In short
- Human Benchmark is the reference simple-reaction test; a median in the region of 250–275 ms is commonly reported for it, though the live figure changes and includes device latency.
- Sense Test and Sensorium turn reaction and timing into free, short phone drills; Reaction Speed Test&Train adds dynamic and tracking tests for US$0.99.
- Percore's Reaction and Prediction modules put reaction time next to colour, pitch and space in one self-relative daily record (Pro).
- Practice makes you faster at the practised task; the evidence for transfer to driving, sport or anything else is thin, and no app here should claim it.
- Measure fairly: same device, same time of day, many trials, and read the median, not the best.
How this list was made
Each test and app was checked against its own site or App Store listing on 21 October 2026. Prices are what the listings showed in US dollars on that date. We used no star ratings or download counts. Percore, which publishes this site, is included for the one thing it does here — reaction and anticipation as part of a wider daily measurement — and is not the right choice for someone who only wants the fastest possible number.
What a reaction time number contains
Simple reaction time is one stimulus, one response: the screen goes green, you tap. Choice reaction time adds a decision — tap left for red, right for blue — and is slower by a hundred milliseconds or more, because deciding takes time. Most apps measure the first; a few measure the second; they are not comparable with each other.
Whatever the app, the number is not purely you. Between the pixel changing and the app registering your tap sit the display's refresh and response, the touchscreen's sampling, the operating system and the app's own code. On a phone that chain typically adds tens of milliseconds; on a browser with a mouse it can add more, and a single reading can swing by ±50 ms for reasons that have nothing to do with your reflexes. This is why a number from one device cannot be compared with a number from another, and why the honest use of any of these tests is against yourself, on the same hardware.
- Age — simple reaction time lengthens slowly through adulthood.
- Sleep, alcohol and illness — all slow it, often by more than practice ever speeds it.
- Caffeine and arousal — a modest quickening, with a cost in false starts.
- Attention — expecting the signal is most of the skill; a predictable delay is reacted to faster than a random one.
- Device latency — see above. Hold it constant and it cancels out.
The tests and apps
Human Benchmark — Reaction Time Test
- Best for
- The reference number, with a huge comparison pool
- Platform
- Web
- Price
- Free
The standard. Wait for red to turn green, click, repeat five times, and get your average against a distribution built from a very large number of attempts. A median somewhere in the region of 250–275 ms for the simple visual test is commonly reported; the live figure on the site changes as results come in, and it includes every participant's display and input lag, so treat it as a rough landmark rather than a norm. A free account keeps your history, and the site's other tests — sequence memory, number memory, aim, typing — make it a convenient battery.
It is a test, not a practice: nothing to reproduce, no target to hold in mind. For reaction time that is fine; it is the simplest honest measurement there is.
The limit. Browser and mouse latency are baked into the number, so compare only with your own earlier results on the same machine.
Sense Test – Prove it
- Best for
- Timing judgements, free, alongside seven other senses
- Platform
- iPhone
- Price
- Free
Sense Test does not have a plain green-screen reaction test, and that is why it is here: its timing drills are the ones most apps skip. Time sense asks you to hold a button for a stated duration; tempo sense asks you to keep tapping at a steady rate. Next to them are angle, length, counting, midpoint, colour and percentage judgements, all free, all offline.
Scores feed a global ranking rather than a personal trend, so it tells you where you stand today more than how you are changing.
The limit. No simple reaction test and no self-relative history; it is a ranked test suite rather than a daily practice.
Sensorium: Train Your Senses
- Best for
- A sixty-second daily drill that includes reaction speed
- Platform
- iPhone and iPad
- Price
- Free; Pro US$4.99 once
Sensorium's daily micro-drills cover reaction speed alongside visual discrimination, auditory thresholds and motor control, each scored 0 to 100 and plotted on a multi-axis radar. The free tier is one activation a day with three core drills; Pro — a single purchase — opens the full library, lets you choose a focus and adds a mode that races your personal best live.
For reaction time specifically it is a routine rather than a reference: the value is in the daily series, not in comparing your number with anyone else's.
The limit. Reaction is one axis among several, and the listing does not state what the number includes or how it is computed.
Reaction Speed Test&Train
- Best for
- Static, dynamic and tracking tests in one small app
- Platform
- iPhone and iPad
- Price
- US$0.99, once
A small, single-purpose app with three tests: static (the usual tap-when-it-changes), dynamic (a target that appears in different places) and tracking (follow something moving). The dynamic and tracking modes are closer to what reaction means in sport or driving than a green screen is, and the price is a single dollar with no in-app purchases listed. Haptic feedback and light and dark modes were added in updates.
The listing does not describe the training modes in detail, so take "train" to mean "repeat and keep score" rather than a programme.
The limit. No published method or norms; a measuring tool, with the comparison left to you.
Simple Reaction Time Test
- Best for
- The green-screen test on a phone, nothing else
- Platform
- iPhone and iPad
- Price
- Free with ads; US$2.99 to remove them
Exactly what it says: the screen cycles red and yellow, turns green with a millisecond timer, you tap, and it records the result and your running average. It is the Human Benchmark test in your pocket, with sharing to messaging apps and a paid option to reset your average.
