When Stephen Lisberger was a math major at Cornell, a friend lent him a simple book about neuroscience. "That's exciting—that's what I want to do," he said, even though he'd never taken a biology class or set foot in a lab. Fortunately, Albert Fuchs welcomed him into his group at the University of Washington, where Lisberger began to study how the brain works while it is actually working. "It's fascinating, like figuring out how the computer operates while it is running a program I wrote and doing something I told it to do," he explains. "That's different from opening up the computer and cutting out the pieces."
The brain is very good at learning, but little is known about how it accomplishes that complex task. "If you're a baseball player, you can improve your ability to catch a fly ball by practicing," says Lisberger, an avid athlete and sports fan. "So the brain is designed to profit from its experience. Some of the big questions about learning are How does it work? Where does it happen? And could we help the brain learn better?"
The only way to find out, he says, is to study animals with complex brains, such as monkeys. Although some people object to the use of primates in research, Lisberger notes that stringent regulations govern how the work should be performed. "So the question is not whether this work should be done but whether it is being done right," he says.
Lisberger determines how monkeys learn to make simple eye movements. "You can really control and quantify eye movements and get inside the brain to listen in to the electrical activity of neurons," he says. "That provides unique opportunities for getting very precise details about how and where the brain generates those movements."
The vestibulo-ocular reflex (VOR) is one of the eye movements Lisberger studies. When you turn your head, this reflex moves your eyes in the opposite direction so your field of view doesn't turn as well. But if you suddenly lost half the sensory inputs from the vestibular system (the part of the inner ear that triggers the VOR), you would need to relearn to make the right eye movements. To simulate vestibular disease, Lisberger altered the size of the monkeys' world, using goggles that either doubled the visual field or reduced it to a fourth. The monkeys had to move their eyes twice as fast or four times more slowly to regain normal vision.
The electrical circuit that controls the VOR is in the cerebellum, which is part of the hindbrain. Therefore, Lisberger's group recorded the activity of each of the circuit's neurons before and after the monkeys learned to compensate for a reduced or enlarged world. "We used that information to ask a question: At which synapse in the circuit does it seem like there is something going on that is causing learning?" Lisberger says.
Two parts of the cerebellum are involved, the group discovered. One is the cerebellar cortex, the cerebellum's thin, gray surface layer. The other is at the output of the cerebellar cortex, where the cerebellum sends out instructions that eventually reach muscles. "We figured out how those two sites of learning work together to make the VOR work," Lisberger says. "Our contribution is to say that when the brain is in this dynamic state—when it is learning this new motor skill—this is where the changes are happening."
Lisberger complements his investigations of the VOR, a largely involuntary movement, with studies of smooth pursuit eye movements, which we make to track a moving object such as a ball. One of the group's recent findings is that the parts of the cerebellum that learn smooth pursuit eye movements are the same as those that learn the VOR. Therefore, those sites may be responsible for learning a range of motor skills.
The group is also determining why even professional players can't always hit a tennis ball perfectly or shoot a basketball into a hoop. This annoying variation in performance is caused by noise in the brain—random signals that interfere with useful signals. Smooth pursuit eye movements involve input from the eyes to the visual cortex, the part of the brain that processes visual information. The processed information goes to the cerebellum, which then activates the appropriate muscles. Lisberger has shown that the noise that can sabotage smooth pursuit eye movements doesn't come from the cerebellum and therefore must arise in the visual cortex. "If your visual system could just tell you the same thing every time, you could make the same movement every time," Lisberger explains. "But if the motor system is told the wrong things, you're out of luck."
Lisberger believes that understanding how the brain works will eventually help people with neurodegenerative and neuropsychiatric disorders. He predicts that in a hundred years physicians will use genetic or drug therapies to repair the brain. Then therapies grounded in neuroscience will help the patients learn new behaviors. "Because of the critical importance of behavioral therapy, what we are doing is going to be absolutely crucial for making people better."