Encoding information in neural feedback circuits

Studies the storage of information in recurrent neural feedback circuits modeled by nonlinear delay-differential equations (DDE's). The information is encoded into the initial function from which the DDE evolves into a periodic solution. We find that the Mackey-an der Heiden (MH) model of recurrent inhibition (RI), involving non-spiking neurons, has multistable periodic patterns of voltage, and thus of spike patterns. The spiking solutions of Plant's model, based on the Fitzhugh-Nagumo equations, are also found to be encodable. Our approach offers a simple mechanism for active short-term memory based on the temporal patterns of a simple neural system rather than on the spatial properties of a neural network.