Abstract
What do we remember from an extremely brief snapshot? Typically, the 'gist' of a scene is grasped while relatively little visual detail is perceived and retained in long-term memory. Here, we investigated whether contextual and/or functional associations among objects may reduce stimulus competition and enhance long term memory of visual details when images are merely glimpsed. Participants viewed pairs of contextually-related and unrelated objects (e.g., a kettle and a mug; a shovel and a vase, respectively), presented for an extremely short exposure duration (24 ms, masked). Subsequently, participants performed a memory-recognition test, in which one of two objects within a pair was replaced by a novel object from the same basic category. Participants differentiated old objects from novel object exemplars, while these were presented with their original counterpart pair object. Results demonstrated higher levels of correct recognition for contextually-related than for unrelated object pairs. Furthermore, when object stimuli in the recognition test appeared alone, i.e., without a corresponding pair object serving as a memory-retrieval cue, results remained virtually identical. Namely, memory for specific visual details remained higher for objects initially appearing within contextually-related, than unrelated, object pairs. Finally, the nature of memory enhancement for contextually/functionally related object pairs was examined. Previous research has suggested that objects which are associated by action relations are particularly bound to perceptual integration. Based