arxiv: 2605.10178 · v1 · submitted 2026-05-11 · 🧬 q-bio.NC · cs.LG· cs.NE

Recognition: 2 theorem links

· Lean Theorem

Joint sparse coding and temporal dynamics support context reconfiguration

Faqiang Liu, Hongyi Li, Luping Shi, Mingkun Xu, Pieter M. Goltstein, Qianqian Shi, Rong Zhao, Sandra Reinert, Yue Che

Pith reviewed 2026-05-12 04:43 UTC · model grok-4.3

classification 🧬 q-bio.NC cs.LGcs.NE

keywords sparse codingtemporal dynamicscontext reconfigurationlifelong learningcatastrophic forgettingmedial prefrontal cortexspiking neural networks

0 comments

The pith

Joint sparse coding and temporal dynamics enable context reconfiguration while preserving prior knowledge.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper investigates how neural systems transition between different contexts without losing earlier learned information. It shows that sparse representations of contexts cut down on interference between them, and that changing activity patterns over time keep those contexts distinct. When artificial networks incorporate both of these traits together, they hold onto old knowledge more effectively during continuous learning. This matters for understanding how the brain stays flexible yet stable, and for building AI systems that avoid erasing what they have already learned.

Core claim

Sparsity in context-dependent representations reduces cross-context interference, whereas temporal dynamics within the network activity further enhance context separability across time. Networks endowed with both properties, such as spiking neural networks, exhibit improved retention during lifelong learning without auxiliary heuristics.

What carries the argument

Joint sparse coding and temporal dynamics that together constrain neural activity to maintain context separability and retention.

If this is right

Sparsity limits overlap between representations of different contexts.
Temporal dynamics add separation along the time dimension.
Spiking neural networks that combine both traits retain prior knowledge better without extra protective rules.
The same activity constraints support energy-efficient adaptation in changing environments.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

Similar joint mechanisms could be tested for their role in other brain areas beyond the mPFC.
The combination might guide the design of new continual-learning algorithms that avoid catastrophic forgetting.
Hardware implementations could exploit the energy-efficient nature of these constraints to reduce power use in deployed networks.

Load-bearing premise

The sparsity and temporal dynamics observed in the mPFC are causally responsible for context separability and retention rather than merely correlated with other unmeasured factors.

What would settle it

Disrupting sparsity or temporal dynamics in the mPFC increases cross-context interference and forgetting, or adding these features to non-spiking networks produces no measurable gain in retention.

Figures

Figures reproduced from arXiv: 2605.10178 by Faqiang Liu, Hongyi Li, Luping Shi, Mingkun Xu, Pieter M. Goltstein, Qianqian Shi, Rong Zhao, Sandra Reinert, Yue Che.

**Figure 2.** Figure 2: Fig.2: Context representations in mPFC arise from ensemble-specific recruit [PITH_FULL_IMAGE:figures/full_fig_p012_2.png] view at source ↗

**Figure 3.** Figure 3: Fig.3: Built-in threshold-induced sparsity and temporal coding enable supe [PITH_FULL_IMAGE:figures/full_fig_p014_3.png] view at source ↗

**Figure 4.** Figure 4: Fig.4: Network-wide sparsity improves lifelong learning performance. [PITH_FULL_IMAGE:figures/full_fig_p016_4.png] view at source ↗

**Figure 5.** Figure 5: Fig.5: Temporal dynamics interact with sparsity to enhance context separa [PITH_FULL_IMAGE:figures/full_fig_p017_5.png] view at source ↗

**Figure 6.** Figure 6: Fig.6: SNNs support scalable lifelong learning with low power and reduced [PITH_FULL_IMAGE:figures/full_fig_p018_6.png] view at source ↗

read the original abstract

Adaptive behavior requires the brain to transition between distinct contexts while maintaining representations of prior experience. The ability to reconfigure neural representations without erasing previously acquired knowledge is central to learning in dynamic environments, yet the neural mechanisms that support this balance remain unclear. Understanding these mechanisms is also critical for addressing catastrophic forgetting in artificial systems designed for lifelong learning. Here, we identify joint sparse coding and temporal dynamics in both the mouse medial prefrontal cortex (mPFC) and computational networks as mechanisms that help preserve prior representations during context transitions. Specifically, sparsity in context-dependent representations reduces cross-context interference, whereas temporal dynamics within the network activity further enhance context separability across time. Strikingly, networks endowed with both properties, such as spiking neural networks, exhibit improved retention during lifelong learning without auxiliary heuristics. These findings establish joint sparse coding and temporal dynamics as a core mechanism supporting flexible context reconfiguration in lifelong learning and, through their activity constraining nature, as an energy-efficient architectural principle for stable adaptation. Together, they provide a mechanistic framework for understanding how the brain preserves prior knowledge while flexibly adapting to new contexts.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The paper ties mPFC sparse coding and temporal dynamics to reduced forgetting in SNNs for lifelong learning, but lacks ablations to isolate those factors from other model traits.

