Advanced Theory of Mind in Autism Spectrum Disorder: From Childhood to Adolescence

Background: Age is an important factor in the development of Theory of Mind (ToM) abilities in school-aged children with high functioning autism spectrum disorders (HFASD, Wellman, 1991). However, the development of advanced ToM capacities (Happé, 1994) from childhood to adulthood has been largely neglected in studies on HFASD. In this study we explore the age related change in advanced ToM with a large cross-sectional design (n=175), aiming to assess the validity and sensitivity of advanced ToM tests across a wide age range and examining whether ToM development shows a stagelike or linear change. Furthermore, we will delineate to what extent the advanced ToM test is associated with the severity of the autistic disorder.  
Objectives: Our goal is to explore the influences of age on the advanced Theory of Mind abilities of a large sample of children and adolescents with HFASD and to examine the predictive value of advanced Theory of Mind abilities for the severity of autistic symptoms.

Methods: Participants were 175 children and adolescents with HFASD (aged 6 to 18 years). ADOS scores were obtained from all participants. A five stories version of the Stories of Everyday Life (Kaland, 2002) was used to examine whether the participants were able to infer the intentions or feelings of story characters. The test included stories on second order false belief, emotional display rules, double bluff, violation of a social rules, and irony. Severity of autism symptoms was assessed with the Social Responsiveness Scale (SRS). Verbal IQ was measured with the Peabody Picture Vocabulary Test-III.

Results: Preliminary linear regression analyses on a subsample of participants indicated that when both age and verbal IQ were entered as predictors of Theory of Mind abilities, only age had a (nearby significant) predictive value. This effect became significant when age was entered as a single predictor (β = 0.37, p < .05, n = 37). Despite the wide age range of 6 to 18 years, no floor or ceiling effects were found in performances on the advanced Theory of Mind test. Surprisingly, Theory of Mind abilities were not predictive of symptom severity as indicated by scores on the Social Responsiveness Scale.

Conclusions: Preliminary results suggest age is an important predictor of advanced Theory of Mind abilities. Generally, adolescents perform better than school-aged children on an advanced Theory of Mind test. Yet, there must be other background factors that influence advanced Theory of Mind abilities, because large variance was noted within the age groups. These will be explored in the analysis of the full dataset. Furthermore, the low predictive value of advanced Theory of Mind abilities for symptom severity seems to indicate that as such advanced Theory of Mind abilities are of limited value for understanding social limitations within the group of children and adolescents with HFASD.