Inference Methods for Interaction and Noise Intensities Using Only Spike-time Data on Coupled Oscillators

TL;DR

This paper introduces theoretical methods to infer coupling strength and noise intensity in coupled oscillators solely from spike-time data, applicable to experimental systems without external perturbations.

Contribution

The paper develops novel formulae linking spike-time statistics to coupling and noise parameters, enabling parameter inference without external perturbations.

Findings

Methods successfully applied to FitzHugh-Nagumo and phase models.

Parameter inference achieved using spike-time data alone.

Approach applicable to experimental systems without external perturbations.

Abstract

We propose theoretical methods to infer coupling strength and noise intensity simultaneously through an observation of spike timing in two well-synchronized noisy oscillators. A phase oscillator model is applied to derive formulae relating each of the parameters to some statistics from spike-time data. Using these formulae, each parameter is inferred from a specific set of statistics. We demonstrate the methods with the FitzHugh-Nagumo model as well as the phase model. Our methods do not require any external perturbation and thus are ready for application to various experimental systems.