In the final portion of the chapter, we consider the effects of mirror-reversing a painting and, more generally, left and right in pictorial space. These effects suggest that some pictorial anisotropies, such as profile orientation, reflect the influence of lateralized brain functions, whereas others, such as the tendency to look at a picture from left to right, are cultural conventions.
Figure 6.1
Quartz crystals. From the Nobel Prize address of V. Prelog, 1976.
Figure 6.2
Horns of Marco Polo’s sheep (Thompson, 1969). Reprinted with the permission of Cambridge University Press.
Figure 6.3
Floor pattern from the palace at Tiryns, Greece. Note that the two sides of a vertebrate are enantiomorphs (Swindler, 1929). Used by permission of Yale University Press.
Figure 6.4
Stylized bear from the Tsimshian Indians of the north Pacific coast. Note that the right and left sides of this or any other pattern with a vertical axis of symmetry are enantiomorphs (Boas, 1955).
LEFT, RIGHT, AND COSMOLOGY
Is there a “left” and “right” in the universe? Newton said yes, and Leibnitz said no. Newton thought that the coordinates of space were absolute and “God-given.” Leibnitz attacked this view and argued that left and right were “in no way different from each other.”1 Kant was puzzled by enantiomorphs for decades and this led him to side with Newton.2 For Kant the difference between left and right mirror images was literally “inconceivable.” They could only be distinguished through intuition, through the a priori structure of the mind, and therefore, for Kant, the a priori left-right structure of the universe.3 Until very recently, however, cosmologists agreed with Leibnitz: the universe was symmetrical and right and left were arbitrary human conventions.
The problem was of strictly theoretical interest until many astronomers came to believe that there must be millions of inhabitable planets and therefore probably other beings at least as intelligent as humans. Distinguishing right from left then became of practical importance for developing a method of extraterrestrial communication. A digital code is clearly the best method, since it can be used to transmit pictures as well as messages. In order to decode pictures properly, it is necessary for the receiver to understand the instructions top and bottom, front and back, and left and right. Top and bottom can be described as “away” and “toward” with respect to the center of a planet. Front and back can be described as “near” and “far.” But how does one describe left and right to an extragalactic listener? Describing left and right requires the sender to point to and the receiver to look at one side of an object, but this is impossible intergalactically.4
The solution to this problem came in 1957 from an experiment by Chien-Shung Wu at Columbia. Her experiment shook the very foundations of modern physics; parity had fallen. Madame Wu studied the emission of electrons from cobalt-60, a radioactive isotope of cobalt. Normally, electrons are emitted in all directions from cobalt-60, but, when it is cooled down near absolute zero (–273° C) and placed in a strong magnetic field, the electrons would be expected to line up with the magnetic field and emerge equally from the two poles of the isotope nucleus. What Wu discovered was that more electrons came out from one side than the other of the otherwise uniform nucleus. Thus it was possible to label the poles of the magnetic field and therefore identify right and left in a consistent fashion. Right and left could now be given a meaning beyond human convention. Leibnitz was wrong: the universe is not symmetrical. Now, by describing Wu’s experiment to our extragalactic audience we could tell them about left and right, and about which hand most humans write with.5
WHY ARE MIRROR IMAGES CONFUSING?
Like philosophers and physicists, children and psychologists have been confused about left and right (and cobalt-60 is unlikely to help them). The interest of experimental psychologists in left—right confusion began with the nineteenth-century physicist-philosopher Ernst Mach, who noted that “children constantly confound the letters b and d and also p and q . . . [and] ...adults too do not readily notice a change from left to right.”6 Since then, there has been an enormous volume of literature on letter reversal and more generally on the confusion of mirror images. A practical spur to this research has been the problem of reading disability because of its frequent association with letter reversal. The theoretical interest in mirror-image confusion is that it occurs in a variety of species such as octopus, fish, rats, and monkeys as well as human infants, children, and adults. The ubiquity of mirror—image confusion must therefore reflect something fundamental about how the nervous system processes visual information.7
Mach had suggested that the bilateral symmetry of the brain and body of the perceiver underlay mirror-image confusion and that, to the extent that mirror images could be distinguished, “slight asymmetries . . . particularly in the brain” were responsible. In one version or another, essentially the same explanation of mirror-image confusion has been traditional in psychology ever since. The commonest version has been that of Samuel T. Orton, still the dominant name in the treatment of reading disorders.8 Orton thought that because the two cerebral hemispheres were mirror symmetric, a visual stimulus on the retina would produce patterns of “neural excitations” in each hemisphere that would be mirror images of each other. Therefore, in order for a stimulus to be distinguished from its mirror image, the hemisphere with the veridical excitations would have to “suppress” or “dominate” the hemisphere with the mirror-reversed excitations. Mirror-image confusion, according to Orton, was ascribable to “incomplete dominance” of one hemisphere by the other.
