How to improve reaction time

Your reaction time is made of several steps: seeing the signal, deciding to act, sending the command to your hand and pressing. The body part of that chain is mostly fixed for you. What you can change is how ready you are when the signal comes, how smoothly you press, and how much delay comes from your device and not from you.

So the useful goal is not a record. It is a steady, fair result that you can compare with your own earlier ones, and that shows you whether a change in how you take the test did anything.

• Be rested. A tired mind is slower, and rest is the cheapest thing you can do for your results.
• Use one device every time. A phone, a laptop with a trackpad and a desktop with a mouse each add their own delay, so a change of device looks like a change in you.
• Sit comfortably with the screen straight in front of you and your hand already on the button or the mouse.
• Close other tabs and programs so that the page has the computer's attention and does not stall at the wrong moment.
• Take the test at about the same time of day when you compare days.

1. Open the reaction time test and use its first try, which does not count, to settle in. It is there for exactly that.
2. Take the 5 tries of a run with your eyes on the middle of the screen, and press as soon as you see green, not before.
3. Stop after a few runs. Short sessions on several days beat one long session, because attention and care fade as a session drags on.
4. Look at your middle result across the week and not at your best single try, which is mostly luck.
5. Change one thing at a time, such as your seat or your hand position, and see whether the middle result moves.

Practice at a task usually makes that task quicker, with the biggest gains early on and then a long flat stretch. Do not expect practice on this test to make you quicker at a different task, such as driving or a game with choices in it, because each of those is a different skill.

The most common way to spoil a result is to try to predict the green. If you press when you think it is about to appear, you will sometimes be early, and sometimes a lucky guess will look like a very fast reaction. The test treats a press that comes sooner than 100 ms after the signal as a guess and does that try again, and a press before the signal is done again too.

Because the wait before the signal changes every time, between 1.5 sec and 4.5 sec, a steady rhythm of guesses does not work. React to what you see. Staying relaxed and watching the whole screen works better than leaning forward in tension.

Part of the delay between the signal and your result belongs to the machine. A 144 Hz screen shows a new picture sooner than a 60 Hz screen, and a 1000 Hz mouse tells the computer about your press sooner than a 125 Hz one. Neither makes you quicker. They remove delay that was never yours.

You can look at your own screen and mouse with the refresh rate test and the polling rate test. A web page cannot see the rest of the chain, which is why a result online reads a little higher than a laboratory figure, as the page on the average reaction time explains.

In the cited study, reaction time grew longer with age^[1]. The change from one year to the next was small, but over decades it adds up. If you want to know whether you are improving, compare yourself with your own results from last month, not with someone much younger or older.

Your results will go up and down from day to day, whatever you do, and a run that is quicker than your usual is often just a good moment. Judge a change by the middle of several runs on several days. If a week of steady practice moves that middle a little, you have learned the task. If it does not move, the numbers are showing you your normal range, which is useful to know too.