Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The manifestations of perfectionistic self-presentation in a clinical sample Habke, Amy Marie 1997

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

Item Metadata


831-ubc_1997-250652.pdf [ 6.75MB ]
JSON: 831-1.0099204.json
JSON-LD: 831-1.0099204-ld.json
RDF/XML (Pretty): 831-1.0099204-rdf.xml
RDF/JSON: 831-1.0099204-rdf.json
Turtle: 831-1.0099204-turtle.txt
N-Triples: 831-1.0099204-rdf-ntriples.txt
Original Record: 831-1.0099204-source.json
Full Text

Full Text

THE MANIFESTATIONS OF PERFECTIONISTIC SELF-PRESENTATION IN A CLINICAL SAMPLE by AMY MARIE HABKE B.ScN. University of Alberta, 1983 B.A., University of Calgary, 1990 M. A., University of British Columbia, 1992 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES Department of Psychology We accept this thesis as conforming to thgj required standard THE UNIVERSITY OF BRITISH COLUMBIA September 1997 ©A. Marie Habke, 1997 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of fiy cfe/g^ \j The University of British Columbia Vancouver, Canada •ate Se/rt. /irfa? DE-6 (2/88) Abstract Perfectionism has long been recognized as an important personality trait that has a significant impact on emotional and social well-being. More recently, it has been recognized that there is a stylistic aspect to perfectionism that focuses on a desire to appear perfect. This perfectionistic self-presentation, and in particular, the desire for concealment of imperfections, has been related to psychopathology in past research. However, it is proposed that perfectionistic self-presentation presents a particular concern from a clinical perspective because of it's indirect effects on pathology; a desire to conceal imperfections is especially problematic to the extent that it impacts the experience of therapy and the therapy relationship. The current study examined the cognitive, affective/physiological, and behavioral manifestations of perfectionistic self-presentation in a clinical sample. Ninety clinical subjects completed self-report measures of perfectionistic self-presentation, trait perfectionism, impression management, mood, appraisals, and self-handicapping. A brief structured assessment interview that included a discussion of past mistakes, was conducted by trained clinical interviewers. Physiological monitors recorded heart rate and skin conductance level throughout the interview, and the interview was videotaped. Post-interview measures of mood, appraisals, and self-handicapping, were also completed. Results at the bivariate level showed that the self-protective dimensions of perfectionistic self-presentation were associated with more distress both prior to and following the interview, higher heart rate and greater change in heart rate when discussing mistakes (and greater skin conductance for men), greater claims of disability from self-handicaps, and appraisals of the interviewer as both threatening (wanting more than the participant could provide) and disappointed following the interview. Regression analyses showed that the desire to avoid disclosing imperfections was a unique predictor of appraisals of threat over and above demographics, trait perfectionism, and other measures of distress (interaction anxiety and ii depression) and impression management, and of appraisals of the interviewer as disappointed following the interview, over and above demographics and trait perfectionism. The block change score for perfectionistic self-presentation predicting interviewer satisfaction was marginally significant over and above emotional distress and impression managment. The desire to avoid displaying imperfections was a unique predictor of lower threat appraisals. Perfectionistic self-presentation also predicted higher heart rate when discussing errors, over and above demographics and other measures of distress and impression management, and greater change in heart rate from relaxation; this relation held when controlling for demographics, trait perfectionism, and emotional distress and impression management. Perfectionistic self-presentation did not predict defensive behaviors and was not a unique predictor of self-reported negative affect. The results are discussed in terms of the implications for therapy and the therapeutic alliance. in T A B L E OF CONTENTS Abstract ii Table of Contents iv List of Tables vi List of Figures vii Acknowledgment viii INTRODUCTION 1 THEORETICAL OVERVIEW 3 Conceptualization 3 Dimensions of Perfectionistic self-presentation 4 Relation between perfectionistic self-presentation and trait perfectionism 6 Clinical Implications... 7 Type of presentation 8 Assessment 11 Summary 12 R E S E A R C H OVERVIEW 13 Conceptual Distinctions 13 Dimensions of perfectionistic self-presentation 13 Perfectionistic self-presentation and trait perfectionism 14 Clinical Implications 15 Relations with pathology 15 Impact on therapy 17 Summary 18 MANIFESTATIONS OF PERFECTIONISTIC SELF-PRESENTATION 19 Cognitive aspects of perfectionistic self-presentation 19 Affective /Arousal components of perfectionistic self-presentation 21 Emotion 21 Arousal/stress 22 Behavioral aspects of perfectionistic self-presentation 24 Self-handicapping 25 Defensive Behaviors 26 Summary 28 R E S E A R C H QUESTIONS 29 Overview 29 Hypotheses 31 Exploratory analyses 34 iv METHOD 34 Participants 34 Measures 36 Independent Variables 36 Control Variables 37 Dependent Variables 40 Procedure 51 RESULTS 56 Cognition 60 Affect 65 Arousal 67 Behavior 70 Gender Differences 76 DISCUSSION 79 Cognition 80 Affect/Arousal 84 Behavior 92 Summary 97 Implications for therapy 100 Limitations and suggestions for further research 103 CONCLUSION 105 REFERENCES 106 Appendix A: Study Measures 123 Appendix B: Instructions to Coders 129 Appendix C: Interview 130 v List of Tables Table 1. Univariate statistics 57 Table 2. Correlations between perfectionistic self-presentation and demographics and controls 59 Table 3. Correlations between perfectionistic self-presentation and Behavioral Ratings of Social Competence scores 61 Table 4. Hierarchical regression equations predicting Behavioral Ratings of Social Competence scores 63 Table 5. Correlations between perfectionistic self-presentation and affect by self-report ratings, observer ratings, and physiological measures 66 Table 6. Hierarchical regression equations predicting self-reported negative affect (PANAS) 68 Table 7. Hierarchical regression equations predicting heart rate 71 Table 8. Correlations between perfectionistic self-presentation and behavior 74 Table 9. Correlations between perfectionistic self-presentation and selected study variables for men/women 77 vi List of Figures Figure 1 Heart rate over the interview for the sample as a whole 87 Figure 2. Heart rate over the interview for tercile groups on Nondisclosure 89 Figure 3. Skin Conductance Level over the interview for men, according to tercile groups on Nondisplay 90 vii Acknowledgment Many people have helped me during the production of this research and the process of this degree. First, it is necessary to acknowledge all the participants of the study, who came and shared their lives with us. Many thanks are due to my interviewers who did so much for very little, and to my volunteers who did so much for even less. In the lab, special thanks are due to Aaron Nielsen for being patient, and to Carol Flynn for her well-timed encouragement and help with numerous drafts. And of course, to Paul Hewitt who has been so kind with his resources and his support - having someone who believed in me and this study has meant a lot. On a broader level, no person has been as important to me throughout this as Renee Desjardins. Her support and being grafted into her family, made this experience much more than an academic one. At the same time, many other friends have also been critical to this endeavor and my sanity at it's conclusion. Finally, I would like to acknowledge my long-suffering family, who have always been there to support me in so many ways. viii Perfectionistic Self-presentation/ 1 The Manifestations of Perfectionistic Self-Presentation in a Clinical Sample Over the last several decades, personality researchers have struggled to understand and describe differences among individuals. The research literature is replete with studies of various aspects of personality and their relations to behavior and well-being, and each study offers a glimpse into the human condition. Within the myriad of personality traits that have been proposed, perfectionism currently is receiving attention as a powerful predictor of well-being in both the emotional and social realms (e.g. Hewitt & Dyck, 1986; Hewitt & Flett, 1991a; Hewitt & Flett, 1993; Hewitt, Flett, & Weber, 1994). Recent work demonstrates that perfectionism is multidimensional and includes both personal and social trait components (e.g. Hewitt & Flett, 1991b; Frost, Marten, Lahart, & Rosenblate, 1990). Such traits include unrealistic expectations of the self (self-oriented perfectionism), perceptions that others require perfection for oneself (socially prescribed perfectionism) and the need for others to meet one's own high standards (other-oriented perfectionism; Hewitt & Flett, 1991b). Each of these components has been shown to be stable over time, and each predicts pathology in unique ways (e.g. Hewitt & Flett, 1991a; Hewitt, Flett, & Mikail, 1995; Hewitt, Flett, & Turnbull-Donovan, 1992). A second aspect of perfectionism is particularly relevant for social relationships. Interpersonal problems may be exacerbated when the perfectionist retains a strong motivation to appear perfect to others. More specifically, many perfectionists will engage in self-presentational strategies that include a promotion of perfection or concealment of imperfection (Hewitt, Flett, Fehr, Habke, & Fairlie, 1996). This 'perfectionistic self-presentation' represents a stylistic trait (see Buss & Finn, 1987) that may or may not co-occur with trait perfectionism. That is, it is possible that someone might not have a personal motivation for attaining perfection but still have a desire for others to see them as perfect. The importance of perfectionistic self-presentation in the interpersonal realm has been demonstrated through links with problems such as control, intimacy, and assertiveness (Hewitt, Flett, Fehr, et al., 1996). Perfectionistic Self-presentation / 2 As with many types of self-presentation, we know little about how perfectionistic self-presentation actually manifests within a social interaction. Expanding on the descriptions of this personality style (Hewitt, Flett, Fehr, et al., 1996), it is likely that such individuals will be defensive, with minimal self-disclosure of mistakes or shortcomings, will exhibit more "face work" (Friedlander & Schwartz, 1985), or attempts to manage the perceptions of others by providing excuses or justifications for mistakes or shortcomings, and will have strong emotional and physiological reactions to public errors or self-disclosure. Further, it is likely that they will experience social relationships as threatening, given that being close to someone increases the risk of being known as imperfect. This defensive and interpersonally fearful style has particular significance in a clinical setting. Defensiveness and impression management have important implications for the accurate assessment and appropriate treatment of psychiatric patients (Friedlander & Schwartz, 1985; Taurke et al., 1990). When perfectionists have difficulty in disclosing their distress or accepting responsibility for their mistakes, they are less likely to engage appropriately in therapy. In addition, as many researchers have documented the importance of the therapeutic relationship in therapy outcome (see Frieswyk et al., 1986; Gallop, Kennedy, & Stern, 1994; Krupnick et al., 1994), to the extent that trait or self-presentational aspects of perfectionism are expressed in ways inimical to the therapy relationship, perfectionistic individuals would be expected to do poorly in psychotherapy. The fact that rates of perfectionism and perfectionistic self-presentation are high in patient samples (Hewitt & Flett, 1991a; Hewitt, Flett, Turnbull-Donovan, & Mikail, 1991; Hewitt, Flett, & Ediger, 1996a; Hewitt & Flett, 1996), suggests that a concern over being seen as imperfect is likely to be an important feature for a significant proportion of clients in mental health settings. The study proposed here attempts to examine how perfectionism, manifested as a self-presentational style, is expressed in interpersonal interactions in a therapeutic encounter. Self-report measures, observation of behavior, and physiological monitoring will be used to fill in the picture of the manifestations of the dimensions of perfectionistic self-presentation during a clinical interview necessitating Perfectionistic Self-presentation / 3 self-disclosure. Because the construct of perfectionistic self-presentation is relatively new, the conceptual and empirical bases of the construct and its importance in a clinical context will be discussed in some detail prior to considering the specific cognitive, affective, and behavioral manifestations of this interpersonal style. THEORETICAL OVERVIEW Conceptualization The concept of perfectionistic self-presentation as a pervasive neurotic style reflecting a desire to be seen as perfect or a self-protective desire to avoid being known as imperfect, is just beginning to appear in the literature (Hewitt, Flett, Fehr, et al., 1996; Hewitt, Flett & Ediger, 1995). The construct shares assumptions with the early symbolic interactionists (Cooley, 1902; Mead 1934) and later self-presentation theorists (Schlenker, 1980; Schlenker & Weigold, 1992) about the ways in which we consciously or unconsciously regulate information about the self provided to others. It recognizes that people share a desire to portray images to others, either negative or positive, that will result in personal or interpersonal rewards (Jones & Pittman, 1982). However, unlike much of impression management theory that assumes that the content of the image is highly dependent on situational demands, the image desired by the perfectionistic self-presenter is more likely to be universally one of perfection. This is consonant with Schlenker and Weigold's (1992) observation that, while some individuals present themselves favorably in particular circumstances, others indiscriminately present themselves in a positive light (see also Olson & Johnson, 1991). Self-presentation theory also offers some insight into the motivation behind perfectionistic self-presentation (see Arkin, 1986; Tetlock & Manstead, 1985). In particular, researchers have suggested that self-presentation is motivated intrapersonally by a desire for self-esteem maintenance and enhancement (Schlenker & Weigold, 1992) and interpersonally by a desire to please an audience (Baumeister, 1982) or Perfectionistic Self-presentation / 4 to avoid negative social outcomes (Baumeister & Tice, 1986; Leary & Kowalski, 1990; Schlenker, 1980). It seems reasonable to propose that the perfectionist's already fragile sense of self-esteem (Flett, Hewitt, Blankstein, & O'Brien, 1991) is protected to the extent that he or she is able to avoid criticism or to elicit esteem-enhancing reactions (see Leary & Kowalski, 1990). The strong need for approval from others that has been posited to drive perfectionism (Missildine, 1963; Rothstein, 1991) is also likely to promote a defensive posture that protects the self from being known by others as imperfect. Thus, the individual is motivated to promote a positive image or conceal negative aspects of the self by a need to protect self-esteem or bolster it through securing relationships with others. The emphasis on a motivation to present a perfect self is particularly important, as the ability to present a perfect image does not necessarily follow the desire to do so. That is, the perfectionistic self-presenter is driven to present him or herself as perfect but will not necessarily succeed. In fact, the self-protective nature of perfectionistic self-presentation may actually create a less favorable impression by promoting an avoidance of social interaction, and stimulating defensive behaviors that are viewed negatively (Holtgraves & Srull, 1989; Powers & Zuroff, 1988; Robinson, Johnson, & Shields, 1995). This is in contrast to individuals high in self-monitoring (Snyder, 1974; Synder & Gangestad, 1986) who report both an attention to social standards and an ability to modify their behavior to create a favorable impression. Dimensions of Perfectionistic Self-presentation. Self-presentation theory and research emphasizes two means by which a positive image can be promoted (Arkin, 1981; Olson & Johnson, 1991; Wolfe, Lennox, & Cutler, 1986) and these are both relevant for perfectionistic self-presentation. Wolfe et al. (1986) described acquisitive self-presentation as the process of enhancing the self or ensuring positive treatment or opinion in the present or future. Protective self-presentation is described as the process of defending against or avoiding social disapproval (Wolfe et al., 1986) and involves disavowing negative aspects of the self. Other theorists have maintained a dichotomy between self-presentational styles that Perfectionistic Self-presentation / 5 follows along much the same lines. Roth, Snyder, and Pace (1986), for example, describe the tendency to claim positive characteristics as attributive and to reject negative characteristics as repudiative. They demonstrated that these are orthogonal processes involved in presenting a positive self-image (see also Roth, Harris, & Snyder, 1988). Hewitt, Flett, Fehr et al. (1996) describe perfectionistic self-presentation as encompassing both attributive and repudiative dimensions. Perfectionistic Self-promotion reflects acquisitive self-presentation and is captured in a tendency to promote one's accomplishments, successes, or good qualities. This style results in attempts to secure the admiration of others by convincing them of one's flawlessness. This may be achieved by appearing perfectly groomed, perfectly in control, or perfectly capable. It is similar to the concept o f narcissistic perfectionism' suggested by Sorotzkin (1985) that results in prideful displays of self, and to the "self-promotion" style of Jones and Pittman (1982). In contrast, repudiative or protective self-presentation is reflected in attempts at self-concealment in order to avoid criticism. Perfectionistic self-presenters can accomplish this by two means: a lack of demonstrations of imperfections, known as Nondisplay of Imperfection, manifested in an avoidance of situations where personal flaws or shortcomings might be obvious, or by elaborate attempts to hide mistakes from others, and a lack of verbally admitting imperfections, known as Nondisclosure of Imperfections, manifested in an avoidance of verbal confessions of imperfections. This is reflected in Burns' (1980) description of the perfectionist's "disclosure phobia" and confirmed by research that links perfectionism to a reticence to disclose mistakes (Frost et al., 1995) and to evaluative concerns that reduce self-disclosure (Flett, Hewitt, & De Rosa, 1996). These dimensions are similar in intending self-concealment but differ in the types of situations that are likely to be problematic. For example, while displays of imperfection are more likely to occur in public arenas, under conditions of low volitional control, and with little chance of undoing an impression, disclosures of imperfection are more likely to occur in intimate relations under conditions of high control, and with more opportunities for compensating for a poor impression. Perfectionistic Self-presentation / 6 Relation between perfectionistic self-presentation and trait perfectionism The concept of perfectionism as a maladaptive trait has deep roots in clinical work. Karen Horney (1937) spoke of the neurotic individual who strives to be "unique and be the best in every field he comes in touch with" (pg. 189). She noted that such an individual experiences disappointment at success, because success is inevitably seen as a failure to achieve an even greater goal. Hamachek (1978) described this as "neurotic" perfectionism (in contrast to "normal" perfectionism that involves the pursuit of more reasonable goals and satisfaction with achievement; see Terry-Short, Owens, Slade, & Dewey, 19951). Within this conceptualization, there is a recognition of two components to perfectionism: a cognitive experience captured in many early measures, that includes high self-expectations (Ellis, 1962) and distorted thinking around performance (Burns, 1980), and a drive or motivation to meet these standards (Hewitt & Flett, 1990). The desire or need to be perfect adds the sting to any failure to be perfect in the perfectionist's "all-or-nothing" (Burns, 1980) or "God-or-scum" (Pacht, 1984) way of thinking. This theme of striving after unrealistic standards of performance has continued through the surge of research on perfectionism that has occurred over the last decade but has been expanded to include interpersonal components (Frost et al., 1990; Hewitt & Flett, 1990; Hewitt & Flett, 1991a,b) such as the extent to which an individual experiences others as having unrealistic standards for them (socially prescribed perfectionism) or a desire for others to meet one's own perfectionistic standards (other-oriented perfectionism; Hewitt & Flett, 1991b).2 In contrast to trait aspects that describe the desire to be perfect, perfectionistic self-presentation reflects the desire to appear to be perfect. The conceptual distinction between trait and interpersonal style is well articulated by Buss and Finn (1987). They describe traits as the content or "what" of personality, 1 Many theorists argue that "normal" perfectionism represents a need for achievement that is different from a need to be perfect and note that true perfectionists never achieve a healthy satisfaction from their achievements (Hewitt & Flett, 1996; Pacht, 1984). 2 Because other-oriented perfectionism is not generally associated with emotional distress except in a spouse or partner (Hewitt, Flett, & Mikail, 1995) it will not be discussed further here. Perfectionistic Self-presentation / 7 and contrast this with style which reflects how personality is expressed. As such, the interest is not so much on the level of perfectionism as in how the perfectionism is displayed. However, perfectionistic self-presentation does not exclusively represent an expression of perfectionism. While a need to be perfect will often include a need to appear to be perfect, it is also possible to desire to hide flaws without a desire to be perfect. In this sense, perfectionistic self-presentation is a dynamic trait, or an explanation of "why" we do what we do (Kline, 1993; Thome, 1995). In other words, perfectionism as a content trait is what people have in terms of unrealistic expectations (see Cantor, 1990), whereas perfectionistic self-presentation as a dynamic trait directly reflects the desire to promote one's perfection or conceal one's imperfection. Clinical Implications Recent theorists have proposed that personality factors can have both direct and indirect effects on psychopathology (Contrada, Leventhal, & O'Leary, 1990; Kirmayer, Robbins, & Paris, 1994). Direct effects on symptom experience and expression might include the ways in which personality characteristics generate stress or enhance the aversiveness of stress, that in turn lead to either personal or interpersonal difficulties. Indirect effects focus on the ways in which personality characteristics might interfere with coping with stress or the psychological difficulties that are associated with stress. While recent work on trait perfectionism has focused on its direct effects (e.g. Flynn, 1996; Hewitt & Flett, 1991a; Hewitt, Flett, & Weber, 1994), there are compelling reasons to consider the indirect effects of perfectionistic self-presentation, particularly in regards to an individual's ability to cope with difficulties by accessing professional help. First, the stigma attached to seeking help (see Fischer & Turner, 1970) seems to be more potent for those high in social-evaluative concerns. That is, those who are concerned about how others view them are likely to be more sensitive to the perception that help-seeking is an indication of weakness. Further, a willingness to self-disclose or to reveal secrets about the self is an important precursor to seeking therapy (Kelly & Achter, 1995; Rickwood & Braithwaite, 1994), suggesting again that perfectionistic self-Perfectionistic Self-presentation / 8 presenters may be less likely to seek professional help. This is supported by research that links the dimensions of perfectionistic self-presentation to negative attitudes towards seeking professional help for emotional difficulties (Han & Hewitt, 1996). > Second, there is a strong likelihood that the interpersonally fearful and defensive style of the perfectionistic self-presenter is a precursor of a poor therapeutic relationship. This is supported by research findings that link perfectionistic self-presentation to greater levels of distrust of mental health professionals (Nielsen et al., 1997). Interpersonal difficulties associated with such a distrustful style are often associated with an inability to establish a healthy working relationship with a therapist (Kokotovic & Tracey, 1990; Moras & Strupp, 1982; Muran, Segal, Samstag, & Crawford, 1994; Piper et al., 1991). This is particularly problematic given that the therapeutic alliance has proven to be one of the strongest and most consistent predictors of outcome in psychotherapy (see Frieswyk et al., 1986) and this is true across a broad range of problems (Gallop et al., 1994; see Horvath & Symonds, 1991). Taken together, it seems that perfectionistic self-presenters will be reluctant and fearful help-seekers in times of emotional distress. When they do present, they likely will have more difficulties engaging with the therapist. More specifically, perfectionistic self-presentation may have serious implications for the ways in which clients present in therapy, both nonverbally in terms of interaction styles, and through self-report of emotional status. Type of presentation. The idea that self-presentational concerns are relevant for psychotherapy is not new. Strong (Strong, 1982; Strong & Claiborn, 1982) suggests that psychotherapy is a process of social influence in which both parties use strategic moves to ensure particular reactions from the other. This is consistent with the many theorists who have proposed that psychiatric symptoms are used to manipulate the social environment in order to gain interpersonal rewards such as attention, sympathy, or reassurance (Bonime, 1960; Fenichel, 1945; see Snyder & Smith, 1982). Friedlander and Schwartz (1985) have extended this idea and describe a variety of strategic self-presentation styles that may be adopted in Perfectionistic Self-presentation / 9 therapy to manipulate the attributions and emotional responses of the therapist. Two are particularly relevant for the current discussion. First, Friedlander and Schwartz describe the strategy of "Self-Promotion." This style includes performance accounts, or direct claims of competence and entitlements, or the enhancement of responsibility, value, or ability for a positive outcome (cf. Jones & Pittman, 1982; Tedeschi & Norman, 1985). The intent of this style is to evoke feelings of deference or respect and it is often associated with ambivalence about therapy (Friedlander & Schwartz, 1985). "Facework," the second style described by these authors, involves efforts to avoid disapproval or blame through the use of excuses, justifications, disclaimers, or apologies. The concept of facework is based on the work of Goffrnan who emphasizes the importance of such corrective practices to manage disruptions in identity (see Semin & Manstead, 1983). Such strategies may have the added benefit of fostering empathy and support from the therapist. According to Friedlander and Schwartz, this protective style is likely to be elicited in identity-threatening situations that spoil or tarnish a positive identity. Based on the above descriptions, it seems likely that the motivation to display oneself as perfect (i.e. Perfectionistic Self-promotion) would correspond to a self-promotional style in therapy. Performance accounts and entitlements would both serve to foster the image of perfection. In contrast, the use of facework is more consistent with the perfectionistic self-presenter who desires to conceal his or her imperfections. The use of excuses, justifications, and disclaimers, will shift blame for any revealed or implied imperfections away from the self. Where perfectionistic self-presentation differs from the descriptions of Friedlander and Schwartz is in an emphasis on the motivation for the particular styles in the first place. The concept of perfectionistic self-presentation is integrally tied to the desire to be seen as perfect and not to be seen as imperfect, and less on the desire to manipulate another for specific gains. Further, Friedlander and Schwartz are less likely to see these styles as pervasive and long-standing. In essence, the styles suggested by these authors Perfectionistic Self-presentation / 10 are expressions of pathology, whereas perfectionistic self-presentation is, in its own right, a psychological problem causing emotional and interpersonal distress. Further, Friedlander and Schwartz acknowledge that such presentations will change with different levels of pathology, suggesting a degree of flexibility in styles within and outside of therapy that is inconsistent with perfectionistic self-presentation as proposed here. An example of such inflexibility comes from contrasting perfectionistic self-presentation with the typical self-presentational style of depressives. Corresponding to the "supplication" style of Friedlander and Schwartz (1985), depressives tend to present as self-denigrating or self-critical (see Seligman, Abramson, Semmel, & von Baeyer, 1979; Sweeney, Anderson, & Bailey, 1986) and this is often strongest in a social context. For example, depressives describing stressful life events under public conditions report more uncertainty, greater negative impact and more internal attributions than do depressives under private conditions (Elbin & Weary, 1987). Sacco and Hokanson (1982) found that depressives show more self-reinforcement in private than in public, whereas the non-depressed show the opposite pattern. This suggests that depressives are attempting to portray themselves as helpless and needy. The nature of attempts at self-handicapping also involve assuming a negative role. Depressed subjects are more likely to fail on a task when told that success would lead to a second task (Weary & Williams, 1990). Baumgardner (1991) suggests that even 'pointing out' a negative mood can serve as a self-handicap. She demonstrated that depressives endorsed more negative moods when depression could be used as an explanation for failure. Without negative mood being labeled a handicap, depressives presented higher mood ratings in public than in private. This self-critical and imperfect presentation seems in sharp contrast to the self-protective style of the perfectionistic self-presenter. It is difficult to imagine that someone who desires to avoid being seen as imperfect would so easily provide evidence of imperfection. Even when depressed, the drive to preserve the opinion of the other is likely to preclude self-denigrating behaviors. This would seem particularly true when the failure to maintain a positive image or avoid a negative one is likely to have further implications Perfectionistic Self-presentation / 11 for emotional distress. Thus, despite depression, the perfectionistic self-presenter will likely engage in facework rather than supplication. Assessment. Psychological symptoms are typically exposed and explored in interpersonal contexts (Hill, Weary, & Williams, 1986). Even self-report measures have an interpersonal connotation as they are completed with the assumption that the information will be shared; self-report measures thus serve as a means of communicating distress to professionals. All major clinical personality assessment tools (e.g. MMPI) include a way of assessing the likelihood that the patient is presenting him or herself in a way that is more negative than is warranted. Similarly, a measure of the tendency to deny symptoms or to endorse items in a socially desirable manner has been a standard scale in inventories in recognition of a desire to deny problems. Measures of social desirability typically provide a series of statements about behaviors or experiences that are socially undesirable but universal; high scores imply that the individual is attempting to portray him or herself in a particularly favorable light. The concern about self-presentation influences on self-report is supported by research in personality assessment (e.g. Paulhus, Bruce, & Trapnell, 1995). Clearly, people can respond to personality inventories in ways that are consistent with an intent to manage the impression of others. The attempt to promote an overly positive image (impression management) has been contrasted with an overly positive but honestly held image of the self (self-deception; Paulhus, 1986). Self-deception is considered an unconscious defense against distressing or threatening information. There is ample evidence that there exist many people who are in a state of blissful unawareness regarding their mistakes or internal motivations and that this may actually be associated with better mental health (Lane et al., 1990; see Paulhus, 1985). In contrast, impression management refers to cases when the individual is aware of the act of presenting a falsely positive image but does so for their own purposes (Paulhus, 1986). This pattern seems more likely with a perfectionistic self-presenter. Unlike a self-deceiver who is not aware of any difficulties that need to be shielded from others, the self-protective nature of perfectionistic self-Perfectionistic Self-presentation/ 12 presentation suggests that these individuals are not unaware of their shortcomings, even though they do not want others to know them. The individual who is high on perfectionistic self-presentation experiences a dilemma when they finally seek help and face an assessment. On the one hand, they are aware of the fact that they must convince the professional involved that their distress is genuine; denying their problems would be counterproductive to an attempt to get help. On the other hand, their defensiveness will create a strong desire to minimize symptoms in order to hide the fact that they are imperfect. While there is no reason to suppose that a strong motivation to present perfection will resolve this debate in a consistent manner, an important distinction may lie in the contrast between self-report and public confession. In an anonymous/private situation, a tendency towards self-awareness (low self-deception) may make it more likely that people will not deny their weakness. Thus, on self-report, the perfectionistic self-presenter will be geared toward an honest admission of their current status; as Burisch (1984) notes, this situation does not have the personal threat that might bias self-report. In a public situation, particularly one which involves a personal connection with the possibility of rejection, admitting to weaknesses should be more difficult. This is consistent with an emphasis on perfectionistic self-presentation as a presentational style, that is manifest in interactions with others. In other words, if personality style is a manifestation of "how" an individual is in their world, then it must be considered when he or she is in motion, during an interaction, as well as on self-report. Summary The concept of perfectionism involves both self-directed and social trait components, and an interpersonal expression of perfection in a consistent style. The desire to promote perfection or to conceal imperfection represent attributive and repudiative components of perfectionistic self-presentation. While self-oriented and socially-prescribed trait components reflect the struggle for self-definition, perfectionistic Perfectionistic Self-presentation / 13 self-presentation is an expression of concerns around interpersonal relationships. These different facets of perfectionism have important implications for the individual's presentation in assessment and therapy. RESEARCH OVERVIEW Conceptual Distinctions Dimensions of perfectionistic self-presentation Recent research has demonstrated