The flanker effect does not reflect the processing of "task-irrelevant" stimuli: evidence from inattentional blindness.

It is often the case that stimuli (or aspects of a stimulus) are referred to as being "task-irrelevant." Here, we recount where this label originated and argue that the use of this label is at odds with the concept of "relevance" that has arisen in the contingent-capture literature. This is not merely a matter of labels, but a matter of inference: When people describe a flanker stimulus as being "task-irrelevant," they may be (and sometimes are) tempted to infer that the conditions that were studied in the flanker task generalize to other tasks and other types of stimuli. Here, we show that this generalization is not warranted. The flanker effect exists not because attention has failed at selecting only the target from the display, but rather, the effect arises precisely because attention succeeded at selecting target-like (i.e., attentionally relevant) stimuli from the display. As a result, the flanker effect should not be used to infer how stimuli that are entirely unrelated to a participant's main task would be processed. We propose the use of a new terminology to replace this potentially misleading label.