible constructs were defined as those given first to each of the four affect questions (like, seek out, dislike, avoid), and first and second on the frequency question. In part B, the same procedure was used except the subjects were asked to nominate traits in situations that were associated with anxiety. There may be pertinent constructs that are readily accessible when the individuals are in situations they find anxiety provoking. These constructs might have clinical significance for the maintenance of anxiety. These situations were derived from clinical criteria in the ADIS-R for agoraphobia or social phobia  27 (see appendix 2). Two situations were chosen to be relevant for agoraphobics: a) accompanying you to a shopping mall, and b) travelling with you on a bus. Two of the situations were chosen to be relevant for social phobics: c) discussing different points of view with, and d) accompanying you to a party. Scoring Procedure: The measurement was scored by the investigator. Responses from both Part A and Part B (social phobic situations and agoraphobic situations) were clustered according to thematic dimensions. The traits nominated by the groups were grouped by identical or synonymous matches and their verbal opposites (bipolar dimensions). Word synonyms and antonyms were determined using a thesaurus. The proposed hypotheses were tested by comparing the proportion of individuals from the different groups that demonstrated a particular accessible thematic dimension. Bipolar dimensions were used for two reasons.  First, there is evidence for the  viability of bipolar dimensions from the field of linguistics. The assumption that thinking in terms of opposites is 'natural' for the human species has long been shared by linguists (Osgood, 1990). Osgood (1990) states that the "notion of logical opposition has always had a fundamental and primitive status in Western philosophical thought" (p. 229). Osgood (1990) investigated the use of verbal opposition across different languages and cultures and found that individuals share a tendency to "utilize meaningful opposition as a pillar of their logical constructions" (p. 229). Osgood, Suci and Tannenbaum (1957) used scales anchored by verbal opposites (bipolar scales) to divide semantic space under 3 empirically derived factors. Second, in the analysis of the pilot data, it was also observed that subjects would tend to use  28 both a trait and it's verbal opposite in the Higgins et al.'s (1982) procedure (i.e. subjects would say they liked people who were honest, and disliked those who were dishonest).  Structured measurement of Salient Features of Others  The Others Questionnaire was designed to straightforwardly assess the most salient traits used by individuals when they interact with another individual (see appendix 3). Subjects were asked to consider traits (with a behavioural definition included), and asked to rate how likely they were to notice or pay attention to these traits when they were interacting with another person. This was done using a 10-point scale (1 = would definitely pay attention to; 10= would not pay attention to). There were 12 trait words and definitions to be rated. These were grouped in pairs of antonyms, giving 6 trait dimensions. Salience of the traits was operationalized as the rating scores of the trait words, with higher ratings corresponding to more salience. Of these 6, two of the trait dimensions were derived rationally according to current hypotheses on the most important interpersonal traits for social phobia and agoraphobia. The traits mentioned in the literature was 'critical' for social phobics and 'bossy and supportive' for agoraphobics. Although the literature does not talk in terms of bipolar dimensions, verbal opposition has been found to be important for sematic meaning (Osgood, 1990). For this reason, the verbal opposite of the traits was added to allow bipolar dimensions to be compared. The two dimensions were supportive-critical and enabling-bossy. The four remaining trait dimensions were suggested by a pilot study with University students (see  29 Appendix 1). The two dimensions that were related to social fears in this student sample were sincerity-insincerity and reliability-unreliability. The two dimensions that were related to agoraphobic fears were selfishness-considerateness and conceit-modesty. Scoring Procedure: With the structured measure, the rating scores for pairs of antonyms were added to provide a summary score that represented the bipolar dimension. These scores were analyzed in a one way (groups) MANOVA. Hypotheses derived from the pilot study and those derived from the literature were tested.  B. Supplementary Measures  Beck Depression Inventory (BDI) Both agoraphobic and socially phobic individuals are known to experience negative affect. The BDI was administered so that the relative amount of depressive symptomatology could be compared to see if there were any differences between the groups that could have been responsible for the results obtained regarding social constructs. The BDI is a 21-item inventory that assesses cognitive, somatic, and behavioural symptoms of depression (Beck, Ward, Mendelsohn, Mock, & Erbaugh, 1961). The BDI has been shown to have acceptable psychometric properties (Beck, Steer, & Garbin, 1988). Testretest reliabilities range from .48 to .86 depending on the interval between retesting and type of population, and measures of internal consistency of .73 to .92 (Beck, et al., 1988). Concurrent validity is suggested by high to moderate correlations (.55 to .96) with clinical  30 ratings for psychiatric patients, and moderate correlations with similar scales that also measure depression (i.e. Hamilton Psychiatric Rating Scale for Depression (.73) (Beck et al., 1988).  The Body Sensations Questionnaire (BSQ)  The BSQ and it's companion questionnaire, Agoraphobics Cognitions Questionnaire (described below), are inventories that have been developed to measure two components of what has been labelled 'Fear of Fear' (Chambless, Caputo, Bright and Gallagher, 1984). 'Fear of fear' is the fear of panic attacks and their real or imagined consequences. These measures have been found to be an important distinguishing characteristic among clients with anxiety disorders (Chambless & Gracely, 1988). The BSQ is a 17 item scale comprised of items concerning sensations associated with anticipatory anxiety or anxiety in phobic situations (e.g. increased heartbeat). The scale has been found to be highly internally consistent (Cronbach alpha=.87) and have moderately good test-retest reliability (r=.67) (Chambless et al., 1984). Construct validity has been suggested by significant correlations with self-reported avoidance behaviour (.25), STAITrait anxiety (.21), and BDI (.36) (Chambless & Gracely, 1988). In addition, the BSQ has shown highly significant changes with treatment (Chambless et al., 1984).  The Agoraphobic Cognitions Questionnaire (ACQ)  31 The ACQ is a 14 item scale comprised of items concerning negative consequences of experiencing anxiety (e.g. "I'll die or go crazy"). The scale has been found to be highly internally consistent (Cronbach alpha=.80) and have good test-retest reliability (r=.86) (Chambless et al., 1984). Construct validity has been supported in that the measure displays significant correlations with the BSQ (.67), self-reported avoidance behaviour (.37), STAITrait anxiety (.35), and BDI (.38) (Chambless & Gracely, 1988). In addition, the ACQ has showed highly significant changes with treatment (Chambless et al., 1984). The ACQ has two 7 item factors concerning thoughts of physical catastrophe due to anxiety symptoms(e.g. "I'm going to have a stroke") or of social or behavioural disaster from loss of control (e.g."I'm going to babble or talk funny" or "I am going to go crazy"). Cronbach alphas for the factor scores are .65 for the physical concerns factor and .76 for the loss of control factor.  Clinical Background Sheet  Subjects were asked questions regarding their history of panic attacks, their history of depressed mood, and their past and current use of psychiatric medication (see Appendix 4).  Procedure  Subjects were contacted by a member of the Health Psychology Clinic or Se-Cure program. Of the fifteen people asked to participate from the Health Psychology Clinic, two  32 individuals diagnosed with social phobia declined to participate. Both individuals could not arrange transportation to the clinic. The response rate from Se-Cure cannot be reported since the exact number of individuals contacted is unknown. Information about the study was presented by leaders of Se-Cure groups in the community to their group members. Nonanxious controls were recruited through university classes. Of the twenty-two students who completed the questionnaires, 12 did not meet the required criteria (5 were too young to match clinical group, 3 had significant agoraphobic fears, 3 had significant social fears, and 1 had a history of panic attacks). Clinical Subjects were scheduled for two appointments. During the first appointment, subjects completed the questionnaire series. The construct measures were given first to avoid priming any constructs by the other measures. At the second appointment, individuals were given feedback based on their responses to the clinical measures. When necessary for confirmation of the psychiatric diagnosis, the ADIS-R was administered. The control subjects were tested over one appointment. They completed the questionnaire series in the same order as the clinical subjects, and then were administered the ADIS-R.  33 VII. Results  A. Overview  In univariate and multivariate analyses, significant main effects were followed by post hoc analyses (Student Newman Keuls) conducted at .05 significance level. In chi-square analyses, tables that evidenced significant differences in proportions were followed up with 2 by 2 chi-square analyses. The assumptions for these statistical analyses are discussed in appendix 5.  