System identification for chaotic integrate-and-fire dynamics

We discuss applications of the fact that dynamical state information can be reconstructed from a series of interspike interval (ISI) measurements. This system analysis allows system identification and prediction from spike train history. Secondly, using this reconstruction, unstable periodic trajectories of the underlying system can be controlled by small changes in a system parameter. The underlying assumption is an integrate-and-fire model coupling the dynamical system to the observable spike train. © 1997 John Wiley & Sons, Inc.