Training Excitatory-Inhibitory Recurrent Neural Networks for Cognitive Tasks: A Simple and Flexible Framework
Open accessThe DOI matches its Crossref record; 91 of the 91 references it cites were found in a registry or an index.
Abstract
The ability to simultaneously record from large numbers of neurons in behaving animals has ushered in a new era for the study of the neural circuit mechanisms underlying cognitive functions. One promising approach to uncovering the dynamical and computational principles governing population responses is to analyze model recurrent neural networks (RNNs) that have been optimized to perform the same tasks as behaving animals. Because the optimization of network parameters specifies the desired output but not the manner in which to achieve this output, "trained" networks serve as a source of mechanistic hypotheses and a testing ground for data analyses that link neural computation to behavior. Complete access to the activity and connectivity of the circuit, and the ability to manipulate them arbitrarily, make trained networks a convenient proxy for biological circuits and a valuable platform for theoretical investigation. However, existing RNNs lack basic biological features such as the distinction between excitatory and inhibitory units (Dale's principle), which are essential if RNNs are to provide insights into the operation of biological circuits. Moreover, trained networks can achieve the same behavioral performance but differ substantially in their structure and dynamics, highlighting the need for a simple and flexible framework for the exploratory training of RNNs. Here, we describe a framework for gradient descent-based training of excitatory-inhibitory RNNs that can incorporate a variety of biological knowledge. We provide an implementation based on the machine learning library Theano, whose automatic differentiation capabilities facilitate modifications and extensions. We validate this framework by applying it to well-known experimental paradigms such as perceptual decision-making, context-dependent integration, multisensory integration, parametric working memory, and motor sequence generation. Our results demonstrate the wide range of neural activity patterns and behavior that can be modeled, and suggest a unified setting in which diverse cognitive computations and mechanisms can be studied.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Figures



![Fig 4 Context-dependent integration task. (A) Psychometric function, showing the percentage of trials on which the network chose choice 1 as a function of the signed motion (upper) and signed color (lower) coherence in motion-context (black) and color-context (blue) trials. (B) Average population responses in state space during the stimulus period, projected to the 3-dimensional subspace capturing variance due to choice, motion, and color as in [5]. Only correct trials were included. The task-related axes were obtained through a linear regression analysis. Note that “choice” here has a unit-specific meaning that depends on the preferred choice of the unit as determined by the selectivity index (Eq 30). For both motion (black) and color (blue), coherences increase from light to dark. Upper plots show trials during the motion context, and lower plots show trials during the color context. (C) Normalized responses of four recurrent units during the stimulus period show mixed representation of task variables. Solid lines indicate the preferred choice and dashed lines the nonpreferred choice of each unit. (D) Denoised regression coefficients from the linear regression analysis. By definition, the coefficients for choice are almost exclusively positive.](https://figures.cochat.ai/paper-figures/ca3a2b623c1e4966aef22596dc894038/training-excitatory-inhibitory-recurrent-neural-networks-for-cognitive-tasks-a-figure-4.jpg)





Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The cited source says what the claim attributes to it.
In the citing paper
“and which can be viewed as an approximation to continuous time firing rate models of neural circuits (Song et al., 2016).”
From Remapping in a recurrent neural network model of navigation and context inference (2023) · cites this paper as Song et al., 2016
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Hochreiter, S., & Schmidhuber, J. (1997) · Neural Computation · doi:10.1162/neco.1997.9.8.1735 · Cited by 102,318 · cited 1×
“For instance, machine learning applications typically employ a type of RNN known as Long Short-Term Memory (LSTM), which uses multiplicative gates to facilitate learning of long-term dependencies and currently represents one of the most powerful methods for solving sequence-related problems [72].”
MethodReinforcement Learning: An Introduction
Sutton, R. S., & Barto, A. G. (1998) · IEEE Transactions on Neural Networks · doi:10.1109/tnn.1998.712192 · Cited by 27,225 · cited 1×
“Combining RNN training with reinforcement learning methods [75 – 77] will be essential to bringing network training closer to the reward-based manner in which animals are trained.”