read the letter

The main thing to know is that this work reports sparse, context-specific activity with temporal structure in mouse mPFC, then shows that spiking networks using both properties retain prior contexts better during sequential learning without replay or other add-ons. The abstract frames this joint mechanism as the reason for improved separability and stability across context switches. It is a direct attempt to turn a biological pattern into a constraint for continual learning models. That link is the clearest new angle, even if sparse coding and dynamics have been studied separately before. The biology side appears grounded in recordings, and the model comparison to non-spiking baselines is at least a starting point for showing the practical payoff. The soft spots sit mostly in the causal step. The stress-test concern holds: without ablations that add equivalent sparsity and dynamics to standard networks while fixing parameter count, learning rules, and task details, any retention gain could trace to other SNN features like event-driven updates or surrogate gradients. The claims stay tied to observed correlations in the fitted behaviors rather than independent tests. No equations or derivations are visible in the provided text, so the 'support' language does not reduce to falsifiable predictions outside the specific simulations. This is the sort of paper a computational neuroscience or continual-learning reading group might discuss for the idea, even if the controls need tightening. It is not yet strong enough to cite as settled evidence, but the question it raises is worth referee time to check the methods and see whether the results survive better isolation of the claimed factors. I would send it out for review rather than desk reject.

Referee Report

2 major / 2 minor

Summary. The manuscript claims that joint sparse coding and temporal dynamics, observed in mouse mPFC during context transitions, reduce cross-context interference and enhance separability; computational networks (especially spiking neural networks) endowed with both properties show improved retention in lifelong learning tasks without auxiliary heuristics, providing a mechanistic and energy-efficient framework for context reconfiguration in biological and artificial systems.

Significance. If the causal role of these joint properties is isolated, the work would link specific neural coding features to stable lifelong learning, offering a biologically grounded principle for mitigating catastrophic forgetting in AI and a testable account of mPFC function in dynamic environments. The integration of in vivo observations with network simulations is a strength.

major comments (2)

[Abstract] Abstract: The central claim that 'networks endowed with both properties, such as spiking neural networks, exhibit improved retention during lifelong learning without auxiliary heuristics' requires explicit ablation experiments that add equivalent sparsity and temporal dynamics to non-spiking baselines while holding fixed parameter count, learning rule, and task structure; without such controls, retention gains cannot be attributed to the claimed joint properties rather than correlated SNN features (e.g., event-driven updates or surrogate-gradient training).
[Model experiments] Model experiments section: The assertion that sparsity reduces interference and temporal dynamics enhance separability is presented as causal support for reconfiguration, yet the manuscript provides no independent benchmarks or falsifiable predictions outside the fitted network behaviors to demonstrate that these factors are operative rather than epiphenomenal.

minor comments (2)

[Methods] Clarify the precise definition and measurement of 'temporal dynamics' (e.g., time constants, recurrence) and 'sparse coding' (e.g., L0 vs. L1, population vs. single-unit) when comparing mPFC data to network models.
[Figures] Figure legends should explicitly state the number of animals, trials, and statistical tests used for the mPFC context-separability analyses.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We are grateful to the referee for their insightful comments on our manuscript. We address the major comments point-by-point below and outline the revisions we plan to make.

read point-by-point responses

Referee: [Abstract] Abstract: The central claim that 'networks endowed with both properties, such as spiking neural networks, exhibit improved retention during lifelong learning without auxiliary heuristics' requires explicit ablation experiments that add equivalent sparsity and temporal dynamics to non-spiking baselines while holding fixed parameter count, learning rule, and task structure; without such controls, retention gains cannot be attributed to the claimed joint properties rather than correlated SNN features (e.g., event-driven updates or surrogate-gradient training).

