Can we both learn and unlearn while we sleep? A new study suggests that we can. Both processes occur during different phases of sleep, the research shows.
Most of the wonderful work that our brains do is invisible, and what goes on under the hood has preoccupied neuroscientists for at least the past two centuries.
It is a known fact that sleep and memory are deeply connected. For instance, studies have shown that neuroplasticity – that is, the brain’s ability to retrace new connections between neurons and to forge new pathways that enable us to learn new information – is heavily dependent on sleep. It is during sleep that our synapses relax and regain their plasticity.
Despite some of these studies suggesting that our brains have the ability to learn while asleep, the scientific literature available shows mixed results. Some studies managed to produce evidence in favor of this theory, while others did not.
This is why a team of scientists based in Paris, France, set out to examine in more depth whether or not learning occurs during sleep. They hypothesized that perhaps the reason why different studies produced different results is that they studied different sleep phases, each with a different effect on learning abilities.
The researchers – from the École Normale Supérieure (ENS) and the Paris Descartes University – were led by professor Sid Kouider, head of the Brain and Consciousness lab at ENS – and the findings were published in the journal Nature Communications.
Thomas Andrillon, Ph.D., of the Département d’Études Cognitives at ENS, is the first and corresponding author of the study.
To test their hypothesis, Dr. Andrillon and team played sequences of sound to 28 participants while they were asleep. During sleep, their brain activity was monitored with the help of an electroencephalogram (EEG), which records the brain’s electrical activity.
Using the readings from the EEG, Dr. Andrillon and colleagues looked at three sleep phases: rapid eye movement (REM) sleep; light non-REM (NREM) sleep; and deep NREM sleep.
REM sleep accounts for around 25 percent of any sleep cycle, tending to occur between 70 and 90 minutes after a person falls asleep. Likewise, NREM sleep also has several substages. Because of the different brain waves associated with these different sleep phases, an EEG can detect when a person is going through a specific sleep phase.
Additionally, an EEG can also decide whether the brain is responding to new auditory information or to information that has already been learned.
The researchers examined EEG data. They also asked participants, upon waking, to recognize the sounds that they were played while asleep, and the team measured their learning performance in a series of tests.
Finally, the participants were repeatedly played some of the same sound sequences, in an attempt to see how easily they relearned information that had previously been presented to them.
Overall, the study revealed significant differences between the sleep phases. When participants heard the sound sequences during REM sleep or during light NREM sleep, they were better able to recognize the sounds. By contrast, when they were exposed to the new sounds during deep NREM sleep, they performed significantly worse at the recognizing tests when awake.
These findings were confirmed by EEG markers. And quite surprisingly, the experiments revealed that during deep NREM sleep, the brain seems to not aid learning as well as suppress it.
After waking, not only did the participants find it difficult to recognize the sounds that were played to them while asleep, but they also found it even harder to (re)learn them than entirely new sounds. The role of deep NREM sleep, therefore, seems to be to suppress previous learning.
Speaking to Medical News Today about the findings, the study’s first author said:
“[The] biggest surprise came from brain’s ability to unlearn. Thus, it seems that during sleep, we can either form new memories, learn, or do the reverse: suppress memories and unlearn.”
The findings are significant because they help to harmonize two previously discordant theories. One theory referenced by the authors in the study suggests that sleep’s main function in memory is to consolidate newly acquired information. The other theory sees sleep as a way to discard useless information that would otherwise back up and overwhelm our brains’ capacities.
When asked about the possible mechanisms that could explain the findings, Dr. Andrillon pointed to the neuromodulator acetylcholine as a potential key player.
“Interestingly, the memories formed during light NREM sleep were erased as sleepers transitioned toward deep NREM sleep. We interpret this reversal as the effect in brain chemistry during sleep,” he told MNT.
“Indeed, sleep is characterized by large changes in the concentration of neuromodulators,” he continued. “Acetylcholine in particular is high during both wakefulness and REM sleep but is low during deep NREM sleep.”
“Crucially, acetylcholine can modulate synaptic plasticity. Under high concentration, the activation of a given memory will lead to its strengthening (due to an increase in the strength of synaptic contacts between neurons). Under low concentration, the reverse would occur,” explained Dr. Andrillon.
Based on this, a possible direction for future research would “target the precise mechanisms of learning and unlearning during sleep [focusing on] a model in which the neuromodulator acetylcholine plays a central role in determining brain’s ability to learn or unlearn.”