Читать книгу A Symphony in the Brain - Jim Robbins - Страница 12
ОглавлениеCHAPTER TWO
That Special Rhythm
Tom was an eccentric young man, and it fell to Barry Sterman, who was a graduate student in psychology at the University of California, Los Angeles, to tutor him. Test scores showed that Tom, an eleventh-grader, was extremely intelligent, yet he could manage schoolwork at only an eighth-grade level. He had an odd, stiff gait, his face was pale, and he seldom showed emotion. When he spoke, it was in single syllables, such as “hi” or “yeah.” A staff psychologist at the learning center had diagnosed Tom as having schizophrenic tendencies and warned Sterman that he could be delusional. This tutoring experience would ultimately lead Sterman to a series of discoveries that transformed our understanding of the power of the brain.
At the time, Sterman was in the midst of a biology class on endocrinology, the study of the body's pituitary, thyroid, and other glands and the essential functions, physical and emotional, that they fulfill. “I was studying for my midterm, and part of that was thyroid function, and all of this stuff was running around in my head,” Sterman says. “So I came from class, took a look at this kid, and said, ‘My God, he's a walking definition of hypothyroidism.’” He recommended to the young man's parents that Tom have a basal metabolic rate test. A few days into a prescription for thyroid medication, Tom walked in for his tutoring appointment and called out with a smile, “Hi, Barry! What are we going to be doing today?”
Sterman was floored, and as he recounts the episode that happened more than four decades ago, he is still in awe of the difference in the teenager. “He had color in his face, he was less rigid, and he talked to me like a regular kid. It was a defining moment. I knew then I had to study physiology if I was going to work with human beings.” Psychology by itself, he felt, just wasn't enough, for the physiology of the brain and the body were somehow intertwined with thoughts and emotions. It was an idea that was ahead of its time, for the Cartesian notion of a separate mind and body was still firmly entrenched. After he received an undergraduate degree in psychology, Sterman studied neurology and psychology at UCLA for his Ph.D. and did his dissertation in sleep research. (Later, as part of his postdoctoral work, he taught psychology at Yale, in 1964.) After he earned his Ph.D., Sterman accepted a job as a sleep researcher at the Veterans Administration Hospital at Sepulveda, California. It was here, doing sleep research on cats and monkeys in a two-story, red-brick laboratory, that a couple of fortuitous accidents would lead Sterman to the discovery of the power of teaching the brain to produce certain frequencies.
Now in his seventies, Maurice “Barry” Sterman is an affable man with a keen sense of humor and a deep, raspy smoker's voice. His career is a distinguished one, and he's well respected by his peers. He is a professor emeritus in the departments of Neurobiology and Psychiatry at UCLA. He taught physiology at the School of Medicine and the School of Dentistry, and also taught seminars on inhibition in the central nervous system and a survey of the neurosciences at the university's prestigious Brain Research Institute. Sterman is also a member of the UCLA Academic Senate as a representative of the Department of Anatomy. He has written or coauthored some two hundred scientific papers.
Sterman has a reputation in the field as a tough, no-nonsense researcher who does not tolerate claims or statements based on what he thinks is anything less than rigorous science. His work has stood out in a field that has been accused of poor research methods and making claims beyond what the science can support. And he is not shy about voicing his opinion; some say he's arrogant and something of an intellectual tyrant. He has stood up at scientific meetings, in the middle of research presentations, loudly interrupting and challenging the methods and results of colleagues.
Sterman's most valuable tool in his studies was the electrical “fingerprint” of the human brain, the electroencephalogram, or EEG. Decades of experience with the EEG have made Sterman an expert on the electrical signals of brain and mind. “I can tell from an EEG whether someone's paying attention, and if they are, if they are paying attention to me or what they did last night. You can tell whether someone is mildly retarded from an EEG. Or whether someone is hyperaroused and can't relax, because there's very little in the way of rhythmic activity in the EEG, very little alpha. Everything depends on the topographical distribution” of the electrical frequency.
In 1963 Sterman was using the EEG as he retraced the footsteps of Ivan Petrovich Pavlov, the Russian physiologist who, using dogs as subjects, discovered the principle of classical conditioning. The son of a priest, Pavlov did wide-ranging research, studying the central nervous system, the cardiovascular system, and especially the digestive system, which won him the Nobel Prize in 1904. Later, Pavlov was in need of saliva for his work on digestion, great quantities of it, and to produce it he stimulated the flow of saliva in a dog by injecting meat powder into the dog's mouth. Before long, Pavlov noticed that the dog didn't need the injection to salivate; the animal's mouth began watering in anticipation of the powder as soon it could hear footsteps in the hall.
