Verbal problem-solving, executive functioning and language development in autism spectrum disorders
Alderson-Day, Benjamin David
Day, Benjamin David Alderson
Autism spectrum disorders (ASDs) are primarily defined by problems with social interaction and communication, but they are also associated with a complex cognitive profile. One area of difficulty for children and adults with ASD is problem-solving, or the process of identifying a solution to a puzzle or question where the answer is hidden. This can be seen on the Twenty Questions Task (TQT), a commonly-used measure of verbal problem-solving and executive functioning. Children with autism are consistently less efficient than typically-developing children in their questioning on the task: for instance, rather than ask a general, category-based question (e.g. “Is it a living thing?”) they may ask about single items (“Is it the dog?”) or very restricted groupings (“Is it something you wear on your feet?”). This has previously been interpreted as an example of a concept formation deficit in autism, deriving from underlying difficulties with complex and integrative information processing. However, success in problemsolving relies on a number of cognitive and linguistic processes that may be impaired in ASD. This thesis attempts to identify which of these may better explain autistic problem-solving performance, using the TQT as a specific example. The first experiment presented here examines the role of executive functioning difficulties in this profile. The performance of 22 children with ASD and 21 age- and IQ-matched typically-developing (TD) children was compared on a version of the TQT adapted to assess planning skills prior to problem-solving and selective attention during the task. Compared to controls, ASD participants were less efficient in their planning of questions, although not all ASD participants had difficulty constructing a plan. No specific effects of selective attention were evident. The second and third experiments explore the importance of atypical language development to this profile, using the example of deafness. Experiment 2 compares the performance of deaf (n = 9) and hearing (n = 27) adults on the TQT, replicating prior evidence of less efficient problem-solving in deaf graduate students. Experiment 3 contrasts TQT performance in 13 deaf schoolchildren with the ASD and TD data acquired in experiment 1. Like ASD children, deaf children were less efficient in their questioning than TD participants, even when controlling for cognitive ability differences. Both autism and deafness are associated with delays in early language development, whereas Asperger Syndrome (AS) is not. To test whether language delay explains autistic problem-solving difficulties, experiment 4 compares TQT performance in 15 children with autism, 15 AS children and 15 age- and IQ- matched typically-developing controls. Participants with autism asked less efficient questions than both AS and TD participants, between whom no differences were observed. This suggests that the problem-solving profile in autism may be better explained as a consequence of atypical language development, rather than other aspects of information processing or executive dysfunction.