Behavior is governed by the coordination of multiple learning and memory systems. A core challenge in autism spectrum disorder (ASD) research is to determine which neural processes are disrupted and give rise to abnormal behavior and which are relatively intact and can be leveraged for therapeutic intervention. New computational and neurobiological tools can enable us to disentangle the component cognitive processes and neural circuits and assess their relevance to ASD symptomatology.

Linda Wilbrecht proposes that an inability to appropriately update behavior based on probabilistic feedback represents a core learning deficit in ASD. Furthermore, she proposes that the function and/or development of specific frontal-striatal circuits is altered in ASD and contributes to ASD-associated learning deficits. Wilbrecht’s hypotheses are based on 1) findings that individuals with ASD exhibit impairments in probabilistic reinforcement learning, probabilistic inference, statistical speech perception and social learning (which is probabilistic by nature) and 2) the fact that frontal-striatal circuits are implicated in learning and balancing flexible goal-directed behavior versus inflexible habitual behavior.

Wilbrecht will test these ideas in two genetic mouse models of ASD, by comparing the development and function of two major frontal-striatal cell types that show different trajectories of maturation and that may play different roles in learning and flexible behavior. These experiments will help to determine if ASD risk genes affect specific cell types and/or specific forms of learning more than others. The findings from this work may also highlight particular aspects of learning that could be targeted for functional behavioral interventions.