Pavlov began studying reflexes. He divided them into two categories: Unconditioned reflexes are those that animals and humans come into the world with, which enable us to go about life. Sleeping and eating are good examples. Conditioned reflexes are learned responses. They are strengthened by use and diminished by disuse. The ebb and flow of these behaviors allow an organism to react to a changing world around it. Pavlov built a special laboratory in Leningrad to study conditioned reflexes in dogs, one that excluded all stimuli save those he introduced. In it he built a special two-part chamber. A dog was kept in a fixed, upright position by a harness, isolated on one side of the chamber, while an experimenter sat on the other side. Tubes led from the dog's glands into the experimenter's side of the chamber so he could witness secretions. Then the researcher introduced various kinds of sensory stimuli and measured the results. A tray with food on it was swung into the animal's view. A light was flashed into the dog's eyes. A bell was rung. Smells were released. A metronome was played at various speeds. Pavlov conducted hundreds of experiments. The one he is most famous for paired the ringing of a bell with the delivery of meat powder. After a while the bell would cause the dog to salivate even if the meat powder was withheld—a conditioned response. In another experiment he attached an electrode to a dog's hind leg and explored the animal's response to “negative stimuli” or pain. After a buzzer sounded, he would shock the dog. As with the bell and the meat powder, the dog associated the buzzer and the shock and displayed physiological signs of distress when the buzzer sounded alone. What, Pavlov wondered, would an animal do if he sent it mixed signals, sounding the positive stimulus, the tone that caused salivation, and then a short time later sounding the negative stimulus, which was associated with a shock?
At first there were two separate responses, salivation and then distress. As he moved the two contrary signals closer and closer together, however, the dog was confounded and responded in a totally unexpected way. As the stress became unbearable, the animal simply checked out of the experiment altogether and fell asleep. Pavlov called this response internal inhibition, which meant the animal could voluntarily close down its system to escape the stress.
A different kind of conditioning was described in 1911 by E. L. Thorndike, who locked a cat in a puzzle box. Once the cat figured out a way to open the box, it could undo the latch faster each time. Later, the Harvard psychologist B. F. Skinner built on the concept of conditioning with the creation of the Skinner box. A rat was placed in a box with a lever. The animal would press the lever a certain number of times by chance. If pressing the lever led to a food pellet's dropping out, the probability that the rat would do it again increased. If it led to a negative reward, the probability that the rat would do it again decreased. This added the element of active involvement and is called operant conditioning.
These are the origins of the behavioral school of psychology, the study of how organisms react to external events. It was an approach that disdained the subjective inner states of emotions and feelings and sought to bring measurement and objectivity to the study of the mind. Behaviorism is responsible for a great deal of our understanding of how people move about in the world. Conditioning is a fundamental element of human behavior. At a simple level, if a bakery makes our favorite kind of doughnut, says Sterman, we will go out of the way every morning to take that route on the way to work. If a cop gives us a ticket for speeding on the way to get the doughnut, it is hoped that the negative stimulus will condition us to slow down. Most theories of learning and parenting are based on behaviorism. “You take operant conditioning plus genetic endowment, you put them together, that pretty much explains who we are by the time we're twenty,” Sterman says.
The application of behaviorism took off in this country as a result of World War II. A shortage of physicians led to the drafting of experimental psychologists to deal with the emotional problems of returning veterans. Once relegated to research labs, behavioral psychologists began to develop a practical application for their work.
Conditioning voluntary responses is one thing. The big surprise in behaviorism came in the late 1950s and early 1960s. Neal E. Miller, a researcher at Yale, proved that by using operant conditioning he could teach laboratory animals to alter their autonomic functions, functions that supposedly weren't changeable. His first experiment involved carefully observing a dog. Whenever the animal began to salivate on its own, he rewarded it with a drink of water. Eventually, the dog learned to salivate whenever it wanted a drink. Miller moved on to rats, to see if they could control their heart rate. To avoid the possibility that they were somehow using the muscles in their chest around the heart to slow the heartbeat, he injected them with a dose of curare, the poisonous plant extract used by natives in South America to paralyze their prey. Then he stuck an electrode into the pleasure center of a rat's brain, and every time it lowered or raised its heart rate—depending on what he was training the rat to do—he rewarded it with a dose of current. Within ninety minutes he had taught the animals to raise or lower their heart rates by 20 percent. Dr. Miller moved on to humans in 1969 and successfully taught a group of patients with chronic tachycardia, an abnormally fast heartbeat, to slow their hearts. Whenever a patient's heartbeat fell below a certain level, he was rewarded with a pleasant electronic tone. Further studies confirmed Miller's work, including one at Harvard Medical School, where researchers taught a group of male subjects to raise or lower their blood pressure. The reward was a five-second look at a centerfold from Playboy magazine. These discoveries would lead to the notion that people could control their own brain waves.
