Enhancement of gamma oscillations in E/I neural networks by increase of difference between external inputs (Q2055162)

The goal of the paper is to investigate the change of gamma oscillations in a neural network system versus variation of external inputs. In fact, there has been experimentally observed that if one sends two types of external inputs to a neural network, the greater the difference between the two inputs is, the stronger the gamma oscillations of the system are. The study starts with the analysis of gamma oscillations in a small excitatory/inhibitory neural network. The excitatory and inhibitory neurons are modeled as Integrate-and-Fire neurons with conductance-based synapse. The network consists of 400 excitatory and 100 inhibitory neurons with all-to-all connections. Different external stimuli S1 and S2 are input to the network and gamma oscillations are observed. Network simulation as well as data analysis are performed with MATLAB 2012a. The simulation results show the sensitive dependence of gamma oscillations from the changes of inputs S1 and S2. Gamma oscillations get stronger when the difference in the inputs increases. The third section of the article is dedicated to the study of gamma oscillations in a large-scale excitatory/inhibitory neural network of integrate-and-fire neurons with synapses described by the conductance-based synapse model. For simulation a CUDA-based algorithm is developed. One shows that also in this case, the obtained results are consistent with the biological experimental observations.