Using illusory line motion to explore attentional capture

I explored endogenous control of attentional capture using performance and phenomenology. Trials began with four figure-8s presented above, below, left and right of fixation. One figure-8 (or none) brightened uninformatively (cue) before either a digit target (2 or 5 made from a figure-8 by deleting line segments) calling for a speeded identification, or a line connecting adjacent figure-8s calling for a motion judgment. Such lines are seen drawn away from a cue when the cue is adjacent to the line. Digit targets appeared only on the horizontal or vertical axes (counterbalanced across observers), encouraging endogenous attention to 2 of the 4 figure-8s. Attention was captured efficiently by exogenous cues at task-irrelevant locations unless the line task was omitted or the line was presented connecting task-relevant locations. ILM was weakly experienced moving away from attended locations in lines not adjacent to a cue, suggesting endogenous generation of illusory line motion.