Towards a Better Understanding of Early, Parallel Attentive Processing in Vision
Alejandro Lleras
University of Illinois at Urbana-Champaign

The goal of the vast majority of visual attention theories has been to understand how the capacity limitations of focused/spatial attention impact our interactions with the visual world. Our work is aimed at complementing these theories by providing a more precise understanding of how parallel, unlimited capacity attentive processing interact early on with the visual input to create the representations that will guide the behavior of focused attention (such as priority maps). Our theoretical work uses mathematical models to simulate various cognitive architectures and test those models using human behavioral data. In this talk, I will give some examples of our work where we have successfully used this approach to estimate behaviorally for the first time the extent of inter-item interactions (such as distractor-distractor suppression effects) and also to suggest that priority/salience maps can guide attention to likely target locations without explicitly representing salience (or target-distractor similarity).