You're exhausted by 3pm. You yawn through dinner. You count down the hours until you can get into bed. And then - the moment your head hits the pillow - your brain switches on. Thoughts start circling. Your body feels tired but your mind is running. Sleep doesn't come for an hour, or two, or more.
This experience is so common it has a name: wired but tired. And despite how familiar it feels, most people don't understand what's actually causing it - which means they keep trying solutions that address the wrong thing entirely.
This isn't about screen time before bed. It isn't about not winding down properly. It's a specific biological state, and once you understand what's driving it, the path out becomes much clearer.
The wired-but-tired state is the result of two systems being out of sync: your cortisol rhythm and your sleep drive.
Your sleep drive - technically called adenosine pressure - builds throughout the day as a chemical byproduct of being awake. By evening, it's genuinely high. Your body is tired in a real, measurable sense. This is the "tired" half of the equation.
The "wired" half comes from cortisol. In a healthy circadian rhythm, cortisol peaks in the morning to get you up and running, then gradually declines through the day, reaching its lowest point in the late evening to allow sleep. This decline is what lets adenosine do its job.
When cortisol fails to follow that declining pattern - when it stays elevated into the evening or spikes unexpectedly at night - it directly suppresses the sleep-onset signal. Your body is exhausted. Your nervous system is alert. Both are true simultaneously, which is why the experience feels so contradictory and so frustrating.
Cortisol activates the same arousal pathways that keep you alert during the day. When evening cortisol remains elevated, it chemically overrides adenosine pressure - your body's natural sleep signal. The tiredness is real. The cortisol just won't let you act on it.
This is the question most sleep advice completely ignores. It assumes the problem is behavioural - not winding down, too much blue light, not meditating enough. Sometimes those factors contribute. But for most people experiencing chronic wired-but-tired, the underlying driver is physiological.
Cortisol is produced by a system called the HPA axis - the hypothalamic-pituitary-adrenal axis. This system is supposed to be self-regulating: cortisol rises when needed, then triggers a feedback mechanism that dials it back down.
That feedback mechanism depends heavily on magnesium. Magnesium acts as a natural brake on the HPA axis - it sits at multiple points in the cortisol production pathway and limits how much cortisol gets produced. When magnesium is depleted, that brake weakens. The HPA axis produces more cortisol than it should, for longer than it should, and the evening decline that's supposed to prepare you for sleep becomes shallow or delayed.
Here's what makes this particularly persistent: stress depletes magnesium, and magnesium depletion makes you more reactive to stress. Every cortisol spike burns through magnesium faster. The lower your magnesium, the higher and longer your cortisol runs. The more your cortisol runs, the more magnesium you lose.
The wired-but-tired cycle is not a willpower problem. It's a mineral depletion problem that feeds itself the longer it runs. You cannot meditate your way out of a physiological deficit.
This is why people in this pattern often find that stress management techniques help somewhat but never fully resolve it. The techniques reduce the cortisol input, but if the underlying magnesium deficit remains, the HPA axis stays dysregulated.
There's a second mechanism running in parallel. GABA - gamma-aminobutyric acid - is your brain's primary inhibitory neurotransmitter. It's what allows your nervous system to actually stop firing. The transition from wakefulness to sleep requires a meaningful increase in GABA activity.
Magnesium is a required cofactor for GABA receptor function. Without adequate magnesium, GABA receptors become less responsive - meaning even when your brain tries to shift into sleep mode, the signal doesn't land properly. The result is a mind that won't switch off, regardless of how tired the body underneath it feels.
Melatonin signals when to sleep - it's a timing hormone. It does nothing to address elevated cortisol or impaired GABA function. Taking melatonin while cortisol is still elevated is like pressing the accelerator and brake simultaneously. The signal arrives; the system can't respond to it.
Wired but tired presents in a specific and recognisable way. The following checklist isn't diagnostic - but if most of these apply to you, you're looking at HPA axis dysregulation rather than simple insomnia.
The last point is particularly telling. If your stress response feels amplified - if small things create a level of physiological arousal that takes hours to settle - that's a hallmark of a depleted HPA brake, not a character flaw.
The wired-but-tired cycle has three points of intervention. Addressing all three produces faster, more lasting results than targeting any single one.
This is the root intervention. The specific forms that matter for this pattern are magnesium bisglycinate - which crosses the blood-brain barrier and activates GABA receptors directly - and magnesium taurate, which specifically buffers the cortisol stress response at the HPA axis level.
