Running head: SELF-FOCUSED ATTENTION AND EVALUATION

This study investigated the effects of self-focused attention and evaluation on the perceived performance of individuals high in social anxiety. High and low socially anxious participants (n = 92) were asked to give four short speeches under different conditions of increased self-focused attention (in front of a mirror) or decreased self-focused attention (in front of a poster), and/or under conditions of increased evaluation (by the examiner) or decreased evaluation (no examiner evaluation). Results indicate that high socially anxious participants rated their performance to be worse in all conditions compared to low socially anxious participants. Additionally, evaluation appears to have a slightly negative effect on high socially anxious performance ratings, but has a slightly positive effect on low socially anxious performance ratings. No evidence was found for main effects of self-focused attention or evaluation. Results are discussed in light of possible threat/challenge effects in individuals high and low in social anxiety. Threat or Challenge 3 Threat or Challenge? Self-Focused Attention and Evaluation on Perceived Performance in Social Anxiety Interest in social anxiety disorder has grown tremendously over the past 20 years due in large part to it being one of the most common mental disorders in the population (Kessler, Berglund, Demler, Jin, Merikangas, & Walters, 2005). Social anxiety is an experience of fear or apprehension about a social situation and the prospect of being evaluated by others. People with social anxiety generally expect to be evaluated negatively and this tends to further heighten their sense of fear (Clark & Wells, 1995). In some cases an individual‟s social anxiety may become so severe that he or she may also avoid certain social situations, be more likely to abuse alcohol or drugs, and be at risk for suicide (Bolton, Cox, Clara, & Sareen, 2006; Stopa & Clark, 2000). In these cases, it can be diagnosed as social anxiety disorder, also known as social phobia. (DSM-IV; American Psychological Association, 2000). This study hopes to investigate two possible reasons why social anxiety may have such a profound effect on one‟s self-image in social contexts: self-focused attention and fears of negative evaluation. Self-Focused Attention in Social Anxiety Current cognitive models propose that when socially anxious individuals are in social situations they tend to shift their attention to an intense monitoring of themselves, known as selffocused attention (Brown & Stopa, 2007). During a social interaction a person‟s attentional focus may be on any number of things, including internal issues (i.e., on thoughts, sensations, and emotions relevant to the self) or on external issues (such as the other person, a task at hand, or the environment). A social interaction in which an individual is appropriately balancing internal and external attentional focus will be regarded as more successful and rewarding to the individual (Bögels & Lamers, 2002). Focusing entirely on oneself creates the feeling of being a Threat or Challenge 4 detached object in social interactions rather than partaking in the interaction (Duval & Wicklund, 1975). This is evident in Clark and Wells‟ (1995) theory that socially anxious individuals use self-focused attention to evaluate what other people think about them. Taking a step back from the conversation as an outside observer allows the individual to see themselves from a detached point of view, and since their interpretation of their own behavior is generally negative, they assume that the other individual is also rating them negatively. Albright and Malloy (1999) looked at self-observation of social behavior in unselected participants to investigate whether or not participants could accurately determine how other people in a social interaction judge them. Results showed that when self-focused attention was manipulated (using a video camera), participants in the high self-focused attention group had higher correspondence with observer judgments of their own behavior when compared to other group member‟s ratings. Therefore, we predict that when under conditions of high self-focused attention, individuals high in social anxiety will judge their social performance as negative, given their tendency to assume a critical outsider‟s view of their performance. Because individuals with social anxiety overestimate how negatively other people judge their performance, exclusive monitoring of the self should perpetuate their own negative selfperceptions. Mellings and Alden (2000) asked low and high socially anxious participants to have a ten-minute interaction with a confederate. Following the interaction, participants completed measures of self-focused attention and anxiety-related behaviors. Results showed that socially anxious individuals endorsed significantly more self-focused attention items as compared to low anxious participants. Mellings and Alden found a judgmental bias for socially anxious participants as well, with socially anxious participants