Targeting diverse operational regimes in recurrent spiking networks DOI Creative Commons
Pierre Ekelmans, Nataliya Kraynyukova, Tatjana Tchumatchenko

и другие.

bioRxiv (Cold Spring Harbor Laboratory), Год журнала: 2022, Номер unknown

Опубликована: Апрель 22, 2022

Neural computations emerge from recurrent neural circuits that comprise hundreds to a few thousand neurons. Continuous progress in connectomics, electrophysiology, and calcium imaging require tractable spiking network models can consistently incorporate new information about the structure reproduce recorded activity features. However, it is challenging predict which connectivity configurations properties generate fundamental operational states specific experimentally reported nonlinear cortical computations. Theoretical descriptions for computational state of are diverse, including balanced where excitatory inhibitory inputs balance almost perfectly or inhibition stabilized (ISN) part circuit unstable. It remains an open question whether these co-exist with they be recovered biologically realistic implementations networks. Here, we show how identify patterns underlying diverse such as XOR, bistability, stabilization, supersaturation, persistent activity. We established mapping between supralinear (SSN) allowed us pinpoint location parameter space regimes occur. Notably, found biologically-sized networks have irregular asynchronous does not strong excitation-inhibition large feedforward input showed dynamic firing rate trajectories precisely targeted without error-driven training algorithms.

Язык: Английский

Regulation of circuit organization and function through inhibitory synaptic plasticity DOI
Yue Kris Wu, Christoph Miehl, Julijana Gjorgjieva

и другие.

Trends in Neurosciences, Год журнала: 2022, Номер 45(12), С. 884 - 898

Опубликована: Окт. 28, 2022

Язык: Английский

Процитировано

62
Top–down modulation in canonical cortical circuits with short-term plasticity DOI Creative Commons
Felix Waitzmann, Yue Kris Wu, Julijana Gjorgjieva

и другие.

Proceedings of the National Academy of Sciences, Год журнала: 2024, Номер 121(16)

Опубликована: Апрель 9, 2024

Cortical dynamics and computations are strongly influenced by diverse GABAergic interneurons, including those expressing parvalbumin (PV), somatostatin (SST), vasoactive intestinal peptide (VIP). Together with excitatory (E) neurons, they form a canonical microcircuit exhibit counterintuitive nonlinear phenomena. One instance of such phenomena is response reversal, whereby SST neurons show opposite responses to top–down modulation via VIP depending on the presence bottom–up sensory input, indicating that network may function in different regimes under stimulation conditions. Combining analytical computational approaches, we demonstrate model networks multiple interneuron subtypes experimentally identified short-term plasticity mechanisms can implement reversal. Surprisingly, despite not directly affecting activity, PV-to-E depression has decisive impact We how reversal relates inhibition stabilization paradoxical effect several demonstrating coincides change indispensability for stabilization. In summary, our work suggests role generating makes testable predictions.

Язык: Английский

Процитировано

9
Functional subtypes of synaptic dynamics in mouse and human DOI Creative Commons
John Beninger, Julian Rossbroich, Katalin Tóth

и другие.

Cell Reports, Год журнала: 2024, Номер 43(2), С. 113785 - 113785

Опубликована: Фев. 1, 2024

Synapses preferentially respond to particular temporal patterns of activity with a large degree heterogeneity that is informally or tacitly separated into classes. Yet, the precise number and properties such classes are unclear. Do they exist on continuum and, if so, when it appropriate divide functional regions? In dataset glutamatergic cortical connections, we perform model-based characterization infer characteristics functionally distinct subtypes synaptic dynamics. rodent data, find five clusters partially converge transgenic-associated subtypes. Strikingly, application same clustering method in human data infers highly similar clusters, supportive stable clustering. This nuanced dictionary shapes dynamics provides lens basic motifs information transmission brain.

