The slump that comes with or without lunch
Everybody blames lunch for the afternoon fog. The evidence says lunch is not the whole story.
Here is a scene you probably know. It is about two or three in the afternoon. You ate lunch an hour or two ago, and now the words on your screen slide around, small decisions feel heavy, and the clock seems to be moving through syrup. You decide, as nearly everyone does, that lunch is to blame. A sandwich, a bowl of pasta, a few biscuits. Surely food made you sleepy. If you skip lunch tomorrow, the fog will lift.
Now the puzzle. Researchers have tested that idea, and it does not hold up as the whole story. In a review of the evidence on what is called the post lunch dip, the sleep researcher Timothy Monk wrote in 2005 that the dip can occur even when a person has not eaten and does not know the time of day. In a 1996 study of eight sprinters, who stayed on a fixed meal schedule for five days, a dip in performance showed up at 3 in the afternoon on one of the days even though lunch was not served until 4. The authors concluded that the dip is not directly tied to when lunch happens, although lunch might make it worse.
Think about what that means. If the dip comes at about the same time with food, with a late lunch, or with liquid snacks in place of lunch, then lunch cannot be the main cause. Something else is making the early afternoon sleepier than the late morning, and food is at most an amplifier. The question for this lesson is what that something is, how much lunch adds, and what, if anything, helps.
This is a lesson about a common experience and what research says about it. It is not medical advice.
Look at the curve above, and remember it is a schematic. Sleep research describes a pattern in which alertness and body temperature peak twice a day. There is a deep low before dawn, a flat spot or shallow dip in the early afternoon, and a higher peak in the early evening. The early afternoon flat spot is what people call the post lunch dip, though lunch did not create the shape.
You can run a rough check on yourself, though it is not a proper experiment. For a week or two, note how sleepy you feel at around 2 or 3 in the afternoon on days when you ate a big lunch, a small lunch, or a late one. Real studies control the sleep people got, the task they did and the time they ate, and you will not. But if the dip shows up every day regardless of what you ate, that matches what the research describes. If it only appears after a particular kind of meal, that is worth noticing too.
There is even an old human habit built around it. The Spanish word siesta comes from the Latin phrase for the sixth hour of the day, counting from dawn, which puts it around noon. The custom of a midday rest goes back to Roman times along the Mediterranean coast. English language writing often mixes up the siesta with the long lunch break of Spanish and Portuguese working life, but the nap and the meal are different things that happen to sit near each other in the day.
So lunch gets the blame for a dip that was already on its way. That is easy to understand. We notice the sleepiness after eating, we remember what we ate, and we draw a line between the two. It is the same logic as noticing that every time you carry an umbrella, it rains. The two things occur together, but one did not cause the other.
In the next six parts we will look at the two clocks inside you, what lunch actually does, why the night before matters, what coffee and naps can and cannot do, how much we really know, and how to plan an afternoon around what the evidence supports. We will say where the evidence is thin, and in this field it often is.
Lunch gets the blame for a dip that was already on its way.
The early afternoon dip in alertness is real, and it appears even when people skip lunch. Food can make it worse, but it is not the root cause.
Two clocks running against you
Inside you are two separate timers. The afternoon is where they cross.
To see why the afternoon is a bad time to be alert, you need to know about two systems in your body that control sleepiness. A sleep researcher, Alexander Borbely, proposed a way to think about them in 1982, called the two process model, and a version of it is still used today. A review by Borbely and colleagues in 2016 says it successfully simulates the timing and intensity of sleep in many different experiments.
The first process is sleep pressure. It rises during every hour you are awake and falls while you sleep. Think of it as a tank filling with sleepiness. The longer you have been up, the fuller the tank, and the more your brain wants to sleep. Overnight the tank drains, so that in the morning you start near empty. Sleep scientists call this process S, for sleep.
