Activitydriven Computational Strategies of a Dynamically Regulated Integrateandfire Mode.pdf

Activity-driven Computational Strategies of a Dynamically Regulated Integrate-and-fire Model Neuron Michele Giugliano, Marco Bove, Massimo Grattarola N.B.T. – Neural and Bioelectronic Technologies Department of Biophysical and Electronic Engineering University of Genoa Via Opera Pia 11A, I - 16145 Genoa – ITALY email: michi@dibe.unige.it url: http://www.bio.dibe.unige.it (to appear on the Journal of Computational Neuroscience, Vol. 7, No. 3, 1999) Activity-dependent slow biochemical regulation processes, affecting intrinsic properties of a neuron, might play an important role in determining information processing strategies in the nervous system. We introduce second-order biochemical phenomena into a linear leaky integrate-and-fire model neuron together with a detailed kinetic description for synaptic signal transduction. In this framework, we investigate the membrane intrinsic electrical properties differentiation, showing the appearance of activity-dependent shifts between integration and temporal coincidence detection operating mode, for the single unit of a network. Key Words: integrate-and-fire, dynamic regulation, neural code, integration, coincidence detection. INTRODUCTION Two main competing proposals are presently debated by the neuroscience community, in order to explain information representation strategies