Noise-induced interspike interval correlations and spike train regularization in spike-triggered adapting neurons

Abstract

Spike generation in neurons produces a temporal point process, whose statistics is governed by intrinsic phenomena and the external incoming inputs to be coded. In particular, spike-evoked adaptation currents support a slow temporal process that conditions spiking probability at the present time according to past activity. In this work, we study the statistics of interspike interval correlations arising in such non-renewal spike trains, for a neuron model that reproduces different spike modes in a small adaptation scenario. We found that correlations are stronger as the neuron fires at a particular firing rate, which is defined by the adaptation process. When set in a subthreshold regime, the neuron may sustain this particular firing rate, and thus induce correlations, by noise. Given that, in this regime, interspike intervals are negatively correlated at any lag, this effect surprisingly implies a reduction in the variability of the spike count statistics at a finite noise intensity.

7 pages, 6 figures, published in Epl

References in corpus (6)

• Adaptation Reduces Variability of the Neuronal Population Code
• Patterns of interval correlations in neural oscillators with adaptation
• Onset of negative interspike interval correlations in adapting neurons
• Survival probability and first-passage-time statistics of a Wiener process driven by an exponential time-dependent drift
• Series solution to the first-passage-time problem of a Brownian motion with an exponential time-dependent drift
• First-passage-time statistics of a Brownian particle driven by an arbitrary unidimensional potential with a superimposed exponential time-dependent drift

Cited by in corpus (2)

• Evolution of moments and correlations in non-renewal escape-time processes
• Interspike interval correlations in networks of inhibitory integrate-and-fire neurons