Timing matters: 60 to 90 minutes before bed gives bisglycinate time to reach the brain before sleep onset. Results are gradual - most people notice the wired feeling beginning to quiet around day 8 to 12, with more substantial change by day 21.
For a full breakdown of which forms do what and why, see the guide to why magnesium may not have worked before - the form problem is usually what's standing between people and results.
While magnesium addresses the physiological root, reducing the triggers that spike evening cortisol accelerates the process. The most impactful are:
Beyond magnesium bisglycinate, two things support the GABA shift into sleep. First, consistent sleep timing - the same bedtime within 30 minutes every night trains the GABAergic system to anticipate the transition. Second, a genuine wind-down window of 45–60 minutes - not because winding down is morally virtuous, but because GABA activity increases in response to a consistent pre-sleep behavioural sequence. The sequence itself doesn't matter much; the consistency does.
This depends entirely on how long the cycle has been running and how depleted the underlying magnesium stores are. For most people who implement all three interventions:
| Timeframe | What typically shifts |
|---|---|
| Days 1–7 | Sleep environment and behavioural changes take effect. Evening cortisol input reduced. Little change in the wired feeling itself yet. |
| Days 8–14 | First signs of change - the wired feeling at bedtime begins to feel slightly less intense. Mind still active but settles faster. |
| Days 14–21 | Meaningful shift in most people. Falling asleep within 30 minutes rather than 60–90. Fewer early-morning wake-ups. |
| Day 21–30 | Baseline changed. The second wind in the evening begins to flatten. Sleep feels qualitatively different - deeper, less fragmented. |
| Month 2–3 | HPA axis rhythm restored in most cases. The wired-but-tired pattern becomes the exception rather than the default. |
The temptation is to evaluate too early. If you're on day 6 and nothing has changed, that's completely expected - the cellular magnesium rebuild takes time that cannot be shortcut. The full explanation of how the cortisol awakening response works covers why this timeline is physiologically inevitable, not a sign the approach isn't working.
Wired but tired is one of the most common sleep complaints and one of the least well-addressed - because most interventions treat the symptom rather than the system. The system is the HPA axis. The lever is magnesium. The timeline is weeks, not nights.
The second wind is cortisol. When the HPA axis is dysregulated, cortisol doesn't follow its natural evening decline - instead it stays elevated or spikes again in the late evening, creating a temporary surge in alertness that overrides your sleep drive. It's not energy - it's stress hormone activity. The body is genuinely tired; the cortisol is simply preventing the sleep signal from landing. Rebuilding magnesium stores, which regulate the HPA axis, is the most direct way to flatten this pattern over time.
They share physiological overlap but are not the same thing. Anxiety is a psychological state with a physiological component. Wired but tired is a physiological state that can produce psychological symptoms - racing thoughts, a sense of alertness or unease at night - without an underlying anxiety disorder. The distinction matters because the interventions differ. Addressing HPA axis dysregulation through magnesium and cortisol management often resolves the nighttime mental activity without any psychological treatment needed. If you have clinical anxiety, addressing the physiological root can still significantly reduce its nighttime manifestation.
Yes - and this is actually the more common presentation. The initial cortisol decline allows sleep onset, but if the HPA axis is dysregulated, the early-morning cortisol rise (the cortisol awakening response) fires too early - typically in the 2am–4am window - pulling you out of sleep feeling alert and unable to return. This is the wired-but-tired pattern manifesting in the middle of the night rather than at bedtime. The same mechanism, the same fix.
Sleep restriction - deliberately limiting time in bed to build sleep pressure - can help with sleep-onset insomnia driven by conditioned arousal. It does not address HPA axis dysregulation. If your primary issue is elevated cortisol preventing sleep, sleep restriction may help you fall asleep faster but will not resolve the underlying pattern. It can also backfire by increasing cortisol further through sleep deprivation. Address the physiological root first; sleep restriction is a tool for a different problem.
Insomnia is a broad umbrella term covering any persistent difficulty sleeping. Wired but tired is a specific subtype characterised by the simultaneous presence of physical fatigue and cognitive or physiological alertness. Classic insomnia can have many causes - conditioned arousal, circadian misalignment, hyperarousal, sleep apnoea. Wired but tired specifically points to HPA axis dysregulation as the primary driver. Identifying it as a distinct pattern matters because it leads to a more targeted intervention than general sleep hygiene advice.