Regensburg 2004 – wissenschaftliches Programm

Bereiche | Tage | Auswahl | Suche | Downloads | Hilfe

DY: Dynamik und Statistische Physik

DY 24: Dynamic Instabilities in Biophysics

DY 24.7: Vortrag

Dienstag, 9. März 2004, 16:15–16:30, H2

Spectra and Waiting-Time Distribution of Stochastic FitzHugh-Nagumo Neurons — •Tatiana Verechtchaguina, Lutz Schimansky-Geier, and Igor M. Sokolov — Institut für Physik, Humboldt Universität zu Berlin, Newton Str. 15, 12489 Berlin

A reponse of a neural cell to an external stimulus can follow one of the two patterns: Cells of the first type (non-resonant neurons) monotonously relax to the resting state. Neurons of the second type (resonant neurons) show subthreshold oscillations leading to selective response to periodic pulse sequences or noisy signals. We discuss how do these resonant properties of neurons affect power spectra of their response to a random (noise-like) signal.

The dynamics of a neuron is modelled by a FitzHugh-Nagumo (FN) system; the two types of neurons differ in the values of parameter є, characterizing separation of time-scales of the slow and the fast variables. We show that corresponding spectra obtained numerically can be well described as spectra of sequences of pulses of the same form following a given waiting-time distribution, and can be calculated using the Stratonovich formula for spectra of point processes. Thus, the probability density function of interspike intervals contains practically the whole information about the response of a neuron, and can in some cases be more useful than spectral characteristics.