First Activity Map of a Mammalian Brain Completed

In a pioneering pair of studies, researchers have created the first comprehensive activity map of a mammalian brain, fundamentally altering our understanding of decision-making processes.

This large-scale project involved collaboration among a dozen laboratories and analyzed data from more than 600,000 individual mouse brain cells, covering over 95% of the brain. The findings, published in two papers in the journal Nature, indicate that decision-making might engage much more of the brain than previously thought.

The initiative was spearheaded by the International Brain Laboratory (IBL), a collective of experimental and theoretical neuroscientists from Europe and the U.S. They shared a common concern about the existing methods in science.

Matteo Carandini, a neuroscientist at University College London and a core member of the IBL, expressed, “We had a problem with the way science was done.” In earlier studies, different labs tackled the same significant questions about brain function using various behavioral tasks and distinct mice, which led to inconsistencies. Adding variable definitions of brain regions, these differing approaches made it hard to interpret results.

“We wouldn’t know whether we actually agree or disagree because so many things were different,” Carandini noted.

To tackle this, the IBL designed a unified, large-scale experiment, which no single lab could manage alone. The study utilized precision measuring tools and standardized analysis methods, aiming to produce consistent results and address a long-standing challenge in neuroscience: understanding how variations in neural systems relate to behavioral differences.

The project involved 139 mice from 12 labs worldwide, equipped with brain-recording devices called Neuropixels probes, capable of recording activity from around 1,000 neurons simultaneously. The behavioral task was simple: mice would face a screen where a black-and-white striped marker flashed either left or right. If they moved a wheel in the right direction, they received a reward.

Based on conventional neuroscience teachings, one might expect a clear, linear sequence of brain activity: first, the visual cortex would activate, then other areas like the prefrontal cortex, which deals with decisions. This could later connect with memory-related activity before prompting muscle responses.

However, the researchers found that while the visual cortex did indeed activate first, the announcement of decision-related signals was more widespread than anticipated. “We found decision signals and signals related to prior information in way more brain regions than we might have thought,” Carandini remarked, indicating that nearly all studied regions could contribute to whether a mouse received a reward.

In some trials, the on-screen marker was very faint, requiring mice to guess their responses. The second paper focused on how mice drew on past expectations to make these guesses. Surprisingly, the brain activity exhibited a broader distribution across regions than expected.

The IBL’s strategy mirrored significant scientific initiatives in other fields, like particle physics at CERN or the Human Genome Project. Carandini likened the project’s impact to astronomy, where early astronomers could see stars but lacked detail. “Previous neuroscience work was like focusing on just one galaxy while different researchers looked at others and declared their own findings,” he explained. “This project offers a comprehensive view of the night sky all at once.”

Recent technological advancements and enhanced collaboration among labs made this work feasible, but there are still many questions left unanswered. The current findings are correlational, leaving unclear whether the observed brain activities directly cause decisions or merely correlate with the process.

“I think that’s the next frontier,” Carandini said, “is to add causality to the study.”