Tired but Can't Sleep: Why Exhaustion Doesn't Make You Sleepy

You are wrecked. You got into bed because you could barely keep your eyes open on the sofa. And now you are flat on your back, wide awake, with a brain that has decided this is the moment to review your entire week.

Most sleep advice assumes the problem is that you are not tired enough. Cut the caffeine, move more, get up earlier. But you are tired — that is the whole complaint. The advice is aimed at a problem you do not have.

Tired and sleepy are not the same thing

This sounds like splitting hairs. It is the most useful distinction in the entire subject.

Sleepiness is how likely you are to fall asleep. Fatigue is how worn out you feel. They travel together often enough that we use the words interchangeably, but they are measurably different. A 2024 study of 606 sleep disorder patients found them only moderately correlated, r = 0.334 — related, clearly, but nowhere near the same measurement.

What predicted each one differed sharply too. For fatigue, the strongest predictor by a distance was depression, at 42.7 % of the explained variance, with insomnia severity second at 12.3 %. For sleepiness, insomnia symptoms came first. Two complaints, two different underlying patterns.

Ordinary speakers already know this without the research. When 96 people rated the words sleepy, drowsy, tired and fatigued, all four differed significantly from one another, with “sleepy” the most positive and “fatigued” the most negative. The authors concluded that clinicians and researchers should stop treating them as synonyms.

So when you say *I am so tired, why can't I sleep*, there may be no contradiction in it at all. You can be thoroughly fatigued and not remotely sleepy.

What actually puts you to sleep

Two systems, not one. The framework sleep research has used for three decades is the two-process model: a homeostatic process, Process S, which builds the longer you are awake, and a circadian process, Process C, run by your internal clock. Sleep arrives when S is high and C permits it.

Notice what is not in that model: how hard your day was. Being worn out by work, a difficult conversation or three hours of childcare adds to your fatigue. It does not add to Process S, which tracks time awake rather than effort spent. That is why you can finish a brutal day feeling destroyed and still arrive in bed with an entirely ordinary amount of sleep pressure.

The third force, and the one that is actually your problem

Sleep drive and the clock explain when sleep is *available*. They do not explain why it fails to turn up when both are in your favour. For that, the research points at arousal.

A 2025 study in the Journal of Sleep Research compared the sleep architecture of 1,710 people with insomnia and 1,455 without. The insomnia group had a higher probability of transitioning into wakefulness from every single sleep stage, and a lower probability of transitioning out of wakefulness into any sleep stage. The authors name the mechanism directly: a dysfunction in sleep quality “with a continuous hyperarousal”.

Take that as a description rather than a diagnosis and it maps onto the experience exactly. There is nothing wrong with your supply of sleepiness. Something keeps overriding it.

It is also why “just relax” is useless as an instruction while being accurate as a description. Arousal is not a decision.

About cortisol, before you buy anything

Search this topic for five minutes and something will offer to lower your cortisol. The underlying research is real, and it is far thinner than the marketing suggests.

The landmark study is from 2001. Vgontzas and colleagues monitored insomniacs and controls in a sleep laboratory for four nights, then measured ACTH and cortisol around the clock. Both were higher in the insomnia group, and — the detail that matters — “the greatest elevations were observed in the evening and first half of the night”. Precisely when you are lying there.

Their conclusion deserves quoting, because it is the reverse of how cortisol is usually sold to you. The findings were “consistent with a disorder of central nervous system hyperarousal rather than one of sleep loss”. Cortisol was a marker of the arousal, not a deficiency to be corrected. And the target they named was “the overall level of physiologic and emotional arousal, and not just to improve the nighttime sleep”.

Now the part nobody repeats. That study had eleven insomniacs and thirteen controls. The ACTH difference reached significance at P = 0.04. The cortisol difference did not: P = 0.07. A 2022 open-access review in Sleep Science restates the finding, and it has held up well enough to stay in the literature — but a confident claim that your cortisol is what is keeping you awake rests on twenty-four people and a borderline result.

None of that means a supplement will fix anything, and none of it is a reason to buy one. I have no idea what your cortisol is doing, and neither does anyone selling you something to change it.

What follows from this

Stop going to bed earlier. It is the obvious response to feeling exhausted, and it makes this particular problem worse. Less time awake means less Process S, so you arrive with even less sleep pressure and more time to lie there noticing it.

Separate the two states before you act. Worn out but alert is not the same as fighting to keep your eyes open. If you are the first, bed is the wrong tool at that moment, and the useful move is to be somewhere else until the second one shows up.

Work on the evening, not the last ten minutes. The arousal blocking sleep at eleven has been building since about seven. A wind-down that begins when you switch off the light begins far too late.

And know where the better evidence sits. The technique with the strongest research behind it for not lying awake is getting out of bed, which I covered in stimulus control and CBT-I rather than repeat here. If the wired feeling follows a drink or a hard late session, that has its own mechanism.

When this is the wrong article

Go back to that 2024 finding: depression was by far the strongest predictor of fatigue. If what you mainly recognise is being exhausted in a way sleep does not touch, that is worth raising with someone qualified, and it is a different conversation from sleep technique.

The same holds if you snore heavily, wake gasping, or have been told you stop breathing — that points towards sleep apnea. And if this has run for months rather than weeks, ask about CBT-I by name.

What we could not verify

Nothing here was tested. This comes from published research, not from anything I measured.

Three of the six sources were read as abstracts only. The Vgontzas cortisol study, the 2024 fatigue-and-sleepiness study and the two-process model reappraisal are paywalled. The 2025 sleep architecture study, the 2022 word-meaning study and the 2022 Sleep Science review are open access and were read in full.

The cortisol evidence is thinner than this article can make it sound even while trying not to. Eleven insomniacs, thirteen controls, and a cortisol result at P = 0.07.

The word-meaning study was 96 undergraduates, not patients, and in one language. It shows the words are understood differently. It does not establish that everyone means the same thing by them.

The practical section reasons from mechanism, not from trials. No study ranked going to bed earlier against getting out of bed against starting the wind-down sooner. Where a technique does have trial evidence behind it, I have said so and linked to it.

I am not a doctor, a sleep scientist or a therapist. Insomnia that has settled in is a diagnosable condition with a recommended treatment, and the useful version of that treatment comes from a person.