Huge new dataset pushes the boundaries of neuroscience

Zoom Juan Gaertner/Getty Images

A video is played in almost every introductory neuroscience course. It doesn't sound like much - a bar of light moving and rotating across a black screen while background audio pops and crackles like the sound of distant fireworks. Dry stuff, until you learn that the pops represent the firing of a single neuron in a cat's brain, watching the bar move across the screen. When the bar hits a specific spot and is at a particular angle, the popping explodes into a grand finale of frantic activity. The message is clear: this neuron really, really cares about this bar.

The experiment shown in the video was performed by David Hubel and Torsten Wiesel in the 1960s and helped scientists deduce basic principles about how the visual system works. For decades, neuroscientists have stuck thin metal electrodes into the brains of mice, finches and monkeys to spy on individual neurons and figure out what triggers them. There are neurons that react to specific colors or shapes; or at particular places in space or in the direction of the head; or whole faces or individual features.

A driver as powerful as single-cell analysis has proven, "Everyone has always wanted more neurons," says Anne Churchland, professor of neurobiology at the University of California, Los Angeles. Part of the reason was simple statistics: more observations are always better, regardless of the experiment. But scientists have also hit analytical walls when looking at individual neurons. In the prefrontal cortex, the region at the front of the brain that plays a major role in planning, decision-making and social behavior, neurons respond to such a diversity of things - visual characteristics, tasks, decisions - that researchers have been unable to assign any particular role to them, at least individually. Even in the primary visual cortex, the area far at the back of the brain where Hubel and Wiesel made their recordings, only a fraction of the neurons actually fire when the animal looks at angled bars.

With the techniques of Hubel and Wiesel, it was impossible to examine more than a handful of neurons at a time. But engineers have pushed and pushed this capability, culminating in the development of the Neuropixels probes in 2017. A centimeter long and made of silicon, a single probe can listen to hundreds of neurons at once and is small enough for neuroscientists to knock in several. the brain of an animal. At the Allen Institute, a nonprofit research institute started by Microsoft co-founder Paul Allen, they used six Neuropixel probes to simultaneously record from eight different regions of the mouse's visual system. In August, the institute released data from 81 mice, including the activity of about 300,000 neurons. The data is freely accessible to all researchers who wish to use it.

As the largest dataset of its kind ever collected (three times larger than the previous record holder), this release allows researchers to observe huge groups of neurons acting in concert. This unprecedented scale may open up opportunities to understand parts of cognition that have previously eluded the understanding of the scientific community. "We want to understand how we think, see and make decisions," says Shawn Olsen, a researcher at the Allen Institute who played a central role in the project. "And it just doesn't happen at the level of single neurons."

The challenge now is how to analyze all this data. Gargantuan data sets are not easy to manage; even sharing and uploading them can be difficult. But as tricky as the analysis can be, working with such datasets is worth it for many researchers because it allows them to study the brain on its own terms.

For Hubel and Wiesel, the brain was like an assembly line: groups of neurons, each specialized for a specific role, dividing and conquering each task. Show someone a red balloon, and the red-sensitive and circle-sensitive neurons will respond independently. But this approach has never really adapted to the real functioning of the brain - it is so densely wired that no neuron ever acts...