About ReactionTestPro
ReactionTestPro is a small, free collection of 10 browser-based tests for reaction time, attention, memory, vision and hearing. No accounts, no paywall, no data collection.
Last updated:
Who runs this site
This site is built and maintained by [PLACEHOLDER: your name, or the business name you publish under].
[PLACEHOLDER: two or three sentences about you: your background, what you do, and why you started building reaction tests — real experience only, no invented credentials]
A note on expertise
This site is built by a developer, not a clinician. The screening tests are based on published, publicly documented instruments and paradigms, and each one names its source on its own page — but none of them is administered, reviewed or validated by a medical professional here. Where a page carries a reviewer byline, that person is named explicitly; where it does not, none exists.
Why it exists
Most online reaction tests are either buried in ads, demand an account before showing your result, or quietly send your answers somewhere. This site was built as the opposite: open the page, take the test, see the number, close the tab. Nothing is stored anywhere except your own browser, and only if the test keeps a personal best.
The second reason is that reaction time is genuinely interesting. It sits at the boundary between perception and movement, it varies with sleep, age, caffeine and attention, and it is one of the few cognitive measures you can meaningfully test on a laptop.
How the tests are built
Every test is plain client-side JavaScript. There is no backend, no database and no API call — the code that scores your test is the code your browser downloaded.
- Timing uses
performance.now(), a high-resolution monotonic clock, sampled at the moment the stimulus is painted and again on your input. Animations are driven byrequestAnimationFrameso they stay aligned with your display's refresh cycle. - Delays are randomised before each trial, so you cannot learn the rhythm and anticipate the signal. Reacting before the signal is detected and rejected as a false start rather than counted as a fast time.
- Audio is generated with the Web Audio API — tones are synthesised in the browser rather than downloaded, and no microphone access is ever requested.
- Screening tests follow the structure of published, publicly documented instruments, cited on each test page. They reproduce the format, not a clinical calibration.
The honest limitation
Your measured time always includes your hardware. Display latency, refresh rate, browser scheduling and input device all sit between the signal and the number you see, and can easily account for 20–50 ms of difference between two setups. That is why this site shows ranges and tiers rather than claiming precision it cannot deliver, and why comparing your own results over time on the same device is more meaningful than comparing against anyone else.
What this site will never do
- Ask you to create an account to see your own result.
- Send your test answers to a server. They are computed and discarded in your browser.
- Publish invented statistics, fabricated leaderboards or made-up user counts.
- Claim a test result is a diagnosis.
The full detail is in the Privacy Policy and the Terms of Use.
Get in touch
Corrections are welcome — if a source is wrong, a test behaves oddly on your device, or a claim looks overstated, please say so on the contact page.