@article{21895, abstract = {The mammalian brain organises knowledge about entities in the world and relationships between them using cognitive maps. When forming a cognitive map, there is a necessary trade-off between extending the map to make novel inferences, and storing a veridical copy of past experience. However, the neural mechanisms that control this trade-off remain unknown. Using a cross-scale approach that combines a pharmacological intervention in humans with neural network modelling, we show that the neuromodulator noradrenaline elicits a significant ‘spread of association’ across hippocampal cognitive maps. This neural spread of association can be explained by changes in synaptic plasticity that predict overgeneralisation in behaviour. Thus, elevated noradrenaline during learning increases the ‘smoothing kernel’ for plasticity across the cognitive map, allowing disparate memories to become linked and distorted.}, author = {Koolschijn, Renée S. and Parthasarathy, Prakriti and Browning, Michael and Przygodda, Xenia and Capitão, Liliana P. and Clarke, William T. and Vogels, Tim P and O’Reilly, Jill X. and Barron, Helen C.}, issn = {2041-1723}, journal = {Nature Communications}, publisher = {Springer Nature}, title = {{Noradrenaline causes a spread of association in the hippocampal cognitive map}}, doi = {10.1038/s41467-026-70659-x}, volume = {17}, year = {2026}, } @article{21378, abstract = {From insects to mammals, essential brain functions, such as forming long-term memories (LTMs), increase metabolic activity in stimulated neurons to meet the energetic demand associated with brain activation. However, while impairing neuronal metabolism limits brain performance, whether expanding the metabolic capacity of neurons boosts brain function remains poorly understood. Here, we show that LTM formation of flies and mice can be enhanced by increasing mitochondrial metabolism in central memory circuits. By knocking down the mitochondrial Ca2+ exporter Letm1, we favour Ca2+ retention in the mitochondrial matrix of neurons due to reduction of mitochondrial H+/Ca2+ exchange. The resulting increase in mitochondrial Ca2+ over-activates mitochondrial metabolism in neurons of central memory circuits, leading to improved LTM storage in training paradigms in which wild-type counterparts of both species fail to remember. Our findings unveil an evolutionarily conserved mechanism that controls mitochondrial metabolism in neurons and indicate its involvement in shaping higher brain functions, such as LTM.}, author = {Amrapali Vishwanath, Anjali and Comyn, Typhaine and Mira, Rodrigo G. and Brossier, Claire and Pascual-Caro, Carlos and Faour, Maya and Boumendil, Kahina and Chintaluri, Chaitanya and Ramon-Duaso, Carla and Fan, Ruolin and Ghosh, Kishalay and Farrants, Helen and Berwick, Jean-Paul and Sivakumar, Riya and Lopez-Manzaneda, Mario and Schreiter, Eric R. and Preat, Thomas and Vogels, Tim P and Rangaraju, Vidhya and Busquets-Garcia, Arnau and Plaçais, Pierre-Yves and Pavlowsky, Alice and de Juan-Sanz, Jaime}, issn = {2522-5812}, journal = {Nature Metabolism}, number = {2}, pages = {467--488}, publisher = {Springer Nature}, title = {{Mitochondrial Ca2+ efflux controls neuronal metabolism and long-term memory across species}}, doi = {10.1038/s42255-026-01451-w}, volume = {8}, year = {2026}, } @article{20986, abstract = {During complex vocal interactions, different features of acoustic stimuli are integrated to produce appropriate vocal responses,1 such as copying sounds during vocal matching behavior in some animals.2,3,4,5,6,7,8,9,10,11,12 However, little is known about the interplay and possible trade-offs between the different temporal and spectral acoustic features during these vocal exchanges.2,13,14 Nightingales can flexibly match the pitch of their tonal “whistle songs” in real time during counter-singing duels.15,16 Here, we show that the syllable duration of whistle playbacks could alter the song responses of wild nightingales, causing their whistle duration distribution to shift toward the presented stimulus duration. When exposed to whistle playbacks featuring unnatural combinations of pitch and duration, nightingales demonstrate a flexible trade-off between pitch matching and temporal imitation, yet they are constrained by their vocal repertoire. They selectively adapted their vocal responses to approximate these novel stimuli, aligning them with their natural whistle repertoire. We developed a computational model of nightingale whistle-matching behavior that revealed a hierarchical organization of acoustic feature production. During whistle matching, the feature integration process is constrained by the duration of syllables, and pitch matching follows within this temporal framework, forcing a trade-off between the two features. Our findings reveal a complex interplay between the spectral and temporal domains that shapes song-matching behavior.