The relationship between surface potentials and the number of active motor units

One of the assumptions inherent in a technique recently devised for enumerating motor units in human muscles is that the surface potentials from active motor units summate in a linear fashion. We present an electrical model of a muscle which predicts that a linear relationship between the number of active units and the electrical response recorded at the surface overlying the muscle would not be expected. The extent of the non-linearity, and hence the error in the calculation of the number of motor units in a given muscle, depends upon the ratio between the mean conductance of the motor units themselves and that of the external conduction pathway through which the electrical signal is fed (Gu/Ge). The extent of non-linearity is assessed experimentally in human hypothenar muscles using a "collision" technique. The average underestimate introduced into the calculation of the number of motor units in this particular case was concluded to be 26%. The value of Gu/Ge derived from these experiments, in 2 subjects, was checked by simulating an intramuscular action potential and determining the attenuation at the surface. The 2 independently obtained values were sufficiently close to suggest that the model may be a valid one. We conclude that caution should be employed in the interpretation of experiments which purport to determine the number of motor units in a muscle by means of surface recordings.