Non-visually-guided distance perception depends on matching torso fluctuations between training and test

Abstract

Blindwalking to replicate an instructed distance requires various sensory signals. Recent evidence in movement science across many organisms suggests that multifractal organization of connective tissue supports the use of these signals. Multifractal structure is a multiplicity of power laws defining distribution of proportion across many time scales that helps predict judgments of the objects’ length. Present work tests whether the multifractal structure in postural accelerometry during blindwalking predicts blindwalking distance replications. Ten undergraduate student participants each completed 20 trials of distance-perception each comprising two laps. On each Lap 1, experimenters led participants to walk on any of five prescribed distances, randomly assigning half to walk Lap 1 with eyes open and another half to walked Lap 1 with eyes closed. On Lap 2, all participants walked with eyes closed to replicate instructed distances from Lap 1. We collected postural accelerometry from the torso during each lap. Regression modeling showed that multifractality of postural accelerometry on both Lap 1 and Lap 2 contributed significantly to Lap-2 blindwalking responses. According to this model, more accurate Lap-2 replications of Lap-1 distance came from eyes-closed participants whose posture had comparable multifractality on both laps. Multifractality provides insights into the sequence of exploratory behaviors for blindwalking responses to distance perception.