The Milinda Panha is a Buddhist text written more than 2,000 years ago. It takes the form of a dialogue between Indo-Greek King Menander I and a Buddhist sage.
At one point, the king is interested in the meaning of dreams. The sage talks about the possible influences of the waking life on dreams and then mentions that they occur in the period between falling asleep and deep sleep. This insight is fascinating, because scientists weren’t able to make much more progress than this until the 1950s. For something so universal, it’s odd that scientists really didn’t start to understand what goes on when we sleep until relatively recently.
The breakthrough came when American physiologists Eugene Aserinsky and Nathaniel Kleitman were making observations of sleeping infants and recording different phases of sleep. In particular, they noticed that the eyes would sometimes jerk and twitch under the eyelids. The pair named this “rapid eye movement” (REM) sleep and later found that it didn’t just occur in babies but in people of all ages. We’ve seen it since in all mammals and in birds, too.
You probably know that REM sleep is closely associated with dreaming. Wake someone during REM sleep and they will almost always be able to tell you about the dream they were having. It turns out that dreams are quite taxing in terms of energy, as measured by oxygen use and glucose metabolism in the brain. The brain uses up at least as much energy — and sometimes more — when in REM sleep compared to when we are awake.
Now, it seems, our understanding of sleep has taken another important step forward. Yu Hayashi of the International Institute for Integrative Sleep Medicine at the University of Tsukuba and colleagues have discovered the identity of the neurons that control REM sleep. “We found a brain region that controls REM sleep, and we hope to discover drugs that affect the activity of this area,” Hayashi says. “This should lead to an increase in REM sleep and, hence, improve sleep quality.”
Hayashi’s colleague, Shigeyoshi Itohara, of the RIKEN Brain Science Institute says there will be ethical concerns to take into account. “However,” he says, “if our understanding reaches a certain level, we might be able to create a novel way to manipulate sleep in humans.”
The pair studied sleeping mice to identify the neural circuit in the brain that regulates REM sleep. They also found that REM sleep controls the physiology of the other major sleep phase, the imaginatively named non-REM (NREM) sleep.
Neuroscientists suspected that REM sleep was controlled from a part of the brainstem called the pons. Hayashi and Itohara observed that many cells in the pons originated from a distant brain area called the rhombic lip during early embryonic development. The team used a new technique to label cells from the rhombic lip in order to track their migration through the brain. They could then activate the cells artificially.
They used a technique ominously called “DREADD,” which stands for “Designer Receptors Exclusively Activated by Designer Drugs.” They used mice that were genetically engineered to have special rhombic lip cells, which could then be controlled by giving the mice a drug. Hayashi’s team were able to suppress REM sleep, leading to an increase in the dreamless blank of NREM sleep.
The scientists then investigated the role of REM sleep in controlling the other phases of sleep. If you record brainwaves at different stages of sleep, you find that NREM sleep has large, slow waves of activity moving across the brain, whereas REM sleep has lots of unsynchronized activity.
Using their master control switch — the DREADD system — the team could shorten or lengthen REM sleep. What they discovered was that size of the slow waves that followed in the next bout of NREM sleep were correspondingly smaller or larger. It’s the first time that anyone knew that there was a hierarchy of sleep, and that NREM was controlled by REM sleep.
“We found that REM sleep enhances slow waves during NREM sleep. Slow waves are important for memory consolidation,” Hayashi says. “Thus, we think one role for REM sleep is to control NREM sleep quality.”
It’s important to understand how and why our sleep patterns evolved. Reptiles, amphibians and fish don’t have two phases of sleep in the same way that mammals and birds do. We know that new experiences we have during the day are stored in the brain that night during NREM sleep. Until the work of Hayashi and Itohara, we didn’t know that REM sleep had such an important role in this consolidation. “I hypothesize that REM sleep has crucial roles in nondeclarative memory and emotional memory,” Itohara says. Nondeclarative memory, also known as procedural memory, relates to unconscious memories and skills, such as knowing how to ride a bike.
REM sleep occurs in bursts, getting longer throughout the night. By the end of the night, adults will have spent 90-120 minutes in REM sleep, totalling 20-25 percent of total sleep time. However, people with depression have more REM sleep, meaning they wake up still feeling tired.
Sleep scientists try to understand the function of REM sleep by waking people up when their eyes start to flicker and their brain waves change. In this way researchers have built up a large set of dream diaries, and they found gender differences in dream subjects. There are obviously a huge variety of dreams but, generally, scientists have found that men tend to dream about fighting other men, whereas women commonly dream about talking with men and other women.
A better understanding of how and why we sleep won’t just help us all feel more refreshed, it will help us understand how the brain works — and eventually help us answer those questions about dreams that Menander was so keen to understand.
Rowan Hooper is the news editor of New Scientist magazine. The second volume of Natural Selections columns translated into Japanese is published by Shinchosha. The title is “Hito wa Ima mo Shinka Shiteru” (“The Evolving Human”). Follow Rowan on Twitter @rowhoop.