As a sleep researcher, Sterman was interested in the big question that plagued the field at the time: Was going to sleep something a person chose to do? Or was it an automatic shutoff, something the brain imposed regardless of choice? Pavlov's work indicated that it could indeed be a choice, for his confused dog had decided to nod off. This is what Sterman believed. To prove it Sterman set out to replicate Pavlov's experiments. He had one big advantage over the Russian: an EEG instrument, which provided detail on how the brain was responding. Was the EEG of nighttime sleep, for example, the same as “internal inhibition"? Or was it another kind of sleep altogether? This was pure behavioral research, an investigation into one of nature's curiosities with no particular goal or economic application.
In the fall of 1965 Sterman began the experiment, bringing thirty cats to his lab, where they were kept in cages and deprived of food. Cats, from the pound or laboratory supply houses, were routinely used in this kind of research for several reasons. One, they are euthanized by the thousand every year because there are so many. Two, they are inexpensive. But most important, it is because cats—as opposed to dogs—are all roughly the same size and have a uniform-sized brain, which makes an atlas of brain sites applicable for the whole species. It is much easier, therefore, to be sure the electrodes are implanted in the same place in the brain in all the cats.
The first part of the experiment was basic behavioral training. A cat was taken out of its residential cage and placed in a two-foot-by-two-foot, aluminum, soundproof experimental chamber. Every time the animal pressed a lever, a ladle would dip into a reservoir of milk mixed with chicken broth (cats don't find milk tasty enough to work for it alone), and the ladle arm would swing to a hole in the cage big enough for the cat to stick its head through and drink. “These were happy and healthy cats, and they loved working with us,” Sterman says. “We'd let them out of their cage, and they would run down the hall to the training facility and say, ‘Let's go to work.’” Among the researchers working with Sterman was Wanda Wyrwicka, a physiologist who had defected from Poland. Sterman was placing the cats in the cage and waiting for the animals to figure out that pressing the lever would deliver a reward. Nonsense, said Wyrwicka. That takes too long. “She said, ‘Don't wait for the cat to figure it out,’” Sterman remembers. “She put the cat's paw on the lever and said, ‘Here, stupid, press this.’” The cats learned quickly. The team fastened stainless-steel screws into the cats’ skulls so they could attach leads to the EEG instrument, which read the brain wave activity across the cat's sensory motor cortex, the strip of the brain across the middle of the skull that governs the cat's motor activity and its sensory processing.
After the cats were thoroughly conditioned to press the lever and get their reward, a new element was introduced: a tone. If the cat pressed the lever while the tone sounded, the dose of chicken broth and milk would not be delivered. It had to wait until the tone stopped before it could press the lever and get the reward. So the cat would sit in the box, press the lever a couple of times, and get its reward. Then the tone would come on, and the cat would have to wait. Sterman expected the cat to go into a state of internal inhibition, possibly into a “microsleep,” which should be reflected in the EEG. But the cat didn't go to sleep. The animal entered a unique state—it remained absolutely still, though extremely alert, waiting for the tone to end. It is the same state a house cat waits in, feigning heavy-lidded indifference, as a bird makes it way near enough to be pounced on. Accompanying this motor stillness was an EEG “spindle"—that is, the scribbled display of the brain's rhythmic electrical signal on the EEG paper that was unlike any seen previously. “It was a clear, rhythmic change,” says Sterman. “It was fascinating. We had never encountered this EEG rhythm before, and it didn't exist in the literature of the time.” Later, Sterman found passing reference to the frequency, but there had been no systematic study. Sterman named the frequency sensorimotor rhythm—a rhythmic signal peaking in the range of 12 to 15 hertz. It is a beta frequency, but over a specific part of the brain, the sensorimotor cortex, and so it is called SMR.
