Electromyographic Data Reduction
To analyze the levels of EMG activity during stretching, TA and soleus muscle EMG data were transmitted from FM tape through a full-wave rectifier and a low-pass filter (25-msec time constant, cutoff frequency 6 Hz)28 and then analog/ digital converted(sampling rate 50Hz) for input into a PDP 11-34 computer. Baseline EMG levels with the ankle
resting at 90 degrees were sample for a 1-second period at the beginning of each trial. The first four H-reflex stimulus artifacts of each trial (ISI = 10 sec) served as a signal for the computer to begin sampling EMG activity during muscle stretching. No EMG activity was sampled 1 second before or 1 second after the stimulus artifact. Digitized and smoothed EMG activity then was sampled for an 8-second period. Thus, four8-second periods of both soleus and TA muscle EMG activity were sampled for each trial. A second computer program was used to integrate the sampled rectified and filtered EMG data over 1-second time segments. The integrated value of the baseline EMG sample was subtracted
from the integrated values of the EMG activity sampled during the stretch. Only the last seven 1-second time segments of each period were included; the first was omitted because it was frequently an extreme value.