The question always comes back in the same form: how many hours should one sleep? And the honest answer begins with a disappointment: there is no single number. The need varies with age, it varies from person to person, and the average repeated everywhere precisely crushes what matters.
What is measurable, however, are the major trends by age group, how sleep debt sets in, and the signs that indicate a lack of it. This is what this article is about.
An average that hides individuals
For adults, the generally accepted range is around seven to nine hours. The important word is range. Some people genuinely function well with a little less, others need more without it being an anomaly. These are constitutional differences, like height or sensitivity to cold.
The right question is therefore not "do I sleep eight hours?" but "do I wake up rested, and can I get through the day without an uncontrollable energy slump?". It's a subjective criterion, and yet it's the most reliable.
To this is added a temperament variable, the chronotype: about four out of ten people are morning types, three out of ten are evening types, the rest are somewhere in between. A pleasant evolutionary hypothesis suggests that a group whose members do not all sleep at the same times reduces its collective vulnerability window, with someone always watching while others sleep. Chronotype does not change the necessary duration, it shifts the window in which it occurs.
The newborn, or sleep without a clock
Newborns sleep sixteen to seventeen hours a day, divided into short episodes. They do not yet have a circadian rhythm: their internal clock is not yet set, and their tiny stomach necessitates frequent awakenings. It is therefore neither a whim nor a parenting issue, it is a clock that does not yet exist.
Another peculiarity: the proportion of active sleep, considered the precursor to REM sleep, is proportionally enormous in newborns. This is consistent with the intensity of learning during this period. Remarkably, this alternation of activity and rest already exists in the fetus, from the twentieth week, on a cycle of about forty-five to fifty minutes, independent of the mother's.
Children: naps that disappear one by one
In children, the need decreases in stages, and naps disappear in a fairly regular order. The late afternoon nap goes away around one year, the morning nap around eighteen months, the early afternoon nap between three and six years. After that, bedtime moves back by fifteen to thirty minutes per year.
This is also the age for nocturnal phenomena that worry parents but are nevertheless benign: night terrors, frequent between three and eight years old, or sleepwalking, much more common in children than in adults, and which usually disappears with age.
Adolescents: more sleep, and later
Adolescents need a lot of sleep, and their clock provides it at the wrong time relative to school schedules. Their melatonin is secreted later than that of children or adults, which mechanically pushes back sleep onset. The shift is biological, not laziness.
This results in a debt that accumulates five days a week and that a too-long weekend only half repays. We dedicated an entire article to this topic and what parents can do about it: sleep by age.
Thirty to fifty years old: the age when we sleep the least
This is the least known, and probably the most telling, result. The period when sleep is shortest throughout life is neither adolescence nor old age: it is the thirty to fifty year old bracket. A large cohort study places the minimum around age thirty, with an increase after fifty.
Nothing mysterious about the causes: young children, maximum professional workload, commutes, evenings eaten up by everything that couldn't be done during the day. What matters is knowing it. Many people of this age believe they are bad sleepers when they are simply sleepers who have had hours taken from them.
After sixty: it's not the need that decreases
The idea that we need less sleep as we age is false, and it persists. What changes is the quality: sleep becomes lighter, more fragmented, with more awakenings, and the clock advances, leading to earlier bedtimes and wake-up times.
Two levers effectively act on this clock at this age: bright daylight, every morning, and physical activity. Regarding sleeping pills, whose benefit-risk ratio deteriorates with age, particularly due to the risk of falls, the only useful answer is the family doctor.
Sleep debt: what can be caught up and what cannot
After a few short nights, sleeping more recovers a good part of the fatigue. This is recovery sleep, and it works. However, chronic debt accumulated over months or years cannot be repaid by a lie-in. Regularity always takes precedence over compensation.
That sleep is not optional has been shown in the harshest way possible. Allan Rechtschaffen's protocol in rats, in 1983, designed to separate the effect of sleep deprivation from that of stress, led to death in ten to fifteen days, faster than food deprivation. In humans, the best-documented case remains that of Randy Gardner, eleven days of wakefulness in 1964 under medical supervision: mood, memory, and concentration disorders, episodes of paranoia and hallucinations, then recovery after a few nights. These records are no longer sanctioned, precisely because they are dangerous. They are cited for what they show, never as a challenge.
How to know if you're sleeping enough
A few clues are better than a number. Do you wake up without an alarm when you have the opportunity, and at roughly the same time? Can you make it through the mid-afternoon without irresistible drowsiness? Do your weekends resemble your weekdays, or do you need three extra hours on Saturday to function?
A final, counter-intuitive clue, concerns the memory of dreams. Good sleepers remember their dreams less, simply because they wake up less and fall back asleep faster. Saying "I never dream" therefore often signals good quality sleep, not an anomaly. What these invisible awakenings do to our nights is explained in our micro-awakening fact sheet.
What about sleep before midnight?
The saying goes that it counts double. It is poorly phrased, but it contains a useful truth: slow-wave sleep, the most physically restorative, is concentrated at the beginning of the night. Going to bed early therefore mechanically provides more deep sleep. What matters is not the time displayed on the clock, but the position in your night. For other beliefs of this kind, see our sleep true or false.
Ultimately, the right amount of sleep is what allows you to be yourself the next day. It is found by observing, not by counting. And if nights remain insufficient despite regular hours, it is a sleep professional who should be consulted, not an average.
Content for informational purposes only, not a substitute for medical advice.
