Implicit motor learning tasks typically involve comparisons of subject responses during a sequence versus a random condition. In neuroimaging, brain regions that are correlated with a sequence are described, but the temporal relationship of sequence versus nonsequence conditions is often not explored. We present a functional magnetic resonance imaging (fMRI) study describing activation related to sequential predictability in an implicit sensorimotor learning task and the history (context) dependence of these effects. Participants regarded four squares displayed horizontally across a screen and pressed a button when any one of the four targets was illuminated in a particular color. A repeating spatial sequence with varying levels of predictability was embedded within a random color presentation. Both the right dorsolateral prefrontal cortex (R DLPFC) and right caudate displayed a positive correlation to increasing predictability, whereas the left posterior parietal cortex (L PPC) displayed a negative correlation. However, the activation changes within the caudate were significant when transitioning from high predictability to low predictability but not for the reverse case, suggesting a sensitivity not only to predictability but to order effects as well. These results support the hypothesized relationship between basal ganglia and visuomotor sequential learning, but demonstrate the importance of context upon sequence learning.

We propose a systems-level computational model of the basal ganglia based closely on known anatomy and physiology. First, we assume that the thalamic targets, which relay ascending information to cortical action and planning areas, are tonically inhibited by the basal ganglia. Second, we assume that the output stage of the basal ganglia, the internal segment of the globus pallidus (GPi), selects a single action from several competing actions via lateral interactions. Third, we propose that a form of local working memory exists in the form of reciprocal connections between the external globus pallidus (GPe) and the subthalamic nucleus (STN). As a test of the model, the system was trained to learn a sequence of states that required the context of previous actions. The striatum, which was assumed to represent a conjunction of cortical states, directly selected the action in the GP during training. The STN-to-GP connection strengths were modified by an associative learning rule and came to encode the sequence after 20 to 40 iterations through the sequence. Subsequently, the system automatically reproduced the sequence when cued to the first action. The behavior of the model was found to be sensitive to the ratio of the striatal-nigral learning rate to the STN-GP learning rate. Additionally, the degree of striatal inhibition of the globus pallidus had a significant influence on both learning and the ability to select an action. Low learning rates, which would be hypothesized to reflect low levels of dopamine, as in Parkinson's disease, led to slow acquisition of contextual information. However, this could be partially offset by modeling a lesion of the globus pallidus that resulted in an increase in the gain of the STN units. The parameter sensitivity of the model is discussed within the framework of existing behavioral and lesion data.