Nicolas Brunel
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Nicolas Brunel has a total of 91 co-authors. Here are the top co-authors by number of shared publications:
- Vincent Hakim
Janelia Research Campus
11 shared publications - Alessandro Sanzeni
Bocconi University
9 shared publications - Mark H. Histed
Intramural Program, National Institute of Mental Health, National Institutes of Health, Bethesda, MD, USA
7 shared publications - Maxwell Gillett
Department of Neurobiology, Duke University
6 shared publications - Johnatan Aljadeff
Department of Neurobiology, University of California San Diego, La Jolla, USA
4 shared publications Ulises Pereira
Department of Statistics, The University of Chicago, Chicago, IL 60637
3 shared publications
Latest Publications
- 18 August 2026
- Published by American Physical Society (APS) in Physical Review Letters
- Vol. 137 (8) , 088401
Paradoxical Increase of Capacity due to Spurious Overlaps in Attractor Networks
In Hopfield-type associative memory models, memories are stored in the connectivity matrix and can be retrieved subsequently thanks to the collective dynamics of the network. In these models, the retrieval of a particular memory can be hampered by overlaps between the network state and other memories, termed ``spurious overlaps'' since these overlaps collectively introduce noise in the retrieval process. In classic models, spurious overlaps increase the variance of synaptic inputs but do not affect the mean. We show here that, in models equipped with a learning rule inferred from neurobiological data, spurious overlaps collectively reduce the mean synaptic inputs to neurons, and that this mean reduction causes in turn an increase in storage capacity through a sparsening of network activity. Our Letter demonstrates a link between a specific feature of experimentally inferred plasticity rules and network storage capacity.
- 18 August 2026
- Published by American Physical Society (APS) in Physical Review Letters
- Vol. 137 (8) , 088401
Paradoxical Increase of Capacity due to Spurious Overlaps in Attractor Networks
In Hopfield-type associative memory models, memories are stored in the connectivity matrix and can be retrieved subsequently thanks to the collective dynamics of the network. In these models, the retrieval of a particular memory can be hampered by overlaps between the network state and other memories, termed ``spurious overlaps'' since these overlaps collectively introduce noise in the retrieval process. In classic models, spurious overlaps increase the variance of synaptic inputs but do not affect the mean. We show here that, in models equipped with a learning rule inferred from neurobiological data, spurious overlaps collectively reduce the mean synaptic inputs to neurons, and that this mean reduction causes in turn an increase in storage capacity through a sparsening of network activity. Our Letter demonstrates a link between a specific feature of experimentally inferred plasticity rules and network storage capacity.
- Open Access
A Spatially Structured Spiking Network Model of Beta Traveling Waves and Their Attenuation in Motor Cortex
Beta-band oscillations in primate motor cortex propagate as planar traveling waves whose amplitude attenuates with spatial gradients across the cortical sheet just before movement onset. How local excitatory–inhibitory (E–I) interactions and spatial connectivity jointly generate these waves, their attenuation patterns, and their stereotyped rostro–caudal bias remains unclear. Here we address this question by implementing a spatially structured network of leaky integrate-and-fire neurons with distance-dependent connectivity, conduction delays, and realistic synaptic dynamics. Through linear stability analysis and large-scale simulations validated against macaque electrophysiology, we show that planar beta waves emerge as Turing–Hopf spatiotemporal instabilities, where global beta oscillations coexist with irregular single-neuron firing. When the network operates near the boundary between oscillatory and asynchronous regimes, internally generated fluctuations produce the irregular, transient beta bursts characteristic of single-trial local field potentials. A rapid, spatially homogeneous increase in external drive pushes the circuit into an asynchronous state, reproducing the beta power reduction and spatial attenuation gradients seen at movement onset, alongside the irregular spatiotemporal dynamics of movement execution. By introducing anisotropic excitatory-to-excitatory connectivity, we recover the observed rostro–caudal propagation bias. Our results suggest that motor cortical traveling waves are intrinsic dynamical modes of local E–I circuits, recruited and modulated by behaviorally relevant inputs to organize movement initiation.