}, author = {Calderon Garcia, Juan Sebastian and Costalunga, Giacomo and Vogels, Tim P and Vallentin, Daniela}, issn = {1879-0445}, journal = {Current Biology}, number = {3}, pages = {791--798.e6}, publisher = {Elsevier}, title = {{Interplay between syllable duration and pitch during whistle matching in wild nightingales}}, doi = {10.1016/j.cub.2025.12.025}, volume = {36}, year = {2026}, } @article{22295, abstract = {Despite the functional diversity of over 100 causal genes1,2,3, phenotypic convergence across models may reveal common neurobiological processes in autism spectrum disorder (ASD). Here we profiled 251 samples from 11 monogenic mouse models of ASD using single-nucleus multi-omic sequencing across three developmental stages, both sexes and two brain regions. Despite genetic heterogeneity, ASD-linked mutations converged on perturbations of the radial glial cell lineage. These alterations reflect a transient developmental delay rather than lasting lineage misspecification and resolve by postnatal stages. Molecularly, the largest transcriptional differences emerged in neurons at early postnatal stages. These changes included downregulation of synaptic and ion channel-related genes, consistent with homeostatic adaptation or delayed maturation. Network analysis showed molecular convergence across models within each developmental stage, suggesting that diverse mutations linked to ASD impinge on common, stage-specific processes. Convergence becomes less pronounced by postnatal day 14, highlighting the dynamic nature of ASD-associated changes. Cross-genotype heterogeneity is superimposed on stage-specific effects. Electrophysiology corroborated this pattern: mutants generally showed altered neuronal excitability and synaptic properties with model-specific nuances. Our study also highlighted sex-specific gene expression alterations, with female mice often displaying larger effect sizes than male mice. Together, our findings provide a comprehensive view of developmental cellular and molecular dynamics across models of ASD.}, author = {Schwarz, Lena A and Dotter, Christoph and Isaev, Sergey and Lisi, Michela and Malzl, Daniel and Büschl, Christoph and Ladstätter, Sabrina and Oliveira, Bárbara and Barel, Matteo and Basilico, Bernadette and Chintaluri, Chaitanya and Gorkiewicz, Sarah and Goudarzi, Mohammad and Belinova, Tereza and Reichl, Stephan and Sendžikaitė, Gintarė and Arcot Jayaram, Satish and Koppensteiner, Peter and Sommer, Christoph M and Vogels, Tim P and Menche, Jörg and Adameyko, Igor and Kharchenko, Peter Vasili and Bock, Christoph and Novarino, Gaia}, issn = {1476-4687}, journal = {Nature}, publisher = {Springer Nature}, title = {{Cortical development dynamics across autism spectrum disorder mouse models}}, doi = {10.1038/s41586-026-10679-1}, year = {2026}, } @article{19640, abstract = {Synaptic plasticity is a key player in the brain’s life-long learning abilities. However, due to experimental limitations, the mechanistic link between synaptic plasticity rules and the network-level computations they enable remain opaque. Here we use evolutionary strategies (ES) to meta learn local co-active plasticity rules in large recurrent spiking networks with excitatory (E) and inhibitory (I) neurons, using parameterizations of increasing complexity. We discover rules that robustly stabilize network dynamics for all four synapse types acting in isolation (E-to-E, E-to-I, I-to-E and I-to-I). More complex functions such as familiarity detection can also be included in the search constraints. However, our meta learning strategy begins to fail for co-active rules of increasing complexity, as it is challenging to devise loss functions that effectively constrain network dynamics to plausible solutions a priori. Moreover, in line with previous work, we can find multiple degenerate solutions with identical network behaviour. As a local optimization strategy, ES provides one solution at a time and makes exploration of this degeneracy cumbersome. Regardless, we can glean the interdependecies of various plasticity parameters by considering the covariance matrix learned alongside the optimal rule with ES. Our work provides a proof of principle for the success of machine-learning-guided discovery of plasticity rules in large spiking networks, and points at the necessity of more elaborate search strategies going forward.