What Is the Average Reaction Time by Age

Overview

The short answer, with the honesty the numbers deserve: for simple visual reaction time — press a key when the screen changes — the most defensible average for healthy young adults is roughly 200 to 250 milliseconds on calibrated laboratory equipment, modeled from Woods et al. 2015, which measured a 231 ms mean across 1,469 adults. On a browser-based test like ours, typical scores run slower — a large self-selected online sample reported a 301 ms median — because every online score carries device and display latency on top of biology. Both figures are reference estimates from published research, not first-party data, and no online test is a medical device.

Below is the age-by-age picture the literature supports, why online tests read slower than lab gear, what moves your score day to day, and what our own five-round reaction time test actually measures. If you want the score context first, our average reaction time reference and what counts as a good reaction time pages apply these research anchors to your result.

What this page gives you, section by section:

• The age-by-age numbers the published literature supports, from childhood to 70+, with exact millisecond figures and sources.
• Why visual, auditory, and tactile reaction times are different instruments — and why you should never compare across them.
• How much sleep, caffeine, practice, and distraction each move your score, in published effect sizes.
• An honest accounting of why any browser test — ours included — reads slower than laboratory equipment.
• What a good score actually looks like on an online test, and how to interpret your own median without fooling yourself.

Average Human Reaction Time by Age

No single study has tested every age on one calibrated device, so the honest version of an age chart combines laboratory anchors with clearly labeled browser-test benchmarks. Treat every number below as a central tendency: the variation between two people of the same age is far larger than the gap between adjacent age bands.

• Under 10: roughly 2 to 4 times slower than adults on lab tasks, with high variability — neural pathways are still maturing. Compiled browser-test benchmarks cluster near 285 ms.
• 10 to 19: rapid improvement through the teens, approaching adult levels by about age 14 in developmental meta-analyses (Kail, 1991); browser scores typically land around 210 to 240 ms.
• 20 to 29: the peak decade. In Wilkinson and Allison's 5,325-subject study, the 20s were the only decade clearly faster than every other group, and calibrated lab means sit around 210 to 250 ms.
• 30 to 39: nearly flat on simple tasks — a representative sample of 7,130 UK adults found simple reaction time barely changes before about age 50 (Der and Deary, 2006). Decline is detectable mainly in speed-critical tasks.
• 40 to 49: still near-flat on simple reaction time, though choice tasks — decide, then respond — are already slowing.
• 50 to 59: simple reaction time begins a measurable climb of about 0.5 ms per year, while choice tasks slow two to three times faster (Fozard et al., 1994). Browser scores commonly read 260 to 280 ms.
• 60 to 69: a classic synthesis put mean simple reaction time at about 327 ms at age 60 versus 264 ms at age 20 (National Academies, 2004) — roughly 20 to 30 percent slower than the peak, with rising variability.
• 70 and over: continued slowing and the widest individual variation of any group. Health, sleep, and hardware differences matter more here than the age label itself.

One finding worth singling out: an analysis of 3,305 StarCraft 2 players found cognitive-motor speed declining measurably from about age 24, at every skill level (Thompson et al., 2014). That sounds dramatic, but the effect is measured in milliseconds and older players compensated with strategy. The age curve is a gentle slope, not a cliff — and these are population averages, not a verdict on any individual.

The bottom line: reaction time improves steeply through childhood, peaks in the early-to-mid 20s, holds nearly flat until about age 50, and then slows by roughly 0.5 ms per year on simple tasks. The difference between adjacent decades is small; the difference between two well-rested and sleep-deprived versions of you on the same day is larger. Use the age bands as context, and your own median under stable conditions as the real baseline.

Visual vs Auditory vs Tactile: Not All Reactions Are Equal

The sense being tested changes the number as much as age does. Classic laboratory figures for young adults are roughly 180 to 200 ms visual, 140 to 160 ms auditory, and about 150 ms tactile — sound reaches the cortex faster than light does its processing work, with one study measuring cortical arrival at 8 to 10 ms for auditory stimuli versus 20 to 40 ms for visual (Jain et al., 2015).

