What Losing It Actually Does

The central fact about sleep deprivation, and the one with all the practical consequences, is that it does not feel like impairment.

In chronic restriction studies, people held to six hours a night show performance deficits that accumulate steadily across two weeks. Their subjective sleepiness rises for a few days and then levels off. By the end, performance on vigilance and reaction tasks resembles total sleep deprivation, and the subjects report feeling somewhat tired. The gap between how you function and how you think you are functioning widens continuously, and there is no internal signal telling you it is happening.

That is why researchers reached for the comparison with intoxication. It was made deliberately, testing the same people on the same tasks under both conditions, because intoxication is a level of impairment society has already agreed is unacceptable. Around seventeen to nineteen hours awake produces deficits comparable to the drink-drive limit in several jurisdictions. The point is not that sleep loss is like being drunk in any experiential sense (it is not) but that we regulate one and treat the other as a matter of personal stamina.

The mechanism at the sharp end is the microsleep: a brief involuntary lapse of a few seconds during which the brain is effectively asleep with the eyes open. It shows on an EEG and is almost never noticed by the person having it. Frequency rises with accumulated debt and peaks at the circadian low in the early hours. An accident investigation that finds no evidence of distraction is entirely compatible with the operator having been asleep for four seconds.

A fortnight of six-hour nights is an extremely ordinary way to live, which is what makes the chronic restriction result more important than the dramatic total-deprivation studies. It also explains why a long weekend does not restore things the way people expect: the debt accumulated over two weeks is not repaid by two nights, and the subjective recovery arrives well before the objective one.

The individual differences are real and much smaller than the folklore. Genuine short sleepers exist and carry identifiable variants; they are rare. Everybody else who believes they are one is, on the evidence, simply impaired and unaware, which is precisely the report the research predicts they would give, and is therefore not evidence of anything.

All of which has an uncomfortable implication for institutional design. Any system that relies on individuals declaring themselves too tired to continue is relying on a judgement that the condition itself degrades. Hospitals, haulage firms and air traffic control have all discovered this, generally the hard way. The arrangements that work do not ask: hard limits on consecutive hours, mandated breaks, and handover protocols that assume the outgoing person is impaired. None of them depends on anybody's self-assessment at exactly the point where self-assessment stops working.