Can You Still Be Creative on No Sleep? What Research Says About Quick Wit, Caffeine, ADHD Self-Diagnosis and Short Sleepers

You catch a glimpse of someone with long white hair, and your brain fires back almost instantly: "Rashōmon?" (as in the old woman from Akutagawa's story). You weren't trying to be funny. You didn't search for it. It just showed up, and you were running on very little sleep.

That one moment opens a whole can of questions. What is this "it just pops out" ability? Can people who don't have it learn it? What breaks when you're sleep-deprived, and what somehow survives? How much can coffee or an energy drink bring back? Are some of the people online saying "I think I have ADHD" or "I'm probably autistic" (ADHD meaning attention-deficit/hyperactivity disorder, ASD meaning autism spectrum disorder) just running short on sleep? And what do we make of people who claim they only need 30 minutes a night?

Turns out researchers got there first. Again.

1. The 5-second answer: quick creativity has four steps, and sleep is the foundation under all of them

A handy way to break down "it just came to me":

1. Association: zip through your memory, from white hair to old woman to hair to classic literature.
2. Analogy: link what's in front of you to something with the same shape in a completely different area.
3. Selection: out of all the options, keep the one that is both fresh and actually fits.
4. Compression: squeeze the whole chain into one short line that people can unpack on their own.

A 2024 study found that how new an idea is depended more on your ability to make meaning-based associations, while how fitting it is depended more on executive control, the brain's "manager" skills like focus and self-control.^[1] A 2026 study in the journal Cognition found that the right answers to remote-association puzzles were often already there in people's first, fast, gut-level responses.^[2]

So an idea that "just appears" can still be real creative thinking.

Sleep is a separate layer, though. One burst of association can survive while your ability to stay focused, keep steps in order, and make steady decisions is already slipping.

Even a creative genius needs server maintenance.

2. "Association + analogy + picking the winner" is basically the right model

Association is how far and how fast you can travel through what words and ideas mean to you. Most people go from "white hair" to "aging." A wider search goes "old woman," "hair," "classic literature," and lands on "Rashōmon." In the 2024 study, this kind of meaning-based association predicted how new an idea was.^[1]

Analogy isn't just about similar words. It asks: "What has the same structure somewhere else?" A real-life moment can be mapped onto literature, comics, games, quality control at work, traffic or computer code. Being able to carry a structure from a field you know into a new problem is a big part of creative problem-solving.

Selection keeps wide association from turning into random noise. If fifty options show up ("snow spirit, wizard, hermit, silver-haired anime character, Rashōmon..."), the real skill is spotting the one that connects in several ways at once (white hair, old woman, hair) and still lands as a short line in conversation.

Compression is the quiet MVP. "Your visual features remind me of the old woman in an Akutagawa story" is accurate but dead on arrival. "Rashōmon?" lets the listener fill in the rest themselves.

Real-world creativity isn't just coming up with far-out ideas. It's coming up with them, picking the one that's far out but still makes sense, and shrinking it to a size people can actually use.

3. You need some control, but a checkpoint that shows up too early stops the bulldozer before it moves

A 2025 study of 384 people found that both free association and more controlled association helped originality, but the stronger the controlled association, the fewer ideas people came up with.^[3]

Picture two departments:

• idea generation = the bulldozer, clearing possible paths;
• evaluation = the surveyor, deciding which paths are actually usable.

Trouble starts when the surveyor stands in front of the bulldozer yelling "No precedent. Denied." before it has moved an inch.

It's like a kid making clay figures: make one, "nope, wrong," squash it. Make another, squash it. Nothing ever survives long enough to become a finished piece.

A study published online in 2021 compared people who aim for realistically high standards ("excellence") with people who accept nothing less than perfect ("perfectionism"). The excellence group came up with more answers, and more original ones, on open-ended idea tasks.^[4]

But judging your ideas isn't automatically toxic. A preregistered 2026 study with 440 participants found no significant difference in originality or number of ideas between people who rated every idea right away, people who rated them at the end, and people who weren't asked to rate them at all.^[5]

The real takeaway: checking your ideas isn't the enemy; shooting them down before you've explored them can be.

