The Human Brain Project is combining wet anatomy with next-gen scanning, imaging, and networking to give neuroscience a revolutionary new tool - the globally accessible online mind.
The lobby of UCLA'S Brain Mapping Center is a lofty place, an airy, two-story atrium with a polished concrete floor and a narrow balcony that circles the room like an observation deck. It's a quiet, almost meditative space that in the bright Los Angeles sunlight seems as normal as a library. Only farther in do things get weird.
Walk through an unmarked door and a short hallway leads south, past walls crammed with pictures of human brains. A few of them - scanned, digitized, colored, and sliced - have bright lines drawn through their centers that look like bundles of insulated wire. Others, whose blue depths swirl with red and green, resemble shriveled Christmas ornaments. One, seen through the cutaway skull of a live surgical patient, has been affixed with dozens of tiny, numbered squares - troop deployments on a Pentagon war map.
The images end at a room filled wall-to-wall by a giant white cube. A 2-foot-hole bores through the center of the cube. A man's legs stick out of the hole.
Next door, in the control room, a researcher leans into a microphone. "Ready?" she asks the man. "Follow the hands." Inside the cube, wearing a $40,000 pair of virtual reality goggles, Brian (not his real name) sees a pair of videotaped hands lift and move their index fingers; he copies the movements. As he does, a wavy-patterned lilac oval appears on the control-room computer screen. It's a picture of Brian's brain - specifically, one of 92 cross sections recorded along each of three axes by the cube, a functional magnetic resonance imaging scanner.
The cross section - a thin, 3-D slice - is the starting point of a project that aims to radically change the way we understand the brain. Where MRI technology spins water molecules to get high-resolution pictures of soft tissue, fMRI scans - which can be done with the same $3 million machine - record moment-by-moment variations in blood-oxygen levels, which in turn reflect neural activity. Each computer-generated slice contains a hundred thousand voxels, or 3-D pixels. Combine the information in all the voxels and slices and you get a complete picture of the brain in action. Your brain - live, on screen.
With the genome cracked and the universe mapped to its distant reaches, the brain has become one of science's final frontiers: humankind's own black box. We may know how stars burn and black holes collapse, but we still know only scraps about our own heads: why we can remember 10 phone numbers and not a hundred, or why we can recognize faces effortlessly but computers can't. Back in the 1500s, the famous Flemish anatomist Vesalius first guessed that the important parts of the brain weren't the fluid-filled pockets near its core - previously believed to house essential "animal spirits" - but the fleshy folds and wrinkles all around. (Of course, he missed a few calls, too: He asserted that our brains were awash in "fulginous excrement" that needed to be purged.) Since then, we've picked apart countless cortices - and even diced Einstein's brain in search of clues - but have found only tantalizing hints as to why some people are geniuses and the rest of us are not.
"We're like Martians looking at a car," says UCLA neuroscientist John Mazziotta, the 52-year-old director of the Brain Mapping Center. "We've driven the car, and we've taken the car apart, but we don't know how one part is related to the other." All we know is that somewhere in the homogeneous folds of our cortex, tiny aberrations drag us from normalcy into schizophrenia. Or, in rarer cases, endow us with seemingly superhuman powers: the ability to factor huge numbers, memorize a telephone book, or perceive smell as vividly as a dog does.
The past three decades of poking and prodding brought only the realization that the brain was even more complex than we'd originally suspected: 10 billion neurons and 60 trillion synapses communicating through an elaborate system of electrical and chemical signals. Worse, by the 1980s, a number of studies implied that each of our brains might have unique circuitry, with memory and language wired differently from person to person. If that were the case, comparing brains would be like trying to compare anthills, each with different tunnels and streams of information. It was possible, scientists agreed, that understanding the brain could involve mapping not a single, incredibly complex world, but mapping several billion different worlds, many of whose landmarks had yet to be found.
It was a little embarrassing. Stumped by our own brains! But during the past few years, the trackless wilderness has started yielding to advances in neurotechnology. With the help of MRIs, positron emission tomography scanners, and optical and electromagnetic signal imagers, researchers have been able to view brains down to their synapses. More important, they've peered inside the brain as it functions. With fMRI scans, introduced in 1991 by researcher Jack Belliveau and his colleagues at Massachusetts General Hospital, neurologists have begun to tease out the connections between different parts of the organ: how we remember, make associations, concentrate. Meanwhile, transcranial magnetic stimulators have enabled doctors to zap areas of the brain with magnetic pulses sent through the skull - causing zappees to see flickering lights or experience twitches. Stimulating a spot on the left or right frontal lobe has more recently been tried as a treatment for depression, with some success.
