24021
Effects of Social Skills on Anxiety and Parasympathetic Activity Among Youth with Autism Across the PEERS® Intervention
Objectives: To determine if adolescents with ASD showed significant declines in self- and parent-report of anxiety and improvements in respiratory sinus arrhythmia (RSA) across three time points.
Methods: Thirty-nine adolescents with ASD (N=39) aged 11 to 16, IQ≥70 participated in this study. ASD was confirmed using the ADOS. Adolescents comprised the experimental group of a randomized controlled trial and received the Program for the Education and Enrichment of Relational Skills (PEERS®), which focuses on improving friendship quality and social skills among adolescents with ASD. Adolescents completed the Social Anxiety Scale for Adolescents (SAS-A), Youth Self-Report (YSR), and an eyes-open resting-state RSA paradigm (Porges & Bohrer, 1990). Parents/caregivers completed the Social Anxiety Scale for Adolescents, Parent (SAS-P) and the Child Behavior Checklist (CBCL). All data was collected pre-, post-, and six months following PEERS®.
Results: Pearson’s correlations showed some evidence of a relationship between anxiety and RSA, but no consistent pattern across time points. Repeated Measures ANOVAs showed significant declines in self-report of social anxiety, SAS-A (F(2, 76)=9.174, p<.001), parent-report of social anxiety, SAS-P (F(2, 64.34)=11.34, p<.001), and parent-report of general anxiety, CBCL (F(2, 32)=4.476, p=.019), but not self-report of general anxiety, YSR (p>.05). Paradoxically, RSA significantly declined across the three time points (F(2, 55.04)=8.875, p<.001).
The sample was subsequently split into socially anxious (SA) and not (NSA) using the clinical cutoff for the SAS-P at pre-, and the ANOVAs were rerun. Results showed significant differences across time for the SA subgroup (SAS-A, (F(2, 68)=8.35, p=.001); SAS-P, (F(2, 44)=13.92, p<.001); CBCL, (F(2, 24)=5.85, p=.009); YSR, p>.05), but not for the NSA subgroup (all p’s >.05). RSA data continued to show declines across time for the SA subgroup (F(2, 50)=7.35, p=.002), NSA (p>.05).
Conclusions: Results suggest the PEERS® intervention has a lasting positive impact on co-occurring symptoms of general and social anxiety, especially for adolescents who begin with high levels of social anxiety. Likely, as adolescents with ASD utilize social skills, their symptoms of anxiety decline. Interestingly, measures of arousal were not strongly related to anxiety, and implied that employment of social skills leads to heightened dysregulation, contrary to previous findings among neurotypicals (Porges, 2003). Perhaps, for adolescents with ASD in a social skills intervention, outcomes may be better explained by an Information Processing Model of RSA rather than the Social Engagement Model. That is, social interaction may require greater attention and more effort, associated with greater dysregulation and heightened arousal as these skills are employed. Further research is needed to better understand which of these competing models best explains social interaction in ASD across intervention. Longer-term follow up is also necessary to understand how these processes evolve over time.