Abstract
In the present study, we used a spatial cueing task to investigate fMRI activity associated with voluntary and involuntary attention. A spatial cue appeared on either the left or the right side of fixation, above the horizontal meridian (HM). 250 milliseconds later, a target appeared below the HM, either on the cued- or on the uncued side. The target was one of two faces, and the participants‘ task was to indicate which face was presented. The spatial separation of cue and target allowed separate measurements of cue- and target-related fMRI responses in retinotopic visual areas. Attention type was manipulated by varying the proportion of targets appearing on the cued vs. the uncued side. For half of the blocks, the cue was non-predictive of target location, engaging only involuntary attention. In the remaining blocks, the cue was predictive of target location, engaging both voluntary and involuntary attention. All blocks included cue-only trials which allowed measurement of the responses to cue presentation alone in the different attention conditions. Blocks in which the cue preceded target presentation were paired with matched post-cue blocks in which the order of cue and target was reversed, rendering the cue temporally task-irrelevant. Behaviorally, participants were significantly faster when the target appeared in the cued side for both predictive and non-predictive blocks, and this difference was larger in the predictive condition. Importantly, post-cueing resulted in no difference between cued and uncued RTs. fMRI localizers were used to define cue- and target-related ROIs in occipital and parietal cortex as well as ventral face-selective areas and frontal regions implicated in spatial attention. Our experimental design allows a comparison of each trial type in predictive and non-predictive conditions under identical stimulus conditions. In addition, fMRI coherency analysis is being performed to examine the circuitry engaged in voluntary and involuntary attention.
National Eye Institute #016975.