|Ph.D Student||Shlomo Israelit|
|Subject||Analysis of Spontaneous Activity in Large Random Neural|
|Department||Department of Medicine||Supervisor||Full Professor Marom Shimon|
This research is concerned with how the molecular mechanisms manifest itself in the activity of neurons within their networks.
Neurons within networks are driven to fire by connections with other members of the networks, and their firing patterns reflect interactions between the dynamics of the network and their intrinsic cellular excitability properties.
The present study describes the statistical properties of the firing patterns of a single neuron in an ex-vivo developing cortical network. The cortical neurons were taken from newborn rats. The neurons were plated on multi electrode array culture dishes. Each electrode had a diameter of 30 micrometer, with 200 micrometer separating each electrode from the nearest electrodes. The cultures were kept for two weeks before recording. During this period the neurons arranged to form a random neural network with strong connectivity and a rich repertoire of electrical activity.
Stable recording for hours from the electrode array was performed.
Statistical analysis was preformed, in order to reveal long-term spatial correlations in the firing pattern of a single neuron. The analysis used fractal analysis techniques namely Fano factor, Allan factor and multi-scale inter event fluctuation function.
Using these techniques we were able to show that the spontaneous neural activity showed complex (fractal) statistical properties.
These results demonstrate that the pattern of the spontaneous activity, of a large random neural network, is not random, but characterized by complex statistical properties, with self-similarity.
These properties suggest the presence of spatial correlations over many time scales.