printable versionNew center on brain research connects faculty with multidisciplinary interests
BY KIM SEGALL
There are some 100 billion neurons in the human brain, none of which know how to read, recognize or run—at least, not on their own. But somehow, their collective activity gives rise to all the hallmarks of human intelligence. This phenomenon is the focus of the new Center for the Mind, Brain and Computation.
Launched in the fall of 2006 by psychology Professor James McClelland, the center currently consists of a steering committee, three rooms on top of the Psychology Department and a wealth of ideas. McClelland said the goal of the center is to bridge the gap between the theoretical and experimental sides of brain research.
Along with McClelland, the steering committee includes William Newsome (neurobiology), Sebastian Thrun (computer science and electrical engineering), William Mobley (neurology) and Tom Wasow (linguistics and philosophy). Also on the committee is Brian Wandell, chair of the Psychology Department and co-chair of the Stanford Initiative on Human Health.
"Overall, the idea is to bring together people who are concerned with the mind—that is to say, psychologists and linguists—and people who are concerned with the brain, neuroscientists, and people who are good with computations and understanding computational ideas," McClelland said. "The effort to understand the mind and brain presents one of the most exciting challenges in all of science. To understand it, researchers from many different kinds of backgrounds need to work together."
Wandell said the center has momentum even though it is still in its infancy; he pointed to a grant proposal McClelland made to the National Science Foundation that has received first-round approval. The grant would fund a graduate student training program.
That program is intended to be one aspect of the center. Within a year or two, McClelland said he hopes graduate students will be able to enroll in classes affiliated with the center, and that the training program will give them the tools they need to effectively combine experimental and computational research on the brain.
"The primary focus of the computational side of the training program is in understanding the computations the brain is actually carrying out, such as recognizing a face in the crowd, in spite of all the random noise and other stimuli," McClelland said. "Those are hard problems—there are no machines that can look across and say, 'That's Bill Gates standing over there.'"
McClelland is currently collaborating with Newsome to develop models of how cooperation and competition among neurons in the brain give rise to the decisions people make.
"Say you are trying to choose between a cheeseburger and a caesar salad for lunch," McClelland said. "We want to know how your brain combines the different kinds of information to make this decision. Your decision is affected by the taste and aroma of each of the options, how healthy they are, and how much each one costs. It looks like neurons all over your brain are contributing to help you make such decisions."
Newsome's lab seeks to establish where these neurons are in the brain and what each one of them does, while McClelland seeks to use simulation models to understand how they work together to produce decisions.
In another research project that crosses academic disciplines, Krishna Shenoy, assistant professor of electrical engineering and head of the Neural Prosthetic Systems Laboratory, and his colleagues are researching brain signals that are produced when someone intends to make a specific movement. For people who have degenerative diseases such as ALS (amyotrophic lateral sclerosis) or spinal cord injuries, such research could allow them a far greater level of control over their external environment.
Shenoy, a faculty affiliate of the center, is working on developing a chip that would be implanted in the brain and interpret intention to move. A robot arm would then execute the motion.
It is precisely this kind of research—bringing together experts in fields ranging from neurosurgery to mechanical engineering—that the center hopes to facilitate in the years ahead.
In that sense, the center also is an example of the type of multidisciplinary program that Stanford is encouraging across the university.
"The whole purpose of the center is to facilitate this kind of integration," McClelland said.
Increasingly, science funding is being directed at interdisciplinary collaboration. McClelland and Newsome have joined with mathematicians, psychologists, cognitive scientists and neurobiologists at Princeton, Carnegie Mellon and NASA Ames to win a $7.3 million grant from the Department of Defense's Multidisciplinary University Research Initiative Program to study the brain's basis of decision making.
McClelland makes computer models of the brain's thought process to gain insight into the organ's ability to analyze all the information it takes in. To track thought processes, he and other researchers at the center will use magnetoencephalography, a brain-imaging technique that tracks electrical activity at a rapid speed with sensors in a cap.
McClelland came to Stanford in August 2006, recruited from Carnegie Mellon University by Wandell to strengthen Stanford's position as a leader in mathematical psychology. Before McClelland came on board, the university had lost two professors who were prolific in that field.
McClelland said he tries to make his computer models more closely reflect the current understanding of the brain, coming at it from what he calls the "neural network" approach. "That is to say, really thinking of the process as being distributed over large numbers of simple computing elements, all of which do just a little bit of the task," he said.
Through the collaboration that the center will facilitate, research will focus on the activities of neurons and how these minute interactions affect what is happening at the collective level, Wandell said.
"The importance of something like the center," Wandell said, "is that we will make an attempt to see the bigger picture of what are the goals and objectives of many thousands of neurons—the orchestral sound of the molecules as they play with one another to create an action or a thought or a sensation."
Kim Segall is an intern at the Stanford News Service.