Spike Count Correlation Increases with Length of Time Interval in the Presence of Trial-to-Trial Variation

Authors:
Robert E. Kass;Valérie Ventura
Affiliations:
kass@stat.cmu.edu;Department of Statistics and Center for the Neural Basis of Cognition, Carnegie Mellon University, Pittsburgh, PA 15213, U.S.A. vventura@stat.cmu.edu
Venue:
Neural Computation
Year:
2006

Citing 2
Cited 2

Disambiguating different covariation types

Neural Computation
Correlations without synchrony

Neural Computation

Generating spike trains with specified correlation coefficients

Neural Computation
Correlated firing and oscillations in spiking networks with global delayed inhibition

Neurocomputing

Quantified Score

Hi-index	0.00

Visualization

Abstract

It has been observed that spike count correlation between two simultaneously recorded neurons often increases with the length of time interval examined. Under simple assumptions that are roughly consistent with much experimental data, we show that this phenomenon may be explained as being due to excess trial-to-trial variation. The resulting formula for the correlation is able to predict the observed correlation of two neurons recorded from primary visual cortex as a function of interval length.