Circadian Rhythm — Your Body's Internal Clock

Photo: Klub Boks / Pexels
Your body does not run on a single clock. A tiny cluster of brain cells sets the schedule, but most tissues keep their own timers, and when the two drift out of step, as they do after a transatlantic flight or a run of night shifts, the effects show up in sleep, blood sugar, and wellbeing. In a landmark controlled measurement, the human internal day averaged 24.18 hours, not exactly 24. Here is how the circadian rhythm works, what resets it, and where the research on timing and health is solid or still uncertain.
Educational content, not medical advice. This article summarizes published research on body clocks. Anyone with persistent sleep problems, or who works nights and is concerned about health effects, should talk to a clinician.
What is a circadian rhythm?
A circadian rhythm is a biological cycle of about 24 hours that continues even without outside cues. The word comes from the Latin circa diem, "about a day." The "about" matters. Take away every clock and window and a person's rhythm does not stop; it free-runs on its own period, which for most people is a little over 24 hours.
That persistence is the test that separates a true circadian rhythm from a simple reaction to the sun. Plenty of things change between day and night, but only some of them keep changing on schedule in constant darkness. Core body temperature, hormone levels, alertness, and the activity of thousands of genes belong to the second group.
Circadian clocks are not unique to people. Fungi such as Neurospora, single-celled algae such as Chlamydomonas, fruit flies, hamsters, and mice all have them, and much of what is known about the human clock came from studying those simpler organisms first. In Neurospora, for example, clock mutants found in the early 1980s, most of them at a locus called frq, had periods ranging from 16.5 to 29 hours (Gardner and Feldman, 1980). That kind of result told biologists the clock was built from genes.
How long is a human day, really?
The older literature reported human free-running periods much longer than 24 hours, sometimes approaching 25. In 1999, Charles Czeisler and colleagues measured melatonin, core body temperature, and cortisol rhythms in healthy young and older volunteers under tightly controlled lighting and found a tight distribution close to 24 hours. They reported that the intrinsic period of the human circadian pacemaker averages 24.18 hours in both age groups.
An internal day slightly longer than 24 hours explains something familiar. If your clock runs long, you drift later each day unless something resets it, and the resetting signal is mainly light. It also explains why flying east tends to feel worse than flying west: eastward travel asks the clock to move earlier, which works against its natural tendency to run long, while westward travel asks it to stay up later, which fits the tendency. That is an inference from the free-running period, and individual experience varies.
How does the master clock in the brain keep time?
Humans have a central pacemaker in the hypothalamus, a structure at the base of the brain, called the suprachiasmatic nucleus (SCN). The name describes its location: just above (supra) the crossing of the optic nerves (chiasm). The SCN is a small paired structure, but it plays an outsized role. Damage to the area near it can disturb rhythms dramatically. In one published case, a woman with an aneurysm near the SCN showed a free-running rest-activity cycle of about 22.5 hours for 19 days before her rhythm re-entrained to social cues.
What makes the SCN special is not that it is the only place with a clock. It is that its neurons keep better time as a group than any single cell does. Welsh, Takahashi, and Kay reviewed the evidence in 2010 and concluded that individual SCN neurons can act as autonomous oscillators, but that network properties are essential to the SCN's normal function. Synaptic input from neighbors synchronizes the cells, reinforces their cycles, and makes the rhythm more robust and more precise.
Think of a choir. Each singer can hold a pitch alone, but a choir singing together is steadier than any individual, and it is much harder for one off-key voice to pull it astray. The SCN works in a similar way: a crowd of noisy cellular oscillators becomes a reliable beat.
How does light reach the clock?
The SCN is set by light, but the light does not travel through the same cells that give you vision. In 2002, Berson, Dunn, and Takao showed in Science that retinal ganglion cells projecting to the SCN respond to light on their own, even when rod and cone input is blocked. These intrinsically photosensitive retinal ganglion cells (ipRGCs) carry the pigment melanopsin. Their sensitivity, color tuning, and slow response matched the known properties of circadian entrainment, so the authors suggested they may be the primary photoreceptors for it.
