Motor subcircuits mediating the control of movement velocity: a PET study.

The influence of changes in the mean velocity of movement on regional cerebral blood flow (rCBF) was studied using positron emission tomography (PET) in nine healthy right-handed adults while they performed a smooth pursuit visuomanual tracking task. Images of relative rCBF were obtained while subjects moved a hand-held joystick to track the movement of a target at three different rates of a sinusoidal displacement (0.1, 0.4, and 0.7 Hz). Significant changes in rCBF between task conditions were detected using analysis of variance and weighted linear contrasts. The kinematics of arm and eye movements indicated that subjects performed tasks in a similar manner, particularly during the faster two tracking conditions. Significant increases in rCBF during arm movement (relative to an eye tracking only control condition) were detected in a widespread network of areas known for their involvement in motor control. The activated areas included primary sensorimotor (M1S1), dorsal and mesial premotor, and dorsal parietal cortices in the left hemisphere and to a lesser extent the sensorimotor and superior parietal cortices in the right hemisphere. Subcortically, activations were found in the left putamen, globus pallidus (GP), and thalamus, in the right basal ganglia, and in the right anterior cerebellum. Within the cerebral volume activated with movement, three areas had changes in rCBF that correlated positively with the rate of movement: left M1S1, left GP, and right anterior cerebellum. No movement-related sites had rCBF that correlated negatively with the rate of movement. Regressions of mean percent change (MPC) in rCBF onto mean hand velocity yielded two nonoverlapping subpopulations of movement-related loci, the three sites with significant rate effects and regression slopes steeper than 0.17 MPC.cm-1.s-1 and all other sites with nonsignificant rate effects and regression slopes below 0.1 MPC.cm-1. s-1. Moreover, the effects of movement per se and of movement velocity varied in magnitude independently. These results confirm previous reports that movement-related activations of M1S1 and cerebellum are sensitive to movement frequency or some covarying parameter of movement. The activation of GP with increasing movement velocity, not described in previous functional-imaging studies, supports the hypothesis that the basal ganglia motor circuit may be involved preferentially in controlling or monitoring the scale and/or dynamics of arm movements. The remaining areas that were activated equally for all movement rates may be involved in controlling higher level aspects of motor control that are independent of movement dynamics.