Paper detail

Entrainment in up and down states of neural populations: non-smooth and stochastic models

We study the impact of noise on a neural population rate model of up and down states. Up and down states are typically observed in neuronal networks as a slow oscillation, where the population switches between high and low firing rates (Sanchez-Vivez and McCormick, 2000). A neural population model with spike rate adaptation is used to model such slow oscillations, and the timescale of adaptation determines the oscillation period. Furthermore, the period depends non-monotonically on the background tonic input driving the population, having long periods for very weak and very strong stimuli. Using both linearization and fast-slow timescale separation methods, we can compute the phase sensitivity function of the slow oscillation. We find that the phase response is most strongly impacted by perturbations to the adaptation variable. Phase sensitivity functions can then be utilized to quantify the impact of noise on oscillating populations. Noise alters the period of oscillations by speeding up the rate of transition between the up and down states. When common noise is presented to two distinct populations, their transitions will eventually become entrained to one another through stochastic synchrony.