The Story of Psychology

by Morton Hunt

  They had also traced out much of the complex wiring scheme by which the rods and cones send their impulses to the brain. From the retinas, the bundles of optic nerve fibers make their way to the visual cortex, an area at the lower part of the back of the brain. En route, the fibers carrying messages from the left half and from the right half of each eye’s field of vision are sorted out and redirected; the messages from each eye’s right-hand field of vision end up at the left visual cortex, the left-hand field of vision at the right visual cortex. (To this day, no one has the least idea why evolution arranged this crisscross.)

  Many psychologists were long reluctant to accept the evidence that visual functions are centered in the visual cortex; such localization smacked of phrenology. Late in the nineteenth century, however, brain localization—not of the phrenological type, and only of certain functions—gained new credibility after Wernicke and Broca discovered that speech functions are carried out in two small areas in the left half of the brain. This inspired researchers to look for an area where visual messages are received and understood, and through autopsies of brain-damaged human beings and operations on monkeys they identified it, in general terms, as the rear of the brain.

  More precise pinpointing of the visual cortex was a byproduct of weaponry used in the Russo-Japanese War of 1904–1905.11 In that conflict the Russians introduced a new rifle, the Mosin-Nagant Model 91, which fired bullets of smaller diameter and higher velocity than the rifles of earlier wars. The bullets often penetrated the skull without shattering it, and in some cases destroyed, partly or totally, the victim’s vision without killing him. Tatsuji Inouye, a young Japanese army doctor who worked with wounded soldiers, plotted the extent to which each patient’s visual field had been lost by each eye, determined from the site of the bullet’s entry and exit which parts of the brain had been damaged, and, putting these data together, identified the precise location and extent of the visual cortex.

  Among his findings was that the areas of the visual cortex that receive the retinal messages are grossly disproportionate to the areas of the retinal image. A very large part receives impulses coming from the fovea, the small central area of the retina where vision is sharpest, and only a small part from the larger peripheral area. (Later research showed the disparity in proportions to be about 35 to 1.12) That settled one great issue: what arrives at the brain is in no way an image corresponding in layout to the image on the retina.

  The inescapable implication of Inouye’s and others’ findings, gradually accepted over the next several decades, was that the retinal cells are “transducers” that change light signals into a different form of energy— bursts of nerve impulse—and that these “coded” impulses or signals, when received in the brain, are not turned back into images in the visual cortex, even though “seen” there or elsewhere in the brain.13 How they are seen remained a mystery, but perception physiologists did not trouble themselves with this question; their style of looking at looking dealt with the flow of neural impulses and stopped short at the borders of mind.

  Another style of so-called perception research—it was only peripheral to perception—was in the Wundtian tradition. Its practitioners studied sensations (the immediate simple responses to sounds, lights, and touches), which they considered reflexive, elemental, and scientifically investigable, and the perception of those simple sensations. But they ignored all the complex interpretive aspects of perception, which they correctly deemed the result of the mind’s processing of sensations and incorrectly believed to be beyond objective scrutiny. This approach, popular in the early part of the twentieth century, yielded a huge store of data about sensations but added almost nothing to the understanding of the psychology of perception.

  Yet another style of perception research is psychophysics, which also stops short of looking at mental processes. Fechner and his followers, as we saw, measured sensory thresholds (the weakest sound, light, or other stimulus a subject could perceive) and “just noticeable differences” between pairs of stimuli. While such inquiries touched on conscious mental processes, the psychophysicists asked no questions about how the subject noticed a stimulus or judged differences but stuck to objective data—the magnitude of the stimuli and the subject’s statements that he did or did not perceive a stimulus or a difference between two stimuli. Psychophysics was therefore acceptable during the dominance of behaviorism, when perception was otherwise largely ignored because it assumes that a representation of the world exists in the mind, a concept the behaviorists rejected.14

  But psychophysics was plagued by a chronic problem: subjects were inconsistent in their responses. If given the same threshold stimulus a number of times, sometimes they would see or hear it, sometimes not. If a light at an intensity below a subject’s threshold was gradually increased, he might begin to see it at a given level, but if it was presented above that threshold and gradually decreased, he might cease to see it at a somewhat different level.15

  To solve this problem, in 1961 the psychologist J. A. Swets proposed applying to psychophysics the engineering concepts of signal detectability and information theory, which psychologists had come into contact with during World War II. Swets and his collaborators even gave their approach a name betokening the impersonality and objectivity of engineering—the Theory of Signal Detection. It held, first, that there would always be some random variation in the number of neurons excited by any signal and in the amount of “noise” (extraneous or accidental excitations) entering the neural system, and it corrected for these variations by statistical theory. It held, second, that a subject’s response on any trial is partly determined by his expectations and his effort to maximize his rewards and minimize his costs; these variables could be accounted for by decision-making theory.

