Cue-target contingencies modulate voluntary orienting of spatial attention: dissociable effects for speed and accuracy

Voluntary orienting of spatial attention is typically investigated by visually presented directional cues, which are called predictive when they indicate where the target is more likely to appear. In this study we investigated the nature of the potential link between cue predictivity (the proportion of valid trials) and the strength of the resulting, covert, orienting of attention. Participants judged the orientation of a unilateral Gabor grating preceded by a centrally-presented, non-directional, colour cue, arbitrarily prompting a leftwards or rightwards shift of attention. Unknown to them, cue predictivity was manipulated across blocks, whereby the cue was only predictive for either the first or the second half of the experiment. Our results show that the cueing effects were strongly influenced by the change in predictivity. This influence differently emerged in response speed and accuracy. The speed difference between valid and invalid trials was significantly larger when cues were predictive, and the amplitude of this effect was modulated at the single trial level by the recent trial history. Complementary to these findings, accuracy revealed a robust effect of block history and also a different time-course compared to speed, as if it mainly mirrored voluntary processes. These findings, obtained with a new manipulation and by using arbitrary non-directional cueing, demonstrate that cue-target contingencies strongly modulate the way attention deploys in space.