20152
The Role of Language in Second-Order Theory of Mind Reasoning in Children with Autism Spectrum Disorder and Children with Optimal Outcomes
Theory of mind (TOM) refers to the ability to attribute different mental states to oneself and others. While difficulties with TOM reasoning have been found in children with autism spectrum disorder (ASD), studies have also found a subgroup that can pass TOM tasks (Happé, 1995). Language has been proposed as a potential factor in explaining this variability. However, some studies have found a relationship between lexical knowledge and children with ASD’s TOM reasoning (Happé, 1995), while others have found syntactic knowledge to be important (Paynter & Peterson, 2010). Many of these studies have focused on first-order TOM reasoning; the relative importance of lexical and syntactic knowledge on second-order TOM in children with ASD is unclear.
Objectives:
The study investigates the role of language (lexical and syntactic knowledge) on second-order TOM reasoning in children with ASD. Additionally, we compare high-functioning children with ASD (HFA) and those children with a previous history of autism, but who no longer exhibit any symptoms of ASD (i.e., optimal outcome children, OO, Fein et al., 2013).
Methods:
Twenty-three typically developing (TD) children, 27 children with HFA, and 26 children with OO, matched on age (M=13.24, S.D.=2.90 years) and non-verbal IQ (Table 1) participated. Two second-order TOM tasks involving short stories similar to those in Perner and Wimmer (1985) were administered. Each included nonlinguistic and/or linguistic control questions, followed by an ignorance question (IG), false belief question (FB), and a justification question. The CELF-4 and PPVT were also administered to gather standardized language measures.
Results:
Groups did not differ on control items (p=.11) but differed on their performance of IG, FB, and justification questions (ps<.05; Figure 1). Post-hoc comparisons revealed that the HFA group performed significantly more poorly than the TD group on all three TOM questions (ps<01). Performance by the HFA group did not differ from that of OO group on any TOM questions (ps>.05). OO group differed from the TD group only in performing more poorly on justification questions (p<.001).
No significant correlations were found between performances on any TOM questions and scores on CELF-4 subtests examining syntax (ps>.05) for any groups. For the TD and HFA groups, PPVT scores did not correlate with TOM performance (ps>.05). For the OO group, performances on the three TOM questions were significantly positively correlated with PPVT scores (ps<.01, r’s>.40). Additionally, performance on IG questions was positively correlated with CELF-4 receptive word class subtest scores (p<.05, r=.49).
Conclusions:
Consistent with previous literature, some children with HFA continued to have difficulties with TOM reasoning. While children with OO could engage in second-order false belief and ignorance reasoning, difficulties with providing correct justifications for false beliefs remain. Thus, despite no longer exhibiting ASD symptomology, residual difficulties from their previous ASD diagnosis may remain. Lastly, correlational analyses between the OO group’s TOM and language measures suggest that at least for a subgroup of children, lexical but not syntactic knowledge is related to their second-order TOM reasoning. This may reflect the use of a unique compensatory verbal strategy to solve TOM tasks.