When spatial attention cannot be divided: Quadrantic enhancement of early visual processing across task-Relevant and irrelevant locations.

Spatial attention enables us to select regions of space and prioritize visual processing at the attended locations. Previous research has shown that spatial attention can be flexibly tuned to broader or narrower regions in space, and in some cases be split among multiple locations. Here, we investigate how attentional resources are distributed within a visual quadrant when participants are instructed to either focus attention narrowly, broadly, or split attention among two noncontiguous locations. Using a combination of behavior and steady-state visual-evoked potentials (SSVEP), the oscillatory response of the visual cortex to incoming flickering stimuli, we find clear evidence for ineffective splitting of spatial attention within a visual quadrant. Importantly, by assessing visual-cortical processing across locations at a high spatial resolution (by flickering nearby locations at distinct frequencies), our results reveal that attention was distributed in the exact same manner regardless of whether participants were instructed to attend broadly across a large region of space, or divide attention among two noncontiguous locations: In both cases, the intermediate location showed the strongest boost in visual-cortical processing, no matter whether it was the center of the attended region (broad-focus condition), or the uncued, to-be-ignored location (split-focus condition). Thus, the present study provides strong evidence that when trying to attend to multiple separate locations within a visual quadrant, sustained attention inadvertently enhances visual processing at the intermediate location even when it is detrimental to task performance.