Advanced Theory of Mind Assessment in Adults with High-Functioning Autism

Background:  Theory of mind (ToM) is the ability to attribute mental states to one’s self (first-order) & others (second-order) in order to explain and predict behavior.  Typically developed by age 11, this cognitive construct is substantially delayed in children with autism.  There is conflicting data as to whether ToM deficits are present in adults with autism.   This discrepancy has been attributed to the difficulty level of ToM task used, as adults with high-functioning autism (HFA) have been able to pass simple ToM measures developed for children while continuing to experience difficulty in social interactions that require understanding the beliefs and intentions of others.  However, there are few studies that employ the use of advanced ToM measures in adult populations, particularly in adults with autism.  

Objectives:  To compare ToM in adults with high-functioning autism using two advanced theory of mind assessments.  

Methods:  Five adults with HFA (ages 19-40, 80% male) completed a battery of cognitive assessments.  Autism diagnosis was confirmed by ADI-R and ADOS.  Theory of mind measures included the revised adult version of Reading the Mind in the Eyes and Strange Stories Test as modified by Fletcher et al. 1995 and White et al. 2009.  Language abilities, which have been associated with ToM performance, were assessed using the Comprehensive Assessment of Spoken Language (CASL).  

Results:  Despite having normal IQ, participants’ performance on both ToM tasks showed impairment compared to general population norms.  Mean score on the Reading the Mind in the Eyes test in this sample was below that of normal controls and the HFA/Asperger population.  Likewise, mean scores on the modified Strange Stories were lower in this sample than the reported means for normal controls (Fletcher et al. 1995, White et al. 2009) and other adults with autism (White et al. 2009) for all story types (mental, physical, and unlinked sentences).  The pattern of performance on Strange Stories was consistent with reported findings for normal controls, with average scores highest on mental state stories and lowest on the unlinked sentences control task.  Scores on the CASL revealed an overall weakness in general language skills. The strongest deficit was seen in supralinguistic tasks which included tests of non-literal language, ambiguous sentences, meaning from context, and inference.

Conclusions:  Adults with HFA scored significantly below normal controls on both advanced ToM measures in this study, which indicates that the revised adult Reading the Mind in the Eyes and modified Strange Stories tests are sensitive enough to identify ToM deficits in this population.  While participants in this study had lower performance than normal controls and other autism populations, their pattern of results was more consistent with normal controls than other individuals with autism.  Greater sample sizes are needed in studies using advanced ToM measures to determine the prevalence and areas of ToM deficits in this population and to assess the value of language skills as a predictor of ToM performance.