Is the 90-minute work cycle real? The evidence, briefly
By Luis4 min read
The 90-minute work cycle is an extrapolation, not an established finding. The idea comes from Nathaniel Kleitman's "basic rest-activity cycle", a proposed rhythm of roughly 90 minutes that was linked to the stages of sleep. Whether the same fixed rhythm runs through the working day has been disputed since at least the 1970s. We found no controlled trial showing that 90-minute blocks beat other lengths for knowledge work.
This article is written by Luis, who makes Kiri, which includes a 90-minute timer. It sells the same whether you use 25 minutes or 90.
Where the idea comes from
Kleitman, whose laboratory helped discover REM sleep, proposed that a rest-activity rhythm continues through the day as well as the night, and he revisited the idea in a 1982 paper. Popular writing later turned that into advice to work for 90 minutes and then rest.
The step from sleep to office work is the weak one. Kripke and colleagues published a paper in 1977 with the blunt title "There's no basic rest-activity cycle". Later work does describe rhythms in waking alertness, but a regular rhythm in the body is different from a rule about when you should stop writing.
Sources: Kleitman (1982), "Basic rest-activity cycle: 22 years later", Sleep, 5(4). Kripke, Mullaney, Wyborney & Messin (1977), Proceedings of the Second International Symposium on Ambulatory Monitoring, seen only as cited in this summary.What 90 minutes is still good for
A long block can be useful for practical reasons. Writing and coding often need time to load the problem into your head, and a 25-minute timer may ring just as that happens. That is a judgement from experience and not a research result. Try one 90-minute block in the morning, take at least 15 minutes off, and notice whether the last half hour was any good. If it wasn't, try 50 minutes. For which length suits which kind of study, see how long a study session should be. How many hours you can focus in a day covers the wider limit.
Where Kiri fits, and who should not use it
The 90-minute timer on this page runs in a browser tab on a laptop or desktop, with sound in the same tab. The timer and brown noise work in full with no account. Other sounds are 100-second samples on repeat, a free account also plays rain in full, and all ten soundscapes cost €39 once at the Founding price, then €49. There is no phone app, site blocking or native Mac app. If you want to be locked out of distractions for 90 minutes, a blocker such as Freedom does a job Kiri does not do.
Questions
Do humans have a 90-minute focus cycle?
It is not established. Roughly 90-minute cycles are known in sleep. Kleitman proposed that they continue through the day, but that claim has been disputed and it does not show that work should be split that way.
Is working for 90 minutes better than 25?
We found no controlled trial showing that. Longer blocks can suit work that takes a while to get into, while shorter ones can suit tasks that are hard to start.
How long should a break be after 90 minutes?
There is no tested figure. Fifteen to twenty minutes is a common choice, and it is a convention more than a finding.