This is why the clock can respond to the quality and timing of light without any conscious sense of brightness. It also explains why the effect of a light source on your body clock is not the same as how bright it looks to you.
How do genes build a 24-hour oscillator?
Underneath the SCN's network is a clock built from genes. The story began with fruit flies. In 1971, Ronald Konopka and Seymour Benzer reported in PNAS that three mutants with altered 24-hour rhythms (one arrhythmic, one with a 19-hour period, and one with a 28-hour period) all affected the same gene on the X chromosome. That gene was later named period.
Decades of work then showed that the mammalian clock is made of interlocking loops. In 2017, Jeffrey Hall, Michael Rosbash, and Michael Young received the Nobel Prize in Physiology or Medicine for discovering molecular mechanisms controlling the circadian rhythm. Joseph Takahashi's 2017 review in Nature Reviews Genetics describes the modern picture: a cell-autonomous clock built on transcription-translation feedback loops that globally regulates transcription factor binding, RNA polymerase II recruitment, and chromatin remodeling. Its genomic targets are widespread and tied to metabolism, cell growth, and physiology.
The feedback loop in four steps
In simplified form, the core loop runs like this:
- Two proteins, CLOCK and BMAL1, pair up and switch on the genes for PER (Period) and CRY (Cryptochrome).
- The cell makes PER and CRY proteins, which accumulate slowly in the cytoplasm.
- PER and CRY travel back into the nucleus and shut down CLOCK and BMAL1, which turns off their own production.
- Over many hours, the PER and CRY proteins are tagged and broken down. When enough of them are gone, CLOCK and BMAL1 are free again and the cycle restarts.
The analogy is a bathtub with a delayed drain valve. Water (PER and CRY protein) fills slowly; when the tub gets full, a sensor opens the drain; when the tub is nearly empty, the drain closes and filling begins again. The delays in filling and draining are what stretch a quick on-off switch into a slow cycle of roughly a day.
Why it takes a day and not an hour
The slowness is not accidental, and it can be tuned by enzymes that add phosphate groups to PER. In 1988, Ralph and Menaker reported a mutation in golden hamsters called tau: wild-type animals averaged about 24 hours, heterozygotes about 22 hours, and homozygotes close to 20 hours. In 2000, Lowrey and colleagues identified the mutated gene as casein kinase I epsilon, and showed that the mutant enzyme was impaired in phosphorylating PERIOD proteins.
Humans have the same kind of vulnerability. In 2001, Toh and colleagues mapped familial advanced sleep phase syndrome, an inherited condition that shifts sleep and hormone rhythms about four hours earlier, to a serine-to-glycine change in the human PER2 protein at the site that binds casein kinase I. One amino acid is enough to move a whole family's bedtime.
Does every tissue have its own clock?
Yes, and that fact reshapes how to think about timing. In 2000, Yamazaki and colleagues built rats whose cells glowed when the Per1 clock gene was active. Light output from the SCN stayed rhythmic for up to 32 days in culture, while liver, lung, and muscle tissue faded within a few cycles. After a shift in the light cycle, the SCN re-adjusted faster than behavior or peripheral tissues did, which suggests that the central pacemaker keeps peripheral clocks in phase.
The SCN, in other words, is closer to a conductor than a sole timekeeper. It sets the tempo, and the liver, gut, muscle, pancreas, and heart each play their parts with some independence. The metaphor has a limit: the liver can respond to different cues than the SCN does.
Food can move the liver's clock
That limit became clear in a classic experiment. In 2000, Damiola and colleagues restricted when mice could eat. The phase of circadian gene expression in peripheral tissues shifted by up to 12 hours while the SCN was unaffected. Liver reset faster than kidney, heart, or pancreas, but after a week of daytime feeding the peripheral phases converged.
Feeding, in this view, is a second time-giver for the body, distinct from light. When light and food point to different times, which can happen with shift work or eating late at night, the central and peripheral clocks can fall out of alignment. That is the idea behind internal desynchrony, and it is one of the main proposals for why mistimed schedules affect metabolism.