Язык: Английский

Процитировано

4
Self-sustained and oscillatory activity in two types of attractor networks DOI
Tao Wang,

J. Gui,

Ming J. Zuo

и другие.

Nonlinear Dynamics, Год журнала: 2025, Номер unknown

Опубликована: Март 17, 2025

Язык: Английский

Процитировано

0
Top-down modulation in canonical cortical circuits with short-term plasticity DOI Creative Commons
Felix Waitzmann, Yue Kris Wu, Julijana Gjorgjieva

и другие.

bioRxiv (Cold Spring Harbor Laboratory), Год журнала: 2023, Номер unknown

Опубликована: Июнь 14, 2023

Abstract Cortical dynamics and computations are strongly influenced by diverse GABAergic interneurons, including those expressing parvalbumin (PV), somatostatin (SST), vasoactive intestinal peptide (VIP). Together with excitatory (E) neurons, they form a canonical microcircuit exhibit counterintuitive nonlinear phenomena. One instance of such phenomena is response reversal, whereby SST neurons show opposite responses to top-down modulation via VIP depending on the presence bottom-up sensory input, indicating that network may function in different regimes under stimulation conditions. Combining analytical computational approaches, we demonstrate model networks multiple interneuron subtypes experimentally identified short-term plasticity mechanisms can implement reversal. Surprisingly, despite not directly affecting activity, PV-to-E depression has decisive impact We how reversal relates inhibition stabilization paradoxical effect several demonstrating coincides change indispensability for stabilization. In summary, our work suggests role generating makes testable predictions.

Язык: Английский

Процитировано

6
Inhibition stabilization and paradoxical effects in recurrent neural networks with short-term plasticity DOI Creative Commons
Yue Kris Wu, Julijana Gjorgjieva

Physical Review Research, Год журнала: 2023, Номер 5(3)

Опубликована: Июль 12, 2023

This article is part of the Physical Review Research collection titled Physics Neuroscience. Inhibition stabilization considered a ubiquitous property cortical networks, whereby inhibition controls network activity in presence strong recurrent excitation. In networks with fixed connectivity, an identifying characteristic that increasing (decreasing) excitatory input to inhibitory population leads decrease (increase) firing, known as paradoxical effect. However, responses stimulation are highly nonlinear, and drastic changes synaptic strengths induced by short-term plasticity (STP) can occur on timescale perception. How neuronal nonlinearities STP affect effect unclear. Using analytical calculations, we demonstrate implies stabilization, but does not imply Interestingly, transition nonmonotonically between inhibition-stabilization noninhibition-stabilization, paradoxically- nonparadoxically-responding regimes activity. Furthermore, generalize our results more complex scenarios including multiple interneuron subtypes any monotonically nonlinearities. summary, work reveals relationship nonlinearity STP, yielding several testable predictions.Received 22 December 2022Accepted 5 June 2023DOI:https://doi.org/10.1103/PhysRevResearch.5.033023Published American Society under terms Creative Commons Attribution 4.0 International license. Further distribution this must maintain attribution author(s) published article's title, journal citation, DOI. Open access publication funded Max Planck Society.Published SocietyPhysics Subject Headings (PhySH)Research AreasNeurosciencePhysical SystemsBiological networksCortical networksBiological

Язык: Английский

Процитировано

6
Unified control of temporal and spatial scales of sensorimotor behavior through neuromodulation of short-term synaptic plasticity DOI Creative Commons
Shanglin Zhou, Dean V. Buonomano

Science Advances, Год журнала: 2024, Номер 10(18)

Опубликована: Май 3, 2024

Neuromodulators have been shown to alter the temporal profile of short-term synaptic plasticity (STP); however, computational function this neuromodulation remains unexplored. Here, we propose that STP provides a general mechanism scale neural dynamics and motor outputs in time space. We trained recurrent networks incorporated produce complex trajectories—handwritten digits—with different (speed) spatial (size) scales. Neuromodulation produced scaling learned enhanced or generalization compared standard training weights absence STP. The model also accounted for results two experimental studies involving flexible sensorimotor timing. unified biologically plausible control scales behaviors.

