On the back surface of each of your eyes, there is a scrawny, upside-down image with a hole in it, containing about two thirds of the scene in front of you. Over the first fraction of a second after light from the world reaches your eyes, this image is transformed by your brain into a single stable and richly detailed perception of the three-dimensional world. It is this perception that lets you hit a baseball, recognize your friends even at sunset, or upside down, and detect that someone is looking straight at you from across the room.
In spite of its enormous complexity, vision is one of the functions of the brain that modern neuroscientists understand best. We can describe, in astonishing detail, how patterns of light are transformed into electrical signals by photoreceptors in the eye, and how these electrical signals are transformed, reorganized and passed on, step by step, in the brain. A huge segment of the back of the human brain is devoted to this visual processing. Visual information first reaches the cortex at a region called V1, for example. The V1 region on the left side of the brain contains neurons that represent the right half of the visual scene; V1 neurons on the right represents the left half of the scene. In a mature brain, these neurons are laid out in a regular, but incredibly sophisticated, grid based on (for example) where in the scene the light is coming from (top versus bottom, centre versus periphery), whether the neuron can discriminate colours or light or only shapes, and the specific patterns of light the neuron “prefers,” such as vertical lines, horizontal lines, line junctions, and so on.
Our highly detailed models of visual cortex allow us, among other things, to make sense of the losses that follow different kinds of stroke. Damage to a specific piece of V1, for example, leaves people blind to precisely one quarter (for example, the top left) of the visual scene. Other patterns of damage destroy just the person’s ability to see the world in colour, or to see motion, or to recognize faces. Our models of visual cortex predict exactly these patterns.
Given these successes, it is tempting to conclude that V1 is for vision. So when a blind woman arrived at the hospital with an enormous stroke, destroying V1 on both sides of her head, her neurologists initially thought that she was extremely lucky. Most people with this kind of stroke would be blind for the rest of their lives, but she had been blind from birth, so she had nothing to lose. The neurologists were wrong. Following her stroke, the woman lost, forever, the ability to read Braille.
Why would damage to V1—a visual area—impair a person’s ability to read Braille with her fingers? The simple answer is that the brain is enormously plastic, in the technical sense: it is moulded by experience. So if you deprive the brain of visual experience, normally visual regions will take on other functions—including, in this case, reading Braille.
Brain plasticity is the topic of Norman Doidge’s new book, The Brain That Changes Itself: Stories of Personal Triumph from the Frontiers of Brain Science. About half of the chapters describe the basic science of plasticity, profiling major discoveries from the last 50 years and the scientists responsible for those discoveries. Doidge himself is a psychiatrist, so the other half of the chapters contain his reflections on how the idea of plasticity could be translated into psychiatric therapies. The whole book is an easy read. Doidge’s prose is lucid and fast-paced, and he introduces technical ideas carefully, clearly and only when necessary. His overall message is upbeat: because of brain plasticity, damaged brains can be healed, aging brains can be rejuvenated and even normal healthy brains can be made faster and better.
Chapter one, for example, describes Dr. Paul Bach-y-Rita’s inventions that use the brain’s natural plasticity to achieve semi-miraculous results like giving sight to the blind. These machines—called sensory substitution devices—basically do the job of the eye, transforming the pattern of light from a scene (received by a camera) into a pattern of electrical signals on a grid. To get these electrical signals to the brain, though, Bach-y-Rita just laid the grid of receivers on a patch of the person’s skin: on their back, or their forehead, or their tongue. At first, of course, the person just feels patterns of tingling on their skin. With enough practice, though, people stop feeling the tingling and begin, instead, to perceive the scene in front of them. In fact, Bach-y-Rita reported that their perception is so good that these blind people can now pick up a bat and use it to hit a moving ball.
Although these results are amazing, the basic mechanism at work is the same as it was for the blind woman who used her V1 to read Braille, and as it is in every baby’s brain during development. In the presence of systematic patterns of input, the brain organizes itself to process those patterns. Perception of the world depends less on the “channel” by which information gets to the brain, and more on the systematic patterns in the information itself.
This principle shows up all over the brain. Another example comes from the study of sign language. Sign language is completely different from spoken language on the surface. Sign language arrives in the brain through the eyes instead of the ears, as patterns in space instead of in pitch. Nevertheless, the information in sign language is processed by the same brain regions as auditory language, not by the brain regions that are normally responsible for visual recognition. (This was initially discovered by Helen Neville, a Canadian neuroscientist now at the University of Oregon.)
One chapter of Doidge’s book describes another dramatic example: the phenomenon of phantom limbs. When a person’s arm or leg is amputated, the neurons in the brain that used to receive sensory messages (pressure, heat and pain) from that limb don’t get any input anymore. In the absence of regular input, this brain region starts to borrow inputs from its neighbours. In the sensory cortex, for example, the region representing the person’s arm is right next to one representing the face, so following amputation, stroking the person’s cheek can send input to “hand” neurons, causing tingling in the hand that is no longer there.
Sometimes the tingling in a phantom limb escalates to chronic intense pain that feels like a muscle cramp—but how do you release a muscle cramp in a muscle that does not exist? Impressively, in this case, where brain plasticity is part of the disease, it can also be part of the cure. In cases of phantom limb pain, the brain is falsely interpreting activity in, in our hand example, the hand pain neurons as evidence of damage to the hand, so the trick is to counteract that interpretation. Dr. V.S. Ramachandran’s solution was deceptively simple: he used a mirror. The mirror reflects the person’s remaining hand at just the right angle, providing a visual illusion of the missing hand. By gently stretching the mirror-hand, the patient could then provide visual evidence, to his or her own brain, that the missing hand is healthy. Eventually, the visual evidence overrides the sensory interpretation and the pain (in the best-case scenario) goes away.