}, author = {Confavreux, Basile J and Agnes, Everton J. and Zenke, Friedemann and Sprekeler, Henning and Vogels, Tim P}, issn = {1553-7358}, journal = {PLoS Computational Biology}, number = {4}, publisher = {Public Library of Science}, title = {{Balancing complexity, performance and plausibility to meta learn plasticity rules in recurrent spiking networks}}, doi = {10.1371/journal.pcbi.1012910}, volume = {21}, year = {2025}, } @article{8125, abstract = {Biological memory is known to be flexible—memory formation and recall depend on factors such as the behavioral context of the organism. However, this property is often ignored in associative memory models, leaving it unclear how memories can be organized and recalled when subject to contextual control. Because of the lack of a rigorous analytical framework, it is also unknown how contextual control affects memory stability, storage capacity, and information content. Here, we bring the dynamic nature of memory to the fore by introducing a novel model of associative memory, which we refer to as the context-modular memory network. In our model, stored memory patterns are associated to one of several background network states, or contexts. Memories are accessible when their corresponding context is active, and are otherwise inaccessible. Context modulates the effective network connectivity by imposing a specific configuration of neuronal and synaptic gating—gated neurons (synapses) have their activity (weights) momentarily silenced, thereby reducing interference from memories belonging to other contexts. Memory patterns are randomly and independently chosen, while neuronal and synaptic gates may be selected randomly or optimized through a process of contextual synaptic refinement. Through analytic and numerical results, we show that context-modular memory networks can exhibit both improved memory capacity and differential control of memory stability with random gating (especially for neuronal gating). For contextual synaptic refinement, we devise a method in which synapses are gated off for a given context if they destabilize the memory patterns in that context, drastically improving memory capacity and enabling even more precise control over memory stability. Notably, synaptic refinement allows for patterns to be accessible in multiple contexts, stabilizing memory patterns even for weight matrices that alone do not contain any information about the memory patterns, such as Gaussian random matrices. Overall, our model integrates recent ideas about context-dependent memory organization with classic associative memory models and proposes a rigorous theory which can act as a framework for future work. Furthermore, our work carries important implications for the understanding of biological memory storage and recall in the brain, such as highlighting an intriguing trade-off between memory capacity and accessibility.}, author = {Podlaski, William F. and Agnes, Everton J. and Vogels, Tim P}, issn = {2160-3308}, journal = {Physical Review X}, publisher = {American Physical Society}, title = {{High capacity and dynamic accessibility in associative memory networks with context-dependent neuronal and synaptic gating}}, doi = {10.1103/PhysRevX.15.011057}, volume = {15}, year = {2025}, } @article{14841, abstract = {De novo heterozygous variants in KCNC2 encoding the voltage-gated potassium (K+) channel subunit Kv3.2 are a recently described cause of developmental and epileptic encephalopathy (DEE). A de novo variant in KCNC2 c.374G > A (p.Cys125Tyr) was identified via exome sequencing in a patient with DEE. Relative to wild-type Kv3.2, Kv3.2-p.Cys125Tyr induces K+ currents exhibiting a large hyperpolarizing shift in the voltage dependence of activation, accelerated activation, and delayed deactivation consistent with a relative stabilization of the open conformation, along with increased current density. Leveraging the cryogenic electron microscopy (cryo-EM) structure of Kv3.1, molecular dynamic simulations suggest that a strong π-π stacking interaction between the variant Tyr125 and Tyr156 in the α-6 helix of the T1 domain promotes a relative stabilization of the open conformation of the channel, which underlies the observed gain of function. A multicompartment computational model of a Kv3-expressing parvalbumin-positive cerebral cortex fast-spiking γ-aminobutyric acidergic (GABAergic) interneuron (PV-IN) demonstrates how the Kv3.2-Cys125Tyr variant impairs neuronal excitability and dysregulates inhibition in cerebral cortex circuits to explain the resulting epilepsy.}, author = {Clatot, Jerome and Currin, Christopher and Liang, Qiansheng and Pipatpolkai, Tanadet and Massey, Shavonne L. and Helbig, Ingo and Delemotte, Lucie and Vogels, Tim P and Covarrubias, Manuel and Goldberg, Ethan M.