Reaction Time Test
Click when the screen turns green. 12 scored rounds measure your average simple visual reaction time in milliseconds — then a 6-round distractor block tests your impulse control separately.
What this test measures
Simple visual reaction time: the delay between a visual stimulus (the screen turning green) and your motor response (the press). It is one of the oldest paradigms in experimental psychology, used since Donders' pioneering work in 1868. The test runs 20 rounds: the first 2 are unscored practice, rounds 3-14 are classic simple-RT trials whose average is your score, and rounds 15-20 add flashing distractor colors as a separate impulse-control block. Distractor rounds are shown for interest but never mixed into your scored average — color discrimination is systematically slower than simple detection, so pooling them would bias the score against simple-RT norms. Early presses, distractor presses, and presses landing within 100ms of the green signal (anticipation) are rejected, so the result reflects true stimulus-driven responses.
The science behind it
The largest public dataset comes from humanbenchmark.com — over 81 million clicks with a median of 273ms, published on the site's own statistics page. Laboratory studies report simple visual reaction times of 200-300ms for young adults, slowing with age by roughly 2-6ms per decade (Deary, Liewald & Nissan 2011, Behavior Research Methods). Meta-analyses of action video game players show a consistent speed advantage of about 10-12% over non-gamers without loss of accuracy (Dye, Green & Bavelier 2009, Current Directions in Psychological Science). Browsers, displays and input devices add tens of milliseconds of measurement latency (Bridges et al. 2020, PeerJ — 'The timing mega-study' on experiment-software timing precision), which is why online scores run slower than lab-grade photodiode measurements. Because web reaction times are strongly right-skewed — a long tail of slower responses above a physiological floor around 150-180ms — your percentile comes from a log-normal model of the distribution, not a symmetric bell curve.
How to improve your reaction time
Reaction time is partly trainable. Controlled studies show action video game training transfers to faster responses on unrelated tasks (Green & Bavelier 2003, Nature). Practical levers: sleep (deprivation slows RT by 10-30ms), caffeine in moderate doses, warming up before measuring, and using a high-refresh-rate display with a wired mouse. Expect diminishing returns — most healthy adults can shave 10-30ms with practice, but genetics and age set the floor.
Frequently Asked Questions
What is a good reaction time?
The median across 81M+ online tests on humanbenchmark.com is 273ms. Under our log-normal scoring model, under 240ms puts you roughly in the top 28%, under 210ms around the top 12%, and under 180ms around the top 3%. On the slow side, 330ms sits around the bottom 19% and 390ms around the bottom 5%. Lab-measured times are typically 20-50ms faster because online tests include display and input latency.
Why is my reaction time slower than in other tests?
Hardware matters: a 60Hz display adds up to 16ms per frame, wireless mice add 5-15ms, and browsers add processing time. Compare scores only within the same device and site. Our percentile is calibrated against the same style of browser-based measurement.
Do the distractor rounds count toward my score?
No. Only the 12 classic rounds (3-14) are averaged into your scored reaction time and percentile. The final 6 distractor rounds measure a different ability — withholding responses to decoy colors, closer to a go/no-go task — and discrimination responses run systematically slower than simple detection, so mixing them in would unfairly drag your percentile down. Their average is reported separately at the end of the run.
Does age affect reaction time?
Yes. Simple reaction time is fastest in your early 20s and slows by roughly 2-6ms per decade afterward (Deary et al. 2011). The effect is gradual — a healthy 50-year-old typically tests within 20-30ms of their younger self.
Do gamers really have faster reaction times?
On average, yes. Meta-analyses find action video game players respond about 10-12% faster than non-gamers with no accuracy trade-off (Dye, Green & Bavelier 2009). Training studies suggest the games cause part of this difference rather than just attracting fast responders (Green & Bavelier 2003).
Why do early clicks not count?
Clicking before the green signal means you anticipated rather than reacted. So does clicking within 100ms of it — genuine visual reactions are physiologically impossible that fast. Anticipation produces artificially quick times that don't measure your nervous system's true stimulus-response speed, so those rounds are discarded and restarted.
More Cognitive Tests
Keep your cursor on a moving target for 20 seconds. Measures hand-eye coordination as percent time-on-target.
Press for green circles, hold back for red rings — 45 rapid trials (5 practice + 40 scored: exactly 30 green, 10 red) build an urge to press that your score depends on resisting.
Tap the space bar or click in time with a 120 BPM metronome — 4 beats to listen, 8 to practice, 32 scored. Measures your average timing error in milliseconds.