showing a significantly more negative self-perception than participants low in anxiety. Duval and Wicklund‟s (1972) theory Threat or Challenge 5 of objective self-awareness theorizes that this apparent exclusive self-focused attention induces self-evaluative thinking where a person‟s behavior is compared to some standard. Therefore if one believes him or herself to be below this standard, self-focused attention will heighten negative self-evaluation in social situations. However, if one believes him or herself to be above the standard in social interactions (i.e., someone with high self-esteem) then increasing selffocused attention leads those individuals to rate themselves as having better performance (Brockner & Hulton, 1978). Self-focused attention not only causes individuals high in social anxiety to evaluate themselves more, it also causes individuals to be more aware of their arousal and fear levels and possible flaws in their behavior (Bögels, Rijsemus, & De Jong, 2002). This monitoring of one‟s fear levels may also perpetuate it. Thus, this past research appears to indicate that self-focused attention is leading to more monitoring of the negative interpretations that individuals high in social anxiety already have, subsequently leading to the maintenance and increase in anxiety levels. Additionally, increased self-focused attention has been found in a variety of different anxiety provoking tasks related to social anxiety. For example, Daly, Vangelisti, and Lawrence (1989) asked socially anxious and non-socially anxious participants to give a short speech in a classroom with an audience of three people. After completing the speeches, the participants were asked to relate memories of the speech orally, to complete a memory task assessing the participant‟s recall for environmental features, and to complete a questionnaire to assess how much the participant enjoyed the speech, how competent they felt, how nervous they felt, and how much they perceived the audience to have enjoyed their speech. Results showed that the socially anxious participants paid less attention to their environments and had more negative self-focused evaluations (relative to audience ratings) about their performance compared to the Threat or Challenge 6 non-socially anxious group. Bögels and Lamers (2002) found similar results when they asked high and low socially anxious participants to imagine themselves in different social events where they were a hero who was the center of attention. Results indicated that high socially anxious participants reported more self-focusing behavior and more social anxiety than the low socially anxious group. Additionally, research has shown successful results with cognitive-behavioral group therapy in socially phobic individuals where exposure therapy to the external environment results in decreased levels of self-focused attention in social situations (Hoffman, 2000; Woody, Chambless, & Glass, 1997; Wells & Papageorgiou, 1998). Therefore, self-focused attention can be found in socially anxious groups using an assortment of social interactions, both real and imagined. Given these findings for self-focused attention, our study investigates whether increasing self-focused attention through an experimental manipulation may heighten poor selfperceptions of performance in high socially anxious people. Fear of Negative Evaluation in Social Anxiety Other research indicates that the fear of negative evaluation also results in negative selfrelated performance ratings. According to the Clark and Wells (1995) cognitive model, individuals high in social anxiety have an interpretation bias for socially relevant information in which they interpret seemingly benign or ambiguous information as threatening. For example, if a non-socially anxious individual was giving a talk and saw someone in the crowd yawn, he or she may disregard it as the person being tired or even bored and may attempt to be more engaging in his or her talk. However, a socially anxious individual may see this same person and interpret it to mean that they are not successful in their talk and may become very anxious. Thus, while non-anxious participants may interpret evaluation as a challenge, socially anxious individuals may view it as threatening. This notion falls in line with Zajonc‟s (1965) social Threat or Challenge 7 facilitation theory. The social facilitation theory states that people usually show improvement in performance when confronted with an audience or, in the case of this study, an evaluator. According to Zajonc‟s theory, the presence of an audience enhances a dominant response (for example, an individual who already plays tennis well will improve in the presence of an audience). For individuals who are low in social anxiety, their dominant response in a social situation would most likely not be one of fear, as it is in socially anxious individuals. However, the presence of an audience is expected to enhance fear in socially anxious individuals because it is their dominant response to a social situation. Additionally, current cognitive-behavioral models of