How to Play
One target is on screen at a time. It appears dim, drifts across the play area, and only becomes hittable once it "readies up" and brightens — that takes somewhere between 0.35 and 0.65 seconds, and you never know which. Tap it, or press Spacebar, once it's bright and the hit counts. Each target only lives about 1.25 seconds from the moment it spawns, so the window is short and the target is moving the whole time. The clock starts at 12 seconds and every hit adds 0.45 seconds back, up to a 20-second ceiling, so a good run feeds itself — but the cap means you can never bank a comfortable lead and coast. Hitting a target before it's ready costs 0.8 seconds. Letting one expire costs 0.6. Press Enter or Space to start, and again to restart after a run. The run ends when the clock reaches zero, and your best hit count is saved in your browser.
Tips
- Guessing is worse than missing: an early tap costs 0.8s against 0.6s for letting a target expire. On this clock, anticipating is literally the most expensive thing you can do.
- The ready delay is re-randomized on every spawn, which is deliberate — it makes rhythm-based anticipation useless and forces you to genuinely respond to the brightness change rather than to a beat.
- Targets drift and bounce off the edges, so aim for where the target will be when your finger lands, not where it was when you decided to move.
- Target radius varies from about 26 to 40 pixels. On the small end you're fighting precision as well as speed, so park your pointer near the middle of the play area between hits.
- Judge yourself over a run of many attempts, not one. A single reaction measurement is mostly noise; the useful signal is how rarely you have a bad one.
What Counts as a Normal Reaction Time?
Simple visual reaction time — one stimulus, one response, nothing to decide — generally lands somewhere in the range of roughly 200 to 300 milliseconds for healthy adults, with well-rested young adults clustering nearer the fast end of that band. Sounds and touch tend to be processed slightly faster than visual cues, which is one reason a test built on a color change reads a little slower than one built on a beep.
That figure isn't a single measurement of "how fast your brain is." It's a sum of stages. Light reaches your retina and is converted into a signal; the signal travels to visual cortex and is recognized as a change; a motor plan is selected and released; and finally muscle actually moves. Each stage costs time, and the motor portion alone accounts for a substantial share of the total — which is why the same person tested with a finger, a foot, and a voice response produces three different numbers.
Anything that adds a decision makes it slower. Choosing between two possible responses is reliably slower than having only one, and reaction time grows roughly with the logarithm of the number of alternatives — a relationship known as Hick's law. This game stays close to a pure simple-reaction task: one target, one response. But the wait-for-ready rule adds an inhibition component, since you have to withhold a response you have already prepared. Withholding and then releasing is a different, slightly slower operation than simply firing, which is why scores here sit above what a bare click-when-green tester reports.
What Actually Moves Your Number
- Sleep. The largest and most consistently documented factor. Sleep loss raises your average, but the more striking effect is lapses — occasional very slow responses on a task you would normally handle instantly. If your run is mostly sharp with a few inexplicable misses, that pattern says more than your average does.
- Age. Reaction speed tends to peak in early adulthood and then decline gradually across the decades. The change is real but small per year, and in the short term it is comfortably outweighed by how tired you happen to be today.
- Caffeine. Moderate doses can shave a little off reaction time and, more usefully, blunt the slowing caused by fatigue. The effect is modest next to simply sleeping properly, varies a lot between individuals, and is reduced if you already consume a lot of it.
- Attention and arousal. Performance follows an inverted U. Too relaxed is slow; too keyed-up is also slow, and tends to produce early taps rather than late ones. Knowing when a stimulus is coming helps enormously, which is exactly why the randomized ready delay is what keeps this test honest.
- Everything else. Alcohol, illness, dehydration, and being at the end of a long day all push in the same direction, and none of them are subtle.
Why Your Screen Inflates the Result
A browser reaction test does not measure your nervous system in isolation. It measures your nervous system plus the entire input and display pipeline, and that pipeline is not free.
On a 60 Hz display a new frame is drawn every 16.7 milliseconds, so on average the target has already been ready for about half a frame before you can possibly see it — and that is before the panel's own pixel response time. At 144 Hz the per-frame cost drops to roughly 7 milliseconds. Your mouse adds a similar tax: a 125 Hz polling rate can report your click up to 8 milliseconds after you made it, while a 1000 Hz mouse keeps that closer to 1. On top of those sit the operating system's input handling and the browser's own event loop. On a phone, the touchscreen digitiser typically samples more slowly than a wired mouse and adds noticeably more delay than a desktop setup does.
Added together, tens of milliseconds of latency that you did not produce end up credited to you. This matters most when comparing scores: your result on a phone and a friend's result on a high-refresh desktop are not measuring the same quantity. Compare yourself against yourself, on the same hardware, and the number becomes genuinely informative again.
Can You Actually Train Reaction Time?
Partly. The raw physiological floor — nerve conduction, sensory processing, muscle activation — is close to fixed and does not move much with practice. What does improve is everything wrapped around it: familiarity with the stimulus, having a motor response pre-loaded and ready to release, less hesitation about whether the target has genuinely turned ready, and fewer attention lapses.
In practice, improvement shows up as reduced variability rather than a dramatically faster best. Your single fastest response barely budges; your slow responses become rarer. In a run-based test like this one, that is precisely what raises a score, because the clock is unforgiving about the bad attempts and indifferent to the brilliant ones. It is also what people usually mean when they call an athlete's reactions fast — mostly anticipation and pre-loaded responses, reading the cue early enough that the movement has already begun.
The practical version: keep sessions short, since precision degrades with fatigue within minutes; keep your hardware constant; and judge yourself on typical performance across many trials rather than on a lucky peak.
About Reaction Target
Reaction Target is a free online reaction test that measures visual response speed under continuous time pressure rather than in a handful of isolated trials. Most reflex testers give you five clicks and an average. This one puts your reaction time on a clock: hits buy time, mistakes spend it, and the run continues only as long as you keep responding well. That turns a snapshot into a sustained measurement, where consistency counts for as much as peak speed.
The early-tap penalty being larger than the miss penalty is a deliberate design choice. It removes the most common way to cheat a reaction test — spamming clicks in the hope of catching the stimulus — and rewards controlled responses over gambling. Combined with the randomized ready delay, it means a high score reflects reacting rather than guessing. Your best hit count is stored locally in your browser, so nothing is uploaded and there is no account to create.