Noise-induced synchronisation in heterogeneous nets of neural elements

Institute of Applied Physics, Darmstadt University of Technology - 64289 Darmstadt, Germany

Received: 16 October 2007
Accepted: 14 December 2007

Abstract

The interplay of additive noise and additive variability in an oscillatory net of neural elements is studied. With variability the net is a mixture of oscillatory and excitable elements. Even small values of the variability strength induce pattern formation and thus the elements do not oscillate synchronously anymore. In such heterogeneous nets additive noise can induce a regular synchronised net dynamics. This noise-induced synchronisation shows a resonance- like dependency on the noise strength. Furthermore, in the presence of noise, the coherence of the variability-induced patterns exhibits a resonance-like dependency on the variability strength and one finds maximally coherent pattern formation for intermediate values of this quantity.