The impact of the irrelevant: the task environment modulates the impact of irrelevant features in response selection.

Abstract

There has been a lively debate concerning the impact of contingent top-down control settings on the occurrence of stimulus-driven attentional capture. To date, however, most studies have focused on the manipulation of just a single feature. Here, a variant of the response priming task was utilized in which multiple features as opposed to just a single feature were varied. Two types of features were defined: the response feature(s) (i.e., those features that are used to prepare the correct response) and the selection feature(s) (i.e., those features that help to separate the target against the distractor). The response feature was manipulated for the target as well as for the distractor. For a certain experimental block, the selection feature in the distractor was either fixed (not helpful for the separation of the target) or varied (helpful for the separation of the target). Because of this manipulation of the task environment (Experiment 1-3), differences in the size of the compatibility effect (i.e., the difference between compatible and incompatible trials) were observed for perceptually identical distractor-target sequences. In a task environment with varying selection features, a distractor that shared the selection feature with the top-down sets gave rise to larger compatibility effects than exactly the same distractor presented in a task environment where the selection feature was kept constant. However, participants must have explicit knowledge about the variance of the selection feature to implement the selection feature into their top-down sets. A model based on contingent capture was postulated to explain our results because of the feature-overlap between the distractor and the participants' top-down sets.