# Does Sleep Loss Lower Your IQ? What the Research Actually Shows > Sleep loss does not lower intelligence itself, but it reliably lowers performance on cognitive tasks that day. In a study restricting people to six hours of sleep for 14 nights, attention performance declined linearly to a level comparable with one to two nights of total sleep deprivation, while subjective sleepiness plateaued after a few days. Sleep loss on test day shows up as a degraded measurement condition, not a change in ability. - URL: https://brainrank.jp/en/blog/sleep-and-cognition - Published: 2026-08-24 - Publisher: BrainRank Editorial Team (https://brainrank.jp) - Tags: Cognitive abilities, Health Sleep loss does not lower intelligence itself, but it reliably lowers performance on cognitive tasks that day. Two weeks of six-hour nights push attention down to a level comparable with one to two nights of no sleep at all. The awkward part is that people barely notice it happening. ## The classic study: six hours a night for 14 nights The reference point is the sleep restriction experiment by [Van Dongen and colleagues (Sleep, 2003)](https://doi.org/10.1093/sleep/26.2.117). Healthy adults were assigned to eight, six, or four hours of sleep per night, or to total sleep deprivation, and were tested on attention, working memory, and processing speed tasks for 14 days. The result was unambiguous. In the six-hour and four-hour conditions, performance on the attention task (lapses in responding) worsened linearly day after day, with no plateau. By day 14 the decline was comparable to one to two nights of total sleep deprivation. ## People do not notice the decline The second important finding concerned subjective sleepiness. Self-rated sleepiness rose over the first few days, then flattened — and stayed flat while objective performance continued to fall. The feeling of having adapted, in other words, is not evidence that performance recovered; the more parsimonious reading is that the perception dulled. This mismatch between self-assessment and actual performance has the same structure as [the Dunning-Kruger effect](https://brainrank.jp/en/blog/dunning-kruger-iq): judging your own state accurately is hard. ## What degrades, in what order The first thing to break under sleep loss is the ability to hold attention at a steady level. Once that goes, everything built on top of it is affected in turn. - Sustained attention: lapses and misses increase. It appears first and largest - [Working memory](https://brainrank.jp/en/blog/working-memory-training): less is held, and you lose track of what you were doing - [Processing speed](https://brainrank.jp/en/blog/processing-speed-training): the same work takes longer - Knowledge and vocabulary: largely unaffected That last point is why 'your IQ drops' overstates it. Of [fluid and crystallized intelligence](https://brainrank.jp/en/blog/fluid-crystallized-intelligence), sleep loss hits performance on the fluid side, while accumulated knowledge stays put. ## Why 'your IQ drops' is the wrong description IQ is an estimate of relative position within an age group. Take a test while sleep-deprived and the score falls, but the position did not move — the measurement was taken under bad conditions. It is like taking a blood pressure reading immediately after a run: the number changes, what the measure means does not. The distinction matters in practice. Sleep loss on test day makes a score look lower than your actual standing (see [five things to do right before a test](https://brainrank.jp/en/blog/iq-test-practice)). Nor does a well-rested score overstate you. Holding conditions constant is what makes numbers comparable at all. ## Sleep duration in Japan, and the benchmark Japan's [Sleep Guide for Health Promotion 2023](https://www.mhlw.go.jp/stf/seisakunitsuite/bunya/kenkou_iryou/kenkou/suimin/index.html) sets six hours or more per night as the benchmark for adults. It also stresses the sense of being rested by sleep rather than duration alone. In Japan the share of people averaging under six hours is high — around 40% for both men and women in their 40s and 50s. The condition that experiment studied is not an exotic laboratory state; it is close to many people's ordinary weeknight. ## What to do before a test or demanding work 1. Prioritize the previous night's sleep: it beats last-minute cramming 2. Allow three or more hours after waking: performance is poor immediately after getting up 3. Use a short nap: around 20 minutes helps restore attention 4. Know that caffeine masks sleepiness without restoring performance 5. Measure under matched conditions: without the same time of day and sleep amount, comparison is meaningless ## Measure under conditions you control [BrainRank's free IQ test](https://brainrank.jp/en/start) scores 20 questions across spatial reasoning, pattern recognition, logical reasoning, and classification (about 10 minutes) using item response theory (IRT). Comparing domain tendencies on a well-slept day against a short night gives you a sense of how much your performance depends on sleep. Results are estimates for entertainment and self-understanding, not a medical or diagnostic assessment. If sleep problems persist, consult a medical professional. ## Frequently asked questions ### Does sleep deprivation lower your IQ? The relative standing that IQ expresses does not itself drop, but a test taken while sleep-deprived returns a lower score. It is more accurate to treat that as a measurement taken under poor conditions than as a loss of ability. ### How much sleep protects cognitive function? Japan's Ministry of Health, Labour and Welfare sleep guide (2023) sets six hours or more per night as the adult benchmark. Individual needs vary, and the guide emphasizes the sense of being rested by sleep, not duration alone. ### Is it safe if I can tell that I am sleep-deprived? Self-awareness is not reliable. In restriction studies, subjective sleepiness ratings plateaued within a few days while objective performance kept declining. Not noticing the decline is the most dangerous feature of the effect. ## Sources - [Van Dongen and colleagues (Sleep, 2003)](https://doi.org/10.1093/sleep/26.2.117) - [Sleep Guide for Health Promotion 2023](https://www.mhlw.go.jp/stf/seisakunitsuite/bunya/kenkou_iryou/kenkou/suimin/index.html) --- Statistics and percentages are calculated from a normal distribution model with a mean of 100 and a standard deviation of 15. The tests on this site provide estimates for entertainment and self-understanding; they are not medical or clinical assessments, nor official intelligence tests. When citing, credit "BrainRank (brainrank.jp)" and link the article URL.