Given the lack of between-group differences it was not surprising that there were also no betweengroup differences in the peak amplitude of the electromyogram signal, the rate of EMG rise (RER), and mean average voltage (MAV) measured in the time intervals of 0–30, 0–50, 0–100, and 0–200 ms from the superficial quadriceps muscles, or the antagonist biceps femoris. Somewhat surprisingly, however, we found a lack of statistically significant change in these parameters when the data were collapsed across training groups. This finding contrasts with that of Aagaard et al., who reported significant increases in the rate of muscle activation of the quadriceps, as measured by surface EMG, after 14 weeks of heavy strength training in previously untrained individuals. It also differs from the finding of Barry et al., who showed that increases in the RFD of arm flexor muscles occurred simultaneously with an increase in the rate of activation of those muscles after
4 weeks of resistance training, and Del Balso and Cafarelli, who showed a close association between increases in RFD and the rate of muscle activation of the plantar flexors after 4 weeks isometric training. Nonetheless, our results are consistent with those of Narici et al., who found that increases in peak torque and RFD were not associated with a detectable change in the surface EMG signal, and of Holtermann et al., who reported increases in RFD of the plantarflexors with no change in EMG of agonist or antagonist muscles. One similarity between the present study and that of Narici et al. was the use of the knee extension exercise in training, as opposed to Aagaard et al., for example, who utilized several different lower limb training movements. Thus, the possibility exists that there is an increase in the muscle force developed for a given level of muscle activation (i.e., increased neuromuscular efficiency) that allows increases in RFD measured during tasks
that are similar to the training exercise, but increased recruitment of muscle is probably required
for a similar increase in RFD when the training and testing tasks are dissimilar. Alternatively, a lack of significant change in the surface EMG signal could be attributed to amplitude cancellation or a high level of interindividual variability in the change in recruitment improvement combined with the inherent intraindividual variability of the EMG collection and quantification (e.g., see Table 1); that is, the lower reliability of EMG measures reduces the likelihood of detecting change. The variability of EMG measures is an important consideration when examining temporal adaptations in neuromuscular activation.