An Alternative Perspective on Postural Stability and Variability

Abstract

In this issue of Exercise and Sport Sciences Reviews, the article “Task-Dependent Postural Control Throughout the Lifespan” by Haddad et al. (2) promotes two major ideas: 1) Postural control is highly task dependent and 2) Postural variability is functional — it reflects the availability of multiple solutions (resulting from kinematic and neuromuscular redundancy) and may play an exploratory role by generating sensory information. These ideas are central to an ecological-dynamical perspective on postural control. Haddad et al. (2) make important contributions to that perspective by summarizing recent findings on postural control across the lifespan and by emphasizing the relevance of this approach for clinical and rehabilitation applications. Postural stability typically is described in terms of minimizing spontaneous postural sway and counteracting perturbations that displace the center of mass away from a single reference point centered within the base of support. The alternative proposed by Haddad et al. (2) is that successful postural control amounts to achieving body configurations that fall somewhere within a stable subspace of the postural configuration space (depicted graphically in Configuration Space Diagrams). Variability within the stable region of the configuration space is acceptable and will not lead to falls. The boundary of the stable region is the reference for upright stance; there is not just one body configuration that is most desirable or stable. The boundaries are dynamic; they change on a moment-to-moment basis as the actor moves, the environment changes, or task demands arise and dissolve. What is “stable” reflects suprapostural task goals and the behavioral (including social) and environmental contexts. Configuration space diagrams potentially are useful, but there is a need to develop the idea beyond its current qualitative descriptive form. The uncontrolled manifold (4), goal-equivalent manifold (1), and the Tolerance-Noise-Covariation (3) methods may be well suited for that purpose. These methods differ in important ways yet share in common the idea that motor performance can be understood by describing how relatively microscopic motor system degrees of freedom (DF) map onto relatively macroscopic task variables. In redundant systems, many combinations of the microscale DF preserve a value of the task variable. This is equivalent to saying that there is a stable subregion (or manifold) of the configuration space, whose dimensions are the microscale DF, within which the task variable remains constant. Variations within that manifold can remain unchecked without leading to task failure. For quiet stance, joint angles can serve as the microscale DF and center of mass position can serve as the task variable, but expanding this idea to incorporate the approach of Haddad et al. (2) requires defining the task variable in terms of the functional goals of the suprapostural task. Postural variability is informative for postural control; Postural sway generates information (visual, somatosensory, vestibular, etc.) that is specific to the changing position of the body and thus can be useful for controlling posture. A challenge that remains for researchers is to determine whether (and when) postural sway is explicitly exploratory (i.e., it is an intentional information-seeking act) rather than performatory (in the service of a behavioral goal) or unintended noise. At some level, this may not matter; regardless of intent, postural sway may provide information useful for postural control. But the term “exploratory” connotes an intention, and, if this is indeed the case, then efforts to model postural control will benefit from fleshing out this idea more definitely. Lastly, the authors raise important implications for rehabilitation. Models that account for properties of the actor, task, and environment may be more useful in clinical settings than models based on quiet stance. Quiet stance may lack the sensitivity and specificity desirable in a diagnostic tool for clinical applications because the absence of an explicit suprapostural task leaves postural control underconstrained. Potentially fruitful directions for clinical research may be to identify the limits to which an actor’s postural control system is appropriately adaptive so as to support a variety of suprapostural task goals and to track the evolution of the stable region during the course of rehabilitation.