What If You Didn't Need to Sleep?

Why the question matters

The day has 24 hours. We usually get sixteen.

Sleep takes roughly a third of a life. We keep trying to buy those hours back with light, coffee, shifts and discipline. Fatigue can be postponed. The biological work of night still has to happen.

A 2026 study makes the question less fantastical: if part of that work can happen while the brain stays awake, could we eventually sleep less, or one day, not at all? That answer remains far away. The experiment shows something narrower but remarkable: in mice, selected local jobs of sleep can begin without the animal falling asleep.

During deep sleep, large groups of neurons switch on and off together in a slow rhythm. Researchers recreated that rhythm in one small area of awake mouse brains. Later, the area behaved as if it had paid part of its sleep need. In a separate experiment, the same technique preserved the memory of a floor texture learned before sleep deprivation.

The researchers also tried simply reducing neuronal activity, without the rhythm. The effect disappeared. A pause alone was not enough; the alternation between activity and silence mattered.

These results came from three groups of mice and three different protocols. The animals slept after some tests. Dream sleep, whole-brain coordination, immunity, metabolism and long-term effects remained outside the experiment.

What we can say

Local on/off periods fulfilled measured cortical sleep functions in awake mice.

What we cannot say

Humans can remain permanently awake without cognitive or bodily cost.

Open the starting paper

ON / OFF

Three states, one question

Switch states and follow the connection to the environment, neuronal rhythm and sleep debt.

Wake

Environment: Connected to the environment
Dynamics: Usually no synchronized off periods
Sleep pressure: Sleep pressure accumulates

As wake extends, local silences and performance lapses can begin to appear spontaneously.

What the researchers actually did

The study tested the same idea in three ways.

The results do not come from one continuous sequence. The memory test used different mice and lasted one hour. The other two tests examined one brain area after five hours awake.

Common questionCan cortical on/off periods do useful work while the animal remains awake?separate cohorts and protocols

A

Did the stimulated area feel more rested?

5 h awake
unilateral induction
recovery sleep

B

Did connections between neurons change?

5 h awake
unilateral induction
tissue before sleep

C

Could one memory be preserved?

learning
1 h: sleep / deprivation / induction
recall after 24 h

A

Did the stimulated area feel more rested?

Mice were kept awake for five hours. During the final 30 minutes, researchers induced the slow rhythm in one side of the brain, then allowed the animals to sleep.

Measured: They compared activity in the stimulated area with the matching area in the other hemisphere. Eighteen mice were used for most analyses.
Licensed conclusion: During recovery sleep, the stimulated area showed fewer local signs of fatigue. The animal as a whole still slept.

B

Did connections between neurons change?

In another group, researchers repeated the five-hour protocol and examined the tissue before the mice fell asleep.

Measured: They measured GluA1 and pGluA1, two molecular clues to the strength of selected connections between neurons.
Licensed conclusion: The clues fell in the stimulated area. That suggests a local reset; it does not show that every connection was restored.

C

Could one memory be preserved?

Other mice learned to recognize a floor texture. They then slept, stayed awake for one hour, or stayed awake while the slow rhythm was induced in task-related areas.

Measured: After 24 hours, researchers measured exploration of a new texture: 9 rested mice, 13 sleep-deprived mice and 8 mice with induction.
Licensed conclusion: The final group remembered the texture as well, statistically, as the group that slept. This was one task, not memory as a whole.

The study does not show that the mice could continue without sleep. The authors say the brain may still need to disconnect from the outside world to restore itself and organize memories at large scale.

Sleep is a portfolio of jobs

The paper touched three rows in a much larger table.

Filter by level. “Direct” means measured in the 2026 experiments; “indirect” means connected to the result without full replacement; “untested” means the paper cannot answer.

Function	Scale	2026 study	Exact measure or limit
Reduction of local sleep pressure	Local cortex	Directly tested	Slow-wave activity and synchrony later fell in the stimulated area.
Markers of excitatory synaptic strength	Local cortex	Directly tested	GluA1 and pGluA1 fell on the stimulated side in a separate molecular protocol.
One M2/S1-dependent sensorimotor memory	Distributed networks	Directly tested	A floor-texture recognition task was rescued in a separate behavioural protocol.
Episodic memory and systems integration	Whole brain	Untested	Hippocampal, thalamic and neocortical coordination was not replaced.
REM and affective regulation	Whole brain	Untested	The intervention mimicked NREM elements and did not test REM functions.
Long-duration vigilance	Whole brain	Untested	The protocols were short and the animals slept after intervention.
Immunity and inflammation	Whole body	Untested	No immune measures were included.
Metabolism and hormones	Whole body	Untested	Energy balance and endocrine function remained outside the study.
Cardiovascular function	Whole body	Untested	Long-term vascular or autonomic effects were not assessed.
Development and brain maturation	Whole brain	Untested	The experiments used adult mice.
Clearance of brain metabolites	Whole body	Disputed mechanism	The role of sleep in clearance is actively disputed, and the 2026 study did not measure it.
Alignment with the internal clock	Circadian time	Untested	Removing sleep need would not automatically remove circadian rhythms.

