How to Improve Reaction Time

The changes that move the number without changing you

A faster display is the largest single improvement available to most people, and it is entirely equipment. Moving from a 60Hz panel to a 144Hz one takes several milliseconds off a measured score because the stimulus spends less time waiting to be shown, and nothing about your nervous system has been consulted.

A wired mouse polled at a high rate, or a keyboard with a shallow actuation point, does the same on a smaller scale. Wireless peripherals and, on the audio test, Bluetooth headphones move it in the other direction — and Bluetooth can add more delay than the entire quantity being measured.

None of this is cheating, but it is worth being clear about which category an improvement belongs to. If you want to compare yourself against a friend, compare on the same equipment; if you want to track yourself over time, keep the equipment fixed and the refresh rate shown with your result will tell you when it changed.

What genuinely helps, in the order the effects seem to matter

Sleep is the largest behavioural factor by a wide margin. Reaction time and, more noticeably, the consistency of reaction time both degrade with sleep loss, and the effect on the slow tail of your attempts is much bigger than the effect on your best round. Somebody who is tired usually still produces one or two fast rounds and a scattering of very slow ones.

Caffeine produces a small, real improvement for most people, on the order of a few milliseconds rather than tens. It is a genuine effect and a modest one, and it does not compound — a second cup does not buy a second improvement.

Practice on the specific task improves the specific task, quickly at first and then not much. Most of that early gain is learning where to look, how hard to press and how not to anticipate, rather than a faster nervous system, which is why it does not carry over to a test with a different stimulus.

How to stop losing time you are not losing to reflexes

Rest your finger on the button. Any distance the hand has to travel is added to the score, and on a mouse the difference between a hovering hand and a resting finger is measurable. This is also why keyboard scores tend to be slightly faster than mouse scores for the same person.

Look at where the change will happen rather than scanning for it. A stimulus that fills the panel needs no searching, but the eye still costs time if it is in the wrong place, and looking slightly away is enough to add a noticeable amount.

Do not try to predict the delay. Anticipation feels like it is helping and it produces exactly two outcomes: a discarded round when you go early, and a slower round when you have committed to a guess and have to correct. The randomised delay is there to make guessing unprofitable, and a relaxed wait genuinely beats a tense one.

What the honest ceiling looks like

Simple reaction time is not very trainable. It is close to a physical property — the conduction and processing time of a signal — and the range across healthy adults is narrower than the range of scores on any site like this one, most of which is equipment, attention and practice at the task rather than the underlying speed.

Where training does pay is in tasks with a decision in them. Choosing between two responses, tracking a moving target, or reacting to a specific stimulus among others all improve substantially with practice, because what improves is the decision rather than the reflex. That is a different measurement from the one on this site's reaction pages.

So the realistic goal is to remove the losses rather than to raise the ceiling: sleep, a resting finger, an unpanicked wait, and equipment you have not changed since the last time you measured. Do those and the number you get is the one you actually have.

This is a measurement toy, not a medical or psychological test. No score here says anything about your health, attention or cognition, and it is not a screening tool of any kind.

Questions

Can you actually train your reaction time?

Simple reaction time improves only slightly with practice, and most of the early gain is learning the task rather than getting faster. Tasks with a decision or a moving target in them improve considerably more, because what trains is the decision.

Does caffeine improve reaction time?

Modestly — a few milliseconds for most people, not tens — and the effect does not stack with more of it.

What is the fastest way to improve my score on this site?

A higher refresh rate display, followed by resting your finger on the button rather than hovering. Both are worth more milliseconds than anything you can do to your reflexes, which is exactly why the refresh rate is reported with your result.

Why is my reaction time so inconsistent?

Inconsistency is usually attention rather than speed, and it is the first thing sleep loss affects. Compare your median against your mean — a large gap means one or two bad rounds rather than a slow baseline.