Blog
11 guides on what a reaction-time number actually is, where it comes from, and how to read one without fooling yourself. No filler, no invented statistics, and no claim on this site that cannot be traced toa source we publish or to our own documented method.
Measurement
4 articles
How a number from a browser gets produced, and what it can and cannot tell you. Apparatus effects, timing precision, latency, and why browser scores are not laboratory scores.
Interpreting scores
3 articles
What a reaction time actually means once you have one — reference ranges, statistical expectations, and the difference between a change you caused and a change you drifted into.
Test design
4 articles
The scoring ladders, ladders, and rules behind the individual tests, written out so a score can be interpreted rather than merely reported.
Under 250 ms is quick and under 200 ms is genuinely fast on a browser reaction test. See the full scale, why device lag matters, and how to get a fair score.
The same task, measured two ways, produced a 65 ms gap in published work. That difference is larger than most effects people test for — here is where it comes from.
A median of five real clicks, subtracted from every result. What that measures, what it cannot reach, and how to get a clean latency reading on your own setup.
The rank on your result screen is computed from an assumed distribution, not measured from real people. Here is the arithmetic, and why a single percentile should not be read literally.
Serial testing data shows a real 13 ms gain that almost entirely arrives in the first two sessions. What that means for discarding your first attempt.
A randomised 1–3 second delay is what separates a measurement from a guess. How the too-early detector works, and the corner case a pre-timed press can still get through.
Six real limitations of browser-based timing, how large each one is, and why that does not undermine the one comparison that matters: you against yourself.
Jitter, butterfly, and drag clicking — how each produces its clicks, why peak CPS diverges from average, and why some of them are banned.
Miller argued in bits, not digits. What a number memory test actually scores, where the wall arrives, and how chunking moves it.
Your fastest attempt is mostly luck. What averaging five does to the error, why outliers stay in, and when the difference becomes detectable.
Only correct trials count, which is what stops guessing from looking fast. Plus why Hick’s Law is the reason this mode fixes four colours.
Written by the ReactionTimeTest team — the same people who maintain the engine. No named individual author and no reviewer we do not have.
Every number is traceable. If a figure comes from research, it links to the source on the sources page. If it comes from this site's own code, it is documented on themethodology page. Where the site makes a judgement call — a rating band, say — it is labelled as this site's scale rather than dressed up as a research finding.
No invented authors or reviewers. This site has no named author and no medical or scientific reviewer, so none is credited. The publisher is the organisation.
No fabricated data. We do not publish a population median, because your scores stay in your browser and we never aggregate them. There is no dataset to average, and we would rather say so than invent one.
Corrections are published. When a claim here turns out to be wrong, it is corrected on the page and noted in thesources page.Tell us if you find one.