About the Stroop Test
The Stroop Test is a browser test that shows color words printed in a different ink and asks you to pick the ink color while ignoring what the word says. It shows, in milliseconds (ms), how much you slow down when the word and the ink disagree — the Stroop interference effect.
Everything runs inside your browser, and your results are kept only in this device's browser storage. There is no sign-up and nothing is sent to a server.
Key features
- Three conditions, randomly mixed — Congruent, incongruent and neutral (a string that is not a color word) trials each make up one third, ordered so the same correct answer never appears twice in a row.
- Color words in 11 languages — Each language page uses that language's color words (RED, ROJO, 赤, КРАСНЫЙ, أحمر and so on), because interference only shows up properly in a language you read fluently.
- Reverse Stroop mode — Switch the task to “pick the word's meaning” to see whether the ink color gets in the way of judging the word.
- Keyboard and buttons — Answer with the 1·2·3·4 or D·F·J·K keys, or with the on-screen buttons. The button order is fixed, and each button is labeled with its color name so you never have to rely on color alone.
- Result analysis — Mean and median reaction time and accuracy per condition, the interference effect (incongruent − congruent), facilitation and inhibition cost, error rate difference, and a trial-by-trial scatter plot.
- My history — Results and the interference trend across sessions are saved on this device. You can copy your results as text and share the same settings via a link.
How to use
- Choose the task (ink color / word meaning), the number of trials (24, 48 or 72), the time limit (2 s, 3 s or none) and whether to practice first.
- Press the Start button or the Space key. A + sign appears briefly in the center of the screen, then a word appears.
- In the ink-color task, press the button that matches the color of the letters. 1 = RED, 2 = GREEN, 3 = BLUE, 4 = YELLOW, and D·F·J·K follow the same order.
- Answer quickly, but accurately. Wrong answers and timeouts are flagged briefly before the next trial. Press Esc at any time to stop.
- When you finish, your results and scatter plot appear and the session is saved.
Measurement accuracy and limitations
- This is not a medical diagnosis. This test is not a standardized neuropsychological test, and it does not diagnose or rule out any condition related to attention or executive function.
- There are no comparison norms. No validated norms exist for online button-press tasks, so we do not show comparisons such as “top X%”. Compare only your own results measured on the same device.
- Device latency — Keyboard and touch input delay and the screen refresh rate (16.7 ms per frame at 60 Hz) are added to absolute reaction times. The interference effect, being a difference between conditions, is less affected.
- Color display — Screen color settings, night mode or blue-light filters, and color vision deficiency all affect how well you can tell ink colors apart. It is best to turn off screen filters during the test.
- Timing method — Words are drawn in sync with screen updates (requestAnimationFrame), and the time from that frame to the input event is measured with the performance.now() clock. Responses faster than 200 ms are treated as anticipatory and excluded from analysis.
Frequently asked questions
What is a normal Stroop interference effect in ms?
We do not give a “normal range”. The size of the interference varies widely with the response method (voice or button), device latency, language, number of trials and amount of practice, and no validated norms exist for online button-press tasks. Look at the trend in your own results measured on the same device with the same settings.
My interference effect came out as zero or negative.
With few trials, random variation can do that. At 24 trials there are only 8 per condition, so the value is especially shaky. Try again with 48–72 trials and check your accuracy as well. If you took the test on a page in a language that is not your native one, a smaller interference effect is to be expected.
Why is there an XXXX in the neutral condition?
It is a baseline without the meaning of a color word. Comparing against neutral lets you separate the part where congruent trials speed you up (facilitation = neutral − congruent) from the part where incongruent trials slow you down (inhibition cost = incongruent − neutral). In reverse Stroop mode, the neutral condition shows color words in achromatic ink.
Which is more accurate, keyboard or touch?
A wired keyboard generally has shorter and more consistent latency. Touchscreens and wireless keyboards can add more delay. However, that delay is added about equally to every condition, so the interference effect (a difference) stays relatively stable. Don't switch input methods when comparing your results.
Can I take the test if I have a color vision deficiency?
The buttons are labeled with color names, but the ink-color task requires judging the color of the letters, so red-green color vision deficiency can make you slower or cause more errors regardless of interference. The word-meaning task (reverse Stroop) is less affected. This site is not a color vision test.
Why are the words different for each language?
Stroop interference arises when you read a word automatically. Color words in a language you are less familiar with bring their meaning to mind less automatically, so interference is smaller. That is why each language page uses that language's color words. The full list is on the Color words page.
Where are my results stored?
Only in this browser's localStorage; nothing is sent to a server. You can delete them with the Clear history button or by clearing your browser data, and they are not synced to other devices.
Can these results tell me whether I have ADHD or dementia?
No. This test is not a medical diagnostic tool. Clinical Stroop tests follow a standardized procedure, use norms by age and education level, and are interpreted by a professional as part of a broader assessment. If you have symptoms that concern you, please talk to a healthcare professional.
References
Stroop, J. R. (1935). Studies of interference in serial verbal reactions. Journal of Experimental Psychology, 18(6), 643–662.
MacLeod, C. M. (1991). Half a century of research on the Stroop effect: An integrative review. Psychological Bulletin, 109(2), 163–203.
Cohen, J. D., Dunbar, K., & McClelland, J. L. (1990). On the control of automatic processes: A parallel distributed processing account of the Stroop effect. Psychological Review, 97(3), 332–361.
Durgin, F. H. (2000). The reverse Stroop effect. Psychonomic Bulletin & Review, 7(1), 121–125.
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.