Melatonin is often called the body's sleep hormone, but a more accurate description is that it is the body's signal of biological night.
Its production is controlled by the circadian clock in the brain. Information about light reaching the eyes is passed to the suprachiasmatic nucleus - the master circadian clock - which in turn controls melatonin production by the pineal gland. During daylight, circulating melatonin is normally very low. As biological night begins, provided evening light is sufficiently dim, melatonin levels start to rise, remain elevated through much of the night and fall again towards morning. Bright evening and night-time light can suppress that rise.
That distinction matters because melatonin doesn't simply knock the brain into sleep the way a conventional sedative can. Its rising concentration tells tissues throughout the body that night has arrived. Sleep is one part of that larger circadian transition, alongside changes in alertness, body temperature and other biological rhythms.
Externally supplied melatonin works in two different ways. The first is an immediate sleep-promoting effect: melatonin acts principally through MT1 and MT2 receptors, including receptors within the brain's circadian system, reducing the drive for wakefulness and making sleep more likely. This helps explain why melatonin can shorten the time taken to fall asleep even without necessarily moving the circadian clock itself.
The second effect is very different: melatonin can actually move the timing of the internal clock. Taken at certain times - particularly in the afternoon or early evening relative to a person's biological night - it can shift circadian rhythms earlier. Taken at a different biological time, it can shift them in the opposite direction. Human phase-response studies have demonstrated this directly.
This is why when melatonin is taken can matter almost as much as how much is taken. Someone whose body clock is running late has a fundamentally different problem from someone whose circadian timing is normal but who simply has difficulty making the transition into sleep. Melatonin has biological relevance to both, but through somewhat different actions.
The dose story is also more complicated than the familiar idea that more should produce more sedation. A 2024 dose-response meta-analysis examined 26 double-blind randomised controlled trials, representing 1,689 observations. Across the studies, melatonin reduced sleep-onset latency by about 9 minutes and increased total sleep time by about 20 minutes compared with placebo. When the researchers modelled dose, the sleep-promoting effect increased up to roughly 4mg, after which additional dose did not appear to provide a corresponding increase in benefit.
Perhaps even more interestingly, the same analysis found that timing independently influenced the result. Taking melatonin earlier relative to the intended sleep period was associated with a larger reduction in the time taken to fall asleep, and the authors' modelling suggested that administration around three hours before the desired bedtime could produce a stronger effect than the commonly studied pattern of taking it immediately before bed.
An earlier 2022 systematic review reached a related conclusion from 30 studies involving 1,294 healthy participants: melatonin and other melatonergic compounds could measurably advance sleep-wake timing, with their effects varying according to both dose and the time at which they were administered.
That is what makes melatonin fundamentally different from most substances associated with sleep. It isn't merely something that can make a person drowsy; it is one of the chemical signals by which the brain tells the rest of the body what time it is.
