RESUMO
Fractal analysis was applied to human pallidal neuronal spike trains recorded from patients with Parkinson's disease during ablative surgery of the internal segment of the globus pallidus. Fractal dynamics was quantified by computing the scaling exponent with the average wavelet coefficient approach. We observed fractal persistent correlation in the fluctuation of the interspike intervals of neuronal spike trains recorded in the internal segment of the globus pallidus both before and after the administration of dopamine agonist apomorphine. However, there was a significant increase in the scaling exponent during the "on" state after apomorphine administration as compared with the parkinsonian "off" state prior to apomorphine. In addition, we observed a statistically significant decrease in the average firing rate in the transition from the "off" to the "on" state. We conclude that robust fractal dynamics can be observed in single neurons in the human CNS, indicating that human neuronal dynamics of the internal segment of the globus pallidus are essentially a nonlinear and nonequilibrium process, with a long-range correlation or memory extending across many time scales. Accompanying the "on" state after apomorphine administration was an improvement in the long-range persistent correlation as compared with the more random dynamics in the "off" state. A scaling exponent signaling a breakdown or modification in long-range correlation in a single neuron may serve as a useful indicator of a dysfunctional network in the human CNS.