Variability in a dynamic postural task attests ample flexibility in balance control mechanisms.

When humans stand upright on a platform that sinusoidally translates in the anterior-posterior direction, the movements of upper and lower body segments are appropriately coordinated, in order to keep the body within its limits of stability. A significant fluctuation in this behaviour is evident across subjects and perturbation conditions. The inter- and intrasubject variability in the body segment kinematics, as occurs during repeated trials across different conditions, is quantitatively described here. Twenty normal subjects stood upright with eyes open (EO) or eyes closed (EC) on a platform moving to-and-fro in the horizontal plane for 30 s, at a frequency of 0.2 and 0.6 Hz, with a peak-to-peak amplitude of 6 cm. Each subject made two trial repetitions for each visual and frequency condition. The last 20 s of each trial was acquired. The displacement of markers fixed on the lateral malleolus, hip and head was sampled at a frequency of 50 Hz. An index of the 'average' displacement of each marker during the trial was the standard deviation (SD) of its anterior-posterior displacements, calculated across the acquired trial cycles. The cross-correlation (CC) between pairs of marker displacement traces gave an indication of the degree of coupling of the body segments. All subjects showed two basic modes of coping with the perturbation, depending on the availability of the visual input: with EO, they tended to stabilize the head in space; with EC, the head oscillated in the anterior-posterior direction more than hip and platform. Within this general behaviour, the values of the SD of horizontal displacement of head and hip marker traces varied within an ample range during different trials of the same perturbation condition. Even within a single trial there was an ample variability of the body segments' position. In spite of this, neither head nor hip ever bypassed the anterior or posterior limits of stability. Remarkably, the range of variability of the whole population of normal subjects, both across and within trials, was analogous to the range of variability of single subjects across numerous trials. This large variability notwithstanding, it appeared that the relationship of head to hip SD across trials was almost constant, independent of visual and perturbation frequency condition. The results show that there exist a large variety of dynamic postures, rather than one particular configuration, which assure stability. The findings also suggest a neural or biomechanical constraint underlying the operations of the equilibrium control strategy.