
Humans use stereo and haptic distance cues to improve physical object size estimates
Author(s) -
Peter W. Battaglia,
Marc O. Ernst,
Paul Schrater,
Massimiliano Di Luca,
Tonja Machulla,
Daniel Kersten
Publication year - 2010
Publication title -
journal of vision
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.126
H-Index - 113
ISSN - 1534-7362
DOI - 10.1167/8.6.1090
Subject(s) - subjective constancy , haptic technology , observer (physics) , perception , depth perception , computer vision , artificial intelligence , computer science , sensory cue , stereotaxy , apparent size , cognitive psychology , psychology , physics , quantum mechanics , neuroscience
An object's visual image size is an ambiguous cue to its physical size. But if the object's distance is known, the physical size may be disambiguated and more accurately estimated. We asked whether humans use distance cues to improve size judgments. We presented participants with a virtual ball that changed in physical size (imagine a balloon inflating or deflating) as well as distance simultaneously, and asked them to discriminate whether the physical size increased or decreased. With only visual image size information, size-change discrimination was poor. When additional haptic and/or stereo distance-change cues were provided, size-change judgments improved significantly. We conclude that the brain exploits its knowledge of how image size, physical size, and distance are related to improve perceptual size judgments. We compared participants' use of distance cues with predictions of an ideal observer that incorporates distance cues in proportion to their reliability to quantify human behavior. We independently measured participants' stereo and haptic distance discrimination performance, applied these empirical reliability measurements in the ideal model, and found participants use stereo information to a similar degree as the ideal observer, but use haptic information less than the ideal observer. This result was confirmed by an additional conflict condition in which haptic and stereo distance-change cues indicated different values and their relative use could be measured. Lastly, we ran a condition in which participants gripped the object with two fingers, so that a direct size-change cue was available, and found participants integrated direct and indirect size-change cues to improve performance