Charla Invitada. "A Permutation Entropy Approach to characterize the Internal LFP dynamics"

FERNANDO MONTANI

Workshop; XII Latin American Workshop on Nonlinear Phenomena (LAWNP-2011); 2011

We investigate the internal dynamics of local field potentials (LFPs) signals obtained from extracellular multisite microelectrode recordings from the mouse visual cortex. In order to do it so, we perform Brant and Pompe?s permutation entropy estimations to take into account the intrinsic temporal structure and causal information of the signal. Our approach is particularly useful to detect the presence of internal dynamical noise on the basis of counting ordinal patterns included through the probability distribution in the entropy estimations. This allows us first to identify the intrinsic embedded dimensionality of the LFP?s energy spectra considering time intervals between consecutives UP and DOWN switching at different frequency bands. Second, characterize low frequency band (1-10Hz) as an ?order state? and the higher band as ?disorder state?. Finally, we perform an Information theoretical approach to answer the question by how much does observing this response sequence reduce the uncertainty about when the start of the LFP response sequence occurs. Our findings show that the low frequency bands play a crucial role in the rhythmic activity pattern in the brain, and in the processing of sensory information. However, it is important to point out that bi-stable LFP are also prevalent and informative in the gamma frequency band of the energy spectra.