A technology can reduce sleep only as far as the function it replaces. The rest of the debt remains somewhere in the system.

A history of borrowed wakefulness

Humanity gained hours through watchkeeping, light, drugs, discipline and strategic sleep.

The timeline follows strategies, not a linear story of progress. Old and new solutions still coexist in the same night.

Before the city

Published study

The group sleeps while someone remains awake

Among Hadza participants, chronotype variation produced distributed night vigilance: in 99.8% of sampled epochs, at least one person was awake or in light sleep.

1Samson DR et al

1735–1758

Move the clockDocumented history

Sleep becomes a moral test of industry

Benjamin Franklin’s almanac links early bed and early rising to health, wealth and wisdom. Other maxims in the same tradition treat sleeping time as economic waste.

1Benjamin Franklin 2Benjamin Franklin

19th century

Move the clockDocumented history

Electric light extends the sellable day

Industrial time separates work from natural light and turns night into programmable capacity. Edison cultivates a short-sleeper reputation, while historical sources also document frequent brief naps.

1Smithsonian Lemelson Center

1938–1945

Force arousalDocumented history

Pervitin and Benzedrine enter war

Methamphetamine spreads through Nazi Germany in military and civilian life. The RAF moves from prohibition in 1939 to cautious operational approval in 1942. Wakefulness arrives with medical and command concerns.

1Snelders S, Pieters T 2Pugh J 3Pugh J

1964

Force arousalDocumented history

Randy Gardner turns deprivation into spectacle

The teenager remains awake for about 264 hours under observation. The episode becomes record culture, while microsleeps and imperfect verification show why the maximum is not a clean biological limit.

1Ross JJ 2Guinness World Records

1990–2004

Insert recoveryPublished study

Aviation mixes pills with planned sleep

Simulator studies and B-2 mission data show a combined practice: dextroamphetamine, modafinil, caffeine and naps. Drugs attenuate some losses; mission profile determines whether sleep is possible.

1Caldwell JA et al 2Caldwell JA et al 3Caldwell JA et al

2003

Mask pressurePublished study

Six hours a night builds a debt people do not fully feel

Across a 14-day experiment, chronic four- or six-hour sleep opportunities produced cumulative deficits. Subjective ratings lagged behind objective decline.

1Van Dongen HPA, Maislin G, Mullington JM, Dinges DF

2004

Published study

A 24-hour shift becomes a patient-safety problem

Reducing extended physician shifts reduced serious medical errors in intensive care units. Institutions can pay biological debt through someone else’s outcome.

1Landrigan CP et al 2Lockley SW et al

2011–2017

Replace a functionPublished study

The awake brain begins to sleep locally

In rats and then humans, local silences or slow activity preceded performance errors during deprivation. Behavioral wake does not guarantee uniform wake in every circuit.

1Vyazovskiy VV et al 2Nir Y et al

2016

Published study

Frigatebirds sleep in flight, sparingly and selectively

Birds were recorded sleeping unihemispherically and bihemispherically in flight. Total sleep was tiny compared with land, showing that temporary performance does not erase later recovery.

1Rattenborg NC et al

2017

Mask pressureDocumented history

A platform says its rival is sleep

Reed Hastings calls sleep a large pool of time with which Netflix competes “on the margin.” The remark compresses attention economics: growth can come from hours the body reserves for withdrawal.

1Reed Hastings, Netflix Q1 2017 earnings call, reported by The Guardian and NPR: “We’re competing with sleep, on the margin

2025–2026

Force arousalStatus on 17 July 2026

Orexin agonists try to repair the wake system

Oveporexton improved wakefulness, sleepiness and cataplexy in narcolepsy type 1 and entered priority US review. The target is an orexin-deficiency disease, not removal of sleep in healthy people.

