Stable task information from an unstable neural population
Over days and weeks, neural activity representing an animal’s position and movement in sensorimotor cortex has been found to continually reconfigure or ‘drift’ during repeated trials of learned tasks, with no obvious change in behavior. This challenges classical theories, which assume stable engrams...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
eLife Sciences Publications Ltd
2020-07-01
|
| Series: | eLife |
| Subjects: | |
| Online Access: | https://elifesciences.org/articles/51121 |
| _version_ | 1811201303655743488 |
|---|---|
| author | Michael E Rule Adrianna R Loback Dhruva V Raman Laura N Driscoll Christopher D Harvey Timothy O'Leary |
| author_facet | Michael E Rule Adrianna R Loback Dhruva V Raman Laura N Driscoll Christopher D Harvey Timothy O'Leary |
| author_sort | Michael E Rule |
| collection | DOAJ |
| description | Over days and weeks, neural activity representing an animal’s position and movement in sensorimotor cortex has been found to continually reconfigure or ‘drift’ during repeated trials of learned tasks, with no obvious change in behavior. This challenges classical theories, which assume stable engrams underlie stable behavior. However, it is not known whether this drift occurs systematically, allowing downstream circuits to extract consistent information. Analyzing long-term calcium imaging recordings from posterior parietal cortex in mice (Mus musculus), we show that drift is systematically constrained far above chance, facilitating a linear weighted readout of behavioral variables. However, a significant component of drift continually degrades a fixed readout, implying that drift is not confined to a null coding space. We calculate the amount of plasticity required to compensate drift independently of any learning rule, and find that this is within physiologically achievable bounds. We demonstrate that a simple, biologically plausible local learning rule can achieve these bounds, accurately decoding behavior over many days. |
| first_indexed | 2024-04-12T02:19:42Z |
| format | Article |
| id | doaj.art-20628c39d67f481c81e9437b30663da7 |
| institution | Directory Open Access Journal |
| issn | 2050-084X |
| language | English |
| last_indexed | 2024-04-12T02:19:42Z |
| publishDate | 2020-07-01 |
| publisher | eLife Sciences Publications Ltd |
| record_format | Article |
| series | eLife |
| spelling | doaj.art-20628c39d67f481c81e9437b30663da72022-12-22T03:52:09ZengeLife Sciences Publications LtdeLife2050-084X2020-07-01910.7554/eLife.51121Stable task information from an unstable neural populationMichael E Rule0https://orcid.org/0000-0002-4196-774XAdrianna R Loback1Dhruva V Raman2Laura N Driscoll3Christopher D Harvey4Timothy O'Leary5https://orcid.org/0000-0002-1029-0158Department of Engineering, University of Cambridge, Cambridge, United KingdomDepartment of Engineering, University of Cambridge, Cambridge, United KingdomDepartment of Engineering, University of Cambridge, Cambridge, United KingdomDepartment of Electrical Engineering, Stanford University, Stanford, United StatesDepartment of Neurobiology, Harvard Medical School, Boston, United StatesDepartment of Engineering, University of Cambridge, Cambridge, United KingdomOver days and weeks, neural activity representing an animal’s position and movement in sensorimotor cortex has been found to continually reconfigure or ‘drift’ during repeated trials of learned tasks, with no obvious change in behavior. This challenges classical theories, which assume stable engrams underlie stable behavior. However, it is not known whether this drift occurs systematically, allowing downstream circuits to extract consistent information. Analyzing long-term calcium imaging recordings from posterior parietal cortex in mice (Mus musculus), we show that drift is systematically constrained far above chance, facilitating a linear weighted readout of behavioral variables. However, a significant component of drift continually degrades a fixed readout, implying that drift is not confined to a null coding space. We calculate the amount of plasticity required to compensate drift independently of any learning rule, and find that this is within physiologically achievable bounds. We demonstrate that a simple, biologically plausible local learning rule can achieve these bounds, accurately decoding behavior over many days.https://elifesciences.org/articles/51121spatial navigationlearning and memoryneural codingcomputational neuroscienceplasticitysystems modeling |
| spellingShingle | Michael E Rule Adrianna R Loback Dhruva V Raman Laura N Driscoll Christopher D Harvey Timothy O'Leary Stable task information from an unstable neural population eLife spatial navigation learning and memory neural coding computational neuroscience plasticity systems modeling |
| title | Stable task information from an unstable neural population |
| title_full | Stable task information from an unstable neural population |
| title_fullStr | Stable task information from an unstable neural population |
| title_full_unstemmed | Stable task information from an unstable neural population |
| title_short | Stable task information from an unstable neural population |
| title_sort | stable task information from an unstable neural population |
| topic | spatial navigation learning and memory neural coding computational neuroscience plasticity systems modeling |
| url | https://elifesciences.org/articles/51121 |
| work_keys_str_mv | AT michaelerule stabletaskinformationfromanunstableneuralpopulation AT adriannarloback stabletaskinformationfromanunstableneuralpopulation AT dhruvavraman stabletaskinformationfromanunstableneuralpopulation AT laurandriscoll stabletaskinformationfromanunstableneuralpopulation AT christopherdharvey stabletaskinformationfromanunstableneuralpopulation AT timothyoleary stabletaskinformationfromanunstableneuralpopulation |