While we’re asleep at night, our brain is doing something incredible. The hippocampus and the neocortex, two of its key regions, talk back and forth, processing information for long-term storage—what’s known as memory consolidation. Catching Z’s, as it turns out, is critical for building our mental library. “During sleep, a magical process happens,” says Itzhak Fried, a neurosurgeon at the University of California at Los Angeles.
In a study recently published in Nature Neuroscience, Fried and his team have discovered that this process can be hacked. By gently stimulating the brain’s frontal lobe (part of the neocortex) in sync with the electrical waves of the hippocampus during sleep, the team improved the accuracy of recognition memory—the ability to recognize things previously encountered—in patients with epilepsy. They hope that this sort of stimulation might one day help improve memory for people with other brain disorders—like Parkinson’s or Alzheimer’s.
The idea that sleep is important for memory is not new, and was of interest to the likes of Sigmund Freud. In recent years, scientists have used animals to determine how exactly that process might occur. By looking at the mouse brain, they found that the hippocampus, the brain’s memory hub, generates little bouts of high-frequency “ripples” thought to be useful in long-term memory. Likewise, the neocortex (which governs things like movement and language) and the thalamus (close to the brain’s center) emit more prolonged waves—known as slow waves. According to Gyorgy Buzsaki, a neuroscientist at New York University, a synchronized dance between these ripples and slow waves is what generates successful memories during sleep.
The neocortex’s slow wave, Buzsaki says, is an interplay of “up” and “down” states. “Sometimes, the neocortex is active and is ready to take in [information],” he says. “Other times, it’s just dead—in the down state.” If those hippocampal ripples travel to the neocortex during its down state, the messages are not well-received—which is why coordination between the two parts of the brain is so important.
Fried and his team wondered if enhancing this synchronized dance between the hippocampus and neocortex might improve memory consolidation during sleep. To test this hypothesis, they turned to a select group of patients. These people, who had a form of epilepsy unresponsive to drug treatment, already had electrodes implanted in various areas of their brain for clinical reasons. “It’s a very rare opportunity to look at brain activity from the inside with very high precision, since these electrodes are implanted in brain regions that are important for memory and sleep,” says Yuval Nir, a neuroscientist at Tel Aviv University in Israel and a coleader of the study.
The scientists focused on two electrodes inside the brain: one to measure wave activity near the hippocampus and another in the brain’s frontal lobe to deliver the stimulation. During the active (“up”) state of the slow wave, the electrode inside the frontal lobe would deliver a series of brief and gentle pulses. Nir describes this as “listening” to the hippocampus—using its wave patterns to determine when it was trying to communicate information to other parts of the brain. “Then we gave very precise and minor stimuli in the cortex—sort of like a pinch—to wake it up and make it be attentive so that it could receive the message from the hippocampus,” he adds.
The researchers called this type of stimulation “synced.” They also tested another form of stimulation, called “mixed-phase,” where the electrode delivered pulses into the frontal lobe without regard to activity in the hippocampus.
To see if these types of stimulation would affect memory, the scientists used a test in which the patients were introduced to pictures of famous people, paired with pictures of pets. Each patient subsequently spent one night in which stimulation was given while they were asleep, and one night without any intervention. During the mornings following each night, the patients were shown pictures of the famous people they’d been exposed to the night prior, as well as “lure” images of people they had not seen before. The team assessed whether the patient could recognize the famous person, could match that person to the associated pet, and could reject the lure images.
The researchers found that after the synced stimulation, recognition of the previously learned famous people was better than after the night without intervention. This improvement wasn’t seen in patients who had been exposed to mixed-phase stimulation, indicating that the timing of stimulation was critical to boosting memory.
“The most interesting part that we didn’t predict was that we saw an increase in the ability to correctly reject those falsely introduced images,” says Maya Geva-Sagiv, previously a postdoctoral fellow in Fried’s lab and a study coauthor. This meant that after synced stimulation during a good night’s sleep, the patients didn’t fall for the lure images. Altogether, these results pointed to an increase in memory accuracy after synced stimulation.
This increase in memory accuracy was reflected in the brain’s physiology, too. The team found that the synced stimulation caused an increase in sleep spindles—bursts of neural activity (that look, unsurprisingly, like spindles on an EEG) known to play a role in memory consolidation. According to Geva-Sagiv, patients with the most improvement in memory accuracy also had the largest increase in sleep spindles. The team also found that after the synced stimulation, the brain was more coordinated—hippocampal ripples occurred in tandem with slow waves and sleep spindles.
Nir draws an analogy to two children on a swing set: the hippocampus on one swing and the neocortex on the other. “All we did was look at one of the swings, and based on its movement, time some very delicate pushes to the other swing to make them in sync,” he says. “Really, the way I think about it is that we provided some back wind—we were helping the sleeping brain do what it’s doing anyway, more effectively.”
Michael Zugaro, a neuroscientist at the Center for Interdisciplinary Research in Biology at the College de France, who was unaffiliated with the study, had previously seen improvement in memory consolidation after a related form of synced stimulation in rats. “It’s interesting to see that these general principles that we can find in different species also apply to humans,” he says.
For Buzsaki, more work is needed to see whether this memory consolidation process is similar in healthy humans, and whether a similar improvement in memory accuracy can be achieved. He said the question is whether the quality of improvement was due to regularizing something that is “already perfect in your brain but not so perfect in an epileptic patient” or it’s something that can be optimized in everyone. He and Zugaro both note, though, that implanting electrodes in a person’s brain is an invasive procedure that raises serious ethical concerns when done without demonstrated clinical need.
Regardless, Fried is hopeful that these results can help patients with different types of memory disorders. In the future, he wants to develop this technique as a method for amplifying certain types of memories and possibly even eliminating bad ones—which could be useful for something like PTSD. For Geva-Sagiv, the potential to stimulate further advances for patients has made the publication of the study, which was a long time in the making, worthwhile. “I’m happy that we can now add more knowledge to this very important field,” she says.
