January 1970
The Visual Ego
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 References to these essays must cite them explicitly as unpublished manuscripts.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.
Koffka distinguished between the phenomenal field extending forward from the head and that extending backward from the head, and asserted that between the “in front” and the “behind” stood the ego (Principles, p. 322). I have a theory of the visual ego which is consistent with this phenomenological description but which fits better with the facts of optics and of human vision. It derives both the experience of “here” and the experience of “out there” from the fact that the eye-head system samples the total ambient optic array. It lays special emphasis on the temporary boundary between the unhidden portion of the environment and the hidden portion of it (Visual World, p. 228).
The solid-angular sample of the ambient array of light at a point of observation that enters each eye is determined by the posture of the head. The experience corresponding to this sample, or its cross-section, is what I call the “visual field” (Visual World, Ch. 3, and p. 226; Senses Considered p. 254). The boundary of this field is fixed by the bony orbits of each eye, but also by the nose, hands, body, and feet of the observer insofar as they protrude into the visual field. The head of the observer can thus be said to hide a portion of the surrounding environment, and the nose, hands, body, and feet can be said to occlude the surrounding environment. Each of these surfaces has an occluding edge of the same sort that is found in the environment proper. An occluding edge is a visible opaque surface in front of another surface (Gibson, Reynolds, Wheeler, & Kaplan). Note that the eyelids can hide the entering sample of the array.
We can now ask what visual information specifies the ego as distinguished from the environment, and what specifies the experience of “here” at zero distance as distinguished from “out there” at increasing distances. Instead of appealing to a continuous gradient of density of optical texture and similar gradients of optical motion and optical disparity, the explanation is in terms of optical accretion/deletion, and it presupposes an active observer who turns his head, moves his body, and locomotes.
1a. The permanent oval border of the visual field, the transition between the sample of ambient light and the non-sample or, in other words, between the unhidden and the hidden specifies the ego of the observer, more exactly, his head. Similarly, the sweeping of this border over the ambient array, revealing elements at the leading edge and concealing them at the trailing edge, specifies head-turning.
1b. The changeable occluding edges that protrude into the field of view specify the hands, the body, and the feet of the observer. Similarly, the accretion/deletion of optical textures at these edges specifies movement of the hands, the body, or the feet. The more of the visual field occluded by such an edge the nearer it is to the eye (cf. Schiff). A hand can cover the eyes if it is near enough. When the eye-lids cover the whole visual field they are experienced at zero distance from “here”.
2a. The edge of the nose, at the right-hand margin of the left visual field and the left-hand margin of the right visual field is at the extreme of disparity for the two array-samples entering the two eyes. The nearer is an edge in the world the greater is its optical disparity. Hence the nose is specified by disparity information as being at zero distance.
2b. The changing disparities of the hands or other bodily extremities specify the movements of the hands or extremities toward or away from the eyes.
3. The covariation between movements of parts of the body, both active and passive, and changes of occlusion in the optic array constitutes one kind of “visual kinesthesis”. (What gives rise to the perception of motion? Psychol. Rev.). All such changes in the optic array are seen as movements of the body or its parts not as motions of objects. Similarly, locomotion is seen as such, and the transformations in the optic array or the “apparent motions ” in the visual field that samples it are not perceived as motions in the environment. The rule seems to be that any optical change that is always a concomitant of a bodily movement will eventually be seen as having a kinesthetic quality and as pertaining to the self instead of to the world, even if at first the optical change is taken to be the motion of an object or the behavior of another animal.
The ability of an observer, child or animal, to learn to see the cast shadow of his body or, more striking, the image of his body in a pool or mirror as pertaining to himself instead of to another organism may depend on the above fact.