A few weeks ago, I taught my son to tie his shoelaces. At first, he found himself a bit exasperated with the task, and we practiced making, crossing and tightening “bunny ears” right up until his bedtime. The next morning, despite the previous night’s frustration and a 10 hour sleep break, he awoke and successfully tied his shoes on the first try.
It’s no surprise. Scientists have known for years that sleep plays a pivotal role in helping us learn motor tasks more efficiently. New research out of Brown University has teased out some of the neural mechanisms underlying those sleep-related learning gains.
Sleep facilitates learning
The research has been remarkably consistent: If you want to learn something, practice, repeat, and sleep. In study after study, researchers have demonstrated that sleep plays an integral part in learning and memory. Similarly, getting too little sleep results in serious, detrimental effects to cognition, including learning. Yet most research regarding learning focuses on wake states, not slumber. How might sleep help us learn?
Sleep is characterized by two main phases: rapid eye movement (REM) sleep, complete with said tell-tale eye movements, and non-rapid eye movement (NREM) sleep. NREM can be further divided into four sequential stages, starting with two stages of light, drowsy sleep followed by two stages of deep or slow wave sleep (SWS). Jan Born, a pioneering sleep researcher at Germany’s University of Tuebingen, has demonstrated that the two stages of SWS, where neurons fire together in a very slow, oscillating rhythm, seem to be when the brain consolidates and stores the previous day’s learning. His lab has shown that, during SWS, neurons fire at the same spatio-temporal pattern as during wakeful learning, which he thinks codes the information to be accessed for the long-term.
“During NREM sleep, particularly SWS, the brain reactivates the memories that were encoded during wake states,” says Born. “These reactivations induce a kind of transformation or redistribution of those memory representations. In more psychological terms, sleep helps to change the quality of memory, helping it to lose its initial episodic features to become more schema-like so it is easier to store and access later.”
While numerous studies have shown that SWS improves learning, Masako Tamaki and Yuka Sasaki, researchers at Brown University’s Department of Cognitive, Linguistic, and Psychological Sciences, wondered how exactly it was doing so.
“We didn’t know the detailed neuromechanisms underlying what’s happening when people are sleeping and how it helps this kind of learning,” says Sasaki. “So we tried to reveal different kinds of activations and see why it’s so beneficial.”
Three types of brain scans
Tamaki, Sasaki, and colleagues recruited 15 participants and put each through rigorous measurements of sleep and brain activity. Using magnetoencephalography (MEG) and polysomnography (PSG), the researchers first took baseline readings of brain oscillation patterns and sleep phase, respectively. Once baseline readings were established, participants learned a simple finger-tapping task on their non-dominant hand. Nine participants then slept for three hours while being scanned again with MEG and PSG and were asked to perform the tapping task again one hour after waking. A control group of six participants did not sleep after learning the task—they were simply asked to perform it four hours after initial learning. Consistent with past studies, those who slept performed the task with more speed and accuracy than controls.
To localize the origins of oscillations measured by MEG, the researchers later used magnetic resonance imaging to scan the brains of each participant. By combining the data from all three scanning techniques, the group was able to precisely determine where, when, and how sleep was aiding memory consolidation. They found that sleep-related improvement on the tapping task was significantly correlated with changes in both fast-sigma and delta brainwave oscillations during SWS in an area of the brain called the supplementary motor area (SMA), an area implicated in sequential learning, and prefrontal and parietal regions.
Tamaki says the results were surprising but do make sense. The data suggests sleep is activating a large cortical network during SWS to consolidate memories from motor learning. “SMA is very close to the brain area M1, which you see activated when you are acquiring a motor memory when you are awake. But while we’re sleeping, deep sleeping, the SMA is working very hard with other brain areas to make that memory long-lasting.”
Born says these results are very interesting and add to the argument that sleep is critical to learning and memory. “The fundamental idea is that you need sleep. You need that off-time, that time where your brain is offline, to produce the reorganization of memories of what you learn. This type of reorganization in the motor system would never happen if the brain were consciously awake. So you must take it seriously that the brain needs sleep so it can reorganize motor skills and other cognitive skills in a way that is beneficial to your everyday life.”
Of course, there are still many questions. One is whether sleep-associated learning is modal-specific (whether the task is a motor skill or sensory, for example). Tamaki plans to use a similar method to look at the sleeping brains of people after they learn a visual task to see if they show a similar pattern of oscillations during SWS. She hypothesizes that the group will see oscillation changes in visual areas of the brain, but cautions that it’s too early to make any specific predictions. She and Sasaki also hope to look more at the way REM sleep may aid in learning.
“We didn’t investigate REM sleep in this paper. But it’s possible that it is also helping with learning. If we look closer, we may see that the role of REM sleep is different from the role of NREM sleep,” says Sasaki. “Human sleep is long and repeats REM and NREM sleep a couple of times a night. So we think REM also definitely has implications for learning and memory. To really understand what is happening, to answer all these questions about how sleep helps us learn, we need to look at both.”