Abstract
Theories of visual attention claim that representations in visual working memory (VWM) are able to control attention as we search for targets embedded in arrays of other objects. However, recent findings have shown that VWM is only used to control attention when the target is new, whereas long-term memory (LTM) takes control of visual attention after a handful of trials searching for a certain target. This study tested whether VWM would take the control of attention back from LTM if participants were expecting to search for a new target. The task was a visual search task in which participants were shown a target item (a Landolt-C) for 250 ms. After a 900 ms delay, an array of items was presented. Participants were instructed to respond with a button press indicating if the target item was present or absent. The key manipulation was the number of consecutive trials that subjects would be cued to search for the same target. On 75% of same-target runs, the subjects searched for the same object for 4 trials, with only 25% of same-target runs being 5 trials long. As a result, subjects should expect VWM to need to take back attentional control after 4 trials. We used the contralateral delay activity (CDA) and the anterior P1 (or P170) component as a measure of whether the target was stored in VWM or LTM. These event-related potentials replicated previous studies, showing a trade off of attentional control from VWM to LTM during the first four trials. Contrary to the prediction that VWM would take attention control back from LTM when a target change was likely, we found no change in reaction time, CDA, or P170 amplitude between trials 4 and 5. In summary, participants do not switch back to using VWM to control attention until absolutely necessary.
Meeting abstract presented at VSS 2016