Is Early Joint Attention Associated with School-Age Pragmatic Language Among Children with ASD, Infant Siblings of Children with Autism and Low-Risk Controls?

Background:  

Are longitudinal associations between joint attention and structural language (systems of meaning such as syntax or the lexicon; Mundy & Gomes, 1998; Sigman & Ruskin, 1999) attributable to social or representational mechanisms? Reduced joint attention is a core symptom of ASD early in development (Sigman & Ruskin, 1999). While structural language is not always impaired in ASD, atypical pragmatic language (socially appropriate use of language) is characteristic of ASD (Bishop & Baird, 2001; Kjelgaard & Tager-Flusberg, 2001) and the Broader Autism Phenotype (BAP; Bishop et al., 2006; Losh and Piven, 2007).  Both joint attention and pragmatic language may arise from social knowledge. Alternatively, rather than emerging from social understanding, the parallel and distributed processing (PDP) model suggests that joint attention emerges from increasing ability to represent triadic relations  (Mundy and Jarrold; 2010).

 Objectives:

The primary aim of this study was to examine longitudinal relations between joint attention at 18 months and aspects of language that are primarily social (pragmatic language) or primarily representational (structural language) 6 years later in order to evaluate evidence in favor of the social-cognitive (Tomasello, 1995) or the PDP (Mundy & Jarrold, 2010) model of joint attention. A secondary aim was to determine if communicative difficulties were apparent among siblings who did not develop autism.

Methods:  

We examined longitudinal relations between joint attention at 18 months and pragmatic and structural language approximately 6 years later among children with ASD (n =11), siblings of children with ASD who did not develop autism (n =30), and low-risk controls (n=20). Initiation and response to joint attention (IJA and RJA) were assessed with the Early Social Communication Scales (Mundy et al., 2003). A parent-report measure of pragmatic language, the Children’s Communication Checklist-2 (CCC-2), was used (Bishop, 2003b). Structural language was assessed with both the CCC-2 and the Clinical Evaluation of Language Fundamentals-4 (CELF-4; Semel et al., 2003).

Results:  

IJA, but not RJA, was associated with structural (rs(9) = .769, p = .006), but not pragmatic, language on the CCC-2 among participants with ASD. IJA (rs(8) = .667, p = .031), but not RJA, was associated with receptive, but not expressive, language on the CELF-4 for children with ASD. No relations between joint attention and language measures were observed for high- and low-risk children without ASD. Children with ASD exhibited reduced RJA (p = .002), pragmatic language (p= .008), expressive (p = .024) and receptive (p = .001) language. High-risk participants without ASD did not differ from low-risk participants.

Conclusions:  

Despite literature linking joint attention and pragmatic language concurrently among children with ASD (Loveland & Landry, 1986), relations between joint attention and both parent-report and observational measures of structural, but not pragmatic, language were observed.  No evidence of BAP related communicative difficulties was apparent. This study, perhaps the first infant sibling study to assess school-age outcomes in ASD, provides support for the PDP theory of joint attention by demonstrating that commonly observed longitudinal associations between joint attention and later language may arise from representational rather than social aspects of both.