Anticipation and fusimotor activity in preparation for a voluntary contraction

1. Multi-unit and single-unit recordings were made of muscle spindle afferent activity from the pretibial muscles of human subjects who were initially relaxed. The muscles were subjected to a stretching perturbation of 1 s duration, occurring irregularly, on average once every 5 s. In test sequences, an auditory or visual warning was provided 1 . 06 s before some of the perturbations. Subjects were required to oppose every perturbation by contracting the receptor-bearing muscle as rapidly as possible. 2. Following the warning all subjects sometimes tensed the receptor-bearing muscle unintentionally in preparation for the perturbation. In these contractions, the discharge of a spindle ending accelerated only if the contraction strength exceeded the ending's threshold for activation, established in control voluntary contractions performed under isometric conditions. 3. When the receptor-bearing muscle did not contract in the interval between warning and perturbation, there was no detectable change in the multi-unit recordings of spindle activity or in recordings from twelve of thirteen single spindle afferents. The thirteenth spindle afferent discharged prior to the perturbation in the absence of detectable e.m.g. in response to (only) three of twenty-three warning stimuli. However, this ending had been so responsive during isometric voluntary contractions that a contraction level at which it did not respond could not be established, and it is suggested that the findings with this ending resulted from its low threshold rather than from selective activation of the fusimotor system. 4. When subjects were warned of the perturbations, the dynamic response of spindle endings to the perturbations was not increased in size or altered in latency. 5. The motor response to perturbations without warning generally contained only long-latency (volitional) e.m.g. activity occurring 107--200 ms after the onset of the perturbation. When a warning was given, short-latency (reflex) e.m.g. activity was also recorded, beginning 46--76 ms after the onset of the perturbation. 6. It is concluded that anticipation of the need to contract a muscle does not result in selective activation of fusimotor neurones in preparation for the contraction. The change in stretch reflex gain that occurs as a result of 'anticipation' occurs through a central process which does not involve the fusimotor system.