1 min read Papers

Noise-induced patterns in neurobiological networks

Noise-induced patterns in neurobiological networks
Starting from an identical initial condition, a spiral wave is sustained for sufficiently large noise intensity (sigma = 0.45), but it disappears in the absence of noise (sigma = 0)

A new paper by ACDC member Daniele Avitabile and ACDC visitor James MacLaurin has been published on SIAM Journal of Applied Mathematics.

The paper studies large networks of firing-rate neurons, subject to noise and to random synaptic connections. It is proved that, when the number of neurons tend to infinity, the system is well described by a set of equations similar to a neural field, in which the noise intensity appears as a bifurcation parameter.

It is thus possible that spatiotemporal patterns form as the noise intensity is increased: patterns that are absent when the noise intensity is zero can be elicited when noise intensity is sufficiently large.