Differential Predictors of Social Behavior with Unfamiliar Adults Versus Peers in Higher Functioning Children with Autism

Background: There is a wide range of variability in the severity of social deficits displayed by higher functioning individuals with autism.  Syndrome specific and non-syndrome specific individual differences may predict variability in the severity of social deficits.

Objectives: The aim of the current study was twofold: 1) To examine the correlation between Autism Diagnostic Observation Schedule (ADOS) Social-Communication combined scores and an observational measure of social and communication skills obtained through a peer interaction task. 2) To examine and compare the syndrome specific and non-syndrome specific individual difference factors that predict performance on ADOS and during a peer interaction.

Methods: Twenty-nine higher functioning individuals with autism, ages 9 to 17, participated in two laboratory visits, an interaction with an unfamiliar age-, gender-, and IQ-matched typically developing peer and an individual assessment. The dyad was given five minutes during which they were instructed to get to know each other. Eye contact, appropriateness of conversation/interaction, and conversational efficacy were coded from video recordings using 7-point Likert scales. A composite score of Peer Competence was obtained by standardizing and summing scores on the three behavioral codes. In a separate visit, an ADOS was administered to each individual. Of interest was the combined Social-Communication score. Syndrome specific factors (verbal IQ and theory of mind) were assessed using the Wechsler Intelligence Scale for Children-Fourth Edition (WISC-IV) Verbal Comprehension Index (VCI) and the Reading the Mind in the Eyes Test. Non-syndrome specific factors of inhibitory control and social anxiety were assessed with Early Adolescence Temperament Questionnaire (EATQ-R) self- and parent reports and Social Anxiety Scale for Children (SASC) self-report.

Results: Preliminary correlational analyses revealed a strong negative correlation between ADOS Social-Communication and Peer Competence, r (27) = -.55, p = .002, such that higher social-communicative deficit on the ADOS corresponded to lower observed competence during the peer interaction. Despite this strong correlation between the independent indices of social skills, correlations revealed different sets of syndrome specific and non-syndrome specific predictors of behaviors in the ADOS versus peer interactions.

WISC-IV VCI was significantly associated with Peer Competence, r (27) = .40, p = .031, and approached significance for the ADOS, r (27) = -.34, p = .071.  However, Eyes Test, r (25) = .385, p = .047, self-reported Inhibitory Control, r (25) = .41, p = .034, and Fear of Negative Evaluation scores, r (27) = -.365, p = .052, predicted Peer Competence but not ADOS social-communication scores.

Conclusions: The ADOS is commonly used as the primary instrument for assessing the severity of current symptoms in individuals with autism. The results of the current study suggest that supplementing assessments with in vivo observations of peer interactions provide an effective way to capture variation in children’s implementation of social skills.  Higher order cognitive skills including theory of mind and aspects of emotional reactivity (social anxiety) and self-regulation may be particularly important in peer contexts where interactions are highly dynamic and potentially demanding especially during late childhood and adolescence.