Valence-partitioned learning signals drive choice behavior and phenomenal subjective experience in humans
bioRxiv Preprint Server March 17, 2023 L. Paul Sands, Angela Jiang, Rachel E. Jones et al. 4 citations preprint
A hypothesized neurocomputational model, valence-partitioned reinforcement learning (VPRL), maintains separate tracks for appetitive and aversive information, generating independent reward and punishment learning signals. This model predicts changes in human choice behavior, subjective experience, and brain activity in regions including the ventral striatum and ventromedial prefrontal cortex during introspection. The findings suggest that valence-partitioned reinforcement learning provides a neurocomputational basis for investigating mechanisms that may drive conscious experience.