Abstract
Memory can serve as a powerful guide for attention when we encounter objects and environments with which we have prior experience. However, relative to other forms of attentional control based on stimulus salience and intentional goals, memory-guided attention has been relatively less studied. A fundamental question concerns how attention is allocated among items that differ in memory strength, in the absence of differences in salience or goal-relevance. To address this question, we examined how novel scenes were processed in the context of other scenes that had or had not been experienced before. Observers were presented with scenes while performing a cover task (detecting occasional inverted scenes). Each trial consisted of three sequential events: the first two events each contained one scene at fixation, which was either the same across both events or different; the third event contained two scenes placed randomly on either side of fixation, which were either both novel (different-novel and same-novel), or contained one novel scene and the scene from the preceding two events (same-repeated). Observers then completed a surprise memory test for the novel items from the third event of each trial. Notably, novel items from same-repeated and same-novel trials differed only in terms of whether a repeated item was present during encoding. We found that subsequent memory was significantly better for same-repeated than same-novel and different-novel trials (which in turn did not differ). These findings suggest that the presence of a repeated item can bias processing toward novel items competing for attention. Moreover, these results stand in contrast to recent demonstrations that attention is drawn toward items actively maintained in working memory. Thus, novel items may be prioritized over recently encountered items in the absence of working memory demands, allowing other forms of memory to contribute, along with salience and goals, to the control of attention.
Meeting abstract presented at VSS 2012