Recently, Corballis and Beale have produced a new version of the Mach–Orton idea that mirror-image confusion derives from the symmetry of the brain.9 Like Orton, they proposed that mirror-image confusion results from mirror-image representations of a stimulus in the two hemispheres. However, they realized that Orton’s claim that an asymmetric stimulus would produce enantiomorphic patterns of stimulation in the visual areas of each hemisphere is false and that, in fact, the pattern in the two visual areas would be veridical and identical. Instead, they suggest that after the “neural excitation” is stored as a topographic memory in each hemisphere, the “memory representation” would mirror reverse when it transferred across the midline to the opposite hemisphere. (They assume that symmetrical points in each hemisphere are interconnected.) Thus each hemisphere would contain memories of the stimulus in both its veridical and its mirror form. According to their hypothesis, since every stimulus is stored in each hemisphere in its original and mirror-reversed forms, the organism would treat the two enantiomorphs as equivalent and therefore confuse them.
There are several serious difficulties with the various versions of the idea that mirror-image confusion derives from brain symmetry. One is presented by Gerstmann’s disease. This disorder is characterized by extreme left-right confusion and follows damage to the left parietal lobe of the brain.10 In this case an abnormally asymmetric brain increases left-right confusion rather than reducing it, as would follow from the Mach–Orton– Corballis hypothesis. Another argument against the role of brain symmetry comes from a comparison of children and animals. Both have severe difficulty in mirror-image discriminations, but the two hemispheres in children are functionally and anatomically asymmetrical, while in animals (at least below great apes) they appear to be relatively symmetrical.11
As mentioned above, Corballis and Beale realized that the initial representation of a stimulus in the two hemispheres would be veridical, not reversed as Orton had thought, and they suggested that the reversal would occur at a subsequent “memory” stage. However, this idea is also unsupportable by modern visual anatomy and physiology. The left visual half-field is represented in the right hemisphere and the right visual half-field in the left hemisphere as shown in figure 6.5. Moreover, there are multiple representations of the left half of space in the right hemisphere and of the right half of space in the left hemisphere.12 However, the only anatom
ical connections between the representations of the left field in one hemisphere and the right in the other are between the representations of a narrow midline strip of the visual field.13 This interconnection does not carry information about integrated visual patterns.14 At this stage there are no longer any maps of visual space to be reversed. (However, the absence of evidence for an anatomical mechanism for interhemispheric reversal of memory traces is not a conclusive argument against the possibility of such a mechanism. After all, we learn and remember, and the anatomical bases of these phenomena are still totally obscure.)
More damaging for the Corballis and Beale hypothesis is that attempts with both humans and monkeys to demonstrate mirror reversal in interhemispheric memory transfer have failed.15 That is, if a visual stimulus is presented to one hemisphere and the person or monkey is required to match it with a stimulus presented to the other hemisphere, the two hemispheres invariably match the stimuli accurately. They do not equate the stimulus in one hemisphere with its mirror image in the other hemisphere, as would be predicted from the Corballis and Beale idea of “interhemispheric image reversal.”
Figure 6.5
Diagram of the visual pathways from Ramón y Cajal’s classic (1999). The labels have been added. Note that the optics of the eye reverse the image of the arrow on the retinae. The nerve fibers from each retina separate so that messages from the left half of each retina travel to the visual cortex of the left hemisphere, and the messages from the right halves travel to the visual cortex of the right hemisphere. Thus when the center of the arrow is fixated (as shown), information in the left half of space (the arrow head) goes to the right cortex, and information in the right half of space (the feathers) goes to the left cortex. Note further that the two cortical representations are not mirror-reversed with respect to each other.
Finally, a general objection to both Orton and Corballis and Beale is that the existence of a topographic representation of a stimulus (“a picture”) in the brain, whether veridical or reversed, does not constitute an explanation of perception. “Seeing is an interpretive process not a representational one.”16
In summary, the Mach–Orton–Corballis hypothesis that mirror-image confusion derives from the symmetry of the body and brain does not work. Neither Orton’s “incomplete dominance” nor Corballis’s “memory reversal” fits the known facts.
AN EVOLUTIONARY HYPOTHESIS
If the symmetry of the body cannot explain mirror-image confusion, what can? We propose that the answer lies in the nature of the world and the evolution of the vertebrate visual system within it.
The selective pressure of evolution made it advantageous for the visual system to be able to perform certain types of visual processing while other types were unnecessary for survival. In the natural world there are never any mirror images that would be useful for an animal to distinguish. Indeed, with two exceptions, there are virtually never any mirror images at all. One exception is the two sides of a face or, more generally, the two sides of another bilaterally symmetrical animal. But here the two sides are aspects of the same thing, and it would be more adaptive to treat them as the same—not to distinguish them. Another exception is that a silhouette viewed from the back is the mirror image of the same silhouette viewed from the front. Again it would be adaptive to equate, not distinguish, these mirror images. In other words, we propose that the confusion of mirror images is not a “confusion” per se but an adaptive mode of processing visual information. In the natural world the only mirror images that ever occur are aspects of the same thing and therefore need not be distinguished. Rather, it is adaptive to treat enantiomorphs as equivalent to each other. Instead of the “confusion” of mirror images we can speak of their perceptual equivalence.