that perfectionistic self-presentation consists of three components (Hewitt, Flett, Fehr, et al., 1996). First, Perfectionistic Self-promotion includes the importance of appearing perfect such as acting perfectly, appearing to be "on top of things," being perfectly groomed, and appearing to be in control. Second, Nondisplay of Imperfection, is reflected in attitudes such as believing that failing in public is awful, and behaviors such as brooding over public mistakes, attempting to cover up mistakes, and not performing an unmastered task or role in public. Finally, Nondisclosure of Imperfection, focuses on assumptions that problems and faults should be kept private, admitting mistakes to others is awful, and others should not be told how hard you work on something. Factor analytic work provides clear support for the presence and distinctiveness of these three dimensions in various samples (Hewitt, Flett, Fehr, et al., 1996; Hewitt & Flett, 1996). The correlations between these dimensions and measures of goal commitment support the contention that the desire to promote perfection is related to a desire to be seen as competent and successful, and the desire not to display imperfections is related to the motivation to avoid revealing situations and being seen as a failure (Hewitt, Flett, Fehr, et al., 1996). The initial validation study of a measure of perfectionistic self-presentation (PSPS; Hewitt, Flett, Fehr, et al., 1996), provides evidence of construct validity through relations with the dispositional tendency to make excuses for oneself (self-handicapping), the tendency to hide aspects of the self from others (self-concealment) and sensitivity to social comparison cues and to the expressive behavior of others. Although Perfectionistic Self-presentation / 14 the lack of relation between perfectionistic self-presentation and measures of social desirability and of impression management (Hewitt, Flett, & Fehr, 1994) is somewhat counterintuitive, these latter measures typically focus on highly unrealistic behaviors such as always being courteous and never resentful. This suggests that the perfectionistic self-presenter's attempts at impression management are unlikely to include overtly boastful claims and likely reflects the fact that perfectionistic self-presenters are well aware of their shortcomings, even though they experience difficulty in admitting them or displaying them to others. Finally, the desire to promote perfection and attributive self-presentation were positively correlated (but in a range that suggests that the concepts are related but not identical). While the two conceal imperfection scales did not correlate with a measure of repudiative self-presentation, on closer inspection it becomes apparent that this latter scale taps a mixture of negative behaviors that might be difficult to endorse (e.g. I am sometimes rude to other people), and negative feelings that would be consistent with psychological distress (e.g. I worry about things over which I have little control). Thus, scores on this scale are likely to represent a mixture of self-presentation and psychopathology. Perfectionistic self-presentation and trait perfectionism There is clear evidence to suggest a general link between perfectionism and perfectionistic self-presentation. In one study, Frost and his colleagues (Frost et al., 1995) had subjects high and low in a perfectionistic concern over mistakes perform tasks that involved either a high or low number of errors. They found that those high in perfectionism were likely to estimate the reactions of others to be more negative under the high mistake condition, in contrast to the low mistake condition in which they did not differ from subjects low on perfectionism. Further, subjects high in perfectionism consistently endorsed a lower willingness to disclose their performance to others, compared to those low in perfectionism and this was particularly true after having made a large number of mistakes. This supports the idea that perfectionism is associated with both a belief that others will be judgmental of errors, and a desire to conceal one's mistakes or weaknesses. Perfectionistic Self-presentation / 15 More direct evidence is also available to support a connection between trait perfectionism and perfectionistic self-presentation. Correlations between trait and self-presentation subscales range from .59 to .20 (M=-38; Hewitt, Flett, Fehr, et al., 1996) which suggests both common and unique characteristics. The strongest correlation is between self-oriented perfectionism and the desire to promote perfection; it is not surprising that high expectations for the self include high standards for appearance of perfection. The correlations between socially prescribed perfectionism and the desire to hide imperfections (r=.44 for both dimensions) is consistent with the protective nature of this style. Again, it makes sense that the perception of others as critical or overdemanding is associated with hiding one's flaws to avoid censure. Thus, it would appear that research on trait perfectionism can offer some insight into the manifestations of perfectionistic self-presentation. Despite these relations, however, it must be stressed that trait perfectionism and perfectionistic self-presentation are different constructs. First, the correlations between the dimensions as noted above are in a range that clearly indicates unique characteristics. Further, Hewitt and his colleagues (Hewitt, Flett, Fehr, et al., 1996) submitted both trait and self-presentation dimensions to a series of regressions predicting a variety of psychopathological and interpersonal outcomes. They demonstrated that different dimensions of perfectionistic self-presentation predicted significant variance even after controlling for trait perfectionism. Clinical Implications Relations with pathology Because the construct of perfectionistic self-presentation is just beginning to appear in the literature, there is little known about the ways in which it relates to intrapersonal functioning. However, preliminary evidence suggests that the dimensions are significantly related to measures of psychopathology. More specifically, the desire to conceal imperfections is strongly related to high levels of anxiety and depression, and low self-esteem (Hewitt, Flett, Fehr, et al., 1996). This is consistent with other research on Perfectionistic Self-presentation / 16 self-concealment (Ichiyama et al., 1993; Larson & Chastain, 1990) that also has been shown to relate to shyness (Ichiyama et al., 1993). These dimensions are related strongly to eating disorder symptoms (Hewitt, Flett, & Ediger, 1995)3and in clinical samples, the desire to avoid displaying imperfection is uniquely associated with both generalized and social anxiety (Hewitt & Flett, 1996). Other studies report that a defensive or protective style is associated with lower self esteem (see Tice, 1991), more social anxiety (Roth et al., 1986), and higher levels of shyness, whereas an acquisitive style shows the opposite relations with these variables (Wolfe et al., 1986). This suggests that perfectionistic self-presentation is a maladaptive self-presentational style with components that vary in their relation with maladjustment. The evidence for interpersonal difficulties in perfectionistic self-presentation, stressing the interpersonal nature of the style, is both direct and indirect. In initial studies, the desire to conceal imperfections by not displaying them or admitting to them was shown to be associated with interpersonal difficulties in being assertive and sociable (Hewitt, Flett, Fehr, et al., 1996). In addition, the avoidance of displaying imperfections was uniquely related to overcontrol and a lack of intimacy in relationships; this is, perhaps, not surprising given that those high in perfectionistic self-presentation are particularly likely to engage in self-concealment (Hewitt, Flett, Fehr, et al., 1996). Self-concealment has been demonstrated to be negatively related to network support (Larson & Chastain, 1990). The role of perfectionistic self-presentation in interpersonal distress also receives indirect support. For example, research on trait perfectionism shows that perfectionists who are concerned about being evaluated negatively in social situations are less verbally expressive (Flett, Smith, & Hewitt, 1994). There is also strong evidence that the use of defensive self-presentational strategies is not interpersonally successful. In general, people seem to prefer modestly positive self-presentations (Robinson, et al., 1995), especially for good performance (even though self-enhancers are seen as competent, they are not liked; Powers & Zuroff, 1988) and react negatively to excuses for poor performance (Schlenker & Leary, 1982). 3 While the attributive component is also related to these variables, the relations are not as strong. Perfectionistic Self-presentation / 17 Given that self-presentational style is generally mimicked by others (Baumeister, Hutton & Tice, 1989), it is also likely that a defensive or self-promotional style will promote conflict and may preclude intimacy by stimulating a similar defensive response from others. Impact on Therapy Both trait perfectionism and perfectionistic self-presentation seem to be unrelated to measures of socially desirable responding. Trait perfectionism has been examined in several studies which have found no relation between trait dimensions and the lie scale from the Eysenck Personality Questionnaire (Hewitt, Flett, & Blankstein, 1991), the Marlowe-Crowne Socially Desirable Responding Scale (MCSDS; Crowne & Marlowe, 1960; Hewitt, Flett, & Fehr, 1994) cr the Balanced Inventory of Socially Desirable Responding (BIDR; Paulhus, 1984). (Although some of the relations between these measures and socially prescribed perfectionism approached significance, all correlations were negative). The relation between perfectionistic self-presentation and these measures similarly suggests that the perfectionistic self-presenter is not likely to deny distress on self-report. As noted earlier, correlations between perfectionistic self-presentation and the MCSDS and the Impression Management subscale from the BIDR (Hewitt, Flett, & Fehr, 1994) were not significant, although there is a trend towards lower scores on impression management being associated with higher scores on the Nondisplay subscale. This may reflect an awareness of the self-presentational aspects of extreme responding to these items; people who are especially aware of the impressions they are making on others may be particularly careful not to appear overly boastful. However, given the findings that perfectionistic self-presentation is related to distress on self-report, it seems more likely that the perfectionistic self-presenter is simply responding honestly. This supports the notion that conditions of anonymous responding may be quite different from confessing the same things in person (Burisch, 1984) although this contrast needs to be established empirically. Taken together then, it would appear that the perfectionistic self-presenter desires to appear perfect but is capable of honest responding at least on anonymous self-report. This goes a long way in establishing Perfectionistic Self-presentation / 18 that perfectionistic self-presenters are not self-deceiving. Further support is found in the positive relations between perfectionistic self-presentation and Lennox and Wolfe's (1984) measure of cross-situational variability (Hewitt, Flett, Fehr, et al., 1996) that suggests that perfectionistic self-presenters are aware that their appearance can be deceiving to others. Finally, the relations between the social aspects of trait perfectionism and measures of social desirability offer tangential support to the proposition that perfectionistic self-presenters are not self-deceiving. For example, socially prescribed perfectionism is negatively associated with MMPI measures of self-deceptive defensiveness and repression (Hewitt, Flett, & Turnbull, 1992). Further analyses controlling for levels on other perfectionism dimensions support the role of socially prescribed perfectionism as a unique predictor of both low defensiveness and low repression. Summary Perfectionistic self-presentation is a personality style that includes a desire to promote an image of perfection and to avoid the appearance of imperfection. It is clearly maladaptive as it is associated with intrapersonal difficulties such as depression, anxiety and low self-esteem, and interpersonal difficulties in relationship adjustment and intimacy. Such a style presents special challenges in a clinical setting as it is likely to be part of the presentation of a significant proportion of patients, and may interfere with the process of therapy. To date, despite a strong conceptual base and preliminary evidence as to the relations between this motivation and distress, little is known about perfectionistic self-presentation. One particular limitation to the extant work is the paucity of data available on the actual manifestations of such a motivation to present as perfect. Understanding the manifestations of this style will help to clarify the links with distress and offer insight into the difficulties such individuals might bring to a therapy setting, and will help to clarify the distinction between perfectionistic self-presentation as a personality style and trait perfectionism. Perfectionistic Self-presentation / 19 MANIFESTATIONS OF PERFECTIONISTIC SELF-PRESENTATION In a recent article, J. A. Singer (1995) describes a model of personality that consists of a variety of different 'selves' that reflect different roles and expressions of personality. He outlines the role that information arising from five different 'systems of personality' (cognition, affect, behavior, motivation, and psychophysiology) plays in producing an individual's understanding of his or her self in any given context. Although Singer's intention was to outline the sources of information available to the self, this framework suggests a way of organizing information available to others that may aid in understanding an individual. The first three of these systems, cognition, affect, and behavior, have been identified by other theorists as important sources of information about the individual (see Westen, 1995). Considering multiple systems provides a much richer picture of the individual and addresses concerns about an over-reliance on self-report measures that have characterized the area of personality research (Westen, 1995). The ways that perfectionistic self-presentation might be manifest in each of these three systems are reviewed below. Cognitive aspects of perfectionistic self-presentation Information on the cognitive experience of perfectionistic self-presenters can be derived from two primary sources. First, the measure of this personality style presents subjects with a number of items that represent distorted cognitions or dysfunctional attitudes. "I must always appear to be perfect," "It would be awful if I made a fool of myself in front of others," and "Admitting failure to others is the worst possible thing" are examples of rigidity, catastrophizing, and overgeneralization (see McMullin, 1986). These types of cognitions have long been proposed as central to the experience of depression and anxiety (Beck, 1976). Secondly, the protective component to the style of perfectionistic self-presentation suggests that these individuals will appraise the social environment as threatening. That is, they are more likely to make assumptions that others will regard them negatively should they display or disclose their imperfections. Threat has been described as an appraisal of high stress relative to an appraisal of an inability to cope (Tomaka, Blascovich, Kelsey, & Leitten, 1993). Although no research has directly assessed the level of Perfectionistic Self-presentation / 20 threat perceived by perfectionistic self-presenters, the correlations between the protective types of perfectionistic self-presentation and socially prescribed perfectionism suggest that they are more likely to see others as holding high expectations for them and as being critical of their shortcomings. Two studies addressing perceived standards are of particular importance. Wallace and Alden (1991) had subjects participate in a brief interaction with a research assistant, and then asked them to rate three standards on a visual analogue scale of levels of social interaction: the level of performance that the experimenter was expecting of them for an upcoming interaction, the level of performance with which they would feel happy, and the level of performance they expected to achieve. They found that although anxious men did not rate the standards of an experimenter higher than non-anxious men, they rated their own ability and thus their ability to meet the experimenter's standards, much lower than nonanxious men. This suggests that the socially anxious may experience others as threatening because they perceive others as expecting more from them than they can deliver. This finding was replicated in a second study using a similar protocol with socially anxious women (Alden, Bieling, & Wallace, 1994). • The role of perfectionism in appraisal remains unclear. Alden et al. (1994) included the MPS and considered the role of both self-oriented perfectionism and socially prescribed perfectionism in standard setting. They found that self-oriented perfectionists established personal goals that appeared to exceed their ability. However, socially prescribed perfectionism was not related to ratings of others' standards or the discrepancy between these standards and the subject's rating of his or her own ability. It is important to extend these findings to perceptions of the standards of other referents, as in an early study by Hewitt and Flett (1991b) found strong positive correlations between the minimum and ideal standards of "someone close to you" and socially prescribed perfectionism. While it is not clear that this would hold if referencing the perceived standards of high status others, it seems likely that authority figures will create the same feelings of being pressured to perform well (see Leary, 1983a; Frost et al., 1995). Although the Multidimensional Perfectionism Scale (MPS; Hewitt & Flett, 1991b) does not generally specify a Perfectionistic Self-presentation / 21 particular referent in assessing the perceived standards of others, it seems reasonable to assume that people are not referring to a generalized other but rather to a person or persons who has some personal relevance because they are close or important in some way. It is also possible that the important perfectionism dimension for this effect is perfectionistic self-presentation. The self-protective nature of perfectionistic self-presentation is consistent with a concern for the opinions and expectations of others, while at the same time, an awareness of shortcomings will make the individual feel that they will fail to meet these expectations. Expressed in terms of appraisal, this judgment would be reflected in the perception that the other expects more than the self can give. However, the most important role for perfectionistic self-presentation in appraisal may occur after the interaction. Although the perfectionistic self-presenter desires to conceal imperfections, their scores on the self-monitoring scale noted above suggest that they also doubt their ability to do so. This is likely to translate into the belief that the other is able to "see" their imperfections and holds negative views of the self. This is supported by research suggesting that subjects who believe they do not handle social interactions well (and are more socially anxious) believe they are less liked by interaction partners compared to those who are confident in social interactions (Alden, Teschuk, & Tee, 1992). This perception is particularly important in the context of therapy, as the belief that the therapist has negative views of the self will interfere with the therapeutic alliance and may promote early termination. Affective/arousal components of perfectionistic self-presentation Emotion. The research reviewed earlier suggests that perfectionistic self-presentation is associated with negative affect such as depression and anxiety. However, the most profound effect of perfectionistic self-presentation may lie in its relation to social anxiety. In particular, the need to avoid a display of imperfection is significantly related to social interaction anxiety and a fear of public scrutiny in patient (Hewitt & Flett, 1996), and university samples (Hewitt, Flett, Fehr, et al., 1996). This relation holds over and above the relations between trait perfectionism and social anxiety (Hewitt, Flett, Fehr, et al., 1996) and Perfectionistic Self-presentation / 22 is consistent with findings that social anxiety is associated with a protective style of self-presentation during a social interaction (Meleshko & Alden, 1993). A connection between self-presentational concerns and social anxiety has been suggested in the literature, often in conjunction with cognitive appraisals. Schlenker and Leary (1982) propose that people will become anxious in situations where they are motivated to make a particular impression but also feel they will be unsuccessful in doing so. In their model, while the motivation to present in a particular way is related to the importance of the goal, and various audience and situational factors, the distinguishing feature of social anxiety is the pervasiveness of doubts about the ability to create the desired impression. This model has been termed "reactive" in that it posits that social anxiety arises from cognitive and motivational concerns (Shepperd & Arkin, 1990). Even though empirical work has failed to demonstrate a discrepancy between perceived ability and self-standards for the dispositionally socially anxious (Alden et al., 1994; Wallace & Alden, 1991), such doubts may still have a role in the experience of anxiety in a social situation. Thus, the perfectionistic self-presenter who doubts their ability to change their behavior in social situations and fears social disapproval (Hewitt, Flett, Fehr, et al., 1996), is likely to experience anxiety in social interactions; intuitively, it makes sense that someone who is concerned about being seen as imperfect would be anxious in a situation where their imperfections might be noticed. Alden et al. (1992) offer some support for this position by demonstrating that those who doubted their ability to handle an interaction well expressed more negative affect following an interaction than did those who did not doubt their ability. This is in line with the position of Arkin, Lake and Baumgardner (1986) who stress the relation between social anxiety and a fear of social disapproval. Arousal/Stress. A second area that is related to the affective experience of an individual is the physiological arousal or stress experienced in specific situations. While arousal is clearly not the same as Perfectionistic Self-presentation / 23 emotion,4 it responds to emotion (see Walreth & Stem, 1980) and is often studied in the hope that it"... can tell us something about motivation and emotion that is not necessarily obvious from overt behavior" (Fowles, 1980; pg. 93). For example, anxiety has been found to be associated with increases in autonomic system (ANS) arousal measured by cardiovascular (Linden, 1991) and skin conductance activity (see Gale & Edwards, 1986; Weinberger, Schwartz, & Davidson, 1979). There is strong evidence that the connection between emotion and arousal occurs in conjunction with cognitive appraisals of threat and challenge. In a series of studies, Tomaka et al. (1993) assessed the role of primary (how stressful the upcoming task will be) and secondary (how well do you expect to cope with it) appraisals in the subjective and physiological reaction to stressors. For tasks requiring active coping (mental arithmetic), those subjects who saw the stress as high relative to their ability to cope were classified as threatened, whereas those who saw the stress as low relative to their ability to cope were classified as challenged. The results showed that challenge appraisals were associated with less subjective stress but more cardiac reactivity and a drop in peripheral resistance (skin conductance) compared to the threat group. For tasks that precluded active coping (viewing disturbing pictures), the analyses were based on only primary appraisals of threat. Those subjects with high perceptions of threat showed greater cardiac reactivity and an increase in skin conductance activity. These results suggest that cognitive appraisals are important determinants of psychophysiological reactions to stress. There are several stimuli, relevant to perfectionistic self-presentation, that may induce stressful arousal. Most notably, situations in which the individual is required to expose their imperfections are likely to be perceived as stressful. The role of interpersonal influences in physiological arousal is supported by research that demonstrates that being observed by a stranger rather than a friend is associated with greater cardiac response to a stressful task (Snydersmith & Cacioppo, 1992). Further, given that individuals demonstrate physiological changes when talking about stressful situations that are personally 4 Arousal occurs under situations of physical exertion as well as emotional, and is associated with both positive and negative affective states, making it a nonspecific indicator of the state of the individual. Perfectionistic Self-presentation / 24 relevant and potentially embarrassing (see Guest, 1990 for a clinical application; Reidbord & Redington, 1992) and report more arousal in high-intimacy conditions (Meleshko & Alden, 1993), stressors such as the disclosure of imperfections should induce autonomic system arousal in those who are motivated or wish to hide them. Shedler and colleagues (Shedler, Mayman, & Manis, 1993) supported this by showing that subjects classified as having illusory mental health (low distress on self-report but high distress on clinical ratings) responded more defensively to a series of stimulus phrases with distressing themes, and were more physiologically reactive to a variety of different tests compared to those classified as having genuine mental health (low distress on self-report and low distress on clinical ratings). The primacy of an interpersonal stressor in arousal for perfectionistic self-presenters integrates well with the work reviewed earlier on the role of appraisal. It is likely that the confession of difficulties will be perceived as threatening, rather than challenging because of perceptions of the other. That is, if an individual feels he or she is faced with expectations they are not responsible for, it seems reasonable to assume that they would feel less control over the opinion formed by the other. This may be related to an experience of the social interaction as a passive, rather than an active coping task. Consistent with the evidence on self-monitoring (Hewitt, Flett, Fehr, et al., 1996), the perfectionistic self-presenter is likely to feel there is little he or she can do to change his or her behavior in the situation, despite the desire to do so. This is consistent with findings that social evaluation anxiety at the trait level is associated with higher state reports of anxiety in response to evaluation threats (Endler, King, Edwards, Kuczynski, & Diveky, 1983). Behavioral aspects of perfectionistic self-presentation The correlations between the different facets of perfectionistic self-presentation and repudiative self-presentation suggest that many of the behaviors of individuals high on this construct will be 'interpersonally defensive.' That is, they will be geared towards defending the image of the self to other people. Thus, some of the research on self-presentational behaviors in general may be brought to bear on Perfectionistic Self-presentation / 25 the question of behavioral manifestations of perfectionistic self-presentation. For example, self-serving attributions (and excuse making), self-flattering social comparisons and descriptions (Brown & Gallagher, 1992; Greenwald & Breckler, 1985), and self-handicapping (Kolditz & Arkin, 1982) have all been suggested as methods to maintain the image of the self, and could be used to maintain the image of a perfect self. Of these, self-handicapping, and attributions and excuses, have received the most research attention and are most relevant to the current discussion. Self-handicapping. Self-handicapping refers to actions or choices that occur prior to a performance that maximize the chance that failure will be attributed externally and success will be attributed internally (Berglas & Jones, 1978). It is a display of an impediment to performance that should divert the observer from making attributions that would blame the performer (Shepperd & Arkin, 1990). Although self-handicapping often entails the use of attributions or excuses, it may also include behaviors that will explain a poor performance such as drinking alcohol (Tucker, Vuchinich, & Sobell, 1981), choosing music that is described as interfering with ability (Rhodewalt & Davison, 1986), procrastinating (Ferrari, 1992) or failing to practice adequately (Tice & Baumeister, 1990). The claim of emotional distress such as depression to explain upcoming performance has also been suggested to relate to a need for protective self-presentation (Baumgardner, 1991). Snyder and Higgins (1988) extend self-handicapping to include excuses or attributions that occur after a poor performance, that similarly function to provide a reason for the observer to withhold blame for a poor performance. That perfectionistic self-presenters are likely to engage in self-handicapping is supported by two lines of research. First, as mentioned above, the tendency to self-handicap is positively related to the promotion of perfection and concealment of imperfection (Hewitt, Flett, Fehr, et al., 1996). This study used a measure of self-handicapping developed by Rhodewalt (1990). The scale measures the tendency to use excuses ("I tend to make excuses when I do something wrong") and the actual use of excuses ("I suppose I feel 'under the weather' more often than most people"). The correlations between these scores Perfectionistic Self-presentation / 26 and the different dimensions of perfectionistic self-presentation were all substantial (ranging from .31 to .52). Other research has demonstrated that high scores on this scale translate into more self-reported (e.g. reporting high levels of stress) and behavioral (e.g. not practicing a behavior) self-handicapping (Hirt, Deppe, & Gordon, 1991). The second support for self-handicapping in perfectionistic self-presentation is more tangential in that it assesses the role of trait perfectionism in self-handicapping. First, although lacking a strong measure of perfectionism (they used the compulsiveness scale from the MCMI), Organista and Miranda (1991) found that those higher in perfectionistic tendencies were more likely to endorse psychosomatic symptoms under conditions of achievement stress. These authors suggest that such symptoms may offer perfectionistic individuals a protective rationalization for failure.5 Similarly, Hobden and Pliner (1995) gave subjects either an experience of success or failure, and then asked them to choose between performance impairing or enhancing music to listen to while taking a second test. The choice was made under either public or private conditions. The results showed that individuals high in socially prescribed perfectionism self-handicapped more in public conditions than did those low in socially prescribed perfectionism. This pattern did not hold for those high in self-oriented perfectionism. It is suggested that this reflects the greater concern of socially prescribed perfectionists with self-presentational concerns. The connections between socially prescribed perfectionism and perfectionistic self-presentation suggest that this pattern would hold for those high in a need to present themselves as perfect. Defensive Behaviors. The tendency for people to claim credit for success and to disclaim responsibility for failure has been well documented (e.g. Pyszczynski & Greenberg, 1987; Riess, Rosenfeld, Melburg, & Tedeschi, 1981; Schlenker, Weigold, & Hallam, 1990; Weary & Arkin, 1981). Researchers have argued that the most relevant dimension lies in the distinction between internal (attributed 5 While acknowledging that these individuals may actually be experiencing more somatic events because of higher stress, Organista and Miranda point to other findings on subclinical somatization that corresponds to personality factors in the absence of illness, to support their position that these individuals do not actually experience more physical illness. Perfectionistic Self-presentation / 27 to something within the individual) and external (attributed to something external to the individual) attributions (e.g. Bradley, 1978). Self-serving attributions for failure tend to be external and for success, internal (Riess et al., 1981; Schlenker et al., 1990). Attributions are clearly responsive to interpersonal contexts. For example, when positive attributions might be offensive by suggesting egotism, subjects moderate the extent to which they claim personal responsibility for success (Miller & Schlenker, 1985; Tetlock, 1985). This same result has been obtained using self-evaluations in which the individual rates him or herself more modestly in front of an audience than in private (Brown & Gallagher, 1992). Similarly, if there is a risk that the flattering attribution might be falsified in the future, people are more cautious and claim less personal responsibility for a successful event (Schlenker, 1986; Tetlock, 1985). Riess and his colleagues (Riess et al., 1981) manipulated the perceived reliability of feedback to an audience about the subject's "real" feelings about their performance (in comparison to their attributions for their performance). They found that the subject's attributions for failure tended to be external as long as the audience knew the feedback might be false. Although often including causal explanations for events in ways that parallel studies on attributions, a new literature has recently developed around excuse theory. In ways very similar to self-serving attributions, excuses provide a means for us to manage our social identities (Mehlman & Snyder, 1985) and attempt to foster positive social relationships (Weiner, 1992). Like attributions, excuses help us to avoid blame or sanctions for a failure or shortcoming (Weiner, 1992; Whitehead, & Smith, 1986). The act is accepted but the individual's responsibility for the outcome of the act is diffused (Weiner, 1992). Unlike attributions, excuses encompass a range of behaviors that do not focus simply on realigning causal responsibility. For example, excuses can also include behaviors that attempt to manage the impact of the action (derogate the victim, deny knowledge of a problem, or argue that the damage is not that bad), manipulate the proposed standards for performance (such as questioning or reworking standards, or pointing to the good that comes from the failure or mistake) and derogate the sources of the negative Perfectionistic Self-presentation / 28 feedback (Snyder, 1985). In general, excuses function in ways similar to attributions when used to maintain self-esteem, but are more often used to manage interpersonal interactions (Weber & Vangelisti, 1991; Weiner, 1992; Weiner, Figueroa-Munoz, & Kakihara, 1991). Excuse theory also offers us an extensive range of behaviors that occur frequently in the natural environment and are readily observable (Synder, 1985). Thus they are particularly relevant for situations where we are highly motivated to hide imperfections from another. Indeed, as noted above, the dispositional tendency to use excuses to self-handicap has been found to be positively related to all dimensions of perfectionistic self-presentation (Hewitt, Flett, Fehr, et al., 1996). Summary The above discussion outlines the information available from three different systems or expressions of personality to an understanding of perfectionistic self-presentation as a personality style. At the same time, making divisions between the systems imposes an illusion of discontinuity. Rather, cognition influences affect and behavior, affect changes physiology, physiological cues effect cognition, motivation influences behavior, etc., in numerous feedback loops and subroutines (see Miller, 1996). More