Surrogate data test for nonlinearity of the rat cerebellar electrocorticogram in the model of brain injury

Abstract

The surrogate data test for nonlinearity has been used in order to establish the existence of nonlinear dynamics and justify the use of nonlinear tools in time series analysis. We applied Higuchi fractal dimension and third-order correlation function on the rat electrocortical activity as discriminative statistics. Our particular interest in this study was to investigate the nonlinearity of cerebellar electrocortical signals in rat model of acute and repeated cerebral cortical injury. We performed the surrogate data test for nonlinearity by using the algorithm of statically transformed autoregressive process (STAP) to generate the surrogate data. Surrogate data test for nonlinearity indicated that cerebellar cortical signals have mostly nonlinear properties during all experimental conditions in the model of repeated cerebral cortical injury. We conclude that results of testing nonlinearity by Higuchi fractal dimension as discriminative statistic are more stable than those obtained by the third-order correlation.