What is the strongest signal for resetting the clock?
Light is the strongest signal. A zeitgeber (German for "time-giver") is any cue that resets the clock, and light dominates because the SCN receives it directly. But the effect of light depends on when it arrives, a relationship captured in a phase response curve (PRC).
In 2003, Khalsa and colleagues measured a PRC to single bright-light pulses in 21 healthy adults. Each volunteer went through constant routines in dim light before and after a 6.7-hour exposure to bright light, and plasma melatonin marked circadian phase. The result was a type 1 curve with a peak-to-trough amplitude of 5.02 hours. Light before the core body temperature minimum produced phase delays; light after it produced phase advances. There was no stretch of the day in which light had no effect.
Put plainly: light in the evening and early night tends to push the clock later, and light in the late night and morning tends to push it earlier. That one rule explains many everyday observations about late nights and early risers.
How bright is bright?
The reason indoor and outdoor light differ so much for the clock is their intensity. Typical illuminance figures, measured in lux, run like this:
| Setting | Illuminance (lux) |
|---|---|
| Full daylight, not direct sun | 10,000 to 25,000 |
| Overcast day | about 1,000 |
| Office lighting | 320 to 500 |
| Living room lighting (an Australian household survey, 1998) | about 50 |
Source: the figures are drawn from the lux entry on Wikipedia, which compiles them from standard references. The gap between an overcast sky and an office is roughly a factor of two to three, and the gap between daylight and a living room can exceed a factor of 100. A body that spends most of its day indoors receives a much weaker daytime signal than the one its clock evolved with, and a much stronger evening signal than a pre-electric night provided.
What does melatonin do, and what does it not do?
The pineal gland releases melatonin at night, under instruction from the SCN. Researchers often use the dim light melatonin onset (DLMO), the moment when melatonin begins to rise under dim light, as a marker of where a person's clock is. Bright light, especially blue-rich light, suppresses melatonin secretion.
A study that put this to a practical test was Chang and colleagues' 2015 comparison of reading on a light-emitting eReader with reading a printed book before bed. Participants who read on the light-emitting device took longer to fall asleep, felt less sleepy in the evening, had lower evening melatonin, showed a later circadian clock, and were less alert the next morning. The authors concluded that evening exposure to the device delayed the clock and suppressed melatonin.
Melatonin as a timing signal
Melatonin is often described as a sleep hormone, but in clock research it is better thought of as a "dark" signal that also nudges the clock. The question of dose and timing was tested in 2010 by Burgess and colleagues, who built a phase response curve for a 0.5 milligram dose and compared it with an earlier curve for 3.0 milligrams in 34 healthy adults. Advances were largest when the lower dose was taken in the afternoon, roughly 2 to 4 hours before DLMO. When each dose was given at its best timing, both produced similarly sized phase advances and delays.
The practical implication for researchers was that timing matters more than quantity. That finding comes from a single laboratory study in healthy adults. It is not a recommendation to take melatonin, and the long-term safety of supplements in different groups has not been settled by this work.
What else cycles through the day?
Many body functions rise and fall on the clock:
- Core body temperature. It reaches a daily minimum in the small hours of the morning, and the phase of that minimum is used as a reference point in clock experiments like Khalsa's.
- Cortisol. Cortisol rises sharply after waking in what is called the cortisol awakening response (CAR). In a 2009 review, Fries, Dettenborn, and Kirschbaum described the CAR as a sharp rise in cortisol after waking that sits on top of the daily rhythm, influenced by sex, health status, behavior, and perceived stress. They proposed that anticipating the coming day largely determines its size, while noting that its exact function is still unclear.
- Glucocorticoids more broadly. A 2017 review in Endocrine Reviews described the 24-hour rhythm of circulating glucocorticoids as driven by the central and peripheral clocks, with misalignment between them potentially harmful.
- Alertness and sleep pressure. The need to sleep builds the longer you are awake, and the clock determines when that pressure can be discharged. Researchers describe this as a two-process model of sleep regulation, in which a homeostatic drive and a circadian drive interact.
