Towards an Understanding of Social Anxiety and Close Friendships: The Role of Expectations, Negative Self Portrayal Concerns, and the Discrepancy Gap

Abstract

Past research has demonstrated a robust correlation between higher social anxiety (SA) and lower friendship satisfaction, suggesting that the interpersonal difficulties experienced by socially anxious individuals in anxiety-provoking social situations with evaluative others might extend to their close friendships. My dissertation sought to clarify how symptoms of SA manifest within close friendships, as well as the potential cognitive-behavioural mechanisms by which SA symptoms may influence friendship outcomes. Each research question was first investigated within an undergraduate student sample (Studies 1A and 2A), with SA assessed as a continuous variable across the full spectrum of symptom intensity, followed by replication and extension using a clinical sample (Studies 1B and 2B). First, I developed a measure of adult friendship expectations and assessed variability in how individuals with different levels of trait SA believe their close friends ought to behave. Findings showed that, among undergraduate students, ratings of friendship expectations did not vary by participant level of SA; however, among community participants, diagnosis of Social Anxiety Disorder was predictive of significantly lower expectations related to solidarity and agency, but no differences in expectations of symmetrical reciprocity or communion. To assess the extent to which participants were concerned about negative self-portrayal when in the company of their close friends, I adapted the instructions of the Negative Self Portrayal Scale (Moscovitch & Huyder, 2011) to inquire specifically about the extent to which participants have such concerns within close friendship contexts. Consistent with hypotheses, among both undergraduate and community participants, SA was predictive of greater concern about revealing perceived personal flaws to close friends. Next, I used an experimental study design in which discrepancy threat salience was manipulated as a between-subjects variable, to test how socially anxious individuals respond to potential friendship ruptures. While all participants reported more interpersonal threat in response to the Threat versus Control condition, this effect was particularly strong for individuals with social anxiety disorder. Consistently across samples, SA was linked to greater negative emotionality regardless of study condition, while results regarding positive emotionality were more varied, highlighting the importance of assessing positive and negative affect as distinct facets of human experience. Effects of perceived interpersonal threat on self-reported behavioural motivation was also surprisingly inconsistent across studies: although SA symptoms predicted motivation for lower approach and greater avoidance among undergraduate students, a clinical diagnosis of social anxiety disorder (SAD) was unrelated to self-reported behavioural motivations within the community sample. Findings are interpreted within the context of cognitive and interpersonal models of SA, highlighting methodological strengths and limitations of my research, and suggesting future research to build on the present studies that may inform interventions that enhance social connection in individuals with SAD.