Time Varying Coherence Spectrum Analysis of Multichannel Local Field Potentials and Neuronal Ensemble

Abstract

The present study investigated time-varying coherence to analyze the spectrum synchronization pattern among multi-channel local field potentials (LFPs). Spectrogram analysis, Fast Fourier Transform (FFT) and power density estimate on LFPs were performed for characteristic bandwidth as theta-band power. 16-channel’s time-varying spectrum coherence was calculated between each channel and reference channel with 200-ms window sliding and 50% overlap. And the same principles were applied to analyze neuronal population discharges as well. The basic variables for coherence can be acquired after getting the average firing rates of individual neurons around a behavioral event (an auditory stimulation in a behavioral task). The results show that the linear coherence increased at the onset of the stimulation for both LFPs and neuronal population. These results indicate that the time-varying coherence spectrum synchronization analysis is a powerful tool for revealing the dynamic coherence spatial distribution characteristic of electroneurographic signals.