Authors: We concur that rigorous controls are essential to attribute the observed benefits specifically to the combination of sparse coding and temporal dynamics. Accordingly, in the revised manuscript, we will add ablation studies that equip non-spiking baseline networks with comparable sparsity constraints and temporal integration mechanisms. These ablations will keep the total parameter count, learning rules, and task specifications fixed to ensure fair comparison. This will clarify whether the retention improvements stem from the joint properties or other SNN-specific characteristics. revision: yes
Referee: [Model experiments] Model experiments section: The assertion that sparsity reduces interference and temporal dynamics enhance separability is presented as causal support for reconfiguration, yet the manuscript provides no independent benchmarks or falsifiable predictions outside the fitted network behaviors to demonstrate that these factors are operative rather than epiphenomenal.

Authors: Our model experiments systematically vary the degree of sparsity and the presence of temporal dynamics to quantify their impact on cross-context interference and separability, providing direct evidence within the computational framework. To further address this point, we will include additional benchmarks against standard continual learning algorithms and formulate explicit, falsifiable predictions regarding mPFC neural dynamics during context switches that can be tested in future in vivo studies. revision: partial

Circularity Check

0 steps flagged

No significant circularity in derivation chain

full rationale

The paper presents observational and modeling results linking sparse coding and temporal dynamics in mPFC and SNNs to context separability and retention during lifelong learning. No explicit equations, first-principles derivations, or parameter-fitting steps are described that reduce any claimed prediction to the inputs by construction. The central claim rests on empirical correlations and model performance comparisons rather than self-definitional loops, fitted inputs renamed as predictions, or load-bearing self-citations that import uniqueness theorems. The derivation is self-contained against external benchmarks such as biological recordings and network retention metrics.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The central claim rests on the assumption that observed sparsity and dynamics are the operative mechanisms rather than epiphenomena, plus standard neuroscientific assumptions about population coding.

axioms (2)

domain assumption Sparsity in context-dependent representations reduces cross-context interference
Invoked in abstract as the reason sparsity helps preservation
domain assumption Temporal dynamics enhance context separability across time
Invoked as the second supporting property

pith-pipeline@v0.9.0 · 5515 in / 1189 out tokens · 36474 ms · 2026-05-12T04:43:11.379906+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

53 extracted references · 53 canonical work pages

[1]

Task set and prefrontal cortex.Annual Review of Neuroscience, 31: 219–245, 2008

Katsuyuki Sakai. Task set and prefrontal cortex.Annual Review of Neuroscience, 31: 219–245, 2008. doi: 10.1146/annurev.neuro.31.060407.125642

work page doi:10.1146/annurev.neuro.31.060407.125642 2008
[2]

Mechanisms of cognitive set flexibility in the human brain.Journal of Neuroscience, 24(50):11208–11216, 2004

Roshan Cools, Luke Clark, and Trevor W Robbins. Mechanisms of cognitive set flexibility in the human brain.Journal of Neuroscience, 24(50):11208–11216, 2004. doi: 10.1523/ JNEUROSCI.4099-04.2004

work page 2004
[3]

An integrative theory of prefrontal cortex function

Earl K Miller and Jonathan D Cohen. An integrative theory of prefrontal cortex function. Annual Review of Neuroscience, 24:167–202, 2001. doi: 10.1146/annurev.neuro.24.1.167

work page doi:10.1146/annurev.neuro.24.1.167 2001
[4]

Conflict monitoring and cognitive control.Psychological Review, 108(3):624–652, 2004

Matthew M Botvinick, Jonathan D Cohen, and Cameron S Carter. Conflict monitoring and cognitive control.Psychological Review, 108(3):624–652, 2004. doi: 10.1037/0033-295X. 108.3.624

work page doi:10.1037/0033-295x 2004
[5]

Interplay of hippocampus and prefrontal cortex in memory.Current Biology, 23(17):R764–R773, 2013

Alison R Preston and Howard Eichenbaum. Interplay of hippocampus and prefrontal cortex in memory.Current Biology, 23(17):R764–R773, 2013. doi: 10.1016/j.cub.2013.05.041

work page doi:10.1016/j.cub.2013.05.041 2013
[6]

Bidirec- tional prefrontal–hippocampal interactions support context-guided memory.Nature Neu- roscience, 19(8):992–994, 2016

Rebecca Place, Anja Farovik, Melissa Brockmann, and Howard Eichenbaum. Bidirec- tional prefrontal–hippocampal interactions support context-guided memory.Nature Neu- roscience, 19(8):992–994, 2016. doi: 10.1038/nn.4327

work page doi:10.1038/nn.4327 2016
[7]

