Differential influence of sinusoidal and noisy inputs on synaptic connections in a network with STDP

We hypothesize that the type of cortical network activation influences synaptic connectivity in the network, eventually expressed in an altered responsiveness to external stimuli. Our predictions are based on a time discrete canonical model of spike-time-dependent plasticity. The results show that, at a given synaptic connection strength in the network, sinusoidal input to the network can decrease synaptic potentiation whereas uncorrelated noise increases synaptic potentiation, implying that these opposing effects manifest themselves in respective decreases and increases of the network response to an external stimulus. These predictions are in qualitative agreement with visually evoked responses obtained in humans after 9 hour periods of visual deprivation (used to increase sinusoidal EEG alpha-activity in cortical networks) or normal daytime vision (as an approximate of noise input).