Sterman carried the experiment a step further, forging the first, crucial link between the neurological and behaviorist findings of the time. Could a cat be operantly conditioned to create this specific range of brain waves on its own, to willfully alter what was thought to be out of its control? Neal Miller's work indicated that it might be possible. Sid Ross, a technician who worked with Sterman, fabricated an electronic filter that isolated the 12 to 15 hertz in the EEG. The cat sat in the experimental chamber, but no light or lever was used in the conditioning experiment. If the cat created a half-second burst of the unusual 12-to-15-hertz frequency, its brain waves automatically triggered delivery of a shot of broth and milk into the ladle. Over the course of about a year, Sterman and his assistants trained ten cats an hour a day, three or four times a week, and the cats learned to produce 12 to 15 hertz at will. (The researchers also trained a group to inhibit SMR to see what effect a lack of the frequency might have on sleep.) To see the effect of such EEG training, Sterman studied the cats’ EEG during sleep, because when a cat is sleeping there is no possibility that it is trying to please its handler. The sleep EEG had been profoundly altered. Sterman also noticed that the cats slept much more soundly with a marked decrease in the number of times they woke up.
The final step of an operant conditioning experiment is a process called extinction. If the animal is thoroughly conditioned and the reward is withdrawn suddenly and completely, the animal will perform the task repeatedly to regain the reward. Sterman stopped providing the broth and milk mixture. Sure enough, the EEG showed that the cats were producing more SMR than ever. “I like the analogy that goes as follows: Every day you stop at that favorite doughnut shop,” Sterman says. “One day you're on your way to work and you get to the shop, and the door is locked. You pull on the door, knock on the door, and even go around to the back door and pull on it. You try very desperately to get into the place. That's what extinction is.”
The results of the cat training were published in the prestigious journal Brain Research in 1967. Sterman repeated the study with eight rhesus monkeys, using Spanish peanuts instead of milk and broth, and those studies showed the same results.
While Sterman's research results were interesting, he had no idea if the SMR work had any application in the real world. He had answered a basic research question and created a mildly intriguing, if obscure, niche for himself as the first person to isolate the 12-to-15-hertz frequency, observe its properties in cats and monkeys, and train the animals to produce it. No one, save a handful of other sleep researchers, would ever know about the concept of SMR.
Another coincidence, however, ensured that Sterman's research would in fact spread. In 1967, during the SMR research, Sterman got a call from a friend of his, a researcher named Dave Fairchild. It seemed that Gordon Allies, a drug researcher who had gone down in pharmaceutical history as the inventor of amphetamine, had obtained a contract with the army to research monomethylhydrazine, or rocket fuel. The United States was playing catch-up in the space race with the Soviet Union, but the rocket scientists had a problem. Monomethylhydrazine was highly toxic. When workers breathed in the fumes, or came in contact with the substance, it caused nausea, severe epileptic seizures, and eventually death. There was also concern that at lower doses it could bring about hallucinations or disruption of cognitive functions. Astronauts orbiting the earth during the Mercury program had claimed that they had seen natives waving at them as they flew over the South Pacific. That was, of course, impossible, for Earth was miles below. Was monomethylhydrazine leaking into the cockpit somehow? The Defense Department awarded a contract to Fairchild and Allies for research into the toxic effects of rocket fuel. Unfortunately, Allies had tested on himself a chemical compound he was developing, and it had killed him. Fairchild asked Sterman if he was interested in taking on the study, and Sterman accepted.
Sterman brought in fifty cats for the new study. The animals didn't run down the hall to work in this experiment. They were injected with rocket fuel, ten milligrams for every kilogram of body weight. Again, their brains were wired to an EEG to measure their reaction. A few minutes after the injection, all of the cats did the same thing: they vomited, made noises, salivated, and panted. Most of them went into grand mal epileptic seizures after one hour. Most of them, but not all. While a small part of the research population—ten cats—displayed all of the symptoms the other cats did, the onset of seizures was substantially delayed in seven of the population and never happened at all in the other three. “Boy, did we scratch our heads over that,” says Sterman. “We couldn't figure out what the hell was going on. Answering that question defined the next ten years of my life. I forgot all about sleep research.”
Through the course of science, a remarkable number of key discoveries have been abetted by coincidence. This was one of them. The seizure-resistant cats, it turned out, were animals left over from the previous study, the study in which the cats were taught to produce the sensorimotor rhythm. What Sterman had done by teaching the cats to produce SMR, he would come to realize, was to strengthen their brain function at the sensory motor strip, the same way a person builds muscle mass by repeatedly lifting weights. In medical parlance, their “seizure thresholds” had been increased; their brains were now functionally altered so as to resist the spread of slow theta waves across the motor cortex that caused seizures.
