The purpose of this study is to examine the validity of muscle fatigue evaluation using maximum voluntary torque (MVT), and to identify the dependence of individual's tolerance for fatigue on the capacity to exert MVT. In 14 young male subjects (10 regular exercisers and 4 sedentary), MVT was measured during isometric knee extension, and voluntary activation (VA), which reflects motor unit activation, was evaluated using the twitch interpolation technique. In addition, the maximum endurance time (ET) was measured, and behavior of the mean power frequency (MPF) and the average rectified value (ARV) of surface EMGs from the vastus lateralis muscle were analyzed during constant force isometric contractions of 60% MVT (short-duration fatigue task; SDF task) and 20% MVT (long-duration fatigue task; LDF task) . Correlations were examined among these five variables.
The results were as follows:
1) Subjects were divided into a high voluntary activation group (HVA group) and a low voluntary activation group (LVA group) . Four sedentary subjects were included in the latter group.
2) MVT was significantly larger in the HVA group than in the LVA group (p<0.01) . A significant positive correlation (r=0.72) was found between MVT and VA (p<0.01) .
3) A significant negative correlation (r=-0.71) was found between MVT and endurance time (ET) for the LDF task (p<0.01) . The ET was significantly longer in the LVA group than in the HVA group (p<0.01) .
4) The MPF of voluntary EMG decreased consistently, as ARV increased during isometric contraction in both tasks (p<0.01), indicating the development of fatigue in the muscle. The final change of MPF relative to the initial value was significantly greater in the SDF task than in the LDF task (p<0.05) .
5) A significant correlation (r=-0.83) was seen between the relative change in MPF and ARV in the SDF task (p<0.01) .
6) For the SDF task, the final change of MPF and ARV relative to the initial value was significantly greater in the LVA group than in the HVA group (p<0.05) .
These results indicate that tolerance for local muscle fatigue usually evaluated as maximum endurance time, may depend on individual differences in VA, the VA, in turn, depending on adapta-tion to exercise, and that there appears to exist a corresponding adaptative strategy of the neuromuscular system during fatiguing contractions. Usefulness of our procedure using the twitch interpolation technique in evaluating muscle fatigue was also suggested.