Abstract
Attentional capture by an exogenous spatial-cue is generally believed to be mediated by retinotopic mechanisms. Using a Ternus-Pikler display, we show that attentional capture occurs in non-retinotopic coordinates. In a first frame, three squares were presented for 200ms followed by a 70ms ISI and a second frame containing the same three squares shifted laterally by one position. Observers perceived three squares moving laterally as a group. Within the central square of the first frame, a cue was flashed either at the center (neutral cue) or at a peripheral position (retinotopically neutral cue, non-retinotopically 100% valid cue). In the central square of the second frame, a conjunction search display was presented. Subjects had to search for a red tilted bar (target) and indicate its tilt direction. Relative to the surrounding squares, the peripheral position of the cue in the first frame corresponded to the peripheral position of the target in the second frame. On the other hand, because the central square moved, retinotopic positions of the cues were different from those of the search items. A retinotopic account of attention predicts no difference in performance between central and peripheral cue as both cues recruit attention at a location where the target is not presented. In contrast to this prediction, we found faster reaction times for peripherally versus centrally cued trials, indicating that the peripheral cue acts in non-retinotopic coordinates in summoning attention to the target location. These results provide strong evidence for a non-retinotopic component in spatial visual attention.