An Exploration of the Phenotypic and Etiological Relationships Between Autism Spectrum Disorder and Specific Language Impairment

Background: While Autism Spectrum Disorder (ASD) and Specific Language Impairment (SLI) have traditionally been considered as distinct disorders, evidence is emerging which indicates that the boundaries between the two conditions are not clear-cut. Specifically, findings that a subgroup of children with ASD has structural language difficulties similar to those observed in SLI have led some authors to claim that there is substantial aetiological overlap.

Objectives:  Much of the research to date has focused on overlap in the language phenotypes of ASD and SLI and no previous study has investigated potential cognitive overlap. Therefore, the aim of this study was to examine possible points of cognitive similarity in children with ASD, children with SLI, and the parents of both groups of children.

Methods: 32 Children with an ASD (18 with ‘normal’ and 14 with ‘impaired’ language), 19 children with SLI and 61 typically developing (TD) children completed tasks that tapped phonological working memory (nonword and sentence repetition), a known deficit in SLI, and tasks that measured cognitive characteristics of ASD, namely ‘weak central coherence’ (assessed by the Children’s Embedded Figures Test: CEFT) and emotion recognition ability.

We also examined possible shared heritable risk factors for ASD and SLI. 24 ASD parents, 12 SLI parents and 20 parents of TD children completed measures sensitive to the broader SLI (nonword and sentence reptition) and broader ASD (Embedded Figures Test, emotion recognition task) phenotypes.

Results:  ‘Language impaired’ children with ASD and children with SLI performed worse than both ‘language normal’ ASD and TD children on the nonword and sentence repetition tests. In addition, the SLI children performed worse than all other groups on the CEFT. For the emotion recognition task, all clinical groups were less accurate than the TD children across visual and auditory domains. The only significant group differences that emerged for the parents were on the nonword repetition task (ASD parents were worse than the SLI and the TD parents) and the emotion recognition task (ASD parents were faster than the SLI parents and less accurate than the TD parents).

Conclusions: These findings point to areas of overlap, as well as points of divergence in the cognitive profiles of ASD and SLI, and the heritable phenotypes of these conditions. The results have implications for the debate about aetiological overlap in ASD and SLI.