When you reach out to grab a cup of coffee, your brain does something surprising: it uses the same basic recipe for pulling your arm back as it does for stretching it out in the first place. That's the kind of insight emerging from a new study at Northwestern Medicine in Chicago, where scientists have built a tool that acts like a brain signal decoder ring.
Researchers there created a machine learning method called Sparse Component Analysis, or SCA for short. The tool can untangle the messy electrical chatter from hundreds of neurons at once—something that has been incredibly hard to do until now. The work was published in the journal Neuron.
"Historically, people have had an individual neuron-centric view," said Joshua Glaser, an assistant professor in the Ken and Ruth Davee Department of Neurology at Northwestern and senior author of the study. "They asked: what does one neuron do? But over the last couple of decades, there's been a shift toward thinking about groups of neurons working together."
The challenge is that modern technology lets scientists record from many neurons at the same time, but all that data gets tangled up together, making it hard to figure out what's actually happening. Traditional methods often squash this complicated information into simpler signals, hiding the real story underneath.
SCA works differently. Instead of looking at each neuron separately, it finds shared signals that represent underlying brain activities—like ingredients that get mixed together in different recipes. When Glaser and his team tested the method on existing datasets from motor cortex recordings, tiny roundworms called C. elegans, and even artificial neural networks, they kept finding the same pattern: the brain builds complex behaviors from reusable parts.
"What's exciting is that this approach can take these separable computations that are kind of mixed together in individual neurons and actually separate them out," Glaser said. "You can start to discover what I think of as the building blocks of how computations are happening in the brain."
In studies of reaching movements, the team discovered that the same neural components were activated both when extending an arm and when bringing it back. The brain wasn't learning something entirely new each time—it was recombining existing pieces.
The method also showed that different stages of behavior—like planning a movement, executing it, and holding your posture afterward—produce distinct signals in the brain that SCA can separate out. These processes often get blurred together in older analysis methods.
Glaser and his collaborators are already working to expand the approach, hoping to study how these building blocks interact across different brain regions at the same time.
"Our results suggest that a lot of behavior is built out of these reusable pieces," Glaser said. "And this method gives us a way to find them."