Microsaccades track shifting but not necessarily maintaining covert visual-spatial attention

Covert attention enables the brain to prioritise relevant visual information without directly looking at it. Ample studies have linked covert visual-spatial attention to the direction of small fixational eye movements called microsaccades, offering researchers and clinicians a potential non-invasive tool to track internal states of covert attention. However, other studies have reported only a weak link or no link at all. We now show that the link between covert visual-spatial attention and microsaccades critically depends on the stage of attentional deployment. Across two experiments—each employing a distinct but widely adopted approach to fixational control—we show that spatial microsaccade biases were more pronounced (Experiment 1) or even exclusively present (Experiment 2) during the initial stage of shifting covert spatial attention, even when covert attention was subsequently maintained as testified by enhanced visual discrimination. This shows that the involvement of the brain’s oculomotor system in covert visual-spatial attention qualitatively changes over the course of attentional deployment, and how its peripheral fingerprint in the form of microsaccades reliably indexes shifting but not necessarily maintaining covert visual-spatial attention.