Abstract
Attention capture by perceptually salient items and by distractors that have affective value has been reported in the literature. Whether, and under what circumstances, affectively neutral materials, encoded into long-term memory (LTM) might also be subject to attentional prioritization remains relatively unexplored. To address this gap in the literature, participants who completed the current investigation encoded 36 scene-object pairs prior to completing a basic attention task; the objects were simple colored circles. During attention task performance, eye movements were recorded as participants were presented with multiple object displays. Upon presentation of each display, participants were instructed to immediately fixate a target object (e.g. a square), ignoring all of the remaining display elements (i.e. colored circles). Attention task displays were of three types. Baseline trials consisted of 6 objects, none seen during the corresponding encoding phase. No Cue trials consisted of 6 objects, one of which had been studied. Cue trials were preceded by the presentation of a studied scene, and consisted of 6 objects, one of which was the studied associate of the scene cue. It was expected that capture would be greatest following scene cues, which were expected to trigger retrieval and active representation of the associated object. Consistent with this prediction, results indicated that saccades to irrelevant distractors (i.e. encoded and to-be-ignored objects) occurred more often in the presence of scene cues. Furthermore, saccades that went directly to targets, as instructed, were initiated more slowly in the cue condition than in no cue and baseline conditions; differences in saccade initiation time to no cue and baseline trials were not significant. Collectively, these results indicate that long-term memories can capture attention, but that this outcome may be limited to situations in which the memory representation has been recovered and is active prior to presentation of the corresponding attention displays.
Meeting abstract presented at VSS 2016