This error is intent-independent, driven only by the surface features of perceptual input

What it is

The standard first question when triaging an error is "did the person actually intend to do this" — but that question has no discriminating power for data-driven errors. Their defining property is intent-independence: the error can occur even when the person has zero interest in, or has never once thought about, the action that got mistakenly triggered, because what triggers it isn't the person's goal or thoughts but the fact that the data currently being perceived happens to match some well-learned schema's input pattern. Diagnosing this error type needs a different question: "was the intruding content actually present in front of the person's eyes or ears at the time," not "was the person thinking about it."

Why it happens

The activation that drives an action comes from two channels: an internal one generated by the person's current goals and thoughts, and an external one generated by surface pattern-matching against perceptual input. A data-driven error is a case where the external channel alone crosses the activation threshold for an unrelated schema — which means it can happen to a person who is fully absorbed in their actual task, with no stray thoughts at all, simply because a well-learned schema's trigger pattern happened to appear in the input stream. This is why intent-independence isn't a footnote but the most essential feature of this error type: it bypasses the level of intent entirely, jumping straight from perceptual match to action output.

Studying it

Establishing the intent-independence claim requires an experimental design that can separate perceptual triggering from intentional triggering: have participants focus on a task completely unrelated to some well-learned schema, while inserting into the input stream an irrelevant stimulus that happens to match that schema's trigger pattern, and observe whether it induces the schema's associated action even though the participant has no expectation of or interest in it. The value of this kind of design is that it can dissociate "doing something because you were thinking about it" from "doing something because you saw or heard it" within the same person, rather than relying only on retrospective classification of naturally occurring cases.

Where it stops holding

The intent-independence claim only holds for actions that are already highly automatized, where execution doesn't require step-by-step conscious verification — actions that need to be actively constructed and deliberated over give surface pattern-matching in perceptual input no channel to bypass that construction process and trigger the action directly, which is why data-driven errors are far rarer in that setting. This also sets a hard limit on countermeasures: any fix that relies on "getting the user to focus harder" is doomed to fail, because focus changes the strength of internal goal activation, and this error type never passes through the internal-goal level to begin with.

Related

• Same group: A10.10.1 Data-driven errors are triggered by externally presented information hijacking a similar-looking action schema · A10.10.3 Interface elements that resemble a hazard's trigger cues keep inducing this error · A10.10.4 Reducing this error means redesigning the input side, not asking users to focus harder
• Nearby: A10.11 Associative-activation errors · A8.05 Motor programs and movement preparation
• Search terms: data-driven error · bottom-up activation · automaticity