Sleep appears in the brain as slow waves that surge across the surface at a frequency of about one per tenth of a second – or so we thought.
A new study in mice suggests that there are patterns of brain activity associated with sleep that we have overlooked – and that reflect the state of individual brain cells rather than the collective activity of millions or billions of neurons.
Furthermore, by measuring these hyperlocal, submillimeter brain signals with single-wire electrodes, researchers have found that parts of the mammalian brain may nod off for short naps while other regions remain wide awake.
“It was surprising to us scientists to find that different parts of our brain actually take little naps when the rest of the brain is awake,” says David Haussler, a bioinformatician at the University of California (UC) Santa Cruz and lead author of the study.
For about a century, patterns of electrical activity throughout the brain have been used to quantitatively define the difference between sleep and wakefulness. These brain waves are most commonly recorded using an electroencephalogram (EEG) via electrodes on the scalp.
But Haussler and his team questioned the way we measure sleep and distinguish it from wakefulness, even though there is clearly some overlap in the brains of animals that remain awake even while sleeping, an ability known as unimesispheric deep sleep.
In the 1960s, researchers first suspected and discovered that dolphins and other whales can rest half of their brain while remaining active, sometimes keeping one eye open to watch for predators and maintain contact with other animals in their pod.
Seals and birds also exhibit variations of this half-asleep, half-awake rest – a clever compromise between sleep and survival.
Humans can also exhibit temporary asymmetrical sleep patterns that are similar to, but not identical to, those of animals.
In 2016, researchers at Brown University in the USA found that the left side of the brain of people who slept in an unfamiliar place for the first night reacted more strongly to different sounds than the right side. Once we get used to a sleeping environment, this difference disappears.
“It turns out that the human brain is equipped with a less dramatic form of the monohemispheric sleep found in birds and some mammals,” wrote neuroscientist Christof Koch in Scientific American when these results were published.
Based on the mouse brain, the mixing of waking and sleeping states in humans could be a neurological property that we share with other animals.
Haussler and his team spent weeks collecting data from nine mice that had thin wire electrodes implanted in ten different brain regions and fed this data into an artificial neural network that learned to distinguish between sleep and wakefulness.
The recordings were taken from 100 micrometers (one tenth of a millimeter) thick brain tissue, and the algorithm was able to reliably identify sleep-wake cycles based on short “flickers” in brain cell activity lasting only 10 to 100 milliseconds.
These “hyperlocal” signals indicated that part of the animals’ brains were dozing off while other regions remained active and awake. By chance, the researchers noticed that this happened precisely when the mouse stopped for a split second, almost as if it had “drifted off.”
“We were able to look at the individual points in time at which these neurons fired, and it was quite clear that [the neurons] into another state,” explains Aiden Schneider, a computational biologist at Washington University in St. Louis who led the study with David Parks, a computer science student at UC Santa Cruz.
“In some cases, these flickering effects could be limited to a single brain region or even smaller.”
The team believes that their new method of measuring sleep-wake states could reveal new secrets about our sleep if these “flickers” can be observed by other research groups.
“She [the flickers] break the rules you would expect from a hundred years of literature,” says neuroscientist Keith Hengen of Washington University in St. Louis.
The study was published in Natural neuroscience.