MethodReinforcement Learning: An Introduction
(2005) · IEEE Transactions on Neural Networks · doi:10.1109/tnn.2004.842673 · Cited by 24,496
Learning Internal Representations by Error Propagation
Rumelhart, D. E., Hinton, G. E., & Williams, R. J. (1985) · doi:10.21236/ada164453 · Cited by 16,172
Parallel Distributed Processing: Explorations in the Microstructure of Cognition: Foundations
Rumelhart, D. E., McClelland, J. L., & AU (1986) · doi:10.7551/mitpress/5236.001.0001 · Cited by 15,686
Learning long-term dependencies with gradient descent is difficult
Bengio, Y., Simard, P., & Frasconi, P. (1994) · IEEE Transactions on Neural Networks · doi:10.1109/72.279181 · Cited by 8,542 · cited 1×
“Even so, SGD with the objective function given in Eqs 15 and 16 often fails to converge to a solution when the network must learn dependencies between distant time points [60].”
BackgroundOn the difficulty of training recurrent neural networks
Pascanu, R., Mikolov, T., & Bengio, Y. (2012) · arXiv · doi:10.48550/arxiv.1211.5063 · Cited by 4,310 · cited 4×
“To remedy this problem, which is due to some gradient components being too large ( exploding gradients) and some gradient components being too small ( vanishing gradients), we follow [35] in making two modifications.”
BackgroundMethodInfluential(2013) · Cited by 3,740
Spiking Neuron Models: Single Neurons, Populations, Plasticity
Gerstner, W., & Kistler, W. M. (2002) · Cambridge University Press eBooks · doi:10.1017/cbo9780511815706 · Cited by 3,080
Interneurons of the neocortical inhibitory system
Markram, H., Toledo-Rodriguez, M., Wang, Y., Gupta, A., Silberberg, G., & Wu, C. (2004) · Nature Reviews Neuroscience · doi:10.1038/nrn1519 · Cited by 2,954 · cited 2×
“Local cortical circuits have distance [48], layer [26, 49, 50], and cell-type [23, 25, 27, 51] dependent patterns of connectivity and different overall levels of sparseness for excitatory to excitatory, inhibitory to excitatory, excitatory to inhibitory, and inhibitory to inhibitory connections [52, 53].”
Background
Research works printed but not linked: 2
- Bengio Y. 2015 Towards biologically plausible deep learning;Available from: http://arxiv.org/abs/1502.0415.
- DE Rumelhart, 1986, Parallel Distrib. Process, Vol. 1, 318
Cited by
Items 201–273 of 273
A cortical information bottleneck during decision-makingPreprint
Kleinman, M., Wang, T., Xiao, D., Feghhi, E., Lee, E., Carr, N., Li, Y., Hadidi, N., Chandrasekaran, C., & Kao, J. C. (2023) · doi:10.7554/elife.89369.1 · Cited by 2
Wainstein, G., Whyte, C. J., Martens, K. A. E., Müller, E. J., Medel, V., Anderson, B., Stöttinger, E., Danckert, J. A., Munn, B., & Shine, J. M. (2025) · eLife · doi:10.7554/elife.93191.2 · Cited by 3
Teng, X., & Zhang, R. (2024) · eLife · doi:10.7554/elife.99383.2 · Cited by 3
A Recurrent Neural Network for Rhythmic TimingPreprint
Zemlianova, K. O., Bose, A., & Rinzel, J. (2024) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2024.05.24.595797 · Cited by 3
“The network was trained using the Pycog library 55 using stochastic gradient descent algorithm with a learning rate of 0.01.”
BackgroundMethodInfluentialZajzon, B., Bouhadjar, Y., Fabre, M., Schmidt, F., Ostendorf, N., Neftci, E., Morrison, A., & Renato, D. (2025) · JuSER (Forschungszentrum Jülich) · doi:10.48550/arxiv.2512.24977 · Cited by 2
Remapping in a recurrent neural network model of navigation and context inference
Low, I. I., Giocomo, L. M., & Williams, A. H. (2023) · eLife · doi:10.7554/elife.86943.3 · Cited by 1
“53 The number of inputs, , and outputs, , varied depending on the computational task the RNN 𝑀 𝐿 was trained to perform.”
