Distributed Delay and Desynchronization in a Neural Mass Model
Isam Al-Darabsah,Sue Ann Campbell,Bootan Rahman
DOI: https://doi.org/10.1137/23m1618028
2024-12-11
SIAM Journal on Applied Dynamical Systems
Abstract:SIAM Journal on Applied Dynamical Systems, Volume 23, Issue 4, Page 3013-3051, December 2024. Neural mass models offer a robust mathematical framework for studying the collective behavior of neural populations and their role in brain function and dysfunction. In this work, we consider a neural mass model which consists of a network of an arbitrary number of Wilson–Cowan nodes with homeostatic adjustment of the inhibitory coupling strength and time-delayed, excitatory coupling. We extend previous work on this model to include distributed time delays with commonly used kernel distributions: delta function, uniform distribution, and gamma distribution. Focusing on networks which satisfy a constant row sum condition, we show how each eigenvalue of the connectivity matrix may be related to a Hopf bifurcation and that the eigenvalue determines whether the bifurcation leads to synchronized or desynchronized oscillatory behavior. We consider two example networks, one with all real eigenvalues (bidirectional ring) and one with some complex eigenvalues (unidirectional ring). In bidirectional rings, the Hopf curves are organized so that only the synchronized Hopf leads to asymptotically stable behavior. Thus the behavior in the network is always synchronous. In the unidirectional ring networks, however, intersection points of asynchronous and synchronous Hopf curves may occur, resulting in codimension-two Hopf-Hopf bifurcation points. Thus asymptotically stable synchronous and asynchronous limit cycles can occur as well as torus-like solutions which combine synchronous and asynchronous behavior. Increasing the size of the network or a discrete time delay makes these intersection points, and the associated asynchronous behavior, more likely to occur. The effect of distributed time delays is subtle, with some distributions showing asynchronous behavior more likely for small mean delays and less for larger. Numerical approaches are used to confirm the findings, with Hopf bifurcation curves plotted using Wolfram Mathematica. These insights offer a deeper understanding of the mechanisms underlying desynchronization in large networks of oscillators.
mathematics, applied,physics, mathematical