1Dauvilliers Y et al 2Takeda

2026

Replace a functionStarting paper

Researchers pay one local sleep function during wake

Induced on/off periods in mouse cortex reduce local pressure and rescue one memory task. For the first time in this timeline, the intervention aims beyond the feeling of wakefulness.

1Driessen K, Squarcio F, Tononi G, Cirelli C

The arsenal

Each instrument buys a different kind of hour.

Select an intervention. The comparator describes mechanism and limits, with no dosing or personal-use advice.

Caffeine

Common use

What it changes: Blocks adenosine receptors and temporarily reduces sleepiness.
What evidence shows: Can improve vigilance and reaction time; effects depend on dose, timing and tolerance.
What remains unpaid: Does not reproduce NREM, REM, full consolidation or bodily recovery.
Costs and limits: Can delay the next sleep, cause tremor or anxiety and make alertness feel better than it is.

This section is educational. It provides no instructions for dosing, combining or using medicines outside authorized medical care and protocols.

Stories from the edge of night

When sleep has to wait.

Mission94 sorties

35.3 hours inside a B-2 bomber

During Operation Iraqi Freedom, 75 pilots flew 94 sorties. On 16.9-hour missions, dextroamphetamine was used on 97% of sorties and naps on 13%. On 35.3-hour missions, the pattern partly reversed: stimulant on 58%, naps on 94%. When cockpit and mission allowed sleep, sleep remained the dominant countermeasure.

1Caldwell JA et al

Simulator37 h awake

Modafinil keeps the aircraft nearer the line

Ten F-117 pilots received 100 mg after 17, 22 and 27 hours without sleep. Six of eight maneuvers degraded less than with placebo. Accuracy still remained roughly 15–30% below baseline in many conditions. The stimulant bought performance, not normality.

1Caldwell JA et al

War1939 → 1942

A pill becomes command policy

The RAF began the war with a ban and arrived at cautious approval. The question was not only whether Benzedrine kept crews awake, but who decided, when it was justified and what happened to aircrew wellbeing after landing. The same questions follow every wake-extending technology.

1Pugh J

Record11 days

A teenager and 264 hours of wake

Randy Gardner stayed awake in 1964 under observation. Attention, mood and perception changed, followed by recovery sleep. Guinness later stopped monitoring deprivation records for safety reasons. Microsleeps make the maximum difficult to verify in any case.

1Ross JJ 2Guinness World Records

Animal0.69 h/day

Frigatebirds fall asleep above the ocean

On flights lasting up to ten days, frigatebirds slept in short bursts, sometimes with one hemisphere. Average sleep in the air was under one hour per day, about 7% of land sleep. Evolution can compress sleep under extreme conditions; the bird resumes long sleep after return.

1Rattenborg NC et al

Genetics4–6 h

Some people really need less

Rare families with DEC2, ADRB1, NPSR1 and SIK3 variants show natural short sleep. They are not ordinary people trained to ignore fatigue. Their biology starts elsewhere, and each variant describes a molecular route, not a training recipe.

1He Y et al 2Shi G et al 3Chen H et al

Hospital24+ h

Sleep debt enters the patient chart

In a randomized intensive-care study, a traditional schedule with extended shifts produced more serious medical errors than a schedule without shifts of 24 hours or more. An institution can normalize deprivation while the consequence lands in another person’s decision.

1Landrigan CP et al

The myth cabinet

Sleep attracts stories precisely where measurement is hard.

Open each drawer. The verdict separates tradition, experimental result and the unknown.

01“Before electricity, people naturally slept eight uninterrupted hours.”Too simple

Three preindustrial societies measured by actigraphy averaged roughly 5.7–7.1 hours and fell asleep several hours after sunset. Historical and ecological settings differ; there is no single ancestral template.

1Yetish G et al

02“Everyone used to have a first and second sleep.”Disputed

Historian A. Roger Ekirch documented many references to segmented sleep. Critics argue the evidence does not establish that the pattern was predominant. The dispute remains open.

1Boyce N 2Ekirch AR

03“I adapted to six hours.”Self-assessment can lag

Under chronic restriction, performance keeps deteriorating while subjective sleepiness does not keep pace. Felt adaptation and neurobiological adaptation are different measures.

1Van Dongen HPA, Maislin G, Mullington JM, Dinges DF

04“Coffee or modafinil pays the debt.”They sustain wake, not rebuild night

Countermeasures can improve vigilance and selected tasks. The literature does not show that they reproduce NREM, REM, immune, metabolic and cardiovascular effects or systems memory integration.

