Motor control enhancements by sub-threshold mechanical noise applied to foot soles during quiet standing

Abstract

Intervention to improve the balance ability of individuals with impaired balance is needed to prevent falls. While sub-threshold mechanical noise applied to foot soles has been shown to improve balance not only for balance-impaired but also healthy individuals, how calf muscle activity is changed to enhance motor control to achieve improvement has not been explored. To address this issue, we study the calf muscle activity of healthy young adults standing on firm and compliant surfaces, with and without noise applied to their feet. The compliant surface experiment simulates balance impairment. Center of pressure (COP) data was used to assess balance changes, surface electromyography (EMG) recorded muscle activity, and COP-EMG correlations measured muscle contribution to postural control. The Wilcoxon signed-rank test was used to compare the data between the control and noise conditions. On both surfaces, the applied noise enhanced motor control efficiency of all three calf muscle groups studied - the tibialis anterior (TA), lateral gastrocnemius lateralis (LG), and medial gastrocnemius (MG). Noise also increased the contribution of the LG muscle group to postural control in the anteroposterior direction. Our finding suggests that, for balance-impaired individuals with weak calf muscles, higher-frequency noise should be used - this will increase motor control efficiency, i.e., increase posture correction frequency with concomitant reduction in calf muscle contractions, which is well-suited to the weak muscles.

Keywords: Center of pressure (COP); Electromyography (EMG); Motor control; Postural control; Sub-threshold mechanical noise.

Fig. 1

Mechanical noise. a Foam insoles with embedded vibrotactile transducers, b Example vibratory signal of the mechanical noise generated by the transducers, and c its power spectrum. Adapted from Khor et al. [23]

Fig. 2

Experimental setup. a Electrode placement for measurement of the EMG signals from the TA, LG, and MG muscle groups of both legs. b The standing posture required of the subjects during balance trials on the firm surface. This posture was repeated for the balance trials on the compliant surface

Fig. 3

EMG parameter calculation. a The processing pipeline used to calculate the RMS and MF values from the raw EMG signal. b Example of a band-passed EMG signal used to calculate the RMS. c The power spectrum of the band-passed signal in (b), used to calculate the MF

Fig. 4

COP-EMG correlation calculation. a The steps used to process the raw COP and EMG data before calculating the correlation. Example AP sway and MG EMG time-series overlaid on the same plot (b) before; and (c) after cubic-spline smoothing. Spearman’s correlation is 0.798 and 0.894 before and after smoothing, respectively

Fig. 5

Box plots of the COP sway parameters on the (a) firm surface, b compliant surface, with and without noise. The control and noise results are significantly different for only AP range on the firm surface, and for all parameters except AP speed on the compliant surface, as indicated by the P-values in Table 1

Fig. 6

Box plots of the EMG parameters (RMS and MF), with and without noise. RMS: a firm, b compliant surface. MF: c firm, d compliant surface. The control and noise results are significantly different for only RMS on both surfaces, as indicated by the P-values in Table 2

Fig. 7

Box plots of the AP-MG and AP-LG correlations on the (a) firm surface, b compliant surface, with and without noise. The control and noise results are significantly different for only AP-LG correlation on both surfaces, as indicated by the P-values in Table 4.c