To summarize our argument, mirror images are not confused because of the symmetry of the brain. Rather, mirror images are treated as equivalent because in the natural world mirror images are almost always aspects of the same thing.
READING, MIRROR IMAGES, AND DYSLEXIA
The perceptual equivalence of mirror images poses a problem only for humans and only when, for example, they have to learn to distinguish b from d and to write s and not its lateral mirror image in the Latin alphabet. Then children must learn to overcome the natural mode of processing that evolution built in to the brain, namely the perceptual equating of mirror images. On this view, mirror-image confusion is not unique to childhood except in the sense that it is as a child that a person usually has to overcome it. Thus, we would expect that nonliterate adults should show the same confusion as children. Corballis and Beale inadvertently confirm our prediction in reproducing the medieval woodcut shown in figure 6.6.
Reading is a complex skill, and children may have difficulty in learning to read for a great variety of reasons including poor instruction, antagonism between cultures of school and home, sensory and neurological disorders, impaired language development, emotional disturbances, and mental retardation. However, there is a small proportion of children for whom none of these conditions is present and yet they have severe and persistent difficulty in learning to read. Their condition is known as developmental or specific dyslexia or more simply “reading failure of unknown origin.” Specific dyslexics are children of normal intelligence, vision, hearing, motivation, and oral language development who have inexplicable reading problems. These dyslexics can be taught to read eventually, but difficulty in reading, spelling, and learning foreign languages usually persists into adulthood.17
Figure 6.6
Francis I Offers His Heart to Eleonore of Austria, by an anonymous French master, ca. 1536 (Corballis and Beale, 1976). Note that two Ns and two Ss are reversed in the print. Letters in the block for this woodcut should have been incised originally in reversed form. Shapiro (1970) observed that S and N are often reversed in early Medieval Latin inscriptions, as they are by children and unpracticed adults.
Reversal of letters in place (b and d ), reversal of letters in a word (on for no), and failure to progress consistently from left to right are common errors in learning to read and write. However, they are even more common among dyslexic children. Although not universal, mirror reversals and left-right difficulties appear to be the principal distinguishing features of specific dyslexics other than their reading disability.18 Can our explanation of mirror-image confusion also help explain specific dyslexia? Before suggesting how it may, it will be useful to review the difference between the left and right sides of the brain.
It is now well established that the two hemispheres of the human brain are each specialized for different psychological functions.19 In virtually all right-handed people the left hemisphere is specialized for language. Therefore it is often termed the dominant hemisphere; but, strictly speaking, it is dominant only for language.20 The right hemisphere is not just a nonlinguistic version of the left. Rather, the right hemisphere is specialized for and better than the left hemisphere at a range of perceptual functions that do not involve language. These include visual matching, memory for abstract designs, copying and drawing, face recognition, construction of block designs, and map reading. It seems likely, therefore, that the perceptual equating of mirror images is also a right-hemisphere function.21
Thus, specific developmental dyslexia may not reflect a deficit or dysfunction. Rather, we suggest it may indicate a relatively greater importance or dominance of the nonlanguage hemisphere (the right hemisphere in right-handers). This would result in slower acquisition of reading, not only because reading is a left-hemisphere (language) skill but also because mirror-image equivalence would be particularly strong or “good.” A strong tendency to treat mirror images as equivalent would manifest itself in reversal of individual letters, in reversal of the order of letters in a word, and reversed scanning of phrases and lines: all classic symptoms of developmental dyslexia. According to this view, specific dyslexics should be superior to normal children on those perceptual, and perhaps artistic, skills that do not involve language, with one exception. T
he exception is in the discrimination of mirror images: On such tasks they would be much worse than normal children, since we are postulating that mirror-image equivalence (“confusion”) signifies the predominance of the nonlanguage hemisphere.
The repeated reports that dyslexics are unusually artistic and better at drawing even before their reading disability manifests itself support this interpretation.22 Unfortunately, there are very few studies that provide an adequate test of our hypothesis, because most are directed toward looking for deficits rather than superiorities among dyslexics. Furthermore, when perceptual abilities are examined, the results of perceptual tests dependent on language or on right-left discrimination are usually reported together with the results of tests that are not; we would expect dyslexics to be superior only on the latter ones. Finally, most studies of dyslexics include children who clearly have other difficulties in addition to learning to read, or they fail to compare these children with children of matched intelligence, age, and background. Thus it is not surprising that among studies of visuoperceptual abilities in dyslexia, most find dyslexics “unimpaired,” a few find deficits, and only a very few find superiorities.23
If correct, our suggestion that specific dyslexics are superior at a variety of non-language-related perceptual skills may be relevant to techniques of teaching reading. In any case, to view dyslexics as potentially superior in so many other skills than reading and to encourage and develop this potential would certainly reduce the devastating emotional and social consequences of reading disability, which often are more crippling than the disability itself. It may be both more accurate and more helpful to view specific dyslexia as a difference rather than a deficit.
A Hole in the Head Page 11