Clever Quest Lab Tools

Reaction Time Test

2026-10-12

Check how fast you react to a colour change or to a short beep. A run has five valid tries, and the page shows your average, median and best time in milliseconds.

Reaction test

Try 1 of 5

    The time includes delays of your screen, input device and browser, and in sound mode of the audio output (wireless headphones add the most). Results differ between devices. This is not a laboratory or medical measurement.

    History on this device

    No runs saved yet. Finish a run of five tries and it appears here.

    Everything runs in your browser. Nothing you do here is sent anywhere; the history is kept only in this browser on this device, and you can delete it with “Clear history”. The page makes one calm colour change per try, with no flashing.

    How to use the reaction time test

    1. Choose Colour change or Sound. In sound mode, press “Play test beep” first and set your device volume to a comfortable level.
    2. Press the large area to start a try: click it, tap it, or press Space or Enter. It says “Wait...”.
    3. After a random pause of roughly 1.5 to 5 seconds the area turns green and says “Now! Press” (in sound mode a short beep plays instead). React as fast as you can with the same input; your time in milliseconds appears.
    4. Press again for the next try. Pressing before the signal gives “Too soon”; a reaction under 100 ms gives “Too fast”. Neither is counted, and the try is repeated.
    5. After five valid tries you see every try, the average, the median and the best time. “Try again” starts a new run.

    How it works

    Every try draws a new random waiting time between 1.5 and 5 seconds, so there is no rhythm to learn. In colour mode the page changes the colour inside a requestAnimationFrame callback, which runs just before the browser draws a frame, and starts the timer at the beginning of the next frame, close to the moment the new colour is first drawn. A web page cannot know when pixels really light up, so that screen delay stays in your result.

    Your press is read from the pointerdown or keydown event, the moment you press down, not the release. The event timestamp uses the same clock as the timer, so the reaction time is the press timestamp minus the start timestamp, rounded to a whole millisecond. Holding a key down counts once. In sound mode the beep (1,000 Hz, about 0.16 seconds, fixed moderate level) is scheduled on the audio clock about 80 ms ahead, and the start time is that scheduled moment; the delay of the audio output is not compensated.

    A press during the wait cancels the pending signal. A try is also cancelled if you hide the tab or press Esc, and discarded as “No reaction” if nothing happens within 4 seconds. The average is the sum of the five times divided by five; the median is the middle of the five sorted times; both are rounded to a whole millisecond.

    Examples

    Example 1: steady. Tries 245, 232, 261, 228 and 254 ms. The sum is 1,220, so the average is 1,220 / 5 = 244 ms. Sorted: 228, 232, 245, 254, 261, so the median is 245 ms and the best is 228 ms.

    Example 2: one slow try. Tries 310, 285, 402, 295 and 288 ms. The sum is 1,580, so the average is 316 ms. Sorted: 285, 288, 295, 310, 402, so the median is 295 ms.

    Example 3: sound mode. Tries 180, 175, 190, 172 and 183 ms. The sum is 900, so the average is 180 ms. Sorted: 172, 175, 180, 183, 190, so the median is also 180 ms.

    Example 4: rounding. Tries 231, 247, 238, 252 and 229 ms. The sum is 1,197, so the average is 1,197 / 5 = 239.4, shown as 239 ms. Sorted: 229, 231, 238, 247, 252, so the median is 238 ms.

    What reaction time is and what the test measures

    Simple reaction time is the time between a signal and the start of your response when you know what the signal will be and what to do. It adds up from several steps: the sense organ turns light or sound into nerve signals, the brain notices the signal and decides to act, a command travels to the muscles, and the finger presses. Choice reaction time, where different signals need different responses, is longer. The result here also includes the technical delays described below.

    Typical values

    In laboratory studies of healthy young adults, simple reaction time to a light is typically around 180 to 250 ms, and to a sound typically around 140 to 200 ms. Sound is usually faster because the hearing pathway typically needs less time to deliver the signal. Individual values vary a lot, and also from try to try in the same person.

    Online tests typically read higher by tens of milliseconds, because screen, input and browser delays are added to the human part. A colour test on an ordinary computer often lands between 200 and 300 ms.

    Why the number differs between devices

    Compare results only from the same device, input and headphones.

    Average, median and best

    The average uses every try but is pulled by an odd one, such as a try where you looked away. The median is the middle try, so it is steadier. If the two are close, the run was consistent; if the average is much higher, one slow try is the reason. The best time shows what is possible on your setup, but with five tries it is the least stable number.

    What affects reaction time

    How to get a consistent measurement

    Frequently asked questions

    What is a good reaction time?

    There is no single threshold. Typical online results for a colour change are roughly 200 to 300 ms, and sound results are often somewhat lower, but a faster or slower screen or input device changes that.

    Why is my sound result different from my colour result?

    The human part is typically faster for sound, but the audio output adds its own delay, small with wired headphones and large with wireless ones. The modes are not directly comparable.

    Why does the page say “Too fast”?

    A reaction under 100 ms is faster than people can react to an unexpected signal, so it is almost always a guess made before the signal. Such tries are not counted and are repeated.

    Can I improve my reaction time?

    A little. Being rested and focused and doing a few practice runs helps; large gains are unlikely.

    Is this a medical or scientific test?

    No. It is a simple self-test for curiosity and comparison. It cannot diagnose anything; if you are worried about your health, talk to a doctor.

    Is my data sent anywhere?

    No. The test runs entirely in your browser. The history of your last runs is stored in this browser on this device only, and “Clear history” removes it.

    Related tools