}, issn = {1091-6490}, journal = {Proceedings of the National Academy of Sciences of the United States of America}, number = {3}, publisher = {National Academy of Sciences}, title = {{A structurally precise mechanism links an epilepsy-associated KCNC2 potassium channel mutation to interneuron dysfunction}}, doi = {10.1073/pnas.2307776121}, volume = {121}, year = {2024}, } @article{15171, abstract = {The brain’s functionality is developed and maintained through synaptic plasticity. As synapses undergo plasticity, they also affect each other. The nature of such ‘co-dependency’ is difficult to disentangle experimentally, because multiple synapses must be monitored simultaneously. To help understand the experimentally observed phenomena, we introduce a framework that formalizes synaptic co-dependency between different connection types. The resulting model explains how inhibition can gate excitatory plasticity while neighboring excitatory–excitatory interactions determine the strength of long-term potentiation. Furthermore, we show how the interplay between excitatory and inhibitory synapses can account for the quick rise and long-term stability of a variety of synaptic weight profiles, such as orientation tuning and dendritic clustering of co-active synapses. In recurrent neuronal networks, co-dependent plasticity produces rich and stable motor cortex-like dynamics with high input sensitivity. Our results suggest an essential role for the neighborly synaptic interaction during learning, connecting micro-level physiology with network-wide phenomena.}, author = {Agnes, Everton J. and Vogels, Tim P}, issn = {1546-1726}, journal = {Nature Neuroscience}, pages = {964--974}, publisher = {Springer Nature}, title = {{Co-dependent excitatory and inhibitory plasticity accounts for quick, stable and long-lasting memories in biological networks}}, doi = {10.1038/s41593-024-01597-4}, volume = {27}, year = {2024}, } @article{14666, abstract = {So-called spontaneous activity is a central hallmark of most nervous systems. Such non-causal firing is contrary to the tenet of spikes as a means of communication, and its purpose remains unclear. We propose that self-initiated firing can serve as a release valve to protect neurons from the toxic conditions arising in mitochondria from lower-than-baseline energy consumption. To demonstrate the viability of our hypothesis, we built a set of models that incorporate recent experimental results indicating homeostatic control of metabolic products—Adenosine triphosphate (ATP), adenosine diphosphate (ADP), and reactive oxygen species (ROS)—by changes in firing. We explore the relationship of metabolic cost of spiking with its effect on the temporal patterning of spikes and reproduce experimentally observed changes in intrinsic firing in the fruitfly dorsal fan-shaped body neuron in a model with ROS-modulated potassium channels. We also show that metabolic spiking homeostasis can produce indefinitely sustained avalanche dynamics in cortical circuits. Our theory can account for key features of neuronal activity observed in many studies ranging from ion channel function all the way to resting state dynamics. We finish with a set of experimental predictions that would confirm an integrated, crucial role for metabolically regulated spiking and firmly link metabolic homeostasis and neuronal function.}, author = {Chintaluri, Chaitanya and Vogels, Tim P}, issn = {1091-6490}, journal = {Proceedings of the National Academy of Sciences of the United States of America}, number = {48}, publisher = {National Academy of Sciences}, title = {{Metabolically regulated spiking could serve neuronal energy homeostasis and protect from reactive oxygen species}}, doi = {10.1073/pnas.2306525120}, volume = {120}, year = {2023}, } @article{11143, abstract = {Dravet syndrome is a neurodevelopmental disorder characterized by epilepsy, intellectual disability, and sudden death due to pathogenic variants in SCN1A with loss of function of the sodium channel subunit Nav1.1. Nav1.1-expressing parvalbumin GABAergic interneurons (PV-INs) from young Scn1a+/− mice show impaired action potential generation. An approach assessing PV-IN function in the same mice at two time points shows impaired spike generation in all Scn1a+/− mice at postnatal days (P) 16–21, whether deceased prior or surviving to P35, with normalization by P35 in surviving mice. However, PV-IN synaptic transmission is dysfunctional in young Scn1a+/− mice that did not survive and in Scn1a+/− mice ≥ P35. Modeling confirms that PV-IN axonal propagation is more sensitive to decreased sodium conductance than spike generation. These results demonstrate dynamic dysfunction in Dravet syndrome: combined abnormalities of PV-IN spike generation and propagation drives early disease severity, while ongoing dysfunction of synaptic transmission contributes to chronic pathology.}, author = {Kaneko, Keisuke and Currin, Christopher and Goff, Kevin M. and Wengert, Eric R. and Somarowthu, Ala and Vogels, Tim P and Goldberg, Ethan M.