B. Demographical data  Educational level, age, sex and marital status were balanced across groups as represented in Table 1. ANOVAs were conducted on years of education and age. Chi-squares were conducted on sex and marital status. These analyses revealed no significant differences between the groups (Table 1). In addition, all subjects were Caucasian. Thus, it is unlikely that any significant differences between the group in salience of constructs stemmed from differences in these particular demographic characteristics.  34  Table 1. Between-group comparability of demographic characteristics. 1  SP  Ag  N  Chi-  df  P  (n=9)  (n=9)  (n=9)  square  3.99  2  0.14  or F Sex  male  6  2  3  female  3  7  6  16  15  16  0.23  2,24  0.76  1.00  4  0.91  0.28  2,24  0.69  education (yrs)  Marital  Married  5  4  3  Status  Single  3  4  5  Divorced  1  1  1  Age  Mean(SD)  38 (7)  40 (12)  36 (11)  (yrs)  Range  27-52  26-58  24-58  1. SP= Social Phobic, Ag= Agoraphobic, N= Nonanxious control  35 C. Social Phobia and Anxiety Inventory (SPAI) Preliminary analyses were conducted to check on the division of the three diagnostic groups: those individuals diagnosed with social phobia, those diagnosed with agoraphobia, and those who did not meet a psychiatric diagnosis (nonanxious controls). The means and standard deviation of the two subscales scores, Social phobia subscale (SPSS) and Agoraphobia subscale (ASS), as well as the overall Social Phobia and Anxiety Inventory score (SPAI) is shown in Table 2. The overall score represents the difference score of the Social phobia subscale minus the Agoraphobia subscale. An one-way (Diagnostic group) ANOVA was conducted on SPAI scores and revealed a significant difference for groups, F(4,46) = 14.25, p<.001. Multiple comparisons were conducted using the Student-Newman-Keuls technique at the .05 level. This confirmed that all the group means were significantly different from each other, with social phobics scoring the highest, agoraphobics the middle, and nonanxious controls the lowest. The two subscales scores, Social phobia subscale (SPSS) and Agoraphobia subscale (ASS), were analyzed by a one-way (diagnostic Group) MANOVA. This showed significant differences between the groups, F(4,46) = 16.08, p_<.001. This was followed up with univariate analyses. The scores on the SPSS were significantly different between groups, F(2,24)=28.42, p. < .0001. Student-Newman-Keuls conducted at the .05 level revealed that social phobics and agoraphobics scored significantly higher than normals, but not from each other. The scores on the ASS were significantly different between groups, F(2,24) = 14.02, P < .001. Student-Newman-Keuls conducted at the .05 level revealed that agoraphobics scored significantly higher than social phobics who scored significantly higher than normals.  36  Table 2. The Means and Standard Deviations of the Social Phobia and Anxiety Inventory (SPAI) and it's two subscales. Social Phobia (SPSS) and Agoraphobia (ASS).  Social Phobic  Agoraphobic  Nonanxious control  SPAI  103 (26)  64  (29)  34 (14)  SPSS  127 (22)  102 (31)  43 (19)  ASS  24  39  9  (11)  (15)  (8)  37 D. Clinical Background Between group comparison of different background factors shown on Table 3. Medication History A 2 by 3 (Have been prescribed antidepressant medication by Diagnostic category) revealed a significant difference between groups, Chi-square(2)=6.75, p_<.05. Follow-up chi-squares revealed that both social phobics and agoraphobics were significantly more likely to have been on medication for depression than nonanxious controls (Chi-square(l)=5.14, p_< .05 and Chi-square(l) =5.74, p< .05, respectively). Social phobics and agoraphobics were not statistically different from each other (Chi-square(l) = .22, g>.25). A 2 by 3 (Currently on antidepressant medication by Diagnostic category) revealed a trend for a difference between the group, Chi-square(2)=5.20, p_<.10. Follow-up chisquares found that this trend was a result of social phobics being more likely to be on medication than nonanxious controls (Chi-square(l)=5.14, p< .10). Agoraphobics were in the middle, and there was not a trend for them to differ from either social phobics or nonanxious controls (Chi-square(l) = 1.00, p_>.25 and Chi-square(l) = .84,  p_>.25,  respectively). A 2 by 3 (Have been prescribed anxiolytic medication by Diagnostic category) revealed a significant difference between groups, Chi-square(2) = 11.7, p< .005. Follow-up chi-squares revealed that both social phobics and agoraphobics were more likely to have been prescribed anxiolytic medication than nonanxious controls (Chi-square(l)=5.14, p_< .05 and Chi-square(l) = 10.74, p_<.01, respectively), but did not differ from each other (Chisquare(l)=2.10, p_>.10).  38 Table 3. Between group comparability on Clinical Background Factors.  Medication History  Depression History  History of panics  Social Phobic (n=9)  Agoraphobic (n=9)  Normal (n=9)  Prescribed antidepressant in the past  4  5  0  Currently on antidepressant medication  4  2  0  Prescribed anxiolytic medication in the past  4  7  0  Currently on anxiolytic medication  1  5  0  Episode of depressed mood in past  9  8  6  Sought treatment for depressed mood  4  7  3  Experienced a panic attack in the past  5  9  0  Experienced a panic attack in the 3 weeks prior to assessment  0  2  ,0  39 A 2 by 3 (Currently on anxiolytic medication by Diagnostic category) revealed a significant difference between groups, Chi-square(2)=9.95, p_< .01. Follow-up chi-squares revealed that agoraphobics were significantly more likely than social phobics and nonanxious controls to be currently on medication (Chi-square(l) =4.00, p < .05 and Chi-square(l)=5.75, p_< .05, respectively). Social phobics and nonanxious controls did not differ from each other (Chi-square=1.06, p_>.25).  Depression History There were no significant differences between the groups on their self-report as having experienced a period of low mood, depression or sadness (Chi-square=4.10, p>.10). In addition there was no between group differences in having sought treatment for depression (Chi-square=3.86, p>.10).  History of Panics There was a significant difference between groups on the proportion of individuals who had experienced a panic attack, Chi-square(2) = 18.1, p< .0001. Follow-up chi-squares revealed all groups were significantly different from each other. Agoraphobics were significantly more likely than social phobics to have experienced a panic attack, Chisquare^) =5.14, p_<.05. Social phobics were in turn more likely to have experiences one than nonanxious controls, Chi-square(l)=6.92, p<.05. No differences were found in the frequency of panics during the last 3 weeks (F(2,23) = 1.43, p>.25). Only 2 agoraphobic subjects reported experiencing panic attacks in this time period.  40  E. Other Clinical Measures  A one-way (Diagnostic group) MANOVA was conducted on the clinical measures: Beck Depression Inventory, Body Sensations Questionnaire and Agoraphobic Cognitions Questionnaire. This analysis revealed a significant main effect for diagnostic group, F(6,44) = 16.5, p<.0001. Significant multivariate effects were followed with univariate analyses. Means and standard deviations for the clinical measures are shown in Table 4.  Beck Depression Inventory (BDI)  A one-way (Diagnostic group) ANOVA was conducted on the BDI scores. This analysis revealed a significant effect for Diagnostic Group, F(2,24) = 14.84, p< .0001. This was followed up with Student-Newman-Keuls at the 0.05 level to determine which groups differed significantly from the others. This revealed that social phobics and agoraphobics differed significantly from nonanxious controls, but not from each other.  Fear of Fear Assessment  a. Body Sensation Questionnaire (BSQ) A one-way (Diagnostic Group) ANOVA on the BSQ scores revealed a significant effect for Diagnostic Group, F(2,24)=58.7, g< .0001. Student-Newman-Keuls performed at  41  Table 4. Means and standard deviations on clinical measures for Diagnostic Groups. 1  Social Phobic  Agoraphobic  Nonanxious control  BDI  12 (5)  15 (7)  2  BSQ  2.72 (0.49)  3.50 (0.53)  1.27 (0.27)  ACQ  2.03 (0.46)  2.44 (0.40)  1.09 (0.16)  BDI=Beck Depression Inventory, BSQ=Body ACQ=Agoraphobic Cognitions Questionnaire  Sensations  (3)  Questionnaire,  42 the 0.05 level revealed that all the means were significantly different from each other with the agoraphobics scoring the highest, social phobics in the middle, and nonanxious controls scoring the lowest.  