How fast is your brain?

The tests

• Reaction Time TestClick the moment the panel turns green. Five rounds, median scored.~30 sec
• Aim TestThirty targets, one at a time. Average time and accuracy.~60 sec
• Number Memory TestA number flashes, then you type it back. One digit longer each round.~60 sec
• Sequence Memory TestSquares light up in order. Repeat the pattern as it grows.~60 sec
• Visual Memory TestSquares flash all at once. Find every one, in any order.~60 sec
• Chimp TestNumbers flash, then hide. Click them in order.~60 sec
• Verbal Memory TestWords appear one at a time. Call each one SEEN or NEW.~60 sec
• Color Match TestOne tile is a slightly different shade. The gap shrinks every level.~60 sec
• Focus TestFind 1 to 25 on a scrambled grid, in order, as fast as you can.~40 sec
• Stroop TestPick the color of the ink, not the word it spells.~30 sec
• 10 Second ChallengeThe clock hides after 3 seconds. Stop it at exactly 10.00.~40 sec
• Typing Speed TestCopy the sentence for 30 seconds. Only correct letters count.~30 sec
• F1 Reaction TestGo the instant the five red lights go out, like a real Formula 1 start.~30 sec

▶ Full Brain Test Five brain areas in one sitting, with a combined score out of 1000.

Guides

• ≶What is a good reaction time?Every score from under 150 to over 360, on the screen scale and the lab scale, with the badge each one earns.scores~9 min
• ⏲How accurate is a browser reaction time test?A browser adds roughly 80 ms nobody spent. Where it comes from, and what we measured while building this site.measurement~7 min
• ⌛Average reaction time by ageLab studies say 0.55 ms a year, browser data says 7. Why both are right, and what to expect at your age.age~7 min
• ⇄Reaction time on a phone vs a mousePhones add 35 to 140 ms of screen and touch delay. Which score to trust, and how to compare fairly.devices~6 min
• ↗How to improve your reaction timeSleep, time of day, warm-up and hardware do most of the work. Supplements do not.practice~7 min
• ⚖Do brain training games actually work?You get better at the game itself. Whether that carries over to anything else is still argued over.evidence~8 min
• ⌫How we screen leaderboard nicknamesA slur got through our filter. What replaced it, and why over-blocking costs more.moderation~8 min

Each guide names its sources, and anything we measured ourselves is marked as ours. See all guides.

How to read your score

Each result comes with a grade (from needs practice up to superhuman), a one-line verdict, and your history on this device. Every grade line has a stated reason, which you can find in the grade table on that test's page. Add a nickname and the score goes on that game's leaderboard.

Each finished run also sends one anonymous record to our server: the test, the difficulty, the score, touch or mouse, the language, the date and whether it was the daily. Aim runs add accuracy, and reaction runs add the number of voided clicks. There is no account and no identifier.

The raw number also moves more than people expect. A browser adds roughly 80 milliseconds that nobody spent. A 60 Hz screen costs about 8 more than a 144 Hz one, and your own runs scatter by around 40 milliseconds. So the comparison that holds up is you against your own history on the same device. That is what the score panel shows you.

What we measured while building this

The figures below come from building this site, not from a research paper. We share them because nobody can check a test that does not say how it measures, and most test sites do not say.

Timing starts the moment your finger or mouse button goes down. It does not wait for the click event, which fires after you let go. A reaction score is the median of five rounds, not the average.

These two choices alone change the numbers. A site that times the click event and averages every round will give the same person on the same hardware a different result. That is why comparing millisecond scores across sites tells you very little.

A click within 100 milliseconds of the green signal is marked invalid. Track and field uses the same line to call a false start. Anything slower counts, whether it was a fast reaction or a lucky guess. A machine cannot tell those two apart, so this test only filters out times that are physically impossible.

The wait before the signal is random, anywhere from 1.5 to 7 seconds. With a narrow window, people can tap to a rhythm, and the score stops showing real skill. Making the window wide worked better than trying to catch rhythm tapping afterward.

We do not subtract hardware delay from your score. Studies put the delay from browsers and devices at roughly 71 to 87 milliseconds, and it changes from one setup to the next. One flat correction for everyone would claim a precision nobody has. The research figures are on the methodology page for reference.

Browsers slow down timers in background tabs, enough to ruin a measurement. When we tested with the tab hidden, rounds that normally take 2 seconds ran past the 25-second limit. So every timed test here refuses to start while the tab is hidden and tells you why on screen. We think that beats quietly saving a wrong number.

On a 320-pixel screen, a 6x6 grid cannot give every square a 44-pixel target. So square sizes are worked out from the panel's actual width. When the sizes used to be fixed, squares overlapped their neighbors by 3 pixels, and a tap near an edge hit the wrong tile. That bug is why grid sizing now lives in one file.