{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,5]],"date-time":"2026-03-05T16:31:13Z","timestamp":1772728273174,"version":"3.50.1"},"reference-count":41,"publisher":"MIT Press","issue":"12","content-domain":{"domain":["direct.mit.edu"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2021,11,12]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Recurrent neural network (RNN) models trained to perform cognitive tasks are a useful computational tool for understanding how cortical circuits execute complex computations. However, these models are often composed of units that interact with one another using continuous signals and overlook parameters intrinsic to spiking neurons. Here, we developed a method to directly train not only synaptic-related variables but also membrane-related parameters of a spiking RNN model. Training our model on a wide range of cognitive tasks resulted in diverse yet task-specific synaptic and membrane parameters. We also show that fast membrane time constants and slow synaptic decay dynamics naturally emerge from our model when it is trained on tasks associated with working memory (WM). Further dissecting the optimized parameters revealed that fast membrane properties are important for encoding stimuli, and slow synaptic dynamics are needed for WM maintenance. This approach offers a unique window into how connectivity patterns and intrinsic neuronal properties contribute to complex dynamics in neural populations.<\/jats:p>","DOI":"10.1162\/neco_a_01409","type":"journal-article","created":{"date-parts":[[2021,10,28]],"date-time":"2021-10-28T18:29:53Z","timestamp":1635445793000},"page":"3264-3287","update-policy":"https:\/\/doi.org\/10.1162\/mitpressjournals.corrections.policy","source":"Crossref","is-referenced-by-count":7,"title":["Learning the Synaptic and Intrinsic Membrane Dynamics Underlying Working Memory in Spiking Neural Network Models"],"prefix":"10.1162","volume":"33","author":[{"given":"Yinghao","family":"Li","sequence":"first","affiliation":[{"name":"Computational Neurobiology Laboratory, Salk Institute for Biological Studies, La Jolla, CA 92037, U.S.A. yil574@ucsd.edu"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Robert","family":"Kim","sequence":"additional","affiliation":[{"name":"Computational Neurobiology Laboratory, Salk Institute for Biological Studies, La Jolla, CA 92037, and Neurosciences Graduate Program and Medical Scientist Training Program, University of California San Diego, La Jolla, CA 92093, U.S.A. r8kim@health.ucsd.edu"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Terrence J.","family":"Sejnowski","sequence":"additional","affiliation":[{"name":"Computational Neurobiology Laboratory, Salk Institute for Biological Studies, La Jolla, CA 92037, and Institute for Neural Computation and Division of Biological Sciences, University of California San Diego, La Jolla, CA 92093, U.S.A. terry@salk.edu"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"281","published-online":{"date-parts":[[2021,11,12]]},"reference":[{"key":"2021112221524633500_B1","article-title":"Lon short-term memory and learning-to-learn in networks of spiking neurons.","volume":"21","author":"Bellec","year":"2018","journal-title":"Advances in neural information processing systems"},{"issue":"42","key":"2021112221524633500_B2","doi-asserted-by":"publisher","first-page":"16344","DOI":"10.1073\/pnas.0807744105","article-title":"Efficient coding in heterogeneous neuronal populations","volume":"105","author":"Chelaru","year":"2008","journal-title":"Proceedings of the National Academy of Sciences"},{"issue":"6","key":"2021112221524633500_B3","doi-asserted-by":"publisher","first-page":"915","DOI":"10.1016\/0893-6080(95)00033-V","article-title":"Chaotic simulated annealing by a neural network model with transient chaos","volume":"8","author":"Chen","year":"1995","journal-title":"Neural Networks"},{"issue":"3","key":"2021112221524633500_B4","doi-asserted-by":"publisher","first-page":"99","DOI":"10.1016\/0166-2236(90)90185-D","article-title":"Intrinsic firing patterns of diverse neocortical neurons","volume":"13","author":"Connors","year":"1990","journal-title":"Trends in Neurosciences"},{"issue":"13","key":"2021112221524633500_B5","doi-asserted-by":"publisher","first-page":"R496","DOI":"10.1016\/j.cub.2007.04.024","article-title":"Recurrent neuronal circuits in the neocortex","volume":"17","author":"Douglas","year":"2007","journal-title":"Current Biology"},{"issue":"1","key":"2021112221524633500_B6","doi-asserted-by":"publisher","first-page":"149","DOI":"10.1137\/S0363012992238369","article-title":"The minimization of semicontinuous functions: Mollifier subgradients","volume":"33","author":"Ermoliev","year":"1995","journal-title":"SIAM Journal on Control and Optimization"},{"key":"2021112221524633500_B7","doi-asserted-by":"crossref","DOI":"10.1017\/CBO9781107447615","author":"Gerstner","year":"2014","journal-title":"Neuronal dynamics: From single neurons to networks and models