"These new techniques," says Michael Huerta, associate director of neuroscience research at the National Institute of Mental Health, "are providing windows on the essence of human beings."
The upshot is that the secrets of our thoughts and talents aren't just hidden in dead cells under a microscope but in our own buzzing, rushing minds. This, at least, is the belief of Mazziotta, who, along with fellow UCLA neurologist Arthur W. Toga, is among nearly 200 researchers currently undertaking one of the largest projects in the history of neuroscience: an effort so vast and far-reaching that it's known simply as the Human Brain Project.
Launched in 1993 by the National Institute of Mental Health and four other federal agencies, the Human Brain Project aims to build an omnidimensional, computerized database that synthesizes all the subspecialties of neurological research, from the shape of synapses up through chemistry and gross anatomy. The unique effort, which is designed to help everyone from doctors treating epilepsy to researchers testing new Alzheimer's drugs, is being carried out at 19 universities and 6 hospitals in 10 countries.
The whole project may take decades, but Mazziotta and Toga's contribution at UCLA will be finished much earlier, around 2004. Their plan, underwritten by one of the first Human Brain Project grants, is to build a map that quantifies the range of variation in the human brain - and helps researchers determine whether we do, indeed, think differently.
Once finished, the UCLA brain map will represent the most comprehensive picture ever produced of the "normal" (read: healthy) human brain. Researchers who now spend days searching for information will be able to go online and find it in minutes. Physicians who have no ready basis of comparison for a patient's puzzling brain scan will be able to call up 3-D images, check for discrepancies, and target the problem. "We're trying to build a representative atlas of the human brain, similar to the one we might have for the Earth," Mazziotta says. "Except instead of looking up average rainfall and population, we'll be looking up average blood flow and neurotransmitter density."
To get the core data, Mazziotta and Toga enlisted 7,000 volunteers, ages 17 to 80, all of whom remain anonymous. Of these, 5,800 provided DNA samples and all completed background questionnaires and submitted to a 50-minute anatomical MRI test. It was, by far, the largest number of scans ever assembled - merely compiling the information took the better part of a decade, with the last scan completed in October 2000. And the process isn't over.
While the first part of the project looked at anatomy, part two, slated to begin this summer, is an attempt to map brain function. A series of nine fMRI scans will be performed on 1,000 of the volunteers, charting their brain activity. The part two information will then join the 100 terabytes of data already in storage on six servers inside UCLA's Reed Building - enough to encode all the books in the Library of Congress five times over.
Ultimately, the atlas will be wedded to the even larger Human Brain Project, along with hundreds of other studies. And as more and more segments of the HBP go online - there is no official launch date yet, since the effort is constantly evolving - neuroscientists will be able to review and combine increasing amounts of data, enhancing their ability to diagnose and fight disease. Doctors could use the data to plan surgeries, or to simulate how a Parkinson's drug affects brain cells, or, in the far future, to monitor regions where patients are likely to develop a problem.
As the vast effort grinds on, what's already certain is that the HBP will dramatically accelerate our ability to decipher brain disorders - and understand how we think. "Within 10 years," Mazziotta predicts, "these databases will have become an integral part of how neuroscience is done."
The Human Brain Project has never lacked for ambition. Back in 1982, the Department of Defense approached a UC San Diego anatomist named Robert Livingston with a wad of cash and a plan to build a computer to evaluate brain function - so that soldiers could be tested for mental strengths, among other things.
"That was the seed - futuristic stuff," says Stephen H. Koslow, Human Brain Project coordinator and associate director of neuroinformatics for the National Institute of Mental Health. Livingston organized a three-day conference at Texas A&M in College Station, which Koslow attended. "We realized that, given the complexity of the brain and the size of the image files, the computer resources you'd need would be enormous," Koslow recalls. "This was 1982, and there was just no way to do this." Soon after, Livingston gave up on the project, and so did the Army. But by 1993, things had changed. Although the military's hoped-for "ability-meter" remained far out of reach, computers were gradually making it possible to connect isolated areas of brain research. Databasing the brain suddenly seemed not just possible, but vital.
"We were getting buried by data," Koslow says. Mazziotta agrees. "Back in 1993, nobody wanted to do this work. 'This is just computer stuff,' they said. 'We want to keep working in the lab.'" In the end, frustration drove Mazziotta to the project: He couldn't stand how unwieldy brain research had become. "Have you ever been to a neuroscience conference?" he asks. "Two thousand articles are presented. You walk away with a huge book of papers, but there's no way to combine studies into something you can use."
