Structured world knowledge plays a central role during decision-making in many settings. For example, when choosing ingredients for a sandwich, most people with reasonable experience eating this kind of food are able to make choices that will increase the expected value of the final product. This kind of decision relies on schema-level information about the value of items in this context that has been abstracted over multiple experiences over the course of a lifetime. In contrast, other decisions rely on value information that was learned during a particular episodic experience: For example, remembering the pleasant taste of a particular sandwich topping from last week. Despite the centrality of these kinds of choice problems in real life, little is known about the neural systems involved in sampling and evaluating such information during decision-making. These different forms of memory depend on distinct regions of the brain, and damage or pathology within these areas could relate to specific forms of decision-making impairment in neurodegenerative disease.
The aim of this proposal is to test the involvement of different neural systems in sampling schematic and episodic memory in support value-based decision-making. I hypothesize that two partially overlapping neural systems support sampling memory for value information, and describe two functional neuroimaging experiments that test this hypothesis. The first experiment tests the functional contributions of a few key brain regions in sampling episodic and schematic memory for value information. By explicitly crossing the need for schema-level information and expected value, this experiment can disentangle the involvement of different brain areas in these processes. The second experiment will use representational similarity analysis to test how the brain organizes different categories of stimuli based on their expected value in different contexts. This experiment will examine how such contextual information is used to dynamically determine the values of stimuli. These experiments test key questions at the understudied intersection of value-based decision-making and declarative memory. My hope is that this research will shed new light on the interaction of memory and decision-making systems that may given insight into how decisions are made in ecological settings, with meaningful translational implications to understanding the role of memory impairments in decision behavior.
This project aims to investigate the neural systems involved in sampling memory to support value-based decision-making. Studying this circuitry has immediate relevance for understanding how decision-making may be affected in healthy aging and neurodegenerative diseases, and how to bias retrieval toward adaptive information for goal-directed behavior. This work will bridge research in memory and decision-making, and will make important conceptual advances for these fields.
