A deep bout of depression that isn’t responding to potent antipsychotic drugs prompts doctors to run more tests on a terrified 13-year-old boy diagnosed with schizophrenia. A battery of EEGs, CAT scans, sleep tests, and neuropsychological exams returns a new verdict.
When he gets home from the hospital, Jeff Hudale opens his dictionary and flips to the strange word he has never heard before. Autism. An incurable illness marked by a morbid tendency to daydream, the definition reads. The scant description is unsatisfactory to Hudale, who hungers for more answers as to why he doesn’t feel and behave like other kids.
“I always knew something was wrong with me and that people perceived me as somewhat unusual,” says Hudale, now 35. “But back then, not as much was known about autism.”
First described in 1943 by American psychiatrist Leo Kanner, autism is a complex, sometimes devastating brain disorder that strikes an estimated one in 150 children born in the United States. People with autism, such as Hudale, often have trouble with social interaction and communication, exhibit repetitive behaviors such as rocking or spinning, and pay obsessive attention to detail. The signs usually develop before age three and can range from mild to disabling.
Some people with autism never learn how to talk or hold a conversation, so they become cut off from the rest of the world. Others avoid eye contact and cannot read people’s moods. They sometimes repeat the same sentence again and again to calm themselves. They might flap their arms to tell you they are happy or hurt themselves to let you know they are not. Although genetics plays a role, the cause of the mysterious disorder isn’t entirely clear, and no cure exists.
Carnegie Mellon neuroscientist Marcel Just didn’t set out to study autism—or the brain, for that matter—but rather to exercise his own mind. As a mathematics major at McGill University in his hometown of Montreal, Canada, Just spent the bulk of his time working incredibly complex, highly impractical algebra problems. A turning point came when he took a required social science course taught by D.O. Hebb, sometimes called the founder of cognitive neuroscience, for whom Just’s endowed chair at Carnegie Mellon is named today.
“There I am, forced to do homework on trivial algebra problems, while this guy is lecturing about really fundamental aspects of the human mind,” says Just, whose office is decorated with pictures of Hebb and the forefathers of artificial intelligence, Herbert Simon and Allen Newell.
Intrigued by the links Hebb drew between the brain and behavior, Just shifted his academic focus and earned his doctorate in psychology from Stanford University. He joined Carnegie Mellon’s psychology faculty in 1972, drawn to the University by its willingness to disregard disciplinary barriers and its emphasis on computational approaches to research.
Just, 59, has thick hair and a winsome smile and owes his trim build to countless miles logged on his recumbent bike. Cycling on trails, he says, is “the closest I get to a meditative state.” He lives in the close-knit Squirrel Hill section of Pittsburgh and has two grown sons following in his scientific footsteps: Adam (CS’98), 29, a computer scientist, and Allan, 24, a graduate student in environmental health.
Just has dedicated his career to figuring out how we know what we know, securing millions of dollars in federal grants and recognition from the American Association for Advancement of Science and the National Institute of Mental Heath for his pioneering work in cognitive neuroscience.
But a gap always remained in his theories of cognition—what actually goes on in the brain when a person processes information?
Answers to the fundamental question of biology began to surface in the early 1990s with the advent of functional magnetic resonance imaging, fMRI technology, which allows scientists to see which brain structures are activated during particular mental operations. fMRI works by measuring the increase in blood flow necessary to keep up with the high oxygen demand in busy parts of the brain.
“It is really like walking into the future—someone enters a room, and seconds later you can see their brain working,” says Just, who directs Carnegie Mellon’s Center for Cognitive Brain Imaging. “It is incredibly dramatic.”
A decade ago, University of Pittsburgh research neurologist Nancy Minshew approached Just about using his newfound fMRI expertise to investigate the neurobiological underpinnings of autism. He agreed without hesitation, eager to explore the brain’s frontiers while serving others in a more tangible way.
“Research can be wonderful, elegant, beautiful, and satisfying in that sense, but it is different when you know it is going to help someone,” Just says. “Then it becomes incredibly motivating.”
In 1997, Just and Minshew launched one of nine Collaborative Programs of Excellence in Autism, an international network formed with $42 million in funding from the National Institutes of Health with the goal of finding a cure for the disorder.
Progress was tough going and slow. The researchers had to set up the fMRI scanner paid for by a $2 million grant from the National Science Foundation, refine their experimental procedures, and develop computational methods to extract meaning from the gigabytes of imaging data they were collecting.
To provide accurate fMRI results, patients must lie completely still on a gurney inside the donut-shaped bore of the scanner’s powerful superconducting magnet. In Just’s experiments, research subjects are asked to perform various high-level thinking tasks such as reading sentences or evaluating geometric shapes, while the machine monitors brain activity. The scientists then analyze the timing of an activity and determine whether it is synchronized among brain regions.
