Theory of Mind in Autism:

How Comparisons with Down Syndrome Have Changed

Premack and Woodruff's seminal study in 1978 introduced the concept of theory of mind (ToM), revolutionising our understanding of cognition, social behaviour, and interpersonal relationships. Their article, provocatively titled ‘Does the chimpanzee have a “Theory of Mind”?’, posed a fascinating question that inspired decades of research into the development of theory of mind in humans.

Baron-Cohen, Leslie, and Frith (1985) extended this line of inquiry by reframing Premack and Woodruff’s question to focus on autistic children: ‘Does the autistic child have a “Theory of Mind”?’ Using a model of metarepresentational development, they hypothesised that autistic children lack this capacity, preventing them from attributing beliefs to others or predicting others’ behaviour.

To test their hypothesis, the researchers employed Wimmer and Perner's puppet play paradigm, commonly known as the “Sally-Ann test.” Their study compared autistic children to typically developing children and children with Down syndrome as control groups. The findings revealed that, even though the mental age of the autistic children was higher than that of the controls, they alone failed to impute beliefs to others. This deficit was interpreted as a specific cognitive impairment independent of general intellectual ability, as children with Down syndrome—despite more severe intellectual impairments—performed near ceiling on the task. Specifically, 80% of autistic children exhibited impairments in the lowest level of ToM (i.e., first-order belief attribution) compared to non-autistic mentally handicapped controls.[1]

Baron-Cohen (1989) extended this research to the 20% of autistic children who passed first-order belief tasks, examining their ability to attribute second-order beliefs. Compared to typically developing children and children with Down syndrome, these autistic children were severely impaired at the higher level.

More recent research challenges the original comparison between autistic children and those with Down syndrome. Shojaeian et al. (2022) investigated the role of IQ and ToM understanding in children with an autism spectrum disorder (ASD), Down syndrome, and typically developing controls, aged 6–12. Across ToM tasks, ASD children demonstrated better first-order ToM understanding than those with Down syndrome. While typically developing children consistently outperformed both clinical groups, there was no significant difference between ASD and Down syndrome groups on second-order tasks. However, on third-order tasks, children with ASD performed significantly better than those with Down syndrome.

Could it be possible that one explanation for this is the narrower diagnosis of autism in the 1970s-1980s, as compared to the broader diagnosis of autism spectrum disorder in current diagnostic classifications? During that time, it was believed that the majority of autistic children were mentally retarded[1] (DeMyer et al., 1974; Wing et al. 1976). This shift in diagnostic criteria may explain the observed differences in ToM performance across decades.

_______________________

[1] This text uses terminology reflective of the time