The practical rule: never compare scores across modalities. A 170 ms reaction to sound is ordinary; the same number to a visual signal is fast. This is also why sprinters can legitimately react to the gun in about 85 to 100 ms while no one can do that to a light — the IAAF sprint-start research project recorded reactions below 80 ms, and the rules still treat anything under 100 ms as a false start — and why our auditory reaction test produces lower numbers than the standard visual test. Choice tasks add another step entirely: four-choice visual reaction time averages roughly 480 to 640 ms in young adults, about double simple reaction time, which is why our choice-color reflex test reads slower by design.

What Moves Your Score: Sleep, Caffeine, Practice, Distraction

Day-to-day state swings your score more than most people expect — and more than most permanent traits. The published effect sizes, in milliseconds:

• Sleep loss is the biggest everyday factor. Mean vigilance-task reaction time rises from about 250 ms rested to over 300 ms after one night without sleep, and the landmark Van Dongen et al. (2003) study found two weeks of six-hour nights degraded performance as much as one to two nights of total sleep deprivation — while subjects rated themselves only mildly sleepy.
• Caffeine helps modestly: about 10 to 12 ms faster than placebo in controlled EEG work (Saville et al., 2018), with the largest benefit appearing precisely when people are sleep-deprived. More caffeine is not more speed.
• Practice shaves about 13 ms across ten sessions of a clinical reaction test, then plateaus (Del Rossi et al., 2014). Action video game players respond about 11 percent faster across dozens of task conditions without losing accuracy (Dye, Green and Bavelier, 2009) — real but bounded gains.
• Distraction is worse than any of them: adding a phone conversation raised brake initiation time by 148 to 221 ms in a driving-simulator study (Mohebbi et al., 2009) — several times the total effect of caffeine or practice.

The pattern running through all of this: the biggest levers on your score are state and setup, not raw talent. Our reaction training plan is built around that reality — consistency and conditions first, grinding second.

Why Online Tests Read Slower: The Science and Its Limits

Every computerized reaction test measures biology plus apparatus, and we build one, so we will be blunt about the apparatus part. A 60 Hz screen redraws only every 16.7 ms; keyboards, mice, and touchscreens add polling and debounce delay; the browser and operating system add compositing latency on top. The Bridges et al. (2020) timing mega-study measured response-time lags of roughly 8 to 67 ms across browser-based experiment platforms, and Anwyl-Irvine et al. (2020) found average lags around 80 ms on realistic home hardware — sometimes 100 ms.

That is the whole story behind the gap between headline datasets: calibrated lab systems report about 231 ms (Woods et al.), while the browser crowd reports medians near 273 to 301 ms. The roughly 40 to 70 ms difference is mostly hardware, not slower people. The research literature is clear on the consequence: web-based scores are valid for tracking your own trend on one device, not for absolute comparisons between people on different machines.

What Is a Good Reaction Time?

Against the browser-test context — where your score lives — a five-round median of 250 to 300 ms is typical, and staying under 250 ms is on the faster side. Treat anything far below 150 ms on a visual browser test with suspicion: after device delay is accounted for, it implies near-floor biology, and it usually means an anticipatory click rather than a genuine reaction. Our full breakdown lives on the good reaction time page.

Where the milliseconds actually matter, the stakes get concrete. At 90 km/h, every extra 100 ms of brake reaction time carries a car about 2.5 meters further before the brakes engage — so the 221 ms distraction penalty measured above is worth roughly 5.5 meters, often the difference between a near miss and an impact. Highway design standards assume brake reaction times of 1.64 to 3.5 seconds for exactly this reason. In esports, peer-reviewed measurement puts elite competitors around 175 to 193 ms on visual tasks (Luu et al., 2021) — fast, but far from the mythical 150 ms average circulating online.

How to Interpret Your Score on This Site

Our test runs five rounds with a randomized wait, rejects clicks that jump the signal, and reports your median — the protocol is documented in full on our methodology page. What it gives you is a consistent, repeatable browser measurement. What it cannot give you is a lab-grade absolute number, because no browser test can subtract your specific hardware's latency. A few rules of thumb:

• Compare only against yourself on the same device and browser — a phone tap and a desktop click are different measurements.
• Trust the five-round median, never a single round; one lucky guess is anticipation, not reaction.
• Test under similar conditions — rested, same time of day — before reading anything into a change.
• Treat all published averages, ours included, as reference estimates modeled from published research, not first-party data or medical assessment.

