A robust model of Stimulus-Specific Adaptation validated on neuromorphic hardware

AbstractStimulus-Specific Adaptation (SSA) to repetitive stimulation is a phenomenon that has been observed across many different species and in several brain sensory areas. It has been proposed as a computational mechanism, responsible for separating behaviorally relevant information from the continuous stream of sensory information. Although SSA can be induced and measured reliably in a wide variety of conditions, the network details and intracellular mechanisms giving rise to SSA still remain unclear. Recent computational studies proposed that SSA could be associated with a fast and synchronous neuronal firing phenomenon called Population Spikes (PS). Here, we test this hypothesis using a mean-field rate model and corroborate it using a neuromorphic hardware. As the neuromorphic circuits used in this study operate in real-time with biologically realistic time constants, they can reproduce the same dynamics observed in biological systems, together with the exploration of different connectivity schemes, with complete control of the system parameter settings. Besides, the hardware permits the iteration of multiple experiments over many trials, for extended amounts of time and without losing the networks and individual neural processes being studied. Following this “neuromorphic engineering” approach, we therefore study the PS hypothesis in a biophysically inspired recurrent networks of spiking neurons and evaluate the role of different linear and non-linear dynamic computational primitives such as spike-frequency adaptation or short-term depression (STD). We compare both the theoretical mean-field model of SSA and PS to previously obtained experimental results in the area of novelty detection and observe its behavior on its neuromorphic physical equivalent model. We show how the approach proposed can be extended to other computational neuroscience modelling efforts for understanding high-level phenomena in mechanistic models.

Download Full-text

Different theta frameworks coexist in the hippocampus and are coordinated during memory-guided exploration and novelty detection

10.1101/418434 ◽

2018 ◽

Cited By ~ 2

Author(s):

Víctor J. Lopez-Madrona ◽

Elena Pérez-Montoyo ◽

Efrén Álvarez-Salvado ◽

David Moratal ◽

Oscar Herreras ◽

...

Keyword(s):

Theta Rhythm ◽

Novelty Detection ◽

Molecular Layer ◽

Sensory Information ◽

Neuronal Firing ◽

Theta Oscillations ◽

Lower Phase ◽

Guided Exploration ◽

Memory Representations ◽

Freely Behaving

SummaryHippocampal firing is organized in theta sequences controlled by internal memory-related processing and by external sensory cues. How these computations are segregated or integrated, depending on the cognitive needs, is not fully understood. Although theta activity in the hippocampus is most commonly studied as a unique coherent oscillation, it is the result of a complex interaction between different rhythm generators. Here we investigated the coordination between theta generators as a possible mechanism to couple or decouple internally and externally driven computations. We separated and quantified three different theta current generators from the hippocampus of freely behaving rats, one originating in CA3 with current sinks in CA1 str. radiatum and two with current sinks in CA1 str. lacunosum-moleculare and dentate molecular layer, mainly driven by entorhinal cortex (EC) layers 3 and 2, respectively. These theta generators followed non fully coherent dynamics and presented epochs of higher and lower phase coupling, suggesting a flexible interaction between them. Selective optogenetic inhibition in CA3 depressed the str. radiatum generator without affecting the EC-driven theta oscillations, indicating that theta rhythm generators can be modulated independently. In addition, band-specific gamma interactions with theta oscillations selectively occurred with the corresponding pathway-specific theta current generator, supporting the existence of different theta-gamma coding frameworks to organize neuronal firing in the hippocampus. Importantly, we found that epochs of highly synchronized theta rhythmicity across generators preferentially occurred during memory-guided exploration and mismatch novelty detection in familiar environments, two conditions in which internally generated memory representations need to be coordinated with the incoming sensory information about external cues. We propose a mechanism for segregating and integrating hippocampal computations based on the coexistence of different theta-gamma frameworks that flexibly couple or decouple accommodating the cognitive needs.

Download Full-text

Nonlinear collision between anisotropic propagating waves in mouse somatosensory cortex

10.1101/2021.01.04.425215 ◽

2021 ◽

Author(s):

M. Di Volo ◽

I. Férézou

Keyword(s):

Somatosensory Cortex ◽

Mean Field ◽

Sensory Information ◽

Temporal Patterns ◽

Inversion Method ◽

Model Parameters ◽

Lower Sensitivity ◽

Mean Field Model ◽

Linear Interaction ◽

Spatio Temporal

AbstractHow does cellular organization shape the spatio-temporal patterns of activity in the cortex while processing sensory information? After measuring the propagation of activity in the mouse primary somatosensory cortex (S1) in response to single whisker deflections with Voltage Sensitive Dye (VSD) imaging, we developed a two dimensional mean field model of S1. We observed that, for strong enough excitatory cortical interactions, whisker deflections generate a propagating wave in S1. We developed an inversion method that reconstructs model parameters from VSD data, revealing that a spatially heterogeneous organization of synaptic strengths between pyramidal neurons in S1 is likely to be responsible for the anisotropic spatio-temporal patterns of activity measured experimentally. Finally, we report that two consecutive stimuli activating different spatial locations in S1 generate two waves which collide sub-linearly. In the model, such sub-linear interaction is explained by a lower sensitivity to external perturbations of neural networks during activated states.