b. Agoraphobic Cognitions Questionnaire A one-way (Diagnostic Group) ANOVA on the ACQ scores revealed a significant effect for Diagnostic Group, F(2,24)=32.2, p_< .0001. Student-Newman-Keuls preformed at the 0.05 level revealed that all the means were significantly different from each other with the agoraphobics scoring the highest, social phobics in the middle, and nonanxious controls scoring the lowest. Two factor scores of the ACQ, ACQ-physical and ACQ-social/behavioural, are shown in Table 5. A one-way (Diagnostic Group) MANOVA was performed on the factor scores. It revealed a significant effect for diagnostic category, F(4,46) = 13.81, p< .001. This was followed up with univariate analyses. A one-way (Diagnostic Groups) ANOVA on ACQ-physical scores revealed a significant difference between groups, F(2,24) = 13.34, p < .0001. A Student-Newman-Keuls performed at the .05 level revealed that agoraphobics scored higher than social phobics and nonanxious controls, who didn't differ significantly from each other. A one-way (Diagnostic Groups) ANOVA on ACQ-social/ behavioural scores revealed a significant difference between groups, F(2,24) = 16.43, p < .0001. A Student-Newman-Keuls performed at the .05 level revealed that agoraphobics and social phobics scored significantly higher than nonanxious controls, but did not differ significantly from each other.  43  Table 5. Means and standard deviations of the two factors of the Agoraphobic Cognition Questionnaire: ACO-physical and ACO-social/behavioural for Diagnostic Groups.  Social Phobic  Agoraphobic  Nonanxious control  ACQ-physical  1.36 (.40)  2.21 (.75)  1.05 (.07)  ACQ-social/  2.70 (.72)  2.67 (.86)  1.13 (.28)  behavioural  44  F. Main Analyses: Accessible Constructs Unstructured measurement of construct accessibility No relationship between diagnostic group and accessible constructs nominated on the Impression of Others Questionnaire Part A , the Higgins et al. (1982 ) task, or Part B, the social phobic or agoraphobic situations, could be found. The constructs were compared to those dimensions hypothesized from the pilot data and the literature. The proportion of individuals in the three groups with those constructs were compared and no trends could be seen (See Table 6-8). It is not likely that this lack of result was due to insufficient power. 2  As a hypothetical exercise, we tested the proportional differences at double the sample size and still found no statistical differences (using a bonferonni corrected alpha for multiple statistical tests).  Structured measurement of salient features of others As noted before, six bipolar dimensions were compared by the Other's Questionnaire. The scores of each trait and it's verbal opposite were summed for a dimension score. Four bipolar dimensions were derived from the pilot data: reliable-irresponsible and sincerehypocritical for social phobics and appreciative-conceited and considerate-selfish for  To insure that other patterns were not missed, an exploratory analysis looking for shared constructs other than those hypothesized was conducted. This did not reveal any patterns in the shared accessible constructs of the groups. In addition, traits were grouped according to the three major factors of semantic meaning that divide up semantic space that were empirically derived by Osgood et al. (1957). These factors are Evaluation, Potency, and Directed Activity. The published bipolar scales that underlay the factors were used to place the traits nominated by subjects under the appropriate factor (Osgood et al., 1957). No patterns could be detected within the groups. 2  45 Table 6. Number of subjects in each diagnostic group who nominated traits under the hypothesized dimensions on the Impression of Others Questionnaire: Part A .  Social Phobic  Agoraphobic  Nonanxious  (n=9)  (n=9)  control (n=9)  Sincere-Hypocritical  3  2  0  Reliable-Irrespon sible  0  0  0  Appreciative-Conceited  1  ; i  2  Considerate-Selfish  3  4  2  Enabling-Bossy  4  4  4  Supportive-Critical  2  2  1  46  Table 7.Number of subjects in each diagnostic group who nominated traits under the hypothesized dimensions on the Impression of Others Questionnaire: Part B - Social Phobic Situations.  Social Phobic  Agoraphobic  Nonanxious  (n=9)  (n=9)  control (n=9)  Sincere-Hypocritical  1  0  0  Reliable-Irresponsible  0  1  0  Appreciative-Conceited  0  0  0  Considerate-Selfish  0  1  0  Enabling-Bossy  0  1  1  Supportive-Critical  1  2  0  47  Table 8. Number of subjects in each diagnostic group who nominated traits under the hypothesized dimensions on the Impression of Others Questionnaire: Part B - Agoraphobic Situations.  Social Phobic  Agoraphobic  Nonanxious  (n=9)  (n=9)  control (n=9)  Sincere-Hypocritical  0  0  0  Reliable-Irresponsible  0  1  0  Appreciative-Conceited  0  0  0  Considerate-Selfish  0  2  0  Enabling-Bossy  1  0  0  Supportive-Critical  0  1  0  48 agoraphobics. Two dimensions had been derived from the clinical literature. Supportivecritical and enabling-bossy were consistent with current conceptualizations of agoraphobics. Supportive-critical was also consistent with the clinical literature on social phobics. The means and standard deviations of the six bipolar dimensions (sincere-hypocritical, reliableirresponsible, appreciative-conceited, considerate-selfish, enabling-bossy, and supportivecritical) are shown on Table 9. A one-way MANOVA was done on the 6 trait dimensions and revealed significant group differences, F(12,38)=2.09, g< .05. This was followed up with univariate analyses. One-way (Diagnostic Groups) ANOVAs revealed that two dimensions showed significant group  differences:  Supportive-Critical, F(2,24)=4.38,  p_<.05,  Enabling-Bossy,  F(2,24)=3.80, p_<.05 and there was a trend on another, appreciative-conceited, F(2,24)=3.09, p_<. 10. Student-Newman -Keuls comparison of means was performed at the .05 level on the dimensions with significant group differences. Agoraphobics were found to endorse significantly higher for the dimension supportive-critical than social phobics and nonanxious controls, who did not differ significantly from each other . For Enabling-Bossy, 3  The dimension supportive-critical was derived from the literature of both social phobics and agoraphobics. The trait critical has been postulated to be important for social phobics and the trait supportive for agoraphobics. To ensure that social phobics were not differentially endorsing critical, but not its verbal opposite supportive, the poles were analyzed separately. A one-way (diagnostic group) MANOVA on the two single traits revealed a significant effect for group, F(4,46)=3.17, p_<.05. This was followed with univariate analysis. One-way (Diagnostic Groups) ANOVAs revealed that this difference was accounted by the trait Supportive, F(2,24)=6.33, p_<.005. There were no significant differences between the groups on critical, F(2,24)=2.01, p_=. 16. Student-Newman -Keuls comparison of means was performed at the .05 level on Supportive. Agoraphobics were found to endorse significantly higher than social phobics and nonanxious controls, who did not differ significantly from each other. There was no evidence that social phobics were differentially endorsing the trait critical. 3  49  Table 9. Means and standard deviations of bipolar dimensions on the Others Questionnaire by diagnostic group.  Social Phobic  Agoraphobic  Nonanxious control  Sincere-Hypocritical  17.0 (2.2)  19.4 (3.5)  17.1 (4.4)  Reliable-Irresponsible  14.9 (3.6)  16.1 (3.5)  15.2 (6.3)  Appreciative-Conceited  15.8 (3.2)  17.8 (2.5)  19.0 (2.6)  Considerate-Selfish  17.2 (2.5)  19.8 (2.6)  18.0 (3.2)  Enabling-Bossy  16.3 (3.2)'  19.8 (2.0)  18.2 (2.6)  Supportive-Critical  16.1 (4.0)  20.1 (1.3)  15.0 (5.1)  50 Agoraphobics endorsed significantly higher than social phobics, but nonanxious controls were not significantly different from either group. Student-Newman-Keuls comparison of means was performed at the .10 level for the dimension considerate-conceited. This revealed that there was a trend for nonanxious controls to notice this dimension more than social phobics. Agoraphobics scored in the middle, and there was no trend for them to differ from either group. There may not have been adequate power to statistically detect the smaller group differences on the other dimensions. Sincere-hypocritical and considerate-selfish had smaller effect sizes (approximately 1/2 standard deviation) than those of dimensions where group differences were found. The univariate tests for sincere-hypocritical and considerate-selfish were at a power of .27 and .36, respectively (derived from an Edgeworth-type normal approximation to the non-central beta distribution). Reliable-irresponsible had a very small effect size (less than 1/4 standard deviation), and the corresponding univariate test had small power, .07. The small effect size for the dimension reliable-irresposible , however, suggests that the difference would not be clinically significant.  Supplementary Analysis  During the administration of the questionnaires, the subjects were encouraged to ask questions when anything was unclear. Four of the individuals with agoraphobia expressed that they noticed all of the traits, and that it was hard to discriminate. None of the individuals with social phobia, and only one of the controls expressed the same sentiment. The overall  51 mean was compared between the groups to see if there was a tendency for agoraphobics to obtain high scores on all traits. Table 10 shows the means and standard deviations for the mean of all twelve items. A one-way (Diagnostic Group) ANOVA on a sum of the 12 items of the Others Questionnaire revealed a trend for group, F(2,24)=2.83, p < . 