---------- Forwarded message ----------
From: Tiago Pereira <tiago....@imperial.ac.uk>
Date: 23 October 2012 08:19
Subject: DynamIC seminar: Natalia Janson (Wednesday, 24 October 2012, 14:00)
To: imperial-dyna...@imperial.ac.uk
Dear All,
please notice that this seminar is on Wednesday.
------------------------------------------------------------------------------------------------------
DynamIC: Dynamical systems at Imperial College London
http://www2.imperial.ac.uk/~jswlamb/DynamIC
------------------------------------------------------------------------------------------------------DynamIC Seminar
Speaker: Natalia Janson (Loughborough University)Title: Self-organisation and synchronization in stochastic neuron-like networks
Date: Wednesday, 24 October, 2012
Time: 14:00-15:00
Room: Huxley 139Abstract:
A single neuron is modelled as a highly non-linear dynamical system. Its crucial feature is excitability: when there is no input, or the input is below a threshold, the neuron is silent, and when the input is above the threshold, the neuronfires. The noise can play a highly counter-intuitive role in such systems: with the increase of the amount of noise, the amount of order in the system grows. Thus, the noise-induced spiking in such neurons becomes almost regular at the optimal intensity of the stimulus. Many neurons are coupled together in a network by various means, to model a biological neural network.
Their collective behavior ranges from independent spiking, through partly and fully synchronised spiking, to the lack of any spiking, depending on the parameters of the coupling. Synchronized spiking in the brain is associated with epilepsy, Parkinsonian disease and tremor, so the ability to eliminate synchronization by non-invasive weak stimulation could be essential in treating such conditions. It appears that in the model of the stochastic neural network it is possible to partly control the collective behavior by a specially constructed feedback.