The Schröder staircase: A new perspective

Abstract

A range of variants of the Schroder staircase was generated by computer, giving quantifiable changes of perspective and the option of adding appropriate texture to the steps of the staircase. Increases in perspective and the addition of texture gradients led to increases in the proportion of viewing time during which the percept appropriate to the perspective was reported. It is suggested that this might be a useful method of studying the processing of depth information.There are several reasons for supposing that the perceptual alternations of a class of reversible figures, which includes the Necker cube, Schroder staircase, and Mach card, are linked with depth processing mechanisms. First, subjects often describe such an alternation as a reversal of the apparent depth relationships within the figure. Secondly, adding depth cues to such figures can alter both the rate of alternation and the relative durations of the various percepts. A third line of evidence is much more indirect: although the alternation rate of a number of multistable figures changes with the subjects' viewing distance, that of depth reversible figures does not (Atkinson, Campbell, Fiorentini, & Maffei, 1973).The evidence that depth cues affect reversible figures is somewhat variable in its quality, ranging from brief anecdotal reports, through incidental observations, to experiments in which attempts were made to quantify both the depth information added to the reversible figure and the resulting changes in perceptual alternation. For example, Ogle (1962) mentions that the addition of perspective to the Necker cube and Schroder staircase produces a tendency to one percept. Orbach and Zucker (1965) report that removing those edges from a Necker cube which would be hidden if it were solid has a similar effect. It was found by Howard (1961) and Virsu (1975) that adding disparity to a skeletal wire cube and Schroder staircase, respectively, increases the initial period before a perceptual reversal. In a study of the Necker cube, Cormack and Arger (1968) reported that increases in disparity up to 45 min of arc both increased the durations of the percepts appropriate to the disparity and decreased the alter-. nation rate, but that a further increase to 90 min of arc had little additional effect. This last study, if I thank the MRC for support; Nick Markovits for his programs, advice, and help, without which the study would have been impossible; Ian Low for building the timers and printer; Philip Clark for photography; and Richard Gregory, David Rose, and Priscilla Heard for helpful comments. not unique, must be one of the few investigations in which the added depth cue has been systematically varied.These studies are usually said to tell us something about perceptual multistability. But it also seems plausible to argue that they tell us something about the way in which depth information is processed by the visual system. There are some assumptions necessary for this argument, which will now be considered. The first is that the kinds of perceptua...

The local character of perspective organization

“…These results confirmed that adaptation is a powerful tool that can be used to manipulate perspective orientation of reversible figures. Similar effects have been obtained before (Carlson, 1953;Harris, 1979Harris, , 1980Hochberg, 1950;Virsu, 1975), but the effect had to be established for the present situation and stimuli.…”

As the cube turns: Evidence for two processes in the perception of a dynamic reversible figure

“…Moreover, such a two-process model of reversible figures can incorporate a wide range of empirical findings that otherwise present serious difficulty for those investigators favoring either satiation or learning as the single dominant process underlying perceptual reversals. For example, how is a strict learning/decisional model able to reconcile the finding that adaptation to an unambiguous version of a normally reversible figure results in a greater probability that an observer, upon being then presented with the standard reversible figure, will report the opposite version of the figure (e.g., Carlson, 1953;Harris, 1979Harris, , 1980Hochberg, 1950;Virsu, 1975)? The satiation model, on the other hand, predicts this precise result because of the fatiguing of a particular channel by the prior adaptation.…”

Making ambiguous displays unambiguous: The influence of real colors and colored aftereffects on perceptual alternation

“…Virsu (1975), for example, made the Schroeder staircase display unambiguous by presenting it to subjects dichoptically with appropriate binocular disparity allowing only one perceptual organization to occur. Harris (1979) added texture gradients to the Schroeder staircase to produce a similar effect with binocular viewing. Lawson, Packard, Lawrence, and Whitmore (1977) varied the disparity information available during dichoptic viewing of Figure lb.…”