Relation between single neuron and population spiking statistics and effects on network activity

Hideyuki Cĝteau, Alexander Reyes

Research output: Contribution to journalArticle

Abstract

To simplify theoretical analyses of neural networks, individual neurons are often modeled as Poisson processes. An implicit assumption is that even if the spiking activity of each neuron is non-Poissonian, the composite activity obtained by summing many spike trains limits to a Poisson process. Here, we show analytically and through simulations that this assumption is invalid. Moreover, we show with Fokker-Planck equations that the behavior of feedforward networks is reproduced accurately only if the tendency of neurons to fire periodically is incorporated by using colored noise whose autocorrelation has a negative component.

Original languageEnglish (US)
Article number058101
JournalPhysical Review Letters
Volume96
Issue number5
DOIs
StatePublished - 2006

Fingerprint

spiking
Population Characteristics
neurons
poisson process
statistics
Neurons
Fokker-Planck equation
spikes
autocorrelation
Noise
tendencies
composite materials
simulation

ASJC Scopus subject areas

  • Physics and Astronomy(all)
  • Medicine(all)

Cite this

Relation between single neuron and population spiking statistics and effects on network activity. / Cĝteau, Hideyuki; Reyes, Alexander.

In: Physical Review Letters, Vol. 96, No. 5, 058101, 2006.

Research output: Contribution to journalArticle

@article{0ddff3e6dfcb4d3a9a38466f7c9601a0,
title = "Relation between single neuron and population spiking statistics and effects on network activity",
abstract = "To simplify theoretical analyses of neural networks, individual neurons are often modeled as Poisson processes. An implicit assumption is that even if the spiking activity of each neuron is non-Poissonian, the composite activity obtained by summing many spike trains limits to a Poisson process. Here, we show analytically and through simulations that this assumption is invalid. Moreover, we show with Fokker-Planck equations that the behavior of feedforward networks is reproduced accurately only if the tendency of neurons to fire periodically is incorporated by using colored noise whose autocorrelation has a negative component.",
author = "Hideyuki Cĝteau and Alexander Reyes",
year = "2006",
doi = "10.1103/PhysRevLett.96.058101",
language = "English (US)",
volume = "96",
journal = "Physical Review Letters",
issn = "0031-9007",
publisher = "American Physical Society",
number = "5",

}

TY - JOUR

T1 - Relation between single neuron and population spiking statistics and effects on network activity

AU - Cĝteau, Hideyuki

AU - Reyes, Alexander

PY - 2006

Y1 - 2006

N2 - To simplify theoretical analyses of neural networks, individual neurons are often modeled as Poisson processes. An implicit assumption is that even if the spiking activity of each neuron is non-Poissonian, the composite activity obtained by summing many spike trains limits to a Poisson process. Here, we show analytically and through simulations that this assumption is invalid. Moreover, we show with Fokker-Planck equations that the behavior of feedforward networks is reproduced accurately only if the tendency of neurons to fire periodically is incorporated by using colored noise whose autocorrelation has a negative component.

AB - To simplify theoretical analyses of neural networks, individual neurons are often modeled as Poisson processes. An implicit assumption is that even if the spiking activity of each neuron is non-Poissonian, the composite activity obtained by summing many spike trains limits to a Poisson process. Here, we show analytically and through simulations that this assumption is invalid. Moreover, we show with Fokker-Planck equations that the behavior of feedforward networks is reproduced accurately only if the tendency of neurons to fire periodically is incorporated by using colored noise whose autocorrelation has a negative component.

UR - http://www.scopus.com/inward/record.url?scp=33144489051&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33144489051&partnerID=8YFLogxK

U2 - 10.1103/PhysRevLett.96.058101

DO - 10.1103/PhysRevLett.96.058101

M3 - Article

VL - 96

JO - Physical Review Letters

JF - Physical Review Letters

SN - 0031-9007

IS - 5

M1 - 058101

ER -