It is the right choice if all you want is the same measurement every day on the same device — which, as the section above explains, is the only way the number is comparable.
The limit. One test, ads in the free version, and nothing to practise beyond repetition.
Senaptec Sensory Station
- Best for
- Athletes with access to a sports-vision programme
- Platform
- Clinic and training-facility hardware, with companion apps
- Price
- Quoted per installation; strobe eyewear sold separately
Not an app you download, but the system most sports-vision programmes use, and the one a reader searching for athletic reaction training will meet. The station assesses ten visual and sensorimotor skills in about twenty-five minutes — visual clarity, contrast sensitivity, depth perception, near-far quickness, hand-eye coordination, multiple-object tracking, reaction time and perception span — and feeds a training programme; the company's strobe eyewear and a phone app for controlling it extend that to home practice. It is installed in professional and university facilities, including UFC Performance Institutes.
It is here so the comparison is honest: this is what reaction training looks like when it is instrumented properly, and it is not something a phone app replaces.
The limit. Institutional pricing and access; the home component is an accessory to a facility programme, not a stand-alone product.
Percore — Reaction and Prediction
- Best for
- Reaction and anticipation in the same self-relative record as colour, pitch and space
- Platform
- iPhone
- Price
- Free daily practice; the Reaction and Prediction modules are in Pro (US$39.99 a year or US$4.99 a week)
Percore's free daily practice measures colour, pitch, duration and shape. Its Timing sense has four modules in total, and two of them are about reaction: Reaction — respond the instant a signal appears — and Prediction — anticipate a moment before it arrives, which is the skill a green-screen test leaves out and the one that catching, cutting and conducting actually use. Both are practice modules, so they are part of Pro, alongside Duration (hold a press for a remembered interval) and Tempo (match a pace).
The point of having them here rather than in a dedicated app is the record: every module feeds a rolling average compared with your own recent weeks, confidence falls when a sense is left alone rather than the score falling, and there is no leaderboard. What is Percore? covers the rest.
The limit. Reaction is two modules of seventeen and sits behind Pro; for a free daily reaction number, use one of the tests above.
What practice changes, honestly
Repeat any of these tests daily and your number will fall — for a while, then plateau. Most of that gain is task-specific: you learn the rhythm of the test, you stop false-starting, you attend better to the moment the signal tends to come. The research on reaction-time and speed-of-processing training finds reliable improvement on the trained task, smaller improvement on closely related tasks, and little evidence of transfer to everyday performance. That is the same pattern found across perceptual learning generally, and it is not a reason not to train; it is a reason not to believe an app that promises you will drive better. Sleep, as it happens, moves the number more than any app.
How to measure yourself fairly
- Same device, same app, every time. Latency differs between phones, browsers and even cases of the same model; keep it constant and it cancels.
- Same time of day. Reaction time has a daily rhythm. Morning-after-coffee and late-evening are different people.
- Many trials, and read the median. Your best is a lucky guess; your mean is pulled by one lapse. The median is you.
- Note the context. Short sleep, a cold, a drink the night before — write it down, so a slow day is explained rather than alarming.
- Judge the trend over weeks. Any single day is noise; the series is the signal.
Which one, for which need
| You want to… | Start with |
|---|---|
| Get the standard number and compare with a large pool | Human Benchmark |
| Test timing judgements free, with seven other senses | Sense Test |
| Do a sixty-second drill a day and pay once | Sensorium |
| Add moving and tracking targets to the test | Reaction Speed Test&Train |
| Take the green-screen test on your phone every day | Simple Reaction Time Test |
| Train reaction as an athlete, instrumented | Senaptec, through a facility |
| Keep reaction and anticipation in one record with your other senses | Percore (Pro) |
Measure four senses, free, and add reaction when you want it.
Download on the App StoreFrequently asked
What is a good reaction time?
For a simple visual test on a typical device, a median in the region of 250 ms is commonly reported for adults, with practised gamers and athletes often faster. The number includes your device's latency, so the only comparison that means much is with your own earlier results on the same hardware.
Can you train reaction time?
You can get faster at the test you practise, and the gain is real. Evidence that it carries over to driving, sport or everyday life is thin. Sleep, attention and expectation change the number more than practice does.
Why is my reaction time different on my phone and my laptop?
Because the measurement includes the display, the input device and the software on each. A touchscreen and a mouse add different delays, and a single reading can vary by tens of milliseconds for hardware reasons alone. Pick one device and stick to it.
What is the difference between simple and choice reaction time?
Simple reaction time is one signal, one response: tap when it turns green. Choice reaction time adds a decision — which colour, which side — and is slower, usually by a hundred milliseconds or more. Apps that measure one are not comparable with apps that measure the other.
Is a free reaction time test accurate?
Accurate enough to track yourself, not accurate enough to compare with a laboratory figure. Free tests cannot control display and input latency; a lab can. Use them for the trend.