Hippocampal–prefrontal interactions in cognition, behavior and psychiatric disease.Frontiers in Systems Neuroscience, 9:190, 2016

Torfi Sigurdsson and Sevil Duvarci. Hippocampal–prefrontal interactions in cognition, behavior and psychiatric disease.Frontiers in Systems Neuroscience, 9:190, 2016. doi: 10.3389/fnsys.2015.00190

work page doi:10.3389/fnsys.2015.00190 2016
[8]

LeCun, Y

Yann LeCun, Yoshua Bengio, and Geoffrey Hinton. Deep learning.Nature, 521(7553): 436–444, 2015. doi: 10.1038/nature14539

work page doi:10.1038/nature14539 2015
[9]

Continual lifelong learning with neural networks: A review.Neural networks, 113:54–71, 2019

German I Parisi, Ronald Kemker, Jose L Part, Christopher Kanan, and Stefan Wermter. Continual lifelong learning with neural networks: A review.Neural networks, 113:54–71, 2019

work page 2019
[10]

A continual learning survey: Defying forgetting in classification tasks.IEEE transactions on pattern analysis and machine intelligence, 44 (7):3366–3385, 2021

Matthias De Lange, Rahaf Aljundi, Marc Masana, Sarah Parisot, Xu Jia, Aleˇ s Leonardis, Gregory Slabaugh, and Tinne Tuytelaars. A continual learning survey: Defying forgetting in classification tasks.IEEE transactions on pattern analysis and machine intelligence, 44 (7):3366–3385, 2021

work page 2021
[11]

Hybrid neural networks for continual learning inspired by corticohippocampal circuits.Nature Communications, 16(1):1272, 2025

Qianqian Shi, Faqiang Liu, Hongyi Li, Guangyu Li, Luping Shi, and Rong Zhao. Hybrid neural networks for continual learning inspired by corticohippocampal circuits.Nature Communications, 16(1):1272, 2025

work page 2025
[12]

The Journal of Physiology160(1), 106–154 (1962)

David H. Hubel and Torsten N. Wiesel. Receptive fields, binocular interaction and func- tional architecture in the cat’s visual cortex.Journal of Physiology, 160(1):106–154, 1962. doi: 10.1113/jphysiol.1962.sp006837

work page doi:10.1113/jphysiol.1962.sp006837 1962
[13]

Gradient-based learning applied to document recognition.Proceedings of the IEEE, 86(11):2278–2324, 1998

Yann LeCun, L´ eon Bottou, Yoshua Bengio, and Patrick Haffner. Gradient-based learning applied to document recognition.Proceedings of the IEEE, 86(11):2278–2324, 1998. doi: 10.1109/5.726791

work page doi:10.1109/5.726791 1998
[14]

Neocognitron: A self-organizing neural network model for a mecha- nism of pattern recognition unaffected by shift in position.Biological Cybernetics, 36(4): 193–202, 1980

Kunihiko Fukushima. Neocognitron: A self-organizing neural network model for a mecha- nism of pattern recognition unaffected by shift in position.Biological Cybernetics, 36(4): 193–202, 1980. doi: 10.1007/BF00344251. 19

work page doi:10.1007/bf00344251 1980
[15]

Daniel L. K. Yamins, Ha Hong, Charles F. Cadieu, Ethan A. Solomon, Darren Seibert, and James J. DiCarlo. Performance-optimized hierarchical models predict neural responses in higher visual cortex.Proceedings of the National Academy of Sciences, 111(23):8619–8624,

work page
[16]

doi: 10.1073/pnas.1403112111

work page doi:10.1073/pnas.1403112111
[17]

Neural mechanisms of selective visual attention

Robert Desimone and John Duncan. Neural mechanisms of selective visual attention. Annual Review of Neuroscience, 18:193–222, 1995. doi: 10.1146/annurev.ne.18.030195. 001205

work page doi:10.1146/annurev.ne.18.030195 1995
[18]

Gomez, Lukasz Kaiser, and Illia Polosukhin

Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N. Gomez, Lukasz Kaiser, and Illia Polosukhin. Attention is all you need. InAdvances in Neural Information Processing Systems, volume 30, 2017

work page 2017
[19]

Brain-inspired replay for continual learning with artificial neural networks.Nature communications, 11(1):4069, 2020

