| To behave adaptively animals must maintain an internal representation of the value of competing alternatives and link that representation to the neural processes responsible for decision making and motor planning. The neural substrates of these evaluative and decision making processes have recently attracted attention as targets for neuroscientific investigation, motivated by the hope that understanding these underlying mechanisms may help illuminate the pathophysiological basis of various cognitive disease states. Despite this growing interest, however, our understanding of how the brain represents the 'value' of an object or action remains rudimentary.;Here we develop an approach to explore the neural basis of value-based decision making that involves three closely interrelated parts. First, and central to the endeavor, is a decision making task that is complex enough to require subjects to track the values of alternatives in a dynamic and uncertain environment, while simple enough to yield robust and reproducible behavior. Second, is a quantitative behavioral model that accurately predicts individual subject behavior in the context of this foraging task. Third, is an analytic approach that leverages the key internal variables of this quantitative model as probes with which to search for and interpret neurophysiological and brain imaging data collected from subjects engaged in the task.;This dissertation describes how we have successfully applied this approach in conjunction with single unit electrophysiological recordings in awake behaving monkeys to isolate the specific contributions of different brain areas to value based decision making. Specifically, we report the existence of a spatial representation of the value of competing behavioral alternatives in the parietal cortex of the rhesus monkey that is appropriate for selecting the animal's next operant response. We further report complementary signals within orbitofrontal cortex (OFC) that are not selective for the specific sensory or motor characteristics of the alternatives on offer, but instead reflect the process of updating the value of the chosen alternative to reflect the outcomes of ongoing behavior. These OFC signals suggest a role for midbrain dopamine neurons in constructing the representation of value that guides behavior in this task, as well as a possible neural basis for the marginal value theorem of animal foraging; we discuss how these ideas can be tested in future experiments. |
