Does contingent capture occur in driving scenes?

Contingent capture theory suggests that only stimuli consistent with the observer’s internal goals will capture attention. For example, when looking for a red target, red stimuli should automatically capture observers’ attention. It is unclear whether contingent capture occurs in complex real-world scenes. Arexis et al. (2017) investigated contingent capture using a search task where participants viewed photographed driving scenes in which a single red letter appeared at random locations. Their task was to decide whether the red target letter was a T or an L. A GPS navigation system image was also shown in the bottom right corner of the display. The GPS either had a blank screen, a red-coloured route (goal-relevant distractor colour) or a green-coloured route (goal-irrelevant distractor colour). The GPS appeared 1 second in advance of the driving scene and letter, acting as a pre-search display. If contingent capture occurs, participants should be slower to respond when the GPS showed a red route (goal-relevant distractor colour). However, goal-relevance had no effect, perhaps because the GPS pre-search display appeared so far in advance of the search display (1 s). In the present study, we manipulated the presentation duration of the pre-search display (0 ms, 100 ms, 1 s) and instructed participants to ignore the GPS. We predicted that goal-relevance would have an effect, but only when there was insufficient time to disengage attention from the GPS distractor before search display onset (the 0 and 100 ms conditions). Results call into question the contingent capture hypothesis within the context of real-world scenes.