Journey To The Center Of The Mind

Functional MRI is yielding a clearer picture of what thoughts look like

The clanking from within the giant white magnetic resonance imaging (MRI) scanner sounds like somebody banging a wrench on a radiator. "Tommy," a healthy 8-year-old, is halfway inside the machine's round chamber, and his little white-sweat-socked feet keep time with the noise. A mirror on a plastic cage around his head will allow him to see images and video.

During the next 45 minutes, Dr. Golijeh Golarai, a researcher at Stanford University, will ask Tommy to hold his feet still, as she directs a computer to flash pictures at him, including faces of African American men, landscapes, faces of white men, then scrambled faces in a cubist redux. When the boy thinks he sees the same image twice, he pushes a button. The machine is tracking the blood in his brain as it flows to the neurons he is using to perform the assigned task.

When Golarai's software is done analyzing the data, she'll have nothing less than a set of snapshots of the boy's thoughts, pinpointing exactly what part of his brain recognizes faces. "There go those feet again," chuckles Tommy's father, watching from the control room.

There was a time not long ago when an MRI scan of a child's brain would not be a lighthearted affair. The technology has been used since the 1980s to detect injury or disease in patients suffering from symptoms such as seizures, paralysis, or severe headaches. But in just the past few years, manufacturers have developed stronger MRI magnets and more sophisticated software that can sort through a flood of subtle signals the scans collect.

The upshot is that this imaging technology has leaped far beyond its roots looking for lumps and shadows. Psychiatrists are now studying the mental activities of patients suffering from depres- sion and other emotional ills. Basic researchers are rolling thousands of healthy subjects like Tommy into MRI machines in order to explore the very essence of mind, asking them to think, decide, feel, and learn inside the scanners. Pharma companies hope the new "functional" MRI (fMRI) technology will enhance drug development. Law enforcement experts hope it could become a more accurate lie detector. Even our most private tastes and impulses are under scrutiny as so-called neuromarketing takes off.


The benefits in psychiatric diagnosis and treatment alone may be revolutionary. Psychiatrist Ned H. Kalin, director of HealthEmotions Research Institute at the University of Wisconsin at Madison, has been imaging the brains of depressed patients who are taking the antidepressant Effexor. Kalin and his colleagues have found that the drug works specifically on the anterior cingulate, a region of the brain involved in focused attention and conflict. They also found that depressed patients whose anterior cingulates were more active before taking the drug responded better to it.

That's powerful information, since it can take weeks for antidepressants to show an effect and about 30% of patients never benefit. "Physicians in the future will be able to predict which patients will be the best candidates for antidepressants simply by looking at brain scans," says Kalin. Adds Lindsey Carver, Global Manager of MR at General Electric Co. (GE ): "There is an opportunity for fMRI to play a major role in diagnosing psychiatric disorders and monitoring drug treatments." GE, along with Siemens and Philips Medical Systems, make the multimillion-dollar scanners that make fMRI possible.

The same kind of excitement is palpable in the field of developmental neuroscience. Unlike other brain-scanning technologies, such as positron emission tomography (PET), which exposes patients to radiation, fMRI simply tracks the response of brain tissue to magnetic fields. It's noninvasive and believed to be harmless. That means even very young children can be scanned -- and scanned repeatedly as they grow older.

In her current study, Golarai is scanning kids 8 to 18 to examine what part of the brain they use to recognize faces and whether this changes over time. Golarai is also trying to determine if children, like adults, recognize faces of their own race more accurately. For the first time, it's possible for researchers to examine biological evidence about whether humans are hardwired to recognize their own racial characteristics or whether they learn to do it. "It's at the heart of the nature-nurture debate," she says.

Teasing apart the mechanics of learning will be another important contribution of fMRI. Last year, researchers in Stanford psychologist John D.E. Gabrieli's lab showed that dyslexic children have different brain patterns from kids who read normally. But their brains can be rewired to perform better with training. It costs more than $500 to do a scan in the research setting, but "if imaging could really predict who would struggle at reading, it might not be so extravagant" for children to be scanned at very young ages, observes Gabrieli.

Almost weekly, scientific journals trumpet new findings across a wide swath of brain-related turf. In February, Columbia University psychologist Tor D. Wager produced imaging evidence of mind over matter in the journal Science. Researchers have long been divided over the "placebo effect" and whether some patients just report feeling better when they really don't. Wager says when subjects in his imaging trial were given a placebo but believed they had been given a pain-relieving cream, there was less activity in the pain-perceiving areas of their brains after a small electric shock and more in the subjects' lateral prefrontal cortex, a region associated with self-control. That region could represent a target for researchers trying to develop therapies that "engage our own willpower," he explains.

As the technology is applied more broadly, the business of peering into the intricacies of thought and emotion raises ethical concerns. One intriguing yet controversial use of fMRI is probing consumer preferences -- a technique sometimes called neuromarket- ing. At California Institute of Technology, researcher Steven R. Quartz is using fMRI to explore how the brain perceives a cool product vs. an uncool one. Among portable MP3 players, "the [Apple] iPod is by far the market leader. What about that gives us a different kind of signal in the brain?" he asks. Quartz also has formed a company that will offer a service to Hollywood studios, imaging the brains of test audiences as they view movie trailers to see which generate the most brain buzz.

BrightHouse is an "ideation consultancy" in Atlanta that used fMRI facilities at Emory University's neuroscience labs to conduct a study on how consumer preferences for different kinds of products -- ranging from broccoli to trucks -- track with activity in different parts of the brain. CEO Joey Reiman insists the company never scanned people reacting to client products. Nevertheless, consumer watchdog Gary Ruskin, executive director of Commercial Alert, complained to Emory that "it's wrong to use a medical technology for marketing, not healing." Emory officials responded that the research would be made public, and was of "fundamental scientific interest."


There are other ethical concerns -- chief among them privacy. Stanford neurobiologist Donald Kennedy, former head of the Food & Drug Administration and current editor of Science, has urged researchers to tread carefully in collecting brain data. "Far more than our genomes," Kennedy told the Society for Neuroscience last fall, "our brains are us, marking out the special character of our personal capacities, emotions, and convictions...As to my brainome, I don't want anyone to know it for any purpose whatsoever."

Inevitably, insurers and employers will want access to scan data. And so may law enforcement. At the University of Pennsylvania, psychiatrist Daniel D. Langleben has found that specific areas in the front of the brain light up when subjects knowingly tell falsehoods. It's too soon to know how reliable fMRI would be as a truth machine.

One limitation of fMRI today is that most studies are based on small numbers of patients whose results are averaged. That's because there isn't enough of a data repository on how most "normal" brains behave to say definitively what any individual scan means. In other words, instant diagnoses -- not to mention mind-reading -- is many years in the future.

Still, it's thrilling to consider what will become of the scanning information provided by the healthy little boy in the white sweat socks. Researchers believe the data are contributing to a knowledge base that someday will help doctors diagnose and treat a child who is not so healthy. Or it may yield fundamental insights into such phenomena as racism. That potential is what's driving these reconnaissance missions deep inside the human mind.

By Joan O'C. Hamilton in Palo Alto, Calif.

Before it's here, it's on the Bloomberg Terminal.