  Although “decision making” sounds mental, the Theory of Signal Detection remains outside the mind; it predicts the likelihood of correct and incorrect responses according to purely mathematical parameters. Signal detection theory was a major advance in psychophysics and is a part of the standard repertoire of experimental methods today, but it concerns only certain objective results of perception and casts no light on how perception takes place.16

  All along, however, a small number of psychologists had been exploring the internal or cognitive aspects of perception. They were mentalists, but not in the metaphysical sense; rather, in the tradition of James, Freud, and Binet, they believed that higher mental processes are the heart of psychology and can be experimentally investigated.

  In 1897, even as Thorndike and others were beginning to turn toward animal experimentation and what would become behaviorist psychology, an American psychologist named George Stratton undertook a perception experiment of a human and distinctly cognitive kind. For a week, allowing himself no respite, he wore lenses that turned his view of the world upside down. At first he had so much difficulty getting about and reaching for objects that he would often close his eyes and rely on touch and memory. But by the fifth day he was automatically making the right movements and by the end of the week felt that things were where he saw them and even, at times, that they “seemed upright rather than inverted.” When at last he removed the lenses, everything was bewildering. For several hours he found that he was reaching for objects in the wrong direction; then he relearned where things were when seen normally. The experiment dramatically showed that spatial perception, at least in human beings, is in some part learned and can be relearned.17

  Striking as these findings were, the outlook of most psychologists in the early decades of the century was so antimentalist that there was little follow up of Stratton’s work and almost no cognitively oriented perception research until half a century later. But by the 1940s several unrelated strains of cognitively oriented psychology—Freudian, Gestalt, personality research, and the nascent discipline of social psychology— were gaining strength, and psychologists who found any of them congenial took a wholly different approach to perception research from that of psychophysiologists an
d psychophysicists.

  Some, in America and elsewhere, rediscovered Stratton’s work and conducted new optical-distortion experiments. In 1951, Ivo Kohler, an Austrian psychologist, persuaded volunteers to spend fifty days seeing the world through prism goggles that displaced part of their visual field 10 degrees to the right and made straight vertical lines slightly curved. For days his subjects found the world unstable and had difficulty walking and performing even simple tasks, but after a week to ten days most things looked normal to them, and after a few weeks one volunteer was even able to ski. Like Stratton, they felt disoriented when they finally removed the prisms but soon adapted to normal vision.18

  Other psychologists revived the long-neglected study of illusions, and by the 1950s it was again a topic of active research. The remarkable subjective triangle shown in FIGURE 25, created in 1950 by Gaetano Kanizsa, an Italian psychologist, was only one of many new illusions used to investigate mental processes of vision. A special kind of illusion was used to explore the mind’s interpretation of ambiguous figures. The following classic example, created in 1930 by Boring, can be seen at will either as an old hag turned partly toward the viewer or as a young woman turned partly away.

  FIGURE 26

  What kind of woman is this? That depends on what you choose to see.

  The capacity to see either of two different images in ambiguous figures like this one or in figure-ground reversal patterns like the Rubin vase (Figure 12) cannot be explained by any known physiological mechanism, the British psychologist Stuart Anstis has said, but is the result of higher perceptual processes.*19 The same is true of the mind’s acceptance of, or bewilderment by, the “impossible objects” created by perception researchers in the 1940s and 1950s, of which these are classic examples:

  FIGURE 27

  Two “impossible objects”

  It is the mind, not the retina, optic nerves, or specialized cells of the neural cortex, that interprets the cues in such a figure as a picture of an object and simultaneously realizes that no such object could exist in the real world.

  Another cognitive approach to perception was that of a number of American psychologists who, beginning in the 1940s, sought to discover how needs, motivations, and mental sets affect perception. Jerome Bruner and Leo Postman of Harvard, two leaders in this endeavor, showed toys and plain blocks, all three inches in height, to young children, and asked them to judge the objects’ size; the children thought the toys were taller. In an extension of the experiment, they told the children they would get to keep the toys, but temporarily broke their promise. When the toys seemed unavailable, the children judged them as even larger than they had previously. Other researchers asked hungry and not-hungry subjects to estimate the size of food items; the hungry ones saw them as bigger than did the others. These and similar experiments demonstrated that need, desire, and frustration influence perception.20

  So do certain traits of personality, according to other studies of that era. By means of a written test and an interview, Else Frenkel-Brunswik, a psychologist who was trained in Vienna and had immigrated to America, rated a group of children on ethnic prejudice, a trait she associated with the rigid “authoritarian personality pattern.” She then showed the children a picture of a dog, followed by a series of transitional pictures in which the image gradually became that of a cat. Those who had scored high on prejudice tended to see the image as a dog longer than the unprejudiced, who were more flexible. Much the same was true when she asked children to identify the color of a series of cards that changed from one hue to another.21

