Mechanical oscillations contributing to the segmentation of the reflex electromyogram response to stretching human muscles.

1. Brisk joint displacements and tendon taps were applied to voluntarily contracting wrist flexor muscles in subjects who did not attempt to react to them. Different types of mechanotransducers, some of them attached to an intramuscular needle, were used to detect mechanical oscillations in the wrist flexors, resulting from the imposed impacts. The transducer responses to the perturbations were compared with simultaneously recorded reflex electromyogram (e.m.g.) responses. Experiments were also carried out on a rubber band model, exposed to similar mechanical stimuli.2. During brisk ramp wrist extensions the transducers signalled damped muscular oscillations at 30-50 Hz. The oscillations grew in amplitude with increasing speed of onset of the stretch movement and at angular accelerations exceeding about 2 x 10(4) deg/sec(2) the e.m.g. response changed from a non-segmented to a progressively more pronounced segmented pattern. Peak intervals in the segmented reflex e.m.g. responses were similar to those of the mechanical oscillations and did not change significantly with small or moderate variations in background contraction force. Latencies from successive deflexions in the accelerometer records to corresponding deflexions in the e.m.g. were 20-25 msec.3. Damped muscular oscillations in the 30-50 Hz range were also initiated by sudden halts of voluntary wrist movements, by electrically induced twitches, and by voluntary brisk contractions. In these instances too, the mechanical oscillations were reflected in the shape of the succeeding e.m.g. response.4. The interval between the two initial peaks in the accelerometer records was always shorter with tendon taps than in ramp stretch experiments. A corresponding difference was noted in the intervals between the following two peaks in the reflex e.m.g. response.5. The initial peak in the accelerometer records could be ascribed to a wave propagated at about 40 m/sec in the wrist flexor muscles. Inconclusive results were obtained in attempts to determine whether the subsequent oscillations represented reflexions of the propagated wave at the ends of the muscle.6. The muscles were also exposed to recurrent stretch stimuli (torque pulses) with a repetition rate varying between 15 and 100 Hz. At 30-50 Hz the intramuscular oscillations reached their maximal amplitude, and such repetition rates were also most efficient in producing synchronized e.m.g. bursts, time-locked to the oscillations. The mechanical responses of the wrist flexors to single or recurrent perturbations were to a large extent mimicked by a rubber band model with a longitudinal resonance frequency of about 40 Hz.7. It is concluded that segmentation of reflex e.m.g. responses to sudden joint displacements and other types of brisk muscle perturbations to a large extent depends on the inherent resonance characteristics of musculo-tendinous structures. Primary spindle endings with their high vibration sensitivity and their segmental projections to alpha-motoneurones are believed to be the receptors primarily responsible for reflex entrainment of the motor impulses.