This site is not a medical device, and reaction time scores — here or anywhere — are not a health diagnostic. If you have concerns about your cognition or processing speed, that is a conversation for a qualified professional, not a browser game.

FAQ

What is the average human reaction time?

For simple visual reaction time, roughly 200 to 250 ms on calibrated lab equipment — the Woods et al. (2015) mean was 231 ms in 1,469 adults — and roughly 250 to 300 ms on browser tests, where device and display latency are included. There is no single universal average; the number depends on the task, the modality, and the hardware.

At what age is reaction time fastest?

The early-to-mid 20s. In the largest decade-by-decade study (5,325 subjects), the 20s were the only age group clearly faster than all others. Simple reaction time then stays nearly flat until about age 50 and slows by roughly 0.5 ms per year after that — a gentle slope, not a cliff.

What is a good reaction time score on an online test?

A five-round median of 250 to 300 ms is typical for a browser test, and under 250 ms is on the faster side — device latency included. Be skeptical of sustained scores far below 150 ms on a visual test; they usually reflect anticipation rather than genuine reaction.

Why is my online score slower than the 231 ms lab average?

Because your browser score includes your hardware. Display refresh, input-device latency, and operating-system compositing add roughly 10 to 80 ms depending on the setup — timing studies have measured it directly. The lab figure had that delay calibrated out; yours does not. Compare your trend over time on one device, not your absolute number against a lab mean.

Does reaction time really decline with age?

Yes, but slowly and later than most people assume. Simple reaction time is essentially flat from the 20s to about age 50, then rises by about 0.5 ms per year; choice tasks decline two to three times faster. Individual variation within any age group dwarfs the difference between adjacent decades.

Can I improve my average reaction time?

Within limits. Practice improves scores by about 13 ms before plateauing, sleep and alertness move the number more than any training does, and caffeine helps modestly when you are tired. What practice cannot do is push you below the physiological floor — expertise mostly buys milliseconds through prediction and consistency, not faster nerves.

Is this site's test a medical measurement?

No. This site is not a medical device, and no browser reaction test is a clinical instrument. Our score is a consistent browser measurement that includes your hardware's latency, and the averages on this page are reference estimates modeled from published research, not first-party data or a health assessment. If you are worried about your processing speed or cognition, see a qualified professional.

Will coffee before the test make my score better?

It can lower your score modestly — controlled studies put the effect at roughly 10 to 12 ms versus placebo — but mostly when you are tired; well-rested participants benefit less, and higher doses do not help more. Caffeine masks fatigue rather than fixing it, so it is not a substitute for sleep, and it will not move a well-rested score from average to elite.

Sources

• Woods et al. (2015), Frontiers in Human Neuroscience — calibrated lab mean of 231 ms (N=1,469), age slope of 0.55 ms/year, and the 200–250 ms young-adult range this article is modeled from.
• Wilkinson and Allison (1989) — decade differences in 5,325 subjects; the 20s as the fastest decade.
• Der and Deary (2006) — 7,130 representative UK adults; simple reaction time flat until about age 50.
• Fozard et al. (1994), Baltimore Longitudinal Study of Aging — simple reaction time +0.5 ms/year, choice tasks two to three times faster.
• Thompson, Blair and Henrey (2014), PLoS ONE — cognitive-motor decline detectable from age 24 in 3,305 StarCraft 2 players.
• Jain et al. (2015) — visual versus auditory versus tactile reaction times and cortical arrival latencies.
• Bridges et al. (2020), PeerJ and Anwyl-Irvine et al. (2020) — measured timing lag of browser-based testing, 8–80 ms depending on setup.
• Mohebbi et al. (2009) — phone conversation adds 148–221 ms to driver brake initiation.
• Best Reaction Test methodology — how our timer, randomized delay, and five-round median work, and where browser measurement ends.