Why Practice Improves Your Score, Then Stops

This site used to claim that repeated attempts strengthen the neural pathway between stimulus and movement. That was an overstatement, and the published data says something more interesting.

Practising a reaction time test does make your score better. But the improvement is front-loaded, and it plateaus almost immediately. Here is the actual data.

The study

Forty-three participants completed ten sessions of a ruler-drop reaction time task over five weeks, ten trials per session. Each session mean is reported below. These are the paper's numbers — we did not run this study and we have no comparable dataset of our own, because your scores never leave your browser.

What the numbers show

Source: Del Rossi, Malaguti & Del Rossi (2014), Journal of Athletic Training 49(3), 356–359.

Three things to take from it:

1. Session 1 to session 2: -7.2 ms faster. Notice that this single step did not reach statistical significance on its own — the paper marks it as a non-significant comparison. Even the largest visible improvement in the series is within noise.
2. Session 1 to session 10: -12.9 ms in total. That is a real and measurable change, and it is the entire size of the effect.
3. Sessions 2 through 10: they span only about 12.9 ms end to end, and no two consecutive sessions differed significantly. Whatever was improving stopped improving by the second session.

Why the first session is slower

The authors' explanation is visual feedback: you see how the task is supposed to go, and you learn to recognise which of your own responses were the slow ones. The benefit is knowledge about the task, not a change in nerve conduction. One study of a related task put it more bluntly — the improvement comes from learning to avoid your own slow responses, not from reacting faster in any physiological sense.

This is also why the same effect appears inside a single sitting. Your first attempt is the unfamiliar one. Attempts two through five are largely noise around whatever your actual baseline is.

What we changed on this site

The previous version of our tips section said: "Repeated attempts strengthen the neural pathway between stimulus and movement." That was not supportable, and it is now replaced with an instruction that the evidence does support — discard the warm-up and measure from the second attempt. Ourimprovement tips now lead with it, and the same point appears in the FAQ on this article's subject.

We kept the finding that aerobic fitness is associated with faster simple reaction times, because that has reasonably consistent support behind it — while being explicit that it is a months-long change in baseline rather than something a session of clicking produces.

A practical protocol

1. Do one attempt and discard it.
2. Run five attempts. Average those.
3. Repeat that whole sequence on the same setup at a similar time of day, over several days.
4. Compare your own numbers across sessions. Expect the first real improvement in week one, then a plateau — and treat anything after that as measurement noise rather than progress.

If you are trying to detect a change of a few milliseconds, note that the session-to-session spread in that study was roughly ±17 ms. A 5 ms difference between two days is not distinguishable from ordinary variation. This is why the site stores your history and shows you the trend rather than pushing you toward a single number.

Source: Del Rossi, Malaguti & Del Rossi (2014), Journal of Athletic Training 49(3), 356–359, 43 participants, ten sessions over five weeks. All session means quoted here are the paper's. This site ran no study of its own and publishes no aggregated user data.