Fall 1962
Old and New Assumptions for a Theory of Visual Perception
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 References to these essays must cite them explicitly as unpublished manuscripts.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.
Old:
1. Every point of light that stimulates a receptive element in a retina gives rise to a visual sensation.
2. Every such visual sensation gives rise to a perception of a subjective visual direction (local sign, either innate or learned).
3. A subjective visual field consists of a pattern of such visual directions, i.e., a mosaic of color sensations.
4. A form in a visual field is an association (or an organization) of colored spots.
5. A motion in a visual field is a set of displacements of spots (changes of direction).
6. A disparity between binocular fields (with stimulation of non-corresponding points on the two retinas) causes a doubling of sensations. When these are nevertheless fused “in the brain”, the result is depth perception.
7. Depth perception may also result from a process of the “interpretation” of the retinal pictures. In gestalt theory this becomes a process of the organization (or self-distribution) of the excited processes in the brain. Constancy is a matter of inference, or of field-processes, but in either case is an overcoming of sensory data.
8. Higher forms of perception and cognition are the result of imposing internal intellectual processes on the deliverances of sense.
New:
1. The stimulus for an eye (potential stimulus) is not a light-ray, or a set of rays, but the array of ambient light. The texture and structure of such an array (regardless of intensity) contain information about environmental surfaces and edges.
2. Each eye explores the information in its optic array by fixations and adjustments. The binocular system registers simultaneous disparity. Each eye registers successive transformation due to head-movement. (The static retinal picture is not relevant to perception).
3. The binocular disparity of two arrays, and the monocular transformation of each array (by the geometry of parallax) provide ordinal stimulus-information about slant and depth of surfaces. (The assumption of punctate stimulus-information leads to insoluble paradoxes).
4. Visual sensations, considered either as spots of color or as patches of color in a visual field are irrelevant to the visual perception of the world, not the basis of the perception of the world.
5. The invariant properties of an optic array specify the permanent properties of the outer world; the variant or changing properties of an optic array specify the changes in the world and the movements of the observer himself. The invariants in stimulation explain the “constancies” in perception.
6. The basic problem of perception is how the active ocular system of an observer selects the genuine stimulus-information from the stimulus flux of ambient light. The stimuli may be unfamiliar, obscure, subtle, conflicting, or impoverished. With laboratory-controlled stimuli, perception may be reduced to guessing. But in any case, the crux of perception is attention and the education of attention.