Open AccessA Spatially Structured Spiking Network Model of Beta Traveling Waves and Their Attenuation in Motor Cortex
Beta-band oscillations in primate motor cortex propagate as planar traveling waves whose amplitude attenuates with spatial gradients across the cortical sheet just before movement onset. How local excitatory–inhibitory (E–I) interactions and spatial connectivity jointly generate these waves, their attenuation patterns, and their stereotyped rostro–caudal bias remains unclear. Here we address this question by implementing a spatially structured network of leaky integrate-and-fire neurons with distance-dependent connectivity, conduction delays, and realistic synaptic dynamics. Through linear stability analysis and large-scale simulations validated against macaque electrophysiology, we show that planar beta waves emerge as Turing–Hopf spatiotemporal instabilities, where global beta oscillations coexist with irregular single-neuron firing. When the network operates near the boundary between oscillatory and asynchronous regimes, internally generated fluctuations produce the irregular, transient beta bursts characteristic of single-trial local field potentials. A rapid, spatially homogeneous increase in external drive pushes the circuit into an asynchronous state, reproducing the beta power reduction and spatial attenuation gradients seen at movement onset, alongside the irregular spatiotemporal dynamics of movement execution. By introducing anisotropic excitatory-to-excitatory connectivity, we recover the observed rostro–caudal propagation bias. Our results suggest that motor cortical traveling waves are intrinsic dynamical modes of local E–I circuits, recruited and modulated by behaviorally relevant inputs to organize movement initiation.
- Open Access
- 10 October 2025
- Published by National Academy of Sciences in Proceedings of the National Academy of Sciences
- Vol. 122 (41)
Inter- and intrahemispheric sources of vestibular signals to V1
Head movements are sensed by the vestibular organs. Unlike classical senses, signals from vestibular organs are not selectively conveyed to a dedicated cortical area but are broadcast throughout the cortex. This distributed processing pattern reflects the fundamental role of vestibular information in contextual modulation across diverse cortical computations. Surprisingly, the routes taken by vestibular signals to reach the cortex are still largely uncharted. Here, we show that the primary visual cortex (V1) receives real-time head movement signals—direction, velocity, and acceleration—from the ipsilateral pulvinar and contralateral visual cortex (VC). The ipsilateral pulvinar provides the main head movement signal, with a bias toward contraversive movements (e.g., clockwise movements in left V1). Conversely, the contralateral VC provides head movement signals during ipsiversive movements. Crucially, head movement variables encoded in V1 are already encoded in the pulvinar, suggesting that those variables are computed subcortically. Thus, the convergence of inter- and intrahemispheric signals endows V1 with a rich representation of the animal’s head movements.
- 10 October 2025
- Published by National Academy of Sciences in Proceedings of the National Academy of Sciences
- Vol. 122 (41)
Open AccessInter- and intrahemispheric sources of vestibular signals to V1
Head movements are sensed by the vestibular organs. Unlike classical senses, signals from vestibular organs are not selectively conveyed to a dedicated cortical area but are broadcast throughout the cortex. This distributed processing pattern reflects the fundamental role of vestibular information in contextual modulation across diverse cortical computations. Surprisingly, the routes taken by vestibular signals to reach the cortex are still largely uncharted. Here, we show that the primary visual cortex (V1) receives real-time head movement signals—direction, velocity, and acceleration—from the ipsilateral pulvinar and contralateral visual cortex (VC). The ipsilateral pulvinar provides the main head movement signal, with a bias toward contraversive movements (e.g., clockwise movements in left V1). Conversely, the contralateral VC provides head movement signals during ipsiversive movements. Crucially, head movement variables encoded in V1 are already encoded in the pulvinar, suggesting that those variables are computed subcortically. Thus, the convergence of inter- and intrahemispheric signals endows V1 with a rich representation of the animal’s head movements.
