New research from Carnegie Mellon University and the University of Pittsburgh disrupts what was previously assumed about the brain's flexibility when learning a new task.
In the study, participants were hooked up to a brain-computer interface, which allowed them to control the cursor of a computer with their thoughts, while also tracking neural activity in the brain.
"We have a subject use that for a while, until they get really good at it, and then we change the mapping on them," Steve Chase, associate professor of biomedical engineering at CMU said. "And we watch what happens as the subject learns to regain control of the cursor movement to make the cursor go where they want it to go."
They found the brain recycled old neural patterns while learning within the span of a few hours, rather than creating new, better patterns.
"None of us predicted this outcome," Matthew Golub, a postdoctoral researcher in electrical and computer engineering at CMU, said in a press release. "Learning is far more limited on the scale of a few hours than any of us were expecting when we started this."
Chase says that learning is extremely complex, and scientists still don't know exactly how the process works. But unveiling one clue at a time can help unlock strategies for teaching people how to relearn motor functions after a stroke, for example.
"We'd like to be able to design rehabilitation strategies that would lead to better learning, faster learning," Chase said. "Any time you learn something new, we'd like to be able to structure training experiences that enables the fastest learning possible."
Other researchers involved in the study were Byron Yu, associate professor of biomedical, electrical and computer engineering at Carnegie Mellon, and Aaron Batista, associate professor of bioengineering at the University of Pittsburgh.