An inevitable result of a prolonged voluntary muscle contraction is fatigue. During sustained skeletal muscle activation, factors that serve to increase force (e.g., postactivation potentiation) compete with factors that decrease force production. Ultimately, however, a loss in force-generating capacity ensues, and the ability to maintain a given force declines. As muscle fatigue progresses, adaptations in the way in which the muscle is activated by the nervous system occur; in particular, there are fatigue-related changes in the motor unit discharge rate.
Marsden et al. (14) developed the “muscular wisdom hypothesis” that proposed that the decrease in the motor unit discharge rate during prolonged contractions served to minimize fatigue. Their studies examined the mechanisms for the slowing of the discharge rate of motor units in primarily sustained maximal voluntary contractions (MVCs). Bigland-Ritchie et al. (1) related the reduction of the motor unit discharge rate during fatiguing MVCs to the slowing of relaxation rate of the whole muscle. They suggested that, with fatigue, the slowing of the motor unit discharge rate would result in a fully activated muscle because of the reduced fusion frequencies associated with the prolonged relaxation times.
The purpose of this review is to provide an overview of the muscular wisdom hypothesis, commonly called the muscle wisdom hypothesis, and to examine its applicability for minimizing muscle fatigue across different types of contraction that have been tested since its conception.
MUSCULAR WISDOM HYPOTHESIS
The muscular wisdom hypothesis stemmed from prior work that sought to mimic the decrease in force evident during MVCs using maximal electrical stimulation. Jones et al. (12) found that no single frequency of stimulation could reproduce the pattern of force loss observed during a 1-min MVC of adductor pollicis muscle (Fig. 1). In Figure 1A, the pattern of force loss during an MVC was fairly linear over time. Force produced by an initially high frequency of stimulation that was progressively reduced to lower frequencies (60 Hz to 20 Hz) imitated the force loss of a sustained MVC. In Figure 1B, it can be seen that the application of 80 Hz of stimulation was required to produce force similar to that at the beginning of the MVC, but constant stimulation at this rate resulted in a more rapid loss of force than in the MVC (possibly due to activation failure). Conversely, a constant 20 Hz stimulation resulted in an increase in force during the 1 min but failed to produce the high forces evident at the beginning of the MVC. Thus, the data acquired through electrical stimulation suggested that there would be an advantage in progressively reducing the motor unit discharge rate during a sustained MVC for maintaining high levels of force production.
The main premise of the muscular wisdom hypothesis is that there is a matching of the motor unit discharge rate with the fatigue-related contractile properties of the muscle. The concurrent slowing of the twitch relaxation of whole muscle as the muscle fatigues (1) has been observed also in single motor units and muscle fibers (6). Slowing and decreasing the amplitude of the twitch contraction with fatigue may not result in a loss of peak tetanic force, although the force-time integral is smaller (3). It is thought that sensory afferent nerves originating within the fatigued muscle provide a signal to the motoneuron pool to induce a lower discharge rate of the motor units (1,6). The reduction in discharge rate that appeared to match the requirements of the muscle was predicated on fatigue. That is, slowing of the relaxation time of the muscle by cooling the hand did not have an effect on the motor unit discharge rate in the adductor pollicis muscle during MVCs (14). Similarly, Bigland-Ritchie and colleagues (2) changed the contractile properties of the human tibialis anterior muscle by shortening the muscle length. They found no difference in the motor unit discharge rates despite differences in contractile properties regardless of whether subjects were performing maximal or submaximal contractions. Thus, manipulating the contractile properties of the muscle alone, in the absence of fatigue, did not induce a concomitant change in the motor unit discharge rate.
