The Sustained Attention to Response Task (SART) is the standard laboratory measure of sustained attention and the everyday attention lapses that accompany it (Robertson, Manly, Andrade, Baddeley & Yiend, 1997). Single digits from 1 to 9 are presented one at a time in a fast, monotonous stream. Participants make a speeded key press to every digit (the frequent Go response) but must withhold the response whenever the digit is a 3 — the rare no-go target (~11% of trials).

Because the Go response is highly prepotent and the target is infrequent, participants slip into an automatic, rhythmic responding mode and periodically fail to withhold — pressing on the 3. These commission errors (SART errors) are the paradigm's signature measure and are interpreted as momentary lapses of sustained attention / mind-wandering. Trial-to-trial reaction-time variability on Go trials is a second, complementary index of attentional instability.

To prevent participants from solving the task with a low-level perceptual template (e.g. "withhold to a small numeral"), each digit is drawn at one of five randomly varying font sizes, forcing genuine semantic processing of the numeral. Each digit is shown briefly (250 ms) and immediately backward-masked by a ring, with the response accepted across the full 1150 ms stimulus-onset asynchrony.

This implementation ships with 9 practice trials (one full 1–9 cycle, with correct/incorrect feedback) followed by a 45-trial main block (a full crossing of 9 digits × 5 font sizes; the target 3 appears 5 times). All timing, trial counts, the target digit, and every participant-facing string are exposed as re-usable parameters so a researcher can adjust or translate the task in the Open Lab study builder without editing individual screens. Set _main_repetitions to 5 to reproduce the full 225-trial Robertson protocol.