Interaction

Winter 2007

An entirely new perspective

Courtesy Sean Mackey

Patients in Sean Mackey’s laboratory – in this case, someone with chronic neck and arm pain – learn to control a specific region of their brain as they watch real time fMRI feedback, with a resulting change in their pain.

The human brain comprises 25 billion neurons that communicate through more or less 25 trillion specialized junctions called synapses.

Try taking a picture of that.

The explosion of interest in neuroscience over the past decade is due largely to advances in imaging technology, which enables scientists to do their work and unites researchers from different areas who discover they can all profit from the same gadget.

Brian Wandell

“A vast amount of both invention and research involves being able to see stuff,” said psychologist Brian Wandell, co-director of the Neuroscience Institute at Stanford (NIS) and chair of the Psychology Department. “Visualization is an enormous help. Stanford has been a world leader in this, but now we’re seeing ways that will allow us to be even better.”

After the conversations that led to “Neuro-X” showed that imaging was a priority across the board, NIS launched a neuro-modeling lab.

“This lab should function as a crossroads, a watering hole, a place for people to come together casually and share expertise and enjoy coffee and mingle, which we hope will lead to a new curriculum in computational neuroscience,” said Stephen Smith, a professor of molecular and cellular physiology at the School of Medicine who works on the brain’s synaptic circuitry.

“Steve Smith and I image things at opposite ends of the size scale,” Wandell said. Smith looks at lab samples; Wandell looks at living humans. “You need to work your way up and down the scale. Using MRI, we’re working hard to link the nano slices used in pathology to the larger images, a few millimeters wide, of whole healthy people, creating an integrative imaging program to teach all those techniques to students.”

Real-time functional magnetic resonance imaging (rtfMRI) is being applied to a host of neuroscientific problems. Among them: the alleviation of chronic pain and the ability to make good or bad decisions.

See your pain

Pain is not just a medical condition; Stanford’s online Encyclopedia of Philosophy devotes pages and pages of philosophical analysis to the subject. It is both a biochemical and a profoundly subjective experience. A pain questionnaire widely used by physicians offers patients some 100 adjectives (pulsing, drilling, wrenching, scalding, taut, unbearable, nagging, torturing) to help them describe their condition.

RtfMRI is finally allowing medical and engineering researchers to get close to this intersection of subjectivity and physicality.

Sean Mackey, co-director of the Pain Working Group at NIS, said, “I knew in grad school [studying electrical engineering] that I was going into medicine.

“This is a natural meld of medicine and high tech. I did early work in cardio, then anesthesia, then pain management. I looked at what people were doing and I said, wow, we’re in the dark ages. We’re still fusing people’s backs! We’re giving them the same drugs as 20 years ago! So I got into imaging and systems neuroscience and networking.”

Most famously, Mackey’s group has enabled people to literally see their pain as they are experiencing it while inside an MRI scanner.

“I don’t want to sensationalize the clinical applications of this,” he cautioned. “We’re not selling snake oil. There’s lots of work ahead before we can envision a therapeutic tool.”

But that said, who wouldn’t be excited by preliminary results that show that by seeing their pain, people to some extent can control it?

The ethical pitfalls of such work, however, are legion.

“Think about it,” Mackey said. “People are rewiring their own brains. You could build up soldiers’ capacity to absorb pain. You could improve memory so students would score better on tests. You could alter the cognitive development of autistic kids. These projects all have very significant ethical ramifications. That’s why we need smart people discussing these things, and so far we’re doing very basic science, with no application.”

Pain research at Stanford unites scientists, physicians and engineers, and “the collaboration is fascinating,” Mackey said. “Pain has gotten so complex, it’s impossible for one person to understand it all. Disparate fields are necessary, with everyone thinking outside the box, and remarkable concepts emerge.”

One of Mackey’s occasional collaborators is Fumiko Hoeft, a senior research scientist at the Center for Interdisciplinary Brain Sciences Research and the Department of Psychiatry and Behavioral Sciences, who started off working with imaging as a potential clinical tool. But fMRI (and rtfMRI) has limits. For one thing, it relies on what Hoeft called unreal situations. You can’t create images of how athletes successfully suppress pain, for example, because they’re on the playing field, not in a lab, as they do it. Moreover, there are so many chemicals coursing through their bodies as they play that figuring out cause and effect is practically hopeless.

“RtfMRI is very expensive to run, it relies on unreal situations, it requires lots of training, and health insurance won’t pay for it because it’s experimental,” she said. But she added that functional MRI (though not real-time functional MRI, the kind Mackey uses for his pain research) for pre-surgical planning can finally be reimbursed by health insurance, which she called very exciting news.

“So the cost has to come down,” she said. “Policy has to change in order for this to be applied.”

At Stanford, there is a good chance of planting the seeds of such change because policy and medical researchers often interact.

“There’s a very, very collaborative atmosphere here, more so than at the other places I’ve been,” Hoeft said. “At the rtfMRI meetings we have radiologists, psychiatrists, psychologists, neuroscientists, anesthesiologists, experts on depression, experts on decision-making ... and all these people have the same goal—to make an impact in clinical settings.

“The big thing is that here everyone seems willing to offer their expertise. It’s easier to get in touch with people you don’t know, to get advice from them. I never thought I’d be working on pain. And now I’m collaborating with Sean.”

Risky investments

Brian Knutson

Another of Hoeft’s colleagues is Brian Knutson, assistant professor of psychology. He is part of a new group of researchers called neuroeconomists—psychologists, economists and neurologists who essentially investigate decision-making. The results, they say, can illuminate not only why people make unwise investments but also the origins of certain mental health disorders, including addiction. (It is a field not without controversy, with some economists alleging that proponents misunderstand classical economic theory.)

“In my work, we deal with people with learning disabilities and developmental problems, which led me to rtfMRI, which gave me an excuse to talk with Brian,” Hoeft said. “I had heard he was interested in rewards and punishment, so I thought, hmmm, this might be useful for studying learning. I thought his work sounded absolutely fascinating. So we’re hoping to collaborate. That was sort of unexpected, and I hope it goes well.”

Knutson began working with fMRI in the mid-1990s, one of very few people doing so then. In fact, it was suggested to the young psychologist that he was working hard at not getting a good job. But he landed at the National Institutes of Health, where he found a mentor who was interested in emotions. How to elicit emotions? the mentor wondered. Money, Knutson replied.

“Technology finally caught up with me,” Knutson said, reflecting from his perch in the country’s top psychology department.

Psychologists, of course, are interested primarily in individual human beings, not the aggregate, so Knutson’s collaboration with economists—like Brian Wandell’s with Steven Smith—is a case of technology helping aligned disciplines answer similar questions. Imaging technology, science and policy can thus intersect; the visible brain reactions of someone choosing a risky investment, deciding to skateboard off a building or shooting heroin may be similar and could help in the search for viable solutions to widespread health-related social problems.

“The imaging is getting better,” Knutson said. “I have faith that the brain is not a processing device; it’s a valuation device, and the molecular level of analysis is not necessarily going to give us the best functional prediction.

“Money influences the brain and vice versa; there’s cross-talk. The issue is which of these links—which are potentially infinite—are most important.”

And those links are finally visible, just as the intensity and shape of our pain is beginning to be visible. If seeing is believing, perhaps too it is the first step toward treating conditions that cost us resources and cause us suffering.