The multitasking myth and the cost of switching
Why doing two things at once costs you up to 40 % of your capacity, and what is really happening in your head.
You never do two things at once
The human brain does not process two demanding tasks in parallel. It alternates. Very fast, but it alternates. What you call multitasking is in fact a succession of context switches.
Every switch has a cost: you have to unload the rules of task A, load those of task B, and find where you were. The work of Rubinstein, Meyer and Evans, published in 2001, measured that cost at between 20 and 40 % of total time when the tasks are complex.
In practical terms: a day spent alternating between email, instant messaging and deep work can cost you the equivalent of three hours, without your having done anything wrong.
Attention residue
The most insidious phenomenon was described by Sophie Leroy in 2009: attention residue.
When you move from task A to task B, part of your attention stays hooked on A, especially if A was unfinished. You are physically on B, but running at 70 %.
That is why a thirty-second glance at your messages costs far more than thirty seconds. You open an unresolved thread, you close it, and it keeps running in the background for fifteen minutes.
The operational consequence. What matters is not the number of interruptions, but the number of open loops left unclosed. An interruption you leave with a closed answer costs little. An interruption that leaves a question hanging costs a quarter of an hour.
The two networks competing for your brain
- The executive network: the one that runs directed tasks. Expensive, limited, tires quickly.
- The default mode network: the one that switches on when you are doing nothing directed. It ruminates, it plans, it wanders.
Concentrating is not forcing. It is keeping the executive network active for longer than the default network manages to take back control. The whole module is about making that switch easier: through physiology in chapter 2, conditioning in chapter 3, suggestion in chapter 4, and architecture in chapter 5.
What this module does not claim to do
It will not make you capable of eight straight hours of deep focus: nobody is, and the research on expertise puts the human ceiling at around three to four hours a day, even among the most trained.
It will not replace sleep: a five-hour night cancels most of what you install here. It does not treat an attention disorder, though it usefully accompanies treatment.
What it does, on the other hand: take you from around ten minutes of continuous focus to an hour and a half, and cut your start-up time from twenty minutes to under three.
The real diagnosis
On a sheet of paper, across one working day, mark every time you leave the task in hand. One stroke. Nothing else.
The average observed among knowledge workers is one interruption every three to eleven minutes, and half of them are self-generated: nobody interrupted you, you left on your own.
That is excellent news. Whatever comes from you can be reprogrammed.
What an interruption costs, depending on what it leaves behind
Time lost · Far beyond the length of the interruption
The length of the interruption predicts almost nothing. What predicts the cost is the number of loops left open.
Exercises2
Understanding produces no change. These exercises are what makes the difference: each one has a measurable success criterion.
Counting your drop-offs
Get your starting figure, without which no progress can be measured.
Equipment : A sheet of paper beside you, a pen. No app.
- 1.Pick an ordinary working day, and above all not a good one.
- 2.Every time you leave the task in hand, draw a stroke. Including mental drift, when you wander off into your thoughts.
- 3.Correct nothing, judge nothing. You are measuring, that is all.
- 4.In the evening, count the strokes and divide by the number of hours worked.
Success looks like : You have a number of drop-offs per hour, written down somewhere, dated.
The classic mistake : Holding back because you know you are being watched. A flattered baseline is useless: at day 21 it will make you believe the protocol achieved nothing.
Going further : Count again at day 21, under the same conditions. The average typically goes from 8 to 15 down to 2 to 5.
An inventory of your open loops
See how many unclosed subjects are occupying your attention in the background.
- 1.Set a timer for five minutes. Write down everything crossing your mind that is unfinished: tasks, worries, messages pending, decisions postponed.
- 2.Count the lines.
- 3.Circle the ones where you do not know what the next concrete action is.
Success looks like : You have a number, and you notice that the circled lines, the ones with no defined action, are the most numerous.
The classic mistake : Limiting yourself to work. Personal loops consume exactly the same attentional resource.
This lesson is part of Instant Focus
17 lessons, 4 resources and the complete protocol to install the method in 30 days.