10. Comparison of means by a Student-Newman-Keuls at the .10 level revealed that the trend was a result of a difference between social phobics and agoraphobics. Nonanxious controls scored in the middle, and there was no trend for a difference between them and either anxiety group. In a nonstatistical comparison of the means (see Table 11), agoraphobics obtained higher scores than social phobics and nonanxious controls on 9 out of the 12 traits. In contrast, social phobics scored the lowest on 7 out of 12 traits, and never scored the highest.  52  Table 10. Means and standard deviations by diagnostic group for the mean of a total score for all items on the Others Questionnaire.  All items  Social Phobic  Agoraphobic  Nonanxious control  8.1 (1.2)  9.4 (1.0)  8.5 (1.3)  53  Table 11. Means and standard deviations of traits on the Others Questionnaire by diagnostic group.  Social Phobic  Agoraphobic  Nonanxious control  Sincere  9.7 (0.9)  9.9 (1.9)  9.3 (2.1)  Hypocritical  7.3 (2.2)  9.6 (1.9)  7.8 (3.3)  Reliable  7.7 (2.0)  9.0 (1.7)  7.9 (3.7)  Irresponsible  7.2 (2.0)  7.1 (2.4)  7.3 (3.1)  Appreciative  8.4 (1.8)  9.1 (1.9)  8.8 (2.1)  Conceited  7.3 (3.0)  8.7 (2.3)  10.2 (1.1)  Considerate  9.8 (0.7)  10.4 (0.6)  9.3 (1.6)  Selfish  7.4 (2.9)  9.3 (2.3)  8.7 (2.6)  Supportive  8.4 (1.7)  10.4 (0.9)  7.3 (2.9)  Critical  7.8 (2.9)  9.8 (1.0)  8.0 (2.6)  Enabling  8.1 (1.7)  10.0 (1.0)  8.0 (2.3)  Bossy  8.2 (2.9)  9.9 (1.2)  10.0 (1.1)  54 VIII. Discussion  Comparability of subjects in the present study to those in previous research This study compared three groups of individuals: those who met the DSM-III-R criteria for social phobia, those who met DSM-III-R criteria for panic disorder with agoraphobia, and nonanxious controls who did not meet either criteria. With small numbers of clinically diagnosed subjects, it was important to confirm the comparability of the groups in this sample to those in past research. In the present study, this was accomplished by comparing the present subjects' responses on the clinical measures to those subjects in past studies. The analysis of the Social Phobia and Anxiety Inventory (SPAI) confirmed that we had three distinct groups. SPAI scores differed significantly between groups with individuals with social phobia scoring the highest, those with agoraphobics scoring in the middle, and nonanxious controls scoring the lowest. As expected, agoraphobics scored significantly higher than social phobics on the agoraphobic fear subscale. In turn, the social phobics scored significantly higher than nonanxious controls. There was no significant difference between agoraphobics and social phobics on the social fear subscale. This was expected since agoraphobics often report a high number of social fears (Turner et al., 1989; Liebowitz et al., 1985, Pollard & Cox, 1988). This replicates the results found by Turner et al.(1989). The groups were also compared using Chambless's instruments, Body Sensations Questionnaire (BSQ) and Agoraphobic Cognitions Questionnaire (ACQ), which assesses 'fear of fear'. The BSQ measures the fear of body sensations that are associated with high arousal  55  or panic, such as rapid heart rate. In this study, the BSQ discriminated between agoraphobics, social phobics, and nonanxious controls. All groups differed significantly from each other with agoraphobics scoring the highest, and nonanxious controls the lowest. This agrees with Chambless and Gracely's ( 1989) earlier finding. The total score for the ACQ which measures maladaptive thoughts about the consequences of anxiety also was found to discriminate between the groups with the same pattern. Craske, Rachman, and Tallman (1986) had found the same general pattern, but with agoraphobics only having a trend to score higher than social phobics. Chambless and Gracely (1986) felt that the total score for ACQ might not always discriminate between social phobics and agoraphobics because they share a common fear of losing of control. Chambless and Gracely felt that the subscale scores: social /behavioural concerns (ACQ-S) (e.g. loss of control) and physical concerns (ACQ-P) (e.g. heart attack, fainting) would more likely reveal differences between these two groups. In this sample, the ACQ-S was found to discriminate both anxiety groups from nonanxious control but not from each other. The ACQ-P, however, was found to discriminate agoraphobics from social phobic. Social phobics were not found to score significantly different on the ACQ-P from nonanxious controls. Chambless and Gracely (1986) did not have a normal comparison group, but found the same pattern of similarity and difference between social phobics and agoraphobics. These analyses reveal that the agoraphobic, social phobic and nonanxious subjects responded to these questionnaires very similarly to those groups reported in published studies. This increases our confidence that we are dealing with clinical groups similar to those involved in past research.  56 Comparison between diagnostic groups on potentially confounding variables  Clinical groups often differ on variables other than those relevant to clinical symptoms. Because of this, it is important in comparative studies to determine whether subjects are matched on possible confounding variables. In this study, the three groups were contrasted on a number of background factors and clinical measures. The groups were found to be balanced for sex, age, marital status, number of years in school, and ethnic background. Depression was also investigated as a possible confound in the study. There was no difference found between the groups on self report of past depressive episodes or on seeking treatment for depression. There was a difference between the groups on self report of current depressive symptomatology. Agoraphobics and social phobics reported more symptoms than nonanxious controls, but they did not differ from each other. Amies et al. (1983) and Solyom et al. (1986) also found no difference in depressive symptomatology in their comparative studies. Other clinical background factors were also studied. Agoraphobics were significantly more likely to have had a panic attack in the past, than social phobics. Social phobics were , in turn,more likely than nonanxious controls to have experienced a panic attack. All the agoraphobics in this sample had experienced a panic attack in the past, 55 % of the social phobics had, and none of the nonanxious controls. This is consistent with the clinical picture of agoraphobia. Cottraux et al. (1988) emphasized the importance of panic symptoms for differentiating agoraphobics from social phobics. Due to the small number of subjects who  57 had experienced a panic attack in the three weeks previous to assessment, no differences were found in panic frequency. Only 2 out of the 9 agoraphobics experienced panics in that time period. None of the social phobics or controls experienced any panics. Although differential medication use was found between the groups in this sample, it was concluded that this was not likely to have contributed to the differences found in the salience of specific traits. There was a trend for a greater proportion of social phobics to be currently on antidepressants than nonanxious controls. However, they were no more likely to be medicated than agoraphobics. The side effects of dry mouth, possible agitation, and nausea could lead to less attention, but this does not fit the pattern of results. Agoraphobics were equally likely to be on antidepressant medication, and they were more attentive to certain traits. Although they were both equally more likely than nonanxious controls to have been prescribed anxiolytic medication, agoraphobics were more likely to be on medication presently than either social phobics and nonanxious controls. It is not clear how current anxiolytic use could lead to greater attention to signs of supportiveness and enabling behaviour. The side effects of lethargy and sleepiness would seem to predispose one to be less attentive. Hence it is difficult to see how differential medication use could lead to the pattern of trait salience that was found. In summary, these groups did not differ on any of the major demographic characteristics investigated in this study. In addition, clinical factors were either balanced over all three groups (i.e. history of depressed mood, seeking treatment for depressed mood, and current panic history) or balanced between the two anxious groups (i.e. depressive symptomatology). When group differences did arise, it was difficult to see how they would  58 explain the differences that emerged in trait salience. Therefore although some other characteristic of these groups might mediate or explain the differences found in salience, it is not likely to be those background and clinical factors investigated in this study.  