Social Environment Influences on Mental State Understanding in Children with or without Autism

Background:  How much the development of mental states understanding benefits from environmental input is a controversial issue. In favor of such effects stand repeated findings showing that having an older siblings helps performance in classical theory of mind tasks (Perner, Ruffman & Leekam, 1994). This is believed to happen because the presence of older siblings creates more opportunities for confronting beliefs and reality and for parental intervention to discuss mental states or intentions. Such environmental effects are welcome in populations that struggle with understanding mental states, as is the case with children with autism. 

Objectives:  We investigated the impact of having a typically developing sibling or not on performance in an implicit false belief task (Southgate, Senju & Csibra, 2007) in children with or without ASD-like social and communicative difficulties. Because some of these children had developed autism themselves yet others were developing typically we could observe the contribution of both social environment (siblings status) and clinical profile on performance in this task.  

Methods:  Forty-seven 3-year-old children participated in this task and either had a diagnosis of ASD (Sib- ASD, n=17), manifested subclinical ASD-like characteristics (Sib-Other, n=12) or were developing typically (Sib-TD, n = 18). All children had an older sibling with ASD. Children were classified as either having no typically developing siblings or having more than one sibling, which in this case meant that they also had typically developing siblings. The false belief task was similar to Sally-and-Ann with the difference that instead of asking where the character would search for an object we measured children’s anticipatory looking towards the location where the character would search, using an eye-tracker (Tobii 120).

Results:  Because Sib-ASD and Sib-ATY performed similarly we analyzed them as one group (Sib-Aty, n=29).  Group (Sib-Aty and Sib-TD) and Number of TD siblings (None, One or more) were entered in a logistic regression. The model including the interaction between Group and Siblings best predicted performance (Rsq = .139; Group x Siblings b = -3.4, sig = .025). This was due to a significant effect of the number of siblings on performance for the Sib-TD group (b = 2.5, sig = .04), only Sib-TD with a TD sibling performing above chance. The number of TD siblings was non-consequential for Sib-Aty, which performed equally poorly.

Conclusions:  It was speculated that implicit measures of mental state understanding do not depend on language proficiency, therefore they might be less prone to environmental influences of the kind that would be created when interacting with siblings. We found that having an older typically developing sibling does improve performance for typically developing children but, interestingly, not for children that manifested themselves clinical or sub-clinical ASD like symptoms. The environmental enrichment provided by siblings is therefore mediated by their phenotypic profile, the  ASD phenotype making children less likely to benefit from sibling social interaction. Future studies should investigate whether children with ASD require longer time to show beneficial environmental effects.