Neural Mechanisms of Reinforcement Learning in the Basal Ganglia The brain's sensitivity to unexpected outcomes plays a fundamental role in an organism's ability to adapt and to learn new behaviors. Emerging research suggests that midbrain dopaminergic neurons encode these unexpected outcomes, yet the central role these neurons play in mediating reinforcement learning has not been established in humans. The goal of this proposal is to use data obtained from microelectrode recordings during deep brain stimulation surgery (DBS) for the treatment of Parkinson's disease (PD) and obsessive compulsive disorder (OCD) to formally investigate learning mechanisms within the human basal ganglia. We propose to examine neural activity in the human substantia nigra (SN), subthalamic nucleus (STN), and nucleus accumbens (NA) while patients undergoing DBS are awake and engage in a probabilistic learning task.
Our aim i s to establish the functional relationship between activity in dopaminergic SN neurons and prediction error, between activity in STN neurons and the temporal dynamics of reinforcement learning and decision conflict, and between activity in NA and unexpected rewards. In addition, we will examine the possibility of augmenting and replacing reward signals in the basal ganglia with electrical stimulation to recapitulate activity within the basal ganglia. By demonstrating that cognitive feedback elicits changes in the activity of these basal ganglia structures, we hope to establish the central role the basal ganglia plays in mediating reinforcement learning in humans. And by demonstrating that electrical stimulation can modulate these learning processes in the basal ganglia, we hope to establish a causative role for activity within these structures, and to establish a means by which electrical stimulation can be used to counteract some of the cognitive deficits seen in patients with disorders of the basal ganglia.
Emerging research suggests that basal ganglia nuclei encode unexpected rewards, yet the central role the basal ganglia play in mediating reinforcement learning has not been established in humans. The goal of this proposal is to use microelectrode recordings and microstimulation during deep brain stimulation surgery (DBS) for the treatment of Parkinson's disease (PD) and obsessive compulsive disorder (OCD) to formally investigate learning mechanisms within the human substantia nigra (SN), subthalamic nucleus (STN), and nucleus accumbens (NA) while patients are awake and engage in a probabilistic learning task.
| Zaghloul, Kareem A; Weidemann, Christoph T; Lega, Bradley C et al. (2012) Neuronal activity in the human subthalamic nucleus encodes decision conflict during action selection. J Neurosci 32:2453-60 |
