The criterion for a reflex EMG burst was the signal magnitude surpassing a threshold equal to three times the standard deviation of the baseline. The baseline EMG (EMGBG) was estimated from the first 200 ms of the trial, prior to initiation of the stretch,17 and the reflex EMG activity was estimated from a 100 ms window centered on the peak EMG for the period between the start and the end of rotation. Analogous to the torque analysis, the EMG value was normalized to the maximal EMG activity, which was recorded during the trial and demonstrated the maximum torque for each subject.