Why you stay up even when you’re exhausted

You wanted sleep at three in the afternoon. You wanted it while answering the final message, clearing the kitchen or travelling home. Then the day released you, the room became quiet, and sleep began to feel like an interruption.

One episode becomes another. One video becomes the discovery that somebody else has been awake and posting. You are tired enough to dislike what you are doing, but not quite willing to stop.

The easiest explanation is a failure of discipline. You knew tomorrow would be harder. You stayed awake anyway.

There is truth in that description, but not much help. It treats bedtime as a single decision made by a rested person under neutral conditions. Often it is the last negotiation in a day that has already made many claims.

To change the night, it helps to ask what staying awake is doing for you.

A particular kind of delay

Researchers use bedtime procrastination to describe going to bed later than intended when no external circumstance prevents it. The definition matters. A parent caring for a child, a person working a night shift, someone kept awake by pain, a person with insomnia and someone who freely watches another episode are not having the same problem.

The boundary is not always clear. Work can be externally required and still expand because the laptop is nearby. A person may technically be free to stop scrolling while feeling too activated to settle. Sleep timing is influenced by biology, health, environment, responsibilities and opportunity as well as choice.

In early studies, Floor Kroese and colleagues found that bedtime procrastination was associated with experienced insufficient sleep and with lower self-regulation. One study used a representative sample of 2,431 Dutch adults. Because the data were largely self-reported and observational, they cannot show that poor self-regulation alone causes the delay. They established the behaviour as a useful subject to investigate, not a complete explanation for it.

Later daily-life studies have continued to associate bedtime procrastination with less favourable aspects of sleep health. The consistent practical point is unsurprising but important: when wake time cannot move, delaying bedtime removes sleep opportunity.

For most healthy adults, a joint expert panel from the American Academy of Sleep Medicine and Sleep Research Society recommends regularly obtaining at least seven hours of sleep. Individual needs and circumstances vary, and duration is only one dimension of sleep. Timing, regularity, quality and sleep disorders matter too.

Knowing this does not necessarily make closing the day easier.

The night may be protecting something

Late leisure can provide something the day did not: autonomy.

Nobody is asking for an answer. The work has either ended or become socially unacceptable to continue. The entertainment is chosen, the silence is yours, and tomorrow has not yet begun. Going to bed can feel like surrendering the only unclaimed hour.

This experience is sometimes called “revenge bedtime procrastination.” It is a vivid phrase, but it is not a medical diagnosis or a single established scientific explanation. People delay bedtime for different reasons. In diary and qualitative research, participants report losing track of time, continuing work or study, using electronic leisure, socialising, seeking personal time and simply not feeling ready for sleep.

These motives can overlap. A phone offers stimulation, escape, connection and choice in the same small object. Removing it may reduce one route into delay without answering the need that made the route attractive.

Ask what the late night is currently providing:

• Freedom: Is this the first time today when nobody directs you?
• Completion: Are you trying to reach a point where the day feels properly finished?
• Relief: Is entertainment helping you avoid an uncomfortable thought or tomorrow’s demands?
• Connection: Are other people available only at this hour?
• Stimulation: Are you tired but not sleepy, or reluctant to exchange input for quiet?
• Momentum: Have you passed the point at which one more choice is easy to resist?

The answer may change from night to night. Its purpose is not to excuse the delay. It identifies what a workable plan must compete with.

“Go to bed earlier” is not yet a plan

A bedtime intention often has no clear beginning. “Tonight I will be sensible” must eventually become a particular action taken at a particular moment.

In a 2025 study, 119 university students reported their bedtime plans over two to four weeks while wearable devices estimated sleep. Participants planned a specific bedtime on fewer than one night per week at the median, and planned bedtimes were often exceeded. On nights with a plan, bedtime was about twelve minutes earlier and estimated total sleep about twelve minutes longer than on nights without one.

This was a small, culturally specific student sample, and the association does not prove that writing down a time will improve sleep for everybody. The size of the difference was modest. Still, the study exposes a common gap: wake times are often fixed by the world, while bedtimes remain a hope.

Instead of choosing only a time to be asleep, choose the first visible action in the transition:

At 10:40, after this programme ends, I will put the phone on charge outside reach and prepare what I need for the morning.

The time and action are examples, not a prescription. The useful features are that the cue is recognisable and the first step can actually be performed. “Sleep now” is not fully voluntary. Beginning a landing sequence often is.

Give autonomy an earlier appointment

If the late night is the only part of the day that feels self-directed, a stricter bedtime can feel like one more demand. Where circumstances permit, protect a smaller period of chosen time before exhaustion.

Twenty minutes used deliberately may feel more like leisure than an hour entered accidentally. The activity does not have to be improving: a game, a conversation, music, a bath or an undistinguished television programme can all be legitimate choices. The distinction is between leisure you chose and time you discover has disappeared.

Name the period before it begins: “From nine to nine-thirty, nothing in this house is a task unless it is urgent.” If your responsibilities make that impossible, do not turn the suggestion into another standard you are failing. The problem may be a genuine lack of time and support, not poor bedtime technique.

It is also worth asking whether work is borrowing from both ends of the day. A person cannot routinely extend work into the evening, rise at the same time and solve the resulting shortage entirely through better self-control at bedtime.

Add friction at the point where choice disappears

Digital media is not the only cause of bedtime delay, but it can remove natural stopping points. A book ends a chapter. A broadcast ends a programme. An infinite feed continuously supplies a reason not to decide yet.

Useful friction is small and specific:

• charge the device somewhere that requires standing up to reach it;
• turn off automatic continuation where the service allows it;
• decide the last episode before pressing play on the first;
• use a lamp, alarm or ordinary event as the cue to begin winding down;
• leave tomorrow’s first necessary item ready, so unfinished preparation does not re-enter the night.

Friction should interrupt automatic continuation, not make essential communication inaccessible. People who are on call, managing health needs or supporting others will need a different arrangement.

Do not make sleep another performance

Once sleep becomes a test, concern about succeeding can become its own form of wakefulness. A wearable score, perfect routine or precise bedtime may provide useful information for one person and increasing anxiety for another.

The aim is not to produce an immaculate evening. It is to make adequate sleep easier to approach more often.

Review a week with curiosity:

1. On which nights did bedtime move later than intended?
2. What was the delay providing in that moment?
3. What cue or stopping point was missing?
4. What need could be met earlier, even in a smaller form?
5. What one change is realistic for the coming week?

Persistent difficulty falling or staying asleep, loud snoring or breathing interruptions, overwhelming daytime sleepiness, or sleep problems that significantly affect daily life deserve assessment by an appropriate healthcare professional. A behavioural experiment cannot rule out a sleep disorder or resolve conditions outside your control.

You may not be staying awake because you do not value sleep. You may be staying awake because the night contains freedom, completion or relief that the day did not provide. A better bedtime begins by taking that need seriously—then finding a way to meet it that does not charge tomorrow for the entire cost.