Misalignment effects in 3-D versions of Poggendorff displays

Abstract

Strong misalignment effects are found in three-dimensional (3-D)versions of Poggendorff displays viewed binocularly. The components of the standard 2-D Poggendorff figure-the parallels and the oblique segments-were presented in 3-Ddepth as a flat rectangular object with occluding edges and an oblique line situated behind the object. Three experiments investigated the misalignment effects under three different observation instructions: Subjects were told to look at the oblique (Experiment 1), at the rectangle (Experiment 2), or at the background (Experiment 3). Experiments 1 and 2 examined the effects on judgments of alignment of varying the distance in depth that separates the oblique from the rectangle. Experiment 3 examined the effects of varying the distance between the fixated background and the 3-DPoggendorff figure. Both standard and reversed misalignment effects were obtained. When the viewing condition produces crossed disparity for the oblique, perceived misalignment occurs in the usual Poggendorff direction, but it is reversed with uncrossed disparity. Moreover, the amount of misalignment is related to the amount of disparity, and it can be much stronger than is usual in the 2-D versions of the Poggendorff. The misalignment effects can be explained by binocular integration to produce a single cyclopean image.In the Poggendorff illusion (Figure 1), the oblique lines erroneously appear to be misaligned. What we show here is that misalignment effects also appear in solid (three-dimensional, or 3-D) binocular versions of Figure 1. We show how the effects depend on factors arising during binocular integration. There is a variety of contradictory reports on binocular versions of the Poggendorff in the literature, and we show how each of the varied reports can be obtained and explained by standard principles of stereopsis.While sometimes it has been considered to be due to depth perception, Poggendorff misalignment has only rarely been associated with binocular vision. Fisher and Lucas (1969), Schiffman (1990) report that pictures of real-life settings, such as an oblique tree behind an upright tree, create the same effect as the Poggendorff figure. However, the observations were undertaken only with flat pictures, and possible binocular versions, though noted by these authors, were not assessed. Pieron (1955) tried explicitly to observe the Poggendorff effect in a fully 3-D world. He observed that the effect failed there! He stated that if the two segments ofthe oblique line are made ofrope passing behind a column and serving to pull a bucket out of a well, the illusion disappears completely. Julesz (1971) examined several illusions in random-dot stereograms. In particular, he observed that when the Poggendorff rectangle is sepWe would like to thank P. Bennett for his comments on an early version of this paper. Correspondence should be addressed to C.

The Pursuit of Leonardo's Constraint

“…The studies reported in the literature, however, show that these two possible consequences are limited to visual direction and do not occur for visual shape and visual alignment in 3-D perception (Ohtsuka and Yano 1994;Ono et al 1998;van Ee and Erkelens 2000). In 3-D space a square behind an occluder is seen as a square (Ohtsuka 1995a;Ono et al 1998) and a line behind an occluder is seen aligned (Liu and Kennedy 1995;Ohtsuka 1995b;Drobnis and Lawson 1976;Gyoba 1978). To demonstrate for yourself that the expectations illustrated in figure 7B do not occur, see demonstration 5 on the website.…”

Sex Differences in the Poggendorff Illusion: Identifying the Locus of the Effect

“…sion with 3 -dimensional stimuli depends upon the actual distance between the oblique and the inducing element (Liu & Kennedy, 1995). Thus, it would be interesting to compare sex differences in 2-dimensional and 3-dimension-a1 Poggendorff contexts to identify whether sex differences would be reduced in the latter context, one in which depth information is directly available to the participants.…”

Leonardo's constraint: Two opaque objects cannot be seen in the same direction.

“…It is similar to Gregory's (1963) hypothesis about the inappropriate constancy scaling as an explanation for the Mueller-Lyer illusion (see Day & Knuth, 1982;Robinson, 1972). Moreover, the ideas presented here address theoretical puzzles, namely, (a) why shape perception remains veridical in three-dimensional perceptual space despite visual direction being nonveridical (Ohtsuka & Ono, 2002;van Ee & Erkelens, 2000); (b) why in three-dimensional perceptual space, a square behind an occluder is seen as a square (Ohtsuka, 1995a;Ono et al, 1998) and an oblique line behind an occluder is seen aligned (Drobnis & Lawson, 1976;Gyoba, 1978;Liu & Kennedy, 1995;Ohtsuka, 1995b); and (c) why the end of an oblique line that abuts an occluder in a drawing appears more misaligned than does its far end (Wenderoth, 1980). To propose that the underlying mechanism for visual direction examined in this study is different from that for shape and alignment is plausible but ad hoc.…”