The proposed study investigates the hypothesis that the deficit people with autism show on the Wisconsin Card Sorting Test is driven by an inability to use feedback from other people to motivate correct responding, rather than by the executive dysfunction proposed in the executive function theory of the core deficit in autism. The executive function theory, like other explanations of the cause(s) of autism, does not take into account the potential effects of deficits in other areas of functioning on the tasks used to test it. Indeed, the Wisconsin Card Sorting Test, one of the main tasks used in assessing executive function deficits, requires the test-taker to use verbal feedback delivered by the examiner to guide responding. Given the significant social deficits in autism, this task requirement may in fact underlie the difficulties people with autism have on this and similar executive tasks. Specifically, this study will examine whether the properties of the feedback given in tasks significantly alters the performance of people with autism. Two feedback properties will be manipulated on the Wisconsin Card Sorting Test and a nonverbal paired associates memory task in a 2x2 design: the method of feedback, delivery (by, another person versus by a computer), and the motivational value of the feedback (right/wrong versus giving/taking away a dime). The performance of 40 individuals with autism, ages 8-55 years, will be compared to that of 40 individuals without autism who will be matched to participants with autism on chronological age, receptive language ability, and gender. It is hypothesized that the performance of the people with autism on the two tasks will improve under the nonsocial feedback delivery conditions and the higher value conditions, but that the manipulation of delivery will have a greater impact on performance. It is further hypothesized that the comparison group will be affected by the value manipulation, but not the method of delivery manipulation.
