Why the word gets in the way
J. Ridley Stroop published the effect in 1935. Naming the ink of clashing colour words took his participants roughly 74 percent longer than naming plain colour patches, and nine decades of replication have never made it go away.
The reason is that reading is automatic for a practised adult. You cannot see the word RED and not read it, so when the ink is blue, two answers arrive and the wrong one arrives first. Naming the ink means holding that one back. The time it takes is the interference, and it is what this test scores.
Why your score survives a change of device
A reaction time test measures raw speed, and raw speed in a browser is partly a measurement of your screen, your mouse and your connection. The same person can score tens of milliseconds apart on two different laptops.
This test times you on both kinds of round, matching and clashing, in equal numbers and shuffled together. Whatever delay your hardware adds, it adds to both, so subtracting one average from the other cancels it. The number you get is the cost of the clash alone.
Only correct answers are timed, and if fewer than 80 percent of your answers are right you get no score at all. Guessing quickly is the easiest way to post a flattering number, and a flattering number that measures nothing is worse than none.
What the comparison is, and is not
The standardised clinical version of this task is a timed paper test read aloud, with published norms behind it. A browser keypress version has no normative sample, so there is no true percentile to give you.
What keypress studies consistently report is the range: most adults show an interference cost somewhere between about 50 and 150 milliseconds. The comparison on your result is built from that range, which is why it says less distracted than about 70% of people rather than quoting a percentile. The first is an estimate and says so. The same standard applies on the visual memory test, while the digit span test quotes a real percentile because published adult norms sit behind it.
Frequently asked questions
What is a good Stroop test score?
Lower is better, because the score is how much the clashing words slow you down. On keypress versions like this one, most adults land somewhere between about 50 and 150 milliseconds. Under about 45 milliseconds is uncommon. A score at or below zero means no measurable interference on that run, which is usually noise from a small number of trials rather than immunity to the effect.
What does the Stroop test measure?
Inhibitory control: the ability to hold back an automatic response in favour of the one the task asks for. Reading is so practised that you cannot switch it off, so naming the ink means overriding it. Inhibition is one of the core executive functions, alongside working memory and switching between tasks.
Is this the same as the Stroop Color and Word Test?
No. The Stroop Color and Word Test is a standardised paper version published by Charles Golden in 1978. You read aloud from three printed cards for 45 seconds each, and the score is how many items you get through, compared against published norms. This is a computerised keypress adaptation of the same effect. It measures the same interference, but the numbers are not interchangeable.
Why is my score a difference rather than a speed?
Because raw speed mostly measures your equipment. Screen refresh, input lag and a slow trackpad all add delay, and they add the same delay to matching and clashing words alike. Subtracting one from the other cancels them out, which leaves the part that is about you: how long it took to override the word.
Does the Stroop test relate to IQ?
Only loosely. Executive functions contribute to reasoning, but inhibition on its own correlates modestly with general ability, and plenty of people with high IQ scores show ordinary interference. A good Stroop score says you block out distraction well. It does not stand in for a measure of reasoning.
Can you get better at the Stroop test with practice?
Yes, somewhat. Interference tends to shrink over repeated sessions as you get used to the task, but it does not disappear, and the gains are largely specific to this task rather than a general improvement in focus.
Can I take it if I am colour blind?
You can, but the result will not mean much. The test depends on telling the four inks apart at a glance, and red-green colour vision deficiency, which affects roughly 1 in 12 men, makes two of them hard to separate. Any extra time spent working out the colour gets counted as interference.