Pragmatic Language and Social Cognitive Overlap in Children with Autism and Fragile X Syndrome

Background: Impairments in pragmatic language and social cognition are core features of autism. Fragile X syndrome (FXS) is the leading single-gene disorder associated with a diagnosis of autism; recent estimates suggest that up to ~50% of males with FXS meet diagnostic criteria for autism on gold standard instruments (Hall et al., 2008). Moreover, many of the pragmatic difficulties attributed to FXS are also observed in autism (Roberts et al., 2007).
Objectives: This study compared the pragmatic language and social cognitive skills of boys with autism and boys with FXS in order to identify precise autism phenotypes that could be linked with the Fragile X Mental Retardation-1 gene (FMR1).
Methods: Pragmatic language and social cognition were examined in boys with autism (n=19) and FXS (n=54), ranging in chronological age from 3-15 years, as well as a typically developing (TD) control group (n=22), ranging in age from 2-6. We administered the Pragmatic Judgment subtest of the Comprehensive Assessment of Spoken Language (CASL) (Carrow-Woolfolk, 1999), a standardized assessment of pragmatic language. Conversational discourse in a more naturalistic context was elicited using the Autism Diagnostic Observation Schedule (ADOS; Lord et al., 1989), and examined for the degree to which the child’s turn was contingent, or topically related to the preceding turn. We examined social cognitive performance using standard theory of mind (ToM) tasks for children (perspective taking, diverse desires and beliefs, and false belief understanding). For some analyses, we divided boys with FXS into two groups based on autism status (FXS-A and FXS-O).
Results: Data were analyzed using a series of ANCOVAs controlling for nonverbal mental age and receptive and expressive language skills. On the CASL, controls scored significantly higher than children with FXS (p<.005) and autism (p<.0001), and the FXS group scored significantly higher than the autism group (p<.01). During conversation, both groups of boys with autism (autism-only and FXS-A) used significantly more noncontingent (off-topic) language than boys with FXS-O (p<.01; p<.0001, respectively) and controls (p values < .0001). Moreover, boys with autism without FXS used significantly more noncontingent language than even the boys with FXS-A (p<.01). Boys with autism-only and FXS-A showed significantly more impairment in ToM than boys with FXS-O and TD (p values < .05). ToM skills were significantly associated with boys’ pragmatic skills measured through the CASL for boys with autism-only (r=.82, p<.001), FXS-A (r=.65, p<.05), FXS-O (r=.54, p<.05), and TD (r=.72, p<.01).
Conclusions: These findings suggest substantially overlapping phenotypes among children with autism and FXS. Moreover, the similarity in performance compared with controls of the autism-only and FXS-A groups and the lack of discourse and ToM impairments found in the FXS-O group are together striking findings which point to a specific role of autism in the discourse and social cognitive impairments in FXS. Results also suggest that ToM ability, impaired in autism-only and FXS-A, may be a key underpinning of pragmatic language deficits.