importantly, these systems work in tandem and information derived from one system will offer insight into other aspects of the individual. For example, if an individual experiences physiological arousal in a certain situation, it is also likely that they are experiencing some affect (positive or negative) and certain cognitions that will also translate into specific behaviors. Thus, in practice it is nearly impossible to isolate one specific system. At the same time, sampling multiple systems maximizes our understanding of the relationship between perfectionistic self-presentation and the experience of a clinical setting. The current study therefore sampled three domains of response to a clinical situation of self-disclosure: cognitive, affective, and behavioral. Psychophysiological measures were taken to assess arousal and tied to self-report and observable manifestations of distress. Perfectionistic Self-presentation / 29 RESEARCH QUESTIONS Overview The overarching goal for the current study is to document the manifestations of perfectionistic self-presentation in a clinical interview requiring self-disclosure. In order to maximize generalizability to this population and to draw some conclusions regarding the patient in therapy, attempts were made to approximate the experience of patients in a clinical setting as much as possible. However, it is recognized that subjects were aware that they were participating in a research project and that this knowledge may have had some effects on their performance during the interview. Having the individual talk about personal life events is a common methodology in clinical research (Donat & McCullough, 1983; Fincham, Beach, & Baucom, 1987; Fletcher & Fitness, 1990; Hewitt, Flett, & Callander, 1994; Holtzworth-Monroe and Jacobson, 1988; Horowitz, Rosenberg & Bartholomew, 1993; Pennebaker, Hughes, & O'Heeron, 1987) that is high in external validity. Stressful interpersonal events have served as a focus for studies of psychophysiological response (Donat & McCullough, 1983), and in the study of dyadic interactions (Fincham et al., 1987; Fletcher & Fitness, 1990). An example of this approach in the perfectionism area is a recently completed research project that assessed the interaction patterns of marital couples (Hewitt, Flett, & Callander, 1994). Using the methodology established by Fletcher and Fitness (1990), couples were asked to discuss current difficulties in their own lives (marital or family) and the verbal content of the interaction was coded along various dimensions, including criticism, justifications, or other-directed blame. In the current study, subjects were asked to discuss their own role in their present difficulties, and their contribution to other difficult situations. This is similar to what would happen in therapy, although it is recognized that such disclosures would probably not occur in the first session, and would occur with less direction on the part of the therapist. While such a design Perfectionistic Self-presentation / 30 precludes causal statements because of a lack of experimental manipulation, it is nonetheless a reasonable first step into a new area of research. In the current study, subjects completed standardized measures of perfectionism, perfectionistic self-presentation, self-presentation, appraisals, and emotional status. They were asked to provide some information on two situations in their life in which they did not cope well and these events formed the backbone of the discussion with a clinical interviewer. Three general areas were considered. First, the cognitive response of perfectionistic self-presenters was assessed by determining their appraisals of the interviewer's expectations and satisfaction regarding performance; high demand (pre-interview) relative to anticipated performance and low interviewer satisfaction (post-interview), would represent a high degree of social threat. Second, the affective response to self-disclosure was assessed to determine if perfectionistic self-presenters find self-disclosure of shortcomings distressing. This was measured by self-report of anxiety and distress, and behavioral observations. Based on the research reviewed above, one would also expect to find differences in arousal to self-disclosure based on perfectionistic self-presentation and therefore heart rate and skin conductance were monitored throughout the interview. Third, the behavioral response to self-disclosure was measured through observing defensive behaviors such as self-handicapping prior to the interview, and external attributions, excuses, justifications, and evasiveness during the interview. The consideration of multiple systems not only provides a more complete assessment of the individual, it also addresses criticisms of relying only on self-report measures to assess a construct (Westen, 1995). The dimensions included in the hypotheses regarding perfectionistic self-presentation are the Nondisclosure of Imperfection, and the Nondisplay of Imperfection. The motivation to promote perfection was not considered as it is less strongly related to symptoms of psychopathology than are the dimensions of hiding imperfections (Hewitt, Flett, Fehr, et al., 1996). Based on past research, the desire not to display imperfections is a stronger predictor of pathology than the desire not to disclose imperfections and thus Perfectionistic Self-presentation / 31 should be a stronger predictor in the current study. However, the nature of an interview situation emphasizes disclosure about personal shortcomings and thus, this factor might be expected to be particularly relevant. At the same time, relations between both dimensions and all the measures in the current study should lie in the same direction. Thus, no distinction was made between the Nondisclosure and the Nondisplay of imperfection in the following hypotheses. Hypotheses The hypothesized relations between perfectionistic self-presentation and cognition, affect/arousal, and behavior were tested at multiple levels. First, a basic bivariate relation was proposed, with perfectionistic self-presentation predicting more negative or threatening appraisals, more negative affect and greater arousal, and more defensive behaviors as outlined below. This relation was tested for a unique contribution of perfectionistic self-presentation over and above control variables. For all equations, controls consisted of a block of four variables. First, gender has been related to levels of self-disclosure and comfort with self-disclosure in other research (see Derlega et al., 1993) and was related to more negative mood following the interview in the current sample. Similarly, treatment-naive subjects should be more anxious during the interview and so past history with a mental health professional was included. Finally, two variables were included because of significant relations with dependent variables in the current sample: Age and education were shown to be related to appraisals (younger and less educated subjects were less optimistic and more threatened) and to be associated with somewhat less positive mood following the interview. All four control variables were included in each regression for consistency. Two additional controls, medication usage and the seriousness of the problem being discussed, were included in the regressions predicting physiological arousal. Although establishing that perfectionistic self-presentation makes a unique contribution to the experience of the interview over and above such variables as age, education, gender, and past experience is important, it is also important to establish that perfectionistic self-presentation adds something above trait Perfectionistic Self-presentation / 32 perfectionism. It is possible that appraisals in particular, are driven by the extent to which someone has unrealistic expectations for the self or reflect only the feeling that others have high expectations; distress may come from perceptions that these expectations are not being met. Thus, a second set of analyses was run testing the hypothesis that perfectionistic self-presentation is not redundant with trait perfectionism but rather adds unique explanation. Finally, there is a distinct possibility that the experience of a clinical interview is driven by the level of emotional distress. As noted above, depression has been linked with self-handicapping and appraisals, and social anxiety clearly has a role for anxiousness in an interpersonal encounter. Particularly given that both dimensions of perfectionistic self-presentation have been shown to be strongly related to these variables, it seems important to establish a unique role for perfectionistic self-presentation. At the same time, it is possible to address one rival hypothesis to observed behavior in particular, and that is that behaviors reflect the tendency towards impression management rather than a desire to avoid being known as imperfect. Thus, in order to test whether perfectionistic self-presentation adds unique variance over and above the contribution of emotional distress (interaction anxiety and depression), and to demonstrate that it differs from impression management, analyses were repeated controlling for these variables. Because including all predictor variables in one equation constricts the final degrees of freedom and results in power estimates between .56 and .72 given the current sample size, these two hypotheses were tested separately. Cognitive manifestations of perfectionistic self-presentation The first area to be examined involved the cognitive processes involved in perfectionistic self-presentation. It was expected that those high in a need to avoid displaying or disclosing imperfection would appraise a social interaction as threatening. More specifically: Hypothesis 1. Perfectionistic self-presentation will be positively related to appraisals of the interviewer as threatening and judgmental (having high expectations, wanting more than the subject is able to give, and being disappointed following the interview). Perfectionistic Self-presentation / 33 Hypothesis la. These relations will hold over and above the influence of trait perfectionism. Hypothesis lb. These relations will hold over and above the influence of emotional distress and a tendency towards impression management. Affective and psychophysiological manifestations of perfectionistic self-presentation The second area involves the affective/physiological experience of those high in perfectionistic self-presentation. It was expected that these individuals would be more anxious when anticipating the interview, and would display more physiological arousal when discussing past shortcomings and mistakes, than would individuals low in this style. More specifically: Hypothesis 2. Perfectionistic self-presentation will be positively related to ratings of negative affect prior to the interview, to observable anxiety during self-disclosure and to self-ratings of negative affect following the interview. Hypothesis 2a. These relations will hold over and above the influence of trait perfectionism. Hypothesis 2b. These relations will hold over and above the influence of emotional distress and a tendency towards impression management. Hypothesis 3. Perfectionistic self-presentation will be positively related to physiological arousal and reactivity during self-disclosure. Hypothesis 3a. These relations will hold over and above the influence of trait perfectionism. Hypothesis 3b. These relations will hold over and above the influence of emotional distress and a tendency towards impression management. Behavioral manifestations of perfectionistic self-presentation The final question to be addressed regards the ways in which individuals with strong motivations to hide their imperfections actually behave in a clinical interview. It was expected that those patients with high levels of perfectionistic self-presentation would present as more defensive and demonstrate more difficulties with self-disclosure. More specifically: Perfectionistic Self-presentation / 34 Hypothesis 4. Perfectionistic self-presentation will be positively related to self-handicapping, and defensive behaviors such as the use of external attributions, excuses, justifications, and evasiveness, in explanations for failure or shortcomings. Hypothesis 4a. These relations will hold over and above the influence of trait perfectionism. Hypothesis 4b. These relations will hold over and above the influence of emotional distress and a tendency towards impression management. Exploratory Analyses Evidence for gender differences in levels of perfectionism is mixed; some studies find that women are lower on other-oriented perfectionism (Flett, Hewitt, Blankstein, & Mosher, 1995) and higher on socially prescribed perfectionism (Hewitt, Flett, Turnbull-Donovan, & Mikail, 1991) while others find no differences (Hewitt, Flett, & Endler, 1995; Flett, Hewitt, Endler & Tassone, 1995). While on the whole, men and women do not differ in perfectionistic self-presentation, there is some evidence that women score higher on the Nondisplay of Imperfection subscale and men higher on the Nondisclosure of Imperfection subscale (Hewitt, Flett, Fehr, et al., 1996). Similarly, there is research that suggests gender differences in the experience of a social interaction, with women being more self-disclosing than men. However, much of the research is unclear. For example, although some studies find that men self-handicap more than women (Hobden & Pliner, 1995), others show no differences on self-reports of self-handicapping (Ferrari, 1992) and still others find that the type of self-handicap used by men and women differs (Hirt et al., 1991). On the whole, we know little about how men and women differ in their presentation to or during a clinical interview. Thus, although no specific hypotheses about gender differences were made, these were explored. METHOD Participants Participants were selected to reflect the mixed diagnostic group typically seen in clinical settings in order to increase the generalizability of the results of this study; such a group is also consonant with the Perfectionistic Self-presentation / 35 view of perfectionistic self-presentation as a personality style, independent of psychiatric diagnosis. Although diagnostic status was not known, participants presented with a range of difficulties such as depression, anxiety, eating disorders, and relationship and work difficulties.6 Participants were excluded if they were suffering from a psychotic disorder. Because issues of self-presentation in different cultures have not been adequately addressed in the research literature, all participants were of European or North American origin. Only one participant spoke English as a second language but had spoken English since adolescence and was clearly fluent. A total of 106 patients participated in the study. Of these, 7 were excluded because of equipment failures of either the physiological monitor or the videotape. Three participants were unable, and 1 unwilling, to complete the study protocol. One subject was inebriated, and one was heavily sedated with anxiolytics and so both were dropped from the analyses. Finally, 3 participants were deemed inappropriate referrals because they either were not seeking treatment in any form or had a history of schizophrenia. The participants that were excluded did not differ significantly from the final group in demographics or perfectionism. Of the final sample of 90 participants (45 women and 45 men), 31 (34%) came from the Affective Disorders Unit at the Vancouver Health Sciences Center, UBC site (VHSC-UBC), 51 (57%) from the Psychology Outpatient Clinic at the University of British Columbia, and 8 (9%) from the Outpatient Psychiatry Units at either the Vancouver Health Sciences Center - UBC Site (VHSC-UBC) or St. Paul's Hospital. This number allows for adequate power for most analyses (Green, 1991).7 Participants ranged in age from 19 to 64 years (M=36.20, SD=11.06; Men =39.24/11.07; Women=33.16/10.29), with an average education of 15.08 years (SD=3.03; Men=15.04/3.16; Women=15.31/2.93). Sixty-one percent of the group had never been married and 23% were currently 6 Approximately 60% of participants described depression as their primary concern, 10% relationship issues, 10% adjustment issues or situational stress, 11% anxiety, and 8% eating disorders. 7 Using the data available on perfectionistic self-presentation to estimate effect size (Hewitt, Flett, Fehr, et al., 1996), power estimates for regressions using 3 steps and 90 subjects are in the range of .86 to .94 (Borenstein & Cohen, 1985). Perfectionistic Self-presentation / 36 married (Men=49% and 33% respectively; Women 73% and 13% respectively). Sixty-four percent were currently employed (Men=60%; Women=69%). There were no differences in age or education based on the site of referral. A similar lack of differences on measures of depression and interaction anxiety, and on the frequency of being on psychoactive medication, suggests that the different sites provided a relatively homogeneous sample of subjects. Men were significantly older than women (t£88) = 2.7, p<01) but did not differ on depression, interaction anxiety, or medication use. Comparisons with other clinical patients suggested that the current sample was quite similar to other clinical samples in levels of depression (Beck & Beck, 1972; Steer, Beck & Brown, 1989) and in line with expectations for interaction anxiety (Leary & Kowalski, 1993).8 Further, comparisons with a large sample of clinic patients who filled out the MPS and PSPS as part of several self-report studies (Hewitt & Flett, 1996) suggested that the current sample was fairly representative of a mixed outpatient sample, as perfectionism scores for the current study were not significantly different from the larger sample. This is important, as it establishes that volunteers in the current study did not reflect a group especially low on self-presentation concerns, as might be expected in those that volunteer for an interview study. Measures (Appendix A) Independent Variables The Perfectionistic Self-Presentation Scale (PSPS; Hewitt, Flett, Fehr, et al., 1996) was developed using the construct validation approach (Jackson, 1970); items reflecting both acquisitive and repudiative self-presentation were generated and tested. Analyses based on a large sample of students confirmed the multidimensionality of the styles and factor analysis supported a three factor structure. Alpha coefficients for the three subscales, Perfectionistic Self-promotion, Nondisplay of Imperfection (referred to as Nondisplay), and Nondisclosure of Imperfection (referred to as Nondisclosure) in the original research showed good internal consistency (.86, .83, and .78, respectively) and test-retest analyses 8 The mean on the IAS in the current study was one standard deviation above those reported in community samples (Leary & Kowalski, 1993) as would be expected in a mixed clinical group. Perfectionistic Self-presentation / 37 showed that the factors were stable over time. As noted earlier, while there is some overlap between such measures of sensitivity to social cues and perfectionistic self-presentation that support the scale's construct validity, the size and pattern of these results suggest that the concepts are reliably distinguishable (Hewitt, Flett, Fehr, et al., 1996). In the current study, the correlation between the scales was .61. In addition to the self-report measure, interviewer ratings were made during the administration of the Interview for Perfectionistic Behavior (see below). These ratings were based on a discussion of the different dimensions and the subject's reactions to making mistakes and disclosing mistakes. Although these ratings represent only a preliminary attempt to validate the self-report scale, the correlations for the Nondisplay and Nondisclose subscales (r=.63 and .61 respectively), offer evidence of convergent validity across different assessment approaches. Control Variables Perfectionism traits. Two aspects of trait perfectionism are important in the current study: self-oriented perfectionism (holding unrealistic standards for the self) and socially-prescribed perfectionism (believing that others hold unrealistic standards for the self). The Multidimensional Perfectionism Scale (MPS; Hewitt & Flett, 1991b) is a self-report scale, developed and validated using university students, clinical and community samples (Hewitt, Flett, Turnbull-Donovan, & Mikail, 1991). Subscale reliability coefficients are reported as .88 for Self-oriented and .81 for Socially-prescribed perfectionism, and test-retest coefficients as .75, and .78 respectively (Hewitt & Flett, 1991b). The subscales demonstrate good concurrent and discriminant validity in clinical samples (Hewitt, Flett, Turnbull-Donovan, & Mikail, 1991). A short 15 item form of the MPS was used; this form has shown good reliability and scores based on the reduced number of items have been demonstrated to show similar relations to other variables as scores based on the full scale. The Interview for Perfectionistic Behavior (IPB; Hewitt, Flett, Flynn, & Nielsen, 1995) was included to confirm levels on the perfectionism variables. This standardized interview includes questions Perfectionistic Self-presentation / 38 assessing each of the dimensions of perfectionism and requires approximately 10 minutes to administer. It has been shown to be predictive of scores from the MPS (range between r=.55 to .67; Hewitt, Flett, Flynn, & Nielsen, 1995). The reliability and validity of this interview is being evaluated through ongoing research. In the current study, the correlations between the MPS and the IPB for both scales were .53; it is possible that this interview is less sensitive to clinical populations although this needs to be explored further. Psychological distress. The Beck Depression Inventory (BDI; Beck, Ward, Mendelson, Mock, & Erbaugh, 1961) is a short self-report measure of depressive symptoms that is commonly used in clinical research. It has been demonstrated to have good reliability and validity (see Beck, Steer, & Garbin, 1988). A short form of the BDI was used (Beck & Beck, 1972) which has shown good reliability and compares favorably with the full scale (Beck, Rial, & Rickels, 1974). In order to address the role of perfectionistic self-presentation over and above social anxiety in the presentation of patients, interaction anxiety was measured. The Interaction Anxiousness Scale (IAS; Leary, 1983) is a 15 item scale that assesses the experience of subjective anxiety in social situations. Although not widely used with patient samples, it has an advantage in the present study in that it excludes measures of inhibited or reticent behaviors that overlap with self-presentation. The reliability for the measure ranges from .87 to .89 across a number of different groups, and the measure demonstrates good construct and criterion-related validity through its correlations with other measures of social anxiousness and shyness (Heimberg, Mueller, Holt, Hope, & Liebowitz, 1992; Jones, Briggs, & Smith, 1986; Leary & Kowalski, 1987) and its ability to predict anxious behavior (DePaulo, Epstein, & LeMay, 1990). Self-presentation. One other measure of the tendency to self-present was included. The Balanced Inventory of Desirable Responding (BIDR; see Paulhus, 1994) produces a measure of impression management that shows good internal consistency (alphas are reported to range from .75 to .86) and positive relations with other measures of impression management (Paulhus, 1991). This 20-item subscale Perfectionistic Self-presentation / 39 (the other subscale measures self-deceptive enhancement and was not included here) can be scored as a mean for all items, or a frequency count of items that are endorsed in the extreme. As Paulhus (1994) suggests that a frequency count provides a better representation of subjects who are high in this style, that is the approach used in the current study. Past experience. While it is likely that the research setting will be different enough from other settings to be somewhat anxiety-provoking, it is possible that any subjects who have been through clinical interviews in the past will experience this situation as less stressful (see Cacioppo, 1994); thus, past experience (or lack of experience) with a mental health professional (MHP) was entered as a control variable. Eighty-four percent of subjects had had such experience. Additional controls for the physiological analyses. Medication status was represented by a dichotomous variable representing presence or absence of a regular psychoactive medication because such medication may have implications for measures of arousal. Of the total sample, 53% were taking regular medication. While the evidence of an effect of antidepressants on reactivity is mixed (Blondin & Waked, 1991; Donat & McCullough, 1993; see Zahn, 1986), the use of anxiolytics should decrease the experience of arousal in the interview (Zahn, 1986). In the telephone contact prior to the interview, participants were asked to refrain from taking any medication that they do not take on a regular basis less than 3 hours prior to the interview; almost all of such medications are short acting anxiolytics (Canadian Pharmaceutical Association, 1996). That is, they were asked to refrain from "pm" or "as needed" medication to eliminate the effects of this type of medication. Only one subject failed to comply with this requirement and his data was dropped from the physiological analyses.9 All other subjects reported taking only their regular medication on the day of the interview. Because caffeine will increase basal arousal level (Lane & Adcock, 1990), subjects were asked not to drink coffee, tea, or cola within 2 hours of the interview. Two subjects failed to comply with this 9 Because removing his data from the analyses of the self-report measures of mood made essentially no difference to the results, his data was not removed from the other analyses. Perfectionistic Self-presentation / 40 request. However, as their data were not univariate or bivariate outliers, and made no difference to the results, they were included for completeness. Finally, smokers were requested to refrain from smoking at least 30 minutes prior to reporting for the study to balance increases in basal heart rate due to nicotine with the increase in arousal associated with nicotine withdrawal (Parrott & Joyce, 1993); all subjects complied with this request. While it is highly likely that discussing the reason why they have sought treatment and any level of self-disclosure around a past failure will be stressful, the two difficult situations provided by the subject will have some influence on how the subject reacts to talking about them. Thus, the seriousness of the situation was assessed by raters based on the subject's initial description of the situation during the interview, on a 9 point Likert scale from not at all serious to very serious. A similar method has been used to rate the level of self-disclosure in other research (Pennebaker et al., 1987). Because it is possible for a situation to be quite serious but involve a low level of culpability for the subject, raters also assessed the extent to which the situation was one for which the subject could be blamed. Correlations between these ratings of .48 and .44 (p<001) for the first and second situations respectively, suggest that these dimensions overlap but are not synonymous. However, because the ratings of seriousness showed much stronger correlations with heart rate and are more consistent with the literature, ratings of seriousness were used as the control variable. Correlations between two independent raters for each situation showed good interrater reliability (r=.85 for the first situation, r=.80 for the second). Dependent Variables Appraisals. As noted by Wallace and Alden (1991) there has been little research examining the role of appraisals in the context of a social interaction. In particular, few studies have assessed standards for an upcoming or recent interaction, focusing rather on global assessments of perceived ability. For example, research has measured the standards of "the typical student" (Ahrens, Zeiss, & Kanfer, 1988) rather than focusing on the perceptions of their expectations for the self. However, the Behavioral Perfectionistic Self-presentation / 41 Ratings of Social Competence scale (BRSC: Lewinsohn, Mischel, Chaplin, & Barton, 1980) has been used to assess specific social behaviors based on self- and other-ratings (Gotlib & Meltzer, 1987; Lewinsohn et al., 1980). Ducharme and Bachelor (1993) used the scale in a slightly modified version to assess subjects' predictions of their behavior in an upcoming interaction. The scale discriminates between depressed and nondepressed subjects (the depressed seeing themselves and those they interacted with as less competent: Gotlib & Meltzer, 1987; Youngren & Lewinsohn, 1980) when rated by subjects (Gotlib & Meltzer, 1987) and observers (Youngren & Lewinsohn, 1980). The scale is comprised of 7-point Likert items that subjects rate according to how characteristic the item is of their actual performance or likely to be of their upcoming performance. The anchors for this scale are from not at all characteristic to extremely characteristic.10 While some studies have used an upper limit of "perfectly" in order to anchor the measurement in extreme standards (Ahrens et al., 1988), this stem may be confusing because it confounds the level on the characteristic with appropriateness; one can be perfectly assertive (for the situation) at moderate levels. It is possible that subjects would also see "extreme" as meaning inappropriate. Therefore, in order to clarify the upper end of the scale, subjects were instructed as follows: "by extremely, we do not mean an excessive, inappropriate, or exaggerated level, but rather one that fits perfectly with the situation." Items were: Friendly, assertive, warm, attractive, communicates clearly, socially skillful, confident, humorous, reasonable, open, and speaks fluently. Six items (popular, interested in other people, has a positive outlook on life, understands what others say, trusting, and notices good experiences) were excluded because they are inappropriate for a structured dyadic clinical interview; low ratings on these items might reflect the lack of information It is acknowledged that these ratings are highly subjective - one subject's "extreme" may correspond to a different actual behavior than another's "extreme." This is the difficulty addressed by the use of a visual analogue scale in the studies by Alden and her colleagues. However, the current study is not concerned with how expectations compare across subjects as much as it focuses on expectations relative to perceived ability (also subjective). Like research in the stress and coping literature that does not equate the severity of the stressor by other than the subject's appraisal of the stressor, what is of interest here is how the subject appraises their social environment as supportive or judgmental. Perfectionistic Self-presentation / 42 available from the interview (where what is talked about is negative and constrained) on that characteristic, rather than a true assessment of low competence in that area. Internal reliability estimates for the scale range from .88 to .97 (Gotlib & Meltzer, 1987; Lewinsohn et al , 1980). In accordance with the method used by Alden (Alden et al., 1994; Wallace & Alden, 1991), the scale was filled out under two conditions to establish appraisals prior to the interview. First, subjects were asked to respond to the questions according to "what point on the scale represents what you believe the interviewer expects from you on this characteristic in the upcoming interview?" To avoid confusion between what the interviewer believes will happen and what she would like to have happen, this was followed by the statement "Not how she thinks you will do, but what level of performance she would like you to achieve." A composite score represented perceived interviewer expectations. However, in order to establish that the interviewer is seen as threatening, it is necessary to demonstrate that the subject sees the interviewer as wanting more than the subject feels he or she can provide. Thus, subjects completed the scale in response to the question "what point on the scale represents how you think you will perform in the upcoming interview?," reflecting anticipated performance. Following the interview, participants again filled out the scale twice, but this time in response to the prompts "what point on the scale represents how you were on this characteristic during the interview?" and "what point on the scale represents how satisfied you think the interviewer is with how you were on this characteristic during the interview?" Composite scores represented perceived performance and perceived interviewer satisfaction. Following the procedure of Alden et al. (1992), subjects were also asked to make a global rating of how likable the interviewer found them (on a 7 point scale from not at all to extremely) and a rating of how likable they believe the interviewer would usually find others on a similar scale. The discrepancy between interviewer standards and anticipated performance was used as a measure of interpersonal threat. Discrepancy scores have been used to assess differences between self-Perfectionistic Self-presentation / 43 efficacy and interpersonal standards (Kanfer & Zeiss, 1983) and differences between perceived demands and ability to meet those demands (Tomaka et al., 1993) in other research. While concerns have been raised regarding the use of change or discrepancy scores by statisticians (see Cronbach & Furby, 1970) who argue that there are no a priori means of establishing the extent to which the variable to be removed should be weighted by other than the value of 1.0 (this being the weight assumed by the standard equation of Y-X=Difference), this is preferable to a covariance approach when there are initial differences between groups (in our case, a possible positive relation between perfectionistic self-presentation and perceptions of the interviewer; Sechrest, 1984). Self-reported and observer-rated distress. Participants were asked to rate their negative mood on the negative affect subscale of the Positive Affectivity Negative Affectivity Scale (PANAS; Watson, Clark & Tellegen, 1988). The scale asks subjects to rate the extent to which 10 negative mood adjectives are descriptive of how they are or were feeling over a given time period, in this case, both prior to and over the course of the interview. Internal reliability is reported as high (consistently above .84) and intercorrelation between the scales is low; these scales also correlate appropriately with other affect measures (Watson et a l , 1988). This scale has been used in numerous studies, with different populations (e.g. Mendolia & Kleck, 1993; Watson, Clark, Mclntyre, & Hamaker, 1992). Although not required for testing the current hypotheses, the PANAS was also completed by the interviewer, in accordance to how they believed the subject felt during the interview. Four additional items, "Defensive," "Frustrated," "Impatient," and "Inhibited" were added based on suggestions from other psychotherapy research on important client variables (O'Malley, Suh, & Strupp, 1983). Many studies have included observer ratings of subject discomfort, using two types of operationalizations. First, some studies measure discomfort in terms of specific behaviors such as smiles or counts of specific types of utterances (e.g. Dow, Biglan, & Glaser, 1985; Katz et al., 1984) or duration and latency of response (e.g. Van Dam-Baggen, Van Heck, & Kraaimaat, 1992). Many of these ratings Perfectionistic Self-presentation / 44 rely on the subject's ability to act freely in the situation, making them less appropriate for interactions that are scripted and other-directed, and are confounded with social skill. Alternately, studies use global ratings of "anxiety" (Dow, Biglan & Glaser, 1985; Falloon, Lloyd, & Harpin, 1981; Mendolia & Kleck, 1993) without reference to specific features of the presentation, an approach suited to the current study. Interviewers were asked to rate observable anxiety over the course of the interview, and two coders were trained to rate observable anxiety over each section of the interview from 1 not at all to 9 extremely anxious. These coder ratings were highly correlated for the "reason here" (r=.85) segment of the interview, and reasonably correlated for the discussion of the "first situation" segment (r=63). While the "second