Point:Counterpoint: The interpolated twitch does/does not provide a valid measure of the voluntary activation of muscle POINT: THE INTERPOLATED TWITCH DOES PROVIDE A VALID MEASURE OF THE VOLUNTARY ACTIVATION OF MUSCLE

Voluntary activation. A muscle is activated voluntarily when force is produced by the recruitment of motoneurons through increased descending drive from the motor cortex. This activation is usually deliberate and is accompanied by a sense of effort. There are many other influences on motoneurons at rest and during voluntary contractions that may alter the response of motoneurons to descending drive. Such influences include excitatory and inhibitory sensory feedback, as well as alterations in motoneuron properties that may make them more or less responsive to synaptic input (21). However, descending drive from the motor cortex is the major determinant of the timing and strength of voluntary contractions. Qualitatively, voluntary activation of a muscle is the recruitment of its motor units by increased descending drive. As a quantity, voluntary activation has been defined as the level of voluntary drive during an effort (12). What twitch interpolation measures is the drive by the motoneurons to the muscle and how this translates into force. Thus measured voluntary activation is the proportion of maximal possible muscle force that is produced during a voluntary contraction. It does not quantify the descending drive reaching the motoneurons or whether motoneuron firing rates are maximal and does not take into account the source of drive to the motoneurons. Principle of twitch interpolation. During increasing strengths of voluntary muscle contraction, motor units are recruited progressively and increase their firing rates. The principle of twitch interpolation is to stimulate the peripheral nerve during a voluntary contraction to add, in most motor axons, an additional action potential to those produced voluntarily (7, 19). If any individual motor unit is not firing fast enough to produce its maximal force then this additional action potential will evoke an increment in force from the muscle fibers of the motor unit. The sum of increments from different motor units produces a twitchlike increment in force from the whole muscle. The more motor units recruited in a contraction and the faster they are firing, the smaller the superimposed twitch. During a contraction that produces maximal muscle force, all motor units in the muscle should be recruited and all firing fast enough to produce fused contractions of their muscle fibers. The principle of twitch interpolation is supported by an inverse relationship between the amplitude of the twitch evoked by an interpolated stimulus and voluntary force output at the moment of stimulation. Such a relationship has been reported for adductor pollicis, biceps brachii, quadriceps, ankle plantarflexors, tibialis anterior, masseter, trapezius, and the diaphragm (e.g., 1, 6, 7, 10, 13, 18, 19, 23). Thus twitch interpolation does give a measure of voluntary activation in many muscle groups. However, easy quantification of voluntary activaton, using the formula, Voluntary activation 100 [1 (superimposed twitch/potentiated resting twitch)], relies on near linearity of the inverse relationship, so that calculated voluntary activation equals the percentage of maximal force that the subject is generating with the tested muscle(s). Model of twitch interpolation. A computer model of a hand muscle with a pool of motor units with realistic motor unit numbers, recruitment, firing rates, and twitch and tetanus properties was developed (15). In the model, superimposed twitches were generated by twitch interpolation, which consisted of addition of an extra action potential to each motor axon that was not refractory at the moment of stimulation. As peripheral nerve stimulation produces antidromic as well as orthodromic potentials, collision with subsequent “voluntary” potentials was incorporated. Although the relationship between input to the motoneurons and force output was sigmoidal so that large increases in “excitation” to the motoneurons were needed to produce small falls in twitch amplitude in nearmaximal contractions, the model predicted a linear inverse relationship between the superimposed twitch and contraction strength. Thus modeling supports the proposal that twitch interpolation can measure the proportion of muscle force recruited voluntarily, i.e., “voluntary activation.” Practical twitch interpolation. In contrast to the model, most published graphs of the amplitude of the superimposed twitch against the force of voluntary contraction show nonlinearities. Belanger and McComas (7) described the curve as S-shaped with a convex up portion at low contraction strengths and a concave up portion at high contraction strengths. Often curves fitted to the relationship approach the x-axis at a very shallow