Correlations between brain activity and components of motor learning in middle-aged adults: an fMRI study

Implicit learning may be shown by improvements in motor performance, which occur unconsciously with practice and are typically restricted to the task that was practiced. The purpose of this study was to examine behaviorally relevant brain activation associated with change in motor behavior during sequence-specific motor learning of a perceptuomotor continuous tracking (CT) task in middle-aged adults. To gain further insight into the neural structures associated with change in motor behavior, overall improvement in tracking (root mean square error; RMSE) was decomposed into two components-temporal precision and spatial accuracy. We hypothesized that individual differences in CT task performance would be evident in unique networks of brain activation that supported overall tracking behavior as well-temporal and spatial tracking accuracy. A group of middle-aged healthy individuals performed the CT task, which contains repeated and random segments for seven days. Functional magnetic resonance imaging (fMRI) data was collected on the first and seventh day while the participants performed the task. Subjects did not gain explicit awareness of the sequence. To assess behaviorally-relevant changes in the blood oxygenation level-dependent (BOLD) response associated with individual sequence-specific tracking performance, separate statistical images were created for each participant and weighted by the difference score between repeated and random performance for days 1 and 7. Given the similarity of performance for random and repeated sequences during early practice, there were no unique networks evident at day 1. On Day 7 the resultant group statistical fMRI image demonstrated a positive correlation between RMSE difference score and bilateral cerebellar activation (lobule VI). In addition, individuals who showed greater sequence-specific temporal precision demonstrated increased activation in the precentral gyrus, middle occipital gyrus, and putamen of the right hemisphere and the thalamus, cuneus, and cerebellum of the left hemisphere. Activation of this neural network further confirms its involvement in timing of movements as it has been previously associated with task performance when individuals are instructed to emphasize speed over accuracy. In the present study, behavioral performance was associated with neural correlates of individual variation in motor learning that characterized the ability to implicitly learn a sequence-specific CT task.