This study aims to investigate the role of fronto-striatal circuits and cognitive control in the perseverative and inflexible behavior that is a defining feature of autism. We hypothesize that deficits in the development of fronto-striatal circuitry may underlie cognitive inflexibility in autism. Specifically, we hypothesize that repetitive, inflexible behavior arises as (1) fronto-striatal systems are capable of learning patterns present in the environment (as in implicit learning paradigms), but are unable to adapt behavior to changing circumstances, related to either (2) decreased ability of basal ganglia to detect violations of expectancy, (3) decreased ability of prefrontal cortex to respond to detected violations, or (4) decreased connectivity of the circuits. We are conducting three functional Magnetic Resonance Imaging (fMRI) studies to address these hypotheses.