Abstract
In a recent developmental study we found that adult depth discrimination was consistently poorer than that of a group of 4-6 year olds. This surprising result was obtained when two grey-scale test stimuli were presented simultaneously in opposite directions in depth, above and below a fixation point (Giaschi et. al., 2008). The present set of experiments investigates the poor performance in otherwise visually normal adult observers. Specifically, we evaluate the hypothesis that this deficit is caused by a perceptual grouping mechanism that integrates disparity signals across space and disparity like that identified by McKee (1983). We measured depth identification in adults (18–33) using a large range of disparities (1 min to 3.5 deg). The stimuli were the same characters used in our original study; we manipulated their number (two or one), their appearance (via phase scrambling), and their alignment and location relative to fixation. On each trial, observers were ask to indicate which of two targets was closer, or when only one stimulus was presented, whether the target was closer or further away than a reference frame. In all conditions we measured proportion correct. We found that adult performance remained poor in all conditions in which two stimuli, offset in opposite directions, were presented. Depth identification was not improved by changing configural or spatial aspects of the stimulus. However, performance improved markedly when the two targets were offset in the same direction, or only one stimulus was presented.
If the adult's poor performance is due to perceptual grouping, our data suggests that it may be involuntary because all tasks could have been performed using one of the stimuli. Further research is required to determine if the absence of this effect in young children reflects developmental changes in perceptual organization.