Motor pool organization in monosynaptic reflexes: responses in three different muscles

Recruitment order of motoneurons was measured as a function of their conduction velocities in the presence of monosynaptic reflexes evoked by dorsal root stimulation. Motoneurons were studied in three cat hindlimb muscles: medial gastrocnemius (MG), plantaris (Pl), and tibialis anterior (TA). A relationship between recruitment order and unit conduction velocity (CV) was clearly seen in all three muscles. The correlation between these two variables was lower than that found in previous studies. A CV-dependent recruitment order was most clearly sen in tibialis anterior motoneuron pool; the relationship was poorest in plantaris. Recruitment order of MG motoneurons was measured and related to their conduction velocities in response to monosynaptic reflexes evoked by L7 + S1 dorsal root stimulation. Recruitment order was then retested in the presence of rostral root stimulation (largely heteronymous) alone and again in the presence of caudal root stimulation (largely homonymous) alone. Changing the composition of the afferent input changed the critical firing levels (rank order) of some motoneurons by as much as 40% and that of others, not at all. Some motoneurons became harder to recruit and others easier; changes in recruitment order were not related to conduction velocities of the units. 1a-afferent inputs are not uniformly distributed to all the motoneurons of a pool. It appears that this nonuniformity is a determining factor in establishing a recruitment order. TA, which receives the most uniformly distributed monosynaptic input, also has the most nearly size-dependent recruitment order. Fractionation of input ca induce additional nonuniformity, and results in recruitment-order changes in some motor units that are independent of their conduction velocity. It is concluded that nonuniformity of afferent inputs, whether present or induced by experiment, can produce large recruitment-order changes among individual motoneurons in a pool and that these individual motoneurons need not share a common property such as conduction velocity or recruitment threshold in response to a control input. Therefore, arguments based on reversals in recruitment order of pairs of motor units or even changes in rank order of individual motor units do not present sufficient evidence for the presence of input specifically directed to motor units sharing a particular property.