Proceedings of The Physiological Society

Physiology 2012 (Edinburgh) (2012) Proc Physiol Soc 27, PC322

Poster Communications

Voluntary vs electrical muscle activation reveals the resonant nature of physiological finger tremor

C. A. Vernooij1, M. Lakie1, R. F. Reynolds1

1. School of Sport and Exercise Sciences, University of Birmingham, Birmingham, United Kingdom.


  • Fig 1. Frequency spectra of tremor acceleration in isotonic conditions (left panels) and tremor force in isometric conditions (right panels). Lighter grey colours represent higher levels of muscular activation. Shaded areas represent standard error.

Human physiological finger tremor is usually described as having two main frequencies, one at ~20 Hz and one at ~10 Hz. The origin of these frequencies is uncertain. Two main views exist: one attributing tremor to neural oscillations and one attributing tremor to mechanical resonance (McAuley and Marsden, 2000). We recently showed the presence of a resonant component in hand tremor that could be reproduced by a very simple model driven by a white noise input (Lakie et al, 2012). This suggests tremor would be similar if the muscle is activated voluntarily by muscle activation or artificially by white noise input (i.e. direct electrical muscle stimulation). In this study, with ethical permission and the subjects' informed consent we examined tremor of the splinted middle finger with voluntary activation or artificial white noise stimulation at five different mean levels. With voluntary activation, surface EMG of the m. extensor digitorum communis was recorded. With electrical activation, white noise sequences of 50 µs current pulses were applied to the skin over this muscle. During both methods of activation, we recorded tremor in two ways. In isometric conditions the finger pressed upwards against a rigid strain gauge device above the nail plate. In isotonic conditions a miniature accelerometer was attached to the finger. The frequency spectra of the isometric tremor showed an exponential decline of force with higher frequency (Fig 1). The effect of different levels of muscular activation changed the amount of force produced, but did not change the shape of the spectrum. However, spectra of the isotonic tremor showed a specific peak frequency, and both frequency and amplitude were dependent on level of activation (Fig 1). For low levels of activation, a peak at ~20 Hz was produced. Increased levels of activation not only increased the acceleration amplitude, but also decreased the peak frequency to ~10 Hz. Importantly, isotonic and isometric tremor spectra show similar profiles with voluntary activation and white noise electrical activation. This implies that the voluntary activation and the artificial random input are equivalent - that is, there is nothing "special" about central drive. We suggest that the peak in the isotonic tremor spectrum represents a resonance of the limb. The resonant frequency decreases with increased activation due to a movement dependent reduction in muscle stiffness. We have recently shown that a drop in muscular stiffness with increased movement leads to a lower tremor frequency in the hand (Reynolds and Lakie, 2010) and we propose the same mechanism to hold for finger tremor. The results show that tremor frequencies of the finger ranging from 10 - 20 Hz can be produced by mechanical resonance and it is not necessary to invoke other causes.

Where applicable, experiments conform with Society ethical requirements