Differential accessibility or salience of interpersonal constructs  The major aim of this study was to compare the accessible or salient constructs of individuals with either social phobia or agoraphobia. In past research, construct accessibility was determined for individuals, and then groups were formed of individuals who shared accessible constructs (e.g. Higgins et al.'s (1982)). In this study, we wanted to look at differences in construct accessibility of preformed groups. Cognitive theory predicts that different construct accessibility develops from an individual's personal developmental history (Wyer & Srull, 1986). From retrospective studies, there is evidence that individuals with social phobia and agoraphobia had different developmental experiences. For example, social phobics remember their parents as more negative, and rated them as more rejective and less affectionate (Persson & Nordlund, 1985), and agoraphobics remember the loss of a love one as a trigger for their subsequent problems (Amies et al., 1983). We had predicted that as a result of different social experiences, these groups would differ on the accessible constructs they use to describe others. In this study, constructs were represented by bipolar (verbal opposites) dimensions. This is compatible with major linguists' conclusions that it is natural for human beings to think in terms of opposites (Osgood et al., 1957; Osgood, 1990). Verbal opposites were seen as semantically meaningful units.  59 From the literature on agoraphobics, it was hypothesized that the dimension supportive-critical and enabling-bossy would be more accessible. Agoraphobics act in a very dependent manner, and theorists have proposed that the presence of a supportive caregiver is very important to them (Beck & Emery, 1985; Oei et al., 1989; Rachman, 1983). In addition, it has been proposed that issues of control are salient for them. Beck and Emery (1985) found that agoraphobics report a strong investment in mobility, and a sensitivity to being directed. Goldberg (1986) theorized that there is a conflict between the desire to be dependent and taken care of, and the wish to be independent and self-directive (Goldberg, 1986). Based on the pilot study using anxious university students, the dimensions appreciative-conceited and considerate-selfish were also hypothesized to be accessible. Derived from the literature on social phobics, it was hypothesized that the dimension of supportive-critical might be of importance. Social phobics have been found to fear negative evaluation (Liebowitz et al., 1985). Theorists have proposed that social phobics are hypervigilant to social cues that may indicate hostile appraisal of their self presentation (Trower & Gilbert, 1989). Leary et al. (1988) found possible evidence for a generalized expectancy for others to be rejecting and critical in socially anxious college students. Based on the pilot study, social phobics were also hypothesized to use the dimensions sincerehypocritical and reliable-irresponsible. The unstructured measurement of construct accessibility (impression of others questionnaire) revealed no differences between the groups. There was no support for the hypotheses derived from the literature or from the pilot study. There was no indication that individuals of one diagnostic category tended to share similar constructs when describing  60 others. There are different possible explanations for the lack of results. One possibility is that it was a result of the small n and with larger samples we might find some pattern. Another is that the individual factors influencing construct accessibility outweighed any experience that was shared within the groups. The lack of replication of the pilot study results, in particular, could be due to the fact that a nonclinically anxious population was used for the pilot study. In addition, the subjects in the pilot and the present study had different demographic qualities. The pilot subjects were younger, were in first or second year university, and were more likely to come from an Asian background. The unstructured questionnaire, on the other hand, did appear to be a clinically meaningful task. Our clinical impression was that subjects took the instrument seriously, and that it was worthwhile to them. Subjects made spontaneous comments indicating interest and personal insight. For example, one agoraphobic said that the type of person he would most like accompanying him was someone with a medical degree who cares a great deal about him. Another example, one person reported that she was surprised at the number of times she used the trait, angry. She said that it made her think about why it bothered her so much. Another individual commented that by consciously thinking about traits of people that she liked and that of people that she disliked, she realized how difficult it would be for anyone to live up to her standards. She stated that she had realized that she was perfectionistic in other areas of her life, but had not thought about her perfectionism in respect to her expectations of others. She felt that this could explain why she often felt let down by others. Thus, this measure may be clinically useful on an individual level. Differences emerged between the groups on the structured task investigating the  61 salience of specific traits, the Other's Questionnaire. This questionnaire asked subjects to rate how likely they would be to notice specific traits in others. These traits were derived from the pilot data and from the literature. Individuals with social phobia were predicted from the pilot data to use the dimensions sincere-hypocritical and reliable-irresponsible. Based on the pilot study, it was also postulated that the dimensions considerate-selfish and appreciativeconceited would be more salient for agoraphobics. There was no support for these hypotheses in this study. In fact, there was a trend for nonanxious controls to report that they would notice the dimension appreciative-conceited more than social phobics. Nonanxious controls were not significantly different from agoraphobics. As explained before, the lack of confirmation of the pilot results could be due to difference in clinical and demographic qualities between the pilot subjects and those in the present sample. Both the pilot sample and the nonanxious subjects in the present study were students. Perhaps being a student primes the dimension appreciative-conceited. One possible priming mechanism is the constant stream of explicit, structured evaluation received by students. Grading is often compared between fellow students, and focuses students attention on their and others comparative 'worth'. From the literature, it was hypothesized that the dimensions supportive-critical and enabling-supportive would be more salient for agoraphobics. There was support for both of these hypotheses. Agoraphobics reported that they would be more attentive to the trait dimension supportive-critical. The salience of this dimension is consistent with the conceptualization of the agoraphobic as an individual who feels they have to be close to others for protection or help to cope with life's dangers (Beck & Emery, 1985, Roth and Argyle, 1988, Zitrin & Ross, 1988). Agoraphobics were also more likely than social phobics  62 to use the dimension enabling-bossy. The focus of this dimension is whether others are going to tell them what to do, or if others will help them do what they want. Specifically, agoraphobics seem to be looking for someone who will help them do what they want in a nondirective manner. This finding is consistent with theorizing in the literature that individuals with agoraphobia want someone to help them, but are sensitive to issues of control (Beck & Emery, 1985; Goldberg, 1986). They are often in catch 22 situations where they want help, but feel that to ask for help will leave them open to be controlled by that person. Based on the descriptions in the literature, it was hypothesized that the dimension supportive-critical would be salient for social phobics. This hypothesis was not supported. In fact, there was a trend for social phobics to notice all the traits less. Taken at face value, the finding that social phobics were non- attentive to signs that someone is critical seems to be inconsistent with previous clinical writings. On closer examination of previous reports, however, this can be resolved. Prior research has revealed that socially phobic individuals expect more negative evaluations from others, are hypersensitive to criticism and hypervigilant to possible social threats (Liebowitz et al., 1985; Smith & Sarason, 1975; Cacioppo, Glass, & Merluzzi, 1979). But this research deals with what social phobics feel will happen, and doesn't really address how they see others. The fact that social phobics expect criticism does not necessarily imply that they think that others are overly critical. Instead, they may feel they are deserving of that criticism. Another problem with existing literature is the absence of empirical data. For example, Beck and Emery (1985) proposed that socially phobic individuals may pay  63 particular attention to whether a potential evaluator seems hostile and rejecting, or warm and accepting. In addition, Trower and Gilbert (1989) theorize that social phobics are hypervigilant to signs that others are critical. However, only one empirical study directly addressed how socially anxious individuals view others. Leary and his colleagues (1988) looked at socially anxious individuals' perceptions of others and found possible evidence for a generalized expectancy for others to be critical. This study however used nonclinically anxious subjects and