situation" anxiety ratings showed a substantially reduced correlation (r=39), the fact that 95% of ratings were within 1 point of the second rater suggests that there was some restriction of range and examination of the data suggests that there was a generally low level of observed anxiety for the second situation. The ratings were also significantly correlated with interviewer estimates of subject's anxiety (ranging from .41 to .35, p<001); 66% of the time, they were within one point of agreement for the first situation, and 68% for the second. More specifically, videotapes of at least one segment from 30 subjects were coded for anxiety by a separate clinical judge. Of the 33 segments coded, 77% were either identical or within one point of the original coders. Of those codes that were not identical, 83% represented cases where the clinical judge coded less anxiety than the original coders. Taken together, this suggests that subjects did indeed display low levels of anxiety overall during the interview. Physiological monitoring. In line with appraisal-based models of emotional reactivity (Mendolia & Kleck, 1993; Smith & Ellsworth, 1985) measures of ANS arousal should reflect distress during self-disclosure. The Davicon M E D A C System/3 provides continuous monitoring of heart rate and skin conductance. This equipment uses a pulse plethysmograph and skin electrodes attached to the fingers. Skin conductance is measured through dry gold-plated sensors with an area of one square centimeter, with Perfectionistic Self-presentation / 45 a constant current density of 3uA/cm2. Participants were informed that the instrument cannot function as a lie-detector, but rather measures levels of relaxation. While many studies to date have focused on heart rate response, researchers have stressed the advantages to measuring multiple channels of reactivity (e.g. Gale & Edwards, 1986; McHugo & Lanzetta, 1983). Skin conductance, or the ease with which an electrical current is conducted along the skin's surface, is influenced by the degree of activity of the eccrine sweat glands and has been reliably used to measure response to stress (Boucsein, 1992). Average skin conductance level (SCL) was chosen in order to maximize comparability to other research. Although generally both these channels should show similar patterns, it is acknowledged that this is not always the case (Dillard & Kinney, 1994; Wallbott & Scherer, 1991). Heart rate has been shown to increase in the absence of skin conductance under conditions of challenge and skin conductance to respond to emotional threat in the absence of heart rate change (Lang, Levin, Miller & Kozak, 1983). Part of the explanation for this divergence may lie in the fact that, while both respond to autonomic system arousal, innervation for the sweat glands is cholinergic rather than the noradrenergic system associated with cardiac activity (Venables & Christie, 1980). However, because there is no way to anticipate divergence between measures on an a priori basis, hypotheses focused on increases in both heart rate and skin conductance when discussing mistakes. Scores on heart rate and skin conductance were calculated for each time period as follows: the last minute of the relax phase (pre-interview), the first minute of the discussion on the reason for the referral, the initial minute following the interviewer's request for the disclosure of each problem, and the final minute of relaxation at the end of the interview. The use of one minute blocks was necessary because of the variability in the length of responses across subjects; all subjects spoke for at least one minute in each section although many subjects spoke for longer. The first minute is also most appropriate as subjects who spoke at length may experience some habituation of the response (Frankish & Linden, 1991). One minute Perfectionistic Self-presentation / 46 segments should minimize the effect of transient changes and give a more reliable and meaningful estimate of arousal than would be obtained over briefer time periods. In addition to considering average levels of heart rate or skin conductance at each time point, it is important to consider the extent to which the subject changed over time; that is, the change in arousal over and above relaxation. As noted earlier for the cognitive threat scores, the use of change scores is also an issue in calculations of psychophysiological reaction. While some studies make use of repeated measures analyses (Buntrock & Reddy, 1992; Donat & McCullough, 1983; Kaiser, Hinton, Krohne, Stewart, & Burton, 1995), others use the discrepancy between baseline measures and measures during stress as an assessment of reaction (Mothersill, Dobson, & Neufeld, 1986; Newton & Contrada, 1992; Pennebaker et al., 1987; Shedler et al., 1993; Weinberger & Davidson, 1994). Some reviewers stress the difficulties inherent in difference scores (Siddle, Turpin, Spinks, & Stephenson, 1980) although again, others note that either an unweighted difference score or a normalized difference score can be appropriate in tests of person variables in different situations (see Ben-Shakhur, 1985; Stemmler, 1992). Recent work predicting change in heart rate from a number of different statistical approaches has suggested that change scores are the most appropriate for this type of analysis (Cribbie & Jamieson, 1997). In order to avoid multiple tests of scores based on the same value (relaxation), it was decided to focus on the arousal or difference in physiological measures from relaxation to the discussion of the first serious mistake, as this is the first and likely the most significant threat to an image of perfection during the interview; having disclosed the most serious mistake, a second mistake should add little in the way of new information to the interviewer. Defensive Behaviors. Several authors have emphasized the distinction between acquired and claimed (Arkin & Baumgardner, 1985) or behavioral and self-reported self-handicapping (Leary & Shepperd, 1986). Acquired or behavioral self-handicapping reflects constructed impediments to success that typically include behaviors such as a lack of practice or an impairing drug, whereas claimed or self-reported handicaps are those that are claimed verbally and can include such things as stress or mood. A Perfectionistic Self-presentation / 47 second and overlapping distinction is the extent to which the handicap is self-evident or requires an avowal from the self-handicapper (Snyder & Smith, 1982). While behavioral handicaps tend to be more observable, most studies remove the need for avowal by defining, a priori, the handicapping nature of a behavior or emotional state to the subject. An exception to this is a study by Hirt et al. (1991) who found evidence for self-reported self-handicapping (high levels of stress) in those high in a tendency to self-handicap, even when stress was not specifically labeled as a potential explanation for failure on the task (see also Smith, Snyder, & Handelsman, 1982). Although not able to address true differences in levels of stress for self-handicappers, this study provides preliminary evidence that subjects who tend to self-handicap (consistent with perfectionistic self-presenters) will spontaneously provide reasons why they might fail at a task. The current study followed this research by measuring self-reported self-handicapping in the absence of delineations to the subject as to what would constitute a handicap in the interview. Participants were provided the opportunity to make avowals prior to the interview of limiting factors on their potential performance during the interview. That is, subjects were asked if there was anything that might affect their performance on the upcoming interview. Providing this opportunity for subjects to explain poor performance by other means early in the questionnaire package should also decrease any need to exaggerate symptoms in order to explain a poor performance. For each reason given, subjects estimated the amount that reason would probably change their behavior on an 11 point Likert scale from (-5) much worse to (+5) much better. The absolute value of negative scores formed the measure of extent of self-handicapping. The highest level of negative impact was used as the self-handicapping score.11 Following the interview, the same procedure was used to allow subjects to explain or excuse their behavior during the interview - subjects were asked if there was anything that may have changed their performance on the 1 1 The maximum response was chosen because it represents the maximum amount the subject hopes to be excused, after which additional excuses become redundant. For example, if the individual gives a reason for poor performance that will make performance "much worse," then other reasons cannot add any additional explanation for their behavior. Perfectionistic Self-presentation / 48 interview and to what extent. It must be recognized that this method of assessing self-handicaps, while following closely from the definition of the phenomena, has not been widely used to date. As a result, the Self-Handicapping Scale (SHS; Jones & Rhodewalt, 1982) was included to provide convergent validity for these scores. This measure of the dispositional tendency towards self-handicapping has been shown to predict self-handicapping behavior (Hirt et al., 1991). A short version (10 items; Strube, 1986) shows adequate internal consistency (alpha = .70),12 and validity (Strube, 1986). However, the reliability of the scale in the current sample was only .47, suggesting that, at least for this group, the scale may be measuring several different constructs. Indeed, the significant correlations with depression (r=.57) and interaction anxiety (r=.43) suggest that this scale may reflect distress in a patient group. Unfortunately, it makes the nonsignificant correlations between this scale and self-handicapping in the current study difficult to interpret. Because a primary interest is in the extent to which the subject tries to avoid blame for their role in difficult situations, raters were asked to judge the extent to which the subject seemed to be attempting to place the cause of their problem outside of him/herself. Using raters to judge Externality of attributions has been shown to be effective in other research (Habke, 1991; Peterson, Bettes, & Seligman, 1985). In order to compare observable externalization with self-report of externalization, as part of the post-interview packet participants were asked to estimate the extent to which the difficult situation was due to something about them or to something about others or the circumstances (from completely due to me to completely due to others/circumstances on a 7 point scale) based on the scoring procedure developed by Peterson, Schwartz, and Seligman (1981). Although these ratings should be correlated (Peterson et al., 1981), in the current study they were not (r=. 17, p_>. 10 for both situations). This may reflect unreliable coder ratings, or may point to a difference between attributions assigned on self-report and implied by behavior. 1 2 This is an improvement over the alpha for the original scale that is reported as .63 (Jones & Rhodewalt, 1982). Perfectionistic Self-presentation / 49 A second form of defensive behaviors arises from this study's focus on personal contributions to current and past stressors. Participants were asked to provide accounts for their difficulties; prompting the subject to voice such attributions should overcome some of the difficulties in coding spontaneous attributions (see Holtzworth-Munroe & Jacobson, 1988). Each reason given following a question regarding the subject's role in the situation, was coded from transcripts according to the account taxonomy used by Gonzales and colleagues (Gonzales, Manning, & Haugen, 1992; Gonzales, Pederson, Manning, & Wetter, 1990; see also McLaughlin, Cody & O'Hair, 1983). These researchers adapted the categories proposed by Schonbach (1980) to include four major categories: Concessions (e.g. explicit acknowledgment of guilt or regret), Excuses (e.g. appeals to own shortcomings or others involved in event), Justifications (e.g. minimization, or appeal to positive intentions), and Refusals (e.g. denial of event or responsibility, other-directed blame). Using two coders, these authors were able to reliably classify spontaneous account statements to an average concordance of .78 (Gonzales et al., 1990) to .86 (Gonzales etal., 1992). Two new categories were proposed for the current study, because of the possibility that subjects would attempt to self-present in ways not necessarily reflected in their explanations for their behavior. First, subjects might engage in Self-promotion as a way of securing the interviewer's good opinion. This would include actions such as drawing attention to other successes or to their own heroic actions to deal with the situation, or attempts to convince the interviewer that they could never make such a mistake now. Secondly, it is possible that subjects might make an appeal for sympathy (Sympathy-seeking) as a way to avoid being judged negatively. That is, while not attempting to avoid their own responsibility for the event, they might refer to they way they have suffered for their actions, or the way that others in the situation made the consequences so much worse. It must be recognized that these new codes are exploratory, and will need further validation. Perfectionistic Self-presentation / 50 In the current study, categorical coding of the transcripts was unsuccessful. Although the trained coders were able to reliably recreate the codes given by the original authors for specific written examples,13 it was not possible to reach an acceptable level of inter-rater reliability with this system on the transcripts. This undoubtedly reflects differences in the nature of the task. Unrestricted verbal responses to a wide variety of highly personal situations are likely to be quite different from written responses to either standardized hypothetical situations (e.g. Gonzales, Haugen, & Manning,, 1994; Sch6nbach,1980) or personal experiences (McLaughlin, et al., 1983), or even verbal responses to standardized situations requiring spontaneous accounts (Gonzales et al., 1990). In this study, it was often difficult to discriminate between subjects' attempts to explain the situation and attempts to excuse themselves for their role in the situation, as well as to determine whether the subject was elaborating on an excuse or using a different excuse. It was decided to modify the coding system into one that was dimensional rather than categorical. Thus, coders rated the extent to which they felt the subject was conceding responsibility or attempting to excuse or justify their behavior during each section of the interview on a Likert scale from 1 or not at all to 5 or extremely. Because very few subjects displayed a refusal to accept responsibility, this dimension was not coded. Means for each of the dimensions were as follows: concessions (M=3.21, SD=98), excusing (M=2.83, SD-89), justifying (M=2.41, SD=97). The mean for self-promotion was 2.42 (SD=.99) and for sympathy-seeking was 2.87 (SD=1.16). (See Appendix B for instructions to coders). Two coders both rated 31% of the transcripts; correlations between coders ranged from .66 to .75. With the exception of excusing during the discussion of the second mistake (65%), ratings were within 1 point on the scale for at least 96% of cases for each behavior at each time point. Ratings on the new dimensions were correlated between raters .77 to .86 for self-promotion and sympathy respectively. While 1 3 Coders rated 20 account statements provided by Gonzales et al (1994); concordance estimates between the codes provided by these authors and our two coders were 75% and 80%. The majority of disagreements focused on the use of refusals - likely the result of a lack of familiarity for our coders as these were extremely infrequent in the present study. Perfectionistic Self-presentation / 51 this system has demonstrated good inter-rater reliability, it must be acknowledged that it represents a departure from the literature and should be interpreted with some caution. The above measures focus on self-disclosures made by participants. However, subjects may also attempt to self-present by avoiding self-disclosure even when the situation pulls for personal information. McCullough and her colleagues (McCullough, 1988; McCullough et al.,-1991) described such behaviors in terms of "intermediate defenses" (consistent with Vaillant, 1971). One such defense consists of excuses and rationalizations and will not be further discussed here. Three others can also clearly serve interpersonal functions by evading self-disclosure (see McCullough et al., 1991): Forgetting ("I forget" or "I don't know" statements), Qualifiers ("maybe" or "possibly" statements that communicate an unwillingness to commit to a position), and Minimization ("not that bad," "a little," or "I suppose" statements that decrease intensity). In order to avoid some of the difficulties with coding as described above, transcripts were examined for the frequency of each of the core phrases - no other judgments as to the use of these defenses outside of the core phrases was made. The frequency of all categories formed a measure of Evasiveness. The entire Psychotherapy Interaction Coding System (PIC System; McCullough, 1988) includes codes for therapist response and other patient variables such as cognition and affect that will not be included here. Although no information is available on the reliability of these intermediate defense categories separate from the whole system, the focus on readily identifiable statements should decrease subjectivity in coding for these categories and these categories typically demonstrate good inter-rater reliability (McCullough, 1988). Such defensive responding (although also including less interpersonal defenses) during therapy has been shown to be negatively correlated with outcome (McCullough et al., 1991). Procedure Residents of the greater Vancouver area referred to each of four sites proceeded through the normal intake procedure at each location. At VHSC-UBC, clients are sent a questionnaire packet prior to Perfectionistic Self-presentation / 52 being seen for a clinical interview at the hospital. A consent to contact form was included in the packet, and those completed and returned to the hospital were forwarded to the study. At the UBC Psychology Clinic, clients were asked for consent to be contacted at the time they inquired about attending the clinic, and were offered a brief interview following the research component to replace the standard telephone screen. The remainder of the participants were recruited through posters or following referral by mental health professionals who asked participants for consent to be contacted by research personnel. It was made clear to each participant that his or her therapy would be in no way dependent on consent to be part of the research. Following verbal agreement to participate during the initial phone contact, participants were asked to refrain from taking any medication that they do not take on a regular basis (i.e. 'as needed' anxiolytics, antihistamines, etc.) within 3 hours of the interview, from drinking coffee, tea, or cola, within 2 hours of the interview, and from smoking within 30 minutes of the interview (this meant an average time delay of 4 hours, 3 hours, and 1.5 hours respectively before being connected to the physiological monitor because of the time involved in completing the questionnaires and IPB). Only one subject had less than 30 minutes acclimatizing to the setting before proceeding to the interview.14 Because of potential influences of menstrual cycle on reactivity measures (Ffastrup & Light, 1984; Polefrone & Manuck, 1988), women were scheduled within the second or third week of their menstrual cycle. Each subject was seen in the Psychology Department at the University of British Columbia. Following a description of the research, written consent, and payment of $10 and transportation costs, participants completed the measures of perfectionism, perfectionistic self-presentation, depressed mood, anxiety, impression management, standards and expectations, as detailed above. In order to emphasize the status of the interviewer, the participants were told that the interviewer was a doctoral student with a great deal of experience. 1 4 This subject was late and the interviewer was not able to reschedule. He completed the questionnaires and the relaxation period within 25 minutes. Perfectionistic Self-presentation / 53 After all of the standard measures were completed, the IPB was administered. While it is possible that the perfectionism interview primed participants to think about perfectionism as a reason for their difficulties, only 3 subjects mentioned this during the interview. Further, although some subjects may have felt more relaxed following the discussion of the different aspects of perfectionistic self-presentation, as it might normalize these concerns, this should have biased against finding differences in distress. At the same time, it is also likely that the change to a neutral interviewer negated the effect of the being 'understood' by the initial interviewer. In any case, it is possible that the same priming could have happened simply as a result of completing the MPS and PSPS as part of the questionnaire packet prior to the interview. Because of time constraints (subjects arriving late or taking more than 45 minutes to complete the questionnaires), the IPB was completed following the interview for a small percentage (< 15%) of subjects. Following the IPB, the Davicon leads were applied to the non-dominant hand. The skin conductance leads were placed on the first and second finger on the palmar side of the medial phalanx, and the pulse plethysmograph was placed on the end and side of either the first or second finger (depending on the quality of the signal) consistent with the equipment manufacturer's suggestions and that of other researchers (Neurodyne, 1994, Venables & Christie, 1980). The hand was secured to a board to limit movement as much as possible. Subjects were told that the physiological monitoring was to assess how relaxed they were while talking to the interviewer. Participants were reminded that the videocamera was off, and were asked to follow a brief relaxation tape for 5 minutes while alone in the room. The clinical interview was videotaped. Because of the extreme discomfort of some individuals regarding videotaping evidenced during pilot testing, the subjects were angled away from the camera, although the view of the subject did allow for assessment of anxious movements.15 One of three trained female interviewers, who were blind to the hypotheses of the research and to the individual's scores on all 1 5 Although this makes coding of anxiety more difficult and probably less accurate, it is difficult to estimate the effect on subject dropout from insisting on full frontal views; certainly, the awareness of the camera might have introduced a hypervigilance in some people that would be quite dissimilar to what would be present in a regular assessment and that would confound the results on arousal. Perfectionistic Self-presentation / 54 measures, conducted the interview; while there is evidence of cross-gender effects in social interactions (see Leary, 1983), there is little evidence to support consistent differences based on therapist-patient gender congruity in the clinical literature (see Garfield, 1994). Interviewers were senior level graduate students, trained in basic interviewing techniques and an open but neutral interviewing style. This was meant to facilitate self-disclosure but also simulate the stance that is most typical of professionals with whom these patients were entering therapy. Additional training on the interview itself was undertaken, and consistency in style between the interviewers was established using taped practice interviews. Ratings of the interviewer in regards to warmth were made to assure consistency across interviewers. These ratings suggested that, despite training, interviewers differed on warmth (F(2,86)=8.65, rj<00T); these differences focused on one interviewer who was more warm, on average, than the other two. However, while this difference is statistically significant, this interviewer only differed by approximately half a rating point (M=5.46 versus M=5.03 and 5.08)) from the other two interviewers. Most importantly, there were no differences between subjects on measures of distress, ratings of performance, ratings of interviewer satisfaction with performance and interviewer liking, defensive behaviors or heart rate, based on which interviewer completed the interview.16 A structured interview was used as a means to present a standardized stimulus for the expression of self-presentational style (included in Appendix C). The interview consisted of three sets of questions. First, subjects were asked about the reason why they have sought treatment at this time. Subjects were asked about their perceptions of what caused their problem, and their contribution to the problem. These questions formed the "Reason Here" section of the interview and were included as an opportunity to have the subject acclimatize to the setting, the interviewer, and the task, as well as to establish the interview as similar to other initial contacts with mental health professionals. Subjects were then asked to think about situations in their life during which they feel they did not cope well; that is, situations in which they made a 1 6 To confirm this, analyses were repeated controlling for interviewer warmth and the results were unchanged. Perfectionistic Self-presentation / 55 mistake or which were made worse by the way they handled it. They were requested to think about the most serious one (that is, the one during which their contribution had the most negative impact for themselves or someone else) and briefly describe it. Subjects were then asked about their perceptions of their contribution to the development of the situation and to its conclusion. These questions formed the "First Situation" section of the interview. Following this description, subjects were asked about a second difficult situation (the next most serious, or another one that was more serious than the first one if such has occurred to them) and the same set of questions applied, forming the "Second Situation" section. Two situations were included to maximize the likelihood of getting at least one that was quite serious, to give additional scope for self-presentation and to allow for an examination of change in arousal over time. Six subjects were not able to provide a second situation, despite interviewer encouragement. These subjects did not differ on perfectionistic self-presentation, trait perfectionism, or impression management, from the sample as a whole. Following the interview, the interviewer thanked the subject and asked him/her to relax for a few minutes. She then left the room, and the subject was alone for a minimum of two minutes. The primary researcher returned and the post-interview questionnaire package was administered. Following completion of these questionnaires, participants were reassured that withholding difficult situations is common and understandable, and were asked to indicate if the situations they described were indeed the most difficult ones they could remember. Twelve subjects indicated that they had held back; these subjects did not differ significantly in perfectionism from those who did not withhold any difficult situation. Subjects also were asked if they felt the interview was at all similar to therapy; while subjects varied considerably in their responses, no one identified the focus on past mistakes as unusual. Finally, participants were thanked, debriefed, and dismissed.17 1 7 Subjects from the Psychology Clinic completed a brief interview with the primary researcher to secure additional information needed for clinic files. Perfectionistic Self-presentation / 56 Two sets of two coders were trained on pilot interviews according to standard procedures to establish adequate inter-rater reliability. One set of coders coded the seriousness of the situations described, and the anxiety and the extent to which the subject was attributing their problems to external sources during each section of the interview. Tapes were transcribed, and the second set of coders made ratings of defensive behaviors. RESULTS A summary of univariate statistics including reliabilities for each standardized scale, is available in Table 1 for the sample as a whole, and separately for men and women. The alphas for the scales ranged from .80 to .92, suggesting good to excellent internal consistency for all measures. There were no significant gender differences except for self-rated mood (PANAS) post interview, when women expressed more negative mood than did men (t(88)=2.06, p<.05). Distributions were examined for skewness, kurtosis, and univariate outliers. Both mood and skin conductance level were distributed significantly different from normal, being positively skewed. Thus, PANAS scores and SCL were transformed to normalize the distributions. However, as subsequent results did not differ from the non-transformed scores, the original scores are presented here for ease of interpretation. Univariate outliers (scores above 3 standard deviations from the mean) were evidenced in measures of skin conductance (n=5) and heart rate (n=3) and these cases were eliminated from the physiological analyses. Two additional subjects had missing data on SCL for one timepoint during the interview and so were dropped from the analysis. Correlations between perfectionism variables and demographics and controls are presented in Table 2. As mentioned earlier, education was significantly related to perfectionistic self-presentation. It is possible that those with more education show more sophistication in completing the measure, responding Perfectionistic Self-presentation / 57 Table 1 Univariate statistics Whole Sample Men Women Mean StdDev Alpha Mean StdDev Mean StdDev Perfectionism Nondisplay Imperfection 45.05 13.05 .91 46.86 11.84 43.24 14.06 Nondisclose Imperfection 23.32 8.38 .82 23.89 8.17 22.77 8.60 Self-Oriented 21.84 7.61 .88 20.98 7.31 22.71 7.89 Socially Prescribed 20.51 6.59 .80 19.33 6.31 21.69 6.73 Self-Rated Negative Affect Pre-Interview 1.98 .83 .89 2.06 .82 1.89 .83 Post Interview 2.09 .78 .88 1.92 .73 2.25 .80 Behavioral Ratings of Social Competence Interviewer Expectations 57.84 11.57 .90 57.56 11.71 58.13 11.55 Interviewer Satisfaction 54.18 11.08 .91 55.29 9.96 53.07 12.10 Anticipated Performance 60.67 10.51 .88 60.53 9.72 60.80 11.36 Perceived Performance 54.65 11.23 .89 55.54 10.71 53.77 11.78 Adjustment Interaction anxiety 49.19 12.41 .92 48.16 12.18 50.23 12.69 Depression 12.63 7.56 .90 13.24 7.83 12.10 7.31 Impression Management 5.04 3.71 .821 4.71 3.63 5.38 3.81 Perfectionistic Self-presentation / 58 Heart Rate (Beats/min) Relax Situation 1 Situation 2 72.58 10.46 74.45 10.85 73.24 10.35 Skin Conductance Level (micromhos) Relax 1.36 .86 First Situation 1.95 1.18 Second Situation 2.02 1.32 72.59 10.92 74.05 10.33 72.35 10.43 1.50 1.00 2.05 1.31 2.18 1.36 72.58 10.14 74.82 11.40 74.06 10.33 1.23 .71 1.87 1.06 1.88 1.00 1 Based on the full Likert version of the scale. Note: First Situation refers to the discussion of the first difficult situation. Second Situation refers to the discussion of the second difficult situation. to a self-presentational concern about admitting to a desire to appear perfect and therefore denying it, or that these concerns are actually ameliorated by more education. Whatever the case, the size of the correlations suggests that education is not a significant confound in the current study but inclusion as a control variable is warranted. Correlations between perfectionism variables and controls showed significant relations between all perfectionistic self-presentation dimensions and both interaction anxiety and depression, ranging between .40 to .51 (p>01). This is similar to other research with the perfectionistic self-presentation scale (Hewitt, & Flett, 1996). Scores on the BIDR failed to relate significantly to any of the perfectionism subscales, supporting the view that perfectionistic self-presentation is distinct from impression management. The data was further examined at both the bivariate and the multivariate level for outliers and violations of assumptions. Examination of Mahalanobis distance and standardized residuals suggest that there were no significant multivariate outliers and that the variables met requirements for regression analyses (Tabachnick & Fidell, 1989). The results are presented here according to the areas of cognition, Perfectionistic Self-presentation / 59 Table 2 Correlations between perfectionistic self-presentation and demographics and controls. Nondisplay Nondisclose Demographics Age .02 -.07 Gender -.14 -.07 Education -.24* -.22* Medication .24* .07 History with a Mental Health Professional .08 .02 Controls Self-oriented Perfectionism 41*** 42*** Socially Prescribed Perfectionism 57*** 49*** Interaction anxiety 51*** 40*** Depression 41*** 40*** Impression Management -.09 .02 Note: Medication is coded as presence or absence of a regular psychotherapeutic medication. *E<05, **p<01, ***p<001 affect, physiological arousal, and behavior. The issue of Type I error deserves special comment. It is acknowledged that each area in the current study is in essence a separate study, each being tested with three hypotheses and with several different variables. In a correlational study of a complex design, it is difficult to know what form an adjustment to significance levels should take, particularly given the necessity of also considering the issue Perfectionistic Self-presentation / 60 of Type II error. Certainly, the requirement of strict probability levels when testing the null hypothesis has been criticized in the recent literature (e.g. Hammond, 1996; Hunter, 1997). However, the following considerations were made while interpreting the results, consistent with Cohen & Cohen (1983). First, the number of variables in each test of the hypothesis was weighed when interpreting the bivariate results. For example, the cognitive hypotheses were tested with interviewer expectations, threat, and interviewer satisfaction which suggests an adjustment to a critical probability level of .016 (or .05/3). Correlations meet the adjusted levels unless noted. The regression analyses were protected in two ways. First, multivariate analysis was pursued only if the bivariate correlation was significant. Second, beta weights for the independent variables were not interpreted unless the block change for both dimensions was significant. Both of these restrictions, although conservative, function in the same way as a protected t-test and do protect against Type I error (Cohen & Cohen, 1983). Cognition. Bivariate statistics. The bivariate correlations between the perfectionistic self-presentation subscales and the cognitive self-report measures of social competence are available in Table 3. The subject's ratings of the interviewer's expectations for behavior were not correlated with either of the dimensions of perfectionistic self-presentation. However, the desire to avoid disclosure of imperfection was positively correlated with the discrepancy between the subject's perceptions of how they would perform and what they felt the interviewer expected of them. That is, subjects high on this dimension were more likely to feel the interviewer wanted more from them than they could manage during the interview. After the interview, both the perfectionistic self-presentation dimensions were negatively correlated with perceived interviewer satisfaction with performance. A second measure of the interviewer's disappointment, the discrepancy between perceptions of what the interviewer expected prior to the interview, and how the subject felt they did during the interview, confirmed this result; the correlation with the Nondisclose subscale was .40 (pK.001). Perfectionistic Self-presentation / 61 Table 3 Correlations between perfectionistic self-presentation and Behavioral Ratings of Social Competence scores. Nondisplay Nondisclose Interpersonal Perceptions Pre-Interview: Interviewer Expectationst -.16 -.04 Threat .02 .31** Post-Interview: Interviewer Satisfactiont -.30*** -.40*** Interviewer Liked Mef -.11 -.17 Interviewer Likes Most Peoplef .15 .10 Personal Judgments Pre-Interview: Anticipated Performance -.20 -.38*** Post-Interview: Perceived Performance -.34*** -.50*** t Refers to subjects' ratings of the interviewer's expectations, satisfaction, and liking. Note: Threat refers to the discrepancy between ratings of what the interviewer expects and what the subject feels he or she will be able to perform. *P<05, **p<.01, ***p<.001 The bottom portion of Table 3 demonstrates that there were also significant relations between Nondisclose and personal judgment of anticipated performance and between both dimensions and perceived performance. Therefore, it could be argued that discrepancy scores do not provide new information about Perfectionistic Self-presentation / 62 interpersonal factors, over and above the subject's own feelings of inadequacy. However, the difference between how liked the subject felt by the interviewer and how they felt the interviewer would like most subjects, produces a very similar pattern of correlations (r=.31, and .30, p<.01 for Nondisclose, and Nondisplay respectively) even though there are no significant relations between the ratings individually and the perfectionistic self-presentation dimensions. The multivariate analyses that follow also support a unique position for the discrepancy score. Taken together, these correlations suggest that the interview was a threatening interpersonal environment for those high on perfectionistic self-presentation. Multivariate statistics. The regression analyses for ratings of social competence are summarized in Table 4. The results show that subjects high in perfectionistic self-presentation were more likely to experience a discrepancy between what they felt the interviewer expects and what they felt they could accomplish, over and above the influence of gender, age, education, and past experience with a mental health professional. The desire to avoid disclosure of imperfection was a unique predictor. Interestingly, the Nondisplay subscale was also a unique predictor but in the opposite direction. That is, after partialing out covariation with Nondisclosure, the desire to avoid displaying imperfection was associated with the belief that personal performance would exceed that expected by the interviewer. Following the interview, perfectionistic self-presentation explains significant variance in ratings of interviewer satisfaction, again uniquely predicted by Nondisclosure. The second regression was consistent with the first; perfectionistic self-presentation explained significant variance over and above the controls and trait perfectionism, and the Nondisclosure of imperfection was a unique predictor, marginally sharing this position with non-display when predicting interpersonal threat. Following the addition of impression management, depression and interaction anxiety, perfectionistic self-presentation added significant additional explanation to the level of interpersonal threat and showed a strong trend for ratings of interviewer satisfaction, with the same unique predictors as Perfectionistic Self-presentation / 63 Table 4 Hierarchical regression equations predicting Behavioral Ratings of Social Competence scores Pre-Interview Post-Interview Threat Interviewer Satisfaction Beta R 2 C h Beta R 2 C h Controlling for demographics Step 1 .02 .11* Gender .00 -.10 Age -.04 .16 Education -.07 .14 History with a MHP .02 -.04 Step 2 .13** .13** Nondisplay Imperfection -.27* -.09 Nondisclose Imperfection 46*** -.30** Controlling for Trait Perfectionism Step 1 .02 .18** Gender .00 -.07 Age -.05 .17 Education -.06 .14 History with a MHP .03 -.07 Self-Oriented -.07 .02 Socially Prescribed .05 -.10 Perfectionistic Self-presentation / 64 Beta R 2 C h Beta R 2 Ch Step 2 .13** .06* Nondisplay Imperfection -.28A -.05 Nondisclose Imperfection 47*** - 29* Controlling for Other Variables Step 1 .13 .25*** Gender -.05 -.10 Age .02 .08 Education -.01 .10 History with a MHP -.09 .07 Interaction anxiety .38** -.19* Depression .16 -.14 Impression Management .07 .10 Step 2 .13** .05A Nondisplay Imperfection -.46*** .04 Nondisclose Imperfection .37** -.27* Note: Threat refers to the discrepancy between ratings of what the interviewer expects and what the subject feels he or she will be able to perform. MHP refers to a mental health professional. Ae<.07, *p<.05, **p<.oi, ***p_<ooi above.18 In order to consider whether the discrepancy score offers information over and above the ratings of anticipated performance, the regressions were rerun using a covariate method. Predicting interviewer 1 8 Entering all control variables together demonstrated similar results for the ratings of threat, although the perfectionistic self-presentation block no longer predicted significant additional variance in interviewer satisfaction. Perfectionistic Self-presentation / 65 expectations controlling for anticipated performance led to a pattern of results that were essentially the same. That is, perfectionistic self-presentation added significant variance as a block for all three regressions, and Nondisclosure uniquely predicted higher perceived interviewer expectations and Nondisplay uniquely predicted lower interviewer expectations, over and above what the subject anticipated for his or her performance. Thus, once subjects were equated on levels of self-efficacy, perfectionistic self-presentation was a significant predictor of interviewer expectations. Affect Bivariate. The correlations between perfectionistic self-presentation and anxiety and arousal are presented in Table 5. There were strong positive correlations between self-ratings of negative affect and both perfectionistic self-presentation dimensions. The relation between Nondisclosure and post-interview mood may be particularly robust, as this dimension was significantly related to post-interview mood over and above negative mood prior to the interview (partial r= .22, p_<.05) whereas the partial correlations between Nondisplay and post interview mood was not. Despite relations with self-reported mood, neither of the perfectionistic self-presentation dimensions were related to observable anxiety as rated by coders. On the other hand, interviewers endorsed higher levels of anxiety for those high in a desire to avoid displays of imperfection.19 This discrepancy raises the possibility that perfectionistic self-presenters provide subtle nonverbal cues of anxiety that were not accessible to the coders or to the clinical judge but were available to the interviewer. Multivariate. The regression analyses for self-report ratings of negative mood are summarized in Table 6. The initial regressions show that perfectionistic self-presentation significantly improves the prediction of negative mood, over and above control variables, both prior to and following the interview. Only the Nondisplay subscale was a unique predictor, and then only of pre-interview negative mood, although there was a trend for Nondisclosure to uniquely predict negative mood post-interview. After 1 9 Although after adjusting the significance level, this is interpreted as a trend. Perfectionistic Self-presentation / 66 Table 5 Correlations between perfectionistic self-presentation and affect by self-report ratings, observer ratines, and physiological measures. Nondisplay Nondisclose Negative Affect (PANAS) Pre-Interview 41*** .33*** Post-Interview .32*** .35*** Observer Ratings of Anxiety Coders: First Situation .06 .09 Second Situation .00 -.02 Interviewer: .23* .17 Heart Rate Relaxation .12 .15 Reason Here .13 .19A First Situation .16 .27** Second Situation .17 .23* Recovery .07 .15 Arousal .11 .31** Skin Conductance Level Relaxation .18 .16 Reason Here .14 .10 Perfectionistic Self-presentation / 67 Nondisplay Nondisclose Situation 1 .05 .03 Situation 2 .08 .05 Recovery .10 .01 Arousal -.13 -.15 Note: First Situation refers to the discussion of the first difficult situation. Second Situation refers to the discussion of the second difficult situation. Arousal refers to the discrepancy between level when discussing the first difficult situation and relaxation. AE<.10, *p<.05, **g<01 controlling for trait perfectionism, and finally after controlling for adjustment and impression management, perfectionistic self-presentation no longer added significant additional explanation. Regressions were run predicting the interviewer's ratings of anxiety over the interview. The perfectionistic self-presentation block did not add significant unique variance to any of the equations. Arousal Bivariate. The correlations for physiological arousal, reported in the bottom portion of Table 5, were somewhat inconsistent. Heart rate was significantly related to the Nondisclosure subscale during the discussion of the first difficult situation, and was also related to arousal, measured as an increase in heart rate over relaxation. This arousal seems to be maintained through the discussion of the second mistake for those high on Nondisclosure, but is reduced by the time of the recovery period. Although the marginally significant correlation between Nondisclosure and heart rate during the 'reason here' section suggests that the discussion of mistakes per se is not uniquely distressing, the correlation with heart rate during the discussion of the first mistake remains significant even when controlling for their response during this preceding section (partial r=.27, p< .03). In other words, even though subjects high on Nondisclosure were somewhat more distressed when talking about why they had come for therapy than Perfectionistic Self-presentation / 68 Table 6 Hierarchical regression equations predicting self-reported negative affect (PANAS) Pre Interview Post Interview Beta R2 Change Beta R2 Change Controlling for demographics Step 1 .09 .09 Gender -.07 .20 Age -.05 -.09 Education .10 -.16 History with MHP -.21 .00 Step 2 .15*** .13** Nondisplay Imperfection .33** .20 Nondisclose Imperfection . 11 .22A Controlling for Trait Perfectionism Step 1 .27*** .22*** Gender -.12 .19 Age -.02 -.08 Education -.13 -.08 History with MHP -.19 .03 Self-Oriented .29* .13 Socially Prescribed .01 .11 Step 2 .03 .03 Nondisplay Imperfection .23A .10 Perfectionistic Self-presentation / 69 Beta R 2 Change Beta R 2 Change Nondisclose Imperfection .03 .16 Controlling for Other Variables Step 1 .40*** .35*** Gender -.05 .25* Age .05 .01 Education -.10 -.07 History with MHP -.29** -.12 Interaction anxiety .10 .13 Depression 45*** .38** Impression Management -.04 -.07 Step 2 .02A .02 Nondisplay Imperfection .16 .03 Nondisclose Imperfection -.00 .14 AP<.06, *p<05, **p_<01, ***p_<001 those low on this dimension, they experienced an increase in heart rate when forced to discuss past serious mistakes. Unlike heart rate, the correlations for skin conductance showed no significant relations to perfectionistic self-presentation. Further, change in skin conductance level during the discussion of the first mistake, over and above relaxation, was not related to either perfectionistic self-presentation dimension. Perfectionistic Self-presentation / 70 Multivariate. Two regressions were run predicting heart rate during the discussion of the first mistake; one to predict mean level, the second to predict increase over and above heart rate at relaxation.20 These results are summarized in Table 7. As discussed earlier; coder's ratings of the seriousness of the problem and medication usage were included in the control block. The results show that perfectionistic self-presentation predicts significant additional variance in heart rate following control variables, and following measures of emotional distress and impression management. Following trait perfectionism, the block of perfectionistic self-presentation variables approaches significance. Predicting arousal over and above relaxation levels, the perfectionistic self-presentation block was significant on all equations. The desire to avoid disclosure of imperfections was consistently a unique predictor of increased heart rate.21 Because of the nonsignificant relations between skin conductance and perfectionistic self-presentation at the bivariate level, multivariate analyses were not pursued. Behavior Twenty-eight subjects (31%) claimed at least one self-handicap prior to the interview, and 30 (33%) subjects claimed at least one excuse following the interview (fourteen claimed at least one at both time points). Self-handicaps prior to the interview included claims of physical concerns such as lack of sleep, coffee, or cigarettes, or other physical ailments (n=20), self-presentational concerns such as being nervous with a videocamera (n=22), present mood (n=10), and life stress outside of the interview (n=10). Following the interview, most subjects claimed factors within the interview (n=35) to explain a poor performance, although some claimed physical concerns (n=6) and mood (n=14). These are consistent with excuse theory and the definition of self-handicapping (Rhodewalt, 1982; Weiner, 1992). Results entering heart rate at relaxation as a step in the regression instead of using a difference score were essentially identical, as were the results using ipsatized scores (although this last approach showed a role for Nondisclosure as a unique predictor for the discussion of the second mistake as well). 2 1 Entering all controls together resulted in similar results for arousal over and above relaxation levels (a significant perfectionistic self-presentation block and Nondisclosure as a unique predictor) although the perfectionistic self-presentation block predicting level of heart rate when discussing the first mistake was significant at g<06 (with Nondisclosure as a unique predictor). Perfectionistic Self-presentation / 71 Table 7 Hierarchical regression equations predicting heart rate First Situation Arousal Beta R2 Change Beta R2 Change Perfectionistic Self-Presentation Step 1 Gender Age Education History with MHP Medication Seriousness of the problem Step 2 Nondisplay of Imperfection Nondisclose Imperfection Controlling for Trait Perfectionism Step 1 Gender Age Education History with MHP Medication .12 .04 .01 .06 .11 .34** .07 .35** .07 .03 .02 .08 .10 .14 .09* .17* .09 .03 .04 .12 .03 .15 •.10 .39** .16 .07 .04 .19 .04 .04 .IV .07 Seriousness of the problem .34** .19 Perfectionistic Self-presentation / 72 Beta R2 Change Beta R2 Change Self-Oriented .10 .20 Socially Prescribed .03 -.34A Step 2 .06A . 12** Nondisplay of Imperfection -.12 .01 Nondisclose Imperfection .31* .41** Controlling for Other Variables Step 1 .15 .06 Gender .11 .08 Age -.06 .04 Education .00 .05 History with MHP .07 -.13 Medication .10 -.03 Seriousness of the problem .36** .16 Interaction anxiety .01 .04 Depression -.05 -.01 Impression Management .04 .00 Step 2 .08* .09* Nondisplay of Imperfection -.06 -.11 Nondisclose Imperfection .36** .39** Note: Arousal refers to the discrepancy between heart rate when discussing the first difficult situation and relaxation. Arj<07, *p<.05, **g<01, ***p<.001 Perfectionistic Self-presentation / 73 Bivariate. The results describing the bivariate relations between perfectionistic self-presentation and behavior, presented in Table 8, are mixed. The perfectionistic self-presentation dimensions did not predict whether or not subjects would claim a self-handicap. However, among those who did self-handicap, there were highly significant relations with the extent of self-handicapping. Prior to the interview, those high on the Nondisclose dimension who stated that their behavior would be negatively affected because of something outside of themselves, had a tendency to claim a greater impact for this reason than did those low on this dimension. Following the interview, Nondisclosure was again related to more negative ratings of impact on performance. Because of the small number of participants who self-handicapped, it was not possible to proceed with this variable at the multivariate level. However, it was possible to address one alternative explanation for the results that arises from the significant correlations between depression and the extent of self-handicapping prior to the interview (r=.54, jK.001) and following the interview (r=.38, p<.05). Partialing out depression, the correlation between Nondisclosure and extent prior to the interview was no longer significant (r=.26, p_>.05) but the post-interview correlation with Nondisclosure remained significant (r=.37, p<05). This suggests that those high on a desire to avoid admitting to imperfection tried harder to excuse their behavior. The observer ratings of defensiveness were also examined. The correlations between perfectionistic self-presentation and externalization, defensive behaviors, and evasiveness over both situations, failed to reach significance. Indeed, the only trends suggest that, contrary to predictions, perfectionistic self-presentation is related to lower rates of externalizing and justifying.22 This pattern of results does not seem likely to be related to the type of problems discussed, as there was no relationship between perfectionistic self-presentation and ratings of the seriousness of the problem or potential culpability.23 However, while subjects high on the Nondisclosure of imperfection did not differ in their observable behaviors, they did speak for a shorter length of time when discussing mistakes; this relation 2 2 After adjusting significance for the number of variables, these cannot be interpreted as trends. Perfectionistic Self-presentation / 74 Table 8 Correlations between perfectionistic self-presentation and behavior. Nondisplay Nondisclose Self-Handicapping Use PreTnterview -.06 -.02 Use Post-Interview .09 .12 Extent Pre-Interviewa .25 .42* Extent Post-Interviewb .28 .46** Behaviors Externalization -18A -.07 Conceding .01 .13 Excusing -.07 -.05 Justifying -.17 -18A Self-promoting -.05 -.10 Sympathy-seeking -.08 -.07 Evasiveness -.16 -.05 Time spent discussing mistakes -.05 -.26** an=28,bn=30 Note: "Use" refers to the claim of any self- handicap, "extent" refers to the maximum rating of incapacitation from a claimed self-handicap. Ap<10, *£<05, **p_<01 2 3 Controlling for seriousness of the mistakes, Nondisplay was associated with lower rates of externalizing (r=-.21, E<05). Perfectionistic Self-presentation / 75 held after controlling for age, education, and gender (partial r=-.21, p_<.05), all significantly related to the length of disclosure. Because it is possible that either the frequency of behaviors or ratings of the extent of different types of behavior are influenced by the length of time spent talking, the correlations between perfectionistic self-presentation and behaviors were repeated controlling for length of disclosure. Only the relation between Nondisplay and justifying (r=-.18, p_<.10) approached significance when considering behavior over the discussion of both problems. A more fine-grained analysis showed that, while the perfectionistic self-presentation dimensions were not related to behaviors during the discussion of the first difficult situation, the extent of self-promotion was related to the Nondisclosure subscale (r=.26, p_<.0T) during the discussion of the second situation after controlling for time spent talking. However, because these relations are isolated, they are not interpreted as significant. In addition to length of time, it is possible that behaviors reflect past experiences with a mental health professional. That is, it is possible that behavior is learned during exposure to the demands of therapy. Participants were classified as having had significant contact with a mental health professional if they had had at least 3 psychotherapy visits at some point in the past (n=62; 69%). Compared with those who had not had such experience, these subjects were more excusing (t (88)=2.61, p_<.01) and had a tendency to justify their behavior more (t (88)=1.84, p<.07). Within the treatment-naive group of twenty-eight subjects, although not reaching statistical significance, there is a tendency (p_<. 10) towards non-display being related to lower concessions (r=-.23) and sympathy seeking (r=-.32). Thus, it would seem that different behaviors may arise from past learning or reactions to past interactions with a therapist or from a desire to self-present as perfect. However, these findings are only suggestive and would require confirmation. In particular, difficulties with the coding scheme described earlier, make these results suspect and in need of replication. Perfectionistic Self-presentation / 76 Gender differences The data was separated by gender and the bivariate analyses repeated. On the whole, the pattern of results for males and females are highly similar, and are similar to the sample as a whole. Because no hypotheses were made about gender differences, it must be remembered that these analyses are post hoc and should be interpreted cautiously. At the same time, some interesting results emerge that may help to understand the results presented above Table 9 provides correlations between the perfectionistic self-presentation subscales and some of the study variables, separately for men and women. On the whole, women seem to show stronger relations between perfectionistic self-presentation and self-report measures of mood and social competence than do men. This seems to be particularly true following the interview. Second, although women do show significant correlations between perfectionistic self-presentation and observable anxiety, this is not as much the case for men. In fact, the negative correlations between observable anxiety and Nondisclosure in particular, suggests that men may go out of their way to appear less anxious if they are high in a desire to be seen as perfect. Finally, it would seem that women high in perfectionistic self-presentation tend to show increases in heart rate, whereas men tend to show increases in skin conductance level. It must be remembered that because of the size of each group, a difference between the correlations of at least .36 is necessary for the difference to be statistically significant (at p_<.05). Because of the post hoc nature of these analyses, as well as the small sample sizes in each group, multivariate analyses were not pursued for comparison between men and women. Perfectionistic Self-presentation / 77 Table 9 Correlations between perfectionistic self-presentation and selected study variables for men/women. Behavioral Ratings of Social Competence Interviewer Expectations Threat (Discrepancy) Interviewer Satisfaction Nondisplay Men/Women .05/-.25 .09/-.02 •.17/-.42** Nondisclose Men/Women .07/-. 14 .39**726 .28/-.51*** Anticipated Performance Perceived Performance Negative Affect (PANAS) Pre-Interview Post-Interview Observer Ratings of Anxiety Interviewer Ratings of Anxiety Heart Rate Relaxation Situation 1 Situation 2 Arousal .15/-.23 24/-.44** 41**/ 40** .24745** -.13/.28 .16/.30* •.01/.29 .06/.28 .07/28 .24/.06 ..31*/-.44** .32*/-.67*** .24/.41** .09/.62*** .26/.52*** .207.33* .18/.21 .22/.40** .22731* .10/.45** Perfectionistic Self-presentation / 78 Skin Conductance Level Relaxation Situation 1 Situation 2 Arousal MenAVomen .33*/.00 31A/-.19 .30A/-.16 .11/-.29A MenAVomen .27A/.01 .19/-. 16 .18/-. 13 -.04/-.25 Note: Highlighted pairs are statistically different at p_<.05 Arj<07, *p<.05, **p<01 Perfectionistic Self-presentation / 79 DISCUSSION The current study was designed to test several hypotheses regarding the cognitive, affective/physiological, and behavioral manifestations of a perfectionistic self-presentational style in a clinical sample. In particular, it sought to examine the impact of a desire to withhold either verbal or non-verbal evidence of imperfection on the experience of a clinical interview. While the results are mixed, in general they suggest that although perfectionistic self-presentation was not related to the use of defensive behaviors, those high in a desire to avoid being known to be imperfect experienced the clinical setting as threatening and experienced emotional distress and physiological arousal. The results assessing a unique contribution of perfectionistic self-presentation suggest that the two dimensions have a role in predicting cognitive appraisals of interpersonal threat and dissatisfaction and increases in heart rate over relaxation when discussing past mistakes, over and above contributions of trait perfectionism. This supports conceptual and empirical distinctions in previous work (Hewitt, Flett, Fehr, et al., 1996). Further, while the current data mirrors previous research in supporting roles for social anxiety (Alden et al., 1994; Wallace & Alden, 1991) and depression (Gotlib & Meltzer, 1987; Youngren & Lewinsohn) in appraisals, and depression in self-handicapping (Weary & Williams, 1990), it suggests that these emotional factors are not the only explanations for the client's experience. The fact that perfectionistic self-presentation predicts appraisals and arousal above and beyond measures of emotional distress provides preliminary evidence for a previously unrecognized presentational style that is particularly relevant for psychotherapy. Given the strength of the relation between perfectionistic self-presentation and these emotional adjustment variables, and the inclusion of multiple controls, it is perhaps even more interesting that the different dimensions provide additional explanatory power. In addition to demonstrating a potential role for perfectionistic self-presentation beyond trait perfectionism and emotional adjustment, the results also make it clear that perfectionistic self-presentation is not equivalent to the traditional means of describing socially desirable responding. Impression managers Perfectionistic Self-presentation / 80 are prone to making exaggerated claims of competence or morality (Paulhus, 1986); in essence, they create an admirable image and place it confidently out into the social world. Scores on impression management did not predict the same internal experience as the self-presenter evidenced in the current study but did predict self-disclosure of problems that were less serious and less blameworthy.24 Given that there is no reason to expect that those high in impression management have consistently made less serious or blameworthy mistakes in their lives, this suggests that they are actively attempting to present a positive image. In contrast, perfectionistic self-presenters neither exaggerate their virtues nor confidently present themselves in a social interaction. Regardless of a desire not to be known as imperfect, they admit to having negative symptoms and to having made serious mistakes, even though they may be distressed doing so. That is, they create an image that is not overly positive, and that is placed in the social world with much fear and trembling. Both the image and the fear are discussed in more detail below. Cognition The current study used subject ratings of social competence to assess the extent to which the subject's appraisals of the interviewer reflected anticipated or perceived judgment. Ratings both prior to and following the interview, referencing both personal evaluations and perceived interviewer evaluations, allow for a thorough examination of how the perfectionistic self-presenter appraises a clinical setting. Several findings stand out from the data. First, perfectionistic self-presentation is clearly related to feelings of inadequacy in a clinical setting. Those high in perfectionistic self-presentation came in with lower expectations for their performance and judged their performance during the interview to be much more negative than those low in this style; supplementary analyses suggest that these relations hold over and above trait perfectionism and emotional adjustment.25 These findings are in line with previous studies that have shown perfectionistic M _r=-.27, p_<01 and -.23, p<05 respectively. 2 5 Repeating the same set of 3 regressions as used for the main analyses, for both anticipated performance and perceived performance, the perfectionistic self-presentation block r2 change scores ranged from .22 to .06, all Perfectionistic Self-presentation / 81 self-presentation to be highly correlated with low self-esteem in general, and more specifically in the realms of social interactions, appearance, and academics (Hewitt, Flett, Fehr, et al., 1996). However, neither perfectionistic self-presentation dimension was correlated directly with subjects' ratings of what the interviewer expected prior to the interview. This is consistent with the findings of Alden et al. (1994) in regard to socially prescribed perfectionism (also not correlated with expectations in the current sample).26 So it would seem that, at least prior to the interview, the interviewer was not seen as being especially critical or judgmental. There was quite a different picture when considering the relation between the subject's own expectations and those perceived to be held by the interviewer, however. Consistent with earlier research showing a relation between social anxiety and a discrepancy between self and others' expectations (r=.30, p<01 in the current study; Alden et al., 1994; Wallace & Alden, 1991), the discrepancy between estimates of what the interviewer wanted and what the subject felt they could provide was positively related to the perfectionistic self-presentation dimension of Nondisclosure. Scores on Nondisclosure predicted ratings of interviewer expectations over and above what the subject expected of his or her own performance. Using the conceptualization of threat by Tomaka et al. (1993) as the comparison of perceived demands and perceived ability to cope with the demands, this discrepancy suggests that those high on Nondisclosure did experience a degree of threat when anticipating the interview. This threat may be a partial explanation of earlier findings that the fear of disclosing imperfections is related to lower confidence in mental health professionals (Nielsen et al., 1977). While one might be justifiably cautious about equating a mathematical model to actual experience of threat, the ratings taken after the interview support the notion of a threatening interpersonal environment. Those high on a desire to avoid disclosing imperfections, compared with those low on this dimension, felt that the interviewer was less satisfied with significant at at least p<.05. The beta weights for Nondisclosure ranged from -.31 to -.44, again all significant at at least p_<05. 2 6 This may be evidence that socially prescribed perfectionists are referring to specific others when they report feeling that others have perfectionistic standards for them, rather than to generalized others. Perfectionistic Self-presentation / 82 their performance27 and felt they were less liked by the interviewer than most people. Given the finding that people with low efficacy feel less liked by their interaction partners (Alden et al., 1992), this may reflect the fact that subjects high on Nondisclosure also felt they did less well during the interview. Indeed, partial correlations between Nondisclosure and ratings of interviewer satisfaction, controlling for perceived performance, were not significant. At the same time, this implies, at the least, that subjects high on Nondisclose expect the interviewer to have the same negative judgment of their performance as they have themselves. There is no grace expected for a poor performance. One intriguing result that was not hypothesized was the finding that Nondisclosure and Nondisplay had very different relations with degree of threat; both dimensions were expected to have similar and positive relations to threat. While the reason for the difference is unclear, this may reflect the nature of the task. Coming into a clinical interview, it is reasonable to expect that you will be required to discuss difficulties, shortcomings, etc.. If you have a strong desire to avoid discussing such shortcomings or imperfections, and fear being judged because of those disclosures, this situation would logically be threatening. At the same time, it is logical to expect that those high in a desire to avoid displaying imperfections might be less threatened by a clinical interview as it would seem less likely that you would encounter a situation by which you would demonstrate your imperfections. In other words, if discussing your inner self is not problematic, then disclosing imperfections should not be related to expectations of judgment or be particularly threatening (although situations that involve performance demands might be more so). Thus, it may be the nature of the task in combination with the core fear of each perfectionistic self-presentation dimension, that is the critical determinant of appraisals. However, although it is logical that threat would not be related to the level of Nondisplay, the data suggest it is related but in the opposite direction; those high in a desire to avoid a display of imperfection " It must be acknowledged that ratings of interviewer satisfaction appear to be influenced by a wide range of variables as the final regression suggests that perfectionistic self-presentation adds only a marginally significant amount of additional variance, over and above adjustment variables. Perfectionistic Self-presentation / 83 (independent of level of Nondisclosure) are more likely to believe that they will perform better than the interviewer is expecting. On the face of it, it would seem then that this dimension may be related to Over-confidence. But how can this be true when Nondisplay is consistently related to negative emotional adjustment in the current and past samples (Hewitt & Flett, 1996; Hewitt, Flett, Fehr, et al., 1996)? Although the current data does not allow for a resolution, it seems there are two possible explanations. First, it is possible that those high in a desire to avoid public displays of imperfection do not anticipate negative social consequences. That is, the higher ratings of negative affect associated with Nondisplay may be nonspecific to levels of social threat, compared to those high on Nondisclosure. This would imply that by demonstrating a confidence in their own ability to perform relative to others' expectations, they are honestly reflecting a lack of social threat that is typical of persons who are low in depression or social anxiety (Lewinsohn et al., 1980; Wallace & Alden, 1991). This explanation seems unlikely in light of the fact that they are high in depression and interaction anxiety, and typically have low self-esteem (Hewitt, Flett, Fehr, et al., 1996). Second, it is possible that those high on Nondisplay do experience the same feelings of threat but more actively resist admitting to them. In other words, they are bluffing. They are either denying the full extent of their feelings of inadequacy, or they are painting a more positive picture of the interviewer than would be expected given their emotional distress. Trends in the data suggest that both may be the case as, unlike depression and interaction anxiety, Nondisplay (controlling for Nondisclosure) is unrelated to expectations for performance (r=05, ns) and marginally related to less negative views of the interviewer (r=-. 17, p<. 