Speech Acts during Spontaneous Peer Conversation in ASD and Typical Development

Background: In typical development (TYP), early peer talk is crucial for pragmatic development. Speech acts – the primary illocutionary values conventionally conveyed by utterances as acts – are a crucial component of children's conversational capabilities. The pragmatic deficit, reflected in remarkably deficient conversational capabilities, is considered the hallmark of the language deficit in autism spectrum disorder (ASD) (Stefanatos & Baron, 2011). Despite its importance for understanding the pragmatic deficit in ASD, preschoolers’ speech acts during spontaneous peer talk has not been explored.

Objectives: The current study’s major aim was to close this knowledge gap about characteristics of the pragmatic deficit in ASD by examining the speech acts of high-functioning preschoolers with ASD (HFASD, IQ>75) versus preschoolers with TYP during spontaneous peer talk. We were also interested in learning about the role of the conversational partner’s friendship and disability status as contributing to pragmatic capabilities. Prior studies reported higher mutual social engagement, responsiveness, and reciprocal verbal exchanges in interactions with friends versus non-friends (Newcomb & Bagwell, 1995). Aadditionally, in the HFASD group, prior research showed higher social complexity and level of play in interactions with a typical partner (mixed dyads) versus a partner with HFASD (non-mixed dyads) (Bauminger et al., 2008).

 Methods: We conducted comparative assessment of speech acts during spontaneous peer talk along 10-minute free-play scenarios in preschoolers (ages 3-6 years) with HFASD (n=21) versus with TYP (n=30). Groups were matched on SES, verbal IQ (VIQ), nonverbal IQ (NVIQ), and CA. Children's conversations were videotaped and coded to tap speech act capabilities. Acts’ coding included: Assertive (e.g., evaluation, declaration); Requestive (e.g., questions, requests); Responsive (e.g., answers to WH questions); Organizational Devices (e.g., request for attention) (Dore, 1979); and also Stereotypic Speech. Correlations with CA, VIQ, and NVIQ were examined. We compared the two groups' interactions with a friend versus non-friend partner; additionally, in the HFASD group, we compared interactions in mixed versus non-mixed dyads.  

Results:  As predicted, analysis of observed speech acts during spontaneous peer conversations revealed fewer speech acts in HFASD versus typical age-mates, mainly in assertive (declarative) acts and organizational devices. Moreover, the HFASD group used stereotypic talk more often than TYP peers. However, in both groups, more frequent and diverse uses of speech acts emerged during interaction with friends versus non-friends, including: assertive (declaration, evaluation); requestive (request); responsive (replies); and organizational devices. Thus, friendship may enable children to converse in a more socially complex way. The comparison of mixed versus non mixed friendship in HFASD did not yield significant differences. CA correlated positively with frequency and variety of speech acts only in the TYP group, whereas VIQ correlated with speech acts’ frequency in the HFASD group while interacting with a non-friend.

Conclusions: Despite the poor performance of speech acts in ASD, involvement in friendship relationships and VIQ were linked to more intact pragmatic capacities. Clinically, emphasis on friendship in early intervention may help elicit more adequate pragmatic capabilities and presumably fuller peer relations in preschoolers with ASD.