The temporal window of attention to self-generated stimuli

Abstract

Perceptual stimuli generated by one's actions (i.e., action-effects) attract attention more strongly than stimuli that occur independently of the action. Whether a stimulus is perceived as an action-effect or not is thought to depend on the temporal proximity of the action and stimulus. Given that, it is possible that the attentional boost to action-effects would have a specific temporal window. Thus, we assessed whether temporal contiguity between the action and effect modulates attention capture. We adopted a spatial orientating task in which a white central fixation and two white placeholders at the left and right locations were presented. We had two conditions; when the fixation turned red, participants in the action condition pressed a middle key while those in the non-action condition did not. Following this, one of two placeholders flashed in red for 50 ms as a cue, and after 50 ms a white circle as a target appeared inside either of the two placeholders. Thus, in the action condition, keypress caused the presentation of the cue while in the no-action condition, it did not. In the action condition, the interval between keypress and cue onset was 50, 750, or 3550 ms and in the no-action condition, the interval between the color change of the fixation and cue onset was 400, 1100, or 3900 ms (including 350 ms as an alternative of keypress time in the action condition). Participants quickly pressed either the left or right key depending on the target location. The results did not support the presence of a strict temporal window; action-effects captured more attention than external stimuli even more than three seconds after the action. This finding implies that the attention to action-effects does not result from time-sensitive sensorimotor processing, but reflects a more general process of action-effect monitoring.