of cognition"},{"issue":"3","key":"2021112221524633500_B8","doi-asserted-by":"publisher","first-page":"477","DOI":"10.1016\/0896-6273(95)90304-6","article-title":"Cellular basis of working memory","volume":"14","author":"Goldman-Rakic","year":"1995","journal-title":"Neuron"},{"key":"2021112221524633500_B9","first-page":"237","volume-title":"A field guide to dynamical recurrent neural networks","author":"Hochreiter","year":"2001"},{"key":"2021112221524633500_B10","first-page":"1433","volume-title":"Advances in neural information processing systems, 31","author":"Huh","year":"2018"},{"key":"2021112221524633500_B11","volume":"7","author":"Kim","year":"2018","journal-title":"eLife"},{"issue":"45","key":"2021112221524633500_B12","doi-asserted-by":"publisher","first-page":"22811","DOI":"10.1073\/pnas.1905926116","article-title":"Simple framework for constructing functional spiking recurrent neural networks","volume":"116","author":"Kim","year":"2019","journal-title":"Proceedings of the National Academy of Sciences"},{"key":"2021112221524633500_B13","author":"Kingma","year":"2014","journal-title":"Adam: A method for stochastic optimization."},{"key":"2021112221524633500_B14","doi-asserted-by":"crossref","DOI":"10.3389\/fnins.2016.00508","article-title":"Training deep spiking neural networks using backpropagation","volume":"10","author":"Lee","year":"2016","journal-title":"Frontiers in Neuroscience"},{"issue":"7474","key":"2021112221524633500_B15","doi-asserted-by":"publisher","DOI":"10.1038\/nature12742","article-title":"Context-dependent computation by recurrent dynamics in prefrontal cortex","volume":"503","author":"Mante","year":"2013","journal-title":"Nature"},{"issue":"3","key":"2021112221524633500_B16","doi-asserted-by":"publisher","first-page":"609","DOI":"10.1016\/j.neuron.2018.07.003","article-title":"Linking connectivity, dynamics, and computations in low-rank recurrent neural networks","volume":"99","author":"Mastrogiuseppe","year":"2018","journal-title":"Neuron"},{"issue":"18","key":"2021112221524633500_B17","doi-asserted-by":"publisher","first-page":"4717","DOI":"10.1523\/JNEUROSCI.3757-16.2017","article-title":"Strength and diversity of inhibitory signaling differentiates primate anterior cingulate from lateral prefrontal cortex","volume":"37","author":"Medalla","year":"2017","journal-title":"Journal of Neuroscience"},{"key":"2021112221524633500_B18","doi-asserted-by":"publisher","DOI":"10.7554\/eLife.20899","article-title":"Biologically plausible learning in recurrent neural networks reproduces neural dynamics observed during cognitive tasks","volume":"6","author":"Miconi","year":"2017","journal-title":"eLife"},{"key":"2021112221524633500_B19","article-title":"Surrogate gradient learning in spiking neural networks.","author":"Neftci","year":"2019","journal-title":"CoRR"},{"issue":"1","key":"2021112221524633500_B20","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-017-01827-3","article-title":"Supervised learning in spiking neural networks with force training","volume":"8","author":"Nicola","year":"2017","journal-title":"Nature Communications"},{"key":"2021112221524633500_B21","doi-asserted-by":"publisher","DOI":"10.7554\/eLife.35869","article-title":"State-dependent cell-type-specific membrane potential dynamics and unitary synaptic inputs in awake mice","volume":"7","author":"Pala","year":"2018","journal-title":"eLife"},{"issue":"5","key":"2021112221524633500_B22","doi-asserted-by":"publisher","first-page":"1731","DOI":"10.1073\/pnas.1109895109","article-title":"Spatiotemporal dynamics of neocortical excitation and inhibition during human sleep","volume":"109","author":"Peyrache","year":"2012","journal-title":"Proceedings of the National Academy of Sciences"},{"issue":"1","key":"2021112221524633500_B23","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevE.82.011903","article-title":"Stimulus-dependent suppression of chaos in recurrent neural networks","volume":"82","author":"Rajan","year":"2010","journal-title":"Physical Review E"},{"issue":"1","key":"2021112221524633500_B24","doi-asserted-by":"publisher","first-page":"128","DOI":"10.1016\/j.neuron.2016.02.009","article-title":"Recurrent network models of sequence generation and memory","volume":"90","author":"Rajan","year":"2016","journal-title":"Neuron"},{"key":"2021112221524633500_B25","author":"Ramachandran","year":"2017","journal-title":"Searching for activation functions."