BackgroundSynergistic Pathways of Modulation Enable Robust Task Packing Within Neural Dynamics
Vedovati, G., & Ching, S. (2025) · Neural Computation · doi:10.1162/neco.a.18 · Cited by 1
Wagner, M., Chen, Y., Karuvally, A., Cameron, M., & Sejnowski, T. (2026) · Philosophical Transactions of the Royal Society B Biological Sciences · doi:10.1098/rstb.2025.0260 · Cited by 2
Navigating uncertainty: Safety solutions for autonomous vehicle integration into mixed-mode mobility
Cooke, J. R. E., Lingam, S. N., Alavian, P., Collast, C., de Croon, G., Franssen, M., Hagenaars, J. J., Kober, J., Morsink, P., Liu, K., et al. (2026) · Transportation Research Interdisciplinary Perspectives · doi:10.1016/j.trip.2026.102141 · Cited by 1
Feasible cognition requires latent state
Howard, I. S. (2026) · Biological Cybernetics · doi:10.1007/s00422-026-01055-3 · Cited by 1
“For example, recurrent neural networks trained on partially observable tasks develop internal dynamics that encode latent variables required for correct behavior (Song et al. 2016).”
Lee, K., Baracco, L., Pennartz, C. M. A., Suzuki, M., & Mejías, J. F. (2026) · Current Research in Neurobiology · doi:10.1016/j.crneur.2026.100155 · Cited by 2
“The membrane time constant ( τ i ) was fixed at 100 ms, which takes into account realistic cortical pyramidal neuron integration times (La Camera et al., 2006) and is consistent with values typically used for firing rate units in network models (Song et al., 2016; Wierda et al., 2025).”
Kleinman, M., Chandrasekaran, C., & Kao, J. C. (2019) · Conference on Cognitive Computational Neuroscience (CCN) · doi:10.32470/ccn.2019.1123-0 · Cited by 1
“We optimized the parameters b, Wrec, Win, and Wout using backpropagation through time with Adam optimization (Pascanu, Mikolov, & Bengio, 2013) to minimize the mean square error between the desired output and the output produced by the network (Song et al., 2016).”
MethodDual sampling neural network: Learning without explicit optimization
Teramae, J.-N., & Tsubo, Y. (2022) · Physical Review Research · doi:10.1103/physrevresearch.4.043051 · Cited by 2
Computational modeling of human multisensory spatial representation by a neural architecture
Domenici, N., Sanguineti, V., Morerio, P., Campus, C., Bue, A. D., Gori, M., & Murino, V. (2023) · PLoS ONE · doi:10.1371/journal.pone.0280987 · Cited by 1
“Differently from canonical machine learning applications and similarly to the approach fostered by Song and colleagues [20], our goal was not to maximize at all costs the model’s performance but to replicate a human-like perceptual behavior.”
BackgroundResultDisinhibitory signaling enables flexible coding of top-down information in cortical networks
Aquino, T. G., Kim, R., & Rungratsameetaweemana, N. (2026) · PLoS Biology · doi:10.1371/journal.pbio.3003831 · Cited by 1
“To preserve biological interpretability, we enforced Dale’s principle by constraining each unit to be either purely excitatory or purely inhibitory, using a sign-constrained weight parametrization [48].”
Kuroki, S., & Isomura, T. (2017) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/181297 · Cited by 1
“Song et al. proposed an RNN model (pycog model) (Song et al., 2016), which comprised separate excitatory and inhibitory neuronal units and involved the use of a simpler optimization method than HF, namely, a modified stochastic gradient descent (SGD) (Pascanu, Mikolov, & Bengio, 2013).”
