BrainScoreDaily

Attention

Why can't you just ignore the word?

You know the rule. You agreed to the rule. The rule is simple. And the word still gets there first. That effect is ninety years old, it has been measured precisely, and the most interesting thing about it is that it only works in one direction.

What Stroop actually did, in 1935

J. Ridley Stroop printed colour words in the wrong ink — red in blue, blue in green — and ran two experiments that are mirror images of each other. They did not come out symmetrically, and that asymmetry is the whole finding.

Stroop (1935), experiments 1 and 2. Times are for 100 responses. The conflicting information is identical in both directions; only the instruction changes.
The taskWhat interferesWhat it cost
Read the words The ink is the wrong colour 2.3 seconds slower across 100 words — which Stroop himself reported as not reliable
Name the ink colours The words say other colours 63.3 seconds became 110.3 — 47 seconds slower across 100 colours

Same conflict, opposite instruction, and one of them barely registers. Colours do not meaningfully slow down reading. Words nearly double the time it takes to name a colour.

Stroop's own answer to why

He put it down to training, and the sentence is worth reading in his words: "The word stimulus has been associated with the specific response 'to read,' while the color stimulus has been associated with various responses: 'to admire,' 'to name,' 'to reach for,' 'to avoid,' etc."

You have read words a very large number of times and always done the same thing with them. You have seen colours at least as often and done a dozen different things with them. So one of those has a single, overwhelming default and the other does not — and the one with the default wins the race every time, whether or not you want it to.

The part that answers the question in the title

Stroop ran a third experiment: eight days of practice, 200 responses a day, naming the ink colour while the words fought back. The practice worked. It made people faster and the interference smaller.

It did not remove it. In his summary: practice "decreased the interference of conflicting word stimuli to naming colors but did not eliminate it."

So the honest answer to why can't you just ignore the word is: you can get better at it, you cannot finish getting better at it, and eight days of deliberate training was not enough to make the conflict go away. That is also, incidentally, why the effect is still usable as a measurement ninety years later — if practice erased it, there would be nothing left to measure after the first go.

Our version uses arrows, not colours — and you should know that before you take it

Ignore the Word puts a direction word across an arrow pointing somewhere else. You answer the arrow. It is a Stroop-family task and it is not the colour Stroop, and we would rather say that plainly than let this page's subject imply otherwise.

The reason is accessibility, and it is not a small one. A colour Stroop cannot be taken honestly by a colour-blind reader — the task would be measuring their colour vision and reporting it as attention. Direction survives that. It costs us the direct comparison with ninety years of published colour-Stroop figures, which is a real loss and is the trade we made.

What "attention" means here, and the three things it does not

This site uses domain to mean something we have a test for. Attention qualifies; several things people mean by the word do not.

If your attention or your concentration is worrying you, an online test is not the tool for that question — and neither is an article. A doctor is. Nothing on this page can tell you whether something is medically wrong, and a low score on any test here does not indicate a condition.

And one admission that belongs on this page rather than in a footnote. Four of our instruments are anchored to the ETS Kit of Factor-Referenced Cognitive Tests, which demonstrates 72 tests against 23 cognitive factors. There is no attention factor among them. The nearest is Perceptual Speed, which is roughly what our symbol match already measures. So this domain does not have the lineage the others have, and it is anchored to the interference literature directly instead.

Why our score is a count and not a reaction time

The best-known modern attention test reports its results as differences between reaction times — how many milliseconds slower you were when something conflicted. We do not publish a difference score as the output of any domain: two noisy measurements subtracted from each other give a third that is noisier than both and looks more precise than either.

In a browser it is worse than that. The effects are tens of milliseconds, and we have published our own measurement of a touchscreen adding four times more latency than the response it was supposed to be measuring.

So we score accuracy — and that forces a deadline. Given unlimited time almost everyone gets almost everything right, and the test says nothing; the published conditional-accuracy work finds congruency effects on accuracy only in responses faster than about 600 milliseconds. Ours is 0.65 seconds per trial, it is a number we chose rather than measured, and the page explaining why two attention tests disagree treats that as the weakest joint in the instrument.

Where to go next

Sources