The cost of switching

Open a maths exercise. Work for four minutes. Check a message. Return to the maths. How long does it take to be properly working again?

The honest answer is longer than the interruption. Laboratory work on task switching finds that performance drops for a period after each switch, because part of the mental setup for the first task has to be rebuilt. The effect is small per switch — often a few seconds to a minute — and it is the frequency that does the damage. Twenty switches in an hour can remove a substantial part of that hour without any single interruption feeling significant.

The effect is worse when the two tasks are similar. Switching between writing an essay and reading a message is harder than switching between writing an essay and standing up to stretch, because both use language and compete for the same mental machinery.

There is a further problem that students report more often than researchers measure. After the interruption you do not return to where you were. You return to where you can most easily re-enter, which is usually something easier than the difficult sentence you were stuck on. Over an evening this quietly converts hard work into easy work.

The fix that survives contact with real life is not heroic concentration. It is removing the possibility of the switch for a fixed, short period: phone in another room for twenty-five minutes, then a real break. Twenty-five minutes is chosen not because it is magic but because almost anyone can agree to it.