Gido M Van de Ven, Hava T Siegelmann, and Andreas S Tolias. Brain-inspired replay for continual learning with artificial neural networks.Nature communications, 11(1):4069, 2020

work page 2020
[20]

Richards, Timothy P

Blake A. Richards, Timothy P. Lillicrap, Philippe Beaudoin, Yoshua Bengio, Rafal Bogacz, Alex Christensen, Claudia Clopath, Rui P. Costa, Archy O. de Berker, Surya Ganguli, et al. A deep learning framework for neuroscience.Nature Neuroscience, 22(11):1761–1770, 2019. doi: 10.1038/s41593-019-0520-2

work page doi:10.1038/s41593-019-0520-2 2019
[21]

Deep neural networks: a new framework for modeling biological vision and brain information processing.Annual Review of Vision Science, 1:417–446,

Nikolaus Kriegeskorte. Deep neural networks: a new framework for modeling biological vision and brain information processing.Annual Review of Vision Science, 1:417–446,

work page
[22]

doi: 10.1146/annurev-vision-082114-035447

work page doi:10.1146/annurev-vision-082114-035447
[23]

H. B. Barlow. Possible principles underlying the transformations of sensory messages. In W. A. Rosenblith, editor,Sensory Communication, pages 217–234. MIT Press, Cambridge, MA, 1961

work page 1961
[24]

David J. Field. Relations between the statistics of natural images and the response prop- erties of cortical cells.Journal of the Optical Society of America A, 4(12):2379–2394, 1987. doi: 10.1364/JOSAA.4.002379

work page doi:10.1364/josaa.4.002379 1987
[26]

Firing patterns of single units in the prefrontal cortex and neural network models.Network: Computation in Neural Systems, 1(1):13–25, 1990

Moshe Abeles, Eilon Vaadia, and Hagai Bergman. Firing patterns of single units in the prefrontal cortex and neural network models.Network: Computation in Neural Systems, 1(1):13–25, 1990. doi: 10.1088/0954-898X 1 1 002

work page doi:10.1088/0954-898x 1990
[27]

Roy, Maximilian Riesenhuber, Tomaso Poggio, and Earl K

Jefferson E. Roy, Maximilian Riesenhuber, Tomaso Poggio, and Earl K. Miller. Prefrontal cortex activity during flexible categorization.The Journal of Neuroscience, 30(25):8519– 8528, 2010. doi: 10.1523/JNEUROSCI.4837-09.2010

work page doi:10.1523/jneurosci.4837-09.2010 2010
[28]

Rikhye, Aditya Gilra, and Michael M

Rajeev V. Rikhye, Aditya Gilra, and Michael M. Halassa. Thalamic regulation of switching between cortical representations enables cognitive flexibility.Nature Neuroscience, 21(12): 1753–1763, 2018. doi: 10.1038/s41593-018-0269-z

work page doi:10.1038/s41593-018-0269-z 2018
[29]

Buonomano and Wolfgang Maass

Dean V. Buonomano and Wolfgang Maass. State-dependent computations: Spatiotemporal processing in cortical networks.Nature Reviews Neuroscience, 10(2):113–125, 2009. doi: 10.1038/nrn2558. 20

work page doi:10.1038/nrn2558 2009
[30]

Murray, Alberto Bernacchia, Nicholas A

John D. Murray, Alberto Bernacchia, Nicholas A. Roy, Christos Constantinidis, Ranulfo Romo, Xiao-Jing Wang, and Amy F. T. Arnsten. Stable population coding for working memory coexists with heterogeneous neural dynamics in prefrontal cortex.Proceedings of the National Academy of Sciences of the United States of America, 114(2):394–399, 2017. doi: 10.1073/p...

work page doi:10.1073/pnas.1619449114 2017
[31]

Tem- poral coding carries more stable cortical visual representations than firing rate over time

Hanlin Zhu, Fei He, Pavlo Zolotavin, Xin Yao, Alexandre Pouget, and Rafael Yuste. Tem- poral coding carries more stable cortical visual representations than firing rate over time. Nature Communications, 16:7162, 2025. doi: 10.1038/s41467-025-62069-2

work page doi:10.1038/s41467-025-62069-2 2025
[32]

Kistler.Spiking Neuron Models: Single Neurons, Popula- tions, Plasticity

Wulfram Gerstner and Werner M. Kistler.Spiking Neuron Models: Single Neurons, Popula- tions, Plasticity. Cambridge University Press, Cambridge, UK, 2002. ISBN 9780521890793. doi: 10.1017/CBO9780511815706

work page doi:10.1017/cbo9780511815706 2002
[33]

