The purpose of this study was to identify consistent characteristic changes of neuronal activity in basal ganglia (BG) nuclei associated with Parkinson's disease (PD) so that a reliable index of PD can be derived. A simple algorithm for automatic identification of firing patterns was devised as an essential tool to achieve this goal. A detailed quantitative analysis of firing patterns as well as firing rate was performed in three BG nuclei: the subthalamic nucleus (STN), the substantia nigra pars reticulate (SNpr), and the globus pallidus (GP). The results showed that the firing rate of STN neurons was not significantly altered in PD model rats. We also did not find a significant alteration in firing rates in the SNpr and GP between normal and PD model rats. In contrast, consistent changes of firing patterns were observed in all three BG nuclei in that the percentage of neurons with a regular firing pattern decreased whereas those with irregular, mixed, or burst patterns increased. This enables a simple algorithm based on burst detection and the shape of the interspike interval histogram to identify whether the neuronal activity is from normal or PD model rats.