Reaction Time Norms by Age — Percentiles & Sources
What's a normal reaction time by age? Grounded percentile distribution for a web/touch simple reaction time test, with sources.
Simple visual reaction time — how fast you tap when the screen turns green — is fastest in your twenties and slows gradually with age. On a web or touch test, the typical median for adults under 50 sits around 271ms once screen and input lag are included.
This 271ms figure is derived two independent ways that agree: lab button-press studies (Woods 2015, ~231ms) plus a disclosed +40ms web/touch latency offset, cross-checked against Human Benchmark's crowd median of ~273ms from millions of web clicks.
Each cell is the tap speed you'd need to be faster than that share of same-age people, on a web/touch test. Lower is faster.
| Age | Faster than 90% | Faster than 75% | Median (50%) | Faster than 25% | Faster than 10% |
|---|---|---|---|---|---|
| Under 50 | 220ms | 244ms | 271ms | 298ms | 322ms |
| 50–64 | 260ms | 284ms | 311ms | 338ms | 362ms |
| 65+ | 310ms | 334ms | 361ms | 388ms | 412ms |
We only use studies that measured the same task — a finger/screen tap in response to a visual signal — and disclose the web/touch calibration. Percentiles are modelled from a normal distribution around the age-band median; because reaction time is mildly right-skewed, the extreme tails are approximate. This is a reference for self-comparison, not a clinical norm.
Want to go deeper than a score? Copy the prompt below into ChatGPT, Claude or another AI assistant to reflect on your results, ask follow-up questions and get practical next steps. AI can be a helpful thinking partner for self-reflection — but it is not a therapist. Please read the note below.
I just completed the Reaction Time Norms by Age — Percentiles & Sources, a self-assessment questionnaire. I'd like to reflect on what my results might mean. Please act as a supportive, evidence-based wellbeing coach: first ask me 2–3 clarifying questions, then explain in plain language what my results could indicate, and suggest small, realistic steps I could try over the next two weeks. Be honest about the limits of a self-test, and tell me when it would be wise to talk to a qualified professional.
AI assistants can make mistakes and are not a substitute for professional diagnosis or care. If you are in crisis or thinking about harming yourself, please contact your local emergency services or a crisis hotline right away.
A reference for typical simple visual reaction time by age: the median and percentile spread for a web/touch tap test, derived from published research.
See what counts as a fast, average or slow reaction time for your age, and where your own result sits.
Find your age band, then read across: a lower millisecond value means you'd beat a larger share of same-age people. The median (50%) is the typical result.
These norms are a reference for self-comparison, not a clinical standard. Web/touch tests include device and screen latency, so treat the numbers as approximate and relative.
| Age | Median reaction time |
|---|---|
| Under 50 | 271ms |
| 50–64 | 311ms |
| 65+ | 361ms |
On a web/touch tap test, the median is about 271ms for adults under 50, ~311ms for 50–64, and ~361ms for 65+. Simple reaction time stays fairly flat until about 50, then slows more steeply (Der & Deary 2006). Lab button-press studies read faster (~230ms) because they exclude screen and touch latency.
Faster than the median for your age is good. For under-50s, tapping around 220ms puts you ahead of about 90% of people, while 322ms sits behind most. On any web test, screen and input lag add a little, so your number is best read against your own age band.
Lab studies use calibrated button boxes and correct for hardware; controlled measurements land near 213–231ms for young adults (Woods 2015). Web and touch tests add display and input latency — often 40ms or more — which is why crowd medians like Human Benchmark's ~273ms are slower. Our norms are calibrated to those web/touch conditions.
We anchor the young-adult mean to Woods 2015 (finger button-press), apply Der & Deary's age slope, add a disclosed +40ms web/touch offset, and model percentiles from a normal distribution around each age median. The under-50 median of 271ms matches Human Benchmark's ~273ms crowd median from millions of web clicks.
Yes. AI assistants like ChatGPT and Claude can be a useful way to reflect on what your results mean and explore next steps, and this tool gives you a ready-made prompt plus a one-click link to start that conversation. Keep in mind that AI is not a licensed professional and cannot diagnose you — for a formal assessment, or if you are struggling, please consult a qualified health professional.