Abstract
Recent work in our lab has shown that participants can adopt an attentional control set (ACS) for 30 visual objects, indicating that the contents of ACSs are stored in long term memory (LTM). This finding raises a question: What is unique about ACS representations in LTM that allows them to influence attentional capture, when most LTM representations do not? One proposition is that ACS representations are stored with greater than normal baseline activation, a state referred to as activated LTM (ALTM). In the present study we evaluated this proposition by testing whether ACS items exhibit a signature of ALTM: an intrusion effect in a working memory change detection task. Specifically, if ACS representations are maintained in ALTM, participants should be slow to correctly reject these items when they appear as the probe on "change" trials during this task. For our study, participants memorized 30 images of everyday visual objects and then completed two tasks, randomly mixed across trials: spatial blink trials (to induce an ACS for the memorized objects and to test for contingent capture), and visual working memory trials (to test for an intrusion effect). Replicating our previous contingent capture findings, on spatial blink trials, ACS objects captured attention more than non-ACS objects. On working memory trials, ACS objects produced an intrusion effect and non-ACS objects did not. This pattern supports the conclusion that the contents of ACSs are maintained in ALTM. More broadly, the present findings add to the growing evidence that LTM has rapid attentional effects during perceptual processing, and that these effects are regulated through differential activation of LTM representations.
Meeting abstract presented at VSS 2017