In my clinical practice, and even more often during forensic evaluations, patients and examinees will mention, almost apologetically, that they “do their best thinking at 2 a.m.”
Some have read the headlines: night owls are smarter. They want to know if that is true, or if it is pop-science folklore. The honest answer is more interesting than either extreme.
After fifty years of research, six replicated studies, one 100,000-person genetic analysis, and a 26,820-participant cognitive study published just last year, we can finally say something defensible about whether the time you naturally fall asleep tells us anything about your IQ.
It does. But probably not what you have been told.
This article walks you through the actual peer-reviewed evidence, sample sizes and effect sizes intact, and separates what the data genuinely supports from what has been inflated by social media and clickbait headlines. By the end, you will know exactly where your chronotype sits on the cognitive map, and what that does and does not mean for your intelligence.
What Chronotype Actually Is
Before we discuss the cognitive findings, we need to be precise about what chronotype is, because the popular conversation conflates several distinct concepts.
Chronotype refers to your stable, biologically determined preference for the timing of sleep and wakefulness within the 24-hour day. It is not a lifestyle choice. It is not a personality trait you can change by setting more alarms. It is a constitutional disposition rooted in your genetic regulation of circadian rhythms, modified by age, light exposure, and social schedules but rarely overridden by them.
Researchers typically classify people into three broad chronotypes. Morning types, commonly called “larks,” naturally wake early, feel most alert in the morning, experience their cognitive peak before noon, and prefer early bedtimes. Their endogenous circadian period is shorter than 24 hours, which biologically pulls them toward earlier sleep and wake times. Evening types, the “night owls,” show the reverse pattern: late natural bedtimes, late natural waking, peak cognitive function in the late afternoon or evening, and a longer endogenous circadian period that pulls them forward. Intermediate types, who make up roughly 40 to 50 percent of the population, sit between these extremes.
Population distributions are remarkably consistent across countries: approximately 25 to 30 percent morning types, 40 to 50 percent intermediate, and 25 to 30 percent evening types. The distribution shifts with age, with adolescents and young adults skewing later and older adults shifting earlier, but the underlying type remains stable for most people across their adult lifespan.
The most important finding about chronotype, and the one that makes all subsequent cognitive research meaningful, is that it is substantially heritable. A 2016 genome-wide association study by Jones and colleagues, published in Nature Communications, analyzed 100,420 participants from the UK Biobank and identified 12 genetic loci robustly associated with chronotype. Twin studies estimate heritability at approximately 40 to 50 percent. Your tendency to be a night owl or a morning lark is, to a meaningful degree, written into your DNA.
This matters for the IQ question because if chronotype were merely a behavioral choice, any correlation with intelligence would likely reflect lifestyle confounders. The fact that chronotype has a substantial genetic basis means that observed cognitive differences between owls and larks reflect something more biologically fundamental than which Netflix show you watch at midnight.
The 1999 Origin: The Roberts and Kyllonen Study
The systematic investigation of chronotype and intelligence began with a 1999 study by Roberts and Kyllonen, published in Personality and Individual Differences. The researchers administered a battery of cognitive tasks to 420 U.S. Air Force recruits, measured their chronotypes using a validated questionnaire, and compared performance across owl and lark groups.
Evening types outperformed morning types on working memory and processing speed measures. The effect was small but statistically robust. Crucially, the testing was conducted in the early morning, the lark’s biological prime time and the owl’s biological worst case. The owl advantage existed despite testing conditions that should have favored the larks.
This methodological detail eliminated one of the most plausible alternative explanations: that night owls only appear smarter because researchers test them when they are at their peak. The Roberts and Kyllonen study showed that the chronotype effect was stable across time-of-testing, suggesting something about evening orientation itself, not just performance timing, was associated with cognitive ability.
The effect size was modest, accounting for roughly 1 to 3 percent of variance in cognitive performance. But the direction was clear, and the methodological design was sound. The finding launched what has become one of the more durable areas of differential psychology research.
