Two-Compartment Stochastic Model of a Neuron With Periodic Input

. A neuronal model made of two interconnected parts - a dendrite and a trigger zone - is considered under the action of white noise and periodic inputs acting on the dendritic compartment. It is shown analytically how the variability of the depolarization potential is decreased from the dendritic to the trigger zone for subthreshold behavior. These filtering properties of the model are also shown when a reset mechanism is included at the trigger zone compartment. 1 1 Introduction The leaky integrate-and-fire model of a neuron appears to be the most common concepts in computational neuroscience as well as in the theory of artificial neural networks. This model is based on very simplified assumption considering all the properties of the neuron concentrated into a single point in space (single-point models). In contrast to this neuron representation, there are computerized multicompartment models or cable models. These spatially complex models usually do not aim at direct descri..