Max Planck Institute for Dynamics and Self-Organization -- Department for Nonlinear Dynamics and Network Dynamics Group
Personal tools
Log in


Tuesday, 08.01.2013 17 c.t.

Chaos and reliability in fluctuation-driven, balanced spiking networks

by Guillaume Lajoie
from Department of Applied Mathematics, University of Washington, Seattle, USA

Contact person: Fred Wolf


Ludwig Prandtl lecture hall


The question of reliability arises for any dynamical system driven by an input signal: if the same signal is presented many times with different initial conditions, will the system entrain to the signal in a repeatable way? Reliability is of particular interest in large, randomly coupled networks of excitatory and inhibitory units. Such networks are ubiquitous in neuroscience, but are known to autonomously produce strongly chaotic dynamics – an obvious threat to reliability. Here, we show that such chaos also occurs in the presence of weak and strong stimuli. However, even in the chaotic regime, intermittent periods of highly reliable spiking often coexist with unreliable activity. We propose a framework to better understand these complex dynamics, leveraging results from random dynamical systems (RDS) theory, by establishing the effect of the underlying chaotic attractor's geometry on output spike trains.

back to overview