Understanding the neurobiological basis of mixed-strategy decision-making in health and disease
During competitive interactions involving multiple agents, such as sporting competitions, the outcome of each individual’s actions is dependent not only on a historical sequence of their own choices, but on those of their opponents as well. Success in such rivalries often requires that individuals adopt a mixed-strategy wherein available actions are chosen dynamically and unpredictably. Action selection in mixed-strategy environments involves the coordination of widespread neural processes, spanning the cognitive, emotional, and limbic domains. More specifically, choice selection involves working memory, valuation and reward processes, reinforcement learning, and execution of motor responses. The three studies in this thesis investigated the neurobiological mechanisms involved in choosing in mixed-strategy environments, and how these processes change throughout the course of neurodegenerative and neuropsychiatric disease. In the first study, we examined the human brain network underlying mixedstrategy decision-making using functional magnetic resonance imaging (fMRI). Using a carefully controlled paradigm, we compared the network underlying strategic decisions to that involved in choosing in nonstrategic environments. This study allowed us to gain insight into the specialized cognitive circuitry involved in choosing dynamically within a strategic context. In the second study, we asked how dopaminergic transmission affected mixed-strategy decisionmaking by investigating how degeneration of the dopamine system in Parkinson’s disease (PD) affected strategic choice patterns. Further, we investigated the hypothesis that cognitive function is deleteriously affected by dopaminergic medication (levodopa and dopamine agonists) in patients with PD, and whether genetic mechanisms controlling dopaminergic transmission could explain susceptibility to medicationinduced cognitive deficits. ii In the final study, we examined mixed-strategy decisions in patients with borderline personality disorder (BPD), a heterogeneous disorder prevalent in adolescent populations. We explored how the core clinical features of BPD, namely impulsivity and emotional dysregulation, affected mixed-strategy decisions in adolescents showing the first signs of BPD. Together, these studies provide a detailed account of how mixed-strategy decision-making is implemented in the brain, and how the cognitive functions required for choosing in such environments become affected throughout the course of neurological illness. This work, as a whole, represents a significant contribution to our understanding of how the brain integrates diverse sources of information to execute mixed-strategy decisions.