When a crowd starts to applaud, each person initially does so to their own rhythm. But in some cases, those claps can synchronize, with hundreds or thousands or millions of hands striking in unison.

Something similar happens in the brain. When a single neuron fires, it sends an electrical pulse down its length. But large networks of neurons can also fire together, creating regular cycles of electrical activity that resemble the synchronized applause of a rapturous crowd. Formally, these are called neural oscillations; more colloquially, they’re brain waves.

These waves are classified by how frequently the neurons fire in a single second. If they fire one to four times, that’s a delta wave, which occurs during deep sleep. If they fire 12 to 30 times, that’s a beta wave, which is typical of normal wakefulness. And if they do so 30 to 90 times, that’s a gamma wave, which has been linked to higher mental abilities, like memory, attention, and perception. It’s no surprise, then, that scientists have seen disrupted gamma waves in many types of brain disorders, including injuries, schizophrenia, and Alzheimer’s disease.

But by studying mice, Hannah Iaccarino and Annabelle Singer have shown that these disrupted gamma waves aren’t just a symptom of Alzheimer’s. By restoring normal gamma waves, Iaccarino and Singer actually managed to counteract a hallmark of the disease. In Alzheimer’s, a protein called beta-amyloid gathers in the spaces between neurons, and creates large, harmful plaques. But gamma can apparently mobilize the immune system to clear these plaques.  

This is still preliminary work, but it heralds a completely new approach to dealing with Alzheimer’s—changing neural activity, rather than delivering drugs or chemicals. “It’s so different from what people have tried, but we are very excited about the possibility of bringing this to human testing,” says Li-Huei Tsai, an MIT researcher who led the study.

“It’s potentially transformative,” and not because of its medical implications, says Vikaas Sohal, from the University of California, San Francisco, who was not involved in the study. “Many neuroscientists, including myself, have traditionally thought about gamma oscillations as having a role in how neurons communicate and process information. We haven't really thought about how they could change the biology of cells. Put it another way: If gamma oscillations are part of the software of the brain, this study suggests that running the software can alter the hardware.”

Indeed, that software sometimes gets ignored. According to Tsai, scientists have made huge progress in understanding the genes and molecules that underlie Alzheimer’s. But there’s been relatively less work on how the disease affects the collective activity of neurons—or vice versa.

Her team began by letting mice run through a maze, and recording the brain waves in their hippocampus—a part of the brain that’s involved in navigation and memory. Typically, when the mice hit a dead end, you’d see a short, sharp burst of gamma waves. But when the team studied a breed of rodents that are especially prone to Alzheimer’s, they saw weaker gamma bursts, and less synchronicity between the firing neurons.

Earlier studies had found gamma disruptions in people and rodents with Alzheimer’s. But with one recent exception, these had always looked at individuals who were already in the late stages of the disease. By contrast, Tsai’s rodents had no large beta-amyloid plaques, and were totally asymptomatic. They were early in their disease, and yet they already had gamma problems. So what would happen, Tsai wondered, if they fixed those problems?

To find out, her team used a technique called optogenetics, in which neurons are loaded with light-sensitive proteins so that they can be activated by flashes of light. By sending 40 such flashes a second, the team could create gamma waves in the brains of their mice. And after doing so for an hour, they found that they had roughly halved the levels of beta-amyloid. “We were very, very surprised,” says Tsai.

The team showed that they had mobilized a class of janitorial cells called microglia. These patrol the brain, cleaning up dead cells and harmful proteins. After the gamma burst, the microglia doubled in both number and size, and started swallowing any beta-amyloid that was lying around.

It’s likely that gamma waves have other benefits beyond clearing beta-amyloid. For example, Jorge Palop, from the Gladstone Institute of Neurological Disease, has shown that enhancing these waves can improve the memories of mice with Alzheimer’s. And Sohal has found that the waves can lead to “profound and long-lasting improvements in learning.”

That’s all encouraging, but it’s only useful to human patients if we can induce gamma waves on demand. Optogenetics probably isn’t the solution: It’s a complicated and invasive technique that is only starting to find its way into human trials. Fortunately, Tsai’s team have developed a much simpler procedure.

By simply exposing mice to lights that had been programmed to flicker at a specific frequency—no protein loading involved—they managed to induce gamma waves, excite the microglia, and reduce beta-amyloid levels. They even managed to clear beta-amyloid plaques in older mice that were further along in their disease. After looking at the flickering lights for just an hour a day, for seven days, the rodent’s plaque counts fell by two-thirds, as did the size of the remaining plaques. “That is the most exciting part of this entire study,” says Tsai.

There are still many unanswered questions, though. It's still unclear why gamma waves are disrupted in Alzheimer’s, or how these waves excite the microglia. And so far, her flickering lights can only induce the waves in the visual cortex—a part of the brain that receives signals from the eyes, and that’s relatively unaffected by Alzheimer’s. The team will need to find ways of stimulating gamma waves in deeper parts of the brain.

And they’re trying. Together with Ed Boyden, one of the creators of optogenetics and a contributor to the new study, Tsai has founded a company called Cognito Therapeutics to develop devices for visually inducing gamma waves—“googles, or something similar,” she says.

In the meantime, she urges caution. “I worry that people will think they can use a homemade device to treat themselves, and the correct frequency is extremely important,” she says. “If people use an incorrect one, we don’t know if it could be harmful. For human use, we need to do more work.”