Developmental Change In Theory of Mind: Late Onset, Yet Normal Rate of Development

Background: Previous studies have found that individuals with autism spectrum disorders (ASD) have specific deficits and delays in theory of mind (ToM) development. However, few studies have systematically compared ToM development across ASD subtypes. Results regarding ToM development have been contradictory; this may be because individuals of different functioning levels were seen, or because only a limited aspect of ToM ability was assessed. 

Objectives: Using a comprehensive ToM battery, we tested whether development is evident in high and low-functioning autism (HFA and LFA), Asperger's syndrome (AS) and pervasive developmental disorder not otherwise specified (PDD-NOS).

Methods: This study was a cross-sectional comparison of 87 children with ASD (20 LFA, 25 HFA, 22 AS and 20 PDD-NOS) and 30 typically developing (TD) children, aged 4-16 years. Children were tested on 11 ToM tasks ranging in difficulty from imaginary play to interpreting complex social situations, and on several measures of cognitive function. (1) To compare overall group performances on ToM, an analysis of covariance was done with verbal IQ, Verbal Generativity, Inhibition/Set Shifting, Digit Span and Processing Speed as covariates. (2) To explore the development of ToM ability, five separate regressions were performed (TD, LFA, HFA, AS, PDD-NOS), and the rate of development and onset of ToM was compared.

Results: (1) Even after controlling for possible confounding cognitive factors, significant differences in ToM remained between the groups, F (4, 69) = 9.02, p < .001. Of particular interest here is the large difference in ToM between the HFA and AS groups (M_HFA = 63.95, M_AS = 112.38; p < .001). (2) Regarding ToM development, the HFA, PDD-NOS and TD groups showed statistically significant increases in ToM with age (all p <.01). The groups also did not differ in their overall rate of ToM development, F (4, 107) = 2.01, p = .098. However, the LFA group showed significantly slower ToM development than the TD group, F (1, 46) = 7.39, p = .009. All the ASD groups except for the AS group showed delayed ToM skills at age 48 months, F (4,111) = 63.53, p < .001. ToM onset was especially delayed in HFA and LFA.

Conclusions: These results support a delayed development hypothesis: ToM does develop in certain ASD groups, and seems to develop at a normal rate, but the onset of the ToM abilities measured in this study is much later than in typical development. The fact that the LFA group showed no increase in ToM with age suggests that certain language and cognitive skills may be necessary for ToM development, or that compensatory skills play a large role in the observed increases in ToM in the other ASD groups. The differences found in ToM are pertinent to the HFA/AS debate, and highlight that some distinction between different levels of ability needs to be made for teaching and intervention purposes if these diagnostic categories are to be grouped into one, and that ToM may be a good way to do so.