Reaction Time Test — Standard Scores & Reference Values

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.

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.

The cleanest lab measurement is Woods et al. 2015 (n = 1,469, ages 18–82, hardware-calibrated): mean simple reaction time 231 ms, 213 ms after hardware correction, slowing by about +0.55 ms per year of age — while raw stimulus detection (131 ms) did not change with age at all (Frontiers in Human Neuroscience: https://www.frontiersin.org/journals/human-neuroscience/articles/10.3389/fnhum.2015.00131/full). The aging cost is in the response, not the seeing.

The shape of the curve comes from Der & Deary 2006 (n = 7,130, representative UK sample): simple reaction time stays nearly flat until about age 50 and accelerates afterwards, while choice reaction time slows linearly across all of adulthood (Psychology and Aging: https://doi.org/10.1037/0882-7974.21.1.62). We cite the qualitative shape only — their per-decade millisecond tables sit behind a paywall we could not verify.

The web-data counterpoint is Talboom et al. 2021 (n = 75,666 online participants): reaction time slowed by about +7 ms per year (npj Aging: https://doi.org/10.1038/s41514-021-00067-6) — ten times the lab slope, because uncontrolled device heterogeneity inflates it. That gap between +0.55 and +7 ms per year is the standing lesson of this whole page: a web test and a lab test are different instruments, and slopes mix biology with hardware.

Orient your median against these sourced anchors. Sport’s hard floor: World Athletics rules treat any start reaction under 100 ms after the gun as a false start — below that, you anticipated rather than reacted (Technical Rule 16.8; see the federation’s explainer: https://worldathletics.org/news/news/iaaf-sprint-start-research-project-is-the-100). The web anchor: Human Benchmark’s 273 ms median across 81M+ clicks (https://humanbenchmark.com/tests/reactiontime/statistics). The lab anchor: 213 ms hardware-corrected (Woods 2015, linked above). And the live one: this site’s community distribution in the strip under the reference table — the only anchor measured on the same task and input mix as your score.

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 →