Stochastic and reduced biophysical models of synaptic transmission.

Abstract

Stochastic and reduced biophysical models of synaptictransmission are formulated and evaluated. Thesynaptic transmission involves presynapticfacilitation of neurotransmitter release, depletionand recovery of the presynaptic pool of readilyreleasable vesicles containing neurotransmittermolecules and saturation of postsynaptic receptors ofboth fast non-NMDA and slow NMDA types. The models areshown to display the principal dynamicalcharacteristics experimentally observed of synaptictransmission. The two main types of neural coding,i.e. rate and temporal coding, can be distinguished bymeans of different dynamical properties of synaptictransmission determined by initial neurotransmitterrelease probability and presynaptic firing rate. Fromthe temporal evolution of the postsynaptic membranepotential response to a train of presynaptic actionpotentials at a sustained firing rate, in particularthe steady-state amplitude and steady-state averagelevel of postsynaptic membrane potentials aredetermined as functions of both initial releaseprobability and presynaptic firing rate. The modelsare applicable to studies of the primary stages oflearning processes and can be extended to incorporateshort-term and long-term potentiation in memoryconsolidation processes.