The study demonstrated a clear connection between mind and physiology. Simply guiding a cat into a specific mental state to strengthen a cluster of neurons in the brain in turn prevented motor seizures. And there was more than just a change in the brain; there were physiological changes from the top to the bottom of the cats and monkeys Sterman had studied. “Very specific, measurable changes,” says Sterman. “In the brain, cell firing patterns changed and cells in the motor pathway reduced their rate of firing. Circuit patterns changed. And we found changes in the body. Respiration stabilized. Heart rate went down. Muscle tone in antigravity muscles decreased. Reflexes diminished.”
These animal studies clearly demonstrate, says Sterman, that the effects of SMR neurofeedback are physiological and not, as some critics say, a placebo. Placebo is Latin for “I shall please.” It refers to a common phenomenon in the medical world in which people who are part of a test on a new drug, for example, unknown to them, are given a sugar pill instead of the drug, but their health still improves, some times dramatically. Somehow they have fooled themselves and gotten better on their own. Placebo effects can be dramatic but are transient. They usually occur at a rate of about 30 percent in a given study, but they can be higher. “But there is no placebo effect in cats or monkeys,” Sterman says.
Sterman would have been content to keep experimenting with animals, he says, but Wyrwicka, his assistant, adamantly insisted that it was his moral responsibility to use the powerful technique on humans. Sterman wasn't sure that humans had the same signature rhythm. With the equipment of the time there was concern that the EEG wasn't advanced enough to accurately pick up brain waves through the human skull, and one couldn't put stainless-steel screws into the human brain. Sterman tried a different approach. A neurologist referred several patients to him who, because of cancer in the skull, had had a portion of the bone removed. Without interference, Sterman says, “Their EEG was beautiful," smiling at the clarity of the frequency as if he were recalling a breathtaking mountain vista. “And there it was—SMR.” The existence of SMR in humans was confirmed.
Then another stroke of serendipity came into play. For the first human subject he would teach to produce SMR, Sterman had to look no farther than his own research lab: a computer coder who worked for a colleague. Twenty-three-year-old Mary Fairbanks, who had suffered from a major motor seizure disorder since she was eight years old, was the perfect test subject. Two or more times per month she went into severe grand mal seizures, violently shaking and passing out. Drugs could not control them. Over the years she had meticulously kept a log of her seizures, detailing their severity and frequency, and that documentation was invaluable. Her seizure disorder had also been thoroughly recorded by the National Institutes of Health and by medical researchers at the University of Wisconsin.
Epilepsy is an invasion of an unwanted frequency in the brain. When a person is walking, talking, and engaging in everyday life, the brain is operating in the higher frequency ranges, called beta, from 12 to 18 hertz. As stated earlier, in many ways the normal waking brain is like a symphony perfectly timed to create the simplest tasks. In epileptics a portion of the brain is unstable, or hyperexcited, and it can't resist as slow theta waves, in the 4 to 8 hertz range, start to creep in. This in turn recruits other areas of the brain to produce the abnormally low frequency. During an epileptic seizure it's as if all of the musicians in a symphony are playing at once, but without a score or a conductor. Normal motor function is disrupted. If the animal research held up, the brain would be made stronger with SMR, better able to resist the unwanted theta, and seizures should diminish or be prevented.
So in 1971 Sterman wired up his first human subject to a biofeedback instrument that he had instructed his technician, Sid Ross, to build. The unit was nothing more than a simple black electronic box with two lights on it, red and green. As in the device wired to the cats, a filter separated out the 12 to 15 hertz. When Fairbanks produced SMR, she was rewarded. Instead of a shot of chicken broth and milk in a ladle, however, a green light came on. When she was not in that range, and the low-frequency brain waves that caused the seizures were dominant, a red light came on. Her task was to keep the green light on and the red light off so as to encourage the high-frequency waves and simultaneously inhibit the low frequency. The SMR frequency was not really new to Fairbanks; like everyone else she passed through it all the time, spending a split second in it here and perhaps a few seconds in it there. What the biofeedback equipment did was help her dwell in that state for longer periods of time. And dwelling there is what teaches the cortex how to maintain stability.
For the purposes of measuring EEG signals, the human head has been divided into nineteen different sites in an international system of measurement called the “ten twenty” system. Sterman trained just two sites, called C-3 and T-3, which lie between the top of the left ear and the very center of the top of the head. Almost all neurofeedback done in the early days was carried out at C-3.
