A range of different stretch reflex responses in the human thumb.

1. Imposed sinusoids were used to assess the resistance to movement at the thumb interphalangeal joint.2. The resistance to high-frequency movements (> 12 Hz) increased when the subject exerted a large voluntary flexing force; this increase was attributable to a greater non-reflex resistance of the contracting flexor muscles. This resistance was essentially ;visco-elastic', and the force was phase-advanced on joint position. At moderately large forces (up to half maximal), however, the resistance changed with changing frequency, and over a range 4-12 Hz the vectors which represented joint stiffness described the wide path that is characteristic of an active stretch reflex (Brown, Rack & Ross, 1982a). At frequencies between about 4 and 6 Hz the force was sometimes phase-delayed on position, and the joint exhibited a negative viscous stiffness. When the voluntary flexing force was very large the reflex contributed less to the resisting force, which was then phase-advanced on position at all frequencies of movement.3. Large amplitude movements did not generate correspondingly large reflex responses; as the amplitude of movement was increased, the reflex component of the resisting force became relatively smaller and the total resisting force was then phase-advanced on joint position at all frequencies.4. The reflex component of the resisting force (as indicated by the excursion of the joint stiffness vectors) varied from subject to subject and from time to time; the reflex usually became more active late in an experiment when the subject had exerted flexing forces against the imposed movement for some minutes. Extreme fatigue, however, diminished the amount of reflex force.5. In some subjects the joint-stiffness records indicated a particularly vigorous reflex response at 8-11 Hz, in contrast to a rather feeble response at 6 or 7 Hz. It is suggested that the reflex pathways then had a relatively low impedance to afferent signals that were modulated at 8-11 Hz, related perhaps to the firing patterns of the most recently recruited motoneurones.6. Under the conditions of these experiments, it appears that the stretch reflex has too small a gain to function as a very effective error-controlled position servo-mechanism.