Abstract
Attention naturally fluctuates when it needs to be sustained for long periods of time. These fluctuations have immediate consequences for performance on current tasks, but their consequences for the encoding of current information into visual long-term memory (VLTM) are less well understood. In particular, although the effects of selective attention and divided attention on memory have been explored, the relationship between sustained attention and memory is less clear. We hypothesized that, in a sustained attention task, being in an attentive state just prior to the presentation of an image would lead to better subsequent VLTM for that image. Based on previous studies of sustained attention, we operationalized attentiveness as slower and less variable response times (RTs). In a first incidental encoding phase of the experiment, observers categorized trial-unique pictures of scenes as indoor or outdoor by pressing one of two buttons. Critically, the scenes were selected to be 90% from one subcategory and 10% from the other, which encouraged observers to lapse into making the more frequent response habitually, thereby inducing fluctuations in sustained attention. In the second phase of the experiment, observers performed a surprise recognition memory test in which they were shown old images (that had appeared in the first phase) and new images. They were asked to decide whether they had seen each image before, as well as to report their confidence. Using these memory responses, we returned to the sustained attention task data and found that stimuli that were subsequently remembered vs. forgotten were indeed preceded by trials with slower and less variable RTs. These results suggest that the dangers of attentional failures may not only be realized in the moment, but also in the future when we need to access memories from those episodes.
Meeting abstract presented at VSS 2015