Phantom limb pain and its cure are both evidence of the same basic principle: brain cells, even in adults, can change their functions in response to changes in their input and experience.
This message is particularly important for the patients, the families and the doctors struggling with recovery from stroke. According to Doidge, the old dogma was that after the first few months of acute recovery, the damaged brain could no longer be helped. Whatever loss of function persisted into the chronic phase would last forever; there was no point in continuing with rehabilitation. That old dogma is falling. Inspired by evidence of plasticity in the adult brains, many hospitals are experimenting with intensive training to help patients recover functions months and years after their stroke—and it is working.
The best things about Doidge’s book are his enthusiastic, readable introductions to the science of plasticity and the message of hope they convey. Unfortunately, in his enthusiasm, Doidge’s account of modern neuroscience has become lopsided. The new science of brain plasticity is only half of the story; the other half of the story is the biology of the brain.
To understand the more complete story, it will help to work through a specific, and well-characterized, example: Mriganka Sur’s “re-wiring” experiments on ferrets. Doidge mentions these experiments briefly, but they deserve more attention. As in humans, ferrets’ brains contain separate channels by which auditory information from the ears and visual information from the eyes normally get to the brain. By very precise surgery in a baby ferret, the nerves from the eye can be induced to grow into the normal auditory channel, so the patch of brain that normally receives signals from the ears now instead receives information from the eyes. When these ferrets are a little bit older, they are taught to look for food near a red light but not a green light. What is amazing is that if the ferrets’ normal visual brain regions are then damaged, in a way that would normally cause total blindness, they continue to perform normally on this visual task—relying exclusively on their rewired “auditory” region. Only if that region of the cortex is damaged do the ferrets become blind, just like the blind women became Braille-blind when her visual cortex was damaged.
Critically, Sur can then study, in microscopic detail, what happened to the auditory cortex that was rewired during development. In a normal ferret or human brain, visual regions are organized in an intricate grid depending on properties of the visual signal, such as the location, shape and colour of the light; auditory regions also are organized in a grid, but organized by properties that are relevant for hearing sound, such as pitch, harmony and rhythm. So what would an auditory region look like, one that developed normally until birth but then received visual inputs? The answer predicted by plasticity is that this rewired auditory region would now be organized like visual cortex—by location, shape and colour of light; and to some degree, that is what happens.
Only to some degree, though. Rewired auditory cortex does contain patches of neurons with preferences for horizontal versus vertical lines; objects at the centre of vision are represented at one end, and objects on the periphery get a response at the other end. On the other hand, the details are different. Rewired auditory cortex retains some elements of auditory structure, like long parallel lines of neurons with similar properties, and the grid of visual properties is more loosely packed than in real visual cortex.
The take-home message is this: the structure and function of a specific brain region are a consequence of the interaction between the patterns of environmental input and biological preparation and control. This interaction explains how rewired auditory cortex successfully processes visual information, but also why rewired auditory cortex still retains features of normal auditory cortex even in the absence of any experience of sound. Of course, the take-home message is not really surprising. Every time biologists ask whether some property of an animal develops under the control of the environment or of biology, the answer is always both. The real question is how does the interaction work, in detail, in this context. Why does auditory cortex retain some properties of an auditory region, and not others?
The same questions apply to the examples that Doidge profiles. The changes underlying phantom limbs are local: sensory inputs from the face colonize the hand region right next door. If plasticity were the whole story, the newly released hand neurons could take on any new role, processing any pattern in the brain’s input, but they do not. The plasticity at work in phantom limbs is conservative, local and bound by prior biological constraints.
Similarly, in the woman who was blind from birth, there are a lot of different jobs that could be assigned to her out-of-work visual cortex. Why does it always become responsible for language and Braille reading? Could some kind of training or medical intervention cause a blind child’s V1 to become assigned to a different function? We don’t know.
Doidge does not tackle these questions. After harshly criticizing neuroscientists for too much attachment to biological factors and too little confidence in plasticity, he himself goes to the opposite extreme.
We do not really know yet how far plasticity can change an adult brain, how much damage can be healed or in what ways the “recovered” functions of plasticity will mimic, or not, the original structure of the brain. But biological limitations will mean that brain plasticity is not the panacea that it sounds like, in Doidge’s treatment. Some kinds of brain reorganization are possible. Other kinds are not. Damaged brains can be healed in some cases; older brains can be rejuvenated to some degree. The real story does not have the utopian ring of Doidge’s claims, but it is very exciting nonetheless. There is so much we still do not know about the limits, and what is possible; many scientists all over the world are working on it right now.
In all, Doidge’s zealous enthusiasm for modern neuroscience is gratifying, and infectious—and he is right: brain plasticity is a hot topic these days. The potential both to discover more about the basic mechanisms of the brain and to develop better medical applications seems enormous. Doidge’s book is a great introduction. But his vision of the science, and of the scientists, lacks nuance. In the profiles he provides, a few exceptional young men make revolutionary discoveries in a week or two, toppling all the old truths, and then preside over the next generation. Science is really much more inclusive, more complex and more fun than that.