One hypothesis about what might be going on in the brain during neurofeedback has to do with the way the cells in the brain connect with one another. Since information travels along the branchlike connections between cells called dendrites, the denser and greater in number these connections are, the better the transfer of information. As frequency increases during a neurotherapy session and the brain is activated, more blood than usual streams to that area of the brain—the nutrients in the blood may be strengthening or reorganizing existing connections, which increases the cells’ ability to self regulate. This is what many scientists think happens during any learning process. (Brain scans show that in people who go blind and learn Braille, the neurons in the area that governs their reading finger become more robust.) The neurofeedback model holds that the brain wave training increases the stability of that area of the brain as well as its flexibility, or its ability to move between mental states (from sleep to consciousness or arousal to relaxation, for example). It allows the players in the orchestra to play their parts better, to find the correct tempo, to come in on time, and to stop playing when they aren't needed. Since every aspect of a person is driven by an assembly of neurons, the healthier those neurons are, the healthier are the functions that they govern.
Describing precisely how it feels to be in the 12-to-15-hertz range in the sensorimotor cortex is difficult. “It's not relaxation, it's not just not moving,” says Sterman. “It's when we will ourselves to be still. It's a standby state for the motor system. You might think of it as a VCR; it's a pause button.” One man who did SMR training, a tennis player, said that producing the rhythm feels to him very much the same as that calm and vigilant split second when he throws a tennis ball into the air to serve and waits for it to fall back down far enough to hit. Imagining that situation, in fact, enabled him to produce SMR in the clinic. And the secret of neurofeedback is that, though it sounds complicated, producing SMR is a simple thing for most people to do. In fact, as with learning to ride a bike, it's easier to do than to describe.
In 1972 Fairbanks trained for an hour a day, twice a week, for three months on the prototype neurofeedback instrument, for a total of twenty-four sessions. “I was very skeptical,” Sterman says. “But the results were remarkable. She went nearly seizure-free in three months. She ultimately got her driver's license. You can see a sudden decline in seizures due to the placebo effect, but if it is a placebo, you usually get breakthrough seizures again after some time. [That is, the seizures return.] For her to be seizure-free three months later was unprecedented in the history of her type of seizure disorder.” Not only did the treatment work, it was amazingly robust. The researchers noted other changes in their subject, including a shift from being “a quiet and unobtrusive individual” to being more outgoing and “showing an increased personal confidence and an enhanced interest in her appearance.” In hindsight, it was vital that someone who responded so well have been the first to try it. If it had been someone whose response was lackluster or nonexistent, Sterman might have ended his work on humans then and there. Sterman wrote a paper on the case, and it was immediately accepted by the top EEG journal, called EEG and Clinical Neurophysiology, and published the same year. Publishing in a major scientific journal is the last big step in a research project and includes a rigorous review by peers. Only those studies that have met strict scientific protocol are published, and acceptance is an imprimatur on the quality of the work. Sterman had finished the first leg of an important discovery.
The paper thrust Sterman into the limelight of behavioral studies. He had discovered a special rhythm: three notes that played a central role in brain self-regulation and could be trained to improve brain function. In 1973 several physicians and neurologists came to work with him; Dr. Joel Lubar, a professor of psychology from the University of Tennessee at Knoxville, who would later become instrumental in the field of neurofeedback, came to Sterman's laboratory at Sepulveda on a nine-month National Science Foundation fellowship to observe Sterman's work after he had replicated the epilepsy work in his own lab. Eventually, Sterman's results were replicated in several other laboratories. This successful exporting of the technique to other labs is critical in demonstrating efficacy.
Four epileptics were recruited for the next phase of research, and the results were very good, a 60-65 percent reduction in grand mal seizures. This time the results were published in Epilepsia, a top journal in the field of epileptic research, in 1974. The grant money rolled in, from the Neurological Disease and Stroke Branch of the National Institutes of Health, and the funding allowed Sterman in 1976 to expand the next test to eight subjects. He made the testing more rigorous. The three-year experiment was constructed with an A-B-A design. As in the first experiment, patients were trained for three months to increase their SMR waves and suppress theta or lower-frequency waves, which cause seizures. The number of seizures dropped dramatically, as predicted. After three months the protocol was reversed, though the subjects weren't told, and they were taught to raise the “bad stuff “ and suppress the good. The subjects started experiencing an increasing number of seizures. Three months later the protocol was reversed again. “They reversed their seizures when they got their SMR up and their theta down,” says Sterman. “We did all-night EEG recordings to keep consciousness out of the picture. And their EEG during sleep showed changes. And when we reversed, their seizures went back to baseline.” All of these changes were made in secrecy—not even the people applying neurofeedback knew who was getting what—to keep any placebo effect out of the picture. An A-B-A design is the most powerful study design of all. The ethics that govern studies no longer allow A-B-A designs, however, because they do harm to a patient by first making them well and then taking them back into their illness. These results were published in Epilepsia in 1978.
