Category: Adult Anxiety - Social

PS10- #A12 - Shame Responses to a Social Exclusion and Inclusion Task for Individuals With Problematic Social Anxiety

Saturday, Nov 18
11:00 AM – 12:00 PM
Location: Indigo Ballroom CDGH, Level 2, Indigo Level

Keywords: Social Anxiety | Cognitive Biases / Distortions | Social Relationships

Individuals with higher social anxiety (HSA) exhibit more negatively biased interpretations of social interactions, compared to individuals with lower social anxiety (LSA; Constans, Penn, Ihen, & Hope, 1999). This may predispose them to experience more shame. Shame serves as a marker of relationship security as it signals devaluation or exclusion by others (Gilbert, 2007). Problematic social anxiety levels may increase shame-proneness as more negative interpretations of social interactions contribute to perceptions of relationships as insecure. Most people are similarly affected by severe ostracism (McDonald & Brent Donnellan, 2012), but when the ostracism is more mild the affective responses of individuals with problematic social anxiety may become more distinct.


We attempted to clarify the relationship between social anxiety and responses to social ostracism (shame and negative affect) using a task called Cyberball (Williams & Jarvis, 2006). It was hypothesised that individuals with HSA would report greater changes in state shame and negative affect after being socially excluded and included.


We recruited 106 participants aged 18 to 83 years old, who played a game simulating mild social exclusion, followed by a game in which they were fully included. Social anxiety correlated positively with participant ratings of feeling more socially excluded in the first game, r(106) = .35, p < .001. Using a subset of participants with HSA and LSA (+/-1SD, n = 30), a repeated measures ANOVA was conducted on state affect at baseline and following each game. HSA participants experienced greater changes in state shame, Wilks’ λ = .717, F (2, 27) = 5.33, p = .011, partial η2 = .28, where shame increased in response to exclusion and decreased in response to inclusion. Contrary to expectations, while HSA participants’ negative affect showed greater change compared to LSA, Wilks’ λ = .57, F (2, 27) = 10.19, p = .001, partial η2 = .43, where it decreased in response to exclusion as well as inclusion. Hierarchical regression analyses of the entire sample found that social anxiety explained 7% of unique variance in state shame levels in response to social exclusion, and 4% in response to social inclusion. Social anxiety made no significant contribution to prediction of changes in state negative affect after levels of perceived social exclusion were taken into account.


Our findings are consistent with the idea that shame serves as a marker of relationship security - greater fluctuations in shame may reflect greater perceived relationship insecurity in individuals with problematic social anxiety. Social exclusion and inclusion may be imbued with greater meaning for individuals with problematic social anxiety, resulting in a more intense emotional response (Frijda, 1986; Wilson & Rapee, 2005).

Christopher Jillard

Ph.D. Student
Swinburne University of Technology
Melbourne, Victoria, Australia

Michelle H. Lim

Lecturer in Clinical Psychology
Swinburne University of Technology

Glen Bates

Pro-Vice Chancellor (Student Engagement)
Swinburne University of Technology