by Morton Hunt
Consider one that the Greek philosophers were the first to ask: How do images of the outside world reach the intellect within?
Plato speculated that the eye actively seeks information by sending forth emanations of some kind that encompass objects—palpating them visually, so to speak. Democritus disagreed, arguing that perception works in the other direction: each object constantly imprints its likeness on the atoms of air, and these replicas, traveling to the viewer, interact with the atoms of the eye and re-create the likeness there, whence it passes to the mind. It was a better guess than Plato’s but wrong in all its details.
In 1604 the German astronomer Johannes Kepler made a leap forward in the understanding of vision. Recent developments in optics and optical instruments enabled him to recognize that the clear body in the front of the eye is a lens that bends rays of light coming from any object, casting an image of the object on the eye’s screenlike retina, from which the resulting nerve impulses are transmitted to the brain.
Ever since, the notion has prevailed that the eye is a kind of camera; the metaphor fits the facts of nearsightedness, farsightedness, and astigmatism, and their correction by eyeglasses. But while it is valid in some respects, it is seriously misleading in many others. Ralph N. Haber, long a leading figure in perception research, has called it “one of the most potent though misguided metaphors in psychology” and the source of much “mischief.”1
What sort of mischief? For one thing, in a camera the image projected by the lens is upside down, and in 1625 Christoph Scheiner, an astronomer, showed that this is also true of the eye. He carefully peeled away the outer coating of the back of an ox’s eye and through the semi-transparent retina saw an upside-down version of whatever he aimed the eye at. But if we see the image that is formed on the retina, why do we not see the world upside down? The question was to plague psychologists for three centuries.2
Another difficulty created by the eye-as-camera metaphor became evident with the advent of photography. To form a sharp image, a camera must be held still during the exposure or, in the case of a movie camera, open and close its shutter many times a second; our eyes, however, constantly jiggle back and forth, even when we look steadily at some point, yet do not produce blurred images. Although we are not aware of and do not normally experience these movements, we can see them by means of a simple procedure. Look steadily at the black dot in the center of the diagram below for about twenty seconds, then quickly shift to the white dot and gaze at it fixedly. You will see an illusory pattern of black lines wavering slightly to and fro. The black lines are an afterimage, due to temporary fatigue of the retinal receptors on which the white lines fell for twenty seconds; the wavering is the never-ending movement in question.
FIGURE 21
Test pattern for perceiving constant eye movement
The meaning of the demonstration is that the eye may be somewhat like a camera, but seeing is nothing like taking pictures.
A second interesting question: Is what we see actually out there? A corollary question: Does it look like what we see? Folk wisdom has always held that we see what exists and what we see is a faithful account of what exists. We see a closed door before us, reach out to the doorknob, and it is where we expected it to be and does what we expected it to do; we lower ourselves onto a chair and it is real and solid, as it appeared to be; we raise a forkful of fettuccini Bolognese to our mouth and it is rich, meaty, and chewy, just as we anticipated. Common sense and philosophy agree that perception is contact with reality. Only a few rare birds, like Bishop Berkeley, have ever doubted that there is a world outside ourselves that corresponds to our perceptions.
But though nearly all of us reasonably assume our perceptions to be truthful, physicists now assure us that the colors we see do not exist as colors outside our heads. The red of a ripe apple, for instance, does not exist as red in the apple; what does is a surface that absorbs all visible light except in the region of 650 nanometers wavelength, which it reflects. When that specific radiation reaches the human eye, the brain perceives it as what we call red. It may be disconcerting to think that the whole splendid colorful world we see on a spring day doesn’t really look like that outside of our own minds. But perhaps we should set aside this philosophic/metaphysical issue and consider a much more approachable problem of vision, namely, that we often have visual experiences we know are misleading or erroneous but cannot will ourselves to correct. The moon, on the horizon, looks huge; we are aware that it does not change size but cannot make ourselves see it as no larger than it is when overhead. We stare at a bright light and, looking away, see an after-image—a perception, but not of anything outside ourselves. We have dreams in which we see persons, places, and actions that are not before us, as they seem to be, or may not even exist.
There are, furthermore, the many illusions that psychologists have studied in the past and the present century. In the following diagram the gray tones of the inner areas look quite different from each other, but actually are identical, as you can determine by cutting a small hole in a piece of paper and centering it over first one and then the other. The mind, or at least the brain’s visual cortex, judges lightness in terms of contrast, not absolute intensity. What you see is not what exists.
FIGURE 22
Which central area is darker? Wrong!
Here are several other classical illusions, each named for its discoverer: (1) the Zöllner, (2), the Poggendorf, (3) the Jastrow, and (4) the Hering: Contrary to what your eyes tell you (and as you can verify with a ruler), the straight lines in (1) are parallel to one another, the angled lines in (2) are aligned, not offset from each other, the figures in (3) are the same size, and the heavy lines in (4) are perfectly straight.