Sterman was well along the way to a major discovery: a nondrug, nonsurgical method of treating epilepsy. He parlayed his latest results into another National Institutes of Health study, with twenty-four patients over three years. Again, the study was carefully and conservatively designed, with three groups of eight subjects: two control groups and one experimental group. If drugs cannot stave off severe seizures, the next step is often surgery, and Sterman's study group included people who were on a waiting list for anterior temporal lobectomy, a procedure that removes the seizure focus—a small piece of damaged tissue that is the source of the low frequency. One control group received no treatment; its members were simply given logs and instructions on how to record seizures. The other two groups were divided into pairs. One person in a pair—let's say Robert—was given true neurofeedback. He was part of the experimental group. His partner, Ralph, a control, was hooked up and thought he was getting true neurofeedback as well, but he was in reality responding to a recording of Robert's EEG signal, which researchers were feeding to him. It was what psychologists call yoked, or sham, treatment. No one, save the administrators, knew who was getting what. It is known as a double-yoked design, because both people are hooked, or yoked, to the same treatment. If controls such as Ralph, who were trained to someone else's EEG, reduced their seizures, researchers knew a placebo effect was at work. But the control group didn't. Those with the real feedback reduced their seizures, while those with the sham treatment did not. “It worked wonderfully,” says Sterman. “We had several patients that went totally seizure-free. The results were unequivocal.” A year later the average rate of severe seizures had dropped by more than 60 percent. And those waiting for surgery? “None of them ever went back,” Sterman says. After the study was completed, those in the control groups were given true neurofeedback training.
I showed Sterman's papers to several psychologists, and all of them feel strongly that Sterman's work is first-class and say his experiments were well and carefully thought out. “It was an elegant design,” said Dr. Chris Carroll, a psychologist in full-time private practice in Glen Cove, New York, and a specialist in special education.”It was very well suited to the questions being asked, and is top-notch work.” Carroll is very familiar with neurofeedback, for he uses it to treat clients. But he is also critical of the lack of double-blind, controlled studies in the field. Sterman's work is one of the few exceptions. If there is a criticism, it is that the sample size—a total of 37 people in Sterman's studies—was small. But Sterman's work was also replicated in several other independent studies in other laboratories. All together 174 subjects were trained, and 142 showed substantial clinical improvement. Five percent went completely seizure-free.
Dr. Alan Strohmayer, a Ph.D. psychologist with a private practice, an assistant research professor of neuroscience and neurology at NYU School of Medicine, and the former director of the biofeedback program at North Shore University Hospital in Manhasset, New York, also said Sterman's work was excellent. “There's no placebo in cats, no expectation, no wanting to perform for the doctor,” said Strohmayer. “The cats statistically demonstrated control over their brain waves. It's the best evidence we have that shows we're capable of training someone else's brain waves.”
As we saw with the electronic brain stimulation experiments in the last chapter, frequency has a dramatic impact on the brain, affecting everything from motor skills to feelings of pleasure and pain. The difference between ESB and Sterman's work is that in neurotherapy, people are learning to generate the electrical current on their own. A number of studies show that the thalamus is the generator of rhythmic electrical activity in the brain—the orchestra conductor. Sterman believes the epileptics he trained were learning, through operant conditioning, to control the thalamic generator in the same way that Neal Miller's subjects learned to alter their heart rates.
While Sterman did not deal directly with the patients very often, a psychologist named Robert Reynolds did. Reynolds is a Ph.D. psychologist in Connecticut, treating traumatic brain injury with neurofeedback. As a psychology student at Cal State Northridge in the 1970s, he worked for Sterman, giving patients a battery of psychological tests as they began the study and testing them again as they left. “These people were having grand mal seizures, multiple grand mals a day,” he said, and the family was usually grim when the person was entering the study. “I would see them four months later, and the family was completely turned around.” Seizures had diminished, and “the people would be laughing and the family was happy. It was remarkable.” Reynolds's experience with neurofeedback led him back to the field, and he now considers it the most important tool in his work with brain-injured and ADD patients.
