Parkinsonian patients develop debilitating motor complications caused by the best treatments of the disease, the levodopa and the dopamine agonists. Among other phenomena, it is known that these complications, named 'dyskinesia', are accompanied by a dramatic reduction in the activity of the structures involved in the control of movement. We have previously characterized some of these disturbances by recording the electrophysiological activity of a key nucleus of the network, the globus pallidus, in the non-human primate model of Parkinson's disease. More recently, it has been proposed that the dyskinesia might also be accompanied by an increased synchronization of firing of the network. This means that, instead of firing independently, all the neurons of the network would fire together, further disturbing the selection of the desired motor program by the network. Our project is thus aimed at identifying in dyskinetic non-human primates the electrophysiological correlates of levodopa-induced and dopamine agonist-induced dyskinesia using multi-single-unit recording with special attention paid to synchronization within and between the nuclei of the whole network. The potential clinical implications of the proposed studies are that once we understand exactly the role of synchronization in the basal ganglia during dyskinesia, it might be possible to devise new rational therapies based upon normalization of this hitherto un-investigated parameter.