The second process is your body clock. It is a roughly 24 hour rhythm, driven by a tiny region of the brain, that raises and lowers the signal that keeps you awake. This process is called C, for circadian. It does not simply rise and fall once. Evidence from many studies shows a two peak pattern, with a second smaller dip in the early afternoon. It is sometimes called the circasemidian rhythm, which means about half a day.
Now put the two together. During the day, sleep pressure climbs steadily. The alerting signal from the body clock is strong in the late morning, and then it weakens for a couple of hours in the early afternoon, before it rises again toward evening. The early afternoon is the point where the pressure to sleep is already substantial and the alerting signal briefly dips. That is the moment the slump lands.
What is sleep pressure made of? One leading idea involves adenosine, a chemical that is produced as brain cells work. In animal studies, adenosine rises in parts of the brain, notably the basal forebrain, during prolonged wakefulness, and blocking the rise reduces the recovery sleep that follows. Caffeine works by blocking the receptors that adenosine acts on. That is the standard explanation for why a cup of coffee makes you feel less sleepy, and we will come back to it.
We should hold this at the right level of confidence. The adenosine account has strong support from animal experiments and from studies of caffeine, but reviews note that measurements of adenosine in the human brain are rare, and that questions remain about where in the brain it acts and through which receptors. It is a good working picture and not a finished one.
There is also a caveat about the dip itself. The circasemidian rhythm has not been found in the cells of the main body clock, the suprachiasmatic nucleus. It may be a kind of echo of the 24 hour rhythm, a harmonic, and not a separate clock of its own. What is well documented is the two peak pattern in many measures of performance and physiology, including in industrial and transport settings where errors tend to cluster twice a day.
This pattern matters enough that it is used in practice. Fatigue modelling tools, such as the Fatigue Avoidance Scheduling Tool used to plan work for people in safety critical jobs, include a 12 hour component in how they predict performance. The reason is simple. If errors really do cluster twice a day, then a schedule that ignores the early afternoon is ignoring a known weak point.
Notice something useful. Because the dip belongs to these two systems, it is not a sign of weakness or of eating badly. It is a feature of how people are built, and it shows up in people who are well rested too. That changes how you can think about it. You do not need to fight a character flaw. You can plan around a known pattern.
The slump is where rising sleep pressure meets a dip in the body clock.
Sleepiness comes from two interacting systems: sleep pressure that builds the longer you are awake, and a body clock whose alerting signal dips in the early afternoon. The dip is part of normal human design.
Does lunch matter at all?
If lunch is not the cause, is it innocent? The evidence says it can make things worse.
It would be easy to swing too far the other way and say food is irrelevant. It is not. Even researchers who stress the body clock find that lunch has an effect, and they have measured it. Monk's review says that the dip is certainly made worse by a high carbohydrate lunch. An earlier finding, by Smith and Miles in 1986 and cited in later papers, reached a similar conclusion: the dip does not depend on food, but performance falls further when a meal is eaten.
The cleanest recent test of meal size is a driving study by Louise Reyner, James Horne and colleagues in 2012. They put drivers in a full size interactive car simulator for two hours in the afternoon after one of two palatable lunches. One was light, about 305 calories. The other was heavy, about 922 calories, with roughly three times the fat and twice the carbohydrate. The researchers measured how often the cars drifted out of their lanes in sleepiness related incidents, how sleepy the drivers said they felt, and brain activity linked to sleepiness.
The heavy lunch made things worse. The drivers who had the larger meal had more lane drifting and more of the brain activity linked to sleepiness, and there was a trend toward higher self rated sleepiness as well. But look at the other half of the result. Sleepiness grew over the drive after both lunches. The light lunch did not protect the drivers from the slump. It only made it a little smaller.
Why would a bigger meal add to sleepiness? Here the honest answer is that nobody knows for sure. The reference summary of the subject says the exact cause is unknown and lists hypotheses, and it adds that very few of them have been tested directly. One idea is that a meal shifts the nervous system toward its resting mode, and that a bigger meal shifts it further. Another involves insulin and the amino acid tryptophan, which is a building block of the brain chemical serotonin. A third involves glucose and the brain cells that help keep you awake. Each idea has some support, and none is settled.