- Open Access
- 16 June 2025
- Published by American Physical Society (APS) in PRX Life
- Vol. 3 (3)
Impact of dendritic non-linearities on the computational capabilities of neurons
How neurons integrate the myriad synaptic inputs scattered across their dendrites is a fundamental question in neuroscience. Multiple neurophysiological experiments have shown that dendritic nonlinearities can have a strong influence on synaptic input integration. These nonlinearities have motivated mathematical descriptions of single neurons as a two-layer computational units, which have been shown to increase substantially the computational abilities of neurons, compared to linear dendritic integration. However, current analytical studies are restricted to neurons with unconstrained synaptic weights and unplausible dendritic nonlinearities. Here we introduce a two-layer model with sign-constrained synaptic weights and a biologically plausible form of dendritic nonlinearity and investigate its properties using both statistical physics methods and numerical simulations. We find that the dendritic nonlinearity enhances both the number of possible learned input-output associations and the learning velocity. We characterize how capacity and learning speed depend on the implemented nonlinearity and the levels of dendritic and somatic inhibition. We calculate analytically the distribution of synaptic weights in networks close to maximal capacity and find that the dendritic nonlinearity increases the fraction of zero-weight (“silent” or “potential”) synapses, compared with the standard perceptron model, when no or weak robustness constraints are present, while the opposite occurs with strong robustness constraints. We test our model on standard real-world benchmark datasets and observe empirically that the nonlinearity provides an enhancement in generalization performance and that it enables to capture more complex input-output relations, compared to the perceptron model.
- 16 June 2025
- Published by American Physical Society (APS) in PRX Life
- Vol. 3 (3)
Open AccessImpact of dendritic non-linearities on the computational capabilities of neurons
How neurons integrate the myriad synaptic inputs scattered across their dendrites is a fundamental question in neuroscience. Multiple neurophysiological experiments have shown that dendritic nonlinearities can have a strong influence on synaptic input integration. These nonlinearities have motivated mathematical descriptions of single neurons as a two-layer computational units, which have been shown to increase substantially the computational abilities of neurons, compared to linear dendritic integration. However, current analytical studies are restricted to neurons with unconstrained synaptic weights and unplausible dendritic nonlinearities. Here we introduce a two-layer model with sign-constrained synaptic weights and a biologically plausible form of dendritic nonlinearity and investigate its properties using both statistical physics methods and numerical simulations. We find that the dendritic nonlinearity enhances both the number of possible learned input-output associations and the learning velocity. We characterize how capacity and learning speed depend on the implemented nonlinearity and the levels of dendritic and somatic inhibition. We calculate analytically the distribution of synaptic weights in networks close to maximal capacity and find that the dendritic nonlinearity increases the fraction of zero-weight (“silent” or “potential”) synapses, compared with the standard perceptron model, when no or weak robustness constraints are present, while the opposite occurs with strong robustness constraints. We test our model on standard real-world benchmark datasets and observe empirically that the nonlinearity provides an enhancement in generalization performance and that it enables to capture more complex input-output relations, compared to the perceptron model.
- Open Access
- 3 June 2025
- Published by MathDoc/Centre Mersenne in Comptes Rendus. Physique
- Vol. 26 (G1) , 463-477
From the perceptron to the cerebellum
The perceptron has served as a prototypical neuronal learning machine in the physics community interested in neural networks and artificial intelligence, which included Gérard Toulouse as one of its prominent figures. It has also been used as a model of Purkinje cells of the cerebellum, a brain structure involved in motor learning, in the early influential theories of David Marr and James Albus. We review these theories, more recent developments in the field, and highlight questions of current interest.
- 3 June 2025
- Published by MathDoc/Centre Mersenne in Comptes Rendus. Physique
- Vol. 26 (G1) , 463-477
Open AccessFrom the perceptron to the cerebellum
The perceptron has served as a prototypical neuronal learning machine in the physics community interested in neural networks and artificial intelligence, which included Gérard Toulouse as one of its prominent figures. It has also been used as a model of Purkinje cells of the cerebellum, a brain structure involved in motor learning, in the early influential theories of David Marr and James Albus. We review these theories, more recent developments in the field, and highlight questions of current interest.
