The body configuration at step contact critically determines the successfulness of balance recovery in response to large backward perturbations.

The mechanical efficiency of stepping to recover balance can be expressed by a biomechanical model that includes the trunk inclination angle and the angle of the leg at the instant of stepping-foot contact. The aim of the present study was to test the hypothesis that this model would accurately predict the successfulness of recovery attempts (recovery vs. falls) following large backward perturbations. Ten young participants were exposed to a series of 12 very large postural perturbations in the backward direction by means of a support-surface translation. At the instant of stepping-foot contact, we calculated the trunk inclination angle and the angle of the stepping leg with the vertical. Reaction time, step duration, step velocity and step length were also determined. A logistic regression analysis revealed that the model with leg and trunk inclination angles accurately predicted successful recovery, with a more forward tilted trunk and a further backward positioned leg increasing the probability of success. The set of spatiotemporal step variables was significantly less predictive. Over the course of the experiment, participants gradually became more successful in recovering balance, which coincided with an increase in leg but not in trunk angles. In conclusion, the body configuration at the instant of first stepping-foot contact accurately predicted successful balance recovery after a backward postural perturbation. Given the observation that participants improved their performance by increasing their leg angles, which suggests that it may be easier to improve this variable, compared to the trunk angle, by exercise interventions.