Functionally Dissociable Influences on Learning Rate in a Dynamic Environment

• Published in: Neuron (Cambridge, Mass.) Vol. 84; no. 4; pp. 870 - 881
• Main Authors: McGuire, Joseph T., Nassar, Matthew R., Gold, Joshua I., Kable, Joseph W.
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 19.11.2014; Elsevier Limited
• Subjects: Adolescent; Adult; Behavior; Brain - physiology; Brain Mapping; Environment; Female; Functional Neuroimaging; Humans; Image Processing, Computer-Assisted; Learning - physiology; Magnetic Resonance Imaging; Male; Medical imaging; Reward; Uncertainty; Young Adult
• ISSN: 0896-6273; 1097-4199; 1097-4199
• DOI: 10.1016/j.neuron.2014.10.013

Maintaining accurate beliefs in a changing environment requires dynamically adapting the rate at which one learns from new experiences. Beliefs should be stable in the face of noisy data but malleable in periods of change or uncertainty. Here we used computational modeling, psychophysics, and fMRI to show that adaptive learning is not a unitary phenomenon in the brain. Rather, it can be decomposed into three computationally and neuroanatomically distinct factors that were evident in human subjects performing a spatial-prediction task: (1) surprise-driven belief updating, related to BOLD activity in visual cortex; (2) uncertainty-driven belief updating, related to anterior prefrontal and parietal activity; and (3) reward-driven belief updating, a context-inappropriate behavioral tendency related to activity in ventral striatum. These distinct factors converged in a core system governing adaptive learning. This system, which included dorsomedial frontal cortex, responded to all three factors and predicted belief updating both across trials and across individuals. •Effective inference in a dynamic environment requires adaptively weighing new inputs•We decomposed this complex process into both task-relevant and incidental factors•Individual factors were represented in distinct brain networks measured via fMRI•These distinct processes converge on a core system that governs adaptive inference Maintaining accurate beliefs in a complex environment requires adapting the rate at which one learns from new experiences. McGuire et al. identify three computationally separable factors influencing learning rate and link these factors to both dissociable and shared brain mechanisms.