Purchase this article with an account.
Simon Watt, Bruce Keefe, Paul Hibbard; Visual uncertainty predicts grasping when monocular cues are removed but not when binocular cues are removed. Journal of Vision 2008;8(6):297. doi: https://doi.org/10.1167/8.6.297.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Grasp apertures are typically wider under monocular than binocular viewing. Several researchers have proposed that this represents an added “margin of error”, given increased uncertainty about object properties (see Melmoth & Grant, 2006). Consistent with this, it has recently been shown that increasing perceptual uncertainty, by presenting objects at eccentric retinal locations, produces systematic increases in grasp apertures (Schlicht & Schrater, 2007). We investigated whether the differences between monocularly- and binocularly-guided grasps are predicted by changes in uncertainty. The stimuli were rectangular objects on a horizontal surface. Grasps were visually open-loop. Experiment 1 measured the effects on grasp kinematics of selectively removing binocular or monocular cues. We varied distance to produce variations in uncertainty (cf. Hillis et al., 2004), which were measured psychophysically using size/distance discrimination. Experiment 2 measured grasps as a function of uncertainty per se, holding the depth cues available, and the required movement, constant. The stimulus was a random-dot-stereogram. We made the object transparent and increased dot density, making it difficult to discriminate the object from the surface. We again measured size/distance discrimination to assess uncertainty. We took measurements at each distance used in Experiment 1, to control for grasp effects due to increased movement variability at farther distances. In Experiment 2 grasp apertures increased systematically with increases in perceptual uncertainty. We used this relationship to generate predictions for the effects of removing binocular and monocular cues in Experiment 1. Grasp apertures increased when monocular cues were removed by a similar amount to that predicted by the change in perceptual uncertainty. The removal of binocular cues, however, resulted in an increase in grasp apertures considerably larger (by a factor of ∼2) than predicted by changes in perceptual uncertainty alone. We conclude that the removal of binocular cues has effects on grasping over and above increasing uncertainty.
This PDF is available to Subscribers Only