It is worth saying that food does appear to affect sleep in other settings. One study, with the title High glycemic index carbohydrate meals shorten sleep onset, looked at how the type of meal eaten in the evening related to how quickly people fell asleep. That is a different question from the afternoon dip, and it concerns an evening meal and bedtime, so it does not show that blood sugar explains the slump. But it is a reminder that what we eat and when we sleep are connected in ways researchers are still working out.
This is where the popular claim about blood sugar belongs. The idea that it is the shape of your blood sugar curve that makes you crash is not an absurd one, and glucose does appear in some of the hypotheses. But as the main explanation of the afternoon slump, it does not fit the evidence, because the dip appears without a meal and at times that do not follow the meal. Blood sugar may play a role in how much lunch deepens a dip. It does not explain why the dip is there.
What can we say with confidence? A very large lunch, in the conditions tested, made sleepiness worse on a task that was monotonous. That is consistent across several studies. The size of the effect for an ordinary desk worker, and the contribution of fat compared with carbohydrate compared with sheer quantity, are less certain. The driving study was a simulator, with a particular group of drivers, and one two hour drive.
So lunch is an amplifier, not the source. You cannot remove the dip by skipping lunch, and you can make it noticeably worse by eating a very large one. That points toward a modest, middle position that is easy to remember: lunch is allowed, and a lighter one is likely to help a little.
Lunch can deepen the dip. It does not dig it.
A very large lunch made sleepiness worse in a driving simulator, but sleepiness grew after the light lunch too. Lunch amplifies the afternoon dip. The exact way it does so is still unknown.
The night before
The same afternoon feels very different depending on what happened at 3 in the morning.
Here is something every tired person already suspects. Some afternoons the slump is a gentle dip, and some afternoons it is a wall. What changes? Often, it is not what you ate. It is how you slept. Reyner and Horne, in the same line of research, describe the post lunch dip as a twice a day phenomenon largely unrelated to lunch, and made worse by a disturbed night of sleep before.
The mechanism follows from the two process model. Sleep pressure rises with every hour awake and drains during sleep. If you sleep less than you need, the tank does not fully empty. You start the day with some pressure already in it. Every hour then adds to a higher baseline, and when the body clock's alerting signal dips in the afternoon, there is less room between you and sleep.
The driving study is a useful example. The participants in it had an ordinary amount of lost sleep, nothing extreme, and the afternoon drive still grew steadily sleepier. The authors stress that the study was realistic: typical lunches, and a typical level of prior sleep loss, and that a heavy lunch added to the sleepiness that was already there. In other words, the underlying sleepiness is part of the story before the first bite of the sandwich.
There is another factor, which is who you are. People differ in whether they are naturally morning people or evening people, which researchers call chronotype. Monk's review says the dip may be more likely to occur in people who are extreme morning types. That fits the picture: someone whose body clock runs early reaches the early afternoon further along in their own day, closer to the point where their alerting signal sags. But note the hedge in his wording: may be more likely.
Notice too what kind of task the drivers faced. The study was about monotonous driving, a long, undemanding road. It is often in dull, repetitive situations that sleepiness shows itself most. A more engaging task may hide the dip for a while, which could be one reason some people feel it in a long meeting and not on a busy day. That is a sensible guess, but it is not something the studies here tested directly.
It is worth knowing that sleep need is not a matter of willpower. If you are chronically short of sleep, then the afternoon slump is probably a symptom of a larger problem, and no trick with lunch will fix it. This is where the simplest advice is also the most likely to work. Protecting the night does more for the afternoon than anything you do at 2 pm.
It is also where we should add a safety note. A slump that is extreme, that comes with falling asleep at dangerous moments such as when driving, or that does not respond to getting enough sleep, can be a sign of a sleep disorder or another medical problem. This lesson cannot tell you which, and the right person to ask is a doctor. Drowsiness behind the wheel in particular is a serious risk.
