Striatal dopamine release in sequential learning

Abstract

Sequential learning is an important aspect of cognitive processing. Neuropharmacological evidence acquired in laboratory animals suggests that striatal dopaminergic mechanisms may be important for processing of this form of learning. However, because experiments conducted on dopamine deficient patients have reported contradictory evidence, the role of dopamine and the striatum remains unclear in human sequential learning. We used a newly developed dynamic molecular imaging technique to determine whether striatal dopamine is released during performance of a sequential learning task. In this study we localized striatal regions where dopamine receptor ligand (11C-raclopride) was displaced from receptor sites, during performance of a motor sequence learning (serial reaction time) task. The results suggest that the task induces release of endogenous dopamine in the posterior two-third of dorsomedial aspect of left putamen and the anterior part of the body of caudate bilaterally. The activations of the left putamen and the right caudate coincided with the activations observed earlier during performance of a motor planning task. Since these activations are associated with the selection and execution of a response, the activation in the left caudate, which was not observed in motor planning, is probably associated with the detection of a change in the 'context', and in the formulation of a new 'rule'. Thus, the results suggest that sequential learning involves two striatal dopaminergic mechanisms, one for the detection of a change in context, and the other for selection and execution of the response.

Figures

Figure 1

Response Time: Mean response time during the SRT task performance. Stimuli were presented randomly in the first 25 min and sequentially thereafter (vertical line).

Figure 2

SRT Experiment (cohort): The figure shows t-maps indicating group difference in the rate of ligand displacement before and after task initiation. It suggests that the rate increased significantly (t>3.0) after the task initiation in the anterior part of the body of the caudate bilaterally, and in the dorsolateral aspect of posterior two-third of the left putamen.

Figure 3

SRT Experiment (single subject): The pattern of activation observed in the data pooled across subjects (Figure 2), was replicated in individual volunteers. This figure shows t-maps drawn from the data acquired in a volunteer (JW). The time activity curves show the concentration histories (open circles) and least square fits (solid lines) for the ligand (11C-raclopride) in the left anterior caudate (left lower panel) and left posterior putamen (right lower panel). The PET concentration history of the reference region (cerebellum) is also shown (lower curves). The rate of ligand displacement did not change significantly during task performance in this region.

Figure 4

Control Experiment: The time activity curves drawn from the left posterior putamen and left anterior caudate (same regions from where curves were drawn in Figure 3) in the control experiment in which subjects did not make any response. There was no significant change in the rate of ligand displacement in this experiment either in the striatal (upper curves) or in the reference region (lower curves).