Dynamic Information Routing in Neuronal Circuits
Jorge Medina Hernández (Advisors: Mirasso, Claudio R.; Eguíluz, Víctor M. )
Master Thesis (2020)
The emergence of flexible channels of information in brain networks is a fundamental issue in neuroscience. Several topological properties have been suggested to define highly influential nodes, the most important being the "hub nodes", that is, nodes (brain regions) with a disproportionate number of connections to other parts of the brain. Nevertheless, the influence of a node does not only depend on its degree and position in the network
but on its internal dynamics. Recently, it has been observed that nodes with an oscillating frequency higher than those of their neighbors can act as functional hubs, redirecting the information. Thus, in this work, the dynamics of a simple neural network have been analyzed to better understand which is the mechanism that transforms a normal node into a highly influential node. Specifically, we have focused on the propagation of signals in chain
networks whose evolution is determined by the Hodgkin-Huxley model. First, we have studied the propagation of sinusoidal signals applied to a high-frequency neuron, verifying the existence of a certain interval of oscillating frequencies that allows the transmission of information across the network, in agreement with the literature. This interval was found to be linked to the capacity of the high-frequency neuron to set optimal relative spiking times
of the other neurons in the network. Then, we considered the competition of two signals when they were applied to a high-frequency neuron and an ordinary neuron, as well as two high-frequency neurons. The results indicate that both the oscillating frequency of the neurons and the frequencies of the input signals determine the information transmission.