was a pencil and paper task. They found that socially anxious individuals thought potential evaluators would rate them, as well as another person, more negatively than nonanxious. Even so, this could be explained by the socially anxious individuals perception that no one performs well on these tasks. Rather than seeing others as critical, social phobics may see themselves as inadequate and deserving of criticism. Social anxiety could result from a diminished sense of selfefficacy in social situations, and not by a perception that others are critical and have unrealistic expectations. Wallace and Alden (1991) compared socially anxious and nonanxious university students and found that their perceptions of others expectations did not differ. However, anxious students were more pessimistic about their abilities to meet these standards. Socially anxious individuals may be focused on their own internal feelings of inadequacy and ineffectiveness, and may in fact be inattentive to other people, at least to cues that they are critical. A number of caveats should be noted regarding these results. The design of this study was quasi-experimental. Diagnostic group was a classification variable, and could not be manipulated. Thus the diagnostic groups may differ on some third factor that might account  64 for any between-group differences observed here (see Ingram, 1989). The groups were contrasted on a number of demographic and clinical variables, and it was concluded that it would be unlikely if these could account for the differences found in the salience of interpersonal traits. Some other characteristic of these groups not studied, however, might mediate or explain the results found. Other limitations of the study include the small sample size. Also, three agoraphobics were included even though they scored higher than the 80 cutoff specified for the SPAI by Turner et al. Ideally these would not have been included or alternatively more agoraphobics who met the criteria but scored higher on the SPAI could have formed a fourth group. These individuals, however, did meet DSM-III-R criteria for agoraphobia and not social phobia according to the ADIS-R. In addition, these results require replication. In conclusion, this research does suggest that there are differences in the salience of certain traits between social phobics, agoraphobics, and nonanxious control. Agoraphobic individuals reported that they would be more attentive to the dimensions supportive-critical, and enabling-bossy. The salience of these dimensions is consistent with current conceptualizations of agoraphobics as having conflicting needs of dependency and autonomy (Beck & Emery, 1985). Social phobics did not seem to be terribly attentive to any of the traits, at least no more than normals and maybe less so. Perhaps instead social phobics focus their attention on internal feelings of inadequacy.  65 Appendix 1: Pilot Study METHOD Subjects The subjects were 67 undergraduate students from the University of British Columbia. Subjects completed a questionnaire drawn from the agoraphobia and social phobia portions of the revised Anxiety Disorder Interview Schedule (ADIS-R) (DiNardo et al., 1983). There is a degree of comorbidity between agoraphobic and social phobic symptoms (Liebowitz, Gorman, Fyer,& Klein, 1987). Using the questionnaire four groups were formed. These were the individuals whose scores were in the top quadrant of scores reported for Social Phobic fears only (N=8), Agoraphobic fears only (N=8), Both Social Phobic and Agoraphobic fears (N=8), or the bottom quadrant for both Social Phobic and Agoraphobic Fears (N=8). Measurement of Construct Accessibility Chronic construct accessibility was measured using the free response measure derived by Higgins, King and Mavin (1982). This procedure involves asking subjects to list up to 10 traits that best describe the following: (a) a type of person they sought out, (b) a type of person they avoided, (c) a type of person they liked, (d) a type of person they disliked, and (e) a type of person they frequently encountered. In accordance with Higgins et al.'s (1982) operationalization of construct accessibility in terms of output primacy, a given respondent's chronically accessible constructs will be defined as those given first to each of the four affect questions (like, seek out, dislike, avoid), and first and second on the frequency question. RESULTS The responses from the Higgins et al. (1982) procedure were grouped according to thematic categories. Using a thesaurus, trait words were grouped into synonymous dimensions. Because most individuals reported antonyms as well (i.e. They liked people who were sincere and didn't like those who were insincere), categories included both poles of a certain trait (i.e. sincere-insincere). The proportion of individuals from the different groups that have a particular thematic type of accessible construct were compared (Table 12). For individuals high in social phobia, the dimensions of sincere-hypocritical were more accessible. These dimensions are semantically related to trustworthiness. A chi-square analysis of the presence of an accessible dimension sincere-hypocritical was significant at the .01 level (chi-square =15.92, df=3). There was also a trend for reliable-irresponsible to be differentially endorsed across the groups (chi-square=6.91, df=3, p<.10). The analysis of the constructs nominated by individuals high in Agoraphobic fears were more likely to include items representing dimensions of appreciative-conceited and considerate-selfish. These are semantically related to selfcentredness. A chi square analysis of the proportion of individuals who nominated appreciative-conceited was significant at the .05 level ( chi-square=8.16, df=3). There was also a trend for considerate-selfish to be differentially endorsed (chi-square=7.64, df=3, p<.10).  66 Table 12. The number of subjects from each group who nominated traits under the following dimensions.  High Social Fears Only (n = 8)  High Agoraphobic Fears Only (n = 8)  High Both Agoraphobic and Social Fears (n = 8)  Low Both Agoraphobic and Social Fears (n = 8)  SincereHypocritical  0  7  3  1  ReliableIrresponsible  0  3  1  0  AppreciativeConceited  6  1  2  2  ConsiderateSelfish  6  2  5  2  67 "Appendix 2: Impression of Others Questionnaire Part A . a) For this exercise, we would like you to list up to 10 traits that best describes a type of person that you LIKE: b) For this exercise, we would like you to list up to 10 traits that best describes a type of person that you DISLIKE: c) For this exercise, we would like you to list up to 10 traits that best describes a type of person that you AVOID: d) For this exercise, We would like you to list up to 10 traits that best describes a type of person that you SEEK OUT: e) For this exercise, we would like you to list up to 10 traits that best describes a type of person that you FREQUENTLY ENCOUNTER: Part B. a) For this exercise, we would like you to list up to 10 traits (characteristics or qualities) that best describes a type of person that you would feel COMFORTABLE accompanying you to a party: b) For this exercise, we would like you to list up to 10 traits (characteristics or qualities) that best describes a type of person that you would make you feel UNCOMFORTABLE travelling with you on a bus: c) For this exercise, we would like you to list up to 10 traits (characteristics or qualities) that best describes a type of person that you would feel COMFORTABLE accompanying you to a shopping mall with: d) For this exercise, we would like you to list up to 10 traits (characteristics or qualities) that best describes a type of person that you would feel UNCOMFORTABLE discussing different points of views with:  68 Appendix 3: Others Questionnaire When you are getting to know someone, you start to form a picture of what the person is like. As you do this, you are more likely to notice some qualities and behaviours than others. For example, you may be more likely to notice whether a person is friendly than whether a person is smart. Individuals differ on the qualities or behaviours that they first pay attention to when meeting another person. The purpose of this exercise is to find out what you are most likely to notice or pay attention to when you meet another person for the first time. Look at the following trait words and their descriptions. Some of these traits are positive and some are negative. We are interested in what you first notice about a person, positive or negative, when you are trying to form an impression of them. For each word, we would like you to rate, using the scale below, how likely you would be to notice or pay attention to the following traits when you are interacting with another person. So imagine that you are meeting with another person. Ask yourself: " Which of these qualities would I most likely notice or pay attention to about that person?."  