In the Human Brain Project's first year, NIMH endowed it with a mere $2.5 million. But as computing power increased, so did the HBP's viability. Last year's budget, a still-meager $12 million - less than one-twentieth of the federal allotment for the Human Genome Project - represented an all-time high, with millions in additional funding brought in from scientists' private grants. HBP supporters believe the money could not be better spent. "It's the fastest way to understanding the brain," Koslow says.
Not everybody buys the rhetoric. Some critics who applaud its goals think the project is overblown and unrealistic - the wishful thinking of neurologists seduced by technology. The creation of a giant pool of freely shared neurological data? Not in a field where competition is stiff and a researcher's results are his only currency. "I've heard some people snicker," admits George Ojemann, a professor of neurological surgery at the University of Washington. And though it's not easy to find neurologists who will publicly fault the effort, some still have doubts.
"A brain database is built on the notion that if you throw all this data together, it will somehow naturally sort itself out in a way that's helpful," argues New York University professor of neural science Tony Movshon. "It's not a bad idea in principle, but in practice it's a complete shot in the dark. I'm just afraid that there's going to be less to this than meets the eye."
So where does the truth lie? The new millennium could be a time of cures and technology-brokered self-understanding. Machines might chart our thoughts; depression might be cured with surgical cortex tweaks; love might be quantifiable. Maybe - and maybe not. There are levels of complexity to the brain that we've only begun to comprehend, let alone manipulate. For starters, the 3-pound organ contains more possible neural pathways than there are atoms in the visible universe - enough to allow us to perform some 20 million billion calculations per second. And while we know that complicated states like self-consciousness arise from this tangle, we don't know which of the billion-billion-billion possible paths combine to create them. Jim Brinkley, a research professor in the University of Washington's Structural Informatics Group, echoes a common sentiment: "Next to databasing the brain, sequencing the human genome is a trivial undertaking." Mazziotta compares the project to "trying to figure out all of the universe and how it interacts."
Of course, this is exactly what makes the HBP so appealing. In an age when we've solved Fermat's last theorem and peered by telescope back to the Big Bang, few things remain as ripe for exploration as the brain. We've already seen the impact of imaging technology, which in the past year has brought us closer to a cure for Alzheimer's and deepened our understanding of schizophrenia, dyslexia, and alcoholism. If it all works out, the HBP may yet save us from, or deliver us to, ourselves.
It's early afternoon and I'm alone in Mazziotta's upstairs office, waiting. The room, like Mazziotta himself, is sleek and a little impersonal. There's a blond-wood desk, a polished concrete floor that he fought the university to get, and a book on architecture by Frank Gehry. The book is called Gehry Talks, and below the title someone has written "too much."
"Gehry wrote that," Mazziotta tells me, coming in with a white lab coat draped over his arm. He sits down and eyes a cup of soup on his desk; judging by its glutinous sheen, it's been waiting awhile. He and Gehry are something like friends, it turns out, although Mazziotta is reluctant to say more. The architect even consulted the doctor about designing a building that somehow approximates a brain.
"Not literally," Mazziotta says. "Just conceptually."
Mazziotta has the no-nonsense look one would want in a neurologist. He's been on call for two days and hasn't had much sleep - but seems perfectly, almost preternaturally, unaffected. He excuses himself to return a call about a woman whose brain is hemorrhaging. I bury my nose in a lurid red-and-black textbook titled Brain Mapping: The Disorders.
Figuring out the brain is like trying to do a crossword puzzle of unknown shape and pattern, whose thousands of clues are hidden around the globe. First there's the matter of finding the clues (how are neurons arranged in the brain?). Then there's the problem of finding answers that force you to look for more clues (why are neurons so densely packed in the cerebellum?). Finally, there's the challenge of answering the hardest clues (how does the density of neurons affect our coordination, musical talent, speech?) in ways that lock all the pieces together.
For now, Mazziotta and his colleagues hope to sort out the link between structure and function - and how it varies. When two people connect the word "cat" to a picture of a cat, do their brains light up identically?
That they would isn't obvious. Cut open the body and the workings are fairly suggestive: a big pulsing heart, long ropy veins, a sacklike stomach full of food. Cut open the brain, and you get nothing. No sparking wires, no tiny gears - just a spongy gray-white ball of tissue that looks, in cross section, like a slab of strudel.