}, issn = {2211-1247}, journal = {Cell Reports}, number = {13}, publisher = {Elsevier}, title = {{Developmentally regulated impairment of parvalbumin interneuron synaptic transmission in an experimental model of Dravet syndrome}}, doi = {10.1016/j.celrep.2022.110580}, volume = {38}, year = {2022}, } @article{12009, abstract = {Changes in the short-term dynamics of excitatory synapses over development have been observed throughout cortex, but their purpose and consequences remain unclear. Here, we propose that developmental changes in synaptic dynamics buffer the effect of slow inhibitory long-term plasticity, allowing for continuously stable neural activity. Using computational modeling we demonstrate that early in development excitatory short-term depression quickly stabilises neural activity, even in the face of strong, unbalanced excitation. We introduce a model of the commonly observed developmental shift from depression to facilitation and show that neural activity remains stable throughout development, while inhibitory synaptic plasticity slowly balances excitation, consistent with experimental observations. Our model predicts changes in the input responses from phasic to phasic-and-tonic and more precise spike timings. We also observe a gradual emergence of short-lasting memory traces governed by short-term plasticity development. We conclude that the developmental depression-to-facilitation shift may control excitation-inhibition balance throughout development with important functional consequences.}, author = {Jia, David W. and Vogels, Tim P and Costa, Rui Ponte}, issn = {2399-3642}, journal = {Communications biology}, publisher = {Springer Nature}, title = {{Developmental depression-to-facilitation shift controls excitation-inhibition balance}}, doi = {10.1038/s42003-022-03801-2}, volume = {5}, year = {2022}, } @article{12084, abstract = {Neuronal networks encode information through patterns of activity that define the networks’ function. The neurons’ activity relies on specific connectivity structures, yet the link between structure and function is not fully understood. Here, we tackle this structure-function problem with a new conceptual approach. Instead of manipulating the connectivity directly, we focus on upper triangular matrices, which represent the network dynamics in a given orthonormal basis obtained by the Schur decomposition. This abstraction allows us to independently manipulate the eigenspectrum and feedforward structures of a connectivity matrix. Using this method, we describe a diverse repertoire of non-normal transient amplification, and to complement the analysis of the dynamical regimes, we quantify the geometry of output trajectories through the effective rank of both the eigenvector and the dynamics matrices. Counter-intuitively, we find that shrinking the eigenspectrum’s imaginary distribution leads to highly amplifying regimes in linear and long-lasting dynamics in nonlinear networks. We also find a trade-off between amplification and dimensionality of neuronal dynamics, i.e., trajectories in neuronal state-space. Networks that can amplify a large number of orthogonal initial conditions produce neuronal trajectories that lie in the same subspace of the neuronal state-space. Finally, we examine networks of excitatory and inhibitory neurons. We find that the strength of global inhibition is directly linked with the amplitude of amplification, such that weakening inhibitory weights also decreases amplification, and that the eigenspectrum’s imaginary distribution grows with an increase in the ratio between excitatory-to-inhibitory and excitatory-to-excitatory connectivity strengths. Consequently, the strength of global inhibition reveals itself as a strong signature for amplification and a potential control mechanism to switch dynamical regimes. Our results shed a light on how biological networks, i.e., networks constrained by Dale’s law, may be optimised for specific dynamical regimes.}, author = {Christodoulou, Georgia and Vogels, Tim P and Agnes, Everton J.}, issn = {1553-7358}, journal = {PLoS Computational Biology}, number = {8}, publisher = {Public Library of Science}, title = {{Regimes and mechanisms of transient amplification in abstract and biological neural networks}}, doi = {10.1371/journal.pcbi.1010365}, volume = {18}, year = {2022}, } @inproceedings{13239, abstract = {Brains are thought to engage in predictive learning - learning to predict upcoming stimuli - to construct an internal model of their environment. This is especially notable for spatial navigation, as first described by Tolman’s latent learning tasks. However, predictive learning has also been observed in sensory cortex, in settings unrelated to spatial navigation. Apart from normative frameworks such as active inference or efficient coding, what could be the utility of learning to predict the patterns of occurrence of correlated stimuli? Here we show that prediction, and thereby the construction of an internal model of sequential stimuli, can bootstrap the learning process of a working memory task in a recurrent neural network. We implemented predictive learning alongside working memory match-tasks, and networks emerged to solve the prediction task first by encoding information across time to predict upcoming stimuli, and then eavesdropped on this solution to solve the matching task. Eavesdropping was most beneficial when neural resources were limited. Hence, predictive learning acts as a general neural mechanism to learn to store sensory information that can later be essential for working memory tasks.