1 2 3 4 5 6 7 8 9 10 11 would not pay attention to  would definitely pay attention to  1. Whether they are sincere } - being honest and straightforward 1  2. Whether they are conceited] - having an exaggerated opinion of one's own abilities, appearance, etc 3. Whether they are supportive'] - being encouraging and willing to assist another 4. Whether they are hypocritical! - being two-faced, and saying things or behaving in a manner that isn't genuine  1 2 3 4 5 6 7 8 9 would not pay attention to  10 11 would definitely pay attention to  5. Whether they are enabling! -respectful of other peoples' opinions and wishes, and empowering others to do what they want 6. Whether they are selfish! - caring only for oneself, and concerned with one's owns interests regardless of others 7. Whether they are reliable! - being able to be relied upon, and counted on 8. Whether they are bossy! - being inclined to want things your own way and order other people around 9. Whether they are appreciative! - appreciating other people's ideas, achievements and true worth 10. Whether they are irresponsible! - being unthinking, without a sense of responsibility 11. Whether they are critical! - inclined to find fault or judge with severity, often too readily 12. Whether they are considerate! - showing kindly awareness or regard for another's feelings, circumstances, etc.  70  Appendix 4: Background Sheet 1. Demographics: Age:  Sex: M/F  Place of Birth:  Total number of years in school, starting at grade 1: Marital Status: Married or Common-Law  Single  Divorced  Widowed  Racial/Ethnic Background :  2. History of panic attacks: We define a panic attack as: (1) a high level of anxiety accompanied by (2) strong bodily reactions (heart palpitations, sweating, muscle tremors, dizziness, nausea) with (3) the temporary loss of the ability to plan, think, or reason and (4) the intense desire to escape or flee the situation (Note, this is different from high anxiety or fear alone.) Have you ever had a panic attack?  yes  no  If you have had a panic attack, please indicate the number of panic attacks you have had in the last 3 weeks: Which situations elicited a panic attack?  3. History of Depression: Did you ever have a period of time when you felt depressed, sad, hopeless or lost interest in almost all of your usual activities?  yes  If you have had a period of depressed mood: a. How long did the longest episode last (in days)? b. How many times in your life have you had episodes of depressed mood? c. Did you seek treatment for the depressed mood? yes  no  4. History of Medication Use: Have you had medication prescribed for depressed mood? yes  no  a. If yes, please specify medication b. Are you currently on medication?  yes  no  Have you had medication prescribed for other emotional problems?  yes  no  a. If yes, please specify medication b. if yes, are you currently on medication?  yes  no  72 Appendix 5: Statistical Assumptions Chi-squares were used to test hypotheses involving proportions. Chi-square test of association is a nonparametric technique. It does not assume that the scores under analysis were drawn from a population distributed in a certain way, e.g., from a normally distributed population (Siegel, 1956). It used to be thought that chi-squares should not be used unless the minimum expected frequencies were 5 or more in each cell (Siegel, 1957). Based on recent Monte Carlo experiments, Glass and Hopkins (1984) concluded that this has been overly conservative. Roscoe and Byars (1971, 1979), Conover (1974), and Camilli and Hopkins (1977, 1979) have shown that the chi-square statistic works well even when the average expected frequency is as low as 2. In this study, it was confirmed that the average expected frequency per cell was 2 or greater for every chi-square table. One-factor analysis of variance (ANOVA) technique involves three assumptions: 1) that there is independence between the observations within a group, 2) that the scores under analysis were drawn from a normally distributed population, and 3) that there is homogeneity of variance among the groups (Glass & Hopkins, 1984). Glass, Peckham, and Sanders (1972) empirically tested the consequences of failure to meet ANOVA assumptions and made the following conclusions: a) that the assumption of independence is necessary for accurate probability statements, and could not be violated, b) that nonnormality has negligible consequences on type-I and type-II error probability with bidirectional (two-tailed) tests, and c) that when n's are equal, heterogeneity of variance among the groups has a negligible effect on type-I or type II error. In this study, the observations within the groups were made individually, and did not influence each other. Thus the assumption of independence of observation was met in all cases in our study. As there was equal cell size, violations of normality and homogeneity of variance would have had negligible effects on the Type I or Type II error rates. One-factor multivariate analysis of variance (MANOVA) has multivariate analogues to the ANOVA assumptions. These are the following: a) that there is independence of observations, b) that there is multivariate normality, and c) that there is homogeneity of the variance-covariance matrix (Bray & Maxwell, 1985). As in the univariate case, other statistical techniques need to be used if the assumption of independence is violated. Although less is known about the consequences of violating assumptions in the multivariate case, Monte Carlo experiments suggest that the MANOVA is robust for nonnormality and heterogeneity of variance-covariance matrices when you have equal cell size (Hakstian, Roed & Lind, 1979; Bray & Maxwell, 1985). As stated above, the assumption of independence was met in our study. As there was equal cell size, multivariate nonnormality and heterogeneity of variance- covariance matrix would have had negligible effects on type I and type II error rates.  73 Bibliography Arkin, R . M . , Lake, E.A., & Baumgardner, A . H . (1986). Shyness and self-presentation. In W.H. Jones, J.M. Cheek, & S.R. Briggs (Eds.), Shyness: Perspectives on research and treatment (pp. 189-203). New York: Plenum Press. American Psychiatric Association. (1987). Diagnostic and statistical manual of mental disorders (rev. 3rd ed.). Washington, DC: Author. Amies, P . L . , Gelder, M . G . , & Shaw, P.M. (1983). Social phobia: a comparative clinical study. British Journal of Psychiatry. 142. 174-179. Arrindell, W . A . , Emmelkamp, P.M.G. & Monsma, A. (1983). The role of perceived parental rearing practices in the aetiology of phobic disorders: A controlled study. British Journal of Psychiatry. 143. 183-187. Arrindell, W.A., Kwee, M.G.T., Methorst, G.J., Van der Ende, J., Pol, E. & Moritz, B.J.M. (1989). Perceived parental rearing styles of agoraphobic and socially phobic inpatients. British Journal of Psychiatry. 155. 526-535. Bargh, J.A., Bond, R.N., Lombardi, W.J. & Tota, M.E. (1986). The additive nature of chronic and temporary sources of construct accessibility. Journal of Personality and Social Psychology. 50, 869-878. Beck, A.T., & Beamsdorfer, T. (1974). Assessment of depression: The depression inventory. Pharmacopsychiatry. 7, 151-169. Beck, A.T. & Emery. G. (1985). Anxiety disorders and phobias: A cognitive perspective. New York: Basic Books. Beck, A.T., Steer, R . A . , & Garbin, M . (1988). Psychometric properties of the Beck Depression Inventory: Twenty-five years of evaluation. Clinical Psychology Review. 8, 77-110. Beck, A.T., Ward, C.H., Mendelsohn, M . , Mock, J.,& Erbaugh, J. (1961). An inventory for measuring depression. Archives of General Psychiatry. 4, 561-571. Bray, J.H., & Maxwell, S.E. 1985. Multivariate analysis of variance. Beverly Hills: Sage Publications. Brown, R., Munjack, D. & McDowell, D. (1989). Agoraphobia with or without current panic attacks. Psychological Reports. 64, 503-506.  74 Buglass, D., Clarke, J., Henderson, A.S., Kreitman, N . , & Presley, A.S. (1977). A study of agoraphobic housewives. Psychological Medicine. 7, 73-86. Butler, G. (1990). Issues in the application of cognitive and behavioral strategies to the treatment of social phobia. Clinical Psychology Review. 9, 91-106. Cacioppo, J.T., Glass, C.R., & Merluzzi, T.V. (1979). Self-statements and self-evaluation: A cognitive response analysis of heterosexual anxiety. Cognitive Therapy and Research. 3, 249-262. Camilli, G., & Hopkins, K.D. (1978). Applicability of chi-square to 2 by 2 contingency tables with small expected frequencies. Psychological Bulletin. 85, 163-67. Camilli, G., & Hopkins, K.D. (1979). Testing for association in 2 by 2 contingency ttfes with very small sample sizes. Psychological Bulletin. 86, 11011-14. Cameron, O.G., Thyer, B.A., Nesse, R . M . , & Curtis, G.C. (1986). Symptom profiles of patients with DSM-III anxiety disorders. American Journal of Psychiatry. 143. 11321137. Cattell, R.B. & Sheier, I.H. (1963). Handbook for IPAT Anxiety Scale Questionnaire. Champaign, Illinois: Institute for Personality and Ability Testing. Cerny, J.A., Himadi, W.G. & Barlow, D.H. (1984). Issues in diagnosing anxiety disorders. Journalof Behavioral Assessment. 6, 301-329. Chambless, D.L., Caputo, G . C , Bright, P.,& Gallagher, R. (1984). Assessment of fear of fear in agoraphobics: the body sensations questionnaire and the agoraphobic cognitions questionnaire. Journal of Consulting and Clinical Psychology. 52, 1090-1097. Chambless, D.L., & Gracely, E.J. (1989). Fear of fear and the anxiety disorders. Cognitive Therapy and Research. 13, 9-20. Chambless, D.L. & Mason, J. (1986). Sex, sex-role stereotyping and agoraphobia. Behavior Research and Therapy. 24, 231-235. Conover, W.J. (1974). Some reasons for not using the Yate's continuity correction on 2 by 2 contingency tables. Journal of the American Statistical Association. 