Estimation of the visual contribution to standing balance using virtual reality

Le résumé fourni par la source

Abstract Sensory perturbations are a valuable tool to assess sensory integration mechanisms underlying balance. Implemented as systems-identification approaches, they can be used to quantitatively assess balance deficits and separate underlying causes. However, the experiments require controlled perturbations and sophisticated modeling and optimization techniques. Here we propose and validate a virtual reality implementation of moving visual scene experiments together with model-based interpretations of the results. The approach simplifies the experimental implementation and offers a platform to implement standardized analysis routines. Sway of fourteen healthy young subjects wearing a virtual reality head-mounted display was measured. Subjects viewed a virtual room or a screen inside the room, which were both moved during a series of sinusoidal or pseudo-random room or screen tilt sequences recorded on two days. In a between-subject comparison of 10x6 min long pseudo-random sequences, each applied at 5 amplitudes, our results showed no difference to a real-world moving screen experiment from the literature. We used the independent-channel model to interpret our data, which provides a direct estimate of the visual contribution to balance, together with parameters characterizing the dynamics of the feedback system. Reliability estimates of single subject parameters from six repetitions of a 6x20-s pseudo-random sequence showed strong to excellent test-retest agreement of all estimated parameters within the three repetitions of the second day and within the last of the three measurements between the days (Intraclass correlation; ICC >0.9). ICCs including the first two stimulus repetitions were consistently smaller for all parameters but the visual contribution and the feedback time delay. Sway responses strongly depended on the visual scene, where the high-contrast, abstract screen evoked larger sway as compared to the photo-realistic room. In conclusion, our proposed virtual reality approach allows researchers to reliably assess balance control dynamics including the visual contribution to balance with minimal implementation effort.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.