What is the average human reaction time?

The short answer, and its two sources

In a large calibrated study, 1,469 adults aged 18 to 65 responded to a simple visual signal with a mean reaction time of 231 ms, and 213 ms once the delay of the computer hardware was subtracted (Woods et al., 2015a). Online, across 70,940 games of the Human Benchmark reaction test, the median is 228 ms (Human Benchmark, 2026).

Two very different settings land on almost the same number, which is reassuring. In everyday terms, a typical adult reacts to a light in a little under a quarter of a second.

By age, about half a millisecond per year

In the same lab study, reaction time rose by about 0.55 ms per year of age between 18 and 65, with no difference by sex or education (Woods et al., 2015a). The authors estimated that simply detecting the signal takes about 131 ms and does not change with age. The slowing comes mostly from starting the movement.

In practice, 0.55 ms per year means about 22 ms over 40 years. It is real, and it is smaller than the difference between two phones.

Simple or choice reaction time

A simple reaction is one signal and one answer, the case of the reaction time test. A choice reaction gives several possible signals with a different answer for each. In 1952 William Hick showed that reaction time grows with the number of possible choices, in a regular way (Hick, 1952). This is why a choice task always gives longer times than a simple one, and why the two should never be compared directly.

The idea of splitting reaction time into steps goes back to Franciscus Donders in 1868, who compared simple reactions, choice reactions and reactions to one signal among several (Donders, 1868).

What distorts online comparisons

Online times include the delay of your device. Conventional touchscreens add 50 to 200 ms of latency (Van Patten et al., 2021), and in 59,587 visitors to an online test platform, people on mobile devices, Android phones especially, were slower than those on computers (Passell et al., 2021). A friend on a gaming mouse and you on an old phone are not playing the same test.

Online data also contain impossible times. The Human Benchmark distribution holds 79 games under 100 ms, which are anticipations rather than reactions (Human Benchmark, 2026). Serious tests reject them, and so does Headgame.

Percentile table

Percentiles from the 70,940 Human Benchmark games (Human Benchmark, 2026). Read P25 as the time a quarter of games are faster than.

Sources

1. Woods D.L., Wyma J.M., Yund E.W., Herron T.J., Reed B. (2015). Factors influencing the latency of simple reaction time. Frontiers in Human Neuroscience 9, 131. doi.org/10.3389/fnhum.2015.00131
2. Human Benchmark, score distributions embedded in humanbenchmark.com, read on 13 October 2026. Reaction time n 70 940, sequence memory n 35 868, chimp test n 27 233, number memory n 24 084. humanbenchmark.com
3. Hick W.E. (1952). On the rate of gain of information. Quarterly Journal of Experimental Psychology 4(1), 11 to 26. doi.org/10.1080/17470215208416600
4. Donders F.C. (1868, translated 1969). On the speed of mental processes. Acta Psychologica 30, 412 to 431. doi.org/10.1016/0001-6918(69)90065-1
5. Van Patten R., Iverson G.L., Muzeau M.A., VanRavenhorst-Bell H.A. (2021). Test-retest reliability and reliable change estimates for four mobile cognitive tests administered virtually in community-dwelling adults. Frontiers in Psychology 12, 734947. PMC8566921. doi.org/10.3389/fpsyg.2021.734947
6. Passell E., Strong R.W., Rutter L.A. et al. (2021). Cognitive test scores vary with choice of personal digital device. Behavior Research Methods 53(6), 2544 to 2557. PMC8568735. doi.org/10.3758/s13428-021-01597-3