Gloria Mark, David Epstein’s explains (in Inside the Box), has been at the forefront of studying what, exactly, workers in a knowledge economy do all day:
Early on, this meant that she and her colleagues were shadowing office workers with stopwatches and logging all of their activity.
[…]
The resulting paper was published in 2004, and for its title she borrowed an emblematic quote from one of the subjects: “Constant, Constant, Multi-tasking Craziness.”
Mark found that the typical office worker switched tasks every three minutes, on average. When her team studied “working spheres”—basically groups of tasks that are connected to the same project—they found that people switched about every twelve minutes and cycled through ten different working spheres per day. Whenever a sphere was interrupted, it took about twenty-five minutes on average to get back to it.
[…]
By 2012, office workers were switching tasks every seventy-five seconds. By 2022, it had stabilized at about every forty-five seconds.
[…]
In simplified form, the shift between tasks actually occurs in two steps: “goal shifting” (switching what you want to do) and then “rule activation” (mentally turning off the rules of one task and on those of another). This takes effort and time, and it is why studies of drivers, in both simulators and the real world, have shown that those having conversations react slowly to hazards, and the impact is the same whether they’re talking on a handheld or hands-free phone.
[…]
Even when people are allowed to switch between tasks at their own discretion, the more they choose to switch, the longer everything takes. As Mark has written: “We find that in real-world work, the more switches in attention a person makes, the lower is their end-of-day assessed productivity.” They also perform worse on important tasks. Multitasking physicians and pilots make more prescribing and in-flight errors, respectively. Famed investor Charlie Munger had it right when he said: “I see these people doing three things at once, and I think, God what a terrible way that is to think.”
[…]
Perhaps the most surprising of Mark’s findings has to do with “self-interruptions.” Self-interruptions are what they sound like: not a phone call or an app notification, but a thought, perhaps about something left undone or what’s happening on social media, that commandeers our attention and leads us to switch goals. Mark found that we are nearly as likely to self-interrupt as we are to be interrupted by some external cue. And here’s the frightening part: We gravitate to a customary level of interruption. If you are disrupted all day, every day by notifications, even if those external triggers magically disappear, you will unconsciously increase your self-interrupting in order to maintain the rhythm of distraction to which you have become accustomed. It is as if we have some sort of internal distraction barometer that, once used to a certain rhythm, will work to maintain it. That is why the mere presence of a smartphone on a desk, or in a pocket—even if it is turned off—has been shown to impair performance on cognitive tests particularly among people who are more phone dependent.
[…]
Background music can have energizing or calming effects that improve performance, but it also has a distracting effect that becomes important when a task is new. Rousing background music has been shown to impair performance among new surgeons learning an unfamiliar task, and a survey of more than two thousand professional software developers found that they tend to turn music off when learning new tools.