}, author = {Van Der Plas, Thijs L. and Vogels, Tim P and Manohar, Sanjay G.}, booktitle = {Proceedings of Machine Learning Research}, issn = {2640-3498}, pages = {518--531}, publisher = {ML Research Press}, title = {{Predictive learning enables neural networks to learn complex working memory tasks}}, volume = {199}, year = {2022}, } @article{10753, abstract = {This is a comment on "Meta-learning synaptic plasticity and memory addressing for continual familiarity detection." Neuron. 2022 Feb 2;110(3):544-557.e8.}, author = {Confavreux, Basile J and Vogels, Tim P}, issn = {1097-4199}, journal = {Neuron}, number = {3}, pages = {361--362}, publisher = {Elsevier}, title = {{A familiar thought: Machines that replace us?}}, doi = {10.1016/j.neuron.2022.01.014}, volume = {110}, year = {2022}, } @article{8253, abstract = {Brains process information in spiking neural networks. Their intricate connections shape the diverse functions these networks perform. In comparison, the functional capabilities of models of spiking networks are still rudimentary. This shortcoming is mainly due to the lack of insight and practical algorithms to construct the necessary connectivity. Any such algorithm typically attempts to build networks by iteratively reducing the error compared to a desired output. But assigning credit to hidden units in multi-layered spiking networks has remained challenging due to the non-differentiable nonlinearity of spikes. To avoid this issue, one can employ surrogate gradients to discover the required connectivity in spiking network models. However, the choice of a surrogate is not unique, raising the question of how its implementation influences the effectiveness of the method. Here, we use numerical simulations to systematically study how essential design parameters of surrogate gradients impact learning performance on a range of classification problems. We show that surrogate gradient learning is robust to different shapes of underlying surrogate derivatives, but the choice of the derivative’s scale can substantially affect learning performance. When we combine surrogate gradients with a suitable activity regularization technique, robust information processing can be achieved in spiking networks even at the sparse activity limit. Our study provides a systematic account of the remarkable robustness of surrogate gradient learning and serves as a practical guide to model functional spiking neural networks.}, author = {Zenke, Friedemann and Vogels, Tim P}, issn = {1530-888X}, journal = {Neural Computation}, number = {4}, pages = {899--925}, publisher = {MIT Press}, title = {{The remarkable robustness of surrogate gradient learning for instilling complex function in spiking neural networks}}, doi = {10.1162/neco_a_01367}, volume = {33}, year = {2021}, } @article{8757, abstract = {Traditional scientific conferences and seminar events have been hugely disrupted by the COVID-19 pandemic, paving the way for virtual forms of scientific communication to take hold and be put to the test.}, author = {Bozelos, Panagiotis and Vogels, Tim P}, issn = {1471-0048}, journal = {Nature Reviews Neuroscience}, number = {1}, pages = {1--2}, publisher = {Springer Nature}, title = {{Talking science, online}}, doi = {10.1038/s41583-020-00408-6}, volume = {22}, year = {2021}, } @article{9228, abstract = {Legacy conferences are costly and time consuming, and exclude scientists lacking various resources or abilities. During the 2020 pandemic, we created an online conference platform, Neuromatch Conferences (NMC), aimed at developing technological and cultural changes to make conferences more democratic, scalable, and accessible. We discuss the lessons we learned.