By the fourth hour of a stretch at the desk, most people don't step away. They push through to the next natural stopping point, a meeting, lunch, the end of the task, whichever comes first.
There's a rule that says they're entitled to something before then. It doesn't say what that something actually does.
The Law Guarantees One
Under Regulation 12 of the Working Time Regulations 1998, once a worker's daily working time passes six hours, they're entitled to a rest break. The rule is specific about the shape of it: an uninterrupted period of not less than 20 minutes, and the worker is entitled to spend it away from their workstation if they have one. One break. Twenty minutes. That's the entitlement, in full.
Unless the Entitlement Was Never Theirs
Regulation 20 removes this break, along with the daily and weekly rest periods, for a worker whose "working time is not measured or predetermined or can be determined by the worker himself," naming "managing executives or other persons with autonomous decision-taking powers" directly. It's a facts-of-the-job test, not a job-title one, but someone who structures their own day with nobody checking when they stepped away fits that description more easily than someone on a fixed rota. For that reader, there was never a 20-minute break to not take. There was just a day with no rule attached to it either way.
What the Best Pooled Evidence Actually Finds
A 2022 systematic review and meta-analysis in PLOS ONE pooled 22 independent study samples on short breaks of ten minutes or less, 2,335 people in total, split roughly evenly between employees in real workplaces and students in lab settings. The result splits in two. Short breaks produced a small but statistically significant improvement in how depleted and how energised people reported feeling afterward. On actual task performance, the improvement was small and not statistically significant overall, and when the researchers separated task types, the pattern held only for less demanding creative and clerical work. For cognitively demanding tasks specifically, there was no significant improvement at all.
Worth stating plainly rather than smoothing over: the same analysis found that longer breaks, within the ten-minutes-or-less range it studied, were associated with bigger performance gains. It didn't test anything longer than that, so it has nothing to say about whether the law's full 20 minutes behaves differently again. That's an open question, not a resolved one.
A Different Mechanism, One Study
A separate, single 2011 laboratory study found something else worth noting, through a different mechanism entirely. Eighty-four people worked through a repetitive computer task for about fifty minutes. Most groups' performance declined over that stretch. One group didn't, the one given two brief, unrelated interruptions to respond to during the task itself, rather than one longer break at a fixed point. The researchers' account is that the interruptions, not the rest, were doing the work, by pulling attention briefly away from the goal and back again. This wasn't one 20-minute break, it was two short diversions scattered through the task. Whether that result would hold at a desk, over a full working day, with an actual job's worth of starting and stopping, wasn't tested here, and the study's own population wasn't specified clearly enough in the sources available to say more about who the 84 people were.
What This Doesn't Tell Us
It doesn't tell us that the law's 20-minute break is too short, too long, or wrongly timed. It doesn't tell us that skipping it costs anything measurable, or that taking it restores anything beyond how someone feels in the moment, which is itself a real and separately useful thing, just not the same claim as restored performance. It doesn't tell us whether the autonomous-worker exemption actually covers any specific reader; that's a question about their own job, not one this piece can settle. And it doesn't tell us anything about hydration, caffeine, or what any drink does during a break. Neither source measured anything of the kind.
What it does say, read together, is that the clean version of the story, step away, come back sharper, isn't the version the pooled evidence actually supports for demanding work. The feeling of recovery and the measurable fact of it aren't automatically the same thing.
Why 0800Labs Is Interested
We're interested in the parts of the working day that come with a rule attached and an assumption nobody's checked. A legal break exists because someone decided people are entitled to one. What it's actually for, in the sense of what it restores and how much, is a separate, less settled question. It's the same kind of question we asked about nap length and about an unbroken run of meetings, two other recovery mechanisms covering a different stretch of the same working day.
Workday Hydration, our own term for the moment we're building for, isn't a claim about what either source found or what any drink does during a break. It's just the name for the stretch of the day this piece has been describing.
If that's useful, the waitlist is below.