Abstract
Is attention automatically captured to the location of salient stimuli, or is capture under our control? The best evidence that capture can be controlled comes from contingent capture in attention cueing tasks: When looking for a visual target (e.g., a red target), distracting stimuli only capture attention if they resemble the target (e.g., a task-irrelevant red pre-cue). Put differently, what observers are doing in the target display determines which types of features will capture attention in the cue display. Here we assessed another level of control. What participants are doing in the target display may also determine where attention goes in response to the cue in the cue display. This prediction draws from a recent demonstration that spatial attention can be cued to arbitrary locations based on implicitly learned associations between features and locations (Girardi & Nico, 2017). In the present experiments, participants completed a cueing task where on every trial a target was presented to the left and right of fixation, and a separate, coloured stimulus indicated which target the participant should report (e.g., red meant report left target; green report right). Thus, the target display created an association between colours and shifting attention to the left or right. Across three experiments, task-irrelevant, non-predictive pre-cues captured attention to the location associated with their colour (e.g., red cues captured attention to the left location) regardless of where the cue physically appeared in space. These experiments also investigated and ruled out two alternative explanations based on spatially specific attentional control settings and colour priming. Instead, the present findings support the conclusion that attentional control settings do not only determine what types of stimuli should capture visual spatial attention, but also define where attention should go when captured.