Correlation between the dischanges of two simultaneously recorded motor units and physiological tremor

Simultaneous recording was performed from pairs of motor units either in a small intrinsic hand muscle (first dorsal interosseus, 6 subjects, 37 records) or in two synergistic calf muscles (gastrocnemius and soleus, 6 subjects, 50 records). The two motor units were recorded by means of two electrodes inserted into the muscle during stationary isometric contractions of different strengths. Cross-correlograms were calculated in two ways: (1) between spike trains of two motor units; (2) between either spike train and the peaks of the tremor record. The cross-correlation between pairs of motor units showed a consistent tendency to synchronization in the hand as well as in the calf muscles. This synchronization was higher than could be expected from the random activity of asynchronously firing motor units. The amount of synchronization was correlated with the amplitude of physiological tremor: the stronger the tremor, the higher the synchronization coefficient. The cross-correlation between a spike train and the peaks of the tremor records showed a higher probability of unit firing 30-60 msec prior to the tremor beats. It is concluded that the tremor force produced by one muscle or a pair of synergistic muscles is the result of the synchronized activity of motor units. This hypothesis is supported by some characteristic changes in the amplitudes and rates of tremor in Parkinson's disease and myopathies. In these diseases the pathological changes in tremor rates simply reflect underlying changes of the motoneuronal firing rates.