Proceedings of the Annual Meeting of the Cognitive Science Society

Attention can be biased by the previous learning and experience. We present an algorithmic-level model of this bias in visual attention that predicts quantitatively how bottom-up, top-down and selection history compete to control attention. In the model, the output of saliency maps as bottom-up guidance interacts with a history map that encodes learning effects and a top-down task control to prioritize visual features. We test the model on a reaction-time (RT) data set from the experiment presented in (Feldmann-Wustefeld, Uengoer, & Schubö, 2015). The model accurately predicts parameters of reaction time distributions from an integrated priority map that is comprised of an optimal, weighted combination of separate maps. Analysis of the weights confirms learning history effects on attention guidance.