1Caldwell JA et al 2State-of-the-art review on the use of modafinil as a performance-enhancing drug in military operationality 3US National Highway Traffic Safety Administration

05“Glymphatic means sleep definitely washes the brain.”Actively disputed mechanism

The influential 2013 study reported increased influx and clearance during sleep. Measurements published in 2024 found reduced clearance for the tracers used. Method and molecule matter; the slogan is more settled than the science.

1Xie L et al 2Miao A et al

06“Open eyes mean an awake brain.”False at local scale

During deprivation, neurons or regions can enter sleep-like periods before an error. Global behavioral state hides an internal geography of fatigue.

1Vyazovskiy VV et al 2Nir Y et al

07“A genius slept four hours, so success requires the same.”Biography selects what it sees

Edison promoted his ability to work through the night, while historical sources document brief naps and cots in the laboratory. A sleepless reputation can conceal distributed sleep.

1Smithsonian Lemelson Center

The economy of eight hours

When night disappears, who gets the time?

Allocate up to eight hypothetical hours. The output does not predict productivity or health; it shows which institutions have motive and power to claim the new resource.

8hours

Available hours0 unallocated

allocated: 8 / 8

More learning time could expand autonomy. Unequal access to the technology would turn time into a new class advantage.

Paid workcontract hours, a second job, availability

0.5

Carechildren, relatives, home and emotional labor

1.5

Learningstudy, art, practice and research

2

Social lifefriends, community, politics and ritual

1.5

Media and consumptionstreaming, games, shopping and platforms

1.5

Protected unproductive timewithdrawal, daydreaming, quiet and nothing scheduled

1

The first sleepless brain may learn to sleep in pieces.

The 2026 paper showed that activity once treated as exclusive to sleep can be moved locally into wake and can perform measurable biological work. The result is narrow, elegant and important enough without additional promises.

The history of stimulants shows how quickly continuity gets confused with restoration. The body can be kept on task, while the invoice arrives as errors, inflammation, metabolism, mood, accidents or later sleep.

A technology that replaced more functions would open a political question before it solved a personal one: do the eight hours become freedom, work, care or market? Sleep is biological infrastructure and the last interval the economy cannot fully occupy.

Anyone who claims to replace sleep must prove two things: every function has been paid, and the night was not sold before it was freed.

Sources, method and limits

Audit every bridge from mouse to history to scenario.

Evidence search: 17 July 2026. The starting paper is an animal study. Oveporexton was investigational and under regulatory review on the evidence date. Claims about historical segmented sleep and brain clearance are presented as active disputes.

This material is educational and does not replace medical advice. No independent clinical review was commissioned for this edition. Seek professional or emergency help in your jurisdiction for involuntary sleep episodes, severe deprivation, suspected sleep disorder, mania, psychosis or driving risk.

Starting paper and mechanism

3 sources

Primary source2026

Driessen K, Squarcio F, Tononi G, Cirelli C. Induction of cortical on/off periods in awake mice fulfills sleep functions. Nature Neuroscience. 2026.

Supports: Local on/off induction in awake mice, reduced local sleep pressure, synaptic markers and rescue of one memory task.
Does not establish: Does not show total sleep replacement, human applicability, non-invasive safety or multisystem benefit.

What if you didn'tneed to sleep?

The day has 24 hours. We usually get sixteen.

Three states, one question

Wake

NREM sleep

Wake + local off

The study tested the same idea in three ways.

Did the stimulated area feel more rested?

Did connections between neurons change?

Could one memory be preserved?

Did the stimulated area feel more rested?

Did connections between neurons change?

Could one memory be preserved?

The paper touched three rows in a much larger table.

Humanity gained hours through watchkeeping, light, drugs, discipline and strategic sleep.

The group sleeps while someone remains awake

Sleep becomes a moral test of industry

Electric light extends the sellable day

Pervitin and Benzedrine enter war

Randy Gardner turns deprivation into spectacle

Aviation mixes pills with planned sleep

Six hours a night builds a debt people do not fully feel

A 24-hour shift becomes a patient-safety problem

The awake brain begins to sleep locally

Frigatebirds sleep in flight, sparingly and selectively

A platform says its rival is sleep

Orexin agonists try to repair the wake system

Researchers pay one local sleep function during wake

Each instrument buys a different kind of hour.

Caffeine

Light and schedule

Strategic nap

Amphetamine / methamphetamine

Modafinil

Orexin agonist

Local off periods

When sleep has to wait.