The last item is why a person can feel wide awake at 3 a.m. after being up all day, or sleepy at 3 p.m. after a good night. For the full sleep picture, see the science of sleep and how sleep affects the brain. The hormone rhythms are covered in how hormones control your body and cortisol and the stress response.
Why are some people larks and others owls?
Chronotype is the tendency to prefer earlier or later hours, and it varies widely. A nationwide Finnish twin study by Koskenvuo and colleagues, covering 8,753 adult twin pairs, estimated that genetic factors account for about half of the variation in diurnal type: a broad-sense heritability of 49.7 percent (95 percent confidence interval 46.4 to 52.8), with the rest coming from environmental factors not shared by siblings.
Genome-wide work shows how many genes are involved. In 2019, Jones and colleagues analyzed data from 697,828 people and raised the number of loci linked to morningness from 24 to 351. The loci tracked with sleep timing recorded by activity monitors, and the genes were enriched in circadian, signaling, and tissue-expression pathways. Their Mendelian randomization analysis suggested that morningness was causally linked to better mental health, but not to body mass index or type 2 diabetes.
Social jetlag
Chronotype collides with the clock on the wall. In 2006, Wittmann, Dinich, Merrow, and Roenneberg introduced the idea of social jetlag, the mismatch between biological and social time, using questionnaires from 501 volunteers. Late chronotypes showed the largest gap between work-day and free-day sleep. The strongest link was with smoking, and associations were strongest in people up to age 25. The authors attributed these correlations mainly to social jetlag rather than chronotype alone, and suggested that work and school schedules could be adapted to chronotype where possible. Social jetlag is an association study, so it cannot show that the mismatch causes those outcomes.
When timing becomes a disorder
Two inherited or clinical patterns show the extremes. Familial advanced sleep phase syndrome is the rare condition Toh described. At the other end, delayed sleep-wake phase disorder is a persistent pattern of late sleep and wake times that does not fit school or work. A 2024 review of adolescents by Narala and colleagues noted that recent studies indicate a prevalence between 1 and 16 percent, with the wide range reflecting differences in definition and population. Diagnosis and treatment are clinical matters, and the condition is often mistaken for ordinary insomnia or a motivation problem.
| Pattern | Typical features | What the clock is doing | Evidence source in this article |
|---|---|---|---|
| Morning type | Prefers early waking and sleeping | Earlier phase | Koskenvuo 2007; Jones 2019 |
| Evening type | Prefers late waking and sleeping | Later phase | Same; Wittmann 2006 (social jetlag) |
| Familial advanced sleep phase | Sleep and hormone rhythms about four hours earlier | Short period or altered PER2 phosphorylation | Toh 2001 |
| Delayed sleep-wake phase disorder | Persistently late sleep that conflicts with obligations | Delayed phase | Narala 2024 (1 to 16 percent, adolescents) |
| Night shift work | Awake and eating at biological night | Central and peripheral desynchrony | Scheer 2009; IARC |
| Eastward jet lag | Clock must move earlier | Acute misalignment | Czeisler 1999 (period 24.18 h) |
What happens when the clock is forced out of alignment?
The cleanest experiment on this question was done in a laboratory. In 2009, Scheer and colleagues put 10 adults through a 10-day protocol that scheduled eating and sleeping about 12 hours out of phase with their body clocks. Misalignment lowered leptin (a hormone that signals fullness), raised glucose despite higher insulin, reversed the daily cortisol rhythm, increased blood pressure, and reduced sleep efficiency. In 3 of 8 participants with usable data, glucose responses resembled prediabetes.
That is a small, short, tightly controlled study, and it shows what can happen in principle. It does not say how much risk a given shift pattern carries over decades.
Shift work and cancer
For long-term shift work, the evidence comes from observational studies and mechanistic research, and it has been reviewed by the International Agency for Research on Cancer (IARC). IARC evaluated shift work involving circadian disruption in 2007 (Straif and colleagues, published in The Lancet Oncology) and re-evaluated night shift work in Monograph Volume 124 in 2019. The Volume 124 working group reviewed cancer studies in exposed people, animal studies of shifted light-dark schedules, and mechanistic evidence. IARC rates night shift work as probably carcinogenic to humans (Group 2A), a category that rests on limited evidence in people combined with stronger evidence from animal or mechanistic studies. The category describes how confident reviewers are that a hazard exists, not how large the risk is for an individual.
