Partly — but not as completely, and not as quickly, as the phrase “catching up” suggests. Recovery sleep helps. What it does not do is settle a debt like an account, and the best-known experiment on the question found three full nights of recovery sleep were not enough to bring moderately sleep-restricted people back to their own baseline.
What happens when you try to catch up
The clearest evidence comes from a 2003 study in the Journal of Sleep Research. Sixty-six healthy volunteers were assigned to 3, 5, 7 or 9 hours in bed a night for a week. Then everyone got three nights of eight hours — a generous version of the weekend lie-in — while their reaction times kept being measured.
| Time in bed, 7 nights | After 3 recovery nights at 8 h |
|---|---|
| 9 hours | Never left baseline at any point |
| 7 hours | No evidence of recovery — stayed at the reduced level |
| 5 hours | No evidence of recovery — stayed at the reduced level |
| 3 hours | Rebounded quickly after one night, but recovery was incomplete |
Belenky et al., Journal of Sleep Research (2003). Performance measured on the psychomotor vigilance task.
The seven-hour row is the one worth pausing on. That is not dramatic deprivation — it is a fairly ordinary week. And three nights of eight hours did not undo it.
Why “stabilised” is not “recovered”
In the five- and seven-hour groups, performance declined and then flattened out. That plateau is easy to misread as adjusting. The authors' interpretation was different: the brain adapts to chronic restriction, and that adaptation is enough to stop the slide while leaving performance at a reduced level. They suggested those adaptive changes persist for several days after normal sleep is restored, which is what delays recovery.
So the feeling of having got used to six hours is consistent with the evidence. It just does not mean what people take it to mean. Getting used to it and being unaffected by it are different things.
What this study does not show
It is one study, with three recovery nights and one measure — a reaction-time task. It does not establish that sleep debt is permanent, or that catching up is pointless, and nobody should read it that way. A longer recovery period might have produced a different result; the study simply did not run one.
What it does establish is narrower and still useful: three nights was not enough for people who had been running two or three hours short. If your plan is a weekend, the evidence says the window you are giving yourself is smaller than the one that already proved insufficient.
The more useful thing to do instead
Raise the weekday number rather than banking on the weekend. An extra 30–45 minutes on five weeknights is more sleep than most weekend lie-ins deliver, arrives while the deficit is still small, and does not swing your body clock by hours in each direction.
For a single bad night, though, take the extra sleep — that is the case where recovery sleep works straightforwardly, and none of the above argues otherwise.
If you want to see the size of the gap rather than guess at it, the sleep debt calculator totals your last seven nights against your own target, and the sleep calculator works backwards from a wake time to a bedtime.
One more finding worth knowing
A companion study published the same year, by Van Dongen and colleagues in Sleep, held people at 4, 6 or 8 hours for fourteen nights. Those at six hours or less built up cognitive deficits the authors described as equivalent to up to two nights of total sleep deprivation — and their own sleepiness ratings suggested they were largely unaware of it.
Taken together the two studies make an awkward pair: the deficit is harder to clear than it feels, and harder to notice than you would expect. That is a good argument for counting the hours rather than trusting the feeling.
This is general information, not medical advice. Persistent poor sleep, loud snoring or daytime sleepiness that does not improve with more sleep are worth raising with a doctor — insomnia and sleep apnoea are treatable, and no amount of catching up addresses them.
Sources: Belenky G, et al. Patterns of performance degradation and restoration during sleep restriction and subsequent recovery. J Sleep Res (2003); Van Dongen HP, et al. The cumulative cost of additional wakefulness. Sleep (2003); CDC/MMWR — Prevalence of Healthy Sleep Duration among Adults (2016).