Scientists have “faked” sleep in mice by artificially creating the on/off neuronal firing pattern similar to that seen in non-REM sleep. This produced sleep-like effects, including improved learning memory [1].
Can sleep be emulated?
Getting proper amounts and quality of sleep is one of the best things for health and longevity [2], but in modern living, this is not always possible. Sleep deprivation has become a global health issue, and scientists are trying to find ways to emulate sleep and its benefits.
The dominant sleep stage, non-REM (NREM) sleep, which makes up about 80%, is defined by a particular pattern of cortical activity in which neurons alternate between on periods (the whole local population fires together) and off periods (the population falls briefly silent). This slow-wave activity (SWA), concentrated mostly at the deep sleep stage, creates the synchronized waves seen on an EEG. A commonly used readout of “sleep pressure,” it spikes after sleep deprivation and decays as you sleep.
The authors of a new study from University of Wisconsin-Madison, published in Nature Neuroscience, have previously proposed the synaptic homeostasis hypothesis: the idea that being awake and learning strengthens cortical synapses, and sleep’s core job is to renormalize (globally weaken) those synapses, preventing saturation, restoring learning capacity, and consolidating memory [3].
A long-standing question within this framework is whether the on/off pattern is merely a symptom or is itself the mechanism that does the renormalizing. If it’s the latter, then sleep benefits might be recapitulated by artificially inducing SWA. In this new study, the team tested their idea by “faking” sleep in mice. Nature itself offers proofs of concept: in dolphins, fur seals, and some birds, one hemisphere sleeps at a time while the other stays wide awake.
Brain on, brain off
Mice were implanted with two recording probes at mirror-image locations in the two hemispheres. One probe carried an optic fiber, so its local network could be manipulated optogenetically. This means that a light-sensitive protein (an opsin, in this case) has been genetically installed in a specific population of neurons, allowing the researchers to impose a chosen activity pattern on it with millisecond timing. The other rode served as the within-animal “contralateral control.”
The researchers used two mouse models to induce off periods, thus creating the on/off pattern (the neurons themselves do the “on” part). In one model (SOM+ mice), the brain’s own off-switch is triggered, which then inhibits surrounding neurons. In the other (ACR mice), the light triggers the excitatory neurons (the brain’s main “chatters”) directly. The two models produce different SWA patterns but showed similar results across the experiments.
In the first experiment, the mice were sleep-deprived for five hours. During the last 30 minutes of sleep deprivation, light pulses induced NREM-like off periods on the optogenetic rode (optrode) side. During the induction, SWA on the stimulated side rose to NREM-like levels. Then, in the first hour of the actual NREM sleep that followed, SWA was reduced on the optrode side relative to its mirror, showing decreased “sleep pressure,” as if optogenetic stimulation diminished the need for sleep.
“What we’re essentially doing is forcing sleep in a local region of the brain. While that part is solidifying memories and restoring learning capacity, other parts stay aware/vigilant and connected to the environment,” said corresponding author Chiara Cirelli, M.D., Ph.D., a professor of psychiatry at the University of Wisconsin-Madison.
The researchers then tried something different: they lowered the overall firing rate, but without creating the rhythmic on/off pattern. This produced no “less need for sleep” effect. Apparently, it is this NREM-like pattern – not a simple reduction in how much neurons fire – that lowers sleep pressure.
Soon after the induction, synaptic terminals from each hemisphere were probed for excitatory strength. It was lower on the optrode side, and the magnitude matched what 6–7 hours of natural sleep produces. Because no sleep followed the induction, the synaptic weakening could only have been caused by the induction itself, providing evidence that on/off activity leads to – not just accompanies – synaptic renormalization.
Let’s not sleep on it
Finally, the mice were given a floor-texture recognition task, a memory test known to depend on the sensorimotor cortex, which the researchers had stimulated. After learning, the animals were split into three groups: allowed to sleep, sleep-deprived for one hour (the amount of sleep deprivation previously shown to affect learning in mice), or sleep-deprived for one hour with concurrent bilateral off-period induction over the relevant cortical areas. Memory was tested 24 hours later.
Sleepers outperformed the sleep-deprived group, but the off-induction group was rescued back to the sleepers’ level. The amount of sleep a mouse got before the task did not predict performance, ruling out that confounding factor..
“This research further decodes why we sleep and how we learn, which brings us a step closer to understanding how to better prevent and treat cognitive decline,” said Amy Bany Adams, Ph.D., acting director of the NIH’s National Institute of Neurological Disorders and Stroke (NINDS).
Literature
[1] Driessen, K., Squarcio, F., Tononi, G., & Cirelli, C. (2026). Induction of cortical ON/OFF periods in awake mice fulfills sleep functions. Nature Neuroscience, 1-12.
[2] Mazzotti, D. R., Guindalini, C., Moraes, W. A. D. S., Andersen, M. L., Cendoroglo, M. S., Ramos, L. R., & Tufik, S. (2014). Human longevity is associated with regular sleep patterns, maintenance of slow wave sleep, and favorable lipid profile. Frontiers in aging neuroscience, 6, 134.
[3] Cirelli, C., & Tononi, G. (2022, May). The why and how of sleep-dependent synaptic down-selection. In Seminars in cell & developmental biology (Vol. 125, pp. 91-100). Academic Press.
View the article at lifespan.io