Jet lag
Jet lag is the same problem on a shorter timescale: the SCN and the local day disagree, and the rest of the body has to catch up. If you travel across time zones, the central clock adjusts at a limited rate. For more detail on how this plays out, including recovery, see why we get jet lag. Researchers have also studied a more extreme version: astronauts on the International Space Station, who see many sunrises and sunsets each day. That research is covered in sleep and circadian rhythms in orbit.
Does when you eat matter?
The mouse experiment of Damiola and the human experiment of Scheer both suggest that meal timing interacts with the clock, and a research field called time-restricted eating has grown up around that idea. Studies so far have been mixed.
In 2020, Wilkinson, Panda, Taub, and colleagues ran a single-arm trial in which 19 adults with metabolic syndrome ate within a self-selected 10-hour window for 12 weeks. The authors reported improved cardiometabolic health. Single-arm means no control group, and 19 people is a small sample.
A randomized trial reached a more cautious conclusion. Also in 2020, Lowe and colleagues compared 16:8 time-restricted eating with eating three structured meals a day over 12 weeks in adults with overweight or obesity. The time-restricted group lost a small amount of weight, but the difference between groups was not significant, and most secondary markers did not differ. The authors concluded that time-restricted eating alone was no more effective for weight loss than eating throughout the day.
A review of timing of eating in adults by Allison and Goel in 2018 reached a similar balance: eating earlier in the day sometimes benefits weight and metabolic outcomes, but results are inconsistent, and the authors called for longer and more rigorous studies. For a wider look at fasting research, see intermittent fasting: what the science says.
Common confusions about body clocks
"Melatonin is a sleeping pill." It is better described as a timing signal. The Burgess study shows that when you take it relative to your clock changes what it does, and that a lower dose can do as much as a higher one.
"Blue light is the only problem." Blue-rich light is potent for ipRGCs, but the effect of light on the clock depends on intensity, duration, and timing, and bright light of any color late in the evening can delay the clock. The Chang study was about a bright, light-emitting reading device, not about every screen at any brightness.
"Being a night owl means being lazy." Twin data suggest genes account for about half of the variation in chronotype, and a single amino acid change can move a whole family's sleep by hours. Preference is not the same as discipline.
"Circadian rhythm is the same as sleep." Sleep is one output of the clock. The clock also schedules body temperature, hormone release, and gene expression in tissues that never sleep, such as the liver.
"Eating late is always harmful." The mechanisms are plausible, and lab studies of misalignment show clear effects. But human trials of time-restricted eating have not shown consistent benefits for weight loss, and conclusions about long-term health require longer studies.
Frequently asked questions
Q: How long is the human circadian rhythm without time cues? A: In the controlled study by Czeisler and colleagues, the human circadian period averaged 24.18 hours in both young and older adults. Because it is slightly longer than 24 hours, the clock drifts later unless it is reset, mainly by light.
Q: Where is the body's master clock? A: The master clock is the suprachiasmatic nucleus (SCN), a small paired structure in the hypothalamus. It receives light information from specialized retinal ganglion cells that contain melanopsin and does not need rods or cones to do so. Most other tissues have their own clocks, which the SCN normally keeps in phase.
Q: Does blue light from screens really delay sleep? A: Evening use of a bright, light-emitting eReader delayed the circadian clock and suppressed melatonin compared with reading a printed book, in a 2015 laboratory study. How much a dimmer screen matters in real life is less certain, because the effect depends on brightness, distance, and duration.
Q: What is the difference between a lark and an owl? A: These are informal names for morning and evening chronotypes. A Finnish study of 8,753 twin pairs estimated heritability of diurnal type at about 49.7 percent, and a large genome-wide study identified 351 loci associated with morningness. Environment and age also contribute.
