Shape Stroop Test

The classic Stroop test is built on colour, which makes it unavailable to a large number of people. This is the same conflict with the colour taken out: the word still fights the thing you can see, but what you can see is a shape.

How to play the Shape Stroop Test

Each trial draws one outline shape with a shape word printed inside it. Your job is to name the outline and choose it from the buttons below. The word is irrelevant, and on most trials it names a different shape entirely.

When the word SQUARE sits inside a circle, the answer is circle. You will read the word before you have decided anything — that is not a failure of concentration, it is how reading works once you are fluent. The task is to stop the word from becoming your answer.

Trials advance as soon as you answer, so a round takes as long as you take. Each trial also has its own time limit, and letting it expire counts as wrong.

Why this exists alongside the colour version

We built the colour Stroop Test first, and it has a real limitation we could not design away: naming the ink is the entire task, so a player who cannot distinguish the inks cannot play it. Every button carries a name and a shape so choosing an answer never depends on colour, but the stimulus is a colour and nothing can change that.

This game replaces the colour dimension with a shape dimension and keeps everything else. There is no colour channel anywhere in it — not in the stimulus, not in the answers, not in the feedback. It is also the version that appears in the Daily Brain Workout, so the daily routine does not contain a station that some players are shut out of.

One honest difference: the interference here is usually a little weaker than in the colour version. A word and the ink it is written in are the same object, while a word and the outline around it are slightly more separable, so the pull is not quite as strong. It is the same effect, somewhat gentler.

What actually helps

1. Look at the edges, not the middle. The outline lives at the border of the figure and the word sits in the centre. Fixing your attention outwards reduces how much the word competes.
2. Name the shape before you look at the buttons. Deciding first and then finding the button is faster than scanning the row while still resolving the conflict.
3. Learn the button positions. The order is fixed for the whole round, so after a few trials your hand can find the answer without a search.
4. Accept the delay on conflict trials. Being slower there is the expected pattern. Rushing produces errors, and an error costs far more than the fraction of a second it saved.
5. Recover immediately. A wrong answer is one trial. Carrying it into the next two turns one mistake into three.

What the difficulty levels change

Three real changes move together. Easy uses three shapes, so there are fewer competing responses to hold ready. Standard and Hard add a fourth. The share of conflict trials rises from 40% to 54% to 70%, so there is progressively less respite. And the per-trial limit tightens from 2.8 to 2.0 seconds.

Hard is not simply Standard sped up: at 70% conflict the matching trials stop working as a rest, and the outline rule has to be actively held for nearly the whole round.

How the score works

Correct answers supply 82% of the score and answer pace the remaining 18%. The pace contribution is gated by accuracy and collapses at the random-choice baseline — one in three on Easy, one in four on Standard and Hard — so tapping quickly through a round earns nothing. Expired trials count as errors.

Your result also reports your interference effect: how much slower your correct conflict answers were than your correct matching ones. It is deliberately not part of your score. Over twenty-odd trials it is far too noisy to rank anyone, and scoring it would reward answering slowly on every trial to flatten the difference. It is shown because it is interesting to see in your own data, not because it measures you.

Records are kept separately for each level, since a 70%-conflict round is not comparable to a 40% one.

How each round is built and checked

Every round comes from its seed, so the same seed always produces the same trials. The generator first builds a balanced list of outline shapes, so each shape is the correct answer a near-equal number of times and no button ever becomes the safer guess. It then selects which trials carry conflict and gives each one a word drawn from the other shapes in play.

The round is then re-read independently: congruency is recalculated from the word and outline alone rather than trusted from the generator, the conflict count is checked against the level, and the balance of correct answers is verified to be within one trial of even. The same outline may not appear more than twice in a row, which would turn the round into a motor repetition rather than a naming task, and no more than four matching trials may run consecutively, because a long stretch without conflict lets the reading habit reassert itself. Long runs of conflict trials are deliberately permitted — they are the difficult part, not an unfairness.

Those generation rules are shared with the colour Stroop Test rather than copied, so a correction to the balance or run logic cannot silently apply to only one of the two games.

Common mistakes

Trying to not read the word

You cannot decide not to read a word you can see. The achievable goal is not suppressing the reading, but declining to act on it.

Rushing the conflict trials

Accuracy is worth 82% of the score and speed 18%, so trading a correct answer for a faster one is always a bad exchange.

Hunting for the button after deciding

If you resolve the outline and then search the row, the search costs as much as the conflict did. The button order never changes within a round.

Reading the answer buttons

The buttons carry shape names, which is itself readable text. Using the drawn shapes and the positions rather than re-reading the labels removes a second, unnecessary reading step.

Shape Stroop questions