The Kanazawa Study: 20,000 Adolescents Tracked Into Adulthood
The most widely cited chronotype-IQ study, and the one responsible for nearly every “night owls are smarter” headline you have ever seen, is Satoshi Kanazawa and Kaja Perina’s 2009 paper in Personality and Individual Differences.
The researchers analyzed data from the National Longitudinal Study of Adolescent Health, a nationally representative U.S. dataset that measured childhood intelligence and tracked participants into young adulthood. Among roughly 20,000 respondents with both childhood IQ scores and adult chronotype data, they found a monotonic relationship: individuals with childhood IQ below 75 went to bed approximately one hour earlier on weekday nights than those with childhood IQ above 125. The pattern held across the entire IQ distribution, controlling for age, sex, race, socioeconomic status, marital status, parental status, religious involvement, and work hours.
The interpretation Kanazawa offered, framed within his “savanna principle,” was that staying awake into the night represents an evolutionary novelty. Our ancestors slept when the sun went down and woke when it rose. The cognitive flexibility required to override this ancestral pattern, to adopt a non-traditional sleep schedule that bears no resemblance to the conditions humans evolved in, may be more available to individuals with higher general intelligence. Smarter brains, in this view, are better at adapting to evolutionary novelty, including the novelty of being awake at hours our ancestors would have spent unconscious.
I want to be transparent about the controversy here. Kanazawa has produced other work that has drawn significant scientific criticism, and his savanna framework is one of several competing theoretical explanations for the chronotype-IQ relationship. Many researchers in differential psychology cite his data while disagreeing with his interpretation, which is intellectually honest.
What I find more important than the theoretical framework is that the empirical pattern Kanazawa documented has been independently replicated by multiple research groups using different samples, different cognitive assessments, and different theoretical lenses. The finding does not depend on whether you accept Kanazawa’s evolutionary explanation.
The Replications That Built the Consensus
Three subsequent studies have given the chronotype-IQ relationship its current scientific standing.
Preckel, Lipnevich, Schneider, and Roberts published a 2011 meta-analysis in Learning and Individual Differences synthesizing the chronotype-cognition literature available at that time. The pooled effect size was small but statistically robust and consistent across studies. Meta-analyses are the gold standard for resolving questions about whether effects are real, and this one confirmed that the chronotype-intelligence relationship was not a statistical fluke from a single dataset.
Piffer, Ponzi, Sapienza, Zingales, and Maestripieri published a 2014 study in Intelligence examining 169 graduate students at a top-ranked U.S. MBA program. This sample is critical because it represents a high-ability population: the class average GMAT score was 702.6, placing them at approximately the 90th percentile of test-takers worldwide. Within this elite cognitive group, night owls scored significantly higher on the GMAT than early-morning types, regardless of sex. Chronotype, sex, and baseline salivary cortisol jointly accounted for 14 percent of variance in GMAT scores. Notably, sleep duration did not correlate with chronotype, ruling out the “night owls just have more time to study” explanation.
The 2014 Piffer finding matters because it shows the chronotype effect operates within high-ability populations, not just across the full intelligence distribution. Even when comparing already-gifted individuals to other already-gifted individuals, the evening types still came out ahead. This argues against the idea that chronotype is simply a marker of broader life circumstances that also produce intelligence.
The most authoritative recent evidence comes from a 2024 study by West and colleagues published in BMJ Public Health, examining 26,820 adults aged 53 to 86 from the UK Biobank. Participants completed a battery of cognitive tests measuring fluid intelligence, pairs matching, reaction time, and prospective memory. Intermediate and evening chronotypes scored significantly higher than morning types on the cognitive composite, after adjustment for sleep duration, sleeplessness, smoking, alcohol use, diabetes, and cardiovascular disease.
