Levels of the Optical Stimulus-array and the Hypothesis of Corresponding Levels of the Exploratory Visual System
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 References to these essays must cite them explicitly as unpublished manuscripts.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 visual equipment is carried around in a body. This includes a head, and the latter contains eyes. All three are mobile for purposes of exploration. The body moves relative to the substratum, the head must move relative to the body, and the eyes have to rotate relative to the head. But a system of automatic postural adjustments is at work such that the eyes can be stabilized (or can move) relative to theenvironment; there is a hierarchy of reflex compensations of the eye-muscles, the neck-muscles, and the body-muscles which enables the eyes to be fixed on the environment. An exploratory head-movement with an eye-movement can thus be a shift from one stable fixation to another. The eyes can fix on the environment only by means of light.
Accordingly there are three levels of the stimulus situation for vision, or three orders of what I call the stimulus-array. The first is a sphere of focusable light, the second is a hemisphere, and the third is acone.
1. The total array for a body. This is explored by head-turning (and subordinately by eye-turning). It is the whole array of ambient light at a station-point, or “viewpoint”, surrounding the individual. It is the available stimulation at a locus in space for the whole eye-head-body system. The coordinates for this array are those of a body in postural equilibrium (cf. “Ecological Optics” in Vision Research, 1961, 1, 253-262).
2. The partial array for a head. This is explored by saccadic eye-movements with head stationary. In man, it is approximately a hemisphere of the total array. It is the available stimulation for the binocular oculomotor system. (It has to be explored or scanned since a foveated ocular system cannot register the interesting details of the array simultaneously but only in successive acts of registration). The coordinate system for this array is that of a head.
3. The partial array for an eye. This is not explored but registered by a fixated eye, the registration being accomplished (in vertebrates) by a retinal image. It is a cone of focusable light, the entering cone for one eye having considerable overlap with the entering cone for the other eye (in primates). It is the stimulus for a retina which has been stabilized (with some residual tremor) by the postural compensations described above. The coordinate system for this array is that of the anatomical entity we call an eye.
It should be noted that each partial stimulus-array is included in the one superordinate to it, although each array has a special coordinate-system appropriate to it. There are three corresponding levels of the visual system, in all probability, both anatomical and physiological. There is the eye, the eyes-in-the-head, and the eyes-in-the-head-in-the-body.
The total array is the available stimulus corresponding to the phenomenal visual world. The head-array is the stimulus corresponding to the binocular visual field, sometimes called the field of view (Blickfeld). The eye-array is the stimulus corresponding to the visual field proper (Sehfeld). Most visual research has so far been concerned with this last stimulus and the resulting sensory impressions.
The visual system presumably responds at three different levels to these three different modes of optical stimulation. The visual process involves more and more exploratory activity the greater the scope of the optical stimulus considered. It is relatively passive, instead of active, only at the lowest level of stimulation, and even then accommodation and pupillary adjustment still operate. There is evidence to show that the lower visual processes are functionally subordinate to the higher ones in the normal use of vision, just as the less inclusive stimulus-array is geometrically subordinate to the more inclusive.
The experimental control of the more inclusive stimulus-arrays for studying the process of visual perception has not been attempted until recently. One method is to modify the field of view of the head (No. 2) by the use of experimental spectacles, as in the Innsbruck experiments. The adaptation that results shows the subordination of the retinal system to the exploratory ocular system (the “conditional” aftereffects). Another method might be to modify the whole optic array (No. 1) by the wearing of a transparent helmet attached to the shoulders. Adaptation to biased stimulation of this type, if it occurred, would show that even the neck muscles can participate in the process of registering the invariant information in an optic array.