Pilot Study of a Peer-Enacted Role Play Measure of Social Skills for Adolescents with High-Functioning Autism and Asperger Syndrome

Background: Social skills have been identified as a high priority target of treatment and research among adolescents with high-functioning autism (HFA) and Asperger’s syndrome (AS) (Tse, Strulovitch, Tagalakis, Meng, & Fombonne, 2007). While numerous social skills interventions have been developed for this population, the development of externally valid social skill assessment measures has not proceeded at the same pace (Matson & Wilkins, 2007). During adolescence, the primary context for social interaction is the peer context, particularly casual conversations (Paul, 2003). One of the most critical skills in successful conversation is responding appropriately to verbal and nonverbal social cues (Turkstra, Ciccia, & Seaton, 2003). Additionally, recognizing and responding appropriately to social cues is a frequent target of social skills interventions for adolescents with HFA and AS (Krasny, Williams, Provencal, & Ozonoff, 2003). Role play measures allow for direct assessment of these critical social interaction skills. While role play measures have been successfully utilized to assess social skills in several populations (Bellack et al., 2006; Blake & Andrasik, 1986), and as a teaching tool in autism (Krasny et al., 2003), a systematic role play measure of social skills has not been developed for ASD.

Objectives: The primary goal of this study was to pilot a peer-enacted role play measure of social skills for adolescents with HFA and AS, focused on the ability to respond appropriately to verbal and nonverbal social cues. Specific objectives included assessing interrater reliability and convergent and discriminant validity via testing in control and HFA/AS samples.

Methods: Participants included 20 participants with high-functioning autism and Asperger’s syndrome and 20 undergraduate controls. Participants completed the Wechsler Abbreviated Scales of Intelligence (WASI), the Awareness of Social Inference Test (TASIT), which assesses theory of mind, and the Social Reciprocity Scale (SRS), in addition to the newly developed peer role play measure of social skills, the Contextual Assessment of Social Skills (CASS). The CASS consisted of two conversations enacted with two different trained undergraduates. In condition 1, the undergraduate demonstrated social interest through a standard set of verbal and nonverbal behaviors. In condition 2, the undergraduate displayed disinterest and boredom through corresponding verbal and nonverbal behaviors. Trained coders rated the participants’ social behaviors in each condition. Participants’ behaviors in each role play and change in social behavior across role plays were examined.

Results: The properties of the role play measure will be evaluated for internal consistency and interrater reliability. Linear regression analyses will be used to examine group differences in social behavior across the Interest and Disinterest conditions. In addition, criterion-related validity will be examined via Pearson correlations with verbal IQ, performance IQ, theory of mind, and SRS scores.

Conclusions: Results of this study will contribute to the field of social skills assessment in high-functioning individuals on the autism spectrum. Issues around utility, validity, and feasibility of this role play assessment of social skills will be examined in relation to existing measures of social skills for this population.