Postural control during upper body locomotor-like movements: similar synergies based on dissimilar muscle modes

Abstract

We studied the organization of leg and trunk muscles into groups (M-modes) and co-variation of M-mode involvement (M-mode synergies) during whole-body tasks associated with large variations of the moment of force about the vertical body axis. Our major questions were: (1) can muscle activation patterns during such tasks be described with a few M-modes common across tasks and subjects? (2) do these modes form the basis for synergies stabilizing M _z time pattern? (3) will this organization differ between an explicit body-rotation task and a task associated with locomotor-like alternating arm movements? Healthy subjects stood barefoot on the force platform and performed two motor tasks while paced by the metronome at 0.7, 1.0, and 1.4 Hz: cyclic rotation of the upper body about the vertical body axis (body-rotation task), and alternating rhythmic arm movements imitating those during running or quick walking (arm-movement task). Principal component analysis was used to identify three M-modes within the space of integrated indices of muscle activity. The M-mode vectors showed clustering neither across subjects nor across frequencies. Variance in the M-mode space across sway cycles was partitioned into two components, one that did not affect the average value of M _z shift (“good variance”) and the other that did. An index was computed reflecting the relative amount of the “good variance”; positive values of this index have been interpreted as reflecting a multi-M-mode synergy stabilizing the M _z trajectory. On average, the index was positive for both tasks and across all frequencies studied. However, the magnitude of the index was smaller for the intermediate frequency (1 Hz). The results show that the organization of muscles into groups during relatively complex whole-body tasks can differ significantly across both task variations and subjects. Nevertheless, the central nervous system seems to be able to build M _z stabilizing synergies based on different sets of M-modes, within the approach accepted in this study. The drop in the synergy index at the frequency of 1 Hz, which was close to the preferred movement frequency, may be interpreted as corroborating the neural origin of the M-mode co-variation.