The 2024 UK Biobank study is the largest and most recent confirmation of the chronotype-cognition relationship to date. It also extends the finding into older adults, demonstrating that the cognitive advantage of evening orientation persists across the lifespan rather than being limited to youth.
The Genetic Evidence: From Correlation to Causation
The most intellectually significant development in the chronotype-IQ literature is the genetic causal evidence, which most popular coverage misses entirely.
The Jones and colleagues 2016 genome-wide association study I mentioned earlier did more than identify genetic loci for chronotype. It also conducted Mendelian randomization analyses to test whether chronotype causally influences educational attainment. Mendelian randomization is a powerful statistical technique that uses genetic variants as natural experiments. Because genetic variants are randomly assigned at conception and cannot be influenced by lifestyle, education, or socioeconomic status, they provide leverage for inferring causation rather than mere correlation.
The results suggested that evening chronotype is causally, genetically related to higher educational attainment. This is meaningfully stronger than correlational evidence. It indicates that whatever genes predispose a person toward eveningness also predispose them toward the kind of cognitive performance that drives educational success.
This finding is the closest thing the chronotype-IQ literature has to a smoking gun. It explains why the correlation persists across sample types, countries, age groups, and cognitive measures: there appears to be a shared genetic architecture between circadian disposition and cognitive ability.
What it does not explain is the mechanism. We do not yet know whether the same genes that produce eveningness also produce cognitive ability through independent biological pathways, or whether one trait causally influences the other at the developmental level. The genetic evidence establishes that the link is real and biologically grounded. It does not establish why.
What This Does Not Mean
I want to be very precise about what the chronotype-IQ research does not support, because this is where popular coverage consistently distorts the science.
First, the effect size is small. Across studies, chronotype accounts for somewhere between 1 and 5 percent of variance in cognitive performance. To translate this into IQ points: the typical difference between a confirmed night owl and a confirmed morning lark, controlling for all the obvious confounders, is on the order of 2 to 7 points. This is statistically robust and meaningful at the population level, but it is not the difference between average and gifted intelligence. A morning person with an IQ of 130 is substantially smarter than a night owl with an IQ of 110, regardless of when they prefer to sleep.
Second, you cannot make yourself smarter by staying up later. The chronotype-IQ correlation operates through shared genetic and developmental factors. Forcing yourself to adopt an evening schedule when your biology favors morning will not raise your cognitive performance. It will mostly produce sleep deprivation, which reduces cognitive performance significantly. The research describes a statistical pattern between two stable traits, not a behavioral intervention.
Third, evening orientation carries real costs that morning orientation does not. Night owls living in a morning-oriented society experience higher rates of mood disorders, substance use, metabolic dysfunction, and cardiovascular disease. They sleep less on average due to social and occupational schedules that misalign with their biology. They report lower life satisfaction and higher psychological distress in multiple large-scale studies. Whatever cognitive advantage they may carry is offset by genuine health costs that the popular “night owls are smarter” framing obscures.
Fourth, the relationship is bidirectional and confounded by educational selection. The 2024 UK Biobank data, like most chronotype research, draws from samples skewed toward higher education. Education itself selects for and shapes evening orientation: university and graduate school operate on schedules that reward late-night work, and academic culture normalizes evening productivity. Some portion of the chronotype-cognition correlation may reflect educational pathway effects rather than constitutional cognitive advantage.
The honest scientific summary is this: chronotype and intelligence are reliably correlated, the correlation is partially genetic, the effect size is small but real, and the relationship cannot be used to predict any individual’s cognitive ability with meaningful precision. Knowing someone is a night owl tells you very little about how intelligent they are. Knowing their actual cognitive profile tells you a great deal.
Why Your Bedtime Is the Wrong Question
After two decades of administering cognitive assessments and reviewing this literature, I have come to believe that the popular question, “Are night owls smarter?”, is the wrong question.
