Do Children with Autism Show Collaborative Competence in Dialogue?

Background: It is widely acknowledged that verbally able individuals with autism have difficulty in adapting their language to pragmatic context (e.g. Gernsbacher, Geye, & Weismer, 2005; Tager-Flusberg, Paul, & Lord, 2005).  In the case of conversations, speakers and listeners need to monitor each other for understanding and interest, interpreting and offering cues as feedback on each other’s contributions.  A variety of non-verbal, verbal and paralinguistic features are used to manage this pragmatic dimension of conversational interchanges so that turns of talk are organized, maintained and shared. Although there is evidence that non-verbal aspects of conversational exchanges such as head-nodding while listening are atypical among individuals with autism (García-Pérez, Lee, and Hobson, 2007), conversational management (or grounding, see Clark, 1996) has received little investigation among individuals with autism.
Objectives: The aims of this study were to evaluate how children with and without autism collaborate in a conversation with someone else using verbal, nonverbal and paralinguistic cues, and to explore what this might reveal about inter-subjective contributions to conversational competence.  To this end, we applied novel ratings of communication grounding to assess collaborative competence in dialogue.

Methods: Participants were 18 verbally able children and adolescents with autism and an age and language-matched comparison group of 18 children and adolescents without autism, all between the ages of 7 and 15.  Participants were administered the Autism Diagnostic Observation Schedule (ADOS), Module 3, a semi-structured, standardized assessment of communication and social interaction (Lord, Rutter, DiLavore & Risi, 2002). This assessment provides 35 – 40 minutes of conversational interaction between the participant and an examiner who presents numerous social presses to elicit conversation and dialogue. Videotapes of the ADOS administration were rated according to a novel rating scale (the Collaborative Competence in Dialogue Scale, CCDS) to assess the presence and quality of seven collaborative features in conversation: continuers, assessments, appropriate next response, try markers, gaze to regulate, gaze to co-regulate, and repairs.
Results: Participants with and without autism were equally likely to include continuers, assessments, appropriate next responses, and try markers in their conversations. However, with the exception of assessments, each of these was rated as markedly atypical in the autism group. Furthermore, participants with autism were less likely to use gaze to regulate, gaze to co-regulate, and repairs.  It was rare for any of the collaborative features to be rated as both present and typical (and to enhance conversation effectively) among the participants with autism. Collaborative cues tended to appear in rote and/or intermittent forms.
Conclusions: Verbally able individuals with autism have pragmatic difficulties that go beyond understanding and producing context-relevant speech.  They also have problems with conversation management and collaboration in dialogue. Drawing on constructs from conversation analysis, this study suggests that limitations in interpersonal engagement may underlie many of the atypicalities in conversation observed in autism.