An often quoted statement about the cost of mental context switching is that it can take on average 23 minutes and 15 seconds to regain deep focus on a task after an interruption.
The paper always cited (if at all) is this one from 2008 by Dr. Gloria Mark: https://ics.uci.edu/~gmark/chi08-mark.pdf
- The number “23” appears nowhere in this paper.
- The abstract says that the only observed effects were “more stress, higher frustration, time pressure, and effort.” There were no observed impacts to quality or time.
- N = 48. The task was asking university students to write emails.
The real source is hinted at in this interview from June 2006, which unfortunately doesn’t link to actual study.
- In the interview Dr. Mark says, “The good news is that most interrupted work was resumed on the same day — 81.9 percent — and it was resumed, on average, in 23 minutes and 15 seconds, which I guess is not so long.”
- N = 36; t = 3 days. “In our study, we observed for a half day, then we shadowed 36 managers, financial analysts, software developers, engineers, and project leaders for three days. We literally followed people around all day and timed every event [that happened], to the second.”
- Most of this interview talks about the frequency of interruptions, not the time to recover.
What’s implied is that, once interrupted, you cannot re-enter deep work for some length of time (averaging 23.25 minutes). But if you read the statement Dr. Mark makes next, you learn that it takes that amount of time to return to the original task that was interrupted. “But the bad news is, when you’re interrupted, you don’t immediately go back to the task you were doing before you were interrupted. There are about two intervening tasks before you go back to your original task, so it takes more effort to reorient back to the original task.”
This study took place two years before the often-cited paper, and studies a different subject with a different number of participants. And even so, from the 2008 paper, “A certain amount of interruptions may be tolerable because people can compensate with a higher working speed.”
I learned that this is called a zombie statistic; a statistic cited frequently that won’t die.
AI usage (Claude) for this post: (1) searching for the phenomenon about incorrectly cited statistics, (2) post title. All other research, analysis, and writing was done by me.