social anxiety propose that a negative interpretation bias and fear of evaluation not only causes individuals to rate their performances more poorly, but will also heighten anxiety and maintain the disorder (Rapee & Heimberg, 1997). Supporting this theory, Stopa and Clark (2000) found that participants with high levels of social anxiety symptoms, compared to non-anxious participants, interpreted ambiguous social events in a negative fashion. They found that these interpretations lead to anxious apprehension and sometimes avoidance of social situations. Daly, Vangelisti, and Lawrence (1989) also found that socially anxious individuals had more negative perceptions of their performance when giving a speech in front of an audience. Furthermore, Brown and Stopa (2007) found similar effects when they asked participants to complete a memory task under either a high social-evaluative condition (in front of a video camera) or a low social-evaluative condition (sitting down at a table). Results showed that under high social-evaluative conditions, participants rated their performance as worse compared to individuals in the low socialevaluative condition. Overall, these results indicate that a fear of negative evaluation is an important component in maintaining social anxiety. Therefore, this study investigates whether Threat or Challenge 8 increased evaluation concerns may be contributing to negative self-reports of performance in socially anxious individuals. Paradoxically, individuals low in social anxiety should show improved self-perceptions of their performance due to social facilitation effects. Current Research While both self-focused attention and evaluation appear to be important factors in cognitive models of social anxiety, no studies could be found investigating their interaction in social anxiety. It appears, given past research, that socially anxious individuals tend to interpret ambiguous social situations as being threatening or anxiety-provoking due in large part to the fact that they are focusing almost all of their attention on themselves and their negative views of their performance (Clark & Wells, 1995; Brown & Stopa, 2007; Rapee & Lim, 1992). This indicates that self-focused attention may be crucial in maintaining the fear and anxiety that is associated with social anxiety, but a sense of being negatively evaluated may be crucial in determining poor self-perceived performance. Thus, neither explanation on its own can fully explain these cognitive effects in social anxiety. Given these findings, this study investigates if self-focused attention or evaluation is detrimental on its own to socially anxious individuals‟ self-perceived performance or if it is in fact the interaction of the two that is especially detrimental. We will ask participants who are either high or low in social anxiety symptoms to give speeches under conditions of increased or decreased self-focused attention and evaluation. Participants will then complete ratings of their perceived performance after each speech. We expect that the high socially anxious group will always have worse perceived performance compared to the low socially anxious group. We also expect that increasing self-focused attention will cause both groups to rate themselves as lower in speech performance, although the high socially anxious group will show a significantly greater Threat or Challenge 9 decrease in perceived performance under high self-focus conditions. Evaluation is also expected to affect the groups differently, such that evaluation will have a detrimental effect on socially anxious participants, but it will have a positive effect on low socially anxious participants given previous research with social facilitation (Zajonc, 1965). Finally, we think that there will be an interaction between evaluation, self-focused attention, and social anxiety group status such that the combination of evaluation and increased self-focused attention will be particularly detrimental for the high social anxiety group. Method Participants All participants were recruited from psychology classes through a set of pre-selection measures. Participants answered questions from the Social Interaction Anxiety Scale and from the Social Phobia Scale (SIAS and SPS; Mattick & Clarke, 1998). Those participants who scored less than or equal to one stand deviation below the undergraduate mean on the SIAS (a score of 9 or under) and did not endorse a fear of public speaking on the SPS were eligible for the low social anxiety group (Low SA). Those participants who scored greater than or equal to one standard deviation above the undergraduate mean on the SIAS and endorsed a fear of public speaking on the SPS were eligible for the high social anxiety group (High SA). Two participants in the High SA group were excluded from analyses because they declined to have their data used after they were informed that the study used deception. One High SA participant was excluded due to prior knowledge of the study‟s intent. Two additional participants (one in each social anxiety group) were excluded from analyses due to