Pragmatic Language Abilities in Siblings of Individuals with ASD

Background: Pragmatic language impairment is a defining feature of autism spectrum disorder (ASD). More subtle pragmatic differences also constitute a core feature of the broad autism phenotype in clinically unaffected relatives (BAP; Landa, 1991, 1992; Losh et al., 2008; Piven et al., 1997). Whereas pragmatic abilities have been studied extensively in parents through both direct assessment and questionnaire, studies of pragmatic skills in siblings have been more limited, relying on parent report and/or focused on early developmental periods when more complex pragmatic skills have not yet emerged. This study assessed pragmatic language abilities in school-age and young adult siblings of individuals with ASD, using direct assessment of conversational interactions.

Objectives: To characterize pragmatic language features of the BAP in siblings of individuals with ASD, and explore how pragmatic abilities aggregate within families.

Methods: Participants included 117 individuals with ASD, 43 siblings, and 46 typically developing controls. Pragmatic language abilities were assessed using the Pragmatic Rating Scale-School Age (PRS-SA; Landa, 2011) during conversational interactions from the Autism Diagnostic Observation Schedule (ADOS; Lord, 2001) or ADOS-2 (Lord, 2012). The PRS-SA is comprised of items that tap five sub-domains of pragmatic skill: presupposition/theory of mind, discourse management, speech and language behaviors, suprasegmental characteristics (e.g., intonation, rate of speech, volume, and rhythm; which are important for mediating expression of communicative intent), and non-verbal communicative behaviors. ANCOVAs were conducted to examine group differences. Sibling correlations were also explored. Analyses controlled for age and IQ.

Results: A stepwise pattern emerged across all domains of the PRS-SA, where individuals with ASD presented with the greatest pragmatic language impairment, followed by siblings, and controls showing lowest rates of pragmatic violations. Whereas individuals with ASD exhibited greater impairment than both sibling and control groups across all domains (ps<.001), siblings differed from controls only in the suprasegmental domain (p=.031). Additionally, pragmatic language ability was positively correlated between the ASD and sibling groups (r=.27, p=.03), with correlations driven by associations in the suprasegmental domain (r=.31, p=.01).

Conclusions: Findings revealed differences in suprasegmental aspects of speech among siblings of individuals with ASD. Within-family associations further suggest that suprasegmentals may comprise a key feature of the pragmatic-related component of the BAP in siblings. This contrasts with studies of the BAP in parents, where more widespread pragmatic difficulties have been observed. However, studies of parents included more naturalistic conversational tasks, which highlights the importance of extending the current study to contexts beyond the ADOS. Finally, results emphasize the need for further exploration into suprasegmental differences observed in siblings. Future studies should investigate suprasegmental use in siblings objectively through acoustic analysis which may reveal key aspects of pragmatic language impacted by genetic liability to ASD.