The double-yoked study had proven the procedure's efficacy beyond any reasonable doubt, Sterman felt, and the next step was to explore long-term management of epilepsy with the technique, with an eye toward a clinical application, something that could be used in a doctor's office or hospital setting. In 1982 Sterman submitted another grant proposed to the NIH and was awarded $70,000 a year for three years. Toward the end of the first year of the study, a strange thing happened. Sterman received a letter from the NIH saying that the grant had been reviewed again and the committee wanted a double-blind aspect to the study. Such midstream changes were unheard of, but Sterman had no choice but to comply. Then, instead of signing off on the already approved grant, the committee sent it back for further review. Then Sterman received another letter, to the effect that the work had achieved its objective and no further research was needed. Sterman was floored. Funding had been pulled out from under him. In the middle of the project, the study was over.
Sterman still simmers about the episode and feels he was a victim of politics, a casualty of an assault by the medical community at NIH. “Doctors want Ph.D.s to be in the laboratory documenting procedures, documenting drugs, documenting the treatments they could apply. And here I was coming up with a protocol for a long-term treatment of epileptics. It was a turf battle. Pure politics.” Almost all of behaviorism was abandoned in favor of pharmaceuticals in the 1970s, and biofeedback is barely a blip on the radar screen of modern medicine.
Conservative in his approach to research, Sterman still wanted more data on the technique. But without funding, there was nothing he could do. He boxed up his files on the SMR training in cats, monkeys, and humans and moved on to other projects, including research for the U.S. Air Force, studying the ways pilots pay attention to their flying, and helping design cockpits for maximum efficiency. Eventually, on a part-time basis, he treated some epileptic patients at Hollywood Presbyterian Hospital with the biofeedback technique.
Elizabeth Kim was among those with grand mal epilepsy treated by Sterman after he finished his research. She started her therapy in 1983 on a fee-for-service basis. Nearly 25 years later, she says that the neurofeedback training changed things for her. “It increased the quality of my life tremendously,” says Kim, who is now 48 and in charge of donor recruitment at a Southern California sperm bank. After several dozen sessions at Hollywood Presbyterian Hospital, she told her neurologist she wanted to begin reducing her medication, and she did. At the same time, her seizures went from an average of one per month to four to six per year and were much less severe. “What was most noticeable,” she says, “were the periods between the seizures. Within a month I may only have one seizure, but in that time I may get an aura and feel like I am going to have a seizure five or six times. It creates a lot of anxiety. I was working full-time, and I would wonder, ‘Am I really going to have a seizure? Should I go home?’ With biofeedback this was reduced tremendously. I didn't have the feelings that I was going to have a seizure and the anxiety that goes with it.” Several years after she stopped doing neurofeedback, Kim was taking one medication instead of the three she was taking before therapy, and her seizures have dropped to one or two a year.
What Sterman and his research team apparently accomplished, even though it wasn't fully appreciated at the time—and still isn't — was enormous. If brain cells are indeed becoming stronger and growing new and permanent connections, as his model holds, he has gone a long way toward proving a concept that has only been accepted by neuroscientists in the last few years: that the brain is a dynamic and extremely plastic organ. He has also found a way to capitalize on that plasticity. He showed that by simply teaching a person how to nudge it in certain directions, the brain is capable of profound change. People with one of the most serious and disabling of medical problems—intractable epilepsy—could be taught to heal themselves. And it wasn't that difficult. Sterman had also provided tantalizing evidence—in the lab, under rigorous scientific conditions—that there was a way to harness the mind-body connection. Simply guiding the way someone thinks can change the structure of tissue in the brain and, subsequently, other key parts of human physiology. Neurons, the work demonstrated, are where mind meets body. The concept was revolutionary, a whole new paradigm. Yet except for a handful of people, the discovery dropped off the map.
Why? Part of the problem was that it had come out of psychology rather than the medical world, and the medical world wouldn't accept it. Another problem, according to Sterman, was the bad reputation he felt biofeedback had earned among scientists. While there was a great deal of careful biofeedback research being done by other bona fide research scientists, including Les Fehmi, Joe Kamiya, Barbara Brown, and others, people were making wildly speculative claims in the 1970s about alpha training with biofeedback, including claims that it provided a shortcut to feelings of transcendence and to enlightenment. Such things went over poorly in the scientific community, and every kind of biofeedback was tarred with the same brush. The bad reputation continues to dog responsible proponents of the practice to this day, but they have made strides toward overcoming it, and they believe fervently that acceptance is just around the corner. For it turns out that early researchers were on to something after all.
