Reaction Time Test — Standard Scores & Reference Values
How fast is a fast reaction time, really? This page collects the published reference values — large-sample medians, the effects of age, sex and education, device and screen latency, fatigue and practice — with the source attached to every figure, so you can read your own score in context.
What simple reaction time is — and how big it is
Simple visual reaction time is the delay between one expected signal (a screen turning green) and one pre-planned response (a click). The largest public dataset is Human Benchmark’s: since the site launched it has recorded over 81 million reaction-time clicks, with a median of 273 milliseconds and an average of 284 milliseconds. Treat those as the everyday reference point for web-based tests like ours. Source: https://humanbenchmark.com/tests/reactiontime/statistics
Biology moves the number more than anything else. Age is the strongest known factor: reaction time slows across adulthood, as shown both in a representative UK sample of over 7,000 adults (Der & Deary 2006, Psychology and Aging: https://doi.org/10.1037/0882-7974.21.1.62) and in two large cohorts analyzed by Talboom et al. 2021 (npj Aging: https://doi.org/10.1038/s41514-021-00067-6). The same Talboom study found men on average 34 ms faster than women, and faster times with more education — 15 ms (some college) to 32 ms (college degree) faster than participants without a high-school diploma (plain-language summary: https://mindcrowd.org/reaction-time-as-a-measure-of-brain-health-mindcrowd-study-findings/).
Your hardware is part of your score. Human Benchmark notes that about 30 ms is a typical input-plus-display lag for a desktop or laptop, that mobile taps run slower than mouse clicks, and that measured times have drifted slower over the years as display technology changed (https://humanbenchmark.com/tests/reactiontime/statistics). Timing research agrees: the PeerJ “timing mega-study” (Bridges et al. 2020: https://doi.org/10.7717/peerj.9414) documents systematic differences between platforms and browsers, and jsPsych’s timing documentation summarizes JavaScript-measured reaction times as typically 10–40 ms longer than dedicated lab systems (https://www.jspsych.org/v7/overview/timing-accuracy/). A 270 ms click on a gaming rig and a 270 ms tap on a phone are not the same biological performance.
State and practice matter too. Fatigue: in the landmark sleep-restriction experiment (Van Dongen et al. 2003, Sleep — full text: https://www.med.upenn.edu/uep/assets/user-content/documents/Van_Dongen_Dinges_Sleep_26_3_2003.pdf), two weeks of six-hour nights degraded vigilance-task reaction performance to levels comparable with total sleep deprivation. Practice: a review of action-video-game research (Dye, Green & Bavelier 2009: https://doi.org/10.1111/j.1467-8721.2009.01660.x) found regular players respond 11% faster across 89 task conditions with identical accuracy (92.76% vs 92.75%), and 50 hours of action-game training cut reaction times by 13% versus 6% in a control-game group — evidence of trainable, but bounded and task-flavored, gains.
Reference values, each with its source
| Measure | Reference value | Source |
|---|---|---|
| Large online aggregate (Human Benchmark) | Median 273 ms · mean 284 ms · 81M+ recorded clicks | https://humanbenchmark.com/tests/reactiontime/statistics |
| Age trend | Reaction time slows across adulthood (representative UK sample, 7,000+ adults; two large cohorts in Talboom et al. 2021) | https://doi.org/10.1037/0882-7974.21.1.62 · https://doi.org/10.1038/s41514-021-00067-6 |
| Sex and education differences | Men on average 34 ms faster than women; some college 15 ms and college degree 32 ms faster than no high-school diploma | https://doi.org/10.1038/s41514-021-00067-6 · https://mindcrowd.org/reaction-time-as-a-measure-of-brain-health-mindcrowd-study-findings/ |
| Device and platform latency | ≈30 ms typical desktop/laptop lag; mobile taps slower than clicks; JavaScript-measured RTs typically 10–40 ms longer than lab systems | https://humanbenchmark.com/tests/reactiontime/statistics · https://doi.org/10.7717/peerj.9414 · https://www.jspsych.org/v7/overview/timing-accuracy/ |
| Fatigue effect (sleep restriction) | 6 h/night for 14 days → vigilance reaction performance degraded to levels comparable with total sleep deprivation | https://www.med.upenn.edu/uep/assets/user-content/documents/Van_Dongen_Dinges_Sleep_26_3_2003.pdf |
| Practice effect (action video games) | Players 11% faster RT across 89 conditions at equal accuracy; 50 h training → 13% RT decrease vs 6% control | https://doi.org/10.1111/j.1467-8721.2009.01660.x |
| Median of 5 rounds on this site | No published norm exists for 5-round web click tests per se | SchulteLab user aggregate (beta — sample still accumulating) |
Honesty note: web-based reaction times include device and screen latency, so compare like with like — same device, same mode, trend over time. There is no official scoring band for browser click tests; where no published norm exists, we label our own user aggregate as beta rather than invent numbers.
Measure offline or take the test
Time your own clicks in five rounds, or print the ruler-drop card and measure reaction time with a partner and a ruler. Open the printable reaction card →
Frequently asked questions
What is a good reaction time?
For web-based click tests, Human Benchmark’s 273 ms median (mean 284 ms, 81M+ clicks) is the everyday reference, so running under about 250 ms is brisk and consistent sub-230 ms is fast for a browser. Lab systems read 10–40 ms lower because they skip device lag. Compare like with like, and weigh your own trend more than any single number.
Does reaction time change with age?
Yes — slowing across adulthood is one of the most replicated findings in the field, seen both in a representative UK sample of over 7,000 adults (Der & Deary 2006) and in two large cohorts (Talboom et al. 2021). The same research reports men averaging 34 ms faster than women and faster times with more education.
Why is my reaction time slower on my phone than my computer?
Mostly hardware, not biology. Touchscreens and mobile pipelines add latency — Human Benchmark notes mobile taps run slower than mouse clicks and that roughly 30 ms is a typical desktop lag, while timing studies find browser-measured reactions run 10–40 ms behind lab equipment. Retake tests on the same device before comparing scores.
Can you train your reaction time?
Yes, within limits. Sleep is the biggest lever — two weeks of six-hour nights impaired vigilance performance like total sleep deprivation in Van Dongen 2003. Practice on the task itself helps (near transfer), and action-game research shows an 11% speed edge at equal accuracy. Expect modest, task-specific gains, not a transformed brain.