Give the creative team freedom, and save the cool-headed judgment for final sign-off.

4. Can you train it? Yes, but it's not a cheat code

A 2024 meta-analysis that pooled 169 creativity-training studies and 844 effect sizes from the past 50 years found an unadjusted effect of 0.53 standard deviations. After correcting for publication bias (the tendency for positive results to get published more), the estimate dropped to about 0.29–0.32.^[6]

So training can work, but "anyone becomes a creative genius in a week" isn't what the evidence says.

A 2025 meta-analysis of 29 studies and 109 effects in university students reported a medium overall effect, Hedges' g=0.628.^[7] Since the field as a whole has publication-bias and method problems, it's smart to keep expectations modest.

If you want to train this specific "it just pops out" skill, a practice set based on the research looks like this:

• Association: take one word and free-associate for 30 seconds without filtering yourself.
• Analogy: ask what today's events would look like in literature, comics, games, work or another field.
• Selection: keep a far-out idea only if it has at least two real points of connection.
• Compression: once you've picked an idea, boil the explanation down to one short punchline.

This exact four-part routine hasn't been tested on its own in a randomized trial. It's a practical version built from the research on each piece.

5. Lack of sleep first hits your ability to stay steadily switched on

Sleep loss doesn't drop your whole brain evenly to a C grade.

A 2024 meta-analysis of 44 studies found that just one night with only 2 to 6 hours of sleep opportunity sharply increased sleepiness and clearly hurt sustained attention, with slower reactions and more "zone-outs."^[8] In that same analysis, some other areas, like working memory and impulse control, didn't show a significant average drop.

So it doesn't always feel like "everything is 30% slower." It's more like:

95 → 20 → 90 → 40.

Your performance gets jumpy.

When sleep loss gets heavier or drags on, the damage spreads. A 2025 meta-analysis of 79 studies found problems with working memory, impulse control and mental flexibility, the core "manager" skills of the brain.^[9]

Chronic sleep loss is the sneaky one. In a classic 14-day experiment, people allowed 4 or 6 hours of sleep a night got steadily worse on tests day after day, while their own sense of sleepiness didn't get worse at the same pace.^[10]

Brain: getting worse.

You: "I think I've gotten used to it."

Those two don't match.

6. And yet a sleep-deprived "Rashōmon?" is totally possible: a 2026 study tested something very close

In May 2026, researchers at Kyushu University split 27 young men into two groups: one slept normally, and one was allowed only 2 hours in bed.^[11]

The short-sleep group got sleepier and their sustained attention dropped. But on a verbal brainstorming task and tangram puzzles that need a flash of insight, there was no clear difference between the groups. Bayesian analysis (a statistical method) also backed the idea that these creative tasks held up under those conditions.^[11]

So "I'm exhausted, but the weird association still shows up instantly" is very plausible.

What this does not show is that two hours of sleep is fine for your health or your job. The paper itself warns that other mental abilities and your health can still take a hit.^[11]

The creative engine may still be running while the brakes, the dashboard and the lane assist are failing.

7. Caffeine power-up: what comes back easily and what doesn't

A meta-analysis of 45 studies and 327 effect estimates found that caffeine after sleep loss improved how fast and accurately people paid attention, information processing and driving performance, and gave a smaller boost to executive functions.^[12]

So caffeine is good at:

• pushing back sleepiness;
• speeding up reactions;
• cutting down on zone-outs;
• keeping simple to medium tasks steady.

But feeling awake is not the same as being fully back.

In an experiment with 276 people, 200 mg of caffeine after an all-nighter improved sustained attention, but barely fixed people's ability to keep track of a multi-step task in the right order.^[13]

Caffeine: Awake! Reactions are back!

