Purchase this article with an account.
Stephen P. Sebastian, Johannes Burge, Wilson S. Geisler; Humans correct contrast for defocus blur: a new kind of contrast constancy. Journal of Vision 2013;13(9):1145. doi: https://doi.org/10.1167/13.9.1145.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Contrast constancy is the perceptual ability to perceive the contrast of targets in the environment as constant, even as sensory constraints reduce the contrast of the image transmitted up the optic nerve to the brain. Here, we demonstrate a new form of contrast constancy where humans correct for the contrast lost via defocus blur. We constructed a custom psychophysical rig capable of presenting light from three different distances simultaneously along the same line of sight. Subjects viewed stimuli monocularly through a 4mm artificial pupil using a bite bar. Each trial began with a focus target presented at 80cm. After the subject focused the target, standard and comparison targets—narrow bandwidth Gabors embedded in a hard-edged disks—were presented simultaneously for 200ms. A narrow band Gabor was used because defocus alters its contrast without altering its spatial wave form. The hard-edged disk provided a cue to defocus. The standard target had a Michelson contrast of 0.6 and was perfectly focused. The comparison target had variable contrast and was presented at defocus levels ranging from 0.00 to 0.75 diopters. In a 2AFC paradigm, subjects judged which of the two targets had greater contrast. If subjects exhibited no contrast constancy, they would match the targets’ retinal contrasts. (Retinal contrasts can be predicted from the optical aberrations of each subject’s eye.) If subjects were perfectly contrast constant, they would match the targets’ on-screen contrast. We found that while subjects were not perfectly contrast constant, the defocus cues produced a significant increase in contrast constancy. Control conditions showed that subject responses were more strongly dictated by retinal contrast when cues to the target’s defocus were removed. These results imply that defocus blur is used to correct contrast. These results demonstrate a new perceptual function for defocus blur, and demonstrate a new form of contrast constancy.
Meeting abstract presented at VSS 2013
This PDF is available to Subscribers Only