The dip has a name
Sleep researchers call it the post-lunch dip. Timothy Monk of the University of Pittsburgh, reviewing the evidence in 2005, described it as a real fall in performance in the mid-afternoon, tied to a rise in how easily people fall asleep at that time of day. It does not affect every person or every task. It is common enough to have its own literature.
Lunch is only an accomplice
The meal is the obvious suspect, and the evidence only partly convicts it. In a 1996 study, Monk and colleagues replaced lunch with small liquid meals served every hour, so there was no midday plate to blame. Volunteers did a monotonous 25- to 30-minute vigilance task every two hours. Five of the twelve still showed a clear afternoon dip.
Those five had something else in common. Their body temperature, which normally climbs through the day, stayed flat between 10am and 3pm. The researchers linked this to a roughly 12-hour component of the circadian rhythm, a smaller second wave riding on the daily one, and concluded that the dip is mainly internal and varies from person to person.
Monk’s later review adds two details. A high-carbohydrate lunch makes the dip worse, and strong morning types may be more prone to it. So the sandwich is not innocent. It just isn’t working alone.
The coffee question
The usual response is caffeine, which brings a timing problem. The European Food Safety Authority puts caffeine’s average half-life in healthy adults at about four hours, with a range of two to eight. A 3pm cup is therefore still partly in circulation at 7pm, and for slow metabolisers well after that.
In a 2013 trial by researchers at the Henry Ford Hospital sleep centre, volunteers took a 400 mg dose of caffeine at bedtime, three hours before it or six hours before it, and each was compared with a placebo. Sleep was measured at home with a portable monitor. All three times disrupted sleep. The authors suggested avoiding substantial caffeine for at least six hours before bed. EFSA separately notes that a single 100 mg dose close to bedtime may affect sleep in some adults.
What the evidence does not say
These are small studies and averages. They do not show that everyone slumps at exactly 3pm, and they do not set one caffeine cut-off that suits every person and dose. Persistent daytime sleepiness can have medical causes that this research does not address.
What the studies do support is narrower and more useful. An afternoon dip in alertness is ordinary timekeeping by the body, it tends to pass, and it is not evidence of a lack of discipline. Planning undemanding work for that window is a reasonable response to a predictable event.