The electrical oscillations we call brain waves have intrigued scientists and the public for more than a century. But their function—and even whether they have one, rather than just reflecting brain activity like an engine’s hum—is still debated. Many neuroscientists have assumed that if brain waves do anything, it is by oscillating in synchrony in different locations. Yet a growing body of research suggests many brain waves are actually “traveling waves” that physically move through the brain like waves on the sea.

Now a new study from a team at Columbia University led by neuroscientist Joshua Jacobs suggests traveling waves are widespread in the human cortex—the seat of higher cognitive functions—and that they become more organized depending on how well the brain is performing a task. This shows the waves are relevant to behavior, bolstering previous research suggesting they are an important but overlooked brain mechanism that contributes to memory, perception, attention and even consciousness.

Brain waves were first discovered using electroencephalogram (EEG) techniques, which involve placing electrodes on the scalp. Researchers have noted activity over a range of different frequencies, from delta (0.5 to 4 hertz) through to gamma (25 to 140 Hz) waves. The slowest occur during deep sleep, with increasing frequency associated with increasing levels of consciousness and concentration. Interpreting EEG data is difficult due to its poor ability to pinpoint the location of activity, and the fact that passage through the head blurs the signals. The new study, published earlier this month in Neuron, used a more recent technique called electrocorticography (ECoG). This involves placing electrode arrays directly on the brain’s surface, minimizing distortions and vastly improving spatial resolution.

Scientists have proposed numerous possible roles for brain waves. A leading hypothesis holds that synchronous oscillations serve to “bind” information in different locations together as pertaining to the same “thing,” such as different features of a visual object (shape, color, movement, etcetera). A related idea is they facilitate the transfer of information among regions. But such hypotheses require brain waves to be synchronous, producing “standing” waves (analogous to two people swinging a jump rope up and down) rather than traveling waves (as in a crowd doing “the wave” at a sports event). This is important because traveling waves have different properties that could, for example, represent information about the past states of other brain locations. The fact they physically propagate through the brain like sound through air makes them a potential mechanism for moving information from one place to another.

These ideas have been around for decades (pdf), but the majority of neuroscientists have paid little attention. One likely reason is that until recently most previous reports of traveling waves—although there are exceptions—have merely described the waves without establishing their significance. “If you ask the average systems neuroscientist, they’ll say it’s an epiphenomenon [like an engine’s hum],” says computational neuroscientist Terry Sejnowski of the Salk Institute for Biological Studies who was not involved in the new study. “And since it has never been directly connected to any behavior or function, it’s not something that’s important.”

The tools researchers use may also have played a part. Today’s mainstream neuroscience has its roots in studying the behavior of neurons one at a time using needlelike microelectrodes. Pioneering researchers in this area noticed the timing of when a neuron fired varied from one trial of an experiment to another. They concluded this timing must not be important and began combining responses from multiple trials to produce an average “firing rate.” This became the standard way to quantify neural activity, but the variability may result from where neurons are in oscillation cycles, so the practice ignores the timing information needed to reveal traveling waves. “The conceptual framework grew out of what a single neuron is doing by itself,” Sejnowski says, but “the brain works through populations of neurons interacting with each other.” Because traveling waves comprise the activity of many neurons spread across the brain, they are invisible to single-neuron techniques. But over the last decade new technologies have appeared that allow many neurons to be monitored simultaneously. “This has given us a very different picture,” Sejnowski says. “For the first time we have the tools and techniques to see what’s really going on—but it’s going to take a generation before it’s accepted by the established neuroscience community.”

Optical methods, like voltage-sensitive dyes, allow researchers to visualize electrical changes in thousands of neurons simultaneously but cannot be used in humans because of the risks they pose. ECoG, however, is commonly used in epilepsy patients to investigate seizures. So the researchers behind the new study recruited 77 epilepsy patients with implanted ECoG arrays and went looking for traveling waves. They first looked for clusters of electrodes displaying oscillations at the same frequency. Nearly two thirds of all electrodes were part of such clusters, which were present in 96 percent of patients (at frequencies from 2-15 Hz, spanning the theta band at 4-8 Hz and alpha band at 8-12 Hz). The researchers next assessed which clusters represented bona fide traveling waves by analyzing the timing of the oscillations. If consecutive oscillations are part of a traveling wave, each will be slightly delayed or advanced, depending on direction of travel. (Think of how people in crowd wave follow one another with a slight delay.) Two thirds of the clusters detected were traveling waves moving from the rear to the front of the cortex. These involved nearly half of all electrodes and occurred in all lobes and both hemispheres of patients’ brains.

The team next gave participants a working-memory task and found traveling waves in their frontal and temporal lobes became more organized half a second after people were prompted to recall information. The waves changed from moving in various directions to mostly moving in concert. Importantly, the extent to which they did this varied with how quickly participants responded. “More consistent waves correspond to better task performance,” Jacobs says. “This suggests a new way to measure brain activity to understand cognition, which can perhaps give rise to new, improved brain–computer interfaces.” (BCIs are devices that connect a human brain to a machine that performs some task, like moving a prosthetic limb.)

These findings should help dispel some researchers’ lingering doubts about the importance of such waves. “The article is a strong contribution to the study of cortical traveling waves, adding to previous work on their role in human cognition,” says psychologist David Alexander of the University of Leuven in Belgium who did not take part in the work. “This really will put to rest any worries that the waves are an artifact of blurring of signal passing through the skull.” He also says the authors make unjustified claims about the novelty of the findings and fail to acknowledge some previous research, however. “Previous work on traveling waves has shown they are evoked during working memory tasks,” he says, pointing to a 2002 EEG study that found the timing of a reversal in direction of theta waves correlated with memory performance. Interestingly, an EEG study Alexander himself published in 2008 found fewer waves moving from the front to the back of the head during a working-memory task in people who had experienced their first episode of schizophrenia, compared with healthy individuals, suggesting differences in traveling wave behavior can be related to psychiatric symptoms. He also claims the methods the team used to assess traveling waves are similar to those he used in a 2016 study. “Alexander’s work is really interesting, but it’s not clear his findings involve the same signals as our paper,” Jacobs notes. “He reported patterns that literally involve the entire brain whereas our findings were limited to particular regions.” Jacobs also points to differences in recording techniques and the nature of recorded signals.

Confirming the importance of traveling waves creates new horizons in neuroscience. “Finding that such a wide range of oscillations are traveling waves shows that they involve coordinating activity across different brain regions,” Jacobs says. “This opens key new areas of research, such as understanding what exactly this coordination consists of.” He thinks the waves propagate information, at least in the context of the current study.

Another idea holds that waves, by repeatedly moving across patches of cortex, modulate the sensitivity of neurons so as to sweep a “searchlight” of attention across, say, the brain’s visual processing area. “The concept of a traveling wave is closely tied up with the issue of how you maintain the cortex in the sweet spot where it’s maximally sensitive to other inputs and able to function optimally,” Sejnowski says. Interest in traveling waves will undoubtedly continue to increase. “What you’re seeing right now is a transformation from one conceptual framework to a completely new framework,” he adds. “It’s a paradigm shift.”