The brain's featurelessness mystified early neurologists, who managed to discover such structures as a visual cortex only by autopsying stroke and tumor patients. As Toga explains to me later, "When a patient had a stroke and suddenly couldn't talk, or could hear but not understand what was said to him, you'd wait until he died, then see which part of his brain blew out."
More recently, it has become obvious that our "featureless" brain actually contains a remarkable microstructure: billions of neurons and synapses stacked in maximally connective ways to create a kind of superparallel electrochemical computer. Every time we read, remember to buy milk, or count change, electrical impulses moving through our neurons launch chemical neurotransmitters toward any of thousands of synaptic receptor sites, which in turn may trigger other neurons. Interrupt the timing or pattern of these circuits, either deliberately (with an electromagnetic pulse), or inadvertently (with a tumor, stroke, or injury), and dramatic things happen. We're suddenly unable to read words on the page in front of us. We don't recognize ourselves in the mirror.
What interests neurologists like Mazziotta is whether normal people - those not suffering brain diseases or mental illness - have their computers wired up in mostly the same way. If they do, it will be possible to establish a normal range of brain appearance and responsiveness. "We're trying to get an idea of how much variation there is," Mazziotta says, as we head downstairs to the lab where the function tests are being prototyped. More important, Mazziotta says, he would like to find out how much variation matters. Consider the folds on the outside of the brain, he says - they're believed to be as unique as fingerprints. But no one knows if that makes a difference to brain function.
Mazziotta and Toga's atlas will be a good place to look for answers. The map could potentially tell doctors which areas of a psychotic patient's brain are failing to activate - or are firing too much. (The voices that a schizophrenic hears, for instance, appear as bursts of activity in the auditory cortex.) Ultimately, such a map may sort out the nature/nurture debate. Perhaps Einstein was a genius because he was born with extra-wide inferior parietal lobes, a feature that has been associated with mathematical skills. Or maybe he widened his lobes through heavy use, the way a weightlifter builds muscle.
The brain-function lab is a tiny, windowless room containing two computers and what looks like an optometrist's chair. Fumiko Maeda, a postdoc, stands inside, preparing for a trial run. The 1,000 volunteers for part two of the mapping project will return for fMRI tests, she explains, as soon as the University of California's human-subjects review board gives its approval, which is expected to happen this summer. The volunteers will repeat a series of exercises inside the cube, like associating a verb with whatever object they see projected in the VR goggles. Maeda hits a button to demonstrate the trial, and pictures flash by: a nose, a chicken, a cigarette, a deer, a ladder, a squirrel, a shirt, a goat. After 30 seconds, the test stops and the viewer is supposed to concentrate on a small black cross in the center of the screen - a control task to help researchers sift out so-called unspecific attentional effects.
At any time, the brain signal that corresponds to, say, a ladder/climb association is deeply buried in background noise: spurious electronic signals, the thunk of a patient's heartbeat, the neural spikes of fleeting thoughts, sounds, and sensations. As a corrective, researchers measure the brain at rest and subtract that image from the test image.
"It's as if you're looking at the Earth but it's covered in fog," Mazziotta explains later, over lunch in the neurology department's Cafe Synapse. "We can dial the fog down a little and make out Everest. Dial it down a little more and we'll see the Himalayas and the Andes." What's different about brain fog, though, is that at some point, the more of it you dissolve, the less you can see of the brain's details.
The ladder/climb association, for example, shows up on the fMRI computer screen as a scattering of 3-D green blobs suspended in our brains like lava-lamp bubbles. If researchers dial down the fog - that is, lower the statistical threshold - they'd see more blobs, but fewer that they could be sure came from the association and not from a glitch in the imager's magnetic field or another extraneous problem.
Still, neuroscientists are encouraged. The verb-association test has been done in 14 languages, and similar parts of the brain have lit up each time. Evidence is mounting that at least some functions in normal brains universally appear in the same place.
People with associative disorders might not connect any verb to the word banana, even though they would recognize a banana and could easily describe one. This seems to indicate that the brain doesn't keep a separate network for ideas about bananas. It uses one network to make all associations.
What such a clue helps reveal is how the brain is organized. One might guess, for instance, that a part of the brain would be reserved for associating words with pictures, but would one expect to find a region responsible solely for recognizing human faces? Strangely, such a thing seems to exist. Patients with similarly located brain lesions suffer from prosopagnosia - a disorder that leaves them able to recognize everything except faces.
"What interests us about people with brain injuries," says Mirella Dapretto, a language-processing researcher collaborating with Mazziotta at UCLA, "is that they begin to show us how the brain categorizes things."
