Reward- and effort-related neuronal activity in the subthalamic nucleus of Parkinson’s disease patients.

Recent studies have suggested that the inadequate evaluation of the cost-benefit ratio of actions in Parkinson Disease (PD), leads to a suboptimal allocation of resources, which would be responsible for the bradykinesia. In addition there is evidence indicating that the Subthalamic Nucleus (STN) is involved in motivated behaviours and in reward-related processes. Starting from these considerations, we tried to determine the extent to which variables that are critical to cost-benefit computation can be identified in the neural activity of STN. We also evaluated how the dopamine level affects the representation of these variables and how these changes relate to the behavioral deficits in PD. To address these issues, we recorded local field potentials (LFP) in the STN of 7 PD patients having benefited from the implantation of deep brain stimulation (DBS) electrodes. During these recordings, the patients were asked to perform a task requiring to squeeze a dynamometer with variable force and with a promise of a variable virtual monetary reward. In each trial, the patients were allowed to choose to either perform or skip the effort proposed as a function of effort intensity and reward amount. These two pieces of information were provided by visual cues at the beginning of each trial. Each patient was tested twice, ON and OFF dopamine replacement therapy. Preliminary results indicate that robust responses to the different visual cues are present in STN and that these responses are strongly modulated by reward, effort and choice variables. The nature of these responses differs according to the frequency bands. Responses in the beta range are inversely proportional to the probability of choosing to execute the effort whereas gamma range signals show the opposite effect. The profiles of these activities are compatible with drift-diffusion models of decision making and suggest that gamma range responses could be involved in cost-benefit computations prior to the decision of whether or not to execute the effort whereas beta range responses would represent inhibitory signals acting to delay the actual response until appropriate. We also found that dopamine replacement therapy increases the probability to accept to execute a trial. In accordance with the above findings, dopamine decreased beta band and increased gamma band activity dramatically. In conclusion, our preliminary data indicate that signals relevant to effort-based decision making can be identified in STN. It remains to be determined whether these signals are causally involved in cost-benefit decisions and whether their perturbation during PD plays a role in the symptoms of the disease.