initial suspicion of the study‟s intent. A total of 48 Low SA participants (28 female) and 44 High SA participants (27 female) were used in the final sample. The mean age of participants was 18.55 Threat or Challenge 10 years (SD = 1.15, range = 17 to 23), and the number of individuals reporting membership in each racial or ethnic group was as follows: 59 Caucasian, 8 African American, 4 Hispanic, 11 Asian, 2 Middle Eastern, 1 Native Hawaiian, and 6 multiple ethnicities or “other.” There were no significant social anxiety group differences in age, t(90) = -1.95, p = .05, d = -0.41; gender,  (4, N = 92) = 0.09, p = .70; or race (comparing Caucasian to all other categories),  (4, N = 91) = 10.80, p = .10. Speech Task Participants were told that the study would examine how students promote various programs that the university offers. They were told that the different programs were either in their piloting stages of marketing research (so that it would not matter how well they did on the speech) or in their final stages of marketing research (so that they would be evaluated on how well they were able to promote the program). Four speeches were done that systematically varied conditions of increased or decreased self-focused attention and high or low evaluation (i.e., low self-focused attention/low evaluation, high self-focused attention/low evaluation, low selffocused attention/high evaluation, and high self-focused attention/high evaluation). To increase self-focused attention, participants were asked to speak in front of a mirror for some conditions. In order to decrease self-focused attention, participants were asked to speak in front of a poster that had various pictures relating to the program topic. Participants were also told in some conditions that they would be evaluated by the examiner to increase evaluation. In these high evaluation conditions, participants were told, “This program is in its final stages of research, so I will be evaluating how well you are able to promote the program. Remember that I am interested in how well you present the aspects of the program that appeal most to you.” While in the low evaluation conditions, examiners prepared for the next task and did not watch Threat or Challenge 11 the participant. Participants in these conditions were told, “This program is in its piloting stages of marketing research, so it doesn‟t matter how well you do; we‟re just interested in the different marketing ideas you may have. While you give your speech I will be getting ready for our next task.” Measures Baseline mood and anxiety measures. Participants were selected for the study and assigned to groups using the Social Interaction Anxiety Scale (SIAS; Mattick & Clarke, 1998), a 20-item questionnaire that assesses anxiety in a variety of social interaction situations. Brown, Turvosky, Heimberg, Harlan, Brown, and Barlow (1997) found that participants with an SIAS score higher than one standard deviation above the mean were likely to meet the criteria for diagnoses of social phobia. Participants also completed the Social Phobia Scale (SPS; Mattick & Clarke, 1998). The SPS is a 20-item questionnaire that, like the SIAS, assesses anxiety in social performance situations. Mattick and Clark (1998) found the SPS to possess a high level of internal consistency as well as test-retest reliability. It was also found to be well correlated with other established measures of social anxiety. Cronbach‟s alpha for this scale was .93. Participants also completed the Brief Fear of Negative Evaluation questionnaire (BFNE; Leary, 1983), an abbreviated version of Watson and Friend‟s (1969) original questionnaire that assessed the anxiety due to negative evaluation from others during social interactions. The original scale of 30 items was reduced to a 12-item scale for the BFNE and items were changed from a truefalse format to a five-point Likert scale, ranging from 1 (not at all characteristic of me) to 5 (extremely characteristic of me). Leary (1983) found the BFNE to be highly correlated to the original scale and to have similar psychometric properties. Cronbach‟s alpha for this scale in the current study was .70. Threat or Challenge 12 In addition to the social anxiety measures, participants completed the Positive and Negative Affect Schedule (PANAS; Watson, Clark, & Tellegen, 1988), a 20-item scale assessing positive and negative affect, at the beginning of the session to check for group differences in mood prior to the public speaking tasks. Cronbach‟s alpha for each subscale was .95. Participants also completed the trait form of the Spielberger State-Trait Anxiety Inventory (STAI-Trait; Spielberger, 1985) to assess general trait anxiety. The STAI-Trait consists of 20 items that ask individuals how they generally feel. Cronbach‟s alpha for this scale in the current study was .93. Manipulation checks. To evaluate the effectiveness of the self-focused attention manipulation, participants completed a slightly modified version of the Focus of Attention Questionnaire (FAQ; Woody, 1996) after each speech. The FAQ is a 10-item scale that includes