Muscle synergies during shifts of the center of pressure by standing persons: identification of muscle modes

Abstract

When a standing person performs a movement such that the center of gravity shifts, the activity of postural muscles adjusts to keep the balance. We assume that such adjustments are controlled using a small set of central variables, while each variable induces changes in the activity of a subgroup of postural muscles. The purpose of this study has been to identify such muscle groups (muscle modes or M-modes) and compare them across tasks and subjects. Four tasks required the subjects to release a load from extended arms leading to a center of pressure (COP) shift prior to the load release. The fifth task required an explicit COP shift by voluntary sway. Electromyographic activity of 11 postural muscles on one side of the body was integrated over a 100-ms interval corresponding to the early stage of the COP shift, and this integrated EMG activity was subjected to a principal component (PC) analysis across multiple repetitions of each task. Three PCs were identified and associated with a "push-back M-mode," a "push-forward M-mode," and a "mixed M-mode." Cluster analysis of the PC vectors across tasks and across subjects confirmed the existence of distinctive push-forward and push-back muscle groups. PC vectors were also compared across tasks and across subjects using cosines as a measure of colinearity between pairs of vectors. In general, M-modes were similar across both tasks and subjects. We conclude that shifts of the COP, whether implicit or explicit, are controlled using a small set of central variables associated with changes in the activity of robust subsets of postural muscles. These results can be used for future analysis of muscle synergies associated with postural tasks.

Learning effects on muscle modes and multi-mode postural synergies

“…The ΔIEMG N indices for each subject were subjected to PCA. Consistent with previous studies (Krishnamoorthy et al, 2003a, Wang et al 2005, across all subjects for three test sessions, we found that the first three PCs (PC 1 , PC 2 and PC 3 ) had at least one muscle index significantly loaded. Principal components from PC 4 onward explained little variance in the ΔIEMG N space and were poorly reproducible across subjects.…”

“…The voluntary sway task was impossible to be performed while standing on the unstable board. Hence, we used a load release (LR) task, as in a number of earlier studies (Krishnamoorthy et al 2003a(Krishnamoorthy et al , b, 2004. Note that the composition of the modes was similar across all the cited studies and methods of their identification.…”

“…The first three PCs for the VS and LR tasks were selected for further analysis. This was based on two criteria, examination of the scree plots (the following PCs accounted for approximately similar amounts of variance per PC) and having at least one muscle significantly loaded per PC (Krishnamoorthy et al 2003a, Wang et al 2005. We are going to address these PCs as muscle modes (M-modes) and assume that magnitudes of (coefficients at) the M-modes are manipulated by the controller to produce COP AP shifts.…”

“…Recently, the notion of muscle synergies has been operationally defined within the framework of the uncontrolled manifold (UCM) hypothesis (Scholz and Schöner 1999;Latash et al 2002;Krishnamoorthy et al 2003a). Within this hypothesis, a neural controller acts in a space of elemental variables and organizes co-variation among those variables to stabilize a value or a time profile of an important performance variable.…”

“…In other words, the controller is assumed to be able to vary elemental variables one at a time. This is not true with respect to activation levels of individual muscles that frequently show patterns of co-variation across wide variations of tasks and conditions (Krishnamoorthy et al 2003a;Danna-Dos-Santos et al 2007). In several earlier studies of postural tasks, elemental variables have been associated with muscle groups within which muscle activation levels scale in parallel with variations of task characteristics; these have been called muscle modes or M-modes (Krishnamoorthy et al 2003a,b;Wang et al 2005Wang et al , 2006Danna-DosSantos et al 2007).…”