Within- versus cross-dimensional capture in fixed-feature visual search

Recent ERP findings indicate that colour singletons fail to capture attention when observers search for a less salient shape that remains fixed across trials. This is consistent with the dimensional weighting account, according to which signals on the relevant (shape) and irrelevant (colour) dimensions are enhanced and suppressed, respectively. The goal of the present experiments was to examine the influence of prior knowledge of a target for over-riding a more salient distractor in a compound search task, when the target and distractor exist within the same visual dimension (colour versus colour; Experiment 1) and across different visual dimensions (shape versus colour; Experiment 2). More interference was found in the within-dimensional condition than the cross-dimension condition (18 msec vs. 7 msec). Both the within- and cross-dimensional targets elicited an N2pc component (indexing selection) and the distractor elicited a PD component (indexing suppression). In the within-dimensional experiment, the amplitude of the N2pc varied as a function of the proximity of the distractor to the target. These results indicate that when dimensional weighting is not a viable selection strategy, the visual system suppresses the location of the distractor while selectively processing the target.