Average Reaction Time

What is the average human reaction time?

It depends on what you react to and what records the response. For simple reaction time (one signal, one response, no decision), the classic averages for college-age adults are about 190 ms for a light, 160 ms for a sound and 155 ms for touch, from Robert Kosinski's literature review. A 2015 study of 120 students also found auditory reactions significantly faster than visual ones.

Calibrated modern equipment agrees: 1,469 adults aged 18 to 65 averaged 231 ms on a visual task, or 213 ms after removing 17.8 ms of monitor and mouse delay. Human reaction time rises when you must choose or are surprised, so an average reaction time in ms needs its task attached.

Why online reaction time tests read slower

Browser tests read higher than the lab, and the gap is mostly equipment. Human Benchmark, with over 81 million clicks recorded, reports a median of 273 ms and a mean of 284 ms. testreaction.com reports a median of 295 ms and blames the screen, the mouse or trackpad, and the page itself. Each link adds time:

• Display. A 60 Hz screen draws a frame every 16.7 ms, so a color change can wait up to a frame to appear. A lab LCD monitor added 11.0 ms in the Woods study.
• Input device. Even a 1,000 Hz gaming mouse added 6.8 ms there. Human Benchmark notes that phone taps tend to be slower than clicks.
• Operating system and browser. Human Benchmark puts typical desktop and laptop lag at about 30 ms. A robot-actuator study of four web experiment platforms on consumer keyboards found responses logged 71 to 87 ms late on average.

Lab results vary too. Large computerized studies report mean visual simple reaction times from 233 ms to nearly 400 ms, and Woods and colleagues note that hardware and software can add up to 100 ms. Compare a browser score with browser medians, not with 190 ms.

Reaction time by age

These means from Woods and colleagues (2015) come from about 1,470 adults on the same calibrated equipment, before their 17.8 ms hardware correction. Age bands are as published.

Simple reaction time rose about 0.55 ms a year and choice about 2.8 ms. The Baltimore Longitudinal Study of Aging found 0.5 ms a year for simple auditory reaction time from about age 20. In 7,130 UK adults, Der and Deary found simple reaction time changed little until around 50, while choice reaction time slowed throughout adulthood.

Ages 12 to 15. No large dataset we found gives a per-year average for 12 to 15 year olds on a comparable test, so we print none. Among 5,325 exhibition visitors grouped by decade, people in their 20s were fastest on average, teens slower and under-10s slower still. A review of 72 studies found speed improves quickly in childhood, then more slowly through adolescence.

Simple vs choice reaction time

Our test measures simple reaction time. In a choice task each signal has its own response, so you identify it before moving. Hick's law (Hick, 1952) says choice reaction time grows with the logarithm of the number of equally likely options: doubling the options adds a fixed step rather than doubling the time.

The West of Scotland Twenty-07 study, ages 16 to the early 60s, averaged 328 ms on a simple task and 638 ms on a choice task. Woods's participants averaged 231 ms and 550 ms.

Reaction time in driving

Braking means noticing a hazard, recognizing it and moving a foot, so it takes far longer than a click. Marc Green's 2000 review estimated about 0.70 to 0.75 seconds when a driver expects the signal, 1.25 seconds for unexpected but common signals such as brake lights, and 1.5 seconds for surprises such as an object entering the lane. Expectation alone changes the time by a factor of two.

US road design allows more. Washington State DOT's Design Manual, citing AASHTO's Green Book, bases stopping sight distance on 2.5 seconds of travel at the design speed and calls the method very conservative. At 60 mph that is 220 feet before braking starts.

Reaction time in gaming and sports

Formula 1 says drivers launch about 0.2 seconds after lights out, on average. Before that, Formula1.com describes a random pause of about 0.2 to 3 seconds (the rulebook sets no range). A race start measures something different from a lab reaction time: the launch means feeding out the clutch while applying throttle, not pressing a button, so the two numbers do not compare directly. Try the sequence on our F1 start test page.

In sprinting, World Athletics start systems flag a possible false start below 0.100 second. The measurement matters: an apparent drop in Olympic sprinters' minimum reaction times from 2004 to 2016 was most likely caused by lower force thresholds in the starting blocks, not better training.

In gaming, a study of 3,305 StarCraft 2 players aged 16 to 44 found in-game response times begin slowing at 24, regardless of skill. Human Benchmark adds that a fast computer and a high-frame-rate monitor help, and some TVs add up to 150 ms.

What is a good reaction time?

Judge a score against the same kind of test. For browser tests, published distributions give these bands:

• Under 243 ms: faster than about three in four testreaction.com results. Around 200 ms in a browser is fast, since the device adds time.
• 243 to 361 ms: the middle half of results there. The medians are 295 ms on testreaction.com and 273 ms on Human Benchmark.
• Over 361 ms: slower than about three in four results. A phone tap, a TV screen or a distracted try often explains part of it.

Our own test labels a five-attempt average with fixed cutoffs at 200, 250, 300 and 400 ms, so the testreaction.com lower quartile of 243 ms reads as Fast there, its median of 295 ms as Typical for an online test and its upper quartile of 361 ms as A little slower than typical.

These bands describe results, not people. Take several tries on our reaction time test and judge your median rather than your single best click.

The fastest reaction time

There is no official record for the fastest human reaction time. Sprint starts give the best-measured floor: the signal is a sound, which people react to faster than light. World Athletics treats anything under 0.100 second as a possible false start, and Devon Allen was disqualified from the 2022 World Championships 110 m hurdles final for a 0.099 second reaction.

That line is debated. Pain and Hibbs found 20 percent of sprint starts in two conditions under 100 ms, and concluded the neuromuscular part of an auditory reaction can be under 85 ms.

A sub-100 ms click on a visual test is different. Reactions to light are slower than to sound, and a typical computer adds about 30 ms, so a result that low almost always means the click began before the screen changed. That is anticipation, not speed.