two 5-item scales that measure the degree to which one focuses their attention on themselves (sample item: “I was focusing on what I should say or do next”) or another person (sample item: “I was focusing on the other person‟s appearance or dress”). In this study, the items that measured other-focus were altered to ask about attention on the poster rather than a conversational partner (e.g., “I was focusing on the appearance of the poster”). Because this study was interested in self-focused attention, only the self-focused attention subscale was used in analyses. Cronbach‟s alpha for this study ranged between .62-.74 across conditions for the self-focus subscale. To assess the effectiveness of the evaluation manipulation, participants were asked to rate on a 0-100 scale to what degree they felt they were being evaluated (where 0 is not at all and 100 is extremely) after each speech. Speech outcomes. Participants were asked to complete a modified version of the Perception of Speech Performance scale (MPSP; Rapee and Lim, 1992). This scale measures Threat or Challenge 13 self-reported performance during a public speaking task and includes six negative (sample item: “said „um‟ a lot”) and six positive (sample item: “appeared confident”) indicators of public speaking performance. This questionnaire was used in the present study to assess participants‟ perceived performance during the public speaking task following each of the four different selffocus/evaluation conditions. Cronbach‟s alpha for this study ranged from .90 to .96 across conditions. Finally, participants rated their anxiety on a 0-100 verbal analogue Subjective Units of Distress Scale (where 0 is completely calm and 100 is extremely anxious; SUDS; Wolpe, 1990) throughout the experiment in order to report their baseline and peak anxiety. Procedure Participants were told the study would investigate ways to advertise several University of Virginia programs and what qualities appeal to typical students at the university. Participants were told that some of the programs were in their piloting stages and thus those speeches would not be evaluated, while other programs were in the final stages of marketing research, and the experimenter would evaluate how persuasive the participants could be when presenting the qualities that appealed most to them. The cover story was used in order to increase the manipulation‟s influence on the public speaking tasks and reduce social desirability effects on the questionnaire tasks. Participants were also told that each speech would be videotaped in order to keep a record of the marketing ideas they had. After informed consent was signed, the experimenter asked participants to rate their current anxiety (using SUDS) and to complete the PANAS and STAI. The four speeches were then completed in counterbalanced order. When participants gave each speech, they were given one of four programs the University currently offers and were told to advertise the program as if they were a spokesperson to an audience. Threat or Challenge 14 Participants were also given a brief paragraph of information about each topic. They were given approximately one minute to prepare each of their speeches. At the end of the planning period, the experimenter asked participants to stand and give their speech in front of a video camera. Participants were asked to speak for one minute. After each speech, participants were asked to rate the highest level of anxiety they experienced during the speech (using SUDS) and also to rate to what degree they felt they were evaluated during their speech (where 0 is not at all and 100 is extremely). Participants were then asked to complete the FAQ and the MPSP to assess their degree of self-focus and perception of their own speech performance after each speech. During the speeches where self-focus was manipulated, participants were asked to speak in front of a mirror, under the cover story that some participants find that a mirror helps them better gauge how they are doing. During the speeches where participants were told that they were being evaluated, the experimenter stood behind the camera holding a clip board with an evaluation form and pretended to be making notes and ratings. Following the speech tasks, participants had a 5-minute relaxation period before completing the other social anxiety measures (SPS and BFNE). At the conclusion of the study, participants went through a funnel debriefing to assess suspicion of the cover story and manipulations. After an exit interview, participants were debriefed with the study‟s true intent. Results Suspicion of Deception During the funnel debriefing at the end of the study, participants were first asked whether they noticed anything strange or suspicious about the study. Next, they were asked whether they suspected that the purpose of the study was not to investigate marketing guidelines. Based on their debriefing answers, participants were placed into the following groups: no suspicion, Threat or Challenge 15 indicated suspicion without being prompted, or indicated suspicion when prompted. As