Average reaction time

The spread is the number that tells you how much people differ. It is the usual distance between one person and the average. If the spread were tiny, everyone would react at almost the same speed; here it is a few hundredths of a second, so two adults can easily differ by that much and both be completely ordinary.

The study measured simple reaction time: one signal, one answer, no choice to make. A target showed up to the left or right of the middle of the screen, and each person pressed the button of a gaming mouse as quickly as they could. They did many practice and test tries, and each person's own average counted once. The time between the target appearing and the button press is the reaction time.

That makes it a measure of how fast you answer something you see, not of how fast you think. It does not say how well you will drive, play a game or catch a ball, because those jobs need choices, planning and practice on top of the plain answer.

The figure in the table is the one the authors printed, with the delay of their own screen and mouse still in it. They also give a second, lower figure that takes that delay out, so the two should not be mixed up when you read about the study elsewhere.

The study used a calibrated set-up built for timing. A web page runs on whatever you own: a screen that lights up a moment after it is told to, a touch screen or mouse that takes a moment to report a press, and a browser that stamps the time slightly late. A page cannot see any of these delays, and they differ from device to device.

Two studies measured this chain with robots that pressed everyday computers and phones the moment the screen lit up^[2][3]. Beyond the delay of the study's own equipment, a keyboard set-up adds 44.1 ms to 115.1 ms, and a touch screen 39.8 ms to 52 ms. No study pressed a mouse, so this site gives a mouse the keyboard's figures, and a pen the touch screen's. That is why a result of the reaction time test is set beside the study as a range, on the quicker side of the time you see, and never as one exact rank. On a computer the range is wide, so it is only a rough guide. How the clock is read, and what a browser cannot see, is written out on the methodology page.

Age is the one effect the study measured: reaction time grew longer as people got older^[1]. Beyond that, the usual suspects are tiredness, how well you are paying attention, whether you expected the signal and how steady your hand is on the device. Your own times will move from one day to the next, and a single try can be far off your usual speed. The guide on how to improve reaction time says what practice can and cannot change.

• The device: a phone, a laptop with a mouse and a keyboard each add their own delay, so compare like with like.
• The signal: a light that you wait for is answered faster than one you do not expect.
• The number of tries: the middle of several tries is steadier than one lucky or unlucky try.
• The day: a rested morning and a tired night can differ more than a week of practice.

Human Benchmark, a popular website, reports a median of 273 ms (reported by the site)^[4] over many millions of clicks. That is a different kind of number from the study's. It counts clicks and not people, so someone who plays a hundred times counts a hundred times, it mixes every kind of phone, screen and mouse, and it has not been through the checking that a published study goes through. We show it as a mark and never as the thing you are compared with.

That is also why the study is the one this page is built on: it says who was measured, how, and how many. A good time for you is a time near your own best on the same device, and the study tells you where adults stand around it.