Abstract
An investigation of the spectral characteristics of physiological tremor and cold-induced shivering tremor in the limbs of healthy subjects was conducted. Recordings of posture-maintaining index finger and hand tremor motion were made using miniature accelerometers, together with emg activity in associated superficial extensor and flexor muscles. The effects on physiological finger and hand tremor spectra of systematically applied external inertial and force loads were studied. Comparison was made of bilateral index finger tremor spectra in three characteristic frequency bands. Estimates were also made of the variation of hand tremor frequency with time. The results showed that the dominant responses to modified limb inertia were representative of those of a second order mass-spring mechanism. In contrast, tremor frequency changes consequent upon externally imposed changes in muscle stiffness were not identifiable with this simple mechanism, suggestive of the existence of muscle tone-mediated action of a reflex loop. Measurement of bilateral finger tremor showed no evidence for a common tremor site at the same segmental level. Significant, random changes with time of postural hand tremor frequency were observed. The variation of frequency with time was smaller at higher levels of externally imposed muscle tension. Using a variety of methods of cooling, shivering was induced in 6 subjects and a descriptive analysis was made of hand and upper arm tremor, both in one arm and bilaterally. An experiment was conducted to detect shivering responses to external localised cooling of the spine. The results showed that synchronous co-contraction of limb flexor/extensor muscles was not universally present during shivering, that shivering was invariably accompanied by an increase in the mean level of muscle activity and that steady preshivering muscle tone was often absent between episodes of shivering. Comparison of bilateral upper arm tremor signals showed that shivering was frequently evident at different frequencies without coherence. No shivering responses were detected when the spine was locally cooled to a temperature comparable with that when shivering was evoked by whole-body cooling. It is concluded that the rhythm of both types of limb muscle tremor is predominantly the result of the excitation and regenerative oscillation of a resonant structure under the influence of the highly adaptive spinal reflex arc. It is also recognised that the observation of wide variability between the subjects' responses may signal the action of additional factors on tremor characteristics.