Stochastic resonance induced by exogenous noise in a model of a neuronal network. (3rd May 2013)

• Record Type: Journal Article
• Authors: Paffi, Alessandra; Apollonio, Francesca; d'Inzeo, Guglielmo; Liberti, Micaela
• Is Part Of: Network. Volume 24:Number 3(2013)
• Page Start: 99
• Page End: 113
• Publication Date: 2013-05-03
• Subjects: Network models -- single neuron computation; Neural computers -- Periodicals; Neural networks (Computer science) -- Periodicals; 006.32
• DOI: 10.3109/0954898X.2013.793849
• Languages: English
• ISSNs: 0954-898X

Abstract

Abstract : This study investigates the possibility of using exogenous noise to restore the processing performances of neuronal systems where the endogenous noise is reduced due to the ageing or to degenerative diseases. This idea is based on the assumption, supported by theoretical studies, that the endogenous noise has a positive role in neuronal signal detection and that its reduction impairs the system function. Results, obtained on a two-layers feedforward network, show the onset of the Stochastic Resonance (SR) behavior, as long as the exogenous noise is properly tailored and filtered. The amount of noise to be furnished from the outside to optimize the system performance depends on the residual level of endogenous noise, indicating that both kinds of noise cooperate to the signal detection. These results support potentially new bioengineering applications where exogenous noise is furnished to enhance signal detectability.