Date of Award
Fall 2023
Document Type
Dissertation
Degree Name
Doctor of Philosophy (PhD)
Department
Interdepartmental Neuroscience Program
First Advisor
Levy, Ifat
Abstract
Rewards and punishments motivate human behavior and guide learning and decision-making processes. Rewarding or punishing feedback leads to updating of our expectations, or subjective values, associated with specific objects or experiences. The construction of subjective value is multi-faceted and can be influenced by various factors, including context, uncertainty, and timing. Many previous studies have predominantly focused on the domain of either rewards or punishments, leaving a significant gap in the literature. Furthermore, these studies have primarily been conducted in the realm of monetary decision making, failing to capture the diverse range of real-life decision outcomes that include diverse categories and are often more difficult to quantify. This dissertation addresses these limitations by examining the learning and decision-making behaviors of human study volunteers across different domains, utilizing behavioral tasks and functional magnetic resonance imaging (fMRI). In Chapter 2, I explore the possible learning phenotypes observed in the general population and discover three generalizable learning profiles for rewards and punishments. Additionally, I describe how an associative learning model can provide insights into the underlying processes driving these learning subtypes. In Chapter 3, I identify brain regions responsible for encoding these learned values. Chapter 4 expands on novel methods to translate conventional economic studies to the real-life domain, where both outcomes and uncertainty around those outcomes are difficult to quantify. I find consistent uncertainty attitudes regarding quantitative monetary rewards and qualitative medical rewards. Together, these results contribute to a deeper understanding of the behavioral and neural foundations of learning and decision making under uncertainty, specifically in the domains of rewards or punishments and quantitative or qualitative outcomes. Moreover, they provide new directions for future investigation.
Recommended Citation
Xu, Chelsea Yi, "Learning and decision making across domains" (2023). Yale Graduate School of Arts and Sciences Dissertations. 1234.
https://elischolar.library.yale.edu/gsas_dissertations/1234