The muscular wisdom hypothesis has become widely accepted and, for example, is appearing in textbooks as a fundamental motor control construct. However, it is important to consider that the conditions under which the muscular wisdom hypothesis was developed were focused on MVC and electrical stimulation protocols. Recently, motor units in the extensor hallucis longus muscle were shown to have no systematic decreases in discharge rate during sustained MVCs (13). This suggests that the muscular wisdom hypothesis may not be applicable to all muscle groups; in this case, a fatigue-resistant postural muscle showed discharge behavior quite distinct from that previously observed in hand muscles. Furthermore, given that most activities performed by humans involve submaximal voluntary contractions, it is critical to consider the validity of the muscular wisdom hypothesis under different conditions.
Subcutaneous fine wire electrodes have been used to investigate the modulation of discharge rate in single motor units throughout prolonged submaximal fatiguing contractions. Typical motor unit behavior in triceps brachii muscle that has been observed during sustained 20% MVC is depicted in Figure 2 (9). The activity of a single motor unit that was active from the onset of the task exhibited a decline in discharge rate, and additional motor units were recruited as the contraction ensued. Overall, there is diversity in the response patterns of motor units during prolonged submaximal fatiguing contractions. That is, some motor units increased in discharge rate and others demonstrated a quasi-constant discharge rate, but the majority of motor units that were active from the beginning of the contraction displayed a decline in discharge rate (8,9). This decline was not accompanied by a slowing of the whole muscle contractile properties. Because low threshold motor units are activated predominantly in low-force contractions, the high threshold motor units (with high force and fast contraction and relaxation times) would be relatively unaffected by fatigue (9). The contribution of the nonfatigued high threshold motor units to the whole muscle twitch could explain the maintenance of the relaxation time. There needs to be further investigation to determine whether the motor units that exhibit a decline in discharge rate during submaximal fatigue also demonstrate a slowing of contractile properties that is not evident when measuring the whole muscle twitch.
The obvious question that arose from the aforementioned data is what predisposed some motor units to demonstrate an increase in discharge rate whereas the majority of others showed a decrease. The argument that a decrease in discharge rate would serve to avoid activation failure does not seem viable in submaximal contractions in which the discharge rates are quite low (10–25 Hz) and unlikely to produce activation failure. The modulation of the motor unit discharge rate during fatigue was not found to correlate with the recruitment threshold of the motor unit. Instead there was a strong relationship between the initial discharge rate of the motor unit and its modulation during fatiguing isometric contractions with a wide range of target forces (10–50% MVC). The initial discharge rate refers to the mean discharge rate during the first few seconds of contraction. Figure 3A demonstrates that a very similar relationship exists in data collected from biceps (8) and soleus muscles (our unpublished data), with biceps motor units tending to discharge at higher rates than soleus motor units (Ybiceps = 5.59–0.46X, R = −0.53, P = 0.0002; Ysoleus = 8.89–1.06X, R = −0.89, P < 0.0001; where X is the initial discharge rate and Y is the change in discharge rate). Those motor units that discharged at initially low rates demonstrated an increase during the contractions, whereas those that discharged initially at high rates showed a decline in rate.
The majority of motor units recorded from triceps brachii muscle during a fatiguing dynamic contraction did not exhibit a decline in the motor unit discharge rate (10,15). Figure 3B depicts the modulation of the motor unit discharge rate as a function of the initial discharge rate for triceps brachii during isometric (9) and dynamic (10) contractions. In the dynamic contractions, subjects performed 100 horizontal flexion and extension movements of the elbow against a preload of 20% MVC that resisted elbow extension. Each movement was separated by a 4-s sustained 20% MVC isometric contraction. The target force in the dynamic protocol was the same as in the isometric protocol, and the same amount of fatigue was achieved (approximately 30% decline in MVC) in both protocols. However, note the difference in slope between the isometric and the dynamic contraction (Yisometric = 4.93–0.64X, R = −0.86, P < 0.0001; Ymovement = 1.62–0.17X, R = −0.50, P = 0.03; where X is the initial discharge rate and Y is the change in discharge rate). Thus, the decline in the motor unit discharge rate during submaximal contractions is task dependent.
