Why Reaction Time Scores Change Between Tests

A reaction-time score can change even when you use the same computer. Attention, task familiarity and the path between the screen and your response all influence an attempt. One unusually fast round is not a reliable baseline. Use the reaction time test with a fixed mode and round count, then compare several sessions under similar conditions.

What is included in the displayed milliseconds?

A browser test measures an interval between its stimulus timing and the response event it receives. Between those points are screen or audio delivery, your perception and movement, the input device, operating system and browser scheduling. It does not isolate biological processing time or tell you precisely how many milliseconds belong to each component.

A precise timer does not remove these other delays. Treat the result as a measurement of this task on this setup. It is useful for repeatable comparisons but is not a neurological diagnosis or a laboratory input-latency test.

60 Hz vs 144 Hz vs 240 Hz

Refresh rate gives a display’s nominal refresh interval. Divide 1,000 milliseconds by the refresh rate to calculate it.

These are calculated intervals, not measured display latency or predicted score improvements. Panel processing, scanout, pixel response and the moment a change becomes ready also matter. You cannot subtract 12.5 ms from every 60 Hz result to predict that person’s score on a 240 Hz display.

The browser’s animation loop is also linked to rendering. The requestAnimationFrame documentation explains that callbacks generally follow display refresh and can pause in background tabs. Keep the test visible and focused.

Average, median and your fastest round

The average adds valid reaction times and divides by the number of valid rounds. The median is the middle value after sorting them. For an illustrative set of 220, 225, 230, 235 and 500 ms, the average is 282 ms and the median is 230 ms. The delayed 500 ms round shifts the average more than the median.

Neither statistic should be chosen simply because it makes your result look better. Use both. The fastest round describes your best observed response, while the other rounds show whether you can repeat it. Record early responses or errors too; do not quietly remove difficult attempts and compare the remainder with someone else’s complete session.

Why mouse, keyboard and touchscreen results differ

Each input involves a different movement and processing path. A finger already resting on a mouse button is not the same task as reaching toward a touchscreen. Keyboard travel, mouse switch behavior and device reporting can affect the interval before the browser sees a response. Keep the input method fixed when tracking progress.

Internet ping is a different measurement. This site’s reaction interaction runs in the browser after loading; the score is not the round trip to a gaming server. A better connection should not be presented as a guaranteed reduction in this test’s measured milliseconds.

A repeatable testing protocol

1. Use the same device, display, browser and input method.
2. Select Visual mode and the same number of rounds for each session.
3. Complete a short familiarization attempt before recording results.
4. Respond to the signal instead of guessing when it will appear.
5. Record average, median, valid rounds and errors.
6. Repeat later under similar conditions and compare the whole session.

When comparing hardware, alternate the order between sessions and change only one component at a time. A small improvement in one run may be ordinary variation rather than evidence that a new device is faster.

Keep speed tests separate

Repeated clicking speed is a different task from responding once to a signal. You can measure mouse repetition with the clicks per second test, but a higher CPS does not establish a lower reaction time. For game-specific context, read how input speed relates to gaming performance. Aim, decisions and timing remain outside a simple browser reaction score.