No abstract

In three experiments, the durations of alternative perceptual organizations of drawings and of a three-dimensional object were used to provide quantitative measurement of local cue strength and of the viewer's intention. In Experiment 1, small (2.2°) line drawings of cubes were constructed so that a local depth cue (occlusion) at an upper intersection specified the cube's orientation while the lower intersection remained ambiguous. On any trial, subjects looked at either the biased or the unbiased intersection, with instructipns to try to hold one organization or another; hence, we call this general procedure the opposed-set method. The stimulus features nearest the viewer's instructed fixation and the viewer's perceptual task both had strong effects on reported durations. Experiment 2 replicated these findings with a real object ( a small three-dimensional moving wire cube) as well as a drawn one. In Experiment 3, the opposed-set methodology was applied to figural completion. The quantitative data provided by this new measurement procedure show that the whole configuration is not the effective stimulus for perception; the data thereby support a constructivist theory by posing grave problems for its strongest present competitors: the global minimum principle and the direct theory.This research was supported in part by National Science Foundation Grants BNS 77-25653 and BNS-82-09470 to Julian Hochberg.We wish to express our appreciation for helpful comments and criticism to Irvin Rock, Daniel Weintraub, an anonymous reviewer, and also to William Epstein, at whose suggestion we performed Experiment 3.Requests for reprints should be sent to

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact

In three experiments, observers who were instructed to perceive one of two alternative depth arrangements of a three-dimensional wire cube fixated near one of two intersections that differed in the degree to which they specified the cube's veridical depth organization. In order to separate perceptual effects from experimenter effects, we measured indirect reports about variables perceptually coupled to perceived depth rather than direct reports about perceived depth. In all three experiments, reversal durations at the two intersections differed, even though the two were parts of a single object. In addition, reversals varied with viewers' intentions. Thus, the unit of perceptual organization may be smaller than the entire object, and viewers' intentions can influence the perception of real moving objects. In additional analyses, reversal durations were separated into two components: nonelective instability and malleability; the question of whether these two components of ambiguity are functionally distinct could not be decided.

No abstract