  Other cognitive studies of perception in the 1940s and 1950s explored “perceptual defenses”—forms of mental resistance to seeing something disturbing. Researchers used tachistoscopes to flash words on a screen very briefly (for a hundredth of a second or so), and found that subjects were less likely to recognize taboo words than neutral ones. The effect was strongest when the subjects were females and the experimenter male. One team used a tachistoscope to display achievement-related words like “compete” and “mastery,” and neutral words like “window” and “article”; subjects who had a strong need for achievement, as measured by Henry Murray’s TAT, could read the achievement-related words more quickly and easily than the neutral ones.22

  Mental set, or the expectation of what one might see, was another topic of this kind of research. Bruner and Postman used the tachistoscope to show subjects very brief views of playing cards, most of which were standard but some of which were not, like a red four of spades. Habit and expectation caused twenty-seven of their twenty-eight subjects to see the abnormal cards as normal, but once the subjects knew about the cards, their mental set was changed and they made far fewer incorrect identifications.23

  By 1949 such studies had become so numerous that psychologists, borrowing a term then current in women’s fashions, spoke of the New Look in perception research. For about a decade, the New Look flourished, amassing data on the extent to which needs, motives, memory, and mental sets affect perception. Then, lacking detailed theory with which to explain the processes through which this took place, the movement lost steam. Much later, perception researchers came up with a cognitive description (rather than a theory) of visual recognition processing, lumpily called “bottom-up, top-down,” that made sense of the accumulated data. In bottom-up processing the mind assembles the various bits of data coming in, achieving higher levels of recognition and meaning as it sees them fit together. But top-down processing, drawing on stored memory, context, and the like, may influence the lower levels of perception, either by making ambiguous information clear or by inducing a conception—or misconception—of what is being reported at the lower level. An often-cited example is the middle letter in each of these words:

  FIGURE 28

  Bottom-up processing alone would leave us unsure what we were seeing; context—top-down processing—leads us to see the first one as an H, the second one as an A, although they are identical. Similarly, in the ambiguous images we saw above in FIGURE 24, it is whatever top-down influence we choose to exert that yields what we finally see.)

  After the New Look petered out, perception research revived with the advent of a new and powerful theory, information processing, which in the 1960s and 1970s was beginning to transform cognitive psychology with its vision of an orderly series of processes by which sensations are transformed into thought, and thought to action. This theory postulates (and provides experimental evidence of) the metamorphosis of sensory input in a sequence of steps, including very brief storage in the sense organ, encoding into nerve impulses, short term memory storage in the mind, rehearsal or linkage with known material, long-term memory storage, retrieval, and so on. The theory made it possible for psychologists to be specific about how the mind handles incoming sensory material, and it revived interest in the cognitive approach to perception. By the 1970s research in the field of cognition was proliferating, as we will see in a later chapter.

  But by then many significant discoveries had been made about the physiology of perception. Ever since, the two styles of looking at looking, the physiological and the cognitive, have existed side by side, seemingly opposed to each other but in reality focused on different aspects of the same phenomena, as discussed from here on.

  Seeing Form

  How do we see the shapes of things? The question may seem absurd— how could we not see them? But the perception of form is neither automatic nor foolproof. We see a shadowy figure in the park at night and cannot tell whether it is a bush or a lurking person; we read a carelessly scrawled signature and do not know whether it starts with C, G, or O; we arrive home exhausted after a long flight, spot our car in the vast airport parking lot, and trudge toward it, only to find as we draw near that it is a lookalike of another make; we enjoy a jigsaw puzzle precisely because we find it challenging and rewarding to locate the piece that will fit into the edge we have just created.

  Research on form perception seeks to identify the mechanisms, both
neural and cognitive, that enable us to recognize shapes—and that sometimes fail us. Much of that research in the past half century has taken the cognitive approach. The Gestaltists and their followers explored the mind’s tendency to group related elements into coherent forms, fill in gaps in what we see, distinguish figures from background, and so on. They and others also said it was inborn higher mental processes that account for the “constancy phenomenon”—our seeing things as unchanged despite distortions of the retinal image, as when we perceive a book lying at an angle to us as having square corners although on the retina, as in a photograph, the book is a rhomboid with two acute and two obtuse angles.

  But such perceptions are results, not processes. By what steps does the mind achieve them? It is one thing to say that we fill in gaps in familiar but incomplete forms that we see; it is quite another to determine by what specific means we achieve this. Many studies exploring cognitive processing of visual information in fine detail have identified some of them. A few examples of the findings:

  —Research on the subjective-contour phenomenon (as in the illusory triangle in FIGURE 25 above) indicates that we create the imagined contours partly through association (the three angles remind us of previously seen triangles) and partly through clues that experience has taught us signify interposition (an object’s obstructing our view of another one). As the perception researcher Stanley Coren pointed out, the gaps in the circles and in the existing triangle suggest that something else—the illusory triangle—is in the way, partially obscuring them. Because of the apparent interposition, the mind “sees” the imaginary triangle.24