FIGURE 23
Four classic visual illusions
Another category of illusion consists of ambiguous figures that we can will ourselves to see as either one or the other of two different things. Two examples:
FIGURE 24
Two reversible figures
In (1) you can will yourself to see the familiar Necker cube as if you were looking down on it, with corner X closest to you, or as if you were looking up at it, with corner Y closest to you. In (2) you can see the handles attached inside the two white sides of the basket—or, if you choose, attached inside the gray sides.
Finally, in the following diagram there appears to be a triangle that is distinctly whiter than the surrounding area, but it was you who created both the triangle and its brightness; no such figure is there, nor is the paper any whiter where the triangle seems to be than in the adjacent background.
FIGURE 25
The triangle that does not exist
As we proceed, we will learn the explanations for some of these illusions; for now, the point is that perception in human beings is not simply a physiological process that transmits representations of outside stimuli to the central nervous system; it often involves higher mental processes that make sense (and sometimes nonsense) of the impulses arriving via the optic nerves.
A third interesting question—Edwin Boring, in his monumental History of Experimental Psychology, calls it “the first mystery of vision”3—is that we have two eyes yet do not see everything doubled. Galen long ago rightly hypothesized that this is because the nerve fibers from both eyes lead to the same part of the brain. But that is only a partial answer. The two retinas receive somewhat different images of all but distant objects, as is easily confirmed by alternately opening and closing each eye while looking at a nearby object. (Each eye sees more of one side of the object than the other, and sees the object in a different relationship to things in the background.) But if these somewhat different images overlap in the brain, why is the result not blurred?
Perception researchers now answer that “fusion” of the dissimilar images takes place in the visual cortex, resulting in a single three-dimensional image. By tracing the axons of the two optic nerves—which are made up of a million ganglion cells—and by using modern bra
in scan techniques to see what brain areas are activated by vision, perception researchers have been able to identify the intricate routing and processing of the incoming neural impulses. Omitting the bewildering details, suffice it to say that the impulses are split up and separated into thirty different pathways to areas of the visual cortex for pattern recognition (how things look), place recognition (where things are), color, and other characteristics. Then these and other arriving data are coordinated through a host of other pathways of the brain’s visual system to yield a final perception of a unified visual scene.4
Another interesting question, one of the most baffling, is how the image on the retina is viewed in the brain. Nerve impulses from the retina travel to the brain’s visual cortex, but what then? No screen exists in the brain on which they can be projected, so how is the incoming flow of data seen? And if it is displayed in some way there or elsewhere in the brain, who or what sees it? The question revives the ancient (and now thoroughly discredited) supposition that there is a homunculus or little man—the “I” of the mind—who perceives what arrives at the cortex. But if the homunculus is seeing that image, with what is it doing so? Eyes of some sort? Then who or what is looking at what arrives at the homunculus’s visual center? And so on, ad infinitum.
Allied to this puzzle is the question of visual memory. Every adult has an immense repertoire of images stored in his or her brain—familiar faces, houses, trees, leaves, cloud formations, beds slept in. They have been recorded, in some fashion, after even a single quick viewing. Though we cannot call all of them clearly to mind, it is by means of them that we recognize something we see a second time. In 1973 a Canadian psychologist, Lionel Standing, a man of great patience, showed ten thousand snapshots of miscellaneous subjects to volunteers at the rate of two thousand a day for five days. Later, when he showed them some of these pictures mixed in with new ones, they correctly identified two thirds of the old ones as pictures they had already seen.5 Where had they stored all the briefly seen images and in what form? When they saw a picture the second time, how did they locate and view the image in memory to compare it with the incoming one? Not by projecting the stored one on a cerebral screen, since none exists. And however they displayed it, what inside them looked at both the stored and incoming images—ah! there’s that troublesome little man again.
(Forget the little man and the screen he’s looking at. Research done in the past two decades has come up with a more realistic but more complicated answer, based in considerable part on studies of people with specific kinds of brain damage due, usually, to strokes. One woman, for instance, when asked to describe a banana, could say that it was a fruit and grows in southern climates but could not name its color. Another patient, asked to describe an elephant, said correctly that it had long legs but incorrectly that it had a neck that could reach the ground to pick things up.
(From many such studies, plus the results of brain scans showing what areas are activated by the effort to visualize something, it has become fairly clear that mental images are not located like filed pictures in any one or several places, but that the components of each image—its shape, its color, its texture, and so on—are filed away separately and that summoning up a mental image uses many of the same processes that perception itself does, calling up and coordinating these several elements into one final, more or less complete image. But not a pictorial image; just as the letters of this sentence symbolize things they don’t physically resemble, the patterns of firing of brain neurons represent objects and events in the outside world.6 Why did evolution devise this scheme? Let the evolutionary psychologists figure out that one.)
