Stroop test

There is no single good number, and this page does not pretend there is. The size of the effect depends on the words and the colours, on how you answer, on your screen and on whether you press a key or touch a button. The nearest published figure comes from a study of 47 students who answered with a key for each colour: the clashing words cost them 78 ms on average^[1]. Their test was much longer and the group small, so the first view of your result shows that average as a mark beside yours, never as a rank. How we measured this also draws your earlier results as short ticks on a ruler.

Almost everyone is slower on the clashing words, so a result above zero is normal. What changes from run to run is how much slower. A smaller effect on a day when you are rested and a larger one when you are tired is a pattern you can see in your own history.

Each word is drawn at the start of a new picture on your screen, and the clock starts with that picture. Your answer is timed from the browser's own record of your tap, click or key, not from the moment our code got round to reading it. Between two words the screen is empty for a time that changes every word, between half a second and nine tenths of one, so you cannot get ready for the next one.

A browser cannot see two small delays: the time your screen needs to show the picture, and the time your touch screen, mouse or keyboard needs to report the answer. They are in both middle times, and so they largely cancel out of the difference between them. That is the reason this test gives a difference and not a single reaction time. If you switch to another tab or app, or the screen stalls while a word is out, that word does not count and it comes again. A key held down repeats by itself: only the first press counts.

A run is compared with nothing if too few of your answers to either kind of word were right, or if too small a share of your answers were right. It is still saved. Your result stays in this browser and nowhere else: it never goes into a link, and it is never sent to us. You can read about us to see who makes the site.

Reading a word is something you learnt so well that it runs by itself, even when you do not want it to. Naming a colour is slower and takes more care. When the word says blue and the ink is red, the two fight, and you have to hold the ink and hold back the word. That small fight costs a fraction of a second. The effect is named after the psychologist John Ridley Stroop, who described it almost a century ago, and it has been used in countless experiments since.

The clash is also why the practice words are there. The first few answers are slow while you find the buttons and settle the rule in your head, and we do not want that in your result.

• Look at the middle of the word, a little through it, rather than at its letters.
• Keep one finger over the buttons or the keys, so that you only have to press and not to find them.
• Do not slow down to be certain. A wrong answer costs less than you fear, and a slow answer costs time on every word.
• Answer with the same hand every time: swapping between a key and a touch changes your times.
• Take the test at a similar time of day if you want to compare runs.

This is one of the few tests where colour is the task itself, so it cannot be done without telling colours apart. We still make sure that no answer depends on colour alone: each button carries the name of its colour and, with a keyboard, a key. If you cannot tell red, green, blue and yellow apart on a screen, the effect this test measures will not show, and another test on this site will serve you better, such as the reaction time test.