Signed and unsigned reward prediction errors dynamically enhance learning and memory
Memory helps guide behavior, but which experiences from the past are prioritized? Classic models of learning posit that events associated with unpredictable outcomes as well as, paradoxically, predictable outcomes, deploy more attention and learning for those events. Here, we test reinforcement lear...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
eLife Sciences Publications Ltd
2021-03-01
|
| Series: | eLife |
| Subjects: | |
| Online Access: | https://elifesciences.org/articles/61077 |
| _version_ | 1811202595802316800 |
|---|---|
| author | Nina Rouhani Yael Niv |
| author_facet | Nina Rouhani Yael Niv |
| author_sort | Nina Rouhani |
| collection | DOAJ |
| description | Memory helps guide behavior, but which experiences from the past are prioritized? Classic models of learning posit that events associated with unpredictable outcomes as well as, paradoxically, predictable outcomes, deploy more attention and learning for those events. Here, we test reinforcement learning and subsequent memory for those events, and treat signed and unsigned reward prediction errors (RPEs), experienced at the reward-predictive cue or reward outcome, as drivers of these two seemingly contradictory signals. By fitting reinforcement learning models to behavior, we find that both RPEs contribute to learning by modulating a dynamically changing learning rate. We further characterize the effects of these RPE signals on memory and show that both signed and unsigned RPEs enhance memory, in line with midbrain dopamine and locus-coeruleus modulation of hippocampal plasticity, thereby reconciling separate findings in the literature. |
| first_indexed | 2024-04-12T02:41:24Z |
| format | Article |
| id | doaj.art-d9d7ee463edd4bd3bebbf1be8603a1e8 |
| institution | Directory Open Access Journal |
| issn | 2050-084X |
| language | English |
| last_indexed | 2024-04-12T02:41:24Z |
| publishDate | 2021-03-01 |
| publisher | eLife Sciences Publications Ltd |
| record_format | Article |
| series | eLife |
| spelling | doaj.art-d9d7ee463edd4bd3bebbf1be8603a1e82022-12-22T03:51:18ZengeLife Sciences Publications LtdeLife2050-084X2021-03-011010.7554/eLife.61077Signed and unsigned reward prediction errors dynamically enhance learning and memoryNina Rouhani0https://orcid.org/0000-0003-2814-0462Yael Niv1https://orcid.org/0000-0002-0259-8371Chen Neuroscience Institute, California Institute of Technology, Pasadena, United StatesDepartment of Psychology, Princeton University, Princeton, United States; Princeton Neuroscience Institute, Princeton University, Princeton, United StatesMemory helps guide behavior, but which experiences from the past are prioritized? Classic models of learning posit that events associated with unpredictable outcomes as well as, paradoxically, predictable outcomes, deploy more attention and learning for those events. Here, we test reinforcement learning and subsequent memory for those events, and treat signed and unsigned reward prediction errors (RPEs), experienced at the reward-predictive cue or reward outcome, as drivers of these two seemingly contradictory signals. By fitting reinforcement learning models to behavior, we find that both RPEs contribute to learning by modulating a dynamically changing learning rate. We further characterize the effects of these RPE signals on memory and show that both signed and unsigned RPEs enhance memory, in line with midbrain dopamine and locus-coeruleus modulation of hippocampal plasticity, thereby reconciling separate findings in the literature.https://elifesciences.org/articles/61077reinforcement learningmemoryreward prediction errorcomputational model |
| spellingShingle | Nina Rouhani Yael Niv Signed and unsigned reward prediction errors dynamically enhance learning and memory eLife reinforcement learning memory reward prediction error computational model |
| title | Signed and unsigned reward prediction errors dynamically enhance learning and memory |
| title_full | Signed and unsigned reward prediction errors dynamically enhance learning and memory |
| title_fullStr | Signed and unsigned reward prediction errors dynamically enhance learning and memory |
| title_full_unstemmed | Signed and unsigned reward prediction errors dynamically enhance learning and memory |
| title_short | Signed and unsigned reward prediction errors dynamically enhance learning and memory |
| title_sort | signed and unsigned reward prediction errors dynamically enhance learning and memory |
| topic | reinforcement learning memory reward prediction error computational model |
| url | https://elifesciences.org/articles/61077 |
| work_keys_str_mv | AT ninarouhani signedandunsignedrewardpredictionerrorsdynamicallyenhancelearningandmemory AT yaelniv signedandunsignedrewardpredictionerrorsdynamicallyenhancelearningandmemory |