I bet you know the third hour of a long planning session quite well. You are trying to hold four or five figures in your head to compare two options and one of them keeps slipping away, and you switch between the spreadsheet and your inbox more often than the work actually needs. Generally speaking, most of us put this down to the meeting itself, to the coffee wearing off, or to a bit of missing discipline. Part of it may be more mechanical than that, and a review published in July in the International Journal of Behavioral Nutrition and Physical Activity gives the clearest numbers so far on how much a short break changes it, and on how little we can yet say for sure. The team, led by Mingnan Zhuang, pooled 21 randomised crossover trials with 433 people in total, aged 13 to 69. Crossover means the same people were tested under both conditions on different days: one session of uninterrupted sitting, and one where the sitting was broken up by standing or light movement. Walking was used in 14 of the 21 trials, the most common rhythm was one break every 30 minutes, and individual breaks lasted anywhere from 30 seconds to 15 minutes. Out of the five areas of thinking they looked at, two showed a change. Executive function, meaning switching between tasks, holding back the first impulse and keeping a goal in view, improved by 0.18 on the standard effect-size scale, which sits just under the 0.2 that statisticians usually call small. Short-term memory, holding and updating information while you work with it, improved by 0.37, somewhere between small and medium. Interestingly enough, attention showed no clear change at all (minus 0.08, from only 5 studies), which is a bit awkward given how many of us take a walk precisely to refocus, and processing speed showed nothing clear either. Don't get me wrong, this is not a strong finding, and the authors say so quite openly. Using the standard GRADE system they rated the certainty of both positive results as low. The trials were small, between 9 and 67 people each, most of them carried some risk of bias, results varied a fair amount from study to study, and the executive function result shows signs that small positive studies may be over-represented, which can mean the true effect is a bit smaller than 0.18. Almost all of it was measured in a lab within a single day, so it says nothing yet about whether doing this for a month changes anything, and the authors are clear that none of this replaces regular exercise. What it does give us is a small, plausible effect at a price close to zero, and that is reasonable to test on yourself.