The point of the diagram is not the exact heights of the lines, which are schematic. It is the shape. A short night lifts the whole curve. The afternoon dip is still there, and it now lands on a higher baseline.
A short night lifts the whole line, and the afternoon dip lands on top of it.
Sleep pressure starts the day higher after a short night, so the early afternoon dip lands on a bigger baseline. Protecting the night is likely to help the afternoon more than any change at lunch.
The coffee and the nap
Two popular fixes are both rooted in the same chemistry. How well do they work, and what do they cost?
Caffeine is the world's most common stimulant, and the reason it works is the story from Part 2. Caffeine looks enough like adenosine to bind to its receptors and block them. With the receptors blocked, the sleepiness signal does not get through as well. It does not remove the pressure, because the adenosine is still there. It hides the message. That is why a coffee can feel wonderful at 2 pm, and why the tiredness can come back when the caffeine wears off.
Here is the cost. Caffeine stays in the body a long time. Its elimination half life in healthy adults is usually quoted as about 3 to 7 hours, with a wider range between individuals, and it is longer in pregnancy and shorter in smokers. A half life of 5 hours means that after 5 hours, half of the caffeine is still there, and after 10, a quarter. Look at the chart. A cup at 2 pm still leaves about 29 percent in the body at 11 pm, if the half life is 5 hours.
Does that matter for sleep? A study led by Christopher Drake and published in 2013 gave people 400 milligrams of caffeine, roughly the amount in several cups of coffee, either at bedtime, 3 hours before, or 6 hours before. The authors concluded that even caffeine taken 6 hours before bed had important disruptive effects on sleep. So an afternoon coffee may help the afternoon and cost some of the night, which then raises tomorrow's sleep pressure. That loop is worth knowing about.
The other classic fix is the nap, and here the evidence is attractive. In the 1990s Mark Rosekind and colleagues at NASA studied crews on long haul flights. Twelve pilots were given a planned 40 minute rest opportunity during the quiet cruise phase. On average they fell asleep in under 6 minutes and slept for about 26 minutes. The study reported up to a 54 percent improvement in a physiological measure of alertness and up to a 34 percent improvement in performance, compared with pilots who did not nap. The wording matters: up to.
Short is the key idea. A nap of about 20 minutes or less ends before the body falls into deep slow wave sleep, which is the stage that is hardest to wake from. Waking out of deep sleep brings on sleep inertia, a groggy, slowed state that is typically felt for 15 to 30 minutes after waking and in bad cases can take much longer. That is why very short naps are recommended for a pick me up, and why a long nap that ends in deep sleep can leave you worse for a while.
Sleep inertia deserves a safety note of its own. It can impair decision making and safety critical tasks soon after waking, and in the majority of cases it lasts about 15 to 30 minutes, though it can last longer. That is why a nap is a poor idea right before something where a slow mind is dangerous, such as driving, and why even a short nap is best taken with a little time to wake properly afterwards.
Be careful with the numbers. The NASA study was on pilots in a cockpit during a flight, with small groups, in conditions that are not like an office. The figures are real, and they come from the report itself, but they are best treated as evidence that a short nap can help alertness, not as a promise of a particular percentage for you. Many popular pages repeat the 54 and 34 as if they were guaranteed.
So both fixes work for the same reason, and both have a cost. Coffee hides the sleepiness signal and can disturb tonight's sleep. A nap reduces sleep pressure, but done wrongly leaves you groggy, and a late or long one can make falling asleep harder at night. Neither is magic, and neither is a replacement for sleep.
Coffee hides the sleepiness signal. A nap drains a little of the pressure behind it.
Caffeine blocks the receptors for adenosine and can still disturb sleep many hours later. A nap of about 20 minutes or less can restore alertness without deep sleep, but the best known figures come from small studies of pilots, and the effect on you is uncertain.
What we do not know
This field runs on small studies. It helps to know exactly how small, and what that means.
