Disentangling the Neural Underpinnings of Risk and Reward in Human Decision Making

Risk-taking is a fundamental human behavior subserved by separable cognitive processes. Understanding how these processes are represented in the brain offers critical insights into decision-making, development, and vulnerabilities to psychopathology. However, functional magnetic resonance imaging (fMRI) research often yokes risk and reward processes during risky decision-making and has limited sensitivity to deep subcortical regions, constraining our contributions. In this study, we present a modified Balloon Analogue Risk Task completed under a multi-echo fMRI protocol meant to enhance subcortical signal. Forty-eight participants inflated virtual balloons across three conditions: risky reward, guaranteed reward, and neutral. GLM analyses revealed increased signal in orbitofrontal cortex, anterior insula (AI), striatum and a brain stem nuclei, ventral tegmental area (VTA), during risky versus guaranteed rewards decision-making conditions. Multivariate analysis identified the AI as a key predictor of the risk condition, surpassing striatal and VTA contributions. These results suggest that neural response to reward-based decision-making is heightened under risk and illuminates putative neurobiological mechanisms which uniquely subserve risk separate from general reward processing. The study also provides a new tool to enhance the resolution of human neuroscience research on risk-taking across the lifespan and vulnerabilities to psychopathology.