10). The possibility that threat is being denied is perhaps most obvious in the pattern of results for men, where those high in Nondisplay did not report high levels of threat but did show heightened skin conductance when discussing mistakes. At the same time, this begs the question of why someone high in Nondisplay would be concerned more about creating a neutral social impression than a similar personal Perfectionistic Self-presentation / 84 impression. After all, ratings of negative mood or emotional distress do not experience the same suppression. This paradox will be discussed further in a later section. In conclusion then, perfectionistic self-presentation does seem to be related to cognitive appraisals in a clinical setting. Earlier research has found that perfectionistic self-presentation is related to interpersonal difficulties (Hewitt, Flett, Fehr, et al., 1996) and difficulties with openness in interpersonal relationships (Nielsen et al., 1997) and current findings suggest that these difficulties are also present in a therapeutic context. The findings also emphasize the importance of including the discrepancy between anticipated and expected performance when assessing standards (see also Kanfer & Zeiss, 1983; Wallace & Alden, 1991). One limitation that could be addressed in future research is the lack of information on the actual performance of the individual on the different social variables. It is possible that those high in perfectionistic self-presentation are accurate in predicting poorer performance during the interview and are not threatened by the interviewer's expectations, although on the whole this is not supported by ratings of observable anxiety or distress. A second suggestion would be to include a direct measure of the subjective experience of interpersonal judgment to supplement the discrepancy scores used here. This may help to tease apart the difference between Nondisplay and Nondisclosure in relation to threat. Finally, this study used expectations of social behavior but there may be other types of expectations or areas of disappointment, or other appraisals, that are also important. For example, one might feel pressured to produce interesting or dramatic life events, or be particularly sensitive to feedback on one's intelligence. Affect/Arousal The current study used self-report ratings of negative mood, observer ratings of anxiety in the interview and physiological measures of arousal to assess the level of affect and stress during the interview. The use of multiple rating systems allows for an extensive examination of the distress associated with self-disclosing in a clinical setting. Perfectionistic Self-presentation / 85 The primary finding in the area of self-reported negative affect is that perfectionistic self-presentation is positively related to distress both prior to and following the interview; these relations hold over and above the influence of demographic variables. Given the ties with competency appraisals discussed above, these results seem consistent with Schlenker and Leary's (1982) proposal that individuals who feel they will be unsuccessful in presenting a desired image will experience anxiety in an interpersonal situation, and are consonant with links between social anxiety and a fear of social disapproval (Arkin et al., 1986). It would seem that the desire to avoid displays of imperfection may be important in elevating distress prior to the interview, and the desire to avoid disclosing imperfections in greater distress after the interview. Nondisclosure was also associated with an increase in distress following the interview at the bivariate level, over and above pre-interview levels. Given that Nondisclosure was unrelated to observable anxiety and skin conductance indicators of threat, it is possible that post-interview ratings reflect a degree of rumination; the delay between the conclusion of the interview and the post-interview measure, although brief, may have allowed subjects to reflect back on their disclosure of serious situations and become more distressed than they were during the disclosure itself. Such a lowering in mood might be expected to the extent that rumination leads someone high in perfectionistic self-presentation to interpret the neutral position of the interviewer as indicating social disapproval; this has been found to be the case for those high in a need for approval who receive feedback indicating social disapproval (Whittal & Dobson, 1991). What is equally clear, however, is that there are multiple determinants of mood ratings. After considering the contribution of trait perfectionism and emotional adjustment, there was no relation between perfectionistic self-presentation and negative affect. Thus, although related to distress, perfectionistic self-presentation is not a strong unique predictor. With the exception of interviewers who may have had access to slightly different cues of anxiety, on the whole, external observers were unable to differentiate levels of perfectionistic self-presentation based on observable anxiety. This may reflect a restriction of range to a certain extent, as overall, subjects did Perfectionistic Self-presentation / 86 not seem to be highly anxious during the study. This low level of observed anxiety is particularly interesting given that some perfectionistic self-presenters did endorse considerable anxiety;28 it seems that they are better at presenting an image of being calm and together (more socially competent) than they anticipated or believe they had been able to achieve. Although not hypothesized, there were interesting differences between men and women that deserve comment. First, women high on Nondisclose seemed to be particularly distressed following the interview; compared to men they rated their mood to be far more negative than those low on this dimension. Similarly, they were more likely to be seen as anxious during the interview. It is possible that men have more difficulty in reporting negative mood when they do feel it, or are more motivated to appear calm and in control compared to women. Men have been shown to be more sensitive to the stigma attached to admitting to a need to seek help (Johnson, 1988) and this may be an extension of that concern.29 Although men might be particularly reluctant to display anxiety in front of a female interviewer, past research findings that both men and women are comfortable confiding in a female therapist (Garfield, 1994) make this unlikely. In any case, these findings suggest that although the interview stimulated similar content from both men and women (there were no differences between men and women in the seriousness of the problems discussed), the process of the interview in terms of verbal or nonverbal confessions of emotion was quite different. The results on physiological arousal during the interview are perhaps the most difficult to interpret but are particularly interesting. While on the whole the hypotheses were supported - perfectionistic self-presentation was related to increased arousal during the discussion of serious mistakes - this is only true when considering heart rate and Nondisclosure. 2 8 Correlations between subject's ratings of negative affect following the interview and observed anxiety were .23, p <05 and .30, p_<.01, for coders and interviewers respectively. Evaluated in terms of a reliability estimate, this suggests that overall, subjects were able to hide their anxiety quite well or perhaps denied anxiety that was obvious to others. 2 9 Although there were no differences in levels of depression or interaction anxiety between men and women, there would seem to be little stigma attached to confessing to being depressed or anxious in a clinical setting. Perfectionistic Self-presentation / 87 Consider first the pattern for the group as a whole. Figure 1 demonstrates that heart rate increased from relaxation when discussing the reason why the subject has come for treatment, was steady for the discussion of the first mistake, but decreased following that, reaching levels below relaxation after the interview. If, as proposed by Lang et al. (1983) heart rate is associated with cognitive processing, this would suggest that the interview presented a task that requires cognitive work. Similarly, Tomaka et al. (1993) found increases in heart rate under conditions of challenge. This is consistent with the conclusions of Fowles (1980) who suggests that cardiac activity reflects a response to motivation under threat; increases in heart rate therefore reflect active (in this case cognitive) efforts to gain internal or external incentives. It is unclear to what extent the decrease over the period of the interview represents the fact that Figure 1: Heart rate over the interview for the sample as a whole. 78 77 --76 --75 --1 69 --68 4 Relax Reason Here First Situation Second Situation Recovery Perfectionistic Self-presentation / 88 the task requires less cognitive attention over time, a drop in motivation, the effects of coping, or the effects of habituation. The patterns appear quite different when one considers heart rate taking into account level of Nondisclosure.30 In order to visualize the relations between heart rate and Nondisclosure, subjects were divided into 3 groups of high, moderate, and low Nondisclosure by a tercile split. Figure 23 1 shows the relation between Nondisclosure and heart rate over each point in the interview. While all subjects showed an increase over relaxation when discussing the reason why they were seeking treatment, those high on Nondisclose showed a significant increase over and above this level when faced with discussing a serious mistake, compared to those low on this dimension who actually began to decrease in arousal. It would appear that for those high on Nondisclose, thinking about a mistake to talk about and actually discussing that mistake, takes more cognitive effort. Given that increases in heart rate have been demonstrated when disclosing traumatic material (Pennebaker et al., 1987), and with exposure to feared objects (Lang et al., 1983) it seems reasonable to posit that these subjects were engaged in attempts to manage a distressing situation, perhaps focusing on attempts to gain the interviewer's approval or manage their disapproval. While not necessarily holding back on the type of situations, they seem to be more fervent in their attempts to handle the task. On the whole, the results for skin conductance did not support a significant relation between perfectionistic self-presentation and levels of arousal. Skin conductance is typically seen as being tied to emotion, increasing in response to threats of punishment (Fowles, 1980; Tomaka et al., 1993) and this lack of relation suggests that perfectionistic self-presenters were not particularly anxious during the interview. However, the analyses based on gender suggest that while this may be true for women, it is not for men. Men high on perfectionistic self-presentation did show higher skin conductance levels 3 0 In fact, the high Nondisplay group shows a similar pattern of heart rate, likely due to the correlation between the dimensions. However, the regressions suggest that, when it is possible to separate these variables statistically, Nondisclosure is uniquely relevant for heart rate. Thus, Nondisclosure and Nondisplay will be discussed as separate dimensions, consistent with the regression results. 3 1 Figures 2 and 3 are provided to aid in interpreting the results only - the data was not tested for significance because of the limitations of imposing group status on a continuous variable (Pedhazur, 1982). Perfectionistic Self-presentation / 89 Figure 2: Heart rate over the interview for tercile groups on Nondisclosure. during the interview, as displayed in Figure 3. This is surprising given the general lack of gender differences in skin conductance in the literature (Gilbert, 1991; Venables & Christie, 1980) and the lack of gender differences in overall SCL in the current study. In addition, it is particularly interesting that this involved the Nondisplay dimension rather than Nondisclosure. While there is little evidence to support gender differences in psychotherapy outcome in general (Garfield, 1994), these results bolster the suggestion by Shay (1996) that men may in fact experience self-disclosure in therapy quite differently, perhaps because of being socialized to resist vulnerability. Further, it is in line with findings that men react more strongly to high emotion conditions compared to low, whereas for women the difference is greater for high cognitive stress (Wallbott & Scherer, 1991). At the same time, it appears that higher arousal in the interview is true for only some men, specifically Perfectionistic Self-presentation / 90 Figure 3: Skin Conductance Level over the interview for men, according to tercile groups on Nondisplay. 3.2 T 1 0.7 -1 1 1 1 1 Relax Reason First Second Recovery Here Situation Situation those high in a desire to avoid displaying imperfections; further, it seems that the arousal is due to the interview situation in general and may not necessarily be tied to the exposure of imperfections. It must be remembered that, on the whole, these men are denying interpersonal threat on self-report of appraisals prior to the interview. This may be initial evidence that Nondisplay represents a self-deception component that is in some ways unconscious (Hewitt & Flett, 1993), as this pattern is similar to that found in the repression literature that links self-deception with increases in arousal under stress (Baumeister & Cairns, 1992; Weinberger & Davidson, 1994). At the same time, while research on repression focuses on denial of intrapsychic distress, nondisplay seems to focus on denials of interpersonal distress. Again, these Perfectionistic Self-presentation / 91 differences between men and women were not hypothesized but do suggest fruitful avenues for future research. Although these results are interesting, several caveats are required. First, this study clearly falls into Venables and Christie's (1980) group of studies (like others in the area of therapy; e.g. Horowitz, Stinson, et al., 1993) that are interested in psychophysiological variables as a behavioral response, rather than being an exhaustive study of the response as a psychophysiological phenomenon. These approaches are quite different in intent and process. While the current approach gives a broad picture of arousal over the course of the interview, providing information about the subject's experience of the interview that would be difficult to get via self-report, clearly there is a wealth of data that could be collected on the phenomenon of the arousal itself. More specific information on response latencies or recoveries, response to different levels of threat, examination of maximal arousals, etc., may all provide interesting and important insight into the phenomenon. For example, it is possible that SCL does correspond to perfectionistic self-presentation for women, but summarizing the results over a full minute, although minimizing random fluctuations, masks the effect. Addressing these limitations would require a degree of control that would not have been appropriate for an analogue study. Further, because of the differences between Nondisplay and Nondisclose, it would seem that this type of research would require fairly clear groups of perfectionistic self-presenters - subjects who were high on one dimension and low on the other. A second caveat is that, overall, correlations between the dimensions of perfectionistic self-presentation and both absolute levels of arousal and response over time, were relatively small. This leaves a great deal of variability in responses to be explained. This is a conclusion and frustration faced by many researchers attempting to link physiological variables to personality (see Gale & Edwards, 1986). Further, it is clear from the heart rate data that subjects did not experience particularly high levels of distress at any point in the interview; the average increase in heart rate between relaxation and discussing the first mistake Perfectionistic Self-presentation / 92 was only 4.14 beats per minute (SD=3.09). While this is reassuring from an ethical standpoint, it is not clear whether these differences are meaningful over and above being statistically significant. Finally, it must be acknowledged that the results are correlational in nature and this leaves open the possibility that the direction of the relation may run from physiological arousal and reactivity to heightened self-presentational concerns. Such a view of the biological bases for personality has been proposed by Eysenck (1967) and elaborated in many studies (e.g. Stemmler & Meinhardt, 1990). Differentiating these two models might involve assessing physiological reactivity under differing conditions of inter- and intrapersonal threat, or considering measures of predispositions to arousability as used in previous research (Coren & Mah, 1993). Finding that perfectionistic self-presentation was associated with arousal specific to interpersonal stress, or that perfectionistic self-presentation predicted physiological arousal over and above predispositions towards arousal, would provide support for a model whereby the desire to avoid being known as imperfect increases physiological stress. With these limitations in mind however, the results do generally support the hypotheses that perfectionistic self-presentation is related to negative affect (although the regression analyses suggest not uniquely, over and above trait perfectionism and other measures of distress) and arousal during the interview, albeit in interesting ways. Again, the Nondisclosure dimension seems the most consistently related to distress in the interview, although Nondisplay may have a role in distress for men. Behavior Tfie use of a structured interview that uses both standard intake questions as well as a request for self-disclosure of past mistakes, provides ample opportunity to observe self-presentational style. The current study included the opportunity to claim self-handicaps prior to the interview and excuse behavior following the interview, and rated overall presentation on the use of interpersonally defensive behaviors. The results suggest that perfectionistic self-presenters are somewhat defensive in their pattern of behavior although the results are mixed. Perfectionistic Self-presentation / 93 Self-handicaps provide an opportunity to avoid negative evaluation by forcing others to attribute negative outcomes to causes outside of the self when anticipating, or following, a negative performance (Berglas & Jones, 1978; Sheppard & Arkin, 1990). In this sample, subjects high on perfectionistic self-presentation were no more likely to claim a self-handicap than those low on these dimensions, even though both dimensions were related to the dispositional tendency to self-handicap as measured by the Self-Handicapping Scale. Given that the SHS itself showed a non-significant correlation with claiming a self-handicap (e.g. post interview, r=.19, p>.10) unlike previous research (Hirt et al., 1991), it is possible that this may have reflected the wording of the question in the current study. Some subjects did indicate that they were confused about the meaning of the question and it is possible that others passed the question for the same reason, without asking for clarification. A more precisely worded question such as "is there anything about you, your circumstances, or even the way your day has been going, that you think will stop you from doing your best during the interview?" or providing plausible reasons (as done by Baumgardner, 1991), might have generated different results. This is particularly likely given that, among those who did endorse the statement, claims of negative impact were strongly associated with perfectionistic self-presentation scores. In other words, perfectionistic self-presenters were more active in their use of the self-handicaps they did claim. Those high on Nondisclose were particularly likely to claim greater incapacitation following the interview. Given that these subjects also have a strong tendency to feel they did poorly on the interview, this incapacitation may serve as a means to explain away their perceived poor performance. Indeed, the interaction between Nondisclosure and anticipated performance predicts a more extreme self-handicap prior to the interview, and the interaction with perceived performance predicts a more extreme excuse after the interview.32 Thus it appears that under conditions of an anticipated or The interaction block predicting pre-interview self-handicapping (after controlling for depression, perfectionistic self-presentation, and anticipated performance), produced an r 2 change of. 16, p<05; the Nondisclose by anticipated performance interaction was significant at p <05. Predicting the extent of post interview self-handicapping, using the same controls, the interaction block r2change was .30, g<01 and the Nondisclose by perceived performance was significant at g<01. Perfectionistic Self-presentation / 94 perceived negative performance, those high on perfectionistic self-presentation try harder to avoid censure for their performance. Contrary to predictions, perfectionistic self-presentation was not related to observable defensive behaviors even though there was a tendency for those high on nondisclosure to limit the length of their disclosure. Perfectionistic self-presenters did not externalize their attributions for their behavior on self-report or during the interaction. Further, while there was evidence for both the presentation styles suggested by Friedlander & Schwartz (1985),33 these were not consistently related to perfectionistic self-presentation. This is somewhat surprising given the self-protective nature of this style. Although it is possible that asking for a serious mistake limited the extent to which excuses or justifications would have been seen as acceptable (McLaughlin at al., 1983), seriousness was in fact positively related to excuse making (r=.21, p<05) in the current sample. Unfortunately, difficulties with coding the transcripts and the necessary changes to the coding preclude concrete conclusions about behavior, as it is not possible to rule out the possibility that this lack of relations may be due to limitations in the coding system itself. At the same time, two lines of evidence offer some, albeit limited, support for the validity of the ratings. Correlations with the BIDR show that impression management was correlated with the extent of conceding (r=-.23, p<05) and had a tendency to be related to the extent of justifying (r=.18, p<10) during the discussion of the first mistake, controlling for the time spent discussing the situation. As noted earlier, the BIDR was also correlated with seriousness and culpability involved in the mistakes, implying that those high on this dimension chose problems that were less likely to promote a negative judgment. As these relations are what would be expected of an impression manager, it seems that coders were somewhat accurate in discerning these defensive behaviors. At the same time, it underscores the difference between impression management as a self-presentational style, and perfectionistic self-presentation. 3 3 As a point of interest, factor analysis suggests there may be one primary behavioral style that corresponds to a mixture of Friedlander and Schwartz's self-promotion and facework styles, including low concessions and high externalization, justifying, excusing and self-promotion. Perfectionistic Self-presentation / 95 A second line of evidence comes from the fact that interviewers seemed to agree that those who were rated as displaying greater levels of defensive behaviors were more defensive. Factor analysis of the PANAS completed by interviewers suggested that defensive, frustrated, and impatient loaded with irritable and hostile to form a defensive/aggressive factor. Although not related to the perfectionistic self-presentation dimensions, scores on this subscale correlated significantly with ratings of conceding (r=-.24, p_<.05), justifying (r=.26, p_<01) and self-promotion (r=.22, p_<05). While these correlations are relatively small, they do offer some preliminary support that the coders were seeing the same types of presentation as the interviewer. Taken together then, the overall pattern of results suggest that it may not simply be a limitation in the coding, but that there is, in truth, little to no relation between observable defensive behavior and perfectionistic self-presentation. This implies that the low correlations between perfectionistic self-presentation and measures of impression management such as the BIDR and the Marlowe-Crowne Socially Desirable Responding Scale (MCSDS; Crowne & Marlowe, 1960) in past (Hewitt, Flett, Fehr et al., 1996) and current research accurately reflect fewer attempts at impression management. Certainly, past research has differentiated between a motivation to self-present and actual self-presentational behavior (Lennox & Wolfe, 1984) and the fact that the desire to avoid displaying imperfections has been shown to be negatively related to a perceived ability to modify behavior in different situations (Hewitt, Flett, Fehr, et al., 1996) suggests that this is true of perfectionistic self-presenters. This conflict between wanting to present as perfect and not feeling able, or being able, to do so may explain much of the distress experienced by our participants. While the data certainly appears to suggest that perfectionistic self-presenters do not engage in overt attempts to avoid social judgment, there are several alternate explanations that should be considered. First, it is possible that the codes used here do not allow for a full exploration of interview behaviors; perfectionistic self-presenters may engage in a combination or pattern of behaviors that was not captured in Perfectionistic Self-presentation / 96 dimensional ratings. For example, it is possible that someone might be honest within the requirements of the task by discussing difficult situations, and concede responsibility but alternate this with excusing or justifying in order to undo the disclosure, with a net effect of appearing to be nondefensive. Secondly, it is possible that the task did not stimulate the defensiveness that truly is part of the presentation of a desire to avoid being known as imperfect. For example, perfectionistic self-presenters may respond to therapist cues that were muted in the current study by the requirement of neutrality and the structured nature of the interview. A therapist who gives a variety of cues that might be interpreted as shock, disapproval, or concern, might have elicited more defensiveness. Such patterns would require a process code similar to the Psychotherapy Interaction System by McCullough (1988) to capture the dynamic nature of the interaction. It is also possible that the current setting was not actually similar enough to a therapeutic encounter to stimulate true defensiveness. In particular, a lack of interpersonal investment - the fact that they would not be seeing the interviewer again - might moderate the expression of perfectionistic self-presentation. Thus, it is possible that participants may have risen to the occasion of a research situation and attempted to present in ways helpful to the researcher or appropriate for the study, even though this was against their natural inclination towards self-protectiveness. At the same time, the extent to which this is also true of the initial stages of therapy, when a client tries hard to be a "good client," is not clear. A final possibility is that subjects were not distressed enough during the interview. It may be in situations of higher arousal that defensive behaviors become more reflexive. Paulhus (1986) has demonstrated that arousal can produce a positive self-presentation style that is described as 'automatic' or reflexive, consonant with Brown and Rogers (1991) who found more self-serving attributions under Perfectionistic Self-presentation / 97 conditions of physiological arousal.34 Again, following individuals through therapy, with its intense emotions, might be particularly enlightening. Summary To date, research on perfectionistic self-presentation has established that those high in a desire to avoid being known as perfect struggle in interpersonal relationships (Hewitt, Flett, Fehr, et al., 1996). In intimate relationships, this relates to lower levels of trust and more withdrawal (Hewitt, Flett, & Callander, 1994). The current study suggests that these interpersonal patterns extend into the therapy arena, offering considerable support to the idea that perfectionistic self-presentation has indirect effects on psychopathology by its effect on therapy. Bringing the above results together, it would seem that on the whole perfectionistic self-presenters did experience the interview as threatening, and were stressed and distressed by the interview. The desire to avoid disclosure of imperfection stands out as a unique predictor of the interviewer as threatening and of increases in heart rate during self-disclosure, and is associated at the bivariate level with post-interview negative mood and the extent of self-handicapping. This pattern is consistent with the nature of the task as a disclosure of one's difficulties, shortcomings, and past mistakes and previous research linking this dimension to difficulties in close relationships (Hewitt, Flett, Fehr, et al., 1996). Nondisplay was a unique predictor of appraisals of the interviewer as less threatening, and associated at the bivariate level with greater negative mood prior to the interview, higher skin conductance for men, and higher visible anxiety during the interview. It is clear that perfectionistic self-presenters are not trying to provide an overly positive image as would an impression manager, but there is some evidence that perfectionistic self-presenters still manage at least part of their image. Supplementary analyses between perfectionistic self-presentation and the An example of this was provided by one subject, high on Nondisplay, who described his reaction to making a mistake in front of someone as involving a "sudden flush," during which his mind goes blank even though he finds himself immediately blaming someone or something else, even if the mistake is clearly his fault. Perfectionistic Self-presentation / 98 MCSDS confirms previous research that perfectionistic self-presenters are not self-deceiving (Hewitt, Flett, Fehr, et al., 1996) but the negative direction (r=-.28, rj<01 and r=-.23, p.<05 for Nondisplay and Nondisclose respectively) implies that they may in fact go somewhat out of their way to avoid appearing grandiose or self-delusional.35 This is consistent with other findings that suggest that people will modify their presentations to avoid appearing boastful (Miller & Schlenker, 1985; Tetlock & Manstead, 1985). There is some preliminary evidence supporting this presentational tightrope walk from the behavioral data; while subjects high in impression management provided more trivial situations as evidence of their shortcomings, and depressed patients provided more serious ones, subjects high in perfectionistic self-presentation consistently did neither. More specifically, while high depressed subjects appeared more self-critical (see also Friedlander & Schwartz, 1985) by providing more serious and more blameworthy situations than low depressed subjects (r= .23 and r=.25, p<05), high perfectionistic self-presentation subjects (either Nondisclose or Nondisplay) did not differ from low. Given that perfectionistic self-presentation is so highly correlated with depression, at the very least, we should perhaps consider the question "why did perfectionistic self-presenters not act as depressed as they were?" Perhaps they were moderating their presentation to reflect that of a 'reasonable person.' Typically, self-presentation styles have been assumed to reflect relative extremes - socially desirable responding as positive, self-denigration as negative. In clinical samples, response scales are used to assess the extent of these styles (e.g. The MMPI includes an estimate of both "faking good" and "faking bad;" Golden, 1990). However, if you consider someone who is emotionally distressed, lacks self-confidence, but does not want to be known as imperfect, you might expect a presentation that is somewhere between positive and self-denigrating, closer to that of a 'reasonable' or 'typical' person. This might in fact be the most protective of their inner reality that they can get. To be overly positive is to risk being seen as arrogant, and may be difficult to pull off for someone who does not feel overly positive or in situations 3 5 Those high on perfectionistic self-presentation tended to score below mean levels found in other research on the MCSDS (Zook & Sipps, 1985). Perfectionistic Self-presentation / 99 where others are seen as 'experts' in understanding people. To be overly self-denigrating is to risk being judged harshly. This implies that the generally flat presentation of our subjects (not too defensive, not too self-denigrating) may be self-protective in its own right. Past research provides support for an ability to present an image that is tailored to a particular situation (e.g. to match different job descriptions; Scandell & Wlazelek, 1996) suggesting that people can be quite sophisticated in their responses to self-presentational demands. Differences between presentations of standards, measures of emotional distress, or observable behaviors may, in fact, reflect differences in the accessibility of the standards for a reasonable person. In clinical settings, it is clearly acceptable to be distressed, but many clients will not have had access to the appraisals and presentational behaviors of the reasonable person during therapy. This implies both a flexibility in presentation in response to the demands of the situation in response to available cues about appropriate responses, and an inflexible adherence to what is to some extent a false representation. The data suggest that the desire to avoid displays of imperfection may be more strongly associated with pursuit of the image of a reasonable person. The low self-image and emotional distress typical of those high on non-display without accompanying appraisals of threat or self-denigrating behaviors supports this contention. Unfortunately, it is not possible with the current data to determine the extent to which someone high on Nondisplay is aware of social threat but downplaying it versus being unaware of the threat until it is experienced. Finding higher skin conductance levels for men high on this dimension during the interview, coupled with a relative denial of negative mood during the interview and lower ratings of threat, could follow from either explanation. While underscoring the conceptual differences between the two dimensions, it also suggests that these two dimensions will function quite differently in therapy. An important goal for further research might be to understand the impact of assumptions about a 'reasonable person' on appraisals and presentation, and to parse out the effects of interpersonal threat. A second challenge would be to attempt to understand the roots of these dimensions - to understand why the desire to avoid displaying imperfection is different from the desire to avoid disclosing imperfection. Perfectionistic Self-presentation/ 100 Implications for therapy In summary then, clinical subjects who are high on perfectionistic self-presentation seem to experience a clinical interview as interpersonally threatening and are more stressed and distressed by the process of the interview. Although this analogue study is somewhat limited in the extent to which it might be generalizable to a true clinical setting (see below), there are several implications for therapy that should be considered and explored with further research. The current study adds to the extant literature regarding the impact of trait perfectionism on therapy outcome. Higher levels of perfectionism have been associated with poor response to treatment on different outcome measures across different types of therapy over and above pre-treatment levels on the depression and adjustment variables (Blatt, Quinlan, Pilkonis, & Shea, 1995). Analyses of data from the large scale National Institute of Mental Health Treatment of Depression Collaborative Research Program (NIMH-TDCRP) suggested that this effect may be at least partially moderated by the quality of the therapeutic relationship. Blatt and his colleagues (Blatt, Zuroff Quinlan, & Pilkonis, 1996) found that perfectionism interacts with a therapeutic or working alliance between patient and therapist, although this is true only for those with moderate levels of perfectionism. In contrast to low levels of perfectionism where patients seem to do well regardless of the quality of the therapy relationship, those with moderate levels of perfectionism did better with a positive relationship and poorer with a less positive therapeutic relationship. At extreme levels of perfectionism, therapy outcome was largely independent of ratings of the therapeutic alliance; these patients did poorly regardless of their ratings of the therapeutic relationship. The current study suggests that rather than pure perfectionism at the trait level, it may be the desire to avoid being known as imperfect that impacts the quality of the therapeutic alliance. Those high in perfectionistic self-presentation arrive with a fair degree of anticipatory anxiety, low expectations of their own abilities, and expectations that they will disappoint their therapist. Coupled with the perfectionistic self-presenter's difficulties with intimate relationships (Hewitt, Flett, Fehr, et al., 1996), this distress and Perfectionistic Self-presentation / 101 distrust of the therapist may interfere with the formation of a working relationship in therapy, as has been clearly demonstrated in previous research (Moras & Strupp, 1982; Muran, Segal, Samstag, & Crawford, 1994; Piper et al., 1991). That is, these patients appear to come into therapy handicapped in regards to forming the early therapy alliance that is an important determinant of outcome (Kokotovic & Tracey, 1990). It is perhaps equally sobering to realize that