One possible explanation for the motor unit behavior during fatiguing dynamic contractions was that the spindle input to the motoneuron pool assisted in the maintenance of the discharge rate in dynamic contractions. To test this hypothesis, vibration was applied for 2 s every 10 s to the triceps brachii muscle during an isometric fatiguing 20% MVC of the elbow extensor muscles. This prevented the decrease in discharge rate in the majority of motor units (11). The muscular wisdom hypothesis ascribes a functional significance to the decline in firing rate, that of minimizing fatigue. If the hypothesis is valid in all situations, then the failure of the discharge rate to decline ought to have resulted in less “minimization” of fatigue. This was not observed in that neither a greater amount nor a faster rate of fatigue occurred (11).
Another limitation of the work on the muscular wisdom hypothesis is that the pattern of activation is usually ignored. That is, changes in mean discharge rate are usually reported as opposed to the instantaneous discharge rate that could take the form of double discharges or very short interspike intervals. Short interspike intervals positioned at the beginning of a spike train can increase the peak force and the rate of rise in force, and reduce the amount of fatigue in human skeletal muscle (3). The number of short interspike intervals (<20 ms) was found to increase in fatiguing dynamic contractions (10), consistent with the notion that patterning of motor unit discharge may serve to compensate for the loss of muscle force production during fatiguing arm movements. Furthermore, the deleterious effects of fatigue evoked by a constant frequency stimulation train can be offset with the use of a short 5-ms interspike interval positioned at the beginning of a stimulation train (3).
Force-frequency curves depict the force output across a wide range of frequencies of stimulation. A typical nonfatigued force-frequency curve is presented in Figure 4 (solid line). A basic tenet of the muscular wisdom hypothesis is that the same amount of force can be generated with a reduction in the frequency of activation (discharge rate) because of the slowing of the contractile properties of the muscle, drawn as a shift in the force-frequency relationship to the left (Fig. 4, dotted line). However, it is important to consider an intracellular property of skeletal muscle (potentiation) that also has the ability to shift this relationship to the left. The force of contraction of an isometric muscle twitch increases after a brief high-frequency conditioning electrical stimulation (posttetanic potentiation) or voluntary contraction (postactivation potentiation). The combination of twitch potentiation and decreases in motor unit firing rate was suggested to be sufficient to maintain a submaximal force in the absence of recruitment of additional motor units (5). Potentiation may be a fundamental mechanism through which force can be maintained despite reductions in the motor unit discharge rate, apart from the muscular wisdom hypothesis.
Dynamic twitch contractions exhibited increases in work and mean power ranging from 25 to 50% after 5 Hz, 20 s conditioning stimuli that were presented during various phases of muscle shortening in the mouse extensor digitorum longus muscle (7). Conversely, the isometric twitch force was potentiated only 14%. If the decrease in firing rate were offsetting the extra force produced by potentiation, then, during concentric conditions, it could be predicted that the motor unit discharge rate could decline to a greater extent than an isometric contraction at a comparable load and still maintain force. This was not found in dynamic fatiguing arm movements in which few motor units in triceps brachii muscle exhibited a decline in discharge rate.
Experimental evidence in the human quadriceps muscle that was fatigued by either electrical stimulation or by voluntary contraction revealed a shift of the force-frequency curve to the right, reflecting low-frequency fatigue (Fig. 4, dashed line). That is, higher frequencies of stimulation (not lower, as predicted by the muscular wisdom hypothesis) were required to produce a 50% MVC after fatigue than before fatigue (4). Additional interactions between potentiation and muscle length, fiber type, sensory feedback from muscle spindles, and temperature to name a few, complicate the relationship between motor unit discharge and contractile properties as fatigue progresses. Consequently, generalizations relating to force optimization during fatigue can become somewhat tenuous. Clearly, more work is required in considering these factors when characterizing motor unit discharge behavior during fatigue.