}, author = {Achakulvisut, Titipat and Ruangrong, Tulakan and Mineault, Patrick and Vogels, Tim P and Peters, Megan A.K. and Poirazi, Panayiota and Rozell, Christopher and Wyble, Brad and Goodman, Dan F.M. and Kording, Konrad Paul}, issn = {1879-307X}, journal = {Trends in Cognitive Sciences}, number = {4}, pages = {265--268}, publisher = {Elsevier}, title = {{Towards democratizing and automating online conferences: Lessons from the Neuromatch Conferences}}, doi = {10.1016/j.tics.2021.01.007}, volume = {25}, year = {2021}, } @inproceedings{11453, abstract = {Neuronal computations depend on synaptic connectivity and intrinsic electrophysiological properties. Synaptic connectivity determines which inputs from presynaptic neurons are integrated, while cellular properties determine how inputs are filtered over time. Unlike their biological counterparts, most computational approaches to learning in simulated neural networks are limited to changes in synaptic connectivity. However, if intrinsic parameters change, neural computations are altered drastically. Here, we include the parameters that determine the intrinsic properties, e.g., time constants and reset potential, into the learning paradigm. Using sparse feedback signals that indicate target spike times, and gradient-based parameter updates, we show that the intrinsic parameters can be learned along with the synaptic weights to produce specific input-output functions. Specifically, we use a teacher-student paradigm in which a randomly initialised leaky integrate-and-fire or resonate-and-fire neuron must recover the parameters of a teacher neuron. We show that complex temporal functions can be learned online and without backpropagation through time, relying on event-based updates only. Our results are a step towards online learning of neural computations from ungraded and unsigned sparse feedback signals with a biologically inspired learning mechanism.}, author = {Braun, Lukas and Vogels, Tim P}, booktitle = {35th Conference on Neural Information Processing Systems}, isbn = {9781713845393}, issn = {1049-5258}, location = {Virtual, Online}, pages = {16437--16450}, publisher = {Neural Information Processing Systems Foundation}, title = {{Online learning of neural computations from sparse temporal feedback}}, volume = {20}, year = {2021}, } @article{8126, abstract = {Cortical areas comprise multiple types of inhibitory interneurons with stereotypical connectivity motifs, but their combined effect on postsynaptic dynamics has been largely unexplored. Here, we analyse the response of a single postsynaptic model neuron receiving tuned excitatory connections alongside inhibition from two plastic populations. Depending on the inhibitory plasticity rule, synapses remain unspecific (flat), become anti-correlated to, or mirror excitatory synapses. Crucially, the neuron’s receptive field, i.e., its response to presynaptic stimuli, depends on the modulatory state of inhibition. When both inhibitory populations are active, inhibition balances excitation, resulting in uncorrelated postsynaptic responses regardless of the inhibitory tuning profiles. Modulating the activity of a given inhibitory population produces strong correlations to either preferred or non-preferred inputs, in line with recent experimental findings showing dramatic context-dependent changes of neurons’ receptive fields. We thus confirm that a neuron’s receptive field doesn’t follow directly from the weight profiles of its presynaptic afferents.}, author = {Agnes, Everton J. and Luppi, Andrea I. and Vogels, Tim P}, issn = {1529-2401}, journal = {The Journal of Neuroscience}, number = {50}, pages = {9634--9649}, publisher = {Society for Neuroscience}, title = {{Complementary inhibitory weight profiles emerge from plasticity and allow attentional switching of receptive fields}}, doi = {10.1523/JNEUROSCI.0276-20.2020}, volume = {40}, year = {2020}, } @article{8127, abstract = {Mechanistic modeling in neuroscience aims to explain observed phenomena in terms of underlying causes. However, determining which model parameters agree with complex and stochastic neural data presents a significant challenge. We address this challenge with a machine learning tool which uses deep neural density estimators—trained using model simulations—to carry out Bayesian inference and retrieve the full space of parameters compatible with raw data or selected data features. Our method is scalable in parameters and data features and can rapidly analyze new data after initial training. We demonstrate the power and flexibility of our approach on receptive fields, ion channels, and Hodgkin–Huxley models. We also characterize the space of circuit configurations giving rise to rhythmic activity in the crustacean stomatogastric ganglion, and use these results to derive hypotheses for underlying compensation mechanisms. Our approach will help close the gap between data-driven and theory-driven models of neural dynamics.}, author = {Gonçalves, Pedro J. and Lueckmann, Jan-Matthis and Deistler, Michael and Nonnenmacher, Marcel and Öcal, Kaan and Bassetto, Giacomo and Chintaluri, Chaitanya and Podlaski, William F. and Haddad, Sara A. and Vogels, Tim P and Greenberg, David S. and Macke, Jakob H.}, issn = {2050-084X}, journal = {eLife}, publisher = {eLife Sciences Publications}, title = {{Training deep neural density estimators to identify mechanistic models of neural dynamics}}, doi = {10.7554/eLife.56261}, volume = {9}, year = {2020}, }