Brain: By the way, we skipped step three.

Creativity splits the same way. In a double-blind trial with 88 people, 200 mg of caffeine improved problem-solving with one right answer, but did nothing significant for open-ended idea generation or working memory.^[14]

So there's no proof that coffee unlocks a "Super Rashōmon Mode."

A better way to put it:

the idea generator that comes up with "Rashōmon": not clearly boosted;

the attention that catches the idea and doesn't drop it: easier to boost.

8. Red Bull vs. coffee: most of those wings are caffeine

As of September 2026, Red Bull Japan's official info says a 250 mL can contains 80 mg of caffeine.^[15]

The Japanese version's official ingredient list names caffeine, arginine, B vitamins, sugars and water as the main ingredients. So research on overseas versions that contain taurine shouldn't be automatically applied to the current standard Japanese can.^[15]

When it comes to your brain, the main thing coffee and Red Bull have in common is caffeine.

And the power-up has a payback. A meta-analysis of 24 studies found that, on average, caffeine cut total sleep by about 45 minutes, lowered sleep efficiency by about 7%, delayed falling asleep by about 9 minutes, added about 12 minutes of being awake after first falling asleep, and reduced deep sleep.^[16]

Its model estimated that to avoid losing total sleep time on average, a coffee with about 107 mg of caffeine (a 250 mL cup) needs to be finished roughly 8.8 hours before bed.^[16] People differ, but "it's just coffee, it's fine in the evening" isn't a safe bet.

Sleep loss → caffeine → shorter sleep that night → more caffeine the next day. Congratulations, you've built a perpetual motion machine.

You got your wings and deleted your landing time.

9. "Are some people online saying they have ADHD or autism just sleep-deprived?" Partly, maybe. But it's not the same thing

This sounds like a joke, but it's actually an important medical point.

The U.S. Centers for Disease Control and Prevention (CDC) says ADHD can't be diagnosed with a single test, because sleep problems, anxiety, depression and learning problems can all cause similar symptoms.^[17] A review on sleep-related breathing problems also notes that poor sleep can make existing ADHD symptoms worse, or cause thinking problems that look like ADHD.^[18]

Sleep loss can cause bits and pieces like trouble focusing, spacing out, forgetting things, more mistakes, slower reactions and a shorter fuse. Plug just those into an online checklist, and it can't tell you the cause.

So:

up late, five hours of sleep → can't focus → online quiz → "ADHD!"

doesn't actually identify the cause.

But ADHD isn't "just being tired" either. A real evaluation looks at how symptoms developed over time and whether they show up consistently across different settings.

Autism is even harder to explain with a bad week of sleep. The CDC's summary of the DSM-5 (the main diagnostic manual) requires ongoing differences in social communication and interaction, plus restricted or repetitive behaviors or interests, with signs present from early childhood.^[19]

Being sleep-deprived and finding people exhausting, misreading emotions or getting irritable for a while is not the same as the developmental pattern of autism.

The accurate way to say it:

sleep loss can amplify bits that look like ADHD or autism, but it doesn't explain the diagnosis itself.

A symptom list is a warning light on the dashboard, not the mechanic's final report.

10. What about people who sleep 30 minutes a night? Separate personal claims from real short-sleeper science

In Japan, Daisuke Hori has gotten attention for saying he sleeps about 30 to 45 minutes a day. In a published interview, he said he originally slept about eight hours, trained himself to sleep less, and at that point was getting by on about 45 minutes of main sleep plus two power naps of about 15 minutes each.^[20]

The key is to keep two things apart: someone publicly claiming it, and 30 minutes being independently proven to be enough sleep.

