Evidence that a disordered servo-like mechanism contributes to tremor in movements during cerebellar dysfunction

The characteristics of discontinuities and tremor that occurred in elbow flexions during cooling of the lateral cerebellar nuclei were investigated in five Cebus monkeys. Discontinuities in movements appeared as rhythmical oscillations (kinetic tremor) when movements were slow or when movements were made with a constant force that loaded the antagonist. These oscillations had similar properties to cerebellar terminal tremor following movements; e.g., their amplitude and frequency were decreased by addition of mass to the handle and they occurred in the absence of visual feedback. The abnormal initial decrease in velocity that initiated oscillations in flexion movements was associated with abnormally early or large antagonist (triceps) electromyogram (EMG) activity. This abnormal EMG activity did not follow the normal inverse relation between initial velocity and antagonist latency from onset of movement. The initial deflection from the expected trajectory was opposed by a second burst of EMG activity in the agonist (biceps). This second burst was not the continuation of a step of EMG activity because its amplitude was often larger than the amplitude of the first agonist burst. The second agonist burst had the properties of a servo-like response: it occurred when biceps shortening was slowed (but biceps was not stretched), its magnitude was proportional to the magnitude or the deflection in velocity, its latency was 50-80 ms from onset of the abnormal decrease in velocity, and it occurred in the absence of visual feedback. However, this servo-like response was disordered because it did not return the limb accurately to the expected trajectory. The servo-like mechanism was studied further by applying torque pulse perturbations during elbow flexions. When the cerebellar nuclei were cooled, agonist responses to the perturbation were proportional to the size of the velocity deflection, but they were prolonged and onset of antagonist activity was delayed. It is suggested that discontinuities and tremor in movements during cerebellar dysfunction result from the same mechanism: alternation between disordered stretch reflexes and disordered servo-assistance mechanisms, both partly involving transcortical pathways.