What this measures

Simple reaction time — how long it takes you to respond once a stimulus appears. The screen sits red, turns green at an unpredictable moment, and you tap. Five rounds; your score is 1000 − average milliseconds, so a 250ms average gives 750.

The delay before the green is randomised precisely so you cannot anticipate it. Tapping early is caught and voids that round.

What counts as fast

Simple visual reaction time in healthy adults typically lands around 200–250ms. That figure is remarkably stable across the research literature, and it is not mostly muscle — the bulk of it is signal transmission and decision, with only the tail end being movement.

Under 200ms — exceptional, and worth double-checking you did not anticipate the change
200–250ms — fast
250–300ms — typical range
300–400ms — slower, very commonly explained by fatigue or input lag
Over 400ms — check your setup before concluding anything

Your equipment is part of the measurement

This is the thing most people miss. A meaningful chunk of what you measure here is not you.

Display refresh rate — a 60Hz screen adds up to 16ms of latency before the green even appears. A 120Hz phone is measurably "faster" with the same person tapping.
Touchscreen sampling — touch panels poll at their own rate, adding several more milliseconds.
Mouse vs touch — different input paths, different latency.
Browser and background load — a busy tab adds jitter.

So comparing your number against a stranger's is close to meaningless. Comparing your own number on the same device across days is genuinely informative.

What actually moves your reaction time

Sleep — the single largest factor. One short night reliably slows simple reaction time, and the effect is large enough to see in five rounds.
Time of day — most people are slowest shortly after waking and in the early hours.
Caffeine — a modest, real, short-lived improvement.
Alcohol — a large and reliable slowdown.
Age — a gradual increase across adulthood, well documented and unremarkable.

Because it is so sensitive to sleep and fatigue, this test doubles as a decent daily condition check. Run it at the same time each day for a week and the pattern usually maps onto how well you slept.

Getting a better score honestly

Watch the whole screen, not one spot. Peripheral detection of a full-screen colour change is fast; hunting for a specific point is slow.
Rest your finger on the surface so the only cost is the press.
Do not try to predict the timing. Anticipation either voids the round or, worse, gives you a fake number.
Stay loose. Tensing up in advance makes people slower, not faster.

Note that practice effects here are small and plateau quickly. Unlike task-based games, there is not much to learn — which is exactly what makes it a useful condition check.

Frequently asked

Q. Why five rounds?
A. One round is dominated by luck. Averaging five smooths it out without making the test tedious. If you want a more stable number, run it three times and take the middle.

Q. I tapped early and it voided the round.
A. Intentional. Without that check, the fastest strategy would be guessing, which measures nothing.

Q. Is this the same as reflexes?
A. No. A reflex bypasses the brain — this involves perception and a decision, which is why it is much slower than, say, a knee-jerk response.

For a broader picture across five cognitive areas, try the full cognitive test.