A person walks into the kitchen, stops in front of the open fridge, and has no idea why they came. Nothing dramatic happened on the way there. No phone rang, nobody called their name. They simply crossed a doorway, and the thought that carried them there quietly dissolved.
The same small blank moment shows up at a desk. A colleague leans in with a ten second question, and the whole afternoon afterward feels shapeless. The interruption was brief. The tiredness that follows is not proportional to it. Something else is doing the draining, and researchers can actually measure it.
The Doorway Moment
Psychologists call the fridge episode the doorway effect.
Photo by Your brain does not store the day as one long ribbon. It chops experience into episodes, and it uses changes in your surroundings to decide where one episode ends and the next begins. Walking from one room into another is a strong signal that something just ended.
When you cross that boundary, the mental workspace holding “go get the scissors” gets filed away with the room you just left. The information is not gone. It is just no longer sitting in front of you, so pulling it back takes an extra beat.
Here is the useful part: this boundary logic is not limited to walls and doors. Your brain treats a shift in task the same way it treats a shift in place. A question about next quarter’s budget, arriving while you are three paragraphs into something else, works as a doorway too.
Not All Interruptions Are Equal
The common assumption is that interruptions are a volume problem: fewer pings, fewer drop-ins, less exhaustion.
Photo by Frequency does matter. More frequent interruptions are linked to greater subjective workload, and the effect is strongest when the underlying task is already complex [NIH].
But frequency is not the whole story, and treating it as the whole story is why so many focus strategies fall short. The evidence keeps pointing to the character of the interruption instead.
One study found that when an interrupting task demanded a different kind of mental effort than the primary task, people reported higher mental workload than when both tasks pulled on the same kind of effort [NIH]. Two interruptions of identical length can cost very different amounts. Being asked to check a figure while you are already working with numbers is cheap. Being asked to mediate a personnel disagreement while you are working with numbers is expensive.
Relevance matters too. Researchers studying what they call illegitimate tasks, meaning interruptions people judge as unnecessary or outside their actual job, found that performance suffers when the disruption feels unwarranted. The same three minutes lands differently depending on whether the request feels like it belongs to you.
Put together, the honest picture looks less like a noise-reduction problem and more like this:
-
Type of demand: does the interruption use the same mental machinery as your current task, or a completely different one
-
Complexity of what you were doing: interruptions during demanding work cost more than the same interruption during routine work
-
Perceived legitimacy: requests that feel irrelevant to your role hurt performance beyond their length
-
Your current fatigue: the same interruption does not always cost the same amount as the day goes on
The Cost Is In The Reload
Here is the mechanism underneath all four of those factors.
When you work on something, your brain holds a temporary setup: the goal, the relevant rules, the half-finished sentence, the three numbers you were comparing. Cognitive psychologists call the price of dismantling that setup and building another one switch cost. Studies on task switching consistently show that people take longer and make more errors when they move between tasks than when they stay with one [NIH].
The tiring part is not the interruption itself. It is the reload afterward: the quiet work of rebuilding everything you were holding before the doorway.
That reload is invisible, which is exactly why we misattribute the fatigue. You remember answering a quick question. You do not remember the ninety seconds of groping around for where you were, and you certainly do not remember doing that forty times in a day.
Fatigue also compounds the problem rather than simply adding to it. Research on interruption under fatigue found that being tired worsened the effects of interruption on working memory and performance, meaning the reload likely gets harder to manage later in a long day.
For a general reader, this suggests the afternoon slump may be less about running out of energy and more about paying a rising toll on every switch.
The Busyness Misread
A day spent handling many small things can feel enormously productive.
Photo by Messages answered, decisions unblocked, forms filed. The visible output is high, and the sense of motion is real.
What that day hides is the accumulated reload tax. Every one of those small completions sat on either side of a context switch, and switching reliably costs time and accuracy. The activity is genuine. The efficiency is partly borrowed against a cost you cannot see on the calendar.
Task switching generally makes people take longer and commit more errors than staying with one task. (NIH)
Shallow days are not wasted, and deep focus is not morally superior to them. Plenty of real work is genuinely made of small pieces. But the exhaustion at the end of such a day is not a sign of weakness or poor discipline. It is the bill for several dozen doorways.
What This Changes In Practice
Once the cost lives in the reload rather than in the interruption itself, some familiar advice starts to look different.
Photo by Batching similar tasks is usually framed as a scheduling trick. Under this lens it is something more specific: grouping work that shares a mental setup, so you pay the reload once instead of six times. Answering all your messages in one stretch is cheap not because it saves minutes, but because your brain keeps the same machinery running the whole time.
The research also suggests some interruptions are worth protecting against more than others. An unrelated request landing in the middle of complex work is a very different creature from a related one landing during routine work, even if both take two minutes. That distinction is something you can actually notice in real time and act on.
Leaving yourself a breadcrumb before you turn away, whether it’s a half sentence or a scribbled next step, builds your own re-entry ramp. You can’t stop the doorway, but you can leave a note on the other side of it.
The person standing in front of the open fridge is not losing their memory. They are watching an ordinary piece of mental machinery do exactly what it is built to do: close one episode, open another, and file the old contents somewhere less immediately reachable. That same machinery runs every time a question arrives mid-paragraph.
The cost sits not in the ten seconds you gave someone, but in the rebuilding afterward, repeated and unnoticed all day. Next time the afternoon feels heavier than the work explains, try counting the doorways instead of the hours.
🔖
- Mental workload higher when interruption and primary task differ in load type
- Interruption frequency and subjective workload in complex tasks
- Illegitimate interruptions reduce productivity in the workplace
- Task switching increases time and errors compared with single-task work
- Fatigue-related effects in the process of task interruption on working memory
