Classification of Spike Wave Propagations in a Cultured Neuronal Network: Investigating a Brain Communication Mechanism

In brain information science, it is still unclear how multiple data can be stored and transmitted in ambiguously behaving neuronal networks. In the present study, we analyze the spatiotemporal propagation of spike trains in neuronal networks. Recently, spike propagation was observed functioning as a cluster of excitation waves (spike wave propagation) in cultured neuronal networks. We now assume that spike wave propagations are just events of communications in the brain. However, in reality, various spike wave propagations are generated in neuronal networks. Thus, there should be some mechanism to classify these spike wave propagations so that multiple communications in brain can be distinguished. To prove this assumption, we attempt to classify various spike wave propagations generated from different stimulated neurons using our original spatiotemporal pattern matching method for spike temporal patterns at each neuron in spike wave propagation in the cultured neuronal network. Based on the experimental results, it became clear that spike wave propagations have various temporal patterns from stimulated neurons. Therefore these stimulated neurons could be classified at several neurons away from the stimulated neurons. These are the classifiable neurons. Moreover, distribution of classifiable neurons in a network is also different when stimulated neurons generating spike wave propagations are different. These results suggest that distinct communications occur via multiple communication links and that classifiable neurons serve this function