Counterfactual Reasoning and False Belief Understanding in Children with Autism

The Sally-Anne task used to assess children’s understanding of false belief has traditionally been conceptualized as a test of mental state understanding in that it asks the child where a protagonist thinks an object is located when the protagonist has a false belief about the object’s location. However, a recent logical analysis by Peterson and Riggs identifies a strategy for such tasks involving a specific reasoning process they term subtractive reasoning. This can be assessed by asking the child a question such as, ‘If the marble had not been moved, where would it be now?’ Studies of typically developing children have shown strong associations between false belief and subtractive reasoning tasks even when verbal mental age is controlled. In the present study we replicated these experiments using children with autism and children with severe learning difficulties. Although significant correlations between the two tasks were found for all three groups, analyses of contingencies between the two tasks and comparison of their respective difficulty for each group suggested that ability in subtractive reasoning was a necessary but not sufficient component of successful performance in the false belief tasks. Our results indicate the presence of a further factor which is required in these tasks, and which is deficient in autism, and we argue that this may consist in a specific type of generativity.