The contribution of visual working memory to priming of pop-out

Priming of Pop-out (PoP);Visual Working Memory (VWM);Visual Attention

Ahn, JeeWon

关键词: Priming of Pop-out (PoP);    Visual Working Memory (VWM);    Visual Attention;   

Others  :  https://www.ideals.illinois.edu/bitstream/handle/2142/44782/JeeWon_Ahn.pdf?sequence=1&isAllowed=y

美国|英语

来源: The Illinois Digital Environment for Access to Learning and Scholarship

【 摘 要 】

Priming of pop-out (PoP) refers to the facilitation in performance that occurs when a target-defining feature is repeated across consecutive trials in a pop-out singleton search task. While the underlying mechanism of PoP has been at the center of debate, a recent finding (Lee, Mozer, & Vecera, 2009) has suggested that PoP relies on the change of feature gain modulation, essentially eliminating the role of memory representation as an explanation for the underlying mechanism of PoP. The current study aimed to test this proposition to determine whether PoP is truly independent of guidance based on visual working memory (VWM) by adopting a dual-task paradigm composed of a variety of both pop-out search and VWM tasks. First, Experiment 1 tested whether the type of information represented in VWM mattered in the interaction between PoP and the VWM task. Experiment 1A aimed to replicate the previous finding, adopting a design almost identical to that of Lee et al., including a VWM task to memorize non-spatial features. Experiment 1B tested a different type of VWM task involving remembering spatial locations instead of non-spatial colors. As a result, only spatial VWM load interfered with PoP, supporting the notion that VWM is involved with PoP. Based on the results from Experiment 1, Experiment 2 observed the magnitude of PoP, with manipulation of the number of items to be remembered in the VWM task. With spatial VWM load, the amount of PoP decreased with the increase of memory load. This result indicates that the concurrent spatial VWM task truly interfered with the occurrence of PoP. Experiment 3 examined whether a VWM task still affects PoP when a strong bottom-up guidance is present, and target ambiguity (Meeter & Olivers, 2006) decreases. Consistent with previous findings, the increase of set size resulted in a decrease of PoP, indicating that there was a strong bottom-up guidance based on the salience of target. Nonetheless, the dual-task interference from the spatial VWM task was consistently observed. The result suggests that the contribution of VWM to PoP is robust, regardless of the presence of other types of guidance. Lastly, Experiment 4 directly investigated which component of a spatial VWM task interacts with PoP. To address this question, Experiment 4 observed the interaction between PoP and the VWM task with and without an inter-trial repetition of the target-related positions. Based on the finding that the repetition of position also elicited inter-trial priming (Maljkovic & Nakayama, 1996), Experiment 4 tested whether the repetition of the relative position of the target affected the effect of a spatial VWM load on PoP. As a result, while PoP interfered with a VWM load, position-based priming appeared to be independent of the VWM load as well as PoP. Such results indicate that spatial VWM exclusively contributes to PoP by swiftly deploying attention to the location of a task-relevant feature held in memory. In sum, across four experiments, PoP was affected by the concurrent spatial VWM task. Together, overall results suggest that the function of VWM contributes to the occurrence of PoP, disagreeing with the previous finding that PoP is due to the increasing gain of perceptual features, as attended in previous trials. Instead, current findings support the theory that PoP is in fact a multi-level process based on the deployment of spatial attention relying on VWM.