The Role of Attention in Basic Ensemble Statistics Processing

The visual system can represent information about multiple objects in the form of ensemble statistics, such as their mean feature. Although ensemble representation is often considered a strategy to deal with attentional capacity limitations, it is under debate whether it requires attention. We investigated this question using two ERP markers, the P3 which is evoked by change detection of attended stimuli, and visual mismatch negativity (vMMN) which is elicited by automatic sensory discrimination under attention diverted from the critical stimulus. In Experiment 1 (attended ensemble changes), observers attended to ensembles and reported rare changes of their mean orientation, while fixating on a central cross. In Experiment 2 (unattended ensemble changes), participants attended to changes in a central cross and ignored the background ensemble stimuli which also sporadically changed mean orientation. When ensembles were attended, changes in their mean evoked the P3 component, a marker of conscious change detection. When the same ensembles were unattended, no evidence for the vMMN, a marker of automatic discrimination, to ensemble mean changes was found. These results let us suggest that attention is critical for ensemble discrimination.