Critical Self-Referent Attributions Potentiate Social Skills Intervention Response in Adolescents with Asperger Syndrome & High-Functioning Autism

Background: A need exists for empirically-validated interventions addressing social skills in adolescents with autism spectrum disorder (ASD) and for a greater understanding of mechanisms of change in interventions (White, Keonig, Scahill, 2007; Kazdin, 2008). Critical self-referent attributions (CSA)—social cognitions representing the tendency to derive pejorative self-evaluations from ambiguous peer cues—have been shown to be associated with actual peer experiences and are thought to be linked to maladaptive social behaviors and depressive symptoms in typical child populations (Prinstein, Cheah, Guyer, 2005), but to our knowledge have never been assessed in ASD populations. Given the high rates of these symptoms in adolescents with asperger syndrome (AS) and high-functioning ASD (HFA) (Williamson, Craig, Slinger, 2008), we expect CSA to be relevant in this population. Given that CSA entails a heightened sensitivity to social failure, we hypothesize that it may facilitate receptiveness to interventions designed to address such failure, thereby potentiating social skills intervention effects.

Objectives:

1. To ascertain whether participation in the six-week Socio-dramatic Affective-Relational Intervention (SDARI; Lerner, Mikami, & Levine, under review; Lerner & Levine, 2007) is effective in improving parent-reported social skills as compared to a matched comparison group.

2. To determine if baseline CSA predicted response within this intervention.

Methods: This study included 35 participants with AS and HFA divided into a SDARI intervention group (n=27. 22 male, 5 female; mean age=13.1) and an age, sex, and diagnosis-matched comparison group (n=8, 6 male, 2 female; mean age=14). Multilevel modeling was used to model change across both the SDARI and comparison conditions. The parent-report Social Responsiveness Scale (SRS; Constantino, 2002), thought to be relevant to ASD and sensitive to change (White et al., 2007), was used to measure treatment response at 3 week intervals beginning six weeks prior to intervention to 6 weeks post-intervention. The Hostile Attribution Questionnaire (adapted from Prinstein, et al., 2005) was used to measure CSA, and was administered during the second week of SDARI.           

Results: When comparing the SDARI and comparison groups on SRS across time, slope was significantly predicted by both time (β=-0.402, p=.002) and intervention status (β=-0.496, p=.017), indicating improvement over time in both groups, but significantly greater improvement in the SDARI group. When entered as a predictor at the random level, CSA significantly predicted slope (β=-0.743, p<.05), indicating greater intervention responsiveness among those with higher CSA at baseline.

Conclusions: Results support the hypothesis that SRS ratings of SDARI participants improve over time relative to comparison condition participants, and that CSA is a significant predictor of these effects. Those with higher CSA seem to benefit more from SDARI than those with lower CSA, indicating either that SDARI may be optimally effective for ASD individuals who are highly self-critical or that such individuals may be more receptive to social skills interventions in general. Further research should assess both the specific efficacy of SDARI for ASD populations and the role that social cognitive factors such as CSA play for this population in terms of both taxonomy and intervention.