August 1967
Note on an Elaboration on the Distinction Between the Proximal and Distal Stimulus
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 References to these essays must cite them explicitly as unpublished manuscripts.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 distinction made by Brunswik, Koffka, and Heider between the proximal stimulus and the distal or distant stimulus is the beginning of wisdom in the study of visual perception. However, it does not go far enough. The distal stimulus is the term applied to the object; the proximal stimulus is the term applied to the retinal image. Heider maintained that the medium in between the two, containing only rays of light, would not convey the organization of the object to the retina. Hence the organization of the object had to be constructed or produced by the brain in a manner suggested by Köhler’s “field theory,” or by some other theory such as Brunswik’s.
The visual situation is actually much more elaborate than this. The object (one that either reflects or emits light) should be called a source of optical stimulation (for it is not a stimulus at all, as I have pointed out). The ambient light at the station point of an eye is a potential stimulus information for the ocular system (the perceptual system). The temporary retinal image considered as light energy is the stimulus for the retinal receptors (rods and cones). The potential retinal image considered as a projection of the array of ambient light is not a stimulus but a “stimulus situation” which is explored by the binocular perceptual system for the information available in it.
These new distinctions presuppose, of course, that normally there is information in ambient light (in disagreement with Heider’s argument) or, more exactly, that information exists when the medium is “clear” (not filled with fog). The information in reverberating reflected light depends on the theory of a steady state in the medium, consisting of a “dense intersecting network of projections” (Ch. 1 and 10 in Senses Considered). This steady-state network is assumed to be independent of the intensity of illumination.
This theory makes a radical separation between stimulus energy and available stimulus information. The latter is invariant with changes of the former. The latter is associated with perceptual systems; the former with receptors. The latter is not quantifiable in metric units; the former is quantifiable in CGS terms. Information has no fixed threshold; energy is an effective stimulus only when it exceeds the threshold of the receptors. Vision fails in darkness because information is lacking by reason of energy being absent; vision fails with dazzling illumination because, although information and energy are both present, the system is swamped by energy. Stimulus energy is a necessary and sufficient cause of excitation; stimulus information is a necessary but not sufficient condition for perception. The term “stimulus.” by itself, should only be used for an event that is a necessary and sufficient cause of some “response.” Instead of continuing to talk loosely about objects as “stimuli,” we should begin to specify the stimulus information in ambient light for the detection of objects. This is the aim of ecological optics.
Heider’s conclusion that the air between the object and the eye could not convey the unity of the object was based on the traditional assumption of physical and geometrical optics that each light ray was independent of every other. The medium, he therefore asserted, contained only “spurious units” (Thing and Medium, p. 7). The perceptual apparatus had to “transform these spurious units again into true units” (p. 24). But if the medium contains projections of the faces and facets of objects, as ecological optics assumes, over and above the reverberating photons described by physical optics, then the old puzzle of Brunswik’s “distal focusing of perception” is resolved.
Brunswik was never convinced by a theory of cortical organization or self-distribution in the brain, feeling that it implied a pre-established harmony between the world and the brain. He preferred unconscious inference to electrical fields as an explanation of the constructive process of perception. This debate is still a stand-off. But if the information for perception is available the process does not have to be one of construction.