},{"issue":"5","key":"2021112221524633500_B26","doi-asserted-by":"publisher","first-page":"1005","DOI":"10.1016\/j.neuron.2018.05.020","article-title":"Flexible sensorimotor computations through rapid reconfiguration of cortical dynamics","volume":"98","author":"Remington","year":"2018","journal-title":"Neuron"},{"issue":"6735","key":"2021112221524633500_B27","doi-asserted-by":"publisher","first-page":"470","DOI":"10.1038\/20939","article-title":"Neuronal correlates of parametric working memory in the prefrontal cortex","volume":"399","author":"Romo","year":"1999","journal-title":"Nature"},{"issue":"3","key":"2021112221524633500_B28","article-title":"Learning representations by back-propagating errors","volume":"5","author":"Rumelhart","year":"1988","journal-title":"Cognitive Modeling"},{"key":"2021112221524633500_B29","doi-asserted-by":"publisher","DOI":"10.3389\/fnins.2019.00095","article-title":"Going deeper in spiking neural networks: VGG and residual architectures.","volume":"13","author":"Sengupta","year":"2019","journal-title":"Frontiers in Neuroscience"},{"issue":"2","key":"2021112221524633500_B30","doi-asserted-by":"publisher","first-page":"298","DOI":"10.1016\/j.neuron.2015.08.039","article-title":"Cell-type-specific sensorimotor processing in striatal projection neurons during goal-directed behavior","volume":"88","author":"Sippy","year":"2015","journal-title":"Neuron"},{"issue":"3","key":"2021112221524633500_B31","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevLett.61.259","article-title":"Chaos in random neural networks","volume":"61","author":"Sompolinsky","year":"1988","journal-title":"Physical Review Letters"},{"issue":"2","key":"2021112221524633500_B32","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pcbi.1004792","article-title":"Training excitatory-inhibitory recurrent neural networks for cognitive tasks: A simple and flexible framework","volume":"12","author":"Song","year":"2016","journal-title":"PLOS Computational Biology"},{"issue":"2","key":"2021112221524633500_B33","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pcbi.1004792","article-title":"Training excitatory-inhibitory recurrent neural networks for cognitive tasks: A simple and flexible framework","volume":"12","author":"Song","year":"2016","journal-title":"PLOS Computational Biology"},{"issue":"4","key":"2021112221524633500_B34","doi-asserted-by":"publisher","first-page":"544","DOI":"10.1016\/j.neuron.2009.07.018","article-title":"Generating coherent patterns of activity from chaotic neural networks","volume":"63","author":"Sussillo","year":"2009","journal-title":"Neuron"},{"key":"2021112221524633500_B35","doi-asserted-by":"publisher","first-page":"47","DOI":"10.1016\/j.neunet.2018.12.002","article-title":"Deep learning in spiking neural networks","volume":"111","author":"Tavanaei","year":"2018","journal-title":"Neural Networks"},{"issue":"6","key":"2021112221524633500_B36","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pcbi.1004895","article-title":"Learning universal computations with spikes","volume":"12","author":"Thalmeier","year":"2016","journal-title":"PLOS Computational Biology"},{"issue":"2","key":"2021112221524633500_B37","doi-asserted-by":"publisher","first-page":"260","DOI":"10.1016\/j.neuron.2016.06.033","article-title":"GABAergic interneurons in the neocortex: From cellular properties to circuits","volume":"91","author":"Tremblay","year":"2016","journal-title":"Neuron"},{"issue":"1","key":"2021112221524633500_B38","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1016\/j.tins.2004.10.010","article-title":"Spike times make sense","volume":"28","author":"VanRullen","year":"2005","journal-title":"Trends in Neurosciences"},{"issue":"2","key":"2021112221524633500_B39","doi-asserted-by":"publisher","first-page":"215","DOI":"10.1016\/j.neuron.2008.09.034","article-title":"Decision making in recurrent neuronal circuits","volume":"60","author":"Wang","year":"2008","journal-title":"Neuron"},{"issue":"10","key":"2021112221524633500_B40","doi-asserted-by":"publisher","first-page":"1550","DOI":"10.1109\/5.58337","article-title":"Backpropagation through time: What it does and how to do it","volume":"78","author":"Werbos","year":"1990","journal-title":"Proceedings of the IEEE"},{"key":"2021112221524633500_B41","first-page":"7800","volume-title":"Advances in neural information processing systems, 32","author":"Zhang","year":"2019"}],"container-title":["Neural Computation"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/direct.mit.edu\/neco\/article-pdf\/33\/12\/3264\/1971998\/neco_a_01409.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/direct.mit.edu\/neco\/article-pdf\/33\/12\/3264\/1971998\/neco_a_01409.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2021,11,22]],"date-time":"2021-11-22T16:53:19Z","timestamp":1637599999000},"score":1,"resource":{"primary":{"URL":"https:\/\/direct.mit.edu\/neco\/article\/33\/12\/3264\/107679\/Learning-the-Synaptic-and-Intrinsic-Membrane"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,11,12]]},"references-count":41,"journal-issue":{"issue":"12","published-online":{"date-parts":[[2021,11,12]]},"published-print":{"date-parts":[[2021,11,12]]}},"URL":"https:\/\/doi.org\/10.1162\/neco_a_01409","relation":{"has-preprint":[{"id-type":"doi","id":"10.1101\/2020.06.11.147405","asserted-by":"object"}]},"ISSN":["0899-7667","1530-888X"],"issn-type":[{"value":"0899-7667","type":"print"},{"value":"1530-888X","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2021,12]]},"published":{"date-parts":[[2021,11,12]]}}}