It is time for a step back. A lesson like this can sound more certain than the evidence allows, so here is an honest look at how much we actually know, and how we know it. The studies we have been quoting are real and carefully done. But look at how many people took part in some of them.
The 1996 sprinters study had eight people. A 1996 constant routine study, in which people were kept in controlled conditions, with lunch replaced by hourly liquid food, and which found a dip in a subset of participants, had a dozen. The NASA nap study gave twelve pilots a rest opportunity. A 2015 experiment on light and the post lunch dip, by Baek and Min, tested twenty healthy adults in four lighting conditions. Small studies are normal in sleep science, because the experiments are expensive and demanding. But they limit what we can say.
What do small studies allow? They allow consistent hints. When several small studies of different kinds, run by different people, agree, such as the dip appearing without a meal, and a heavy meal making it worse, our confidence rises. They do not allow precise numbers. When a popular article says that a nap boosts performance by exactly 34 percent, the honest version is that one small study reported up to 34 percent in one setting.
Take the light study as an example. The researchers found that blue enriched light during the post lunch dip reduced a brain wave pattern linked with drowsiness and improved performance on a task, and concluded that it might help counter the dip. That is interesting. But it was twenty people, in a laboratory, measured over a short period, and I would not stake my afternoon on it. It is a lead worth following, not a prescription.
Here is how researchers try to separate the body clock from the effects of eating. In a constant routine study, people stay awake in controlled conditions, with the same dim light and posture hour after hour, and instead of a lunch they receive small liquid food supplements at regular intervals. If a dip still appears in the early afternoon, it cannot be blamed on a lunch. In the 1996 study of this kind, a dip still appeared in some of the participants. It is a clever design, and it also shows why the studies are small, since keeping people awake under constant conditions for a day is hard to do with many people at once.
There are other gaps. As far as I could find, there is no recent large review that directly compares meals with no meals in a controlled way across many people. We are not certain how much of the dip belongs to a separate half day rhythm and how much is a harmonic of the main 24 hour clock. We do not know precisely how a meal deepens sleepiness. And individual differences are large: some people barely notice a dip, and others struggle.
This is not unusual for science that touches daily life, and it is not a reason to dismiss the field. The broad picture, a body clock and sleep pressure that make the early afternoon a low point, is supported by many kinds of evidence, including studies in people who ate no lunch. What we should be cautious about is the details, and especially any claim that sounds precise, confident and marketable.
A good habit when you read about sleep, food or energy is to ask four questions. How many people were studied? Was it a laboratory, a simulator or real life? Does the claim say up to, or may, or does it sound certain? And who is selling something? A claim that survives those four questions is more likely to be sound.
That same habit applies to this lesson. We have quoted the sizes and the hedges as faithfully as we can, and where the sources disagree or are thin, we have said so. If something here sounds more certain than the evidence we have shown you, hold it loosely.
A small study can tell you which way the wind blows. It cannot tell you how fast.
The research on the afternoon dip, naps, caffeine and light mostly rests on small studies. They agree on the broad picture, and they leave the exact sizes of the effects uncertain.
Designing an afternoon
Put the pieces together and plan around the pattern instead of fighting it.
Now we can bring everything together. Here is the whole mechanism in a breath, which is the test of whether you understand it. Two systems shape your sleepiness. Sleep pressure builds the longer you have been awake, and a body clock lifts your alertness in the late morning and early evening but dips in the early afternoon. The afternoon is where building pressure meets the dip. A short night raises the baseline, and a very large lunch can deepen the dip, but neither creates it.
If that is right, then the advice that follows is not complicated, and most of it is about timing and sleep, not about willpower. Protect your night, since that lowers the baseline you meet the afternoon on. Keep lunch moderate rather than very large, since a heavy meal added to sleepiness in the studies. Expect the dip, and do not read it as a personal failing.
Beyond that, you can plan around the shape. If you can, put the demanding thinking in the late morning and the early evening, when your alerting signal is higher, and leave routine or social tasks for the dip. If you can nap, keep it to about 20 minutes or less, and allow 15 to 30 minutes of grogginess before you need to be sharp. If you use caffeine, remember it lasts, and an earlier cup costs less of your night than a later one.