Natural short sleepers really do exist. Genetic studies have linked inherited natural short sleep to variants involving DEC2/BHLHE41, ADRB1, NPSR1 and GRM1.^[21] But the short sleep described in that research is usually around 4–6 hours, not 30 minutes. That's a whole different league.^[21]

In the public information checked for this article, there was no peer-reviewed study that tracked Hori's sleep around the clock over a long period with brain-wave (EEG) recordings and compared his thinking, metabolism and heart health with normal sleepers.

So, for now, the careful way to describe it is:

an extreme case of someone publicly claiming long-term 30-minute sleep,

not

scientific proof that ordinary people can train themselves down to 30 minutes.

This matters because with chronic sleep loss, your measured performance can keep sliding even while you feel like you've adjusted.^[10]

Japan's Ministry of Health, Labour and Welfare, in its "Sleep Guide 2023 for Health Promotion," recommends that adults get the sleep they need, using 6 hours or more as a general target, while acknowledging that people differ.^[22]

If someone truly stays healthy and sharp on 30 minutes for years, the move isn't to argue about it. It's to get them into a sleep lab.

Sleep scientist: "Great. Let's hook up the EEG."

That's where the science starts.

11. Bottom line: sleep is system maintenance, caffeine is a temporary buff, and creativity is partly its own module

Put it all together and you get this:

Creativity isn't just coming up with far-out associations. It's associating, borrowing structure from other areas, picking the option that fits the moment, and compressing it into one short line.^[1][2]

Control matters, but shooting ideas down before they're even out can shrink your search. At the same time, judging ideas along the way isn't automatically bad.^[3][5]

Sleep loss hits sustained attention and steadiness first, and as it gets worse and lasts longer, it drags in your executive functions too.^[8][9][10]

But some open-ended thinking and insight can hold up under short-term partial sleep loss.^[11] That's why "sleepy, but still instantly funny" is possible.

Caffeine is strong at bringing back alertness, attention and reaction speed. But it doesn't fully restore your ability to manage complex steps, and there's little evidence it boosts open-ended creativity.^[12][13][14]

Self-diagnosing ADHD or autism means you can't ignore that sleep loss can cause look-alike symptoms, but you also can't treat these conditions as the same thing as being tired.^[17][19]

Natural short sleepers are real, but that research doesn't back up 30-minute sleep for regular people.^[21][22]

The final formula:

Sleep = system maintenance.

Coffee = temporary stat buff.

Creativity = a module that can sometimes keep firing even when other systems are already dropping packets.

And coming up with "Rashōmon?" on no sleep doesn't prove the rest of your brain is doing fine.

Even creative geniuses need to sleep.

Sources

^[1]–^[22] refer to the Global References at the end of this page.

---

References (22)

