The preparatory interval has an optimal window; too short or too long is ineffective

What it is

The interval between a warning signal and the target stimulus (the foreperiod) has an optimal window that produces the greatest shortening of reaction time: if the interval is too short, the participant doesn't have time to complete the transition from being warned to being in a prepared state; if it's too long, alertness drifts back down while waiting and the preparation effect gradually fades. Neither extreme delivers the maximum RT benefit — the peak gain sits somewhere in a window between the two.

Why it happens

The prepared state itself is a resource that has to be actively maintained and decays over time — a warning signal takes some time to build up a rise in alertness, and staying at a high level of alertness itself carries a cost that can't be sustained indefinitely at its peak. This naturally produces an inverted-U curve — building up takes time, and maintaining it decays — and the optimal interval falls in the window where the buildup has finished but the decay hasn't yet become significant.

Studying it

By systematically varying the length of the foreperiod (from around 100ms up to several seconds) and measuring RT under the same task at each length, researchers can plot how RT changes with interval length and locate the interval range where RT is shortest. Because the exact optimal value shifts with the task, the sensory modality, and each individual's baseline alertness, most studies can only establish that such a window exists and roughly what shape it takes, rather than giving a single millisecond figure that generalizes across tasks.

Where it stops holding

The exact length of the optimal window varies with task complexity, stimulus modality, and individual differences, so there is no fixed number that can be dropped directly into any interface design scenario. In addition, if the foreperiod is variable rather than fixed (the participant cannot reliably predict when the target will appear), the effect of the optimal window is weakened or reshaped by this added layer of "when will it come" uncertainty.

Applying it

When designing an interaction flow that includes a warning cue, don't assume that "the longer the gap between the warning and the main content, the more prepared the user will be," and don't compress the gap so tightly that the main content appears before the warning signal has even been perceived. Instead, measure directly to find the interval length that produces the shortest RT in this specific product context, and use that as the default — rather than borrowing a value from another product or setting it by intuition. How to check: measure user reaction time under several different warning-to-content interval lengths, and adopt the measured shortest-RT interval as the default parameter.

Related

• Same group: A8.03.1 A warning signal can substantially shorten the reaction time that follows · A8.03.3 A fixed timing schedule induces anticipatory jumping the gun · A8.03.4 An unpredictable stimulus cannot benefit from the preparation effect
• Nearby: A8.01 The structure of reaction time
• Search terms: optimal foreperiod · foreperiod curve · preparatory interval