For people with severe forms of autism, it can be impossible to lie motionless in the confined space of the scanner for long periods or to understand the cognitive tasks they are supposed to perform. Consequently, Just’s studies must be carried out using high-functioning subjects such as Hudale with IQs in the normal range, who represent just 15 percent of all people with autism.
Aside from his raspy voice and tendency to flap his hands when he gets nervous, you might not know upon meeting Hudale that he is autistic. He graduated from the University of Pittsburgh with an engineering degree. He has a knack for rapid mental calculations—he can multiply three-digit numbers in seconds. But finding and keeping a job in his field have been difficult; Hudale now works in a clerical position at a bank. He cannot drive, never dated, often gets teased by strangers, and lost thousands of dollars to a con artist who took advantage of his guileless nature.
“I tell people that autism has been an intellectual boost but a social disaster,” Hudale says.
To recruit enough people like Hudale for their studies, Just and Minshew flew participants to Pittsburgh from across the eastern United States. They also built an fMRI simulator that helps people feel comfortable in the real scanner by rehearsing the experience first. Gradually, the scientists began to amass huge volumes of brain data. From the images, a pattern started to emerge. “You would click on the cerebellum and get some very clear correlations with the prefrontal cortex,” Just says. “Two areas of the brain that were six inches apart were dancing to the same tune, and these correlations were lower in autism.”
Six years after forming their partnership, Just and Minshew proposed their theory of functional underconnectivity to describe what is different in the minds of people with autism. The theory first appeared in the prestigious British journal Brain, where the scientists noted that key processing centers in the brains of high-functioning autistic people don’t communicate together in the coordinated manner required for most advanced thinking.
The article explains that individual areas of autistic brains function properly—in some cases better—but they have trouble collaborating with each other, much like a soccer club filled with highly talented players who don’t work well together as a team. Most often, the poor connectivity affects the networks linking together the frontal cortex (used for problem solving, language comprehension, and other high-level thinking) with more posterior regions of the brain.
The article electrified the scientific community.
“The theory really is the glue that binds together all the other little findings in autism,” says clinical neuropsychologist Susan Bookheimer of the UCLA Center for Autism Research and Treatment.
Viewing autism as a system-wide networking problem represents a paradigm shift in the way scientists think about the disorder. Previously, most researchers tried to find a single brain area that was malfunctioning in autistic patients rather than study the brain as a whole system. “We certainly got everyone’s attention in the autism community, and without a doubt, connectivity is still the buzzword of the day,” Minshew says.
Just says that’s in part because the theory is robust enough to account for many, if not all, of the broad range of behaviors affected in autism, especially the paradox of why people with the disorder frequently excel at details yet struggle “to see the forest for the trees.”
For example, some autistic children might become spelling bee champions yet have difficulty understanding the meaning of a complicated sentence or story. That’s because, Just discovered, the part of the brain called Wernicke’s area needed to process individual words is more active in autistic brains but less synchronized with Broca’s area used to integrate language.
“Better at the pieces, worse at the puzzle,” Just says.
Just and Minshew have since published seven more articles that show similar connectivity problems when autistic subjects performed other high-level cognitive tasks and were at rest.
New research looking at the structure of the brain offers an explanation for why the failures happen. The white matter that makes up the neurological “cables” connecting nerve cells’ bodies to one another is disordered in children who develop autism. And when the wiring in the critical networks goes haywire, the brain can’t send electrical signals in proper synchrony. The next goal, Just says, is to better relate the anatomical results to the fMRI data showing how brain communication breaks down in autistic patients.
Although additional studies are needed to buttress their findings, underconnectivity theory has gained widespread acceptance in the autism community. “What surprises me is how much agreement there is on these fundamental models in autism based on white matter abnormalities and functional connectivity, especially because the research is relatively new. And no one ever agrees in science,” Bookheimer says.
Theories don’t matter much, though, for Hudale and other people afflicted with autism, and therapies cannot wait for researchers to finish mapping out every last corner of the brain. That’s why Just hopes psychologists begin using his theory to design new treatments for autism that train higher-order nerve circuits in the frontal cortex to work together with the back of the brain. Cognitive behavioral therapy that encourages the brain regions to communicate better might even reverse the white matter abnormalities underlying the processing problems, he says with optimism. Ultimately, drugs or gene therapies also might be developed to fix the damaged circuitry—and possibly cure the disorder, Minshew says.
For the hundreds of thousands of children facing lifelong struggles with autism, the advances can’t happen fast enough. Hudale says he hopes his own contributions to Just’s research will lead to therapies that improve his life and also help others with autism. He prays each week that a cure for the disorder will be found soon. “I’d love to be able to be considered a normal human being and to have a normal life,” says Hudale, staring at the floor. “And I’ll do whatever it takes to get there.”
Learn more online: www.ccbi.cmu.edu
Jennifer Bails is a freelance writer and a former award-winning newspaper reporter.