mentioned previously, the three participants (two High SA participants and one Low SA participant) who reported suspicion of the study‟s true intent without being prompted were excluded from further analyses. Of the remaining Low SA participants, 29 indicated suspicion when prompted and 19 reported no suspicion; of the remaining High SA participants, 25 indicated suspicion when prompted and 19 reported no suspicion. The following analyses include the sample of participants who either reported no suspicion or who indicated suspicion only after being prompted that the study involved deception. Descriptive Statistics As determined by pre-selection, the High SA group had a significantly higher mean SIAS score than the Low SA group, t(90) = -21.46 p < .001, d = 1.64. A similar difference was found in SPS scores, with the High SA group having more social performance anxiety than the Low SA group; t(90) = -8.91, p < .001, d = 4.39. Additionally, as predicted, the High SA group had greater fear of negative evaluation on the BFNE compared to the Low SA group; t(90) = -6.22, p < .001, d = 1.30. There were also differences in affect where the High SA group showed lower positive affect, t(90) = 3.69, p < .001, d = .77, and higher negative affect, t(90) = -5.24, p < .001, d = 1.09, on the PANAS. This was also evident in group differences in baseline SUDS ratings with the High SA group reporting significantly more distress than the Low SA group, t(90) = 4.50, p < .001, d = .93. In every condition, the High SA group reported greater peak SUDS than the Low SA group (low self-focus/low evaluation: t = -3.11, p = .002, d = .93; high selffocus/low evaluation: t = -4.08, p < .001, d = .85; low self-focus/high evaluation: t = -5.21, p < .001, d = 1.09; high self-focus/high evaluation: t = -4.51, p < .001, d = .94). In addition to Threat or Challenge 16 reporting high social anxiety, the High SA group reported high trait anxiety on the STAI as well, t(90) = -7.76, p < .001, d =1.64. See Table 1 for descriptive statistics. Manipulation Checks To check the effectiveness of the self-focus and evaluation manipulations, FAQ scores and evaluation ratings were analyzed to see if they showed the expected differences. In combining the evaluation ratings for the two high and two low evaluation conditions, we found that there was a significant difference between the low and high evaluation conditions, t(91) = 12.58, p < .001, d = 2.57, indicating that we able to successfully manipulate evaluative concerns However, in combining the Self-Focus Subscale FAQ scores for the two high and two low selffocus conditions revealed that there were no significant differences between high and low selffocus conditions, t(91) = -.13, p = .90, d = -.02, indicating that the manipulation did not work as expected. This lack of an effect presents a challenge in interpretation of the results; consequently, we will discuss the effects of the mirror manipulation, rather than the effects of “self-focus.” Effects of the Mirror Manipulation and Increased Evaluation on Perceived Performance To test for differences in perceived performance, a 2 (mirror: present, absent) by 2 (evaluation: high, low) by 2 (social anxiety group: High SA, Low SA) repeated measures ANOVA was conducted, with mean MPSP score as the dependent variable. As predicted, there was a main effect for social anxiety group, F(1,90) = 13.91, p < .001, with the High SA group always rating their performance as being lower than the Low SA group regardless of condition. See Figure 1 for a graph of these results. Surprisingly, there were no main effects for mirror presence or evaluation, F(1,90) = .10, p = .76; F(1,90) = .28, p = .60, respectively. Similarly, tests for mirror presence by group, mirror presence by evaluation, and mirror presence by Threat or Challenge 17 evaluation by group interactions were non-significant, F(1,90) = .48, p = .49; F(1,90) = .24, p = .63; F(1,90) = .002, p = .97, respectively. However, the predicted significant group by evaluation interaction was found; F(1,90) = 4.36, p =.04. Although the follow-up t-tests did not reach significance, there was a small effect in which the High SA group rated their performance more negatively in conditions of increased evaluation, t(43) = 1.60, p = .12, d = .39, and the Low SA group rated their performance more positively in conditions of increased evaluation, t(47) = -1.31, p = .198, d = .24. See Figure 2. Discussion This study attempted to investigate two factors involved in the experience of social anxiety for a better understanding of why individuals with social anxiety rate their performance as being poorer than that of their non-anxious peers. Our hypothesis that socially anxious individuals would consistently rate themselves as having performed worse than the low social anxiety group was supported. Additionally, in line with hypotheses, there was a significant group by evaluation interaction, such that the high social anxiety group tended to rate themselves worse in high evaluation conditions, while the low social anxiety group actually tended to rate themselves