Reaction Time vs Attention

Selective attention requires responding to a relevant feature while other information is present. In the site's short Attention Test, a target calls for a response and a no-go cue calls for withholding. A miss and a false alarm are not interchangeable: one omits a required response, the other produces a prohibited one.

Inhibition tasks also expose a strategy tradeoff. Responding aggressively may shorten hit times while increasing false alarms. Waiting for certainty may reduce false alarms while increasing misses or latency. Good interpretation keeps all of those outcomes visible.

A Stroop-style trial asks for the ink color while a readable word may agree or conflict. Congruent, neutral, and incongruent conditions can be summarized separately. The page-defined interference difference compares correct response timing across conditions; it is not a direct measure of a person's executive function.

Language fluency, display color, response mapping, errors, and the number of usable trials all matter. The browser implementation demonstrates rule conflict but does not reproduce a professionally administered assessment.

The Sustained Attention Test presents a stream over minutes and makes targets relatively infrequent. Its useful outputs include hits, misses, false alarms, median hit timing, variability, and a comparison between the first and second halves. One mean reaction time would hide much of the task.

A slower second half can have many explanations, including strategy, interruption, display conditions, or ordinary variation. The page cannot isolate fatigue or diagnose an attention condition from that pattern.

A displayed cue passes through display refresh and rendering. The response passes through a touchscreen, mouse, keyboard, operating system, browser event queue, and JavaScript handler. The recorded interval therefore contains human response plus hardware and software latency. Background load and dropped frames can add variation.

The W3C Event Timing model explicitly distinguishes an event timestamp from the time processing begins. That does not let a normal page subtract every device delay; it explains why a browser number should be treated as end-to-end page timing rather than a laboratory-grade physiological interval.

Phones and computers differ in refresh behavior, input sampling, touch movement, cursor travel, viewport size, orientation, and browser scheduling. A user may also hold a phone at a different distance and use a different posture. Those are changes to the measurement conditions, not noise that can be safely ignored.

Create separate personal baselines for touch and mouse or keyboard. Keep browser zoom and orientation stable. Prefer medians across completed, unflagged runs, and do not compare your result with an online chart whose hardware, sample, and procedure are unknown.

Use Reaction Time when the question is about one known response after one signal. Use Attention for a short target/no-go block, Stroop or Flanker for conflicting information, and Sustained Attention for longer low-frequency monitoring. Use accuracy and error types alongside time on every decision task.

None of these choices yields an ADHD score, clinical reaction-time norm, or general focus grade. They provide different browser interactions whose results remain meaningful only inside their stated rules and conditions.