Neurobiological Foundations of Stability and Flexibility

In order to adapt to changing and uncertain environments, humans and other organisms must balance stability and flexibility in learning and behavior. Stability is necessary to learn environmental regularities and support ongoing behavior, while flexibility is necessary when beliefs need to be revised or behavioral strategies need to be changed. Adjusting the balance between stability and flexibility must often be based on endogenously generated decisions that are informed by information from the environment but not dictated explicitly. This dissertation examines the neurobiological bases of such endogenous flexibility, focusing in particular on the role of prefrontally-mediated cognitive control processes and the neuromodulatory actions of dopaminergic and noradrenergic systems. In the first study (Chapter 2), we examined the role of frontostriatal circuits in instructed reinforcement learning. In this paradigm, inaccurate instructions are given prior to trial-and-error learning, leading to bias in learning and choice. Abandoning the instructions thus necessitates flexibility. We utilized transcranial direct current stimulation over dorsolateral prefrontal cortex to try to establish a causal role for this area in this bias. We also assayed two genes, the COMT Val158Met genetic polymorphism and the DAT1/SLC6A3 variable number tandem repeat, which affect prefrontal and striatal dopamine, respectively. The results support the role of prefrontal cortex in biasing learning, and provide further evidence that individual differences in the balance between prefrontal and striatal dopamine may be particularly important in the tradeoff between stability and flexibility. In the second study (Chapter 3), we assess the neurobiological mechanisms of stability and flexibility in the context of exploration, utilizing fMRI to examine dynamic changes in functional brain networks associated with exploratory choices. We then relate those changes to changes in norepinephrine activity, as measured indirectly via pupil diameter. We find tentative support for the hypothesis that increased norepinephrine activity around exploration facilitates the reorganization of functional brain networks, potentially providing a substrate for flexible exploratory states. Together, this work provides further support for the framework that stability and flexibility entail both costs and benefits, and that optimizing the balance between the two involves interactions of learning and cognitive control systems under the influence of catecholamines