These are but a few of the mysteries of visual perception; perhaps no area of psychology has produced as much research data and as relatively few definitive answers. Some years ago, James J. Gibson, a controversial but noted perception theorist, flatly asserted that most of what perception researchers had learned in the past hundred years was “irrelevant and incidental to the practical business of perception.”7 A trifle more moderately, the perception psychologists Stephen M. Kosslyn and James R. Pomerantz said in 1977 that, despite all the accumulated data, perception is still poorly understood.8 Still, they added, “we do know some things about it.” And, they could add today, they now know a good deal more about it. Indeed, enough to begin to understand it, to answer at least some of the interesting questions, and to discard others in favor of more cogent ones.
Styles of Looking at Looking
For centuries, philosophers debated about whether we are born with mental equipment that makes sense of what we see (the Kantian or nativist view), or must learn from experience to interpret what we see (the Lockean or empiricist view). When psychology became experimental, the findings of perception research not only failed to answer the question but added to the evidence for each side. Although today the terms have been redefined and the hypotheses have become more sophisticated, the debate continues.
Locke, Berkeley, and other philosophers and psychologists sometimes fantasized a test case that would definitively resolve the issue: a person blind from birth who, through an operation or some other intervention, suddenly gains sight. Would he know, without touching what he was looking at, that the object was a cube rather than a sphere, a dog rather than a cat? Or would his perceptions be meaningless until he learned what they meant? Such a person’s experiences might hold the key.
In recent centuries a handful of such cases have, in fact, turned up. The most carefully reported was that of an Englishman with opaque corneas who, in the early 1960s, at the age of fifty-two was able to see for the first time.9 S.B., as he is called by Richard L. Gregory, a British psychologist and perception expert who studied him closely, was an active and intelligent man who had made a good adaptation to his blindness: He was skilled at reading Braille, made objects with tools, often chose to walk without the customary white cane even though he sometimes bumped into things, and would go bicycling with a friend holding his shoulder to guide him.
In S.B.’s middle years, corneal transplants became possible, and he underwent an operation. According to Gregory’s report, when the bandages were removed from his eyes, he heard the surgeon’s voice and turned toward what he knew must be a face. He saw only a blur.
Experience, however, rapidly clarified his perceptions: Within days he could see faces, walk along a hospital corridor without touching the walls, and recognize that the moving objects he saw through the window were cars and trucks. Spatial perception, however, came to him more slowly. For a while, he judged the distance to the ground below his hospital window to be such that he could touch it with his toes if he hung by his hands from the windowsill, although it was ten times that distance.
S.B. was soon able to identify at first sight articles he had known by touch, such as toys, but many objects that he had never touched were mysteries to him until he was told or discovered what they were. Gregory and a colleague took him to London, where he recognized most of the animals at the zoo because he had petted cats and dogs and knew how other animals differed from them. But in a science museum S.B. saw a lathe—a tool he had always wanted to use—and could make nothing of it until, with his eyes closed, he ran his hands over it. Then, opening his eyes and looking at it, he said, “Now that I’ve felt it, I can see.”
Interestingly, when Gregory showed S.B. some illusions, he failed to be misled by them; he did not, for instance, perceive the straight lines of the Hering illusion as curved or the parallel ones of the Zöllner as divergent. Such illusions evidently depend on one’s having learned cues that denote perspective, and those cues, given by the other lines in the illusions, meant nothing to S.B.
The conclusions one can draw from his case are thus disappointingly mixed; some of the evidence favors innateness, some, experience. Besides, the evidence is contaminated: S.B. had had a lifetime of sensory experiences and learning with which to interpret his first visual perceptions, and his story does not reveal the extent to which the mind, be
fore experience, is prepared to understand visual perceptions. Nor is the question answered by developmental research with infants, since it is unclear how much the development of an infant’s perceptual abilities at any juncture is due to maturation and how much to experience. Only impermissible experiments that would deprive an infant of perceptual and other sensory experience could tease the two apart and measure their relative influence.
Making a still worse muddle of the matter is the question of whether perception is primarily a physiological function or a mental one.
The founders of scientific psychology in the nineteenth century and the early decades of the present one tried to evade this issue by asserting that mind was unobservable and perhaps illusory, and by limiting themselves to the study of physical realities. Those who were interested in perception investigated the physiology of the sensory systems, especially the visual one, and over the course of more than a century, a number of them in Europe and America assembled a mass of data on the mechanics of that system. By the early years of the twentieth century they had determined that the retina of each eye, a thin sheet of specialized neural tissue, contains about 132 million photoreceptor cells of two types, rods and cones, both of which convert light into nerve impulses; that the rods, more common in the periphery of the retina, are more sensitive and respond only to very low levels of illumination; that the cones, more common in the center, respond at higher levels of illumination; and that there are three species of cone, one containing primarily chemicals that absorb light of short wavelengths (and thus react to blue and green), another, of middle wavelengths (green), and a third, of longer wavelengths (yellow, orange, and red).10