better in the high evaluation conditions (though the follow-up tests did not reach significance). Contrary to hypotheses, no effects were found for the self-focused attention, mirror manipulation. Interestingly, our results indicate that the manipulations may not have affected the variables they were intended to affect. For example, even though our manipulation for evaluation increased the sense of being evaluated, it also unintentionally increased self-focused Threat or Challenge 18 attention 1 . On the other hand, our manipulation for self-focused attention did not cause greater self-focused attention. There are several possible explanations for the findings of our manipulation checks: first, the mirror may not be a strong enough manipulation to increase selffocused attention when a video camera is also present across self-focus conditions (another common tool for increasing self-focused attention; Albright & Malloy, 1999) as it was in this study. Additionally, it is also possible that the measure used to rate self-focused attention (the Focus of Attention Questionnaire; Woody, 1996) may not have been sensitive enough to show effects of the self-focused attention manipulation. However, it is also possible that evaluation is itself too strong a self-focused attention manipulation. Past research has shown evaluation to be a fairly common method of manipulating self-focused attention in addition to a mirror or video camera (Brown & Stopa, 2007). It is also possible that the effects of evaluation and self-focused attention are so intertwined that it is difficult to tease them apart in a laboratory setting. However, the group by evaluation interaction found in our study provides some support for a threat/challenge hypothesis behind the negative interpretations associated with perception of one‟s own performance in social anxiety. As indicated earlier, our results show a slight tendency for individuals low in social anxiety to rate themselves as doing more positively in high evaluation conditions (perhaps because they see evaluation as a positive challenge), whereas individuals high in social anxiety rate their performances as being more negative in high evaluation conditions (perhaps because they see evaluation as a negative threat). While our results only indicate a small effect, we believe that this evaluation interaction may be the driving force in the negative interpretation bias that individuals with social anxiety often have, and with In combining the evaluation ratings for the two high and two low evaluation conditions, we found that there was a significant difference in self-focused attention, with greater self-focus ratings in the high evaluation conditions, t(90) = -5.65, p < .001, d = 1.16. Threat or Challenge 19 a larger sample size we suspect the follow-up within-group comparisons would be significant. Notably, the effect sizes for our follow-up tests did indicate small to moderate effects. Overall, these results suggest that different interpretations of social performance situations (i.e., as a threat or a challenge) may affect high and low social anxiety groups differently, leading to more negative self-perceptions in anxious individuals. Limitations and Conclusions As stated earlier, a major limitation of this study lies in the inability to both manipulate and possibly measure self-focused attention. The lack of effects indicates that perhaps a mirror is not always a strong enough manipulation, especially when paired with a video camera. It is also possible that participants were not focusing on the mirror when giving their speech (many participants watched the camera even though they were told to look at the mirror). Additionally, the largest limitation in this study may have been habituation effects because participants completed four speeches. Though we attempted to moderate this effect by counter balancing, anxiety may have decreased over the course of the four speeches. This may have contributed to the relatively small effect size we saw in the evaluation by group interaction. A larger sample size where a between-subjects design could be implemented would be valuable for future research. Using the threat/challenge framework, cognitive-behavioral therapies can look at an anxious person‟s thoughts during behavior modification and exposure exercises to change negative interpretations of a situation as being threatening or dangerous into more healthy interpretations of positive challenge. These results suggest that individuals who are high in social anxiety may be struggling with a negative interpretation bias of their performance, and yet every day we humans are forced to interact with one another. One person may see this kind of Threat or Challenge 20 experience as dangerous and even go on to develop social anxiety disorder, with its associated avoidance and distress, while another person will never think twice about social performances. Therefore, it is evident that a greater understanding of these effects will provide us with a greater understanding of anxiety and social interactions in general. Threat or Challenge 21