Multitaper Analysis of Evolutionary Spectral Density Matrix From Multivariate Spiking Observations

Anuththara Rupasinghe, Behtash Babadi

Abstract: Extracting the spectral representations of the neural processes that underlie spiking activity is key to understanding how the brain rhythms mediate cognitive functions. While spectral estimation of continuous time-series is well studied, inferring the spectral representation of latent processes from spiking observations is a challenging problem. In this paper, we address this issue by proposing a spectral estimation methodology that can be directly applied to multivariate spiking observations in order to extract the evolutionary spectral representation of the latent non-stationary processes. We compare the performance of our proposed technique with several existing methods using simulated data, which reveals significant gains in terms of the bias-variance trade-off.