Download PDF (external access)

Current biology

Publication date: 2007-08-01
Volume: 17 Pages: R587 - R589
Publisher: Cell press

Author:

Kempf, Florian
Bruecke, Christof ; Kuehn, Andrea A ; Schneider, Gerd-Helge ; Kupsch, Andreas ; Chen, Chiung Chu ; Androulidakis, Allexandros G ; Wang, Shouyan ; Vandenberghe, Wim ; Nuttin, Bart ; Aziz, Tipu ; Brown, Peter

Keywords:

system, Science & Technology, Life Sciences & Biomedicine, Biochemistry & Molecular Biology, Biology, Cell Biology, Life Sciences & Biomedicine - Other Topics, SYSTEM, Basal Ganglia, Dopamine, Feedback, Psychological, Humans, Levodopa, Motor Activity, Parkinson Disease, Subthalamic Nucleus, 06 Biological Sciences, 11 Medical and Health Sciences, 17 Psychology and Cognitive Sciences, Developmental Biology, 31 Biological sciences, 32 Biomedical and clinical sciences, 52 Psychology

Abstract:

We learn new motor tasks by trial and error, repeating what works best and avoiding past mistakes. To repeat what works best we must register a satisfactory outcome, and in a study [1] we showed the existence of an evoked activity in the basal ganglia that correlates with accuracy of task performance and is associated with reiteration of successful motor parameters in subsequent movements. Here we report evidence that the signaling of positive trial outcome relies on dopaminergic input to the basal ganglia, by recording from the subthalamic nucleus (STN) in patients with nigrostriatal denervation due to Parkinson's Disease (PD) who have undergone functional neurosurgery. Correlations between subthalamic evoked activities and trial accuracy were weak and behavioral performance remained poor while patients were untreated; however, both improved after the dopamine prodrug levodopa was re-introduced. The results suggest that the midbrain dopaminergic system may be important, not only in signaling explicit positive outcomes or rewards in tasks requiring choices between options [2,3], but also in trial-to-trial learning and in reinforcing the selection of optimal parameters in more automatic motor control.