Influence of Prior Visual Information on Exploratory Movement Direction in Texture Perception

Abstract

When humans explore objects haptically, they seem to use prior as well as sensory information to adapt their exploratory behavior [1]. For texture discrimination, it was shown that participants adapted the direction of their exploratory movement to be orthogonal to the orientation of textures with a defined direction [2]. That is, they adapted the exploratory direction based on the sensory information gathered over the course of an exploration, and this behavior improved their perceptual precision. In the present study we examined if prior visual information that indicates a texture orientation produces a similar adjustment of exploratory movement direction. We expected an increase of orthogonal initial exploration movements with higher qualities of prior information. In each trial, participants explored two grating textures with equal amplitude, only differing in their spatial period. They had to report the stimulus with the higher spatial frequency. Grating stimuli were given in six different orientations relative to the observer. Prior visual information on grating orientation was given in five different qualities: 50% (excellent information), 35%, 25%, 15% and 0% (no information). We analyzed movement directions of the first, middle and last strokes over the textures of each trial. The results show an increase in the amount of initial orthogonal strokes and a decrease in variability of movement directions with higher qualities of prior visual information.

Humans flexibly use visual priors to optimize their haptic exploratory behavior

“…We also asked whether this behavior can be further moderated by the quality of the priors and the present task demands, and which mechanism is likely to enable the behavior. Experiment 1 13 demonstrated that indeed, visual priors indicating a texture’s orientation produce orthogonal movement adjustments at initial contact similar to those that typically occur over the course of an exploration 12 . Additionally, it became evident that the higher the priors’ quality, the more pronounced the movement adjustments become.…”

“…In Experiment 1, we studied the role of information quality for the priors’ usage in haptic frequency discrimination 13 . Participants spent on average 5.5 s ( SD = 1.36) per trial on the stimuli, performed 6.6 strokes ( SD = 2.67), switched 1.7 times ( SD = 0.37) between them and gave 91.1% correct responses ( SD = 8.88%).…”

“…The role of information quality for the priors’ usage was in the focus of Experiment 1. Note that most results of this experiment have been pre-published in a conference paper 13 and are being reinvestigated with further experiments in the current study. Oftentimes, sensory input is noisy, lossy, or unreliable due to e.g., memory decay, attentional processes, or physical conditions of the environment.…”