Noise Enhanced Signaling in STDP Driven Spiking-Neuron Network

¹ Lobachevsky State University of Nizhny Novgorod, Gagarin Ave. 23, 603950 Nizhny Novgorod, Russia
² Instituto de Matemática Interdisciplinar, Dept. de Matemática Aplicada, Universidad Complutense de Madrid, Avda Complutense s/n, 28040 Madrid, Spain

^* Corresponding author. E-mail: lobov@neuro.nnov.ru
^** Corresponding author. E-mail: vmakarov@ucm.es

Abstract

Population spike signaling is widely observed both in intact brain and neuronal cultures. Experimental evidence suggests that a locally applied electrical stimulus can shape the network architecture and thus the neuronal response. However, there is no clue on how this process can be controlled. Here we study a realistic model of a culture of cortical-like neurons with spike timing dependent plasticity. We show that a stimulus applied at a corner of the culture can rebuild synaptic couplings. Then the network eventually switches from a turbulent-like asynchronous spiking to an ordered population spike signaling mode. The structural analysis shows that the stimulus potentiates centrifugal couplings, which promotes spiking waves traveling outwards the stimulus location. This phenomenon can be catalyzed by noise of an intermediate strength. We predict that matured cultures with high connectivity are more susceptible to reconfiguration and generation of a population spike response than young cultures with low connectivity. We also report on an intermittent synchronization causing switches between two quasi-stable states: generation of time-locked population spikes and turbulent spiking. In the turbulent mode the stimulus excites patches of spiking activity randomly traveling in the network. Such a regime can be implemented through a large scale looping of couplings backwards to the stimulus location. We anticipate that the robust mechanisms of shaping the network architecture discussed here can also be effective in more complex preparations and studies of the relationship between network structure and function.