Download Full-text

Path Integral Method in the Mean-field Model for the Magnetic Vector Potential

Geomagnetism and Aeronomy ◽

10.1134/s0016793220070300 ◽

2020 ◽

Vol 60 (7) ◽

pp. 989-992

Author(s):

E. V. Yushkov ◽

C. R. Kamaletdinov ◽

D. D. Sokoloff

Keyword(s):

Path Integral ◽

Vector Potential ◽

Integral Method ◽

Field Model ◽

Mean Field ◽

Magnetic Vector Potential ◽

Magnetic Vector ◽

Mean Field Model ◽

The Mean ◽

Path Integral Method

Download Full-text

KDP-type crystal spontaneous polarization behavior better explained by bound charge semiconductor model than by mean field model

Ferroelectrics ◽

10.1080/00150193.2020.1791663 ◽

2020 ◽

Vol 569 (1) ◽

pp. 70-81

Author(s):

V. Hugo Schmidt

Keyword(s):

Spontaneous Polarization ◽

Field Model ◽

Mean Field ◽

Mean Field Model ◽

Polarization Behavior ◽

Type Crystal

Download Full-text

On the Throughput Optimization in Large-Scale Batch-Processing Systems

ACM SIGMETRICS Performance Evaluation Review ◽

10.1145/3453953.3453982 ◽

2021 ◽

Vol 48 (3) ◽

pp. 128-129

Author(s):

Sounak Kar ◽

Robin Rehrmann ◽

Arpan Mukhopadhyay ◽

Bastian Alt ◽

Florin Ciucu ◽

...

Keyword(s):

Large Scale ◽

Mean Field ◽

Queueing Network ◽

Processing System ◽

Batch Size ◽

System Throughput ◽

Throughput Optimization ◽

Mean Field Model ◽

Optimal Batch Size ◽

Globally Attractive

We analyze a data-processing system with n clients producing jobs which are processed in batches by m parallel servers; the system throughput critically depends on the batch size and a corresponding sub-additive speedup function that arises due to overhead amortization. In practice, throughput optimization relies on numerical searches for the optimal batch size which is computationally cumbersome. In this paper, we model this system in terms of a closed queueing network assuming certain forms of service speedup; a standard Markovian analysis yields the optimal throughput in w n4 time. Our main contribution is a mean-field model that has a unique, globally attractive stationary point, derivable in closed form. This point characterizes the asymptotic throughput as a function of the batch size that can be calculated in O(1) time. Numerical settings from a large commercial system reveal that this asymptotic optimum is accurate in practical finite regimes.

Download Full-text

Mean-Field Models for EEG/MEG: From Oscillations to Waves

Brain Topography ◽

10.1007/s10548-021-00842-4 ◽

2021 ◽

Author(s):

Áine Byrne ◽

James Ross ◽

Rachel Nicks ◽

Stephen Coombes

Keyword(s):

Gap Junction ◽

Large Scale ◽

Mean Field ◽

Coarse Grained ◽

Neural Mass Model ◽

Neuron Network ◽

Mass Model ◽

Mean Field Model ◽

Neural Mass ◽

The Rich

AbstractNeural mass models have been used since the 1970s to model the coarse-grained activity of large populations of neurons. They have proven especially fruitful for understanding brain rhythms. However, although motivated by neurobiological considerations they are phenomenological in nature, and cannot hope to recreate some of the rich repertoire of responses seen in real neuronal tissue. Here we consider a simple spiking neuron network model that has recently been shown to admit an exact mean-field description for both synaptic and gap-junction interactions. The mean-field model takes a similar form to a standard neural mass model, with an additional dynamical equation to describe the evolution of within-population synchrony. As well as reviewing the origins of this next generation mass model we discuss its extension to describe an idealised spatially extended planar cortex. To emphasise the usefulness of this model for EEG/MEG modelling we show how it can be used to uncover the role of local gap-junction coupling in shaping large scale synaptic waves.

Download Full-text

Quark mean-field model for single and double and hypernuclei

Progress of Theoretical and Experimental Physics ◽

10.1093/ptep/ptt119 ◽

2014 ◽

Vol 2014 (1) ◽

pp. 13D02-0 ◽

Cited By ~ 11

Author(s):

J. N. Hu ◽

A. Li ◽

H. Shen ◽

H. Toki

Keyword(s):

Field Model ◽

Mean Field ◽

Mean Field Model

Download Full-text

EVOLUTION OF SHELL STRUCTURE IN NUCLEI

International Journal of Modern Physics E ◽

10.1142/s0218301311019581 ◽

2011 ◽

Vol 20 (08) ◽

pp. 1663-1675 ◽

Cited By ~ 4

Author(s):

A. BHAGWAT ◽

Y. K. GAMBHIR

Keyword(s):

Shell Structure ◽

Crucial Role ◽

Field Model ◽

Mean Field ◽

Mass Region ◽

The Other ◽

Mean Field Model ◽

Relativistic Mean Field ◽

Shell Closures ◽

Low Mass

Systematic investigations of the pairing and two-neutron separation energies which play a crucial role in the evolution of shell structure in nuclei, are carried out within the framework of relativistic mean-field model. The shell closures are found to be robust, as expected, up to the lead region. New shell closures appear in low mass region. In the superheavy region, on the other hand, it is found that the shell closures are not as robust, and they depend on the particular combinations of neutron and proton numbers. Effect of deformation on the shell structure is found to be marginal.

Download Full-text