| Paper: | PS-1B.33 |
| Session: | Poster Session 1B |
| Location: | Symphony/Overture |
| Session Time: | Thursday, September 6, 18:45 - 20:45 |
| Presentation Time: | Thursday, September 6, 18:45 - 20:45 |
| Presentation: |
Poster
|
| Publication: |
2018 Conference on Cognitive Computational Neuroscience, 5-8 September 2018, Philadelphia, Pennsylvania |
| Paper Title: |
Neurons of the prefrontal cortex encode a representation of a Bayesian belief during reinforcement learning |
| Manuscript: |
Click here to view manuscript |
| DOI: |
https://doi.org/10.32470/CCN.2018.1269-0 |
| Authors: |
Ramon Bartolo, Richard Saunders, Andrew Mitz, Bruno Averbeck, National Institutes Of Health, United States |
| Abstract: |
In a noisy environment, learning requires organisms to keep track of choices and their associated outcomes across successive decisions to form beliefs about value in the world. This form of reinforcement learning allows them to predict future outcomes and to update their belief after each outcome. The primate prefrontal cortex (PFC) integrates information carried by reward circuits, in addition to its role in working memory. Also, it has been suggested that the prefrontal cortex plays a critical role in inferring the current state of the world under some level of uncertainty. In the present study, we explore the PFC computations related to updating current beliefs in multifaceted reward environments. We conducted high-channel count single-unit recordings in two male macaques while they executed a two-armed bandit reversal learning task. Behavioral analyses showed that they used this prior knowledge to guide their choice preference. We found activity associated to posterior probability estimates. Overall these results suggest that prefrontal neurons encode decisions associated with Bayesian subjective values and highlight the role of the PFC in representing a belief about the current state of the world. |