- Open Access
Inter- and Intrahemispheric Sources of Vestibular Signals to V1
Head movements are sensed by the vestibular organs. Unlike classical senses, signals from vestibular organs are not conveyed to a dedicated cortical area but are broadcast throughout the cortex. Surprisingly, the routes taken by vestibular signals to reach the cortex are still largely uncharted. Here we show that the primary visual cortex (V1) receives real-time head movement signals — direction, velocity, and acceleration — from the ipsilateral pulvinar and contralateral visual cortex. The ipsilateral pulvinar provides the main head movement signal, with a bias toward contraversive movements (e.g. clockwise movements in left V1). Conversely, the contralateral visual cortex provides head movement signals during ipsiversive movements. Crucially, head movement variables encoded in V1 are already encoded in the pulvinar, suggesting that those variables are computed subcortically. Thus, the convergence of inter- and intrahemispheric signals endows V1 with a rich representation of the animal’s head movements.
Open AccessInter- and Intrahemispheric Sources of Vestibular Signals to V1
Head movements are sensed by the vestibular organs. Unlike classical senses, signals from vestibular organs are not conveyed to a dedicated cortical area but are broadcast throughout the cortex. Surprisingly, the routes taken by vestibular signals to reach the cortex are still largely uncharted. Here we show that the primary visual cortex (V1) receives real-time head movement signals — direction, velocity, and acceleration — from the ipsilateral pulvinar and contralateral visual cortex. The ipsilateral pulvinar provides the main head movement signal, with a bias toward contraversive movements (e.g. clockwise movements in left V1). Conversely, the contralateral visual cortex provides head movement signals during ipsiversive movements. Crucially, head movement variables encoded in V1 are already encoded in the pulvinar, suggesting that those variables are computed subcortically. Thus, the convergence of inter- and intrahemispheric signals endows V1 with a rich representation of the animal’s head movements.
- Open Access
- 28 August 2024
- Published by eLife Sciences Publications, Ltd in eLife
- Vol. 12
Dynamic control of sequential retrieval speed in networks with heterogeneous learning rules
Temporal rescaling of sequential neural activity has been observed in multiple brain areas during behaviors involving time estimation and motor execution at variable speeds. Temporally asymmetric Hebbian rules have been used in network models to learn and retrieve sequential activity, with characteristics that are qualitatively consistent with experimental observations. However, in these models sequential activity is retrieved at a fixed speed. Here, we investigate the effects of a heterogeneity of plasticity rules on network dynamics. In a model in which neurons differ by the degree of temporal symmetry of their plasticity rule, we find that retrieval speed can be controlled by varying external inputs to the network. Neurons with temporally symmetric plasticity rules act as brakes and tend to slow down the dynamics, while neurons with temporally asymmetric rules act as accelerators of the dynamics. We also find that such networks can naturally generate separate ‘preparatory’ and ‘execution’ activity patterns with appropriate external inputs.
- 28 August 2024
- Published by eLife Sciences Publications, Ltd in eLife
- Vol. 12
Open AccessDynamic control of sequential retrieval speed in networks with heterogeneous learning rules
Temporal rescaling of sequential neural activity has been observed in multiple brain areas during behaviors involving time estimation and motor execution at variable speeds. Temporally asymmetric Hebbian rules have been used in network models to learn and retrieve sequential activity, with characteristics that are qualitatively consistent with experimental observations. However, in these models sequential activity is retrieved at a fixed speed. Here, we investigate the effects of a heterogeneity of plasticity rules on network dynamics. In a model in which neurons differ by the degree of temporal symmetry of their plasticity rule, we find that retrieval speed can be controlled by varying external inputs to the network. Neurons with temporally symmetric plasticity rules act as brakes and tend to slow down the dynamics, while neurons with temporally asymmetric rules act as accelerators of the dynamics. We also find that such networks can naturally generate separate ‘preparatory’ and ‘execution’ activity patterns with appropriate external inputs.
