Purchase this article with an account.
Monica Gori, Maria Michela Del Viva, Giulio Sandini, David Burr; Six-year-old children do not integrate visual-haptic information optimally. Journal of Vision 2007;7(9):216. doi: https://doi.org/10.1167/7.9.216.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Recent studies suggest that information from our five senses is integrated in a statistically optimal (Bayesian) fashion, weighting each sensory input with an estimate of its reliability (inverse variance of sensory noise). In this study we investigate whether 6-year-old children also integrate visual and haptic information optimally. We asked 3 adults and 11 6-year-olds to judge the width of real plastic blocks (about 5 cm wide) using vision, touch or both. In the bimodal condition the block passed through an opaque barrier that concealed the hands from view, allowing a conflict to be introduced between the visual and haptic modalities while creating the illusion of a continuous block. Subjects reported whether the test blocks of variable widths appeared to be narrower of wider than the probe, producing bias-free psychometric curves from which perceived subjective equality (PSE) and threshold (root-variance) were calculated. The children were significantly worse than adults in both the visual and haptic judgments (1,5 times for vision, 2 times for touch), suggesting that these perceptual capacities were still developing, particularly for touch. Adult observers combined visual and haptic information optimally: the PSEs in bimodal conflict conditions were well predicted by the variance-weighted average of visual and haptic information; and audio-visual thresholds were lower than either unimodal threshold, very close to the Bayesian prediction. Children, on the other hand, combined information sub-optimally, giving far too much weight to touch. This was true both for the PSEs (that tended to follow haptic rather than visual information), and bimodal thresholds that were 1,5 times higher than the vision-only task. In both cases the actual weight for about equal to that of vision, while it should have been 2,2 times lower. These results show that perceptual systems of 6-year-olds are immature, particularly in integrating cues from different modalities.
This PDF is available to Subscribers Only