In some situations, the brain is radically adaptable, able to re-create damaged circuits and route them through wholly different areas. In one famous case in the 1840s, a man named Phinneas Gage continued to function fairly well after an explosion drove a tamping iron all the way through his skull. More typically, though, brain injuries leave permanent aftereffects. Patients become angrier (as Gage did), or more volatile, or abruptly docile, or emotionless. And then there was the British political journalist who recovered from a stroke but developed a sudden obsession with gourmet food. The effect became known as Gourmand Syndrome in 1997, after doctors analyzed 36 patients preoccupied with fine food, 34 of whom had injuries to the same brain region as the journalist.
This is especially eerie because it seems to show that some of our most personal likes and dislikes - our passions - might actually be hardwired. If this is true, it may mean that we will someday be able to fix the "defects" that make us who we are. It may also mean that we will eventually be able to trace our desire for a brain atlas back to its physiological roots: pinpointing the part of our brain that, for some reason, really likes maps.
Arthur W. Toga, associate director of UCLA's Brain Mapping Center and director of the university's Laboratory of Neuro Imaging, is tweedier and more approachable than Mazziotta, and occupies an office that's distinctly less luxe. Slightly peeling wallpaper frames a room dotted with toys that have grotesquely oversized heads. "You want to see brains?" he exclaims at one point. "Do we have brains!"
As a collaborator on the atlas, Toga is responsible for assembling the thousands of high-res anatomical brain scans into a public-access database. Because the matter is thorny - how do you compare one patient's corpus callosum to another's? by thickness? total volume? curvature? - he has enlisted the help of the Mitre Corporation, a government-funded think tank best known for modernizing the Federal Air Traffic Control system. At Toga's behest, Mitre has proposed a five-year plan to create the brain-atlas search software, with a preliminary version of the database scheduled to go online within two years.
At first, the atlas would be limited to anatomy. The more challenging work of integrating the function studies will come later, according to Jordan Feidler, director of Mitre's artificial intelligence division. "The problem is there are so many subtle differences between functional studies," he explains. "Differences in the stimuli, in how the subject is supposed to respond, in how researchers analyze the data. Providing enough detail for someone to correctly interpret the data while keeping the overall complexity of the system manageable is hard."
Still, just being able to search a brain for anatomical anomalies could tell doctors a lot. If, as Mazziotta puts it, you were treating a 28-year-old right-handed woman with seizures, you could ask the database to compare the patient's scan with those of other 20- to 30-year-old right-handed women, and in that way isolate - with high statistical probability - the aberrant fold that was causing trouble.
Back in the office, Toga pulls up a row of brains on his laptop. They are blue, swirled with red and green, and in this instance they show the progression of Alzheimer's. An advancing seep of red scores the normally blue-green field.
"The red shows which areas of the cortex are losing the most tissue compared to a normal brain," Toga says. If it turns out that fMRIs can pinpoint Alzheimer's before it becomes symptomatic, doctors could start treatments earlier, should they become available. The scans might also provide a way for researchers to monitor the effectiveness of Alzheimer's-inhibiting drugs and establish in successive tests whether disease progression has slowed.
Scientists could also use the database to test currently held beliefs about mental disorders and brain anatomy. Some psychiatrists, for example, associate schizophrenia with asymmetry in an area near the front of our cortex called the anterior cingulate gyms. It has already been found that in normal brains, a box drawn around the ACG is always wider than it is high. In schizophrenic brains, the box is taller than it is wide on the left side - the part of the ACG that controls attention processes.
Eventually, Toga and Mazziotta believe we'll enter a golden age of neuroscience, one that will see as many discoveries on the desktop as in the operating room."Fixing broken brains is a big part of brain research right now," says Mazziotta. "But there's a whole untapped world of taking normal brains and trying to make them really good. We have some tools now that can tell us how to do things that might improve the capacity of our nervous system. I view that as one of the great upcoming challenges for those of us studying the brain: not just to fix the problems, but to try to optimize the machinery."
Already, some researchers suspect that taking certain areas of the brain "offline" may foster savant-like talents. They've seen brain injuries that suddenly allow patients to draw things in perfect proportion, or vividly remember long-forgotten scenes from childhood. It may be that we'll someday employ precisely directed electric pulses to optimize a whole array of submerged talents, temporarily turning ourselves into perfect calculators - or, as Aldous Huxley predicted, perfect drones.
