Mixed Mode Bursting Oscillations Induced by Birhythmicity and Noise

TL;DR

This paper investigates how birhythmicity and noise induce mixed-mode bursting oscillations in a simple neural model, revealing how noise intensity and initial conditions influence bursting dynamics.

Contribution

It introduces a new paradigm for MMBO generation driven by birhythmicity and noise in a two-dimensional integrate-and-fire-or-burst model.

Findings

Noise induces random transitions between bursting states.

Transition rates increase with noise intensity.

Bursting dynamics depend on initial conditions at low noise levels.

Abstract

Bursting oscillations are commonly seen as a mechanism for information coding in neuroscience and have also been observed in many physical, biochemical, and chemical systems. This study focuses on the computational investigation of mixed-mode bursting oscillations (MMBOs) generated by a simple two-dimensional integrate-and-fire-or-burst (IFB) model. We demonstrate a new paradigm for the generation of MMBOs, where birhythmicity and noise are the key components. In the absence of noise, the proposed model exhibits birhythmicity of two independent bursting patterns, bursts of two spikes and bursts of three spikes, depending on the initial condition of the model. Noise induces the random transitions between two bursting states which leads to MMBOs, and the transition rate increases with the noise intensity. Our results provide a systematic view of the roles of noise and initial condition: the bursting dynamics produced by the proposed model heavily rely on the initial conditions when noise is weak; while for intermediate and strong noise, the burst dynamics are independent of the initial condition.