The effect of 3-D orientation and stretches on the detection of dotted planes.

Abstract

The present experiment used a masking procedure in which a dotted stimulus form was masked by random dots to explore human form-detection capabilities. The dotted stimulus forms were located in a stereoscopically generated three-dimensional rectangular volume. The form was defined by a set of randomly positioned dots always restricted to a plane. The orientation of the dotted plane varied from frontoparallel to diagonal around a vertical axis of rotation passing through the center of the rectangular volume. The plane always stretched across the region from side wall to sidewall. Since the volume was deeper than it was wide, both the apparent area and the apparent dot density of the plane varied with orientation. The masking dots were distributed at random throughout the volume. For a constant number of stimulus-form and masking dots, however, the detectability of the plane was shown not to depend upon orientation. This counterintuitive result in conjunction with earlier findings suggests that there is a range of conditions over which detectability depends only on the total numbers of stimulus and masking dots and not upon subjective orientation, density, or area of the plane. Therefore, observers respond as if they were sensitive to the retinal two-dimensional density rather than the apparent three-dimensional density.