Язык: Английский

Процитировано

2
Reduced variability of bursting activity during working memory DOI Creative Commons
Mikael Lundqvist, Jonas Rose, Scott L. Brincat

и другие.

Scientific Reports, Год журнала: 2022, Номер 12(1)

Опубликована: Сен. 5, 2022

Working memories have long been thought to be maintained by persistent spiking. However, mounting evidence from multiple-electrode recording (and single-trial analyses) shows that the underlying spiking is better characterized intermittent bursts of activity. A counterargument suggested this activity at odds with observations spike-time variability reduces during task performance. rests on assumptions, such as randomness in timing bursts, which may not correct. Thus, we analyzed and LFPs monkeys' prefrontal cortex (PFC) determine if task-related reductions can co-exist We found it does because both associated gamma were task-modulated, random. In fact, reduction spike could largely explained a related burst variability. Our results provide further support for models working memory well novel mechanistic insights into how reduced cognitive tasks.

Язык: Английский

Процитировано

7
Targeting operational regimes of interest in recurrent neural networks DOI Creative Commons
Pierre Ekelmans, Nataliya Kraynyukova, Tatjana Tchumatchenko

и другие.

PLoS Computational Biology, Год журнала: 2023, Номер 19(5), С. e1011097 - e1011097

Опубликована: Май 15, 2023

Neural computations emerge from local recurrent neural circuits or computational units such as cortical columns that comprise hundreds to a few thousand neurons. Continuous progress in connectomics, electrophysiology, and calcium imaging require tractable spiking network models can consistently incorporate new information about the structure reproduce recorded activity features. However, for networks, it is challenging predict which connectivity configurations properties generate fundamental operational states specific experimentally reported nonlinear computations. Theoretical descriptions state of are diverse, including balanced where excitatory inhibitory inputs balance almost perfectly inhibition stabilized (ISN) part circuit unstable. It remains an open question whether these co-exist with they be recovered biologically realistic implementations networks. Here, we show how identify patterns underlying diverse XOR, bistability, stabilization, supersaturation, persistent activity. We establish mapping between supralinear (SSN) allows us pinpoint location parameter space regimes occur. Notably, find biologically-sized networks have irregular asynchronous does not strong excitation-inhibition large feedforward input dynamic firing rate trajectories precisely targeted without error-driven training algorithms.

Язык: Английский

Процитировано

4
An Agent-Based Model to Reproduce the Boolean Logic Behaviour of Neuronal Self-Organised Communities through Pulse Delay Modulation and Generation of Logic Gates DOI Creative Commons
Luis Irastorza-Valera, José María Benítez, Francisco J. Montáns

и другие.

Biomimetics, Год журнала: 2024, Номер 9(2), С. 101 - 101

Опубликована: Фев. 9, 2024

The human brain is arguably the most complex "machine" to ever exist. Its detailed functioning yet be fully understood, let alone modelled. Neurological processes have logical signal-processing and biophysical aspects, both affect brain's structure, adaptation. Mathematical approaches based on information graph theory been extensively used in an attempt approximate its biological functioning, along with Artificial Intelligence frameworks inspired by functioning. In this article, approach model some aspects of learning signal processing presented, mimicking metastability backpropagation found real while also accounting for neuroplasticity. Several simulations are carried out demonstrate how dynamic neuroplasticity, neural inhibition neuron migration can reshape connectivity synchronise obtain certain target latencies. This work showcases importance remodelling plasticity. Combining mathematical (agents, theory, topology backpropagation) biomedical ingredients (metastability, neuroplasticity migration), these preliminary results prove phenomena reproduced-under pertinent simplifications-via affordable computations, which construed as a starting point more ambitiously accurate simulations.

Язык: Английский

Процитировано

1