Perceptual load and distraction

Why is a nearby distraction easier to ignore during a hard task than during an easy one?

Common sense says that a hard task should be easier to derail. It is using more of you; there is less of you left to resist interruption. Yet the experience is often the reverse. Proof-reading a dense table, you may not register the conversation two desks away. Copying the same figures mechanically, that conversation becomes impossible to shut out.

So the harder task is the more protected one. Before explaining that away, notice how strange it is: the same room, the same voices, the same person — and the only thing that changed was how demanding the thing you were looking at happened to be.

Let us begin with a question that sounds separate but is not. Can you choose not to see something that is in front of your open eyes?

Try it with a word in your own language. You cannot look at it and withhold its meaning; the reading happens before you get a say. That tells us something important — perception is not a tap you open when you want it. It runs on whatever reaches it, without asking.

Now put a second fact beside it: perception is limited. There is only so much you can take in at once. Hold those two together and ask what must follow. If the capacity is bounded, and the processing is not optional, then whatever capacity your current task does not consume does not sit idle in reserve. It cannot — because nothing is holding it back. It spills onto whatever else is in the field of view.

That is the whole mechanism, and it is a conservation argument rather than a psychological one. The budget is fixed and it is always fully spent; the only question is on what. Ask the consequences.

If the task is easy — a single large letter, a familiar layout — how much of the budget does it consume? Very little. And the remainder is therefore spent, involuntarily, on the flashing banner at the edge of the screen. You then have to undo that processing, which is effortful and slow, and you experience the banner as distracting.

If the task is demanding — six similar items to search through, all competing — the budget is exhausted by the task itself. What is left for the banner? Nothing. And notice what this predicts, which is the striking part: the distractor is not resisted, and it is not filtered out by an act of will. It is simply never processed far enough to interfere. Nilli Lavie built her load theory of attention around exactly this in the mid-1990s, and it dissolved a long argument in the field about whether selection happens early or late in the perceptual stream. The answer turned out to be: it depends on what is left over.

Now a test of whether we have really understood it. Suppose I keep the picture simple but make you hold six digits in memory while you look at it. Task harder — so, more protection from the distractor?

No. Distraction gets worse, and this reversal is one of the strongest pieces of evidence for the whole account. The reason is that we have loaded a different resource. Working memory is not what perceives the banner; it is what keeps track of which thing is currently the priority. Occupy it and perception is still free to consume the distractor, while the machinery that would have suppressed the resulting intrusion is busy. So perceptual load protects, and cognitive load exposes — the two pull in opposite directions, which is a strange prediction to make and a good reason to take the theory seriously.

Two honesties before we move on. First, the theory is not uncontested: some researchers argue that adding items to a display dilutes the processing of a distractor simply by competing for it, without any fixed capacity being exhausted, and the dilution account explains a good part of the same data. Second, "capacity" here is a summary of a behavioural regularity, not a measured quantity of anything — nobody has found the tank it is stored in.

Two clarifications.

First, this is not willpower. It is tempting to read "the hard task protected me" as evidence of concentration in the virtuous sense, but the protection is a side effect of the budget being spent, and it appears whether or not you are trying. Whatever your resolve, an easy task leaves capacity, and left-over capacity is spent on something.

Second, the practical advice runs backwards from the intuitive version. When you must work in a distracting place, the useful moves are to keep the perceptual demand of the work high — and, just as importantly, to keep your working memory clear, because a head full of unresolved items is what actually lets the noise in. Removing the distractor from the visual field is more effective still, since a distractor that never enters perception costs nothing to reject.

• The dilution account, and what experiment could separate it from a genuine capacity limit.
• Why load effects are weaker for some distractors — your own name, a face, a sudden movement.
• What this predicts for open-plan offices, cockpit design, and interface clutter.

Every term the collection defines is gathered in the glossary.