Go / No-Go Test
Green: tap. Red: hold. Slips cost lights; the stopwatch decides.
The ladder
Bar widths show roughly how many people land in each band on the provisional curve. Every run waits the same scheduled total between lights, so the clock compares players, not luck.
What moves it
Restraint. Three lights in four are green, so your finger learns to go. A red is the trap, and a slip costs about nine seconds: the red comes back, and four green lights come with it.
Pace. The clock never stops, so a careful-but-sluggish run pays on every green. Speed and discipline are priced in one number, biathlon-style.
Your hardware. Eighteen taps on a clean board, four more for every slip, and each one carries your screen and input lag, so a phone reads slower than a wired mouse. Compare yourself to yourself on the same setup.
What this measures
A go/no-go task measures response inhibition: whether you can stop an action your body has already prepared. Three lights in four are green and want a fast tap, so a rhythm builds in your finger; then a red arrives and the loaded tap has to be cancelled in flight. Psychology has run this design for a century to study impulse control, and the slip, the tap that escapes on a red, is its signature measure.
Here the judge is a stopwatch. The board deals 24 lights, six of them red, exactly two in every eight so no stretch of it is safe, and it does not clear until every light is resolved. A slip sends its red back into the lights you have not seen yet and deals four extra greens with it, about nine seconds of consequence; a green you sleep through simply comes back around, its cost already paid in the waiting. Go and no-go share one 1.2 second window, so hand speed cannot buy back a lack of restraint.
The result prices pace and discipline in one number, biathlon-style: seconds on the clock, penalties folded in. The starting curve is derived from the machine's own timings and an assumed foul rate, with no published norm behind it, which is the most honest label it can carry; the Method page shows the derivation and the measured count.
Questions people ask
A slip count can be farmed by playing slow, and a go-speed average can be farmed by tapping everything. The stopwatch prices both: playing slow loses to the clock directly, and tapping everything buys a few milliseconds per light while costing about nine seconds per red, on a board that never clears until all six reds have been held.
Because they are not the same mistake. A slip is the thing a go/no-go test measures: your finger went when it was told to stay. It puts that red back on the board and adds four green lights, about nine seconds. A miss is a lapse, and the clock has already charged it: the green sat there for its 1.2 seconds, then went back on the board to be answered later, about two and a half seconds in all. The penalty lights are all green, so a fine never breeds a red, and they are dealt back into the lights you have not seen yet rather than piled on the end.
A clean board on this design lands just under fifty seconds for most people, and most runs are not clean: a missed green adds about two and a half seconds and a slip about nine. The curve behind the percentile is provisional until humanlag has its own runs, and everyone plays in one pool, even though the taps carry your hardware's lag.
The behavior is the same: frequent go signals, rarer no-go signals, and the slips on no-go as the measure of inhibition. Labs typically score a slip percentage across a fixed set of trials; this version folds the same behavior into a clock, because a stopwatch is harder to game and better to race. The reference paradigms give both signals the same response window, and so does this board: one 1.2 second window for green and red alike.
It starts as an assumption worth 300 runs, shaped from published norms, and every run played here adds one to it, everyone in one pool. The Method page says which norms, how the percentile is read, and how many runs this test has measured so far.