1. Wang X, et al. “Semantic associative abilities and executive control functions predict novelty and appropriateness of idea generation.” Communications Biology. 2024;7:703. PMID 38849461. DOI: — https://pubmed.ncbi.nlm.nih.gov/38849461/ doi.org
2. Beucler J, De Neys W. “Intuitive insight: Fast associative processes drive sound creative thinking.” Cognition. 2026;271:106422. PMID 41500017. DOI: — https://pubmed.ncbi.nlm.nih.gov/41500017/ doi.org
3. Creative spark or overthinking? The synergistic effects of free and controlled associations on creativity.” Thinking Skills and Creativity. 2025;57:101818. DOI doi.org
4. Gaudreau P, et al. “Is perfectionism a killer of creative thinking? A test of the model of excellencism and perfectionism.” British Journal of Psychology. 2022. PMID 34472100. DOI: — https://pubmed.ncbi.nlm.nih.gov/34472100/ doi.org
5. Does reflecting one's own ideas affect creativity? Metacognitive monitoring activation shows no effect on divergent thinking.” Learning and Individual Differences. 2026;129:102872. DOI doi.org
6. Sio UN, Lortie-Forgues H. “The impact of creativity training on creative performance: A meta-analytic review and critical evaluation of 5 decades of creativity training studies.” Psychological Bulletin. 2024;150(5):554–585. PMID 38635185. DOI: — https://pubmed.ncbi.nlm.nih.gov/38635185/ doi.org
7. Guo H, Zhou Z, Ma F. “Effect of training programs on the creativity of university students: A meta-analysis.” Thinking Skills and Creativity. 2025;56:101779. DOI doi.org
8. Impact of one night of sleep restriction on sleepiness and cognitive function: A systematic review and meta-analysis.” Sleep Medicine Reviews. 2024. PMID 38759474. DOI: — https://pubmed.ncbi.nlm.nih.gov/38759474/ doi.org
9. Cao Y, et al. “The impairments of sleep loss on core executive functions: A systematic review and meta-analysis.” Sleep Medicine Reviews. 2025. PMID 40946426. DOI: — https://pubmed.ncbi.nlm.nih.gov/40946426/ doi.org
10. Van Dongen HPA, et al. “The cumulative cost of additional wakefulness: Dose-response effects on neurobehavioral functions and sleep physiology from chronic sleep restriction and total sleep deprivation.” Sleep. 2003;26(2):117–126. PMID 12683469. DOI: — https://pubmed.ncbi.nlm.nih.gov/12683469/ doi.org
11. Motomura Y, et al. “Partial sleep deprivation with a 2-h time-in-bed may not change divergent thinking and insight problem solving in young males.” Journal of Physiological Anthropology. 2026;45:18. PMID 42169155. DOI: — https://pubmed.ncbi.nlm.nih.gov/42169155/ doi.org
12. Systematic review/meta-analysis of acute caffeine consumption following sleep loss; 45 publications / 327 effect estimates. PMID 31837359 pubmed.ncbi.nlm.nih.gov
13. Stepan ME, Altmann EM, Fenn KM. “Caffeine selectively mitigates cognitive deficits caused by sleep deprivation.” Journal of Experimental Psychology: Learning, Memory, and Cognition. 2021;47(9):1371–1382. PMID 34014758. DOI: — https://pubmed.ncbi.nlm.nih.gov/34014758/ doi.org
14. Zabelina DL, Silvia PJ. “Percolating ideas: The effects of caffeine on creative thinking and problem solving.” Consciousness and Cognition. 2020;79:102899. PMID 32086187. DOI: — https://pubmed.ncbi.nlm.nih.gov/32086187/ doi.org
15. Red Bull Japan. Current Japanese product ingredient/caffeine information. 250 mL standard can: 80 mg caffeine. and https://www.redbull.com/jp-ja/energydrink/questions/how-much-caffeine-is-in-a-can-of-red-bull-energy-drink redbull.com
16. Gardiner C, et al. “The effect of caffeine on subsequent sleep: A systematic review and meta-analysis.” Sleep Medicine Reviews. 2023;69:101764. PMID 36870101. DOI: — https://pubmed.ncbi.nlm.nih.gov/36870101/ doi.org
17. U.S. Centers for Disease Control and Prevention. “Diagnosing ADHD. cdc.gov
18. Ivanov I, et al. Review on sleep-disordered breathing and ADHD risk/symptom overlap. Journal of Attention Disorders. 2024;28(5):686–698. PMID 38353411 pubmed.ncbi.nlm.nih.gov
19. U.S. Centers for Disease Control and Prevention. “Clinical Testing and Diagnosis for Autism Spectrum Disorder. cdc.gov
20. PERSOL. Public interview with Daisuke Hori, 2024: “1日45分睡眠のショートスリーパー堀大輔さんが語る、人生を豊かにする短眠術. studio.persol-group.co.jp
21. Reviews of familial natural short sleep and sleep genetics, including DEC2/BHLHE41, ADRB1, NPSR1, and GRM1-related findings. and https://pubmed.ncbi.nlm.nih.gov/34722012/ pmc.ncbi.nlm.nih.gov
22. 厚生労働省. 「健康づくりのための睡眠ガイド2023」 mhlw.go.jp