A New Way to Understand the Brain's Intricate Rhythm

 Interesting, but someone much smarter than me will have to see how this knowledge can help stroke survivors.

A New Way to Understand the Brain's Intricate Rhythm

Researchers have found evidence in humans that individual neurons time their firing to a deeper beat. But there’s a mystery: What does it mean?

Photograph: Getty Images

Today, when researchers spend long hours in the lab performing tricky experiments, they might listen to music or podcasts to get them through the day. But in the early years of neuroscience, hearing was an essential part of the process. To figure out what neurons cared about, researchers would translate the near-instantaneous signals they send, called “spikes,” into sound. The louder the sound, the more often the neuron was spiking—and the higher its firing rate.

“You can just hear how many pops are coming out of the speaker, and if it’s really loud or really quiet,” says Joshua Jacobs, associate professor of biomedical engineering at Columbia University. “And that's a really intuitive way to see how active a cell is.”

Neuroscientists don’t depend on sound anymore; they can record spikes with precision using implanted electrodes and computer software. To describe a neuron’s firing rate, a neuroscientist will choose a time window—say, 100 milliseconds—and see how many times it fires. Through firing rates, scientists have uncovered much of what we know about how the brain works. Examining them in a deep region of the brain called the hippocampus, for example, led to the discovery of place cells—cells that become active when an animal is in a particular location. This 1971 discovery won neuroscientist John O’Keefe a 2014 Nobel Prize.

Firing rates are a useful simplification; they show a cell’s overall activity level, although they sacrifice precise information about spike timing. But individual sequences of spikes are so intricate, and so variable, that it can be hard to figure out what they mean. So focusing on firing rates often comes down to pragmatics, says Peter Latham, a professor in the Gatsby Computational Neuroscience Unit at University College London. “We never have enough data,” Latham says. “Every single trial is completely different.”

But that doesn’t mean studying spike timing is pointless. Though interpreting a neuron’s spikes is tricky, finding meaning in those patterns is possible, if you know what you’re looking for.

That’s what O’Keefe was able to do in 1993, more than two decades after he discovered place cells. By comparing the timing of when these cells fired to local oscillations—overall wavelike patterns of activity in a brain region—he discovered a phenomenon called “phase precession.” When a rat is at a particular location, that neuron will fire around the same time that other nearby neurons are most active. But as the rat keeps moving, that neuron will fire a little bit before, or a little bit after, the peak activity of its neighbors. When a neuron becomes increasingly out of sync with its neighbors over time, it’s exhibiting phase precession. Eventually, since the background brain activity follows a repetitive, up-and-down pattern, it will get back in sync with it, before starting the cycle again.

Since O’Keefe’s discovery, phase precession has been intensively studied in rats. But no one knew for sure if it happens in humans until May, when Jacobs’ team published in the journal Cell the first evidence of it in the human hippocampus. “This is good news, because things are falling in place across different species, different experimental conditions,” says Mayank Mehta, a prominent phase precession researcher at UCLA, who was not involved in the study.

The Columbia University team made their discovery via decade-old recordings from the brains of epileptic patients that tracked neural activity as the patients navigated a virtual environment on a computer. Epilepsy patients are often recruited for neuroscience research because their treatment can involve surgically implanted deep brain electrodes, which give scientists a unique opportunity to eavesdrop on the firing of individual neurons in real time.

For this experiment, volunteers were asked to “drive” virtual passengers to stores within the game. Even though the people weren’t literally going anywhere, they were imagining themselves navigating through space. Salman Qasim, a postdoctoral fellow in Jacobs’ lab and the study’s lead author, looked for evidence that their place cells would do what rats’ cells do: shift the timing of their firing with respect to their neighbors’ activity as the person “drove” around. 

But that pattern was far from obvious at first. There’s an important reason why no one had found phase precession in humans before, even though the data had been available for years: Brain activity in the hippocampus is messier in people than in rats. Rats show an obvious wavelike pattern that repeats seven or eight times per second. Those oscillations, called “theta rhythms,” are much less clear in humans. “It’s much more sparse, it’s much more fickle,” Qasim says. “It changes a lot in frequency, it pops up here and there.”

While rat researchers can look for a neuron that fires just a bit more frequently than those regular oscillations, the same approach doesn’t work in humans. So rather than looking for activity with a very specific frequency, Qasim screened for irregular oscillations happening between two and 10 times per second, and then looked for neurons that got increasingly out of sync with them. Now the evidence of phase precession jumped out at him. Out of all the hippocampal neurons that were sensitive to location, more than 10 percent showed phase precession—as the person “drove” through space, the timing of their firing shifted progressively earlier, compared with the background rhythm.

Qasim found something else too: Some cells that weren’t sensitive to location also seemed to precess—specifically, when the subject was dropping off passengers at a designated store. In other words, these neurons seemed to be sensitive to goals, not space. He was excited to find phase precession associated with something so abstract, but he wasn’t surprised. “When you think about where you are, you often think about where you are physically in space. But you also think of where you are in a more abstract way: Where are you in life? Where are you on the path to certain goals?” he says. “We do that every day.”

Notably, there was something these cells did not do. Although the goal-sensitive neurons shifted their timing when the person neared their goal, the majority of them didn’t increase their firing rate. Even the place-sensitive ones didn’t show a very clear firing rate pattern, Mehta says. (He wasn’t surprised, either: When his team ran a similar experiment with rats in a virtual environment, phase precession remained intact, even when the firing rate code got fuzzier.) This could indicate that phase precession plays a crucial role in representing location and progress toward an abstract goal, at least in virtual environments.

This was one of the first explanations for the role of phase precession when O’Keefe first discovered it in the ’90s, says Kate Jeffery, a professor of behavioral neuroscience at University College London. A neuron’s firing rate provides coarse, approximate information about an animal’s location—but the exact degree to which a neuron is out-of-sync with the theta rhythm can be a bit more precise. “It’s a way of taking the [firing] rate code of a neuron and chopping it into smaller bits and increasing the resolution,” Jeffery says.

Another theory is that phase precession might be essential to learning. It allows neurons associated with a sequence of locations in space to fire one after another within the timespan of a single theta oscillation—a fraction of a second—as opposed to the several seconds it would take the animal to actually traverse that space. Since the brain’s learning machinery also works on a very fast timescale, phase precession might allow animals to learn sequences.

There’s also the possibility that it doesn’t mean much at all. “Whether the brain uses it or not, we have no idea,” Latham says. Mehta, for example, has shown that if overall brain activity follows an oscillating pattern, and if individual neurons increase their firing rate in particular locations, those neurons can automatically exhibit phase precession. In other words, it could be nothing more than an automatic consequence of the way that a brain works—like the way exhaust is automatically emitted from a car as a function of the engine running, Mehta says.

“If I had to put money on it turning out to be important or not important, I just don't know which way I’d go,” Jeffery says. “I think it’ll probably be another 20, 30 years before we know.”

But Qasim thinks that, given how distinct the pattern is, it’s likely to be serving an important role. “Anyone who looks at brain activity as much as we do knows that it’s often a chaotic, stochastic mess,” he says. “So when you see some order emerge in that chaos, you want to ascribe to it some sort of functional purpose.”

The answer will likely turn out to be a function of what animals, including people, need for survival, Mehta believes: “My philosophy is, if there is something that can be used, you bet evolution will use it.”