| Abstract [eng] |
One of the most active areas in modern network theory is the study of relations between topological network characteristics and synchronization dynamics of these networks. It was shown that synchronization of neurons may depend on a network they are embedded To study the role played by different topologies in the synchronization, four of them were considered: all-to-all, random, small-world, scale-free. The neural networks of different topologies, consisting of one hundred identical neurons with dendritic dynamics were analysed in this work. Three different types of stimulations were applied to those networks, and their effects on robustness to noise, collective firing behaviour and calming rate were investigated. The results of the study showed that each neural network, after being stimulated, can display three different states: quiet, oscillating and mixed, when some of neurons fire and some of them are calm. Also, it was learned that the most robust topology was scale-free, while the fastest calming rate was found in the random network. It is worth mentioning that with some particular parameters random and small-world neural networks can evolve into quiet regime under continued stimulation. |