it may be difficult for therapists to discern who is having difficulty with self-disclosure, as the handicaps appear well hidden. Although those high in perfectionistic self-presentation may be somewhat less talkative and more visibly anxious, they seem to provide fairly typical responses to requests for self-disclosure, and don't appear overly defensive. Such an impairment in the early stages of the alliance is demonstrated in the current study. At the completion of the interview, interviewers were asked how much they liked the subject and how willing they would be to have the subject as a client. Controlling for interaction anxiety (interviewers were more willing to have highly socially anxious subjects as clients compared to low), there were negative relations between the extent the interviewer liked the client and Nondisclose (r=-.24, p<.05) and their willingness to have the subject as a client and both perfectionistic self-presentation dimensions (r=-27, p<.01 for Nondisplay and r=-.29, p<01 for Nondisclose). While the factors that contributed to the interviewers rating these clients less positively are not clear, other research suggests that therapists are particularly sensitive to cues regarding the patient's ability to be genuinely emotionally open (Helstone & Vanzurren, 1996); it is possible that our interviewers were able to distinguish subtle cues that differentiate self-disclosure that includes vulnerability from that which is a mere recitation of facts. In any case, these results provide preliminary evidence that the early therapeutic alliance may be fragile for perfectionistic self-presenters in therapy. In addition to the nature of the relationship between therapist and client, the results suggest that the actual process and content of therapy may require different attention for those high in perfectionistic self-presentation. Certainly, it will be important to pay special attention to the client's response to feedback; the Perfectionistic Self-presentation / 102 negativity in our subjects' appraisals suggests that perfectionistic self-presenters may be prone to mislabeling therapist's comments or lack of comments as reflecting negative judgments. Further, to the extent that perfectionistic self-presentation represents concerns about relationships, an interpersonal approach may be most appropriate. Blatt and his colleagues have established that such patients seem to respond better to interpersonal approaches and demonstrate change in improved social relationships (Blatt, Ford, Berman, Cook, & Meyer, 1988). One of the needs that interpersonal therapy might be particularly suited to meeting is for the perfectionistic self-presenter to be known as imperfect in the absence of interpersonal judgment; while it would be hoped that this is possible in any therapy, interpersonal approaches are particularly designed for understanding the client as he or she goes through that experience (Teyber, 1988). Three closing speculations about therapy: First, it is possible that there will be different events during therapy that will trigger distress for those high in Nondisclosure and Nondisplay. The data provide evidence that the client who wants to avoid admitting to imperfection will experience episodes of disclosure as distressing, and suggest they may ruminate following the session to a greater extent. In contrast, the client who does not want to be seen as imperfect may have particular difficulty in handling visible mistakes, such as being 'caught' in an inconsistency, being late for, or missing, an appointment. Second, there is a group of clients that present to therapy even though they may be highly resistant to seeking help. That is, some clients come at the request of others such as governmental agencies (court-ordered), or family members (for marital or family therapy). When such clients also have a high need to avoid being known as imperfect, it may be particularly difficult to establish a non-threatening, honest relationship. This group of clients may well display the defensive behaviors expected in the current study, as such behaviors would not be outside the presentation of a 'reasonable person' who is not in need of psychological help. Perfectionistic Self-presentation / 103 Finally, it is likely that group therapy will be particularly difficult for this group of patients. The multiple sources of judgment and the inability to assure everyone's positive opinion, as well as the increased likelihood of making what would be judged to be a mistake by some member of the group, might translate into a high level of distress. At the same time, if this distress can be constructively managed, it is possible that the perfectionistic self-presenter would respond well to group therapy. The experience of being known as imperfect without rejection by a greater number of people, as well as modeling of self-disclosure, may be particularly effective. Limitations and suggestions for further research Although many of the limitations of the current study have been mentioned during the different sections discussed above, some deserve special emphasis, and some more global limitations should be pointed out. Many arise from the fact that the study was designed as a global assessment of a number of domains of experience in a setting that is related to therapy but not actually therapy. Such a design provides a valuable initial investigation in a new area of research but more exhaustive examinations of each of the domains of cognition, affect, and behavior, as well as clarification of the links between the domains, is clearly warranted. Obviously, the correlational nature of the design puts certain constraints on interpreting the results. Although the repeated measures pre- and post-interview offer an improvement over a single-timepoint design, it is not possible to draw conclusions regarding causality. Some possibilities for future research might be to manipulate the level of interpersonal threat both prior to the interview (providing profiles of the interviewer that are intimidating or supportive) or demonstrated by the interviewer (cold or warm). Alternately, manipulating views of the 'reasonable person' may clarify the extent to which behavior for perfectionistic self-presenters is deliberate if not extreme. Even within a non-experimental design, screening subjects and choosing those that are high on only one dimension might be helpful in clarifying the distinctions between the two dimensions. At the least, replication of the findings is necessary, particularly Perfectionistic Self-presentation / 104 in regards to those that were unexpected such as the relation between Nondisplay and lower threat, and the gender differences in physiological arousal. Second, it is clear that while there were many similarities between this study and a clinical setting, it was not one. Participants knew that the interviewer was a student, that they would not see her again, and that she was following a set list of questions. This formal structure and the neutrality of the interviewer were the factors most often identified by participants as being unlike what they knew of therapy or believed therapy to be like. Further, it is unknown to what extent the constraints of physiological monitoring or videotaping influenced the results (although most subjects reported that they forgot about both after a short while). At the same time, subjects were recruited from actual patient lists as they approached various institutions for therapy, and reported to a credible setting (a Psychology Department at a well-known university), to interact with Doctoral students in Psychology. These factors should promote generalizability to outpatient settings. Third, it should be made explicit that these results deal with a somewhat limited patient sample. Although the sample was typical of those seen in the outpatient clinics in the area, the sample was restricted in terms of ethnic background, and psychiatric diagnosis. For example, the sample did not contain anyone dealing with anorexia, schizophrenia, or phobias. Indeed, further research would be needed to determine if the results of the study might be particularly applicable to specific diagnostic groups. It is also clear that our participants may have been somewhat different from people dealing with the same difficulties but who have not yet sought treatment. We know nothing of the pre-treatment struggle with the fear of being known as imperfect, only that seeking treatment indicates that there has been at least some measure of victory. Fourth, additional questions arose from the data that suggested useful avenues of future research. The possibility of significant gender differences was raised in the exploratory analyses and seems like an important variable for further consideration. Although the use of female therapists maximizes the potential Perfectionistic Self-presentation / 105 for self-disclosure (e.g., Veroff, Kulka, & Douvan, 1981), we know nothing about the influence of having a male therapist/interviewer, or having one that is of the opposite gender to the subject. Looking at the effect of pre-interview coping, as well as coping attempts within the interview may also have provided useful information (e.g. Zeidner, 1994). Further, it would have been interesting to follow participants for a longer period of time after the interview to examine the long term effects of self-disclosure. In particular, the possibility that perfectionistic self-presenters ruminate about disclosure and experience greater negative mood, particularly if they feel they did poorly, may help to explain at least some of the patients that fail to return to therapy after their first session. Finally, it must be acknowledged that although many of the relations found in the study were statistically significant, many were not particularly large. It is clear that there are still large amounts of variance left to be explained by factors other than perfectionistic self-presentation. The extent to which these relations are meaningful as well as statistically significant remains to be established. This would only seem possible when the study of perfectionistic self-presentation is moved into a real therapy setting. Conclusion Patients come into therapy with a complex set of thoughts, emotions, and fears. Even those who want to avoid being known as imperfect are rarely concerned with only verbal or nonverbal disclosure. At the same time, we see in this current study a picture of how hard it is for some of our clients to be known as imperfect. They are more depressed when they come, and more distressed by coming. They feel particularly vulnerable, assuming they will do poorly and will be judged for that. Despite being able to hold it together on the outside, they are more stressed by the process and feel worse afterwards. They feel they have done poorly and have been a disappointment. In essence, they experience their worst nightmare -exposing their imperfections to another - often without the reassurance of a sympathetic audience. As clinicians, we need to recognize this, and acknowledge their courage. Perfectionistic Self-presentation / 106 References Ahrens, A. H. , Zeiss, A. M . , & Kanfer, R. (1988). Dysphoric deficits in interpersonal standards, self-efficacy, and social comparison. Cognitive Therapy and Research, 12, 53-67. Alden, L. E. , Bieling, P. J., & Wallace, S. T. (1994). Perfectionism in an interpersonal context: A self-regulation analysis of dysphoria and social anxiety. Cognitive Therapy and Research, 18, 297-316. Alden, L. E. , Teschuk, M . , & Tee, K. (1992). Public self-awareness and withdrawal from social interactions. Cognitive Therapy and Research, 16, 249-267. Alden, L. E. , & Wallace, S. T. (1991). Social standards and social withdrawal. Cognitive Therapy and Research. 15, 85-100. Arkin, R. M . (1981). Self-presentation styles. In J. T. Tedeschi (Ed.). Impression management theory and social psychological theory (pp. 311-333). New York: Academic Press. Arkin, R. M . (1986). Self-presentation strategies and sequelae. In S. L . Zelen (Ed.), Self-representation (New York: Springer-Verlag. Arkin, R. M . , & Baumgardner, A. H. (1985). Self-handicapping. In J. H. Harvey, & G. W. Weary (Ed), Attribution: Basic issues and applications (pp. 169-202). San Diego, CA: Academic Press. Arkin, R. M . , Lake, E. M . , & Baumgardner, A. H . (1986). Shyness and self-presentation. In W. H. Jones, J. M . Cheek, & S. R. Briggs (Ed.), Shyness: Perspectives on research and treatment (pp. 189-204). New York: Plenum. Baumeister, R. F. (1982). A self-presentational view of social phenomena. Psychological Bulletin, 91, 3-26. Baumeister, R. F., & Cairns, K. J. (1992). Repression and self-presentation: When audiences interfere with self-deceptive strategies. Journal of Personality and Social Psychology, 62, 851-862. Baumeister, R. F., Hutton, D. G., & Tice, D. M . (1989). Cognitive processes during deliberate self-presentation: How self-presenters alter and misinterpret the behavior of their interaction partners. Journal of Experimental Social Psychology, 25, 59-78. Baumeister, R. F., & Tice, D. M . (1986). Four selves, two motives, and a substitute self-regulation model. In R. F. Baumeister (Ed.), Public self and private self (pp. 63-74). New York: Springer-Verlag. Baumgardner, A. H. (1991). Claiming depressive symptoms as a self-handicap: A protective self-presentation strategy. Basic and Applied Social Psychology, 12, 97-113. Beck, A. T. (1976). Cognitive therapy and the emotional disorders. New York: International Universities Press. Beck, A. T., & Beck, R. W. (1972). Screening depressed patients in family practice: A rapid technique. Postgraduate Medicine,, 81-85. Perfectionistic Self-presentation / 107 Beck, A. T., Rial, W. Y. , & Rickels, K. (1974). Short form of Depression Inventory: Cross-validation. Psychological Reports. 34, 1184-1186. Beck, A. T., Steer, R. A., & Garbin, M . G. (1988). Psychometric properties of the Beck Depression Inventory: Twenty-five years of evaluation. Clinical Psychology Review. 8, 77-100. Beck, A. T., Ward, C. H. , Mendelson, M . , Mock, J., & Erbaugh, J. (1961). An inventory for measuring depression. Archives of General Psychiatry, 4, 561-571. Ben-Shakhar, G. (1985). Standardization within individuals: A simple method to neutralize individual differences in skin conductance. Psychophysiology. 22, 292-239. Berglas, S., & Jones, E. E. (1978). Drug choice as a self-handicapping strategy in response to noncontingent success. Journal of Personality and Social Psychology, 36, 405-417. Blatt, S. J., Ford, W., Berman, B., Cook, B., & Meyer, R. (1988). The assessment of therapeutic change in schizophrenic and borderline young adults. Psychoanalytic Psychology, 5, 127-158. Blatt, S. J., Quinlan, D. M . , Pilkonis, P. A., & Shea, M . T. (1995). Impact of perfectionism and need for approval on the brief treatment of depression: The National Institute of Mental Health Treatment of Depression Collaborative Research Program revisited. Journal of Consulting and Clinical Psychology, 63, 125-132. Blatt, S. J., Zuroff, D C , Quilan, D. M , & Pilkonis, P. (1996). Interpersonal factors in brief treatment of depression: Further analyses of the NIMH Treatment of Depression Collaborative Research Program. Journal of Consulting and Clinical Psychology, 64, 162-171. Blondin, J., & Waked, E. G. (1991). Cardiovascular responses, performance, and mood in heart rate reactive individuals during a challenging cognitive task. Personality and Individual Differences, 12, 825-834. Bonime, W. (1960). Depression as practice: Dynamic and psychotherapeutic considerations. Comprehensive Psychiatry. I, 194-198. Borenstein, M . , & Cohen, J. (1988). Statistical Power Analysis: A computer program. Hillsdale, NJ: Lawrence Erlbaum. Boucsein, W. (1992). Electrodermal activity. New York: Plenum. Bradley, G. W. (1978). Self-serving biases in the attribution process: A reexamination of the fact or fiction question. Journal of Personality and Social Psychology, 36, 56-71. Brown, J. D., & Gallagher, F. M . (1992). Coming to terms with failure: Private self-enhancement and public self-effacement. Journal of Experimental Social Psychology, 28, 3-22. Brown, J. D., & Rogers, R. J. (1991). Self-serving attributions: The role of physiological arousal. Personality and Social Psychology Bulletin, 17, Perfectionistic Self-presentation / 108 Buntrock, C. N. , & Reddy, D. M. (1992). Coping dispositions and the stress appraisal process: The impact of defensiveness on emotional response to threat. Personality and Individual Differences, 13, 1223-1231. Burisch, M. (1984). Approaches to personality inventory construction. American Psychologist, 39, 214-227. Bums, D. (1980). The perfectionist's script for self-defeat. Psychology Today, Nov, 34-52. Buss, A. H., & Finn, S. E. (1987). Classification of personality traits. Journal of Personality and Social Psychology, 52, 432-444. Canadian Pharmaceutical Association (1996). Compendium of pharmaceuticals and specialties (Canada). Toronto: Canadian Pharmaceutical Association. Cantor, N . (1990). From thought to behavior: "Having" and "doing" in the study of personality and cognition. American Psychologist, 145, 735-750. Cohen, J., & Cohen, P. (1983). Applied multiple regression/correlation analysis for the behavioral sciences.(2 ed.). New Jersey: Lawrence Erlbaum Associates Inc. Contrada, R. J., Leventhal, H , & O'Leary, A. (1990). Personality and health. In A. Pervin (Ed), Handbook of personality theory and research (pp. 638-669). New York: Guilford. Cooley, C. H. (1902). Human nature and the social order. New York: Scribner's Sons. Coren, S., & Mah, K. B. (1993). Prediction of physiological arousability: A validation of the Arousal Predisposition Scale. Behaviour Research and Therapy, 31, 215-219. Cribbie, R. A., & Jamieson, J. J. (1997). The measurement of change using structural equation modeling. Poster presented at the Canadian Psychological Association Annual Convention, Toronto, Ont. Cronbach, L. J., & Furby, L. (1970). How should we measure "change" - or should we? Psychological Bulletin, 74, 68-80. Crowne, D. P., & Marlowe, D. (1960). A new scale of social desirability independent of psychopathology. Journal of Consulting Psychology, 24, 349-354. DePaulo, B. M. , Epstein, J. A., & LeMay, C. S. (1990). Responses of the socially anxious to the prospect of interpersonal evaluation. Journal of Personality, 58, 623-640. Derlega, V. J., Metts, S., Petronio, S., & Margulis, S. T. (1993). Self-disclosure in close relationships. London: Sage. Dillard, J. P., & Kinney, T. A. (1994). Experiential and physiological response to interpersonal influence. Human Communication Research, 20. 502-528. Doherty, K., & Schlenker, B. R. (1991). Self-consciousness and strategic self-presentation. Journal of Personality, 59, 1-18. Perfectionistic Self-presentation / 109 Donat, D. C , & McCullough, J. P. (1983). Psychophysiological discriminants of depression at rest and in response to stress. Journal of Clinical Psychology, 39, 315-320. Dow, M . G. , Biglan, A., & Glaser, S. R. (1985). Multimethod assessment of socially anxious and socially nonanxious women. Behavioral Assessment. 7, 273-282. Ducharme, J., & Bachelor, A. (1993). Perception of social functioning in dysphoria. Cognitive Therapy and Research. F7, 53-70. Elbin, S. D., & Weary, G. (1987). Stressful life events, self-presentation, and the depressive cognitive style. Unpublished manuscript, Ohio State University. Ellis, A. (1962). Reason and emotion in psychotherapy. New York: Lyle Stuart. Endler, N . S., King, P. R , Edwards, J. M . , Kuczynski, M . , & Diveky, S. (1983). Generality of the interaction model of anxiety with respect to two social evaluation field studies. Canadian Journal of Behavioral Science, 15, 60-69. Eysenck, H. J. (1967). The biological bases of personality. Springfield, 111.: Thomas. Falloon, I. R. H. , Lloyd, G. G. , & Harpin, R. E. (1981). The treatment of social phobia: Real-life rehearsal with nonprofessional therapists. Journal of Nervous and Mental Disease, 169, 180-184. Fenichel, O. (1941). Problems of psychoanalytic technique. New York: Psychoanalytic Quarterly. Ferrari, J. R. (1992). Procrastinators and perfect behavior: An exploratory factor analysis of self-presentation, self-awareness, and self-handicapping components. Journal of Research in Personality, 26, 75-84. Fincham, F. D., Beach, S., & Nelson, G. (1987). Attributional processes in distressed and nondistressed couples: 3. Causal and responsibility attributions for spouse behavior. Cognitive Therapy and Research, i i , 71-86. Fischer, E. H. , & Turner, J. L. (1970). Orientations to seeking professional help. Development and research utility of an attitude scale. Journal of Consulting and Clinical Psychology, 35, 79-90. Fletcher, G. J. O., & Fitness, J. (1990). Occurrent social cognition in close relationship interaction: The role of proximal and distal variables. Journal of Personality and Social Psychology, 59, 464-474. Flett, G. L. , Hewitt, P. L. , Blankstein, K. R., & Koledin, S. (1991). Dimensions of perfectionism and irrational thinking. Journal of Rational-Emotive and Cognitive-Behavior Therapy, 9, 185-201. Flett, G. L . , Hewitt, P. L . , Blankstein, K. R., & Mosher, S. W. (1991). Perfectionism, self-actualization, and personal adjustment. Journal of Social Behavior and Personality, 6, 147-160. Flett, G. L . , Hewitt, P. L , Blankstein, K. R., & Mosher, S. W. (1995). Perfectionism, life events, and depressive symptoms: A test of a diathesis-stress model. Current Psychology, 14, Flett, G. L. , Hewitt, P. L. , Blankstein, K. R., & O'Brien, S. (1991). Perfectionism and learned resourcefulness in depression and self-esteem. Personality and Individual Differences, 12, 61-68. Perfectionistic Self-presentation / 110 Flett, G. L. , Hewitt, P. L . , & DeRosa, T. (1996). Dimensions of perfectionism, psychosocial adjustment, and social skills. Personality and Individual Differences, 20, 143-150. Flett, G. L. , Hewitt, P. L . , Endler, N. S., & Tassone, C. (1995). Perfectionism and components of state and trait anxiety. Current Psychology, 13, 326-350. Flett, G. L. , Sawatzky, D. L. , & Hewitt, P. L. (1995). Dimensions of perfectionism and goal commitment: A further comparison of two perfectionism measures. Journal of Psychopathology and Behavioral Assessment. 17, 111-124. Flett, G. L. , Smith, J., Hewitt, P. L. (1994). [Perfectionism and social interactions]. Unpublished data. Flynn, C. (1996). Self-oriented perfectionism and stress enhancement: Physiological and emotional reactivity in response to an achievement related stressor. Unpublished master's Thesis, University of British Columbia. Fowles, D. C. (1980). The three arousal model: Implications of Gray's two-factor learning theory for heart rate, electrodermal activity, and psychopathy. Psychophysiology, 17, 87-104. Frankish, J., & Linden, W. (1991). Is response adaptation a threat to the high-low reactor distinction among female college students. Health Psychology, 10, 224-227. Friedlander, M . L . , & Schwartz, G. S. (1985). Toward a theory of strategic self-presentation in counseling and psychotherapy. Journal of Counseling Psychology, 32, 483-501. Frieswyk, S. H . , Allen, J. G. , Colson, D. B., Coyne, L. , Gabbard, G. O., Horwitz, L . , & Newsom, G. (1986). Therapeutic alliance: Its place as a process and outcome variable in dynamic psychotherapy research. Journal of Consulting and Clinical Psychology, 54, 32-38. Frost, R. O., Marten, P., Lahart, C , & Rosenblate, R. (1990). The dimensions of perfectionism. Cognitive Therapy and Research, 14, 449-468. Frost, R. O., Turcotte, T. A., Heimberg, R. G , Mattia, J. I., Holt, C. S., & Hope, D. A. (1995). Reactions to mistakes among subjects high and low in perfectionistic concern over mistakes. Cognitive Therapy and Research, 19, 195-205. Gale, N. , & Edwards, J. A. (1986). Individual differences. In M . G. H . Coles, E. Donchin, & S. W. Porges (Ed.), Psychophysiology (pp. 431-508). New York: Guilford Press. Gallop, R., Kennedy, S. H. , & Stern, ? (1994). Therapeutic alliance on an inpatient unit for eating disorders. International Journal of Eating Disorders, 16, 405-410. Garfield, S. L . (1994). Research on client variables in psychotherapy. In A. E. Bergin, & S. L. Garfield (Ed.), Handbook of psychotherapy and behavior change (pp. 190-228). New York: John Wiley & Sons. Gilbert, B. O. (1991). Physiological and nonverbal correlates of extraversion, neuroticism, and psychoticism during active and passive coping. Personality and Individual Differences, 12, 1325-1331. Golden, C. (1990). Clinical interpretation of objective psychological tests (2nd ed.). Boston: Allyn and Bacon. Perfectionistic Self-presentation / 111 Gonzales, M . H. , Manning, D. L. , & Haugen, J. A. (1992). Explaining our sins: Factors influencing offender accounts and anticipated victim responses. Journal of Personality and Social Psychology, 62, 958-971. Gonzales, M . H. , Pederson, J. H. , Manning, D. J., & Wetter, D. W. (1990). Pardon my gaffe: Effects of sex, status, and consequence severity on accounts. Journal of Personality and Social Psychology, 58, 610-621. Gotlib, I. H. , & Meltzer, S. J. (1987). Depression and the perception of social skill in dyadic interaction. Cognitive Therapy and Research. 11. 41-54. Green, S. B. (1991). How many subjects does it take to do a regression analysis? Multivariate Behavioral Research. 26, 499-510. Greenwald, A. G. , & Breckler, S. J. (1985). To whom is the self presented? In B. R. Schlenker (Ed), The self and social life (pp. 126-145). New York: McGraw-Hill. Guest, H. (1990). Sequential analysis: Monitoring counseling sessions via skin resistance. Counseling Psychology Quarterly. 3, 85-91. Habke, A. M . (1991). The relation of unsupportive spousal actions by the spouse to marital satisfaction. Masters, University of British Columbia. Hamachek, D. E. (1978). Psychodynamics of normal and neurotic perfectionism. Psychology. 15, 27-33. Hammond, G. (1996). The objections to null hypothesis testing as a means of analyzing psychological data. Australian Journal of Psychology, 48, 104-106. Han, H. , & Hewitt, P. L. (1996). Perfectionism and attitudes towards help-seeking. Unpublished manuscript. Hastrup, J. L . , & Light, K. C. (1984). Sex differences in cardiovascular stress responses: Modulation as a function of menstrual cycle phases. Journal of Psychosomatic Research, 28, 475-483. Heimberg, R. G , Mueller, G. P., Holt, C. S., Hope, D. A., & Liebowitz, M . R. (1992). Assessment of anxiety in Social Interaction and being observed by others: The Social Interaction Anxiety Scale and the Social Phobia Scale. Behavior Therapy. 23, 53-73. Helstone, F. S., & Vanzuuren, F. J. (1996). Clinical decision making in intake interviews for psychotherapy: A qualitative study. British Journal of Medical Psychology. 69. 191-206. Hewitt, P. L . , & Dyck, D. G. (1986). Perfectionism, stress, and vulnerability to depression. Cognitive Therapy and Research, 10, 137-142. Hewitt, P. L , & Flett, G. L. (1990). Perfectionism and depression: A multidimensional analysis. Journal of Social Behavior and Personality, 5, 423-438. Hewitt, P. L . , & Flett, G. L. (1991a). Dimensions of perfectionism in unipolar depression. Journal of Abnormal Psychology, 100. 98-101. Perfectionistic Self-presentation / 112 Hewitt, P. L . , & Flett, G. L. (1991b). Perfectionism in the self and social contexts: Conceptualization, assessment, and association with psychopathology. Journal of Personality and Social Psychology. 60, 456-470. Hewitt, P. L . , & Flett, G. L. (1993a). Dimensions of perfectionism, daily stress, and depression: A test of the specific vulnerability hypothesis. Journal of Abnormal Psychology, 102, 58-65. Hewitt, P. L . , & Flett, G. L. (1993b). Perfectionistic self-presentation and maladjustment. Symposium presented at the Annual Convention of the American Psychological Association, Toronto, Ont: Hewitt, P. L . , & Flett, G. L. (1996). Perfectionistic self-presentation in a clinical sample. Unpublished manuscript. Hewitt, P. L . , Flett, G. L. , & Blankstein, K. R. (1991). Perfectionism and neuroticism in psychiatric patients and college students. Personality and Individual Differences, 12, 273-279. Hewitt, P. L . , Flett, G. L. , & Callander, L. (1994). Perfectionism in spousal interactions. Unpublished manuscript, Hewitt, P. L . , Flett, G. L. , & Ediger, E. (1995). Perfectionism traits and perfectionistic self-presentation in eating disorder attitudes, characteristics, and symptoms. International Journal of Eating Disorders,, Hewitt, P. L . , Flett, G. L. , & Ediger, E. (1996a). Perfectionism and depression: Longitudinal assessment of a specific vulnerability hypothesis. Journal of Abnormal Psychology, 105, 276-280. Hewitt, P. L . , Flett, G. L. , & Ediger, E. (1996b). Perfectionism traits and perfectionistic self-presentation in eating disorder attitudes, characteristics, and symptoms. International Journal of Eating Disorders., Hewitt, P. L . , Flett, G. L. , & Endler, N. S. (1995). Perfectionism, coping and depression symptomatology in a clinical sample. Clinical Psychology and Psychotherapy, 2, 47-58. Hewitt, P. L . , Flett, G. L . , & Fehr, B. (1994). [Perfectionistic self-presentation]. Unpublished data . Hewitt, P. L . , Flett, G. L , Fehr, B., Habke, A. M . , & Fairlie, P. (1996). Self-presentational style and maladjustment: Presenting the perfect self. Unpublished manuscript. Hewitt, P. L . , Flett, G. L. , Flynn, C , & Nielsen, A. (1995). Development and validation of an interview measure of perfectionistic behavior. Poster presented at the Annual Convention of the Canadian Psychological Association Charlottetown, PEI: Hewitt, P. L . , Flett, G. L. , & Mikail, S. F. (1995). Perfectionism and relationship adjustment in pain patients and their spouses. Journal of Family Psychology, 9, 335-347. Hewitt, P. L . , Flett, G. L. , & Turnbull-Donovan, W. (1992). Perfectionism and suicide potential. British Journal of Clinical Psychology, 31, 181-190. Hewitt, P. L . , Flett, G. L. , Turnbull-Donovan, W., & Mikail, S. F. (1991). The Multidimensional Perfectionism Scale: Reliability, validity, and psychometric properties in a psychiatric sample. Psychological Assessment. 3, 464-468. Perfectionistic Self-presentation / 113 Hewitt, P. L . , Flett, G. L. , & Weber, C. (1994). Dimensions of perfectionism and suicide ideation. Cognitive Therapy and Research. 18, 439-460. Hill, M . G., Weary, G., & Williams, J. (1986). Depression: A self-presentation formulation. In R. F. Baumeister (Ed.), Public self and private self (pp. 213-235). New York: Springer-Verlag. Hirt, E. R., Deppe, R. K., & Gordon, L. J. (1991). Self-reported versus behavioral self-handicapping: Empirical evidence for a theoretical distinction. Journal of Personality and Social Psychology, 61, 981-991. Hobden, K., & Pliner, P. (1995). Self-handicapping and dimensions of perfectionism: Self-presentation vs. self-protection. Journal of Research in Personality, 29, 461-474. Holtgraves, T., & Srull, T. K. (1989). The effects of positive self-descriptions on impressions: General principles and individual differences. Personality and Social Psychology Bulletin, 15, 452-462. Holtzworth-Munroe, A., & Jacobson, N. (1988). Toward a methodology for coding spontaneous causal attributions: Preliminary results with married couples. Journal of Social and Clinical Psychology, 7, 101-112. Homey, K. (1937). The neurotic personality of our time. New York: W. W. Norton. Horowitz, L. M . , Rosenberg, S. E. , & Bartholomew, K. (1993). Interpersonal problems, attachment styles, and outcome in brief dynamic psychotherapy. Journal of Consulting and Clinical Psychology. 61, 549-560. Horowitz, M . J., Stinson, C , Curtis, D., Ewert, M . , Redington, D., Singer, J., Bucci, W., Mergenthaler, E. , Milbrath, C , & Hartley, D. (1993). Topics and signs: Defensive control of emotional expression. Journal of Consulting and Clinical Psychology. 61, 421-430. Horvath, A. O., & Symonds, B. D. (1991). Relation between working alliance and outcome in psychotherapy: A meta-analysis. Journal of Counseling Psychology. 38, 139-149. Hunter, J. E. (1997). Needed: A ban on the significance test. American Psychological Society. 8, 3-7. Ichiyama, M . A., Colbert, D., Laramore, H. , Heim, M . , Carone, K., & Schmidt, J. (1993). Self-concealment and correlates of adjustment in college students. Journal of College Student Psychotherapy. 7, 55-68. Jackson, D. N. (1970). A sequential system for personality scale development. In C. D. Spielberger (Ed.), Current topics in clinical and community psychology (pp. 61-96). New York: Academic Press. Johnson, M . E. (1989). Influences of gender and sex role orientation on help-seeking attitudes. Journal of Psychology. 122. 237-241. Jones, E. E. , & Pittman, T. S. (1982). Toward a general theory of strategic self-presentation. In J. Suls (Ed.), Psychological perspectives on the self (vol. 1, pp. 231-262). Hillsdale, NJ: Lawrence Erlbaum. Jones, E. E. , & Rhodewalt, F. (1982). The Self-handicapping Scale. Unpublished. (See Strube). Perfectionistic Self-presentation / 114 Jones, W. H., Briggs, S. R., & Smith, T. G. (1986). Shyness: Conceptualization and measurement. Journal of Personality and Social Psychology, 51, 629-639. Kaiser, J., Hinton, J. W., Krohne, H. W., Stewart, R., & Burton, R. (1995). Coping dispositions and physiological recovery from a speech preparation stressor. Personality and Individual Differences, 19, 1-11. Kanfer, R., & Zeiss, A. M. (1983). Depression, interpersonal standard setting, and judgments of self-efficacy. Journal of Abnormal Psychology. 92, 319-329. Katz, M . M. , Koslow, S. H., Berman, N. , Secunda, S., Maas, J. W., Casper, R., Kocsis, J., & Stokes, P. (1984). A multi-vantaged approach to measurement of behavioral and affect states for clinical and psychobiological research. Psychological Reports, 55. 619-671. Kelly, A. E., & Achter, J. A. (1995). Self-concealment and attitudes toward counseling in university students. Journal of Counseling Psychology. 42, 40-46. Kessler, R. C , & Essex, M. (1982). Marital status and depression: The role of coping resources. Social Forces. 61, 484-507. Kirmayer, L. J., Robbins, J. M. , & Paris, J. (1995). Somatoform disorders: Personality and social matrix of somatic distress. Journal of Consulting and Clinical Psychology.. 103. 125-136. Kline, P. (1993). Personality: The psychometric view. New York: Routledge. Kolditz, T. A., & Arkin, R. M. (1982). An impression management interpretation of the self-handicapping strategy. Journal of Personality and Social Psychology. 43, 492-502. Kotokovic, A. M. , & Tracey, T. J. (1990). Working alliance in the early phase of counseling. Journal of Counseling Psychology, 37, 16-21. Krupnick, J. L., Elkin, I., Collins, J., Simmens, S., Sotsky, S. M. , Pilkonis, P. A., & Watkins, J. T. (1994). Therapeutic alliance and clinical outcome in the NIMH Treatment of Depression Collaborative Research Program: Preliminary findings. Psychotherapy, 31, 28-35. Lane, J. D., & Adcock, R. A. (1990). Caffeine effects on cardiovascular and neuroendocrine responses to acute psychosocial stress and their relation to level of habitual caffeine consumption. Psychosomatic Medicine, 52, 320-336. Lane, R. D., Merikangas, K. R., Schwartz, G. E., Huang, S. S., & Prusoff, B. A. (1990). Inverse relationship between defensiveness and lifetime prevalence of psychiatric disorder. American Journal of Psychiatry, 147, 573-578. Larson, D. G., & Chastain', R. L. (1990). Self-concealment: Conceptualization, measurement, and health implications. Journal of Social and Clinical Psychology, 9, 439-455. Leary, M. R. (1983). Understanding social anxiety. Beverly Hills: Sage. Leary, M. R., & Kowalski, R. M. (1990). Impression management: A literature review and two-component model. Psychological Bulletin, 107. 34-47. Perfectionistic Self-presentation / 115 Leary, M . R., & Kowalski, R. M . (1993). The Interaction Anxiousness Scale: Construct and criterion-related validity. Journal of Personality Assessment, 61, 136-146. Leary, M . R., & Sheppard, J. A. (1986). Behavioral self-handicapping versus self-reported self-handicapping: A conceptual note. Journal of Personality and Social Psychology, 51, Lennox, R. D., & Wolfe, R. N. (1984). Revision of the Self-monitoring Scale. Journal of Personality and Social Psychology, 46, 1349-1364. Lewinsohn, P. M . , Mischel, W., Chaplin, W., & Barton, R. (1980). Social competence and depression: The role of illusory self-perceptions. Journal of Abnormal Psychology. 89, 203-212. Linden, W. (1991). What do arithmetic stress tests measure? Protocol variations and cardiovascular responses. Psychophysiology. 28. 