PROJECT SUMMARY/ABSTRACT Decisions in complex environments rely on context-sensitive processes for estimating the subjective value of available alternatives. A course of action that would be beneficial in one context might be disadvantageous in another. Context has an especially pronounced impact on how time and delay factor into decisions, with different levels of patience favored in different environments. Understanding the normative cognitive and neural mechanisms that support context-appropriate calibration of decision processes is a critical step toward understanding the seemingly impulsive or dysregulated decisions that occur in mental health conditions. Ventromedial prefrontal cortex (vmPFC) is thought to contribute to integrative, context-sensitive estimates of subjective value via interactions with brain systems involved in memory, motivation, and time estimation. However, person-to-person variability in the anatomy and functional topography of vmPFC has limited our ability to target precise and meaningful cortical subregions in mechanistic studies of its normative role or in studies of clinical populations. The present project seeks to identify and characterize individual-level cortical subregions engaged during value-based decisions that require event-specific integration of temporal and contextual information. The project is a collaboration among investigators with expertise using neuroimaging and computational methods to study value-based decision making, context representation, and memory. Aim 1 will use a foraging-like experimental paradigm in which the value of engaging with a distinctive item depends on an episodic assessment of how recently it was previously encountered, with either recent (“fresh”) or non- recent (“ripe”) items having greater reward value in different contexts. We will use multi-session fMRI with a vmPFC-optimized acquisition protocol to identify individual-specific cortical regions that show activity modulated by recency-dependent value irrespective of physical item features or recency per se. We hypothesize that valuation effects will emerge in vmPFC subregions that are reproducible within-individual and have a systematic profile of functional connectivity with the medial temporal lobe and other brain structures. Aim 2 will develop a theoretical computational model of cognitive processes for flexibly translating episodic temporal information into decision value. We will test and refine the model on the basis of behavioral data and will test for model-predicted representations in our own fMRI data and other existing data sets. Aim 3 will examine situations in which patterns of context-dependent value must be disengaged and reinstated on the basis of feedback in a changing environment, and will use fMRI to examine signals associated with contextual updating in vmPFC and medial temporal lobes. Results from the project will advance our fundamental understanding of brain processes that support contextually adaptiv...