BackgroundMethodRemapping in a recurrent neural network model of navigation and context inferencePreprint
Low, I. I. C., Giocomo, L. M., & Williams, A. H. (2023) · eLife · doi:10.7554/elife.86943.1 · Cited by 3
Wainstein, G., Whyte, C. J., Martens, K. A. E., Müller, E. J., Munn, B., Medel, V., Anderson, B., Stöttinger, E., Danckert, J. A., & Shine, J. M. (2024) · eLife · doi:10.7554/elife.93191.1 · Cited by 1
Effects of noise and metabolic cost on cortical task representationsPreprint
Stroud, J. P., Wójcik, M. J., Jensen, K. T., Kusunoki, M., Kadohisa, M., Buckley, M. J., Duncan, J., Stokes, M. G., & Lengyel, M. (2024) · eLife · doi:10.7554/elife.94961.1 · Cited by 1
Wainstein, G., Whyte, C. J., Martens, K. A. E., Müller, E. J., Medel, V., Anderson, B., Stöttinger, E., Danckert, J. A., Munn, B., & Shine, J. M. (2025) · eLife · doi:10.7554/elife.93191.3 · Cited by 1
Liao, F., Hu, X., & Song, S. (2018) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/244350
“Song et al. (2016) investigated the connectivity of a recurrent network that is optimized in several cognitive tasks.”
BackgroundTime representation in neural network models trained to perform interval timing tasks
Bi, Z., & Zhou, C. (2019)
Compact task representations as a normative model for higher-order brain activity
Berger, S., & Machens, C. K. (2020) · neural information processing systems · Cited by 3
Teng, X., & Zhang, R. (2021) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2021.10.04.463033
Faculty Opinions recommendation of Frontal cortex and the hierarchical control of behavior.
Fellows, L. K. (2020) · Faculty Opinions – Post-Publication Peer Review of the Biomedical Literature · doi:10.3410/f.732274976.793571333
Robust chaos and multi-stability in piecewise linear recurrent neural networksPreprint
Monfared, Z., Patra, M., & Durstewitz, D. (2022) · Research Square · doi:10.21203/rs.3.rs-2147683/v1
Wainstein, G., Whyte, C. J., Martens, K. A. E., Müller, E. J., Munn, B., Medel, V., Anderson, B., Stöttinger, E., Danckert, J. A., & Shine, J. M. (2023) · Research Square · doi:10.21203/rs.3.rs-2356429/v2 · Cited by 1
A Versatile Hub Model For Efficient Information Propagation And Feature SelectionPreprint
Wang, Z., & Wang, J. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2307.02398
“Connection weights in the three layers could be updated using gradient descent [20], intrinsic plasticity rules [28], synatic plasticity rules [29], or remain unchanged [26, 30].”
BackgroundMethodKleinman, M., Wang, T., Xiao, D., Feghhi, E., Lee, E., Carr, N., Li, Y., Hadidi, N., Chandrasekaran, C., & Kao, J. C. (2023) · eLife · doi:10.7554/elife.89369
Wang, C., Zhang, T., Chen, X., He, S., Li, S., & Wu, S. (2023) · eLife · doi:10.7554/elife.86365.sa2
Nonlinear feedback modulation contributes to the optimization of flexible decision-makingPreprint
Wu, X., & Zhou, Y. (2024) · eLife · doi:10.7554/elife.96402.1
Modeling Normal and Abnormal Circuit Development with Recurrent Neural Networks
Zavitz, D., Ching, S., & Goodhill, G. J. (2024) · Cold Spring Harbor Perspectives in Biology · doi:10.1101/cshperspect.a041507 · Cited by 2
Teng, X., & Zhang, R. (2024) · eLife · doi:10.7554/elife.99383.1
Zhang, Y., Feng, J., & Min, B. (2024) · eLife · doi:10.7554/elife.103636.1
Zhang, Y., Feng, J., & Min, B. (2025) · eLife · doi:10.7554/elife.103636.2
Zhang, X., Altrabulsi, M., Xu, W., Wimmer, R., Halassa, M. M., & Chen, Z. S. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2025.07.11.663676
Nonlinear feedback modulation contributes to the optimization of flexible decision-makingPreprint
Wu, X., & Zhou, Y. (2025) · eLife · doi:10.7554/elife.96402.2
A plastic attractor model of flexible rule-based selective attentionPreprint
Whyte, C. J., Manohar, S., Feredoes, E. A., & Woolgar, A. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2025.09.07.674747
Lang, L., Zheng, C., Blackwell, J. M., Camera, G. L., & Fontanini, A. (2026) · eLife · doi:10.7554/elife.109313.3
Excitatory-Inhibitory Dynamics in Adaptive Decision-Making
Chelu, V., & Precup, D. (2025) · Lecture notes in computer science · doi:10.1007/978-981-95-4381-6_5 · Cited by 1
Emergence of Value and Action Codes in Bimodular Spiking Actor–Critic NetworksPreprint
Antolín, C., Falcó-Roget, J., Serrano-Fernández, L., & Parga, N. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2025.11.26.690655
“Recurrent weights were further rescaled to achieve a spectral radius R = 2, defined as the modulus of the largest eigenvalue [49].”
