Pupil size tracks attentional breadth in the Navon task

Abstract

Processing limitations necessitate the selection and prioritization of parts of the visual input-that is visual attention. Visual attention cannot just shift in space, but also changes in size, so-called attentional breadth. A common paradigm to assess attentional breadth is the Navon task wherein participants are instructed to attend global or local features in ambiguous figures. Differences in response times and accuracy then allow inferences about attentional breadth. Here we tested an alternative, overt-behavior free marker of attentional breadth in the Navon task: pupil size changes. Participants were asked to report the parity of either the global or the local number making up an adjusted Navon stimulus. Global and local numbers differed in luminance. We found no differences in pupil size when either a bright or dark feature was attended. However, we did find a larger pupil size when the global compared with when the local number was attended. This effect could be attributed to multiple factors. First, as accuracy was lower when reporting global compared with local features, task difficulty likely affected pupil size. Second, the observed effect possibly reflects higher effort necessary for a wide compared with a narrow attentional breadth-in our specific task layout. Third, we speculate that attentional breadth may effort-independently contribute to this difference in pupil size. Future work could tease apart these factors by changing task layout and stimulus sizes. Together, our data show that pupil size may serve as a physiological marker of attentional breadth in the Navon task.