Effects of pallidal deep brain stimulation and levodopa treatment on reaction-time performance in Parkinson's disease.

Basal ganglia-thalamocortical circuits play an important role in movement preparation and execution. Tracer, single-cell, and lesion studies in monkeys suggest the existence of topologically segregated motor and nonmotor basal ganglia cortical circuits. In this study we used deep brain stimulation (DBS) of the posteroventrolateral globus pallidus internus (GPi) in patients with Parkinson's disease to elucidate the function of the GPi in human sensorimotor behavior. This question was investigated by comparing the influence of DBS on patients' performance in various reaction-time tasks that differed with respect to cognitive but not motor requirements. As a main result, DBS improved performance on the different tasks independently of the complexity of the involved cognitive processing functions. Furthermore, the observed effects did not depend on the modality of the processed information. These results suggest that the functional state of the posteroventrolateral GPi selectively affects the motor stage in simple sensorimotor acts, because this stage was the only stage involved in all investigated tasks. In addition to DBS, we manipulated the levodopa medication state of the PD patients. In contrast to DBS, levodopa effects on reaction times were less consistent. Levodopa improved reaction times in choice reaction tasks significantly, while affecting reaction times in a simple reaction task to a lesser extent. Error analysis revealed that the medication-dependent reaction-time improvement in the choice reaction tasks was accompanied by an increase in errors, suggesting a shift of the speed-accuracy criteria of the patients. A similar pattern of results was not observed for the DBS effects. Taken together, our data are in agreement with recent findings in monkeys that indicate a topological organization of the GPi in which motor functions are localized in posterolateral regions apart from cognitive regions. Furthermore, our data show a way to uncover the subcortical-cortical circuitry serving human sensorimotor behavior.