Agents Trained through Reinforcement Learning Exhibit Human-Like Decision-Making Flexibility
Zhong, Y., Shi, W., Dong, Z., Carlo, S. D., Song, M., & Jiang, T. (2025) · Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) · doi:10.1109/embc58623.2025.11251764
“Additionally, RNNs have played a significant role in modeling the cognitive [16]–[18] and motor systems [19] of primates.”
Reconstructing Neural Dynamics Underlying Cognitive Flexibility Using Parameter-Evolving RNNs
Thurm, M. I. (2025) · heiDOK (Heidelberg University) · doi:10.11588/heidok.00037427
Shao, L., Xu, Z., Sun, Q., Li, F., & Yang, Q. (2026) · IEEE Transactions on Cognitive and Developmental Systems · doi:10.1109/tcds.2025.3650131
“To examine the temporal evolution of gamma-band activity under time-varying modulatory conditions, we simulated a recurrent excitatory–inhibitory network model subjected to temporally structured noise over repeated trials [25], [26].”
Recurrent Neural Networks models : inference, representations, and the role of regularization
Fanthomme, A. (2021) · theses.fr (ABES) · doi:10.70675/afd73818z7545z4451zb9a3zd32aa98cd0f6
Santhosh, A., & Narayanan, R. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.04.03.716300
Toward a foundational brain model of intelligence
Wang, X. (2026) · Current Opinion in Neurobiology · doi:10.1016/j.conb.2026.103198
Self-citationZhang, Y., Li, X., Shen, X., Li, F., Okazawa, G., Wang, L., Feng, J., & Min, B. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.04.16.718933 · Cited by 1
Self-Organized Neural Integrators in Noisy Spiking NetworksPreprint
Feng, B., Gao, R., Li, N., & Shouval, H. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.05.04.722735
“…to reproduce integrator-like dynamics in service of a task (Sussillo and Abbott, 2009; Wang et al., 2018; Remington et al., 2018; Laje and Buonomano, 2013; Song et al., 2016; Beiran et al., 2023), but at the cost of mechanistic transparency and biological plausibility of the learning procedure.”
Rhythmic temporal structure organizes recurrent dynamics to support sequential working memoryPreprint
Qin, Y., Yang, Y., Yang, Q., Wei, Q., & Zhang, T. (2026) · eLife · doi:10.64898/2026.05.20.726720
“…eventually improves task performance, we trained a biologically constrained excitatory–inhibitory recurrent neural network with short-term synaptic plasticity to perform a se-82 quential delayed match-to-sample task ( Song et al., 2016 ; Barak, 2017 ; Masse et al., 2019 ; Kozachkov et al., 2022 ).”
Modelling chronic stress as excitation-inhibition perturbation in recurrent working-memory networksPreprint
Diaz, M. A., Beyer, M. A., & Hesse, J. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2606.27529
“Following Song et al. [13], recurrent weights were initialised as positive magnitudes and assigned excitatory or inhibitory signs according to Dale’s principle [32, 42, 36], i.e. positive sign for W EE and W IE , and negative sign for W EI and W II .”
InfluentialDynamic coordination and segregation mechanisms in higher cortex for parallel task processing
Wang, S., Zhu, Y., Li, C., Yung, W., & Ya, K. E. (2026) · Neuron · doi:10.1016/j.neuron.2026.06.001
Guglielmi, L., Albertini, D., Vezzani, A., Burioni, R., & Bonini, L. (2026) · iScience · doi:10.1016/j.isci.2026.116776
“Following previous work, 88 Dale’s principle was enforced directly at initialization by constraining each unit to emit either exclusively excitatory or exclusively inhibitory outputs.”
