July 1968
The Contrast Between Physical Motions and Optical Motions
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 References to these essays must cite them explicitly as unpublished manuscripts.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.
Physical optics is a branch of modern physics. Ecological optics is an undeveloped branch of ecology, or ecological physics. The former considers light as radiant energy; the latter treats it as the steady state of projections in the reverberating light between reflecting surfaces. The former assumes that what we see is light; the latter assumes that what we see is objects by means of illuminating light. The former does not distinguish between the physical motions of bodies and the optical motion of particles (or waves); the latter distinguishes sharply between the physical motions of bodies and the optical motions (or transformations) of the projections of bodies illuminated in two motion picture films (Gibson, 1955, 1968).
A consequence of not distinguishing between physical motion and optical motion is Einstein’s special principle of relativity, which takes into account the speed of light in considering the visual perception of very distant events (e.g., their simultaneity). But if the visual perception of terrestrial events depends on changes of steady-state projection in a optic array, not on the propagation of energy from the event to an eye then Einstein’s principle need not apply. In the ambient optic array of projections to a point of observation, all terrestrial events are contemporary. In the rays of luminous energy from the night sky to a point of observation, the celestial events are not contemporary. The occlusion of an ordinary object by an edge, a cessation of projection, is not delayed when the edge is distant. But the occlusion of a radiating source by an edge, an interruption of propagation, is delayed when the edge is distant.
The consequences of distinguishing between the physical motions of bodies and the optical motions in the ambient array of light are far-reaching. The latter constitute the ordinary stimulus information for perceiving the former. The latter are not motions at all, in the sense of the term used by Isaac Newton, but mathematical transformations or transitions. Optical motions are not subject to inertia, or to the “laws of motion”. They are changes in the form, pattern, or structure of an array. Research on the perception of “motion” has suffered from the prevalent confusion between the stimulus and the source of stimulation, between optical transformation and the environmental cause of a transformation.
This distinction also permits research on the visual detection of locomotion, that is, the movement of the observer himself relative to a stationary environment. A physical displacement of the observer causes a corresponding optical motion (transformation) of the whole ambient array at the point of observation. And this fact promises to clarify a set of problems connected with oriented locomotion, with the so-called “cue” of motion parallax, with the difference between proprioception and exteroception, and even problems in connection with skills like automobile driving and the landing of aircraft.
In short, for the psychologist, the physical motion of a body or a particle in space (including the propagation of light itself) has to be distinguished radically from changes in projection — either the projection of a moving surface to a point of observation or the projection of a stationary surface to a moving point of observation. The distinction is necessary if we are to understand the perception of motion and locomotion, although it is unnecessary (and inconvenient) for the physicist’s abstract conception of motion with reference to three coordinate axes in empty impersonal space. But even the physicist has to face the problem of perception in interpreting the results of experimental observations. This problem takes a quite different form when the motions of objects an animals are considered than it does when the motions of stars and particles are considered.