The chronotype-IQ research is genuinely interesting as a window into the biological architecture of cognition. It tells us that certain heritable traits, including ones as seemingly mundane as when you prefer to sleep, share genetic and developmental pathways with intelligence. It contributes to a richer understanding of how cognitive ability is constructed from multiple biological systems.
But for any individual reader, the chronotype finding is too noisy to be practically useful. The effect size means that a typical night owl is perhaps 3 to 4 IQ points higher than a typical morning lark, with enormous individual variation around that average. Within any chronotype, the range of cognitive ability spans from below-average to extraordinarily gifted. Your bedtime is one tiny signal among many, and it tells you less about your own cognitive profile than the average person assumes.
The more useful question is not “what does my chronotype say about my IQ” but “what is my actual cognitive profile.” That is a question chronotype cannot answer. It can only be answered by actually measuring cognitive ability across the domains that matter: verbal reasoning, perceptual reasoning, working memory, processing speed, fluid intelligence.
A well-designed cognitive assessment provides exactly this information. It tells you not just a single number but a profile of your relative strengths and weaknesses across cognitive systems. It places you within population norms with statistical precision rather than evolutionary speculation. It gives you data that is genuinely useful for decisions about education, career, learning strategies, and self-understanding.
The chronotype literature is interesting science. Your actual cognitive profile is actionable information. Knowing the difference is the beginning of using either one wisely.
A Brief Self-Check
If you are wondering where you fall on the chronotype spectrum, here is a clinically informed self-assessment. None of these questions individually identifies your chronotype, but the pattern across them reliably classifies most adults.
When you wake up without an alarm on a free day, what time is it? Morning types typically wake before 7 a.m. Intermediate types wake between 7 and 9. Evening types wake after 9 a.m., often after 10.
When are you naturally most mentally alert? If your peak alertness is before noon, you are likely a morning type. If it is in the late afternoon or evening, you are likely an evening type. If it is variable or moderate throughout the day, you are likely intermediate.
When do you naturally fall asleep on weekends, free of social or occupational obligations? Morning types fall asleep before 11 p.m. Intermediate types fall asleep between 11 and midnight. Evening types fall asleep after midnight, often well after 1 a.m.
How do you feel during the first hour after waking? Morning types report feeling alert and ready to engage almost immediately. Evening types report feeling groggy, foggy, and unable to perform demanding cognitive tasks for one to two hours after waking.
If you needed to perform at your absolute cognitive best for a high-stakes exam or evaluation, at what time of day would you schedule it? Morning types choose 9 to 11 a.m. Evening types choose 4 to 8 p.m. The time you instinctively select reveals your underlying chronotype more accurately than any of the other questions.
Once you have classified yourself, the chronotype literature provides a small piece of context for understanding your cognitive style. But the meaningful next step is not to take more chronotype quizzes. It is to measure your actual cognitive profile against population norms, which gives you information that no bedtime survey can provide.
Final Thoughts
The night owl literature is one of those areas where the popular narrative and the scientific reality have drifted apart in ways that are worth correcting.
The popular narrative is that night owls are smarter than morning people, and this is treated as flattering self-identification for the millions of adults who have always felt at odds with morning-oriented society. Late-night creatives, insomniac professionals, and chronically tired graduate students have embraced the framing as validation for their lifestyles.
The scientific reality is more nuanced. Night owls do score modestly higher than morning larks on cognitive measures, the difference is small and partially genetic, the relationship is real but not strong enough to predict any individual’s intelligence with confidence, and the health and life-satisfaction costs of evening orientation in a morning-oriented society are substantial.
What the science actually supports is not the celebratory “night owls are geniuses” framing but a quieter, more useful insight: chronotype is one of many heritable traits that share biological architecture with cognitive ability, and understanding your own cognitive profile requires direct measurement rather than inference from sleep schedules.
If you have read this far, you are the kind of curious, evidence-driven reader who actually wants the numbers, not the horoscope. Take the test that gives you the numbers. Your bedtime is interesting context. Your measured cognitive profile is the actual data.