69, 374-82. Cook III, E.W., Melamed, B.G., Cuthbert, B.N., McNeil, D.W., & Lang, P.J. (1988). Emotional imagery and the differential diagnosis of anxiety. Journal of Consulting and Clinical Psychology. 56, 734-740.  75 Cottraux, J., Mollard, E., & Duinat-Pascal, A. (1988). Agoraphobia with panic attacks and social phobia: a comparative clinical and psychometric study. Psychiatrie & Psychobiologie. 3, 49-56. Deltito, J.A., Perugi, G., Mareminni, I., Mignani, V. & Cassano, G.B. (1986). The importance of separation anxiety in the differentiation of panic disorder from agoraphobia. Psychiatric Developments. 4, 227-236. de Ruiter, C , & Garssen, B. (1989). Social anxiety and fear of bodily sensations in panic disorder and agoraphobia: a matched comparison. Journal of Psychopathology and Behavioural Assessment. H , 175- 184. DiNardo, P.A., O'Brien, G.T., Barlow, D.H., Waddell, M.T., & Blanchard, E.B. 1983. Reliability of DSM-III anxiety disorder categories using a new structured interview. Archives of General Psychiatry. 40, 1070-1074. Dornbusch, S., Hastorf, A., Richardson, S.A., Muzzy, R., & Vreeland, R.S. (1965). The perceiver and the perceived: their relative influence on the categories of interpersonal perception. Journal of Personality and Social Psychology, i , 434-440. Fodor, I.G. (1974). The phobic syndrome in women: Implications for treatment. In V . Franks & V. Burtie (Eds.), Women in Therapy (pp. 132-168). New York: Brunner/Mazel. Glass, G.V., & Hopkins, K.D. (1984). Statistical methods for education and psychology (2nd ed.). New Jersey: Prentice-Hall. Glass, G.V., Peckham, P.D., & Sanders, J.R. (1972). Consequences of failure to meet assumptions underlying the fixed effect analysis of variance and covariance. Review of Educational Research. 42, 237-88. Goldberg, C. (1986). Agoraphobia: contributing factors, phobic situations, and exposure in vivo. Psychological Reports. 59, 143-160. Goldfried, M.R., Padawer, W. & Robins, C. (1984). Social anxiety and the semantic structure of heterosocial interactions. Journal of Abnormal Psychology. 93, 87-97. Hakstian, A.R., Roed, J.C., & Lind, J.C. (1979). Two-sample T2 procedure and the assumption of homogenous covariance matrices. Pychological Bulletin. 86. 1255-1263. Halford, K. & Foddy, M . (1982). Cognitive and social skills correlates of social anxiety. British Journal of Clinical Psychology. 21, 17-28. Heimberg, R.G., & Barlow, D.H. (1988). Psychosocial treatments for social phobia. Psychosomatics. 29, 27- 37.  76 Higgins, E.T., King, G.A., & Mavin, G.H. (1982). Individual construct accessibility and subjective impressions and recall. Journal of personality and social psychology, 43, 35-47. Hope, D.A., Rapee, R . M . , Heimberg, R.G. & Dombeck, M.J. (1990). Representations of the self in social phobia: Vulnerability to social threat. Cognitive Therapy and Research. 14, 177-189. Ingram, R. (1989). Affective confound in social-cognitive research. Journal of Personality and Social Psychology. 56, 715-722. Kelly, G.A. (1955). The psychology of personal constructs. New York: W.W. Norton. King, R.J., Bayon, E.P., Clark, D.B., & Taylor, C.B. (1988). Tonic arousal and activity: relationships to personality and personality disorder traits in panic patients. Psychiatry Research. 25, 65-72. Leary, M.R., Kowalski, R . M . , and Campbell, C D . (1988). Self-presentational concerns and social anxiety: the role of generalized impression expectancies. Journal of Research in Personality. 22, 308-321. Liebowitz, M.R., Campeas, r., Levin, A . , Sandberg, D. & Papp, L . (1987). Pharmacotherapy of social phobia. Psychosomatics. 28, 305-308. Liebowitz, M.R., Fyer, A.J., Gorman, J.M., Dillon, D., Appleby, I., Levy, F., Anderson, S., Levitt, M . , Palij, M . , Davies, S.O., & Klein, D.F. (1984). Lactate provocation of panic attacks: I. Clinical and behavioral findings. Archives of General Psychiatry. 41, 764770. Liebowitz, M.R., Gorman, J.M., Fyer, A.J., & Klein, D.F. (1985). Social phobia: a review of a neglected anxiety disorder. Archives of General Psychiatry. 42. 729- 736. Markus, H . , Smith, J., & Moreland, R.L. (1985). Role of the self-concept in the perception of others. Journal of personality and social psychology. 49, 1494-1512. Mathews, A. & MacLeod, C. (1989). An information-processing approach to anxiety. Journal of cognitive psychotherapy: An international Quarterly. I, 105-116. Mizes, J.S., Landorf-Fritsche, B., Grossman & McKee, D. (1987). Patterns of distorted cognitions in phobic disorders: an investigation of clinically sever simple phobics, social phobics, and agoraphobics. Cognitive Therapy and Research, i i , 583-592. Munjack, D.J., Brown, R.A., & McDowell, D.E. (1987). Comparison of social anxiety in patients with social phobia and panic disorder. Journal of Nervous and Mental Disease, 175. 49- 51.  77  Myers, J.K., Weissman, M . M . , Tischler,G.L., Holzer, C.E., Leaf, P.J., Orvaschel, H . , Anthony, J.C., Boyd, J.H., Burke, J.D., Kramer, M . & Stoltzman, R. 1984. Six month prevalence of psychiatric disorders in three communities: 1980-1982. Archives of General Psychiatry. 41_, 959-967. Nichols, K.A. (1974). Severe social anxiety. British Journal of Medical Psychology. 47, 301306. Oei, T.P.S., Gross, P R . & Evans, L. (1989). Phobic disorders and anxiety states: How do the differ. Australian and New Zealand Journal of Psychiatry. 23, 81-88. Ohman, A. (1986). Face the beast and fear the face: animal fears as prototypes for evolutionary analyses of emotion. Psychophysiology. 23, 123-145. Osgood, C.E., Suci, G.J., & Tannenbaum, P.H. (1957). The measurement of meaning. University of Illinois Press: Urbana. Osgood, C E . (1990). The cross-cultural generality of visual-verbal synesthetic tendencies. In C.E. Osgood, & O.C.S. Tzeng (Eds.). Language, meaning, and culture (pp. 203-235). New York: Praeger. Ost, L . (1987). Age of onset in different phobia. Journal of Abnormal Psychology. 96, 223229. Persson, G., & Nordlund, C.L. (1985). Agoraphobics and social phobics: differences in background factors, syndrome profiles and therapeutic response. Acta Psychiatrica Scandinavia. 71, 148-159. Pollard, C . A . , & Cox, G.I. (1988). Social evaluative anxiety in panic disorder and agoraphobia. Psychological Reports. 62, 323-326. Rachman, S. (1983). The modification of agoraphobic avoidance behaviour: some fresh possibilities. Behavior Research and Therapy. 21, 567-574. Reich, J., Noyes, R.G., & Troughton, E. (1986). Dependant personality disorder associated with phobic avoidance in patients with panic disorder. American Journal of Psychiatry. 144. 323-326. Reich, J., & Yates, W. (1988). Family history of psychiatric disorders in social phobia. Comprehensive Psychiatry. 29, 72-75.  78 Robins, C.J. (1987). Social perception and heterosexual self-efficacy: A multidimensional scaling analysis. Cognitive Therapy and Research. H , 197-214. Roscoe, J.T., & Byars, J. A. (1971). An investigation of the restraints with respect to sample size commonly imposed on the use of the chi-square statistic. Journal of the American Statistical Association. 62, 626-33. Roth, M . & Argyle, N . (1988). Anxiety, panic and phobic disorders: an overview. Journal of Psychiatric Research. 22, 33-54. Schlenker, B.R.,& Leary, M.R. (1982). Social anxiety and self-presentation:A conceptualization and model. Psychological Bulletin. 92, 641-669. Siegel, S. (1956). Nonparametric statistics for the behavioral sciences. New York: McGrawHill. Smith, T.W., Ingram, R.E. &Brehm, S.S. (1983). Social anxiety, anxious preoccupation and recall of self-relevant information. Journal of Personality and Social Psychology. 44.12761283. Smith, R.E., & Sarason, I.G. (1975). Social anxiety and the evaluation of negative interpersonal feedback. Journal of Consulting and Clinical Psychology. 43, 429. Solyom, L . , Ledwidge, B., & Solyom, C. (1986). Delineating social phobia. British Journal of Psychiatry. 149. 464-470. Thyer, B.A., Nesse, R . M . , Cameron, O.G. & Curtis, G.C. (1985). Agoraphobia: A test of the separation anxiety hypothesis. Behavior Research and Therapy. 23, 75-78. Thyer, B.A., Parrish, R.T., Himle, J., Cameron, O.G., Curtis, G.C. & Nesse, R . M . (1986). Alcohol abuse among clinically anxious patients. Behavior Research and Therapy. 24, 357359. Turner, R . M . , Meles, D., & DiTomasso, R. (1983). Assessment of social anxiety: a controlled comparison among social phobics, obsessive-compulsives, agoraphobics, sexual disorders, and simple phobics. Behavior Research and Therapy. 21, 181-183. Trower, P., & Gilbert, P. (1989). New theoretical conceptions of social anxiety and social phobia. Clinical Psychology Review. 9, 19-35. van Zuuren, F.J. (1987). Personality differences within a diagnostic entity: types of phobias in men and women. Personality and Individual Differences. 8, 101-111.  79 Wallace, S.T. & Alden, L.E. (1991). A comparison of social standards and perceived ability in anxious and nonanxious men. Cognitive Therapy and Research. 15. 237-254. Wyer, R.S. & Srull, T.K. (1986). Human cognition and its social context. Psychological Review. 93, 322-359. Zitrin, C M . & Ross, D . C (1988). Early separation anxiety and adult agoraphobia. Journal of Nervous and Mental Disease. 176. 621-625.  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0302389/manifest

Comment

Related Items