The muscular wisdom hypothesis motivated almost 20 years of experimentation with an attempt to integrate muscle activation with force output. Although it holds up well during MVCs in certain muscle groups, its applicability is likely task and muscle dependent and may be limited during prolonged submaximal isometric or dynamic contractions.
We thank Dr. Tanya Ivanova for her assistance in the preparation of this manuscript.
This work was supported by NSERC Canada through research grants awarded to Dr. Jayne Garland and postgraduate scholarships awarded to Rod Gossen.
1. Bigland-Ritchie B.R., Johansson, R. Lippold, O.C.J.and Woods. J.J. Contractile speed and EMG changes during fatigue of sustained maximal voluntary contractions. J. Neurophysiol. 50: 313–324, 1983.
2. Bigland-Ritchie, B.R., Furbush, F.H. Gandevia, S.C.and Thomas. C.K. Voluntary discharge frequencies of human
motoneurons at different muscle lengths. Muscle Nerve. 15: 130–137, 1992.
3. Binder-MacLeod, S.A., Lee, S.C.K.and Baadte. S.A. Reduction of the fatigue-induced force decline in human
skeletal muscle by optimized stimulation trains. Arch. Phys. Med. Rehabil. 78: 1129–1137, 1997.
4. Binder-MacLeod, S.A., and McDermond. L.R. Changes in the force-frequency relationship of the human
quadriceps femoris muscle following electrically and voluntarily induced fatigue. Phys. Ther. 72: 95–104, 1992.
5. DeLuca, C.J., Foley, P.J.and Erim. Z. Motor unit
control properties in constant-force isometric contractions. J. Neurophysiol. 76 (3): 1503–1516, 1996.
6. Enoka, R.M., and Stuart. D.G. Neurobiology of muscle fatigue
. J. Appl. Physiol. 72: 1631–1648, 1992.
7. Grange, R.W., Vandenboom, R. Xeni, J.and Houston. M.E. Potentiation of in vitro concentric work in mouse fast muscle. J. Appl. Physiol. 84 (1): 236–243, 1998.
8. Garland, S.J., Enoka, R.M. Serrano, L.P.and Robinson. G.A. Behavior of motor units in human
biceps brachii during a submaximal fatiguing contraction. J. Appl. Physiol. 76: 2411–2419, 1994.
9. Garland, S.J., Griffin, L.and Ivanova. T. Motor unit
discharge rate is not associated with muscle relaxation time in sustained submaximal contractions in humans. Neurosci. Lett. 239: 25–28, 1997.
10. Griffin, L., Garland, S.J.and Ivanova. T. Discharge patterns in human
motor units during fatiguing arm movements. J. Appl. Physiol. 85: 1684–1692, 1998.
11. Griffin, L., Garland, S.J. Ivanova, T.and Gossen. R. Muscle vibration sustains motor unit
firing rate during submaximal isometric fatigue in humans. J. Physiol. (Lond). 535: 929–936, 2001.
12. Jones, D.A., Bigland-Ritchie, B.and Edwards. R.H. Excitation frequency and muscle fatigue
: mechanical responses during voluntary and stimulated contractions. Exp. Neurol. 64: 401–413, 1979.
13. Macefield, V.G., Fuglevand, A.J. Howell, J.N.and Bigland-Ritchie. B. Discharge behaviour of single motor units during maximal voluntary contractions of a human
toe extensor. J. Physiol. 528 pt 1: 227–234, 2000.
14. Marsden, C.D., Meadows, J.C.and Merton. P.A. “Muscular wisdom” that minimized fatigue during prolonged effort in man: peak rates of motoneuron
discharge and slowing of discharge during fatigue. Adv. Neurol. 39: 169–211, 1983.
15. Miller, K.J., Garland, S.J. Ivanova, T.and Ohtsuki. T. Motor unit
behaviour in humans during fatiguing arm movements. J. Neurophysiol. 75: 1629–1636, 1996.