Networks of spiking neurons: the third generation of neural network models.Neural Networks, 10(9):1659–1671, 1997

Wolfgang Maass. Networks of spiking neurons: the third generation of neural network models.Neural Networks, 10(9):1659–1671, 1997. doi: 10.1016/S0893-6080(97)00011-7

work page doi:10.1016/s0893-6080(97)00011-7 1997
[34]

Efficient neuromorphic signal processing with loihi 2

Garrick Orchard, E Paxon Frady, Daniel Ben Dayan Rubin, Sophia Sanborn, Sumit Bam Shrestha, Friedrich T Sommer, and Mike Davies. Efficient neuromorphic signal processing with loihi 2. In2021 IEEE Workshop on Signal Processing Systems (SiPS), pages 254–259. IEEE, 2021

work page 2021
[35]

Neuromorphic silicon neuron circuits

Giacomo Indiveri, Bernab´ e Linares-Barranco, et al. Neuromorphic silicon neuron circuits. Frontiers in Neuroscience, 5:73, 2011. doi: 10.3389/fnins.2011.00073

work page doi:10.3389/fnins.2011.00073 2011
[36]

Hebbian deep learning without feedback.arXiv preprint arXiv:2209.11883, 2022

Adrien Journ´ e, Hector Garcia Rodriguez, Qinghai Guo, and Timoleon Moraitis. Hebbian deep learning without feedback.arXiv preprint arXiv:2209.11883, 2022

work page arXiv 2022
[37]

Towards artificial general intelligence with hybrid tianjic chip architecture.Nature, 572(7767):106–111, 2019

Jing Pei, Lei Deng, Sen Song, Mingguo Zhao, Youhui Zhang, Shuang Wu, Guanrui Wang, Zhe Zou, Zhenzhi Wu, Wei He, et al. Towards artificial general intelligence with hybrid tianjic chip architecture.Nature, 572(7767):106–111, 2019

work page 2019
[38]

Spinnaker2: A large-scale neuromorphic system for event-based and asynchronous machine learning.arXiv preprint arXiv:2401.04491, 2024

Hector A Gonzalez, Jiaxin Huang, Florian Kelber, Khaleelulla Khan Nazeer, Tim Langer, Chen Liu, Matthias Lohrmann, Amirhossein Rostami, Mark Sch¨ one, Bernhard Vogginger, et al. Spinnaker2: A large-scale neuromorphic system for event-based and asynchronous machine learning.arXiv preprint arXiv:2401.04491, 2024

work page arXiv 2024
[39]

Questioning the role of sparse coding in the brain

Anton Spanne and Henrik J¨ orntell. Questioning the role of sparse coding in the brain. Trends in Neurosciences, 38(7):417–427, 2015. doi: 10.1016/j.tins.2015.05.005

work page doi:10.1016/j.tins.2015.05.005 2015
[40]

Synaptic reverberation underlying mnemonic persistent activity.Trends in Neurosciences, 24(8):455–463, 2001

Xiao-Jing Wang. Synaptic reverberation underlying mnemonic persistent activity.Trends in Neurosciences, 24(8):455–463, 2001

work page 2001
[41]

A hierarchy of intrinsic timescales across primate cortex.Nature neuroscience, 17(12):1661– 1663, 2014

John D Murray, Alberto Bernacchia, David J Freedman, Ranulfo Romo, Jonathan D Wallis, Xinying Cai, Camillo Padoa-Schioppa, Tatiana Pasternak, Hyojung Seo, Daeyeol Lee, et al. A hierarchy of intrinsic timescales across primate cortex.Nature neuroscience, 17(12):1661– 1663, 2014

work page 2014
[42]

Mouse pre- frontal cortex represents learned rules for categorization.Nature, 593(7859):411–417, 2021

Sandra Reinert, Mark H¨ ubener, Tobias Bonhoeffer, and Pieter M Goltstein. Mouse pre- frontal cortex represents learned rules for categorization.Nature, 593(7859):411–417, 2021

work page 2021
[43]

Loss of plasticity in deep continual learning.Nature, 632(8026):768–774, 2024

Shibhansh Dohare, J Fernando Hernandez-Garcia, Qingfeng Lan, Parash Rahman, A Ru- pam Mahmood, and Richard S Sutton. Loss of plasticity in deep continual learning.Nature, 632(8026):768–774, 2024. 21

work page 2024
[44]

