Could Implants in the Brain Revive Memory?

The Defense Department is funding research to see if “neuroprosthetics” implanted in the brain can heal damaged memory.

Darpa brain image
Scientists are looking to restore memory by stimulating neurons deep in the brain. Image courtesy of DARPA

No question that we’ve learned a lot about the human brain in the past 50 years, but, as neuroscientists are quick to acknowledge, the center of our nervous systems remains largely a mystery.  Near the top of the list of its enigmas: how memory works—particularly how to restore it once it’s lost.

Now, though, through a fast-growing branch of neuroscience known as direct brain recording, scientists can follow brain activity in real time through implanted electrodes. The technique could enable them to map how neurons communicate when memories are formed or recalled, which might then make it possible to develop a device that could mimic the memory-creation process by stimulating those same neurons.

It may seem a bit fanciful, but DARPA, the agency that funds cutting-edge research for the U.S. Department of Defense, believes so much in the technology’s potential that last week it announced grants totaling $40 million to see if such a memory “neuroprosthetic” can be developed in the next four years. 

Other scientists are exploring different ways to unravel the mystery of memory.  In a recent study supported by the National Institutes of Health, Roberto Malinow, of the University of California, San Diego, was able to use precisely targeted light to wipe out, and then restore, memories in genetically engineered mice.  And last summer, researchers at the Columbia University Medical Center announced that they were able to significantly improve the memories of old mice by increasing the level of a particular protein in their brains.

But the idea of using implants to revive memory—considered a radical concept when promoted by the likes of University of Southern California neuroscientist Theodore Berger only a few years ago—has captured the attention of the people at DARPA. They see its potential as an innovative and unusually precise way to help damaged veterans—some 270,000 of whom have suffered traumatic brain injuries, often with debilitating memory loss, since 2000—with few other therapy options.

Making Memories

The research project, called the “Restoring Active Memory” (RAM), is designed around the belief that no matter how sweet or how disturbing, every memory is formed the same way: through a sequential action of many neurons.  Disturb that sequence through a traumatic injury, the thinking goes, and memory function becomes blocked.  But what if scientists, following computer programs of the memory process that they’ve developed, could to use tiny implants to work around a damaged area by sending signals to neurons farther along in the circuit?

That’s essentially the goal of the RAM program, which will involve teams from three institutions: University of California, Los Angeles (UCLA), the University of Pennsylvania and the Lawrence Livermore National Laboratory.  Each will have its own focus.

The UCLA team will concentrate on what’s known as the entorhinal area of the brain. Thanks to previous research, they’ve identified it as the gateway to the hippocampus, the brain region most associated with learning and memory.  To give you some sense of how critical the hippocampus is to what makes us human, consider one of the watershed discoveries in brain science, involving a man in the 1950s who had large portions of his hippocampus removed as a treatment for seizures. After the procedure, he was no longer able to create new memories—he couldn’t recall what happened to him each day afterward. 

To zero in on how the hippocampus turns daily existence into memories, the UCLA researchers will first use data from electrodes already implanted in epilepsy patients to develop a computer model of how neurons in that part of the brain communicate during  memory-making.  From that, they will work with scientists at Livermore to create wireless, implantable devices that can replicate the process by stimulating the appropriate neurons.

In Pennsylvania, meanwhile, researchers will take a broader view of how memories take shape, approaching it as a series of complex interactions among different brain regions.  They will work with patients who already have electrodes implanted in multiple areas of their brains, tracking neural activity as those people play memory games on computers.  The goal again is to identify patterns of neuron behavior when new memories are stored or old ones are retrieved, and also attempt to isolate “biomarkers” for when something goes wrong.

Betting on technology

Still, the project has skeptics. 

“We have to keep reminding ourselves that, no, we are not speaking the brain’s secret language — we’re doing some very crude stimulating,” Dr. Anthony Ritaccio, the director of neurosurgery at Albany Hospital, told the New York Times.  “When working with the brain, you have to keep slapping yourself in the face as a reality check; we still understand so little.”

But Justin Sanchez, manager of the RAM project at DARPA, says it’s time to make a big bet on technology.

“We owe it to our service members,” he said, “to accelerate research that can minimize the long-term impacts of their injuries.” 

And, after all, a working memory helps all of us make better sense of the world, as educational psychologist Peter Doolittle says in this TED Talk.

Peter Doolittle: How your "working memory" makes sense of the world

Get the latest stories in your inbox every weekday.