35.3 hours inside a B-2 bomber

Modafinil keeps the aircraft nearer the line

A pill becomes command policy

A teenager and 264 hours of wake

Frigatebirds fall asleep above the ocean

Some people really need less

Sleep debt enters the patient chart

Sleep attracts stories precisely where measurement is hard.

When night disappears, who gets the time?

A real “no sleep” system needs more than a wake switch.

Claim licensed now

The first sleepless brain may learn to sleep in pieces.

Audit every bridge from mouse to history to scenario.

Starting paper and mechanism

Driessen K, Squarcio F, Tononi G, Cirelli C. Induction of cortical on/off periods in awake mice fulfills sleep functions. Nature Neuroscience. 2026.

Cirelli Sleep and Consciousness Laboratory. OFF_PERIOD_INDUCTION_MATERIALS. Code and statistical source data.

Tononi G, Cirelli C. Sleep and the price of plasticity: from synaptic and cellular homeostasis to memory consolidation and integration. Neuron. 2014;81:12–34.

Sleep, memory and deprivation

Brodt S, Inostroza M, Niethard N, Born J. Sleep: a brain-state serving systems memory consolidation. Neuron. 2023;111:1050–1075.

Vyazovskiy VV et al. Local sleep in awake rats. Nature. 2011;472:443–447.

Nir Y et al. Selective neuronal lapses precede human cognitive lapses following sleep deprivation. Nature Medicine. 2017;23:1474–1480.

Van Dongen HPA, Maislin G, Mullington JM, Dinges DF. The cumulative cost of additional wakefulness. Sleep. 2003;26:117–126.

Dawson D, Reid K. Fatigue, alcohol and performance impairment. Nature. 1997;388:235.

Wüst LN et al. Impact of one night of sleep restriction on sleepiness and cognitive function: a systematic review and meta-analysis. Sleep Medicine Reviews. 2024.

Ballesio A, Fiori V, Lombardo C. Effects of Experimental Sleep Deprivation on Peripheral Inflammation: An Updated Meta-Analysis of Human Studies. Journal of Sleep Research. 2026;35(1):e70099. Epub 2025 Jun 5.

Effects of sleep deprivation on endothelial function in adult humans: a systematic review. 2021.

Al Khatib HK et al. The effects of partial sleep deprivation on energy balance: a systematic review and meta-analysis. European Journal of Clinical Nutrition. 2017.

Xie L et al. Sleep drives metabolite clearance from the adult brain. Science. 2013;342:373–377.

Miao A et al. Brain clearance is reduced during sleep and anesthesia. Nature Neuroscience. 2024;27:1046–1050.

Evolution and animals

Yetish G et al. Natural sleep and its seasonal variations in three pre-industrial societies. Current Biology. 2015;25:2862–2868.

Samson DR et al. Chronotype variation drives night-time sentinel-like behaviour in hunter-gatherers. Proceedings of the Royal Society B. 2017;284:20170967.

Rattenborg NC et al. Evidence that birds sleep in mid-flight. Nature Communications. 2016;7:12468.

Lyamin OI et al. Cetacean sleep: an unusual form of mammalian sleep. Neuroscience & Biobehavioral Reviews. 2008;32:1451–1484.

He Y et al. The transcriptional repressor DEC2 regulates sleep length in mammals. Science. 2009;325:866–870.

Shi G et al. A rare mutation of β1-adrenergic receptor affects sleep/wake behaviors. Neuron. 2019;103:1044–1055.e7.

Chen H et al. The SIK3-N783Y mutation is associated with the human natural short sleep trait. PNAS. 2025;122:e2500356122.

History, war and operations

Snelders S, Pieters T. Speed in the Third Reich: metamphetamine (Pervitin) use and a drug history from below. Social History of Medicine. 2011;24:686–699.

Pugh J. The Royal Air Force, Bomber Command and the use of Benzedrine Sulphate. Journal of Contemporary History. 2018;53:740–761.

Pugh J. “Not … Like a Rum-Ration”: Amphetamine Sulphate, the Royal Navy, and the Evolution of Policy and Medical Research during the Second World War. War in History. 2017;24:498–519.

Caldwell JA et al. Dextroamphetamine use during B-2 combat missions. Aviation, Space, and Environmental Medicine. 2004;75:381–386.

Caldwell JA et al. Modafinil’s effects on simulator performance and mood in pilots during 37 h without sleep. Aviation, Space, and Environmental Medicine. 2004;75:777–784.