Reaction Time Test
The classic reaction time test: wait for the red panel to turn green, then click, tap or hit space as fast as you can. Five trials, timed on the browser high-resolution clock and scored against a reference distribution.
How to use Reaction Time Test
- Choose how many trials you want — three for a quick check, five for the standard benchmark, ten for a steadier average.
- Press Start, then wait while the panel is red. The delay is randomised between 1.5 and 5 seconds so it cannot be anticipated.
- The moment the panel turns green, click, tap or press space. Reacting before it changes counts as a false start and that trial is repeated rather than scored.
- Read your mean, median, best, worst and standard deviation, then compare the mean against the percentile bar.
About Reaction Time Test
Simple visual reaction time is the shortest path from photon to muscle: light hits the retina, the signal travels to the visual cortex, a decision is made that something changed, and a motor command goes back out to the finger. Around 80 to 120 ms of that is pure signal transmission and cannot be trained away. The rest is attention — whether you were genuinely watching the panel or had drifted off — which is why this test is closer to a vigilance measure than a fitness measure.
The design here follows the Psychomotor Vigilance Task used in sleep research: a random foreperiod so the stimulus cannot be predicted, explicit detection of anticipatory responses, and reporting of the whole distribution rather than a single number. False starts are shown but excluded from scoring, because a test that lets you guess is measuring luck. Timing uses performance.now(), a monotonic clock unaffected by system time changes, and the press is captured on pointer-down rather than click so the release stroke is not counted.
Treat the absolute millisecond figure as an upper bound that includes your display and input hardware. What travels well between sessions is the shape of your results: the gap between your best and your mean, the standard deviation, and how those move when you are tired, caffeinated or ill. Repeated runs stored on this device make that trend visible without ever sending a number off your machine.
Frequently asked questions
What is a good reaction time?
For a simple visual task with a mouse, the median adult sits at roughly 270–280 ms. Anything under 250 ms is quick, under 220 ms is genuinely fast, and consistent times below 200 ms usually mean a high-refresh display and a low-latency mouse as much as a fast nervous system. Above about 350 ms is slower than most people, and above 500 ms is worth repeating when you are properly awake.
Why is my reaction time slower here than on other tests?
Absolute numbers differ between sites because of how each one starts and stops its clock, and because your display adds latency. A 60 Hz screen adds up to 16.7 ms just waiting for the next frame, and input plus compositing typically adds another 10–25 ms. The offset is fairly constant for a given setup, so comparisons between your own runs stay reliable even when the absolute figure does not match another site exactly.
Does the device I use change the result?
Considerably. A wired mouse on a 144 Hz monitor can be 30–40 ms faster than a phone touchscreen for the same person, because touch panels sample at a lower rate and add gesture-recognition delay. Compare desktop runs with desktop runs and phone runs with phone runs.
Can you actually improve your reaction time?
A little. Practice, being properly rested, moderate caffeine and warming up all shave a few tens of milliseconds, and reducing display latency helps further. The underlying conduction speed of your nervous system is largely fixed and declines slowly with age, so the realistic goal is consistency — a low standard deviation — rather than an ever-lower minimum.
What does the standard deviation tell me?
It measures how much your trials varied. A low spread means you were reliably attentive; a low mean with a high spread usually means one lucky anticipation and several lapses. Sustained-attention research treats variability as more informative than raw speed, which is why it is reported here alongside the mean and median.
Is my result sent anywhere?
No. Every trial is timed and scored inside your browser, and your attempt history is stored in local storage on this device only. Nothing is uploaded, and clearing your browser data clears the history.
Related tools
- Verbal Memory Test — Seen it before, or is it new?
- Chimp Test — Beat the chimpanzee benchmark
- Aim Trainer — Hit 30 targets, timed to the ms
- Brain Fog Test — Objective Brain Fog Index from 10 timed tasks
- Typing Speed Test — WPM, accuracy and consistency
- Number Memory Test — How many digits can you hold?