Light is a reasonable thing to try. There is a small study suggesting blue enriched light helped during the dip, and daylight and a short break are cheap. But this is a hint, not a proven fix, and whether it works for you is something you can only find out by noticing.
The example day in the chart is exactly that, an example. It starts with light after waking, because the body clock responds to light. It has a moderate lunch, a nap of up to 20 minutes if there is a chance of one, and it respects the second peak in the early evening. It ends with a bedtime that leaves room for a full night, since the night shapes the next afternoon.
A last practical idea, offered as a suggestion and not as a prescription: pay attention for a couple of weeks. Rate your sleepiness a few times a day, and note when you slept, what you ate and when you had caffeine. Look for patterns. Do your worst afternoons follow short nights? Does a lighter lunch seem to help? You are an experiment of one, with all the limits that implies, but patterns across weeks are more informative than a single bad day, and they will tell you which of these ideas matters most for you.
We should also be clear about who this is not for. If your sleepiness is extreme, if you fall asleep at moments when it is dangerous, or if enough sleep does not help, speak to a doctor. Sleep disorders are common and treatable, and an afternoon slump that is really a symptom of one deserves proper attention. This lesson describes research on a common experience and is not medical advice.
There is one more thing to take from the lesson, apart from the practical advice. When you feel something and there is an obvious cause, such as sleepiness after lunch, it is worth asking whether the obvious cause is the real one. Here it turned out to be a partial culprit that took the blame for a pattern that was on its way anyway. That habit, noticing the correlation and testing the cause, is useful far beyond lunch.
Next time the fog rolls in at 2 pm, you can name what is happening. Sleep pressure is high, the body clock is dipping, and the lunch may be adding a little. It is not a flaw. It is a pattern, and now that you know its shape, you can plan around it.
You cannot delete the dip, but you can stop being surprised by it.
Plan around the pattern: protect the night, keep lunch moderate, put hard thinking at the body clock's high points, nap for about 20 minutes or less if you can, and mind how long caffeine lasts. See a doctor if sleepiness is extreme.
- Why the afternoon dip appears with or without lunch, and where it comes from.
- How sleep pressure and the body clock cross in the early afternoon.
- What a heavy lunch adds, and why nobody can yet say exactly how.
- Why a short night makes the same afternoon harder.
- How caffeine and a short nap work, and what each costs.
- Why the research is small, which claims to hold loosely, and when to see a doctor.
Where these facts come from. Timothy Monk, The post lunch dip in performance (Clinics in Sports Medicine, 2005), as summarised in its abstract. Javierre and colleagues, a 1996 study of post lunch dip in sprinters. Louise Reyner, James Horne and colleagues, post lunch sleepiness in prolonged driving and the effect of meal size (Physiology and Behavior, 2012). Alexander Borbely and colleagues on the two process model, including a 2016 reappraisal. Reviews of adenosine, caffeine and sleep homeostasis. Christopher Drake and colleagues, caffeine taken 0, 3 or 6 hours before bed (Journal of Clinical Sleep Medicine, 2013). Mark Rosekind and colleagues, strategic naps in operational settings (Journal of Sleep Research, 1995). Baek and Min, blue light and the post lunch dip (Ergonomics, 2015). Reference summaries of the post meal sleepiness, the circasemidian rhythm, caffeine, naps and sleep inertia. I have read the abstracts and summaries, not the full papers.
Where the evidence is limited. Most studies cited here are small (8 to 20 people in several). The cause of post meal sleepiness is described as unknown, with several untested hypotheses. Whether the afternoon dip is a separate half day rhythm or a harmonic of the daily clock is unresolved. The nap figures are quoted as up to 54 percent and up to 34 percent, from pilots on long flights.
This lesson summarises research on a common experience. It is not medical advice. If sleepiness is extreme or dangerous, see a doctor.