Long latency reflexes to imposed displacements of the human wrist: dependence on duration of movement.

Reflex EMG responses to angular displacements of the wrist joint were recorded from 12 normal human volunteers. A mechanical stop was used to suddenly arrest displacements at varying times following the onset of the stimulus. With unrestricted movement of the handle, the EMG response consisted of an early component (M1) with a latency of 30-35 ms and a long-latency component (M2-3) beginning 55-65 ms after the onset of the displacement. When the displacements were arrested prior to a critical time occurring between 40 and 50 ms after the onset (mean of 44 ms), the M2-3 component of the response was not present. Increasing the duration of the displacement beyond this time resulted in a rapid increase in the size of M2-3. Facilitation provided by volitional intent to oppose the perturbation was not sufficient to generate an M2-3 response to either a brief, low velocity displacement produced by the torque motor or to a phasic, high velocity stretch produced by a tendon tap. The timing relationships between the onset latency of M2 and the minimum duration of displacement required to generate an M2-3 response are not easily reconciled with the notion that the segmentation of the EMG responses into components is mediated by repeated activation of the same central reflex pathway by phasic afferent bursts. Two mechanisms that could account for these results are either inhibition in response to the sudden stop of phasically-active "linking" interneurons which are part of the long latency pathway, or the loss of an essential convergent facilitatory input which serves to monitoring the continuation of the movement.