Can subitizing survive the attentional blink? An ERP study.

This study was done to test whether subitizing versus counting are attention demanding based on whether they can be performed during the attentional blink (AB). ERPs were recorded while participants performed a task requiring them to judge the number of dots presented and this judgment task either followed the presentation of a task-relevant item in a rapid stimulus presentation stream (dual-task) or the potential target was task irrelevant (single-task). The behavioral data demonstrated that T2 accuracies decreased as a function of the number of dots not only in counting range, but also in subitizing range. The ERP results showed a delayed P3 component in the dual-task condition, and this was equally true for both subitizing and counting conditions. Furthermore, the P3 amplitude was reduced during the AB, and this was still equally true for both the subitizing and counting conditions. The present results suggest that both subitizing and counting require attention, and that subitizing is not a purely pre-attentive process.