Reversible Perspective and Reversed Motion
J. J. Gibson, Cornell University
The World Wide Web distribution of James Gibson’s “Purple Perils” is for scholarly use with the understanding that Gibson did not intend them for publication. References to these essays must cite them explicitly as unpublished manuscripts. Copies may be circulated if this statement is included on each copy.
The front-back reversal of the phenomenal object obtained with a drawing such as a Necker cube should not be confused with the front-back reversal of the phenomenal object obtained with a real cube made of wire. In the absence of binocular parallax or motion parallax, both arrays are ambiguous with respect to which is front and which is back, that is, the information is equivocal. (Senses Considered, p. 246 ff). The illusion of “reversible perspective” therefore can arise in both cases.
The reason they should not be confused is that when the observer (or the object) moves it begins to make a difference whether the array comes from a picture or from a real object. There is a kind of transformation of the optic array in the latter case that yields the so-called kinetic depth effect, but only the kind of transformation in the former case that yields foreshortening or change of slant. Whether the former phenomenal object is obverse or reverse makes no difference, but whether the latter phenomenal object is obverse or reverse makes a striking difference.
The optical displacement of “the front” relative to “the back” of a real object is governed by the principle of parallax. If the front and the back of the phenomenal object are reversed, then the relative apparent displacement of its front and back are also reversed. Hence if the material object is turned one way the apparent object must seem to turn the other way or, if the observer moves his head to the right, the front of the apparent object must seem to move to the left. This reversal of apparent motion with reversal of apparent depth has often been observed, and has sometimes been described, but has not been explained. Is the above explanation satisfactory?
Any object whose front and back can be phenomenally reversed without changing occlusion will show the illusion, a paper cup or a glass tumbler for example. It does not have to be an object made of wire, as sometimes assumed.