Bu, X., Jiang, H., Guo, T., Li, H., & Chen, Y. (2026) · Biomimetics · doi:10.3390/biomimetics11080569
“Task-driven paradigms [20,25], currently the more prevalent methodology, train RNN connectivity weights using task conditions from animal experiments as inputs, with behavioral kinematics or targets as outputs.”
Graph-theoretic design of lasing networks for physical visionPreprint
Obernolte, P., Dranczewski, J., Yin, Y., Farchy, T., Ng, W. K., Simonsen, T., Großhauser, E., Arnaudon, A., Peach, R. L., Barahona, M., et al. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2608.13097
Audiovisual sequence memory enabled by a neurodynamic model with primary sensory connections
Lai, R., Li, Y., Huo, S., Li, Y., Grebogi, C., Huang, Z., & Zhang, S.-P. (2026) · Cognitive Neurodynamics · doi:10.1007/s11571-026-10545-w
“…computing models leverage high-dimensional dynamic states to preserve temporal information, offering new insights into multimodal sequence learning (Sporns et al. 2016; Graham et al. 2018; Kanemura and Kitano 2023; Chandrasekaran 2017); and memristor-based neuromorphic circuits enable parallel…”
NeuralRNN: a unified framework for recurrent neural network methods in cognitive neurosciencePreprint
Chen, H., & Ding, N. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.09.14.751453
“Existing tools each cover part of the landscape, including task-optimization frameworks such as PyCog [15] and PsychRNN [26] and general-purpose brain simulators such as BrainPy [27].”
Sergi, I., Rondelli, F., Torricelli, F., Velluto, F., Salvo, A. D., Lauro, M. D., Fadiga, L., & Biscarini, F. (2026) · Advanced Functional Materials · doi:10.1002/adfm.78647
Tuned inhibitory control of neuronal firing thresholds explains predictive sensorimotor behavior
Ahn, J., Kim, S., Park, J., Woo, S., Sohn, H., & Lee, S. (2026) · Communications Biology · doi:10.1038/s42003-026-11082-2
Xiaojun, W., Lin, Z., Deng, B., Liu, X., & Yu, H. (2026) · Neuroscience · doi:10.1016/j.neuroscience.2026.10.016
Nishi, Y., Yamamoto, H., Katori, Y., & Sato, S. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.10.22.755221
Metadata
- Authors
- Song, H. F., Yang, G. R., & Wang, X.-J.
- Affiliations
- New York University (US)
- New York University Shanghai (CN)
- Corresponding author
- Wang, Xiao-Jing
- Venue
- PLOS Computational Biology
- Publisher
- Public Library of Science (PLoS)
- Published
- 2016-02-29
- Volume
- 12
- Issue
- 2
- Pages
- e1004792
- ISSN
- 1553-734X (ISSN-L)
- 1553-7358
- PMC
- PMC4771709
- OpenAlex
- W2280725856
- Cited by
- 281
- References
- 91
- Citations by year
- 2026: 23 · 2025: 33 · 2024: 39 · 2023: 42 · 2022: 26 · 2021: 28 · 2020: 23 · 2019: 27 · 2018: 14 · 2017: 21 · 2016: 5
- FWCI
- 14.21
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- PLoS Computational Biology (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Figshare (submittedVersion, cc-by)
- Funders
- Office of Naval Research (N00014–13–1-0297, N00014)
- Topics
- Life Sciences › Neuroscience › Cognitive Neuroscience › Neural dynamics and brain function
- Physical Sciences › Computer Science › Artificial Intelligence › Neural Networks and Applications
- Life Sciences › Neuroscience › Cognitive Neuroscience › Memory and Neural Mechanisms
- Keywords
- Dale's principle, neural circuit model, cognitive tasks, perceptual decision making, multisensory integration, working memory, neural population dynamics, Theano, automatic differentiation, gradient-based learning
- MeSH terms
- Algorithms, Animals, Haplorhini, Cognition, Computer Simulation, Models, Neurological, Neurons, Computational Biology
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.