Abstract
While there is much disagreement as to what factors determine attentional priority, most theories suggest that attention is deployed to the highest-priority object at any given time. Recently, a new model of attention, the Priority Accumulation Framework (PAF; Lamy, Darnell, Levy & Bublil, 2018), challenged this assumption. According to this model, attentional priority accumulates at each location over time, and detection of the search-relevant context (the search display) triggers the deployment of attention to the location with the highest cumulative priority. Here, we tested new predictions of this model by measuring overt shifts of attention (i.e., eye movements) in a spatial cueing paradigm, in which participants were free to move their eyes. An abrupt-onset cue was followed by a search display containing a perfect circle (the target) among three ellipses (the distractors). On any given trial the distractors could be all similar to the target (difficult-search), all dissimilar from the target (easy-search) or mixed (one similar and two dissimilar distractors). A black dot appeared inside each shape and participants had to report the location of the dot inside the target (left or right). The results revealed that early first saccades, that is, first saccades that occurred prior to processing of the search display, occurred on only one-fourth of the trials, with large individual differences, and were almost exclusively directed to the cue location. Thus, most first saccades were initiated following the search display, that is, when the cue onset had failed to capture overt attention. Crucially, however, the cue still biased the distribution of those later first saccades. Moreover, competition strength (i.e., search difficulty) strongly modulated the cue’s contribution to late saccades’ distribution, with a larger impact of the cue, the stronger the competition. These findings support the Priority Accumulation Framework and challenge the standard interpretations of spatial cueing effects.