Abstract
Three accounts have been proposed to explain the influence of first target (T1) difficulty on the Attentional Blink (AB). The processing speed account (PSA; Visser, 2007) suggests that increasing T1 processing time increases the severity of the AB. Furthermore, the PSA suggests that the inclusion of a T1-mask prevents T1-difficulty from influencing the AB. The resource allocation account (RAA; Shore et al., 2001) suggests that T1-difficulty only influences the AB when resources are allocated prior to the start of the trial, and therefore should only be observed when T1-difficulty is blocked. Finally, the flexible selection account (FSA; Giesbrecht et al., 2007; 2009) suggests that T1-difficulty should decrease processing of subsequently presented information. We test the key predictions of these three accounts in two experiments using a standard RSVP task that evokes an AB (e.g., Chun & Potter, 1995) and in which T1-difficulty was manipulated using noise dots presented simultaneously with T1. In experiment 1, T1-difficulty increased the severity of the AB (F(5,65)=3.166, p = .033) even though T1 was always masked and T1-difficulty was intermixed. This result is inconsistent with both the PSA and the RAA. In experiment 2 we examined the influence of T1-difficulty on lag-1 sparing. While neither the RAA nor the PSA suggest that T1-difficulty should influence lag-1 performance, the FSA predicts that bottom-up differences in T1 processing partially determine the influence of T1-difficulty on the AB. Therefore, differences should be observable as soon as T1-processing commences. As predicted by FSA, increasing T1-difficulty increased the severity of the AB (F(5,100) = 3.122, p = .02) and decreased accuracy at lag-1 (t(20) = 5.016, p <.001). These results support the FSA of T1-difficulty, which suggests that attention during the AB is flexible and that T1-difficulty is one factor that modulates this flexibility.
Meeting abstract presented at VSS 2014