Fig. 8.

Correlation between EMG levels and kinematic variables. A, The potential recorded during a representative reaching movement. The trace is partitioned into bins of 100 msec, with the reference time for the division being the bend time (vertical gray bar in A andB). B, Correlations between peak velocity of the movement and the integrated EMG during the corresponding time bin (bins as in A). The coefficient of correlation (r) is indicated in each graph. C, Standardized integrated EMGs over the 100 msec before the bend time plotted against each of the kinematic variables: peak velocity, local velocity (measured at the bend time), global acceleration (see Results), and local acceleration. Data were pooled from all five experiments. The linear coefficient of correlation (r) is indicated in each graph. Anf test examined the statistical significance of the correlation. The histogram in D shows the number of experiments in which p < 0.01 for each kinematic variable.