Needless to say, microadjusting our brains would pose philosophical problems that make the current eugenics debate look tame. "Who will control this technology? Who will have access to it?" asks Arthur Caplan, a University of Pennsylvania bioethicist who heads a group on the ethical implications of brain imaging. "Will we see some people losing access to the technology while others zoom ahead?"
And who are we, if our most intimate characteristics turn out to be simply chemical? "These technologies are going to raise big questions of personal identity," predicts Caplan. "In Western culture, we are our brains. But if you start to change your brain - modify it, alter it, enhance it - at what point do you know it's still you?"
Neurologists are decades away from optimizing anything so precisely, and in fact they may never succeed. For starters, there's the variety problem. Sophisticated mathematical warping algorithms can overcome the problem of anatomic variability, and some basic functions have been convincingly pegged to certain brain regions, but it's not clear whether higher functions will be easy to locate, let alone generalize. We may be able to map what part of our brain responds when we see a cat and say "cat," but how do we map the conversation we're holding while we're really thinking about something else? Worse, we don't even know whether a function's location on the folds is what matters, or whether the critical correlation lies in the cyto- and chemoarchitecture, the brain's cellular and chemical microstructure. Then, too, there's the question of how things like the cytoarchitecture and the larger folds are related, if at all.
"To a certain degree, this is modern phrenology," Toga says. "We're looking at shapes and structures in the brain and claiming they mean something, but not long ago we were feeling the bumps on people's skulls and claiming the same thing."
For the moment, the rest of the Human Brain Project is no help, either. Mark Ellisman at UC San Diego is building a neuron database, and Gordon Shepherd at Yale is working on chemoarchitecture - but their work is still under way. And there are other problems to overcome: the matter of imaging technology itself, for instance.
The disappointing truth about many of the new imaging machines is that, while revolutionary, they are still far from refined. An fMRI scan measures blood oxygen, not neural firing - the actual microelectric signals that make brains run. Blood oxygenation levels are recorded over seconds, while neurons fire in milliseconds. When we see a picture of a cat, our brain may play a precise neural arpeggio, but an fMRI test will measure it as an averaged-out blob of activity somewhere in the center of the piano.
"Ideally, what you'd want is to combine different scanning techniques in a way that will give you the best spatial and temporal resolution," says John George, a Human Brain Project researcher at Los Alamos National Laboratory. One possibility is to use an EEG, which measures electrical activity in the brain in milliseconds (but can't locate it precisely). Combining the EEG with fMRI and MRI data might create a more complete diagram.
George, like fellow Human Brain Project scientist Peter T. Fox at the University of Texas, has been working on the problem of putting multiple measurements together. "It's hard," Fox says. "With fMRI you get distortions in the magnetic field whose echoes are hard to correct. In EEG, you have electrical impulses that are reflecting and canceling off a very complicated geometry, making the source of those signals - the active parts of our brain - almost impossible to locate. It's a huge mathematical problem."
Instead of a quick transition to a microadjustable brain, Fox envisions a gradual evolution. "The next step for us will be modeling the actual circuits and systems," he predicts. "That will be the next database."
Back in Berkeley a few days after talking to Mazziotta, I see a flyer taped on the door of a cafe. BURN NEURAL RUBBER, it says. LEARN TO SPEED UP YOUR BRAIN. FREE! At the appointed time, I show up at the corner of Cedar and Bonita streets. The seminar is being held in a large, mealy-carpeted room at the back of a church. Forty-eight plastic chairs sit in rows, but only two of them are occupied; as I sit down, one of the visitors grabs his backpack and bolts. Now it's just me, an elderly woman, and the instructor, a young man with a shaved head and a whispery, hypnotic voice. "Feel gooood," our tutor intones, swaying back and forth a little. "Feel like Einstein. Feel gooood."
Ridiculous, embarrassing, perhaps even a little obscene, and yet - I can't quite bring myself to leave. I do want to be able to think faster, or more clearly, or more consistently, or something. And so I stay in my chair, trying to feel good and thereby to will Einstein's spirit to this desolate space.
Einstein doesn't come. Instead I find myself thinking about the little lilac brain that I saw on one subject's fMRI scan. Somehow, over the eons, our brains evolved into that shape, that computer-generated, ever-evolving slice. There are a million bits of information in that slice, and it's still just a tiny part of a single brain recorded at a particular moment in time.
It makes me think of something Toga said as he handed me a copy of one of his papers. "If we had machines that were sensitive enough, we'd see that the brain is always changing: year to year, hour to hour, minute by minute. By the time you finish reading this article, your brain will already be different."
PLUS
