Grant: Project Grant

Deficits in decision-making are a major aspect of many brain disorders. For example, in addition to its motor symptoms, Parkinson's disease often also affects decision-making. Worse still, treatments for the motor symptoms often have side effects leading to decision-making problems including impulsivity and addiction. Understanding the neural bases of decisions is therefore critical to developing more effective therapies that act at the sources of the deficit without disturbing other brain functions. Traditionally, research on decision-making treated it as a purely cognitive process based on abstract representations of costs and benefits. However, growing neurophysiological data suggests that many decisions involve a distributed network of brain regions including those implicated in controlling actions. This makes sense because during the brain's evolution, animal behavior was dominated by decisions about actions in an unpredictable and constantly changing world, such as selecting what to do when searching for food or when faced with a threat. In such natural scenarios, the options themselves as well as their payoffs and risks are constantly changing, and decisions must be made even during ongoing behavior. To understand a brain whose evolution was dominated by the demands of such natural scenarios, we need to study how it deals with real-time, dynamic decision-making. This project will investigate, through a combined experimental and computational approach, how the brain combines sensory information to influence a competition between different potential reaching actions, how it arrives at a commitment to a specific movement, and how it learns from the outcomes of performing that action. The results will lead to a computational framework that can provide a mechanistic link between fundamental mechanisms of adaptive behavior, inter-individual differences in decision-making policies, and the kinds of deficits seen in a variety of brain disorders.