Defining Domains of Social Functioning in Adults with Autism Spectrum Disorder As Targets for Treatment

Background: There is increasing recognition of the need to develop treatments for adults with autism spectrum disorder (ASD) to improve their social functioning, a key factor in adults’ employment, relationships, and overall quality of life (Howlin et al., 2013; Shattuck et al., 2012). However, little is known about the relative level of impairment in various behavioral domains underlying social functioning, or about the relationship among these behavioral domains in adults with ASD. To our knowledge, previous studies have not measured these social behavior domains in the same sample of adults with ASD, which makes it difficult to know which domains should be targeted by treatments and how to best track treatment response. In this study, adults with ASD completed multiple measures in each of four domains underlying social functioning: social motivation, social anxiety, social cognition, & social skills. We hypothesized there would be a high degree of correlation among measures within the same domain (i.e. convergent validity), and more modest degrees of correlation among measures across domains, indicating only a modest relationship among these domains, and the need for multidimensional treatment approaches.

Objectives: To assess behavioral domains underlying social functioning in adults with ASD, and measure correlations among measures within and between these behavioral domains. This analysis will help determine whether multidimensional treatment approaches are warranted, and will provide a baseline profile of these domains in adults that can be used for tracking treatment response.

Methods: Participants were enrolled in a pilot study of a new cognitive behavioral treatment program, TUNE In, Training to Understand and Navigate Emotions and Interactions. At baseline, twenty-nine adult participants with ASD (Table 1) completed multiple measures in each of four behavioral domains, as well as measures of overall ASD phenotype and measures of community social functioning. The raw scores of some measures were inverted so that higher scores indicated greater impairment in all measures. Then, raw scores for each measure were converted into standard scores, and bivariate correlations between scores were analyzed using SPSS. To correct for multiple comparisons, significance thresholds were set at a false discovery rate of 10%.

Results: There were statistically significant, robust correlations among measures within each domain, with the exception of social cognition and community functioning (Figure 1). Measures of social motivation were significantly correlated with measures of all other domains, except for social cognition. Additional significantly strong cross-domain correlations were found between measures of anxiety & ASD phenotype; measures of social skills & community functioning; and measures social skills & ASD phenotype.

Conclusions: There are significant correlations among measures within and between behavioral domains underlying social functioning in adults with ASD. Social motivation, in particular, is highly correlated with all domains, except social cognition. However, the variability among the participants in each measure and the lack of robust correlations between all domains suggest a need for multidimensional service strategies that can target the particular domains of social functioning most in need of improvement in each individual with ASD.