Cognitive Behavioural Theory, Social Anxiety and Autism

Background: Social Anxiety Disorder (SAD) has a lifetime prevalence of 12% in the general population. Studies report 50% of autistic adults meet criteria for SAD. Effective psychological treatment informed by the cognitive behavioural model cites negative self-beliefs, self-focussed attentional processes including observer-perspective self-images and high standards for social performance as significant maintaining factors. A developmental model has been proposed for SAD co-occurring with ASD. High trait anxiety and social interaction difficulties foster an anxious and avoidant style in respect of social situations (Bellini, 2006). Distinct approaches to psychological treatment arise from these different theories.

Objectives: To investigate the relevance of cognitive factors to social anxiety co-occurring with autism.

Methods:

Study 1 Negative performance beliefs: ASD participants (aged 16-21 years) (N=41) were identified as low social anxiety (LSA) (N=19) and High social anxiety (HSA) (N=22) on the basis of the Social Anxiety Scale for Adolescents (SAS-A). A small group discussion task (adapted from Parr and Cartwright-Hatton, 2009) with post-hoc self and objective observer ratings of social performance, revealed significant differences between raters (self vs observer) (F (1, 38)= 8.74, p<0.05), but no effect of group (HSA vs LSA). There were no differences between the groups in observer rating of social performance, but self-rating was significantly lower in the HSA group.

Study 2 Self-focused attention: Participants (n=33) (aged ≥ 16 years) completed the Social Phobia Inventory (SPIN): (HSA N=16; LSA N=17). A small group discussion task was followed by completion of the Focus of Attention Questionnaire (SFA) and Adapted Autonomic Perception Questionnaire (APQ). There was a significant main effect of group (F (2, 30) =.755, partial ƞ2=.34) on SFA scores, which were significantly higher in the HSA group (F(1,31)=14.66, p=.002, partial ƞ2=.32). The majority of participants (82%) noticed changes in physiological symptoms on the adapted APQ. APQ scores were significantly higher in the HSA group than the LSA group (t(32)=-2.28, p=.029, 95% C.I. -.57--.03).

Study 3 Images: Adults (n=62) with clinical diagnosis of ASD recruited via an online survey (SPIN scores: (HSA) (N=31) and (LSA) (N=31)) participated in an adapted Imagery Interview (Hackmann et al., 1998). The interview comprised 3 conditions, Relaxing Image 1, Social Situation Image and Relaxing Image 2. State anxiety ratings and image dimensions: vividness, frequency, distress, and controllability were compared between the groups. State anxiety ratings differed by Image condition but not group. The HSA group reported more frequently distressing images of themselves in social situations with an observer- perspective.

Results: The results of these preliminary studies suggest factors central to CB models of Social Anxiety Disorder (SAD) may have relevance for Autism. There was some tentative evidence of negative performance beliefs but the study was lacking statistical power. Significant differences in self-focussed attention and distressing self-images were reported by participants high in social anxiety.

Conclusions: Larger scale studies should aim to replicate these findings. If established, treatment endeavours should consider how best to adapt cognitive change techniques in interventions adapted for autistic people.