Ternary spike: Learning ternary spikes for spiking neural networks

Yufei Guo, Yuanpei Chen, Xiaode Liu, Weihang Peng, Yuhan Zhang, Xuhui Huang, and Zhe Ma. Ternary spike: Learning ternary spikes for spiking neural networks. InProceedings of the AAAI conference on artificial intelligence, volume 38, pages 12244–12252, 2024

work page 2024
[45]

Three types of incremental learning.Nature Machine Intelligence, 4(12):1185–1197, 2022

Gido M Van de Ven, Tinne Tuytelaars, and Andreas S Tolias. Three types of incremental learning.Nature Machine Intelligence, 4(12):1185–1197, 2022

work page 2022
[46]

Humans and neural networks show similar patterns of transfer and interference during continual learning.Nature Human Behaviour, pages 1–15, 2025

Eleanor Holton, Lukas Braun, Jessica AF Thompson, Jan Grohn, and Christopher Sum- merfield. Humans and neural networks show similar patterns of transfer and interference during continual learning.Nature Human Behaviour, pages 1–15, 2025

work page 2025
[47]

Gradient-based learning applied to document recognition.Proceedings of the IEEE, 86(11):2278–2324, 2002

Yann LeCun, L´ eon Bottou, Yoshua Bengio, and Patrick Haffner. Gradient-based learning applied to document recognition.Proceedings of the IEEE, 86(11):2278–2324, 2002

work page 2002
[48]

Kistler, Richard Naud, and Liam Paninski.Neuronal Dy- namics: From Single Neurons to Networks and Models of Cognition

Wulfram Gerstner, Werner M. Kistler, Richard Naud, and Liam Paninski.Neuronal Dy- namics: From Single Neurons to Networks and Models of Cognition. Cambridge University Press, Cambridge, UK, 2014. ISBN 9781107060838. doi: 10.1017/CBO9781107447615

work page doi:10.1017/cbo9781107447615 2014
[49]

Sensorimotor remapping drives task specialization in prefrontal cortex.bioRxiv, pages 2025–09, 2025

Hugo Tissot, Jeffrey Boucher, Sandra Reinert, Pieter M Goltstein, and Yves Boubenec. Sensorimotor remapping drives task specialization in prefrontal cortex.bioRxiv, pages 2025–09, 2025

work page 2025
[50]

Biological underpinnings for lifelong learning machines.Nature Machine Intelligence, 4(3):196–210, 2022

Dhireesha Kudithipudi, Mario Aguilar-Simon, Jonathan Babb, Maxim Bazhenov, Douglas Blackiston, Josh Bongard, Andrew P Brna, Suraj Chakravarthi Raja, Nick Cheney, Jeff Clune, et al. Biological underpinnings for lifelong learning machines.Nature Machine Intelligence, 4(3):196–210, 2022

work page 2022
[51]

Synaptic computation.Nature, 431(7010):796–803, 2004

Larry F Abbott and Wade G Regehr. Synaptic computation.Nature, 431(7010):796–803, 2004

work page 2004
[52]

A hierarchy of temporal receptive windows in human cortex.Journal of neuroscience, 28(10):2539–2550, 2008

Uri Hasson, Eunice Yang, Ignacio Vallines, David J Heeger, and Nava Rubin. A hierarchy of temporal receptive windows in human cortex.Journal of neuroscience, 28(10):2539–2550, 2008

work page 2008
[53]

Conversion of continuous-valued deep networks to efficient event-driven networks for image classification.Frontiers in neuroscience, 11:682, 2017

Bodo Rueckauer, Iulia-Alexandra Lungu, Yuhuang Hu, Michael Pfeiffer, and Shih-Chii Liu. Conversion of continuous-valued deep networks to efficient event-driven networks for image classification.Frontiers in neuroscience, 11:682, 2017

work page 2017
[54]

Spiking neural net- works and their applications: A review.Brain sciences, 12(7):863, 2022

Kashu Yamazaki, Viet-Khoa Vo-Ho, Darshan Bulsara, and Ngan Le. Spiking neural net- works and their applications: A review.Brain sciences, 12(7):863, 2022. 22 Methods Neural experiment details Animals We analyzed mPFC population recordings from adult female C57BL/6 mice engaged in a rule- based Go/NoGo visual categorization task, using data from a previous...

work page 2022