Abstract
Abstract: :
Purpose: To investigate the maximum binocular disparity (Dmax) at which infants 12_20 weeks demonstrate stereopsis. The psychometric function for stereopsis is hump–shaped, with the rising side defining stereoacuity, and the falling side defining Dmax. Infants are known to have stereoacuity values that start higher than those of adults, but move to lower values (i.e., they improve) rapidly. What about Dmax? Does Dmax also start out higher than that of adults, then progress to smaller , more adult–like disparities? Does the hump–shaped psychometric function mature by sliding from larger disparity values to smaller disparity values? Or, does Dmax start out lower than that of adults, then progress to higher disparity values, as the psychometric function rises from chance to higher levels of performance? Methods: We used forced–choice preferential looking to measure detection of a 3–D target portrayed by horizontal binocular disparity within a random texture, compared to a "catch" stimulus portrayed by an equal value of vertical binocular disparity. Disparities were 0.2?_3?. We used 2AFC to collect control data on adults. Results: 12–wk–olds showed no evidence of binocular stereopsis at any disparity value. Neither infants nor adults correctly detected a disparity of 3?. After age 13 weeks, infant performance between .5?_2? of horizontal binocular disparity improved with age. The psychometric function improved by rising from chance to higher levels of performance. However, infant performance remained modest throughout the 12_20–week age range, in agreement with results reported by many other investigators. Conclusions: Dmax for stereopsis is about the same for infants and adults, starting as soon as binocular stereopsis can be measured. These data provide no evidence that infant stereopsis could be measured at a younger age by using larger values of binocular disparity.
Keywords: visual development: infancy and childhood • binocular vision/stereopsis • infant vision