We have new, strong evidence that REM sleep, the vivid-dreaming phase of rest, plays a key role in the formation of spatial memories. Researchers from McGill University published their findings this month in the journal Science.
The study, performed on mice, offers the first causal evidence for REM's role in memory consolidation, the process of converting newly learned information and experiences into long-term memories. While researchers have known for decades that REM sleep supports memory in some way, they've struggled with the "how" part of the equation. By contrast, scientists have essentially reached consensus over the role of deep sleep, or slow-wave sleep, in memory consolidation. Researchers behind the current study proposed what we can now recognize as a pretty f-ing good hypothesis, writing: "Hippocampal theta rhythms during REM sleep may contribute to memory consolidation by providing a mechanism for strengthening place cells formed during prior wakefulness."
To confirm their hunch, they futzed with rodents' REM sleep. Using a method called optogenetics (lasers, basically), researchers shut off a group of neurons involved in REM sleep called MS neurons. It's not that MS neurons directly control REM, but rather that they regulate activity in the hippocampus, a central region for memory and sleep. Theta waves, one pattern of electrical brain activity, are known to flare up in the hippocampi of mice and men when we're adrift in REM sleep as well as navigating the world around us, be it a wooden maze or concrete metropolis.
To review, the hippocampus plays a huge role in spatial memory. A swath of hippocampal neurons, called "place cells," fire when the rat or human brain identifies its geographic location. Activating place cells is the neural equivalent of dropping a pin on Google maps. Together, place cells and grid cells (another group of neurons in the hippocampus) make up the brain's "GPS System," the discovery of which earned three scientists a Nobel Prize in 2014.
The experiment went like this: First, researchers put mice in an unfamiliar environment and exposed them to two new objects. (Mice, study authors explained in their paper, are intrinsically drawn to unfamiliar sights, smells, sounds and the like.) Then, after introducing mice to their new digs and shiny things, researchers monitored their sleeping brains with EEG. Once the mice entered REM, researchers silenced MS neurons, thereby stonewalling theta waves.
After mice slept, researchers re-tested their object recognition. This time, they left one object in the same place where mice first encountered it. But, they moved the second object to a new spot. If their spatial memory was functioning properly, then the mice should have skipped over the first, un-moved object (because it was familiar) and scurried to the re-located object. But scurry they didn't, suggesting the mice didn't recall making pleasantries with the objects the previous day.
Researchers replicated the experiment outside REM, meaning they silenced MS neurons during other sleep phases and while mice were awake. But, they didn't see the same results. The mice only lost memory of the objects after REM-time neural shuttering.
The outcome? Disrupt REM sleep, forget yesterday's whereabouts. The results, researchers believe, show brain activity during REM as critical to locking in spatial memories.
This study augments research in sleep, neuroscience and cognition. In one 2015 study, for example, cognitive scientists at UCL showed mice a cheese-containing maze before whisking them away for (chemically induced) naps. During naptime, researchers used EEG to record hippocampal brain activity. The next day, the mice returned to the maze and scurried to the previously forbidden cheese.
Not only did the mice know exactly how to reach the treat (based solely on a pre-nap glimpse of the maze); they also exhibited the same pattern of brain activity while navigating towards the cheese as they had during their naps. In this case, researchers didn't zero in on REM brain waves. Rather, they saw their results as evidence of hippocampal place cells playing a role in "future-oriented thinking." The mice, they concluded, hastily mapped out the path-to-cheese, reeplayed their experience while napping and drew on that knowledge the following day. But, in the context of the new McGill study, the mice-dreaming experiment supports the role of REM as important to consolidation of spatial memories.