91-102. McCullough, L . (1988). Psychotherapy Interaction Coding System Manual (PIC System). Social and Behavioral Sciences Documents. 18, 50 (Ms. No. 2881). McCullough, L. , Winston, A., Farber, B. A., Porter, F., Pollack, J , Vingiano, W., Laikin, M . , & Trujillo, M . (1991). The relationship of patient-therapist interaction to outcome in brief psychotherapy. Psychotherapy, 28, 525-532. McHugo, G. J., & Lanzetta, J. T. (1983). Methodological decisions in social psychophysiology. In J. T. Cacioppo, & R. E. Petty (Ed.), Social psychophysiology: A sourcebook (pp. 630-665). New York: Guilford. McLaughlin, M . L . , Cody, M . J., & O'Hair, H . D. (1983). The management of failure events: Some contextual determinants of accounting behavior. Human Communication Research. 9, 208-224. McMullin, R. E. (1986). Handbook of cognitive therapy techniques. New York: W. W. Norton. Mead, G. H. (1934). Mind, self, and society. Chicago: University of Chicago Press. Mehlman, R. C , & Snyder, C. R. (1985). Excuse theory: A test of the self-protective role of attributions. Journal of Personality and Social Psychology. 49. 994-1001. Meleshko, K. G. A., & Alden, L. E. (1993). Anxiety and self-disclosure: Toward a motivational model. Journal of Personality and Social Psychology. 64, 1000-1009. Mendolia, M . , & Kleck, R. E. (1993). Effects of talking about a stressful event on arousal: Does what we talk about make a difference? Journal of Personality and Social Psychology, 64, 283-292. Miller, G. A. (1996). How we think about cognition, emotion, and biology in psychopathology. Psychophysiology, 33, 615-628. Miller, R. S., & Schlenker, B. R. (1985). Egotism in group members: Public and private attributions of responsibility for group performance. Social Psychology Quarterly, 48, 85-89. Missildine, W. H. (1963). Your inner child of the past. New York: Simon and Schuster. Perfectionistic Self-presentation / 116 Moras, K., & Strupp, H. H. (1982). Pretherapy interpersonal relations, patient's alliance, and outcome in brief therapy. Archives of General Psychiatry. 39. 405-409. Mothersill, K. J., Dobson, K. S., & Neufeld, W. J. (1986). The interactional model of anxiety: An evaluation of the differential hypothesis. Journal of Personality and Social Psychology. 51. 640-648. Muran, J. C , Segal, Z. V., Samstag, L., & Crawford, C. E. (1994). Patient pretreatment interpersonal problems and therapeutic alliance in short-term cognitive therapy. Journal of Consulting and Clinical Psychology. 62, 185-190. Newton, T. L., & Contrada, R. J. (1992). Repressive coping and verbal-autonomic response dissociation: The influence of social context. Journal of Personality and Social Psychology, 62. 159-167. Nielsen, A. D , Hewitt, P. L., Han, H., Habke, A. M. , Cockell, S. J., Stager, G., & Flett, G. L. (1997). Perfectionistic self-presentation and attitudes toward professional help-seeking. Poster presented at the Canadian Psychological Association Annual Convention Toronto, Ont: Neurodyne Medical Corporation (1994). Davicon MED AC System/3: Instrumentation User's Guide Olson, K. R., & Johnson, D. C. (1991). Individual differences in self-presentation styles. Journal of Social Psychology. 131. 495-509. Organista, P. B., & Miranda, J. (1991). Psychosomatic symptoms in medical outpatients: An investigation of self-handicapping theory. Health Psychology. 10, 427-431. Pacht, A. R. (1984). Reflections on perfection. American Psychologist, 29, 386-390. Parrott, C , & Joyce, C. (1993). Stress and arousal rhythms in cigarette smokers, deprived smokers, and non-smokers. Human Psvchopharmacology, 8, 21-28. Paulhus, D. (1985). Self-deception and impression management in test responses. In A. Angleitner, & J. S. Wiggins (Ed.), Personality assessment via questionnaires (pp. 143-165). Berlin: Springer-Verlag. Paulhus, D. L. (1984). Two-component models of socially desirable responding. Journal of Personality and Social Psychology. 46, 598-609. Paulhus, D. L. (1986). Self-deception and impression management in test responses. In A. A, & J. S. Wiggins (Ed), Personality assessment via questionnaire (pp. 143-165). New York: Springer-Verlag. Paulhus, D. L. (1994). Balanced Inventory of Desirable Responding: Reference Manual for BIDR Version 6. Vancouver, B.C. Paulhus, D. L., Bruce, M. N , & Trapnell, P. D. (1995). Effects of self-presentation strategies on personality profiles and their structure. Personality and Social Psychology Bulletin. 21, 100-108. Pedhazur, E.J. (1982). Multiple Regression in Behavioral Research: Explanation and Prediction. New York: Holt, Rinehard and Winston. Perfectionistic Self-presentation / 117 Pennebaker, J. W., Hughes, C. F., & O'Heeron, R. C. (1987). The psychophysiology of confession: Linking inhibitory and psychosomatic processes. Journal of Personality and Social Psychology, 52, 781-793. Peterson, C , Bettes, B. A., & Seligman, M . E. P. (1985). Depressive symptoms and unprompted causal attributions: Content analysis. Behavior Research and Therapy, 23, 379-382. Peterson, C , Schwartz, S. M . , & Seligman, M . E. P. (1981). Self-blame and depressive symptoms. Journal of Personality and Social Psychology, 41, 253-259. Piper, W. E. , Azim, H. F., Joyce, A. S., McCallum, M . , Nixon, G. W., & Segal, P. S. (1991). Quality of object relations versus interpersonal functioning as predictors of therapeutic alliance and psychotherapy outcome. Journal of Nervous and Mental Disease, 179, 432-438. Polefrone, J. M . , & Manuck, S. B. (1988). Effects of menstrual phase and parental history of hypertension on cardiovascular response to cognitive challenge. Psychosomatic Medicine, 50, 23-36. Powers, T. A., & Zuroff, D. C. (1988). Interpersonal consequences of overt self-criticism: A comparison with neutral and self-enhancing presentations of self. Journal of Personality and Social Psychology, 54, 1054-1062. Pyszczynski, T., & Greenberg, J. (1987). Self-regulatory perseveration and the depressive self-focusing style: A self-awareness theory of reactive depression. Psychological Bulletin, 102, 122-138. Reidbord, S. P., & Redington, D. J. (1992). Psychophysiological processes during insight-oriented therapy: Further investigations into nonlinear psychodynamics. Journal of Nervous and Mental Disease, 180, 649-657. Rhodewalt, F. (1990). Self-handicappers: Individual differences in the preference for anticipatory, self-protective acts. In R. L. Higgins, C. R. Snyder, & S. Berglas (Ed.), Self-handicapping: The paradox that isn't (pp. 69-106). New York: Plenum. Rhodewalt, F., & Davison, J. (1986). Self-handicapping and subsequent performance: Role of outcome valence and attributional certainty. Basic and Applied Social Psychology, 7, 307-322. Rickwood, D. J., & Braithwaite, V. A. (1994). Social-psychological factors affecting help-seeking for emotional problems. Social Science in Medicine, 39, 563-572. Robinson, M . D., Johnson, J. T., & Shields, S. A. (1995). On the advantages of modesty: The benefits of a balanced self-presentation. Communication Research, 22, 575-591. Roth, D. L . , Harris, R. N , & Snyder, C. R. (1988). An individual differences measure of attributive and repudiative tactics of favorable self-presentation. Journal of Social and Clinical Psychology, 6, 159-170. Roth, D. L . , Snyder, C. R., & Pace, L. M . (1986). Dimensions of favorable self-presentation. Journal of Personality and Social Psychology, 51, 867-874. Rothstein, A. (1991). On some relationships of fantasies of perfection to the calamities of childhood. International Journal of Psychoanalysis, 72, 313-323. Perfectionistic Self-presentation / 118 Sacco, W. P., & Hokanson, J. E. (1982). Depression and self-reinforcement in a public and private setting. Journal of Personality and Social Psychology. 43. 377-385. Scandell, D. J., & Wlazelek, B. G. (1996). Self-presentation strategies on the Neo-5 Factor Inventory -Implications for detecting faking. Psychological Reports. 79, 1115-1121. Schlenker, B. R. (1980). Impression management: The self-concept, social identity, and interpersonal relations. Monterey, CA: Brooks/Cole. Schlenker, B. R. (1986). Self-identification: Toward an integration of the private and public self. In R. F. Baumeister (Ed.), Public self and private self (pp. 21-62). New York: Springer-Verlag. Schlenker, B. R., & Leary, M . R. (1982). Social anxiety and self-presentation: A conceptualization and model. Psychological Bulletin. 92, 641-669. Schlenker, B. R., & Weigold, M . F. (1992). Interpersonal processes involving impression regulation and management. Annual Review of Psychology, 43, 133-68. Schlenker, B. R., Weigold, M . F., & Hallam, J. R. (1990). Self-serving attributions in social context: Effects of self-esteem and social pressure. Journal of Personality and Social Psychology, 58, 855-863. Schonbach, P. (1980). A category system for account phases. European Journal of Social Psychology, L0, 195-200. Sechrest, L. (1984). Reliability and validity. In A. S. Bellack, & M . Hersen (Ed.), Research methods in clinical psychology (pp. 24-54). New York: Pergamon. Seligman, M . E. P., Abramson, L. Y. , Semmel, A., & von Baeyer, C. (1979). Depressive attributional style. Journal of Abnormal Psychology, 88, 242-247. Semin, G. R., & Manstead, A. S. R. (1983). The accountability of conduct: A social psychological analysis. London: Academic Press. Shay, J. J. (1996). Okay, I'm here but I'm not talking - Psychotherapy with the reluctant male. Psychotherapy, 33, 503-513. Shedler, J., Mayman, M . , & Manis, M . (1993). The illusion of mental health. American Psychologist, 48, 1117-1131. Sheppard, J. A., & Arkin, R. M . (1990). Shyness and self-presentation. In W. R. Crozier (Ed.), Shyness and embarrassment: Perspectives from social psychology (pp. 286-314). New York: Cambridge University Press. Siddle, D. A. T., Turpin, G. , Spinks, J. A., & Stephenson, D. (1980). Peripheral measures. In H. M . man Praag, M . H. Lader, O. J. Rafaelsen, & E. J. Sachar (Ed), Handbook of Biological Psychiatry Part II: Brain mechanisms and abnormal behavior - Psychophysiology (pp. 45-78). New York: Marcel Dekker. Singer, J. A. (1995). Seeing one's self: Locating narrative memory in a framework of personality. Journal of Personality. 63. 429-457. Perfectionistic Self-presentation / 119 Slade, P. D., Newton, T., Butler, N. M . , & Murphy, P. (1991). An experimental analysis of perfectionism and dissatisfaction. British Journal of Clinical Psychology, 30, 169-176. Smith, C. A., & Ellsworth, P. C. (1985). Patterns of cognitive appraisal in emotion. Journal of Personality and Social Psychology. 48, 825-838. Smith, T. W., Snyder, C. R., & Handelsman, M . M . (1982). On the self-serving function of an academic wooden leg: Test anxiety as a self-handicapping strategy. Journal of Personality and Social Psychology, 42,314-321. Snyder, C. R. (1985). The excuse: An amazing grace? In B. R. Schlenker (Ed), The self and social life (pp. 235-260). New York: McGraw-Hill. Snyder, C. R., & Higgins, R. L. (1988). Excuses: Their effective role in the negotiation of reality. Psychological Bulletin, 104. 23-35. Snyder, C. R., & Smith, T. W. (1982). Symptoms as self-handicapping strategies: The virtues of old wine in a new wine bottle. In G. Weary, & H. L. Mirels (Ed), Integrations of Clinical and Social Psychology (pp. 104-127). New York: Oxford University Press. Snyder, M . (1974). Self-monitoring of expressive behavior. Journal of Personality and Social Psychology. 30, 526-537. Snyder, M , & Gangestad, S. (1986). On the nature of self-monitoring: Matters of assessment, matters of validity. Journal of Personality and Social Psychology. 51. 125-139. Snydersmith, M . A., & Cacioppo, J. T. (1992). Parsing complex social factors to determine component effects: 1. Autonomic activity and reactivity as a function of human association. Journal of Social and Clinical Psychology. U, 263-278. Sorotzkin, B. (1985). The quest for perfection: Avoiding guilt or avoiding shame? Psychotherapy, 22, 564-571. Steer, R. A., Beck, A. T., & Brown, G. (1989). Sex differences on the Revised Beck Depression Inventory for patients with affective disorders. Journal of Personality Assessment. 53, 693-702. Stemmler, G. (1992). Differential Psychophysiology: Persons in Situations. Berlin: Springer-Verlag. Stemmler, G , & Meinhardt, E. (1990). Personality, situation, and physiological arousability. Personality and Individual Differences, 11. 293-308. Strong, S. R. (1982). Emerging integrations of clinical and social psychology: A clinician's perspective, (pp. 181-213). Strong, S. R., & Claiborn, C. D. (1982). Change through interaction: Social psychological processes of counseling and psychotherapy. New York: John Wiley & Sons. Strube, M . J. (1986). An analysis of the Self-Handicapping Scale. Basic and Applied Social Psychology, 7, 211-224. Perfectionistic Self-presentation/ 120 Sweeney, P. D., Anderson, K., & Bailey, S. (1986). Attributional style in depression: A meta-analytic review. Journal of Personality and Social Psychology, 50, 974-991. Tabachnick, B. G., & Fidell, L. S. (1989). Using Multivariate Statistics(2 ed.). New York: Harper Collins. Taurke, E. A., Flegenheimer, W., McCullough, L., Winston, A., Pollack, J., & Trujillo, M. (1990). Change in patient affect/defense ratio from early to late sessions in brief psychotherapy. Journal of Clinical Psychology. 46, 657-320. Tedeschi, J. T., & Norman, N. (1985). Social power, self-presentation, and the self. In B. R. Schlenker (Ed.), The self and social life (pp. 293-322). New York: McGraw-Hill. Terry-Short, L. A., Owens, R. G., Slade, P. D., & Dewey, M. E. (1995). Positive and negative perfectionism. Personality and Individual Differences, 18, 663-668. Tetlock, P. E. (1985). Toward an intuitive politician model of attribution processes. In B. R. Schlenker (Ed.), The self and social life (pp. 203-234). New York: McGraw-Hill. Tetlock, P. E., & Manstead, A. S. R. (1985). Impression management versus intrapsychic explanations in social psychology: A useful dichotomy? Psychological Review, 92, 59-77. Teyber, E. (1988). Interpersonal process in psychotherapy: A guide for clinical training. Chicago: Dorsey Press. Thome, A. (1995). Juxtaposed scripts, traits, and the dynamics of personality. Journal of Personality, 63, 593-. Tice, D. M. (1991). Esteem protection or enhancement? Self-handicapping motives and attributions differ by trait self-esteem. Journal of Personality and Social Psychology, 60, 711 -725. Tice, D. M. , & Baumeister, R. F. (1990). Self-esteem, self-handicapping, and self-presentation: The strategy of inadequate practice. Journal of Personality and Social Psychology, 58, 443-464. Tomaka, J., Blascovich, J., Kelsey, R. M. , & Leitten, C. L. (1993). Subjective, physiological, and behavioral effects of threat and challenge appraisal. Journal of Personality and Social Psychology, 65, 248-260. Tucker, J. A., Vulchinich, R. D., & Sobell, M. B. (1981). Alcohol consumption as a self-handicapping strategy. Journal of Abnormal Psychology, 90, 220-230. Vaillant, G. E. (1971). Theoretical hierarchy of adaptive ego mechanisms. Archives of General Psychiatry. 33. 111-115. Van Dam-Baggen, R., Van Heck, G., & Kraaimaat, F. (1992). Consistency of social anxiety in psychiatric patients: Properties of persons, situations, response classes, and types of data. Anxiety. Stress, and Coping. 5, 285-300. Venables, P. H., & Christie, M. J. (1980). Electrodermal activity. In I. Martin, & P. H. Venables (Ed), Techniques in psychophysiology (pp. 3-67). New York: John Wiley & Sons. Perfectionistic Self-presentation / 121 Veroff, J. B., Kulka, R. A., & Douvan, E. (1981). Mental health in America: Patterns of help-seeking 1957-1976. New York: Basic Books. 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. Wallbott, H. G. , & Scherer, K. R. (1991). Stress specificities: Differential effects of coping style, gender, and type of stressor on autonomic arousal, facial expression, and subjective feeling. Journal of Personality and Social Psychology, 61, 147-156. Walrath, L. C , & Stern, J. A. (1980). General considerations. In H. M . Van Praag, M . H. Lader, O. J. Rafaelsen, & E. J. Sachar (Ed), Handbook of biological psychiatry: Brain mechanisms and abnormal behavior - psychophysiology (vol. 1, pp. 1-44). New York: Marcel Dekker. Watson, D., Clark, L. A., Mclntyre, C. W., & Hamaker, S. (1992). Affect, personality, and social activity. Journal of Personality and Social Psychology, 63, 1011-1025. Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures of positive and negative affect: The PANAS Scales. Journal of Personality and Social Psychology, 54, 1063-1070. Weary, G , & Arkin, R. M . (1981). Attributional self-presentation. In J. H. Harvey, W. Ickes, & R. F. Kidd (Ed.), New directions in attribution research Vol. 3, (pp. 223-246). Hillsdale, N.J.: Lawrence Erlbaum. Weary, G , & Williams, J. P. (1990). Depressive self-presentation: Beyond self-handicapping. Journal of Personality and Social Psychology, 58, 892-898. Weber, D. J., & Vangelisti, A. L. (1991). "Because I love you..." The tactical use of attributional expressions in conversation. Human Communication Research, 17, 606-624. Weinberger, D. A., & Davidson, M . N. (1994). Styles of inhibiting emotional expression: Distinguishing repressive coping from impression management. Journal of Personality, 62, 587-613. Weinberger, D. A., Schwartz, G. E. , & Davidson, R. J. (1979). Low-anxious, high-anxious, and repressive coping styles: Psychometric patterns and behavioral and physiological responses to stress. Journal of Abnormal Psychology. 88, 369-380. Weiner, B. (1992). Excuses in everyday interaction. In M . L. McLaughlin, M . J. Cody, & S. J. Read (Ed.), Explaining one's self to others: Reason-giving in a social context (pp. 131-146). Hillsdale, NJ: Lawrence Erlbaum. Weiner, B., Figueroa-Munoz, A., & Kakihara, C. (1991). The goals of excuses and communication strategies related to causal perceptions. Personality and Social Psychology Bulletin, 17, 4-13. Westen, D. (1995). A clinical-empirical model of personality: Life after the Mischelian ice age and the . NEO-lithic era. Journal of Personality. 63, 495-524. Whitehead, G. I., & Smith, S. H. (1986). Competence and excuse making as self-presentational strategies. In R. F. Baumeister (Ed), Public self and private self (pp. 161-178). New York: Springer-Verlag. Perfectionistic Self-presentation / 122 Whittal, M . , & Dobson, K. S. (1991). An investigation of the temporal relationship between anxiety and depression as a consequence of cognitive vulnerability to interpersonal evaluation. Canadian Journal of Behavioral Science. 23, 391-398. Wolfe, R. N. , Lennox, R. D., & Cutler, B. L. (1986). Getting along and getting ahead: Empirical support for a theory of protective and acquisitive self-presentation. Journal of Personality and Social Psychology. 50, 356-361. Youngren, M . A., & Lewinsohn, P. M . (1980). The functional relation between depression and problematic interpersonal behavior. Journal of Abnormal Psychology. 89. 333-341. Zahn, T. P. (1986). Psychophysiological approaches to psychopathology. In M . G. H. Coles, E. Donchin, & S. W. Porges (Ed.), Psychophysiology: Systems, processes and applications (pp. 508-610). New York: Guilford Press. Zeidner, M . (1994). Personal and contextual determinants of coping and anxiety in an evaluative situation: A prospective study. Personality and Individual Differences, 16, 899-918. Zook, A., & Sipps, G. J. (1985). Cross-validation of a short form of the Marlowe-Crowne Social Desirability Scale. Journal of Clinical Psychology. 41. 236-238. Perfectionistic Self-presentation / 123 Appendix A Study Measures P E R F E C T I O N I S T I C S E L F - P R E S E N T A T I O N S C A L E Listed below are a group of statements. Please rate your agreement with each of the statements using the following scale. If you strongly agree, circle 7; if you strongly disagree, circle 1; if you feel somewhere in between, circle any one of the numbers between 1 and 7. If you feel neutral or undecided the midpoint is 4. 1 2 3 4 5 6 7 Disagree Neutral Agree Strongly Strongly 1. It is okay to show others that I am not perfect 1 2 3 4 5 6 7 2. I judge myself based on the mistakes I make in front of other people 1 2 3 4 5 6 7 3. I will do almost anything to cover up a mistake 1 2 3 4 5 6 7 4. Errors are much worse if they are made in public rather than in private 1 2 3 4 5 6 7 5. I try always to present a picture of perfection 1 2 3 4 5 6 7 6. I would be awful if I made a fool of myself in front of others 1 2 3 4 5 6 7 7. If I seem perfect, others will see me more positively... 1 2 3 4 5 6 7 8. I brood over mistakes that I have made in front of others .; 1 2 3 4 5 6 7 9. I never let others know how hard I work on things 1 2 3 4 5 6 7 10. I would like to appear more competent than I really am 1 2 3 4 5 6 7 11. It doesn't matter if there is a flaw in my looks 1 2 3 4 5 6 7 12. I do not want people to see me do something unless I am very good at it 1 2 3 4 5 6 7 13. I should always keep my problems to myself 1 2 3 4 5 6 7 14. I should solve my own problems rather than admit them to others 1 2 3 4 5 6 7 15. I must appear to be in control of my actions at all times 1 2 3 4 5 6 7 16. It is okay to admit mistakes to others 1 2 3 4 5 6 7 17. It is important to act perfectly in social situations 1 2 3 4 5 6 7 18. I don't really care about being perfectly groomed 1 2 3 4 5 6 7 19. Admitting failure to others is the worst possible thing 1 2 3 4 5 6 7 20. I hate to make errors in public 1 2 3 4 5 6 7 21. I try to keep my faults to myself 1 2 3 4 5 6 7 22. I do not care about making mistakes in public 1 2 3 4 5 6 7 23. I need to be seen as perfectly capable in everything I do 1 2 3 4 5 6 7 24. Failing at something is awful if other people know about it 1 2 3 4 5 6 7 25. It is very important that I always appear to be "on top of things" 1 2 3 4 5 6 7 26. I must always appear to be perfect 1 2 3 4 5 6 7 27. I strive to look perfect to others 1 2 3 4 5 6 7 Perfectionistic Self-presentation / 124 M U L T I - D I M E N S I O N A L P E R F E C T I O N I S M S C A L E (Short) Listed below are a number of statements concerning personal characteristics and traits. Read each item and decide whether you agree or disagree and to what extent. If you strongly agree, circle 7; if you strongly disagree, circle 1; if you feel somewhere in between, circle any one of the numbers between 1 and 7. If you feel neutral or undecided the midpoint is 4. 1 2 3 4 5 6 7 Disagree Neutral Agree Strongly Strongly 1. One of my goals is to be perfect in everything I do 1 2 3 4 5 6 7 2. Everything that others do must be of top-notch quality r 2 3 4 5 6 7 3. The better I do, the better I am expected to do l 2 3 4 5 6 7 4. I strive to be as perfect as I can be l 2 3 4 5 6 7 5. It is very important that I am perfect in everything I attempt l 2 3 4 5 6 7 6. I have high expectations for the people who are important to me l 2 3 4 5 6 7 7. I demand nothing less than perfection of myself. l 2 3 4 5 6 7 8. I can't be bothered with people who won't strive to better themselves l 2 3 4 5 6 7 9. Success means that I must work even harder to please others l 2 3 4 5 6 7 10. If I ask someone to do something, I expect it to be done flawlessly... l 2 3 4 5 6 7 11. I cannot stand to see people close to me make mistakes l 2 3 4 5 6 7 12. I must work to my full potential at all times l 2 3 4 5 6 7 13. My family expects me to be perfect l 2 3 4 5 6 7 14. People expect nothing less than perfection from me l 2 3 4 5 6 7 15. People expect more from me than I am capable of giving l 2 3 4 5 6 7 Copyright (©) Paul L . Hewitt, Ph.D., & Gordon L. Flett, Ph.D., 1988 I N T E R A C T I O N ANXIOUSNESS S C A L E Please indicate the extent to which each statement is characteristic or true of you, on a scale from 1 to 5: 1 2 3 4 5 not at all slightly moderately very extremely characteristic of you characteristic of you 1. I often feel nervous even in casual get-togethers 1 2 3 4 5 2. I usually feel uncomfortable when I am in a group of people I don't know 1 2 3 4 5 3. I am usually at ease when speaking to a member of the opposite sex 1 2 3 4 5 4. I get nervous when I must talk to a teacher or boss 1 2 3 4 5 5. Parties often make me feel anxious and uncomfortable 1 2 3 4 5 6. I am probably less shy in social interactions than most people 1 2 3 4 5 7. I sometimes feel tense when talking to people of my own sex if I don't 1 2 3 4 5 8. I would be nervous if I were being interviewed for a job 1 2 3 4 5 9. I wish I had more confidence in social interactions 1 2 3 4 5 10. 1 2 3 4 5 11. 1 2 3 4 5 Perfectionistic Self-presentation / 125 12. I often feel nervous when talking to an attractive member of the opposite sex 1 2 3 4 5 13. I often feel nervous when calling someone I don't know very well on the telephone 1 2 3 4 5 14. I usually get nervous when I speak to someone in a position of authority 1 2 3 4 5 15. I usually feel relaxed around other people, even people who are quite different from me 1 2 3 4 5 B A L A N C E D I N V E N T O R Y O F D E S I R A B L E RESPONDING Please rate each statement on a scale from 1 to 7, where 1 is "not true" and 7 is "very true." 1. I sometimes tell lies if I have to 1 2 3 4 5 6 7 2. I never cover up my mistakes 1 2 3 4 5 6 7 3. There have been occasions when I have taken advantage of someone 1 2 3 4 5 6 7 4. I never swear 1 2 3 4 5 6 7 5. I sometimes try to get even rather than forgive and forget 1 2 3 4 5 6 7 6. I always obey laws, even if I'm unlikely to get caught 1 2 3 4 5 6 7 7. I have said something bad about a friend behind his/her back 1 2 3 4 5 6 7 8. When I hear people talking privately, I avoid listening 1 2 3 4 5 6 7 9. I have received too much change from a salesperson without telling him or her 1 2 3 4 5 6 7 10. I always declare everything at customs 1 2 3 4 5 6 7 11. When I was young I sometimes stole things 1 2 3 4 5 6 7 12. I have never dropped litter on the street 1 2 3 4 5 6 7 13. I sometimes drive faster than the speed limit 1 2 3 4 5 6 7 14. I never read sexy books or magazines 1 2 3 4 5 6 7 15. I have done things that I don't tell other people about 1 2 3 4 5 6 7 16. I never take things that don't belong to me 1 2 3 4 5 6 7 17. I have taken sick-leave from work or school even though I wasn't really sick 1 2 3 4 5 6 7 18. I have never damaged a library book or store merchandise without reporting it 1 2 3 4 5 6 7 19. I have some pretty awful habits 1 2 3 4 5 6 7 20. I don't gossip about other people's business 1 2 3 4 5 6 7 B R S C - A N T I C I P A T E D P E R F O R M A N C E Everyone has different expectations of how they will act in certain situations. Please think about the interview that you are going to do in a short while. Below is a short list of descriptions of the way people might behave during a social interaction. For each one, please rate that characteristic from 1 "NOT A T A L L " to 7 " E X T R E M E L Y . " By extremely, we do not mean an excessive, inappropriate, or exaggerated level, but rather one that fits perfectly with the situation. Please think about how you will be in regards to that characteristic in the upcoming interview. 1. 2. Friendly 1 2 3 4 5 6 7 3. Assertive 1 2 3 4 5 6 7 4. Warm 1 2 3 4 5 6 7 Perfectionistic Self-presentation / 126 5. Communicates clearly 1 2 3 4 5 6 7 6. Socially skillful 1 2 3 4 5 6 7 7. Humorous 1 2 3 4 5 6 7 8. Reasonable 1 2 3 4 5 6 7 9. Confident (self-assured) 1 2 3 4 5 6 7 10. Attractive 1 2 3 4 5 6 7 11. Speaks fluently 1 2 3 4 5 6 7 12. Open 1 2 3 4 5 6 7 13. Relaxed 1 2 3 4 5 6 7 S E L F HANDICAPPING S C A L E Read each of the following statements carefully and indicate how characteristic it is of you by circling the appropriate number. 1 2 3 4 5 6 disagree disagree disagree agree agree agree very much pretty much a little a little pretty much very much 1. I tend to make excuses when I do something wrong 1 2 3 4 5 6 2. I tend to put things off until the last moment 1 2 3 4 5 6 3. I suppose I feel "under the weather" more often than most people 1 2 3 4 5 6 4. I always try to do my best, no matter what 1 2 3 4 5 6 5. I am easily distracted by noises or my own daydreaming when I try to read... 1 2 3 4 5 6 6. I try not to get too intensely involved in competitive activities so it won't hurt too much if I lose or do poorly 1 2 3 4 5 6 7. I would do a lot better if I tried harder 1 2 3 4 5 6 8. I sometimes enjoy being mildly ill for a day or two 1 2 3 4 5 6 9. I tend to rationalize when I don't live up to others'expectations 1 2 3 4 5 6 10. I overindulge in food and drink more often than I should 1 2 3 4 5 6 S E L F - H A N D I C A P P I N G Many things can change how well a person does in an interview (that is, how relaxed they are, how well they are able to get their ideas and feelings across, etc.). Is there anything that we should know about that you feel might really change how you do in the interview? No Yes If yes, please describe briefly what that is (or those are). 1. 2. 3. If yes, please estimate how much this/these might change how you do, using the following scale: -5 -4 -3 -2 -1 0 1 2 3 4 5 Much worse No real change Much better First reason: Second reason: Third reason: Perfectionistic Self-presentation / 127 B R S C - I N T E R V I E W E R E X P E C T A T I O N S Again, please think about the interview that you are going to do in a short while. Without looking back at your earlier answers, please think about how you think the interviewer expects you to be in regards to each characteristic in the upcoming interview. Not how they think you will do, but what level of performance they would like you to achieve. For each one, please rate that characteristic from 1 "NOT A T A L L " to 7 " E X T R E M E L Y . " By extremely, we do not mean an excessive, inappropriate, or exaggerated level, but rather one that fits perfectly with the situation. REPEAT BRSC PANAS - Pre Below is a list of different ways that people might feel. Please indicate the extent to which you feel this way right now, on a scale from 1 to 5 as outlined below: 1 2 3 4 5 not at all/very slightly a little moderately quite a bit very much so 1. Scared 1 2 3 4 2. Distressed 1 2 3 4 3. Jittery 1 2 3 4 4. Ashamed 1 2 3 4 5. Nervous 1 2 3 4 6. Upset 1 2 3 4 7. Guilty 1 2 3 4 8. Irritable 1 2 3 4 9. Afraid 1 2 3 4 10. Hostile 1 2 3 4 BDI P A N A S - POST Below is a list of different ways that people might feel. Please indicate the extent to which you felt each way during the interview you just completed (on average or overall), on a scale from 1 to 5 as outlined below: REPEAT PANAS B R S C - P E R C E I V E D P E R F O R M A N C E Everyone has different feelings about how they acted in certain situations. Please think about the interview that you just completed. Below is a short list of descriptions of the way people might behave during a social interaction. For each one, please rate how you were on that characteristic from 1 "NOT A T A L L " to 7 " E X T R E M E L Y . " By extremely, we do not mean an excessive, inappropriate, or exaggerated level, but rather one that fit perfectly with the situation. REPEAT BRSC Perfectionistic Self-presentation / 128 Everyone has some ideas about how they have been seen by others. How much do you think the interviewer liked you by the end of the interview: 1 2 3 4 5 6 7 Not at all Extremely How much do you think the interviewer would like most people who participate in the interview? 1 2 3 4 5 6 7 Not at all Extremely S E L F - H A N D I C A P P I N G - POST Many things can interfere with how well a person does in an interview (that is, how relaxed they are, how well they are able to get their ideas and feelings across, etc.). Is there anything that you feel changed how you did in the interview? No Yes REPEAT SELF-HANDICAPPING B R S C - I N T E R V I E W E R S A T I S F A C T I O N One more time! Please think about the interview that you have just completed. Without looking back at your earlier answers, please think about how satisfied do you think the interviewer is with how you were on each characteristic. For each one, please rate that characteristic from 1 "NOT A T A L L " to 7 " E X T R E M E L Y . " By extremely, we do not mean an excessive, inappropriate, or exaggerated level, but rather one that fits perfectly with the situation. REPEAT BRSC D I F F I C U L T SITUATIONS Q U E S T I O N N A I R E Regarding the first difficult situation you mentioned, 1. On a scale from 1 to 7, how difficult was this situation for you (that is, how distressed were 1 2 3 4 5 6 7 Only a little difficult Extremely difficult 2. On a scale from 1 to 7, how well do you think you managed it overall: 1 2 3 4 5 6 7 Very poorly Extremely well 3. On a scale from 1 to 7, to what extent was the situation caused by something about you or something about others or circumstances outside of your control: 1 2 3 4 5 6 7 Totally due to me Totally due to others/circumstances Regarding the second difficult situation you mentioned, REPEAT ABOVE QUESTIONS .5 Perfectionistic Self-presentation / 129 Appendix B Instructions to Coders Externalization - extent to which subject is attempting to externalize the problem or situation (put the cause outside of themselves or on something outside of their control). Conceding - recognition of own responsibility for actions, own fault for mistake. Excuses - attempts to get you to let them off the hook for their behavior. Examples: True for everyone Didn't know any better Biological factors Things that happened in the past I tried hard to prevent it Justifications - attempts to get you to understand WHY they behaved the way they did (and to see it as a good or rational thing to do). Examples: Fault of others Own rights Positive intentions No damage/not much damage Positive outcomes Self-promotion - attempts to get you to see how wonderful they are, with or without direct reference to their mistake. Examples: I really am a good/able person Could have been much worse I have come so far since then My positive qualities caused the problem Sympathy - direct attempts to get you to feel sorry for them in some way (not just that terrible things have been happening - rather, will feel like an exaggeration of their problems or that they are bringing up things that are not necessary to their story but that make you feel sympathetic). Examples: I have suffered greatly Someone else has wronged me Perfectionistic Self-presentation/ 130 Appendix C Interview Questions 1. First of all, I'd like you to tell me briefly about what brought you to the clinic (hospital). Why did you call for an appointment? (Allow 4 minutes - nonverbal encouragement up to 4 minutes). a) What do you think caused (causes) this/that? (What do you think has made you feel this way?) Anything else? b) Do you think anything you have done, or anything about you, contributes (contributed) to this problem? 2. Now I would like you to think about some of the times in your life that started because of something you did wrong, or times when a problem was made much worse because of the way you handled it. That is, times where you feel you made a mistake. Please try and think about the very worst one, the one that had the most serious consequences to you or to someone else. I'll give you a few minutes to think about the very worst one - tell me when you have thought about it. {Allow up to 4 minutes). Please describe the situation very briefly - not what caused the situation, but just a couple of sentences about what happened. (How did you cope with that?) (Approx 2 minutes). a) What do you think was your contribution (if any) to how that situation began or developed? b) Why do you think that was a mistake/problem? (or: In what ways was it made worse because of what you did to deal with it?) c) What was your contribution - good or bad - to how that situation ended? 3. Now I would like you to think about a second situation that you feel started because of a mistake you made or was made much worse because of how you handled it. It may be less serious than the last one, or it may be one you just thought of that is actually more serious than the last one. Again, I'll give you some time to think of it. Please describe the situation very briefly - not what caused the situation, but just a couple of sentences about what happened. (How did you cope with that?) a) What do you think was your contribution (if any) to how that situation began or developed? b) Why do you think that was a mistake/problem? (Or: In what ways was it made worse because of what you did to deal with it?) c) What was your contribution - good or bad - to how that situation ended? Relax Thank you. Those are difficult questions to answer. Marie is going to come in and ask you a few more questions, but first I would like you to just take a brief minute to relax. Take a couple of deep breaths, and try to let your muscles go loose. 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items