Abstract
The relationship between working memory (WM) and visual search has been heavily debated in the literature. Although behavioral results have been mixed, one proposed role of WM in search is to guide attention toward novel locations through the maintenance of previously rejected distractors (Peterson, et al., 2001). In previous research, we failed to find evidence for this relationship using the Contralateral Delay Activity (CDA) to measure the contents of WM during search for real world objects (Williams & Drew, 2017). However, distractor maintenance might only be necessary in search tasks with targets and distractors that have a high degree of feature overlap. In order to address this possibility, the current study used Landolt C stimuli rather than real world objects. On each trial, participants were shown a novel target for 1000 ms followed by an 800-1000 ms retention interval. The primary task was to indicate if the target was present (50%) or absent (50%) in a lateralized search display. The number of lateralized items in the search array varied between 2, 4, or 6. ERPs were time-locked to the onset of the search array, and the CDA was measured between 300-800 ms. The CDA is an index of the contents of WM and increases in amplitude with the number of items maintained (Vogel & Machizawa, 2004). Therefore, if previously rejected distractors are held in WM, the amplitude of the CDA should increase with the number of distractors in the search array. This pattern should be particularly evident during target absent trials because each item needs to be evaluated before a response can be made. However, there were no differences in the CDA amplitude across set sizes. Together, these studies suggest that, regardless of the level of similarity between targets and distractors, rejected distractors are not maintained in WM during search.
Meeting abstract presented at VSS 2018