Q: Why is night shift work considered a probable carcinogen? A: IARC classified night shift work as probably carcinogenic to humans (Group 2A) based on limited evidence in people, evidence in experimental animals, and supporting mechanistic evidence. Proposed mechanisms include circadian disruption and light at night, but classification does not describe the size of any individual's risk.
Q: Does meal timing matter for health? A: It may. Mouse and human laboratory studies show that eating at the wrong biological time can shift peripheral clocks and worsen glucose responses. Clinical trials of time-restricted eating have been mixed, so it remains an active research question and not established advice.
Q: Can you change your chronotype? A: Partly. The clock responds to the timing of light, and phase response curve research shows that light can shift it earlier or later by hours in the lab. But genetics sets a baseline, and extreme patterns that interfere with daily life are worth discussing with a clinician.
The takeaway
The most useful idea in this field is that timing is a variable in its own right. The same light, the same meal, or the same dose of melatonin can push the clock in opposite directions depending on when it arrives, and the same schedule can be easy for one person and punishing for another because of a handful of gene variants. The clock itself is a feedback loop that takes about a day to turn over, kept in time by a crowd of neurons and repeated in nearly every tissue. Where studies show real harm, as in forced misalignment, the effects are measurable in days. Where the claims are about lifestyle fixes, the trials are smaller and mixed, and the honest answer is that we are still learning how much timing matters outside the extremes.
Sources
- Czeisler CA et al. (1999). Stability, precision, and near-24-hour period of the human circadian pacemaker. Science.
- Berson DM, Dunn FA, Takao M (2002). Phototransduction by retinal ganglion cells that set the circadian clock. Science.
- Takahashi JS (2017). Transcriptional architecture of the mammalian circadian clock. Nature Reviews Genetics.
- Khalsa SB et al. (2003). A phase response curve to single bright light pulses in human subjects. Journal of Physiology.
- Scheer FA et al. (2009). Adverse metabolic and cardiovascular consequences of circadian misalignment. PNAS.
- Koskenvuo M et al. (2007). Heritability of diurnal type: a nationwide study of 8753 adult twin pairs. Journal of Sleep Research.
- Lowe DA et al. (2020). Effects of time-restricted eating on weight loss and other metabolic parameters (TREAT trial). JAMA Internal Medicine.
- IARC Working Group (2020). Night shift work, IARC Monographs Volume 124.
Related reading:
Follow new articles
Add SciHub101 to your feed reader to see new science and technology stories.
About the Author
SciHub101 is an independent science and technology publication founded and edited by Buğra Sözeri, a software engineer and app developer based in Türkiye. AI tools assist with drafting and research. Source documentation in the existing archive is uneven, and articles are being reviewed and corrected. The Editorial Team byline identifies editorial responsibility; it does not certify specialist or independent human review. Where we compare products, anything built by our own team is labelled as such — see the editorial guidelines and About page for how corrections and disclosures work.
Related Articles

Why We Cry — The Biology of Tears
Three types of tears, the chemistry of emotional crying, and why humans are the only species that weeps from feelings. The science of tears explained.

The Skin Microbiome — Your Body's Outer Ecosystem
Your skin hosts ~1,000 bacterial species, fungi, viruses, and mites. How the skin microbiome shapes acne, eczema, odor, and immunity.

Why We Get Jet Lag — Circadian Disruption
Why does jet lag happen and why is going east worse? The circadian science of body clocks, light, melatonin, and evidence-based recovery strategies.

Postprandial Sleepiness — Why Eating Makes Tired
Why do meals make you sleepy? The science of food coma — insulin, orexin neurons, the circadian dip, and the tryptophan myth.

Sleep and Health: How Much You Need and When to Seek Help
Understand adult sleep recommendations, evening exercise evidence, and why persistent insomnia deserves clinical assessment and structured treatment.

Why Do We Sleep: Stages, the Glymphatic System, and Memory
Sleep is not the brain going quiet — it is the brain running a completely different operating mode. During those hours, it flushes toxic proteins, consolidates memories, and resets emotional circuits. Here is what the science actually shows.