- Open Access
- 11 July 2024
- Published by eLife Sciences Publications, Ltd in eLife preprints
- Vol. 12
Dynamic control of sequential retrieval speed in networks with heterogeneous learning rules
Temporal rescaling of sequential neural activity has been observed in multiple brain areas during behaviors involving time estimation and motor execution at variable speeds. Temporally asymmetric Hebbian rules have been used in network models to learn and retrieve sequential activity, with characteristics that are qualitatively consistent with experimental observations. However, in these models sequential activity is retrieved at a fixed speed. Here, we investigate the effects of a heterogeneity of plasticity rules on network dynamics. In a model in which neurons differ by the degree of temporal symmetry of their plasticity rule, we find that retrieval speed can be controlled by varying external inputs to the network. Neurons with temporally symmetric plasticity rules act as brakes and tend to slow down the dynamics, while neurons with temporally asymmetric rules act as accelerators of the dynamics. We also find that such networks can naturally generate separate ‘preparatory’ and ‘execution’ activity patterns with appropriate external inputs.
- 11 July 2024
- Published by eLife Sciences Publications, Ltd in eLife preprints
- Vol. 12
Open AccessDynamic control of sequential retrieval speed in networks with heterogeneous learning rules
Temporal rescaling of sequential neural activity has been observed in multiple brain areas during behaviors involving time estimation and motor execution at variable speeds. Temporally asymmetric Hebbian rules have been used in network models to learn and retrieve sequential activity, with characteristics that are qualitatively consistent with experimental observations. However, in these models sequential activity is retrieved at a fixed speed. Here, we investigate the effects of a heterogeneity of plasticity rules on network dynamics. In a model in which neurons differ by the degree of temporal symmetry of their plasticity rule, we find that retrieval speed can be controlled by varying external inputs to the network. Neurons with temporally symmetric plasticity rules act as brakes and tend to slow down the dynamics, while neurons with temporally asymmetric rules act as accelerators of the dynamics. We also find that such networks can naturally generate separate ‘preparatory’ and ‘execution’ activity patterns with appropriate external inputs.
- Open Access
- 7 February 2024
- Published by Public Library of Science (PLoS) in PLOS Computational Biology
- Vol. 20 (2) , e1011354
Attractor neural networks with double well synapses
It is widely believed that memory storage depends on activity-dependent synaptic modifications. Classical studies of learning and memory in neural networks describe synaptic efficacy either as continuous or discrete. However, recent results suggest an intermediate scenario in which synaptic efficacy can be described by a continuous variable, but whose distribution is peaked around a small set of discrete values. Motivated by these results, we explored a model in which each synapse is described by a continuous variable that evolves in a potential with multiple minima. External inputs to the network can switch synapses from one potential well to another. Our analytical and numerical results show that this model can interpolate between models with discrete synapses which correspond to the deep potential limit, and models in which synapses evolve in a single quadratic potential. We find that the storage capacity of the network with double well synapses exhibits a power law dependence on the network size, rather than the logarithmic dependence observed in models with single well synapses. In addition, synapses with deeper potential wells lead to more robust information storage in the presence of noise. When memories are sparsely encoded, the scaling of the capacity with network size is similar to previously studied network models in the sparse coding limit.
- 7 February 2024
- Published by Public Library of Science (PLoS) in PLOS Computational Biology
- Vol. 20 (2) , e1011354
Open AccessAttractor neural networks with double well synapses
It is widely believed that memory storage depends on activity-dependent synaptic modifications. Classical studies of learning and memory in neural networks describe synaptic efficacy either as continuous or discrete. However, recent results suggest an intermediate scenario in which synaptic efficacy can be described by a continuous variable, but whose distribution is peaked around a small set of discrete values. Motivated by these results, we explored a model in which each synapse is described by a continuous variable that evolves in a potential with multiple minima. External inputs to the network can switch synapses from one potential well to another. Our analytical and numerical results show that this model can interpolate between models with discrete synapses which correspond to the deep potential limit, and models in which synapses evolve in a single quadratic potential. We find that the storage capacity of the network with double well synapses exhibits a power law dependence on the network size, rather than the logarithmic dependence observed in models with single well synapses. In addition, synapses with deeper potential wells lead to more robust information storage in the presence of noise. When memories are sparsely encoded, the scaling of the capacity with network size is similar to previously studied network models in the sparse coding limit.
Mechanisms underlying reshuffling of visual responses by optogenetic stimulation in mice and monkeys