The Theory of Images Transmitted to the Brain
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 References to these essays must cite them explicitly as unpublished manuscripts.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.
Most people cannot conceive of any alternative to the accepted doctrine that the optic nerve relays images to the mind. But the accepted doctrine is full of contradictions when it is considered carefully. And the radical alternative is to assert that the exploratory perceptual system, consisting of the eyes, brain, and ocular adjustments, is attuned to the information in light. This theory rests on ecological optics. It rejects the assumption that what falls on the retina is properly an image, and asserts that what the ocular system does is to pick up the information in an optic array. It says that an observer is an organism, composed of organs, not a little man in a dark skull waiting for images to be relayed to him. It denies that the theory of cameras and optical instruments is a good guide to the understanding of vision and says that a theory of the optimizing or maximizing of awareness is a better guide. This is not a theory of the processing of sensations on the one hand or of behavior on the other; it is a theory of the circular cybernetic adjustment of that perceptual system we call the visual system.
This is not to deny that the human eye can be treated so as to show the symptoms of an image in some respects. The retinal after-image (positive or negative) is analogous to a photographic image on film, in that it seems to be a photochemical effect. It results from a sufficient “exposure” of the retina to light just as the silver halide deposits in film result from a sufficiently long or intense exposure of the film to light. The image has to be “burned into” the photosensitive surface. The question that arises is whether or not the experience of seeing a retinal after-image is like the experience of seeing a photographic image. I think it is not.
The retinal after-image is what I wish to call a pure subjective sensation. It is like a ringing in the ears or an itching of the skin. It’s source in the receptors instead of in the environment is specified by the fact that when the eyes are moved, or the head is turned, or the skin is moved, the sensation does not move the other way. The sight is then necessarily in the eyes; the sound is in the head; the touch is in the skin. An externally caused sight, sound, or touch, an “objective” sensation, is specified by the way in which the eye-movement, head-movement, or limb-movement changes the locus of excitation to the sensory surfaces, that is, it is specified by the feedback of the eye, head, or limb movement. A sensory input not altered by sensory exploration, however, is a subjective afterimage. An object in the world can be explored or scanned; an afterimage in the receptors cannot. (Note that this formula is not that of von Holst; it is contradictory to his, although similar.)
Now is it true that to have an afterimage is somehow to see the after-excitations of the retina? Might it be said that the pattern of after-excitations in the optic nerve, or perhaps the brain, can be attended to? There is something radically wrong with this way of speaking since no observer exists in the head who could possibly see or attend to neurophysiological patterns. The experiencing of afterimages, therefore, must be a kind of experience wholly divorced from normal seeing or attending.
It should be remembered that afterimages in the course of ordinary life are an annoyance only, of not interest to the observer of the world. They are symptoms of having overstrained the capacity of the perceptual system. They become of interest only when men of leisure began to introspect, to describe their sensations, and to realize that their senses could be excited in quite unnatural ways.
The visual afterimage is the only clear case of the supposed transmission of a physiological image from receptors to brain. In ordinary vision a physiological image, if it existed at all, would scarcely last a fraction of a second. Ordinary vision must not entail the transmission of an image from receptors to brain.
It is important not to confuse the physiological image in the retina with the optical image on the retina, although classical optics leads us to do so, and the assumption that the retina is an energy transducter encourages the fallacy. The physiological image in the retina is what might be transmitted to the brain. The optical image on the retina is a temporary sample of the ambient array. Since the retinal afterimage constitutes the main evidence for the existence of a physiological image and since it has nothing to do with normal vision neither does the
A retinal afterimage is said to be “projected out” from the retina into the environment and on whatever surface faces the observer, in accordance with Emmert’s Law. Whether or not this is true (and I doubt it), it provides the basis for the theory that ordinary visual sensations are projected out from the eye, since only thus could we see space. The falsity of this line of reasoning should now be evident.