What is the Stroop test?
In 1935, J. Ridley Stroop published experiments using colour words, such as red, blue and green, each printed in ink of a different colour [1]. In one, 70 college students read the words aloud: reading 100 colour words printed in clashing colours took only 2.3 seconds (5.6%) longer than reading the same words in black, a difference that was not reliable [1]. In another, 100 students named the ink colours instead. Naming the colours of 100 clashing words took 110.3 seconds on average, against 63.3 seconds for 100 colour patches: 74% longer [1].
That extra time is the Stroop effect, also called the interference effect. Stroop's own experiments used printed lists of words and never included matching words such as RED printed in red; those came later [2]. Computer versions typically show one word at a time, and one common score is the difference in response time between mismatching (incongruent) and matching (congruent) words [3].
A 1991 review of about 400 studies found the effect with lists and with single words, with many kinds of response, and in related tasks such as naming pictures with a word printed on them [2].
How it works: the interference effect
The same review listed 18 reliable findings that any explanation of the effect has to account for [2]. Several of them change the size of the number you get on a Stroop test:
| What changes | What happens to interference |
|---|---|
| Matching words (RED in red) | They speed you up a little, but much less than clashing words slow you down |
| Pressing keys instead of speaking | Interference is smaller, but still significant |
| More matching words mixed in | Interference on the clashing words tends to increase |
| Word and colour in separate places | Interference is reduced, not removed |
| Age | It peaks in the early school years as reading develops, falls through adulthood until about 60, then rises again |
All five rows come from the same review [2].
Why does a word you are told to ignore get in the way at all? Stroop's own explanation was practice: we have had far more training in reading words than in naming colours [1]. Later theories focused on how much faster reading is than colour naming, or on reading being automatic. MacLeod found both accounts wanting and argued that theories in which the word and the colour are processed in parallel, and compete, were more likely to succeed [2]. Researchers still debate the mechanism. These theories are background for understanding the task, not evidence about what practising it does.
The reverse effect
Interference in the other direction, when the ink colour slows you down as you read the word, is called the reverse Stroop effect. Stroop's first experiment found almost none [1]. In his third, 32 students practised naming the ink colours of clashing words for eight days, 200 responses a day. The practice reduced the interference on colour naming without removing it, and it introduced interference into word reading [1]. That reverse effect appeared only after considerable training and soon faded; later studies that produced a stable reverse effect usually had to change the task substantially [2].
Interference control is not the same as self-control
The Stroop test is a test of interference control: the ability to keep one feature of what you see, here the meaning of the word, from disrupting your response to another, the colour of the ink [4]. In research, the Stroop task is treated as a measure of response inhibition or executive functioning [3]; our guide to executive functions explains where inhibition fits.
It is tempting to read a good Stroop score as a sign of willpower. The evidence does not support that leap. In a study of 522 online participants who completed a large battery of self-regulation tasks (including Stroop) and questionnaires, task measures and questionnaire measures showed little relationship with each other, and the tasks largely failed to predict self-reported real-life outcomes such as drug and alcohol use, physical and mental health, and socioeconomic outcomes [5].
Why it matters: clinical and research uses
In clinics, the Stroop Color and Word Test is a neuropsychological test used extensively to assess the ability to inhibit cognitive interference [4]. Clinical versions have separate conditions for reading colour words, naming colours and naming the ink colour of colour words, and a systematic review stresses assessing performance in all three [4]. The same review found that none of the many published scoring methods fully captures the Stroop effect [4].
A clinical score means something only against norms. In a study that built norms from 1,856 healthy Dutch-speaking adults aged 24 to 81, the timed Stroop scores were strongly affected by age, sex and education, and the interference score worsened with age more steeply in people with less education [6]. That is why a clinician compares your result with people of your age and education, and why a number from a different version of the test cannot simply be put on the same scale.
In research, the Stroop task is one of the standard measures of inhibition and is also used as a training task. In a randomised study of 102 healthy young adults, three weeks of intensive Stroop practice improved Stroop performance but did not transfer to an untrained inhibition task or to tasks of switching, updating or planning [7].
What a Stroop score can and cannot tell you
The reliability paradox
As a group effect, the Stroop effect is very dependable. In a large project that repeated classic experiments across many labs, it replicated in every attempt, while most other effects replicated far less often [3]. Yet that does not make it a precise measure of one person.
A 2018 study tested the same undergraduates on several classic tasks twice, three weeks apart, with 240 Stroop trials of each kind (matching, mismatching and neutral words) per session [3]. The Stroop cost (response time on mismatching words minus matching words) had a test-retest reliability of 0.60 in one sample of 50 and 0.66 in a second of 62: among the better tasks tested, but no measure in those two studies reached 0.8, the level usually considered excellent or of a clinical standard [3]. The authors explain why. A task becomes popular because almost everyone shows the effect to a similar degree, and that same lack of difference between people makes it hard to rank individuals reliably [3]. Difference scores such as the Stroop cost were also generally less reliable than the response times they are calculated from [3].
Practice changes the score
Doing the task again makes you better at it. In Stroop's practice experiment, interference fell over eight days but did not disappear [1]. Later studies of extended practice often found interference reduced, though not always, and practice on one version did not necessarily transfer to another [2]. More generally, a meta-analysis of 50 studies found that retaking a cognitive ability test raises scores (adjusted effect size 0.26), more so when the identical form is used [8]. So a falling Stroop cost over your first sessions is expected, and on its own it shows that you are getting better at this task.
Online versions show the effect too
The effect does not need a lab. In a browser-based version with 40 anonymous online workers who typed their answers, responses were 293 ms slower on mismatching words than on matching ones, replicating the classic effect [9]. An online test is therefore a fair way to see the effect for yourself. What an online practice test like ours cannot do is give you a clinical score: it is not standardised, it has no norms, and a single short session is a rough reading.
What the research says
How LaserMind's Stroop test works
On LaserMind's Stroop Test, each session has 24 words, half of them printed in a colour that does not match, in random order. A cross shows for 0.4 seconds, then the word, and you have 3 seconds to answer with the D, F, J and K keys or the buttons (Kids mode removes the time limit). Wrong answers, timeouts and answers faster than 150 ms count as errors. Difficulty stays the same throughout a session. There are three modes:
- Colour: name the ink colour and ignore the word (red, green, blue or yellow). The word appears in the language you use the site in.
- Reverse: name the word and ignore the ink. Expect a much smaller effect here, for the reasons above.
- Arrows: an arrow appears on the left or the right; answer the direction it points, whichever side it is on. No colours are needed.
Your main number is the interference cost: your median time on mismatching trials minus your median time on matching ones, counting correct answers only. Lower means the conflict slowed you down less. Because it compares mismatching with matching trials rather than with neutral ones, it combines the slowing caused by clashing words with the smaller speed-up from matching ones [2]. The results screen also shows your accuracy and median response time, says whether your errors came on mismatching trials or elsewhere, and plots your recent sessions in the same mode, with a note on how steady they are. Personal bests are kept separately for each mode and language.
A LaserMind session has 12 trials of each kind, compared with 240 of each in the reliability study above [3], so read any single number as rough and look at the trend over several sessions. If you join the optional leaderboards (adults only, one board per mode, word language and device type), remember that a rank is a practice score on one task, not a measure of ability. How to tell real change from practice and noise is covered in How to measure progress in brain training; how to practise the task is in the Stroop training guide.
Try it on LaserMind
▶ Stroop ChallengeTry Colour mode, then Reverse, and compare your two interference costs.
Try it on LaserMind
▶ Target Watch (Go/No-Go)A different kind of inhibition: press for every letter except X, and hold back when X appears.
What research has not established
- Much of this research used college students, from Stroop's own experiments to the 2018 reliability study [1][3]. Reliability may differ in other groups and with other versions of the task.
- The reliability figures come from long laboratory sessions. They do not tell us directly how stable a short online session is.
- The study linking tasks with real-life outcomes used an online convenience sample and self-reported outcomes, and it tested a battery of tasks rather than the Stroop alone [5].
- Theories of why the effect happens are still debated [2].
- LaserMind's Stroop Test is a practice exercise. It is not standardised, has no population norms and is not a clinical assessment.
Frequently asked questions
What is a good Stroop score? There is no single answer, because the number depends on the version of the test, how you respond, and your age and education [2][6]. On LaserMind, a lower interference cost than your own earlier sessions in the same mode means the conflict slowed you down less on this task.
Can an online Stroop test detect ADHD or dementia? No. Clinical Stroop tests are scored against norms for age and education [6], and researchers still disagree about how best to score them [4]. An online practice version is not a diagnostic test. If you are worried about your attention, memory or thinking, talk to a doctor or another qualified professional.
Why does my score change so much from day to day? Partly because you are learning the task [8], and partly because a single person's Stroop cost is a noisy measurement [3]. A trend over many sessions tells you more than any one result.
What the research says
Last evidence check 2026-10-17The Stroop task (1935) shows that naming the ink colour of a colour word is slower when the word names a different colour. Trained task [1]
Practising the Stroop task improves performance on that task; in a randomised study of healthy young adults, the gains did not carry over to another inhibition task. Trained task [7][11]
It has not been shown to raise general intelligence or everyday self-control. Other abilities [10]
References
- Stroop, J. R. (1935). Studies of interference in serial verbal reactions. Journal of Experimental Psychology, 18(6), 643–662. https://doi.org/10.1037/h0054651
- MacLeod, C. M. (1991). Half a century of research on the Stroop effect: An integrative review. Psychological Bulletin, 109(2), 163–203. https://doi.org/10.1037/0033-2909.109.2.163
- Hedge, C., Powell, G., & Sumner, P. (2018). The reliability paradox: Why robust cognitive tasks do not produce reliable individual differences. Behavior Research Methods, 50(3), 1166–1186. https://doi.org/10.3758/s13428-017-0935-1
- Scarpina, F., & Tagini, S. (2017). The Stroop Color and Word Test. Frontiers in Psychology, 8. https://doi.org/10.3389/fpsyg.2017.00557
- Eisenberg, I. W., Bissett, P. G., Zeynep Enkavi, A., Li, J., MacKinnon, D. P., Marsch, L. A., & Poldrack, R. A. (2019). Uncovering the structure of self-regulation through data-driven ontology discovery. Nature Communications, 10(1), 2319. https://doi.org/10.1038/s41467-019-10301-1
- Van der Elst, W., Van Boxtel, M. P. J., Van Breukelen, G. J. P., & Jolles, J. (2006). The Stroop Color-Word Test: Influence of Age, Sex, and Education; and Normative Data for a Large Sample Across the Adult Age Range. Assessment, 13(1), 62–79. https://doi.org/10.1177/1073191105283427
- Talanow, T., & Ettinger, U. (2018). Effects of task repetition but no transfer of inhibitory control training in healthy adults. Acta Psychologica, 187, 37–53. https://doi.org/10.1016/j.actpsy.2018.04.016
- Hausknecht, J. P., Halpert, J. A., Di Paolo, N. T., & Moriarty Gerrard, M. O. (2007). Retesting in selection: A meta-analysis of coaching and practice effects for tests of cognitive ability. Journal of Applied Psychology, 92(2), 373–385. https://doi.org/10.1037/0021-9010.92.2.373
- Crump, M. J. C., McDonnell, J. V., & Gureckis, T. M. (2013). Evaluating Amazon's Mechanical Turk as a Tool for Experimental Behavioral Research. PLoS ONE, 8(3), e57410. https://doi.org/10.1371/journal.pone.0057410
- Sala, G., & Gobet, F. (2019). Cognitive training does not enhance general cognition. Trends in Cognitive Sciences, 23(1), 9–20. https://doi.org/10.1016/j.tics.2018.10.004
- Simons, D. J., Boot, W. R., Charness, N., Gathercole, S. E., Chabris, C. F., Hambrick, D. Z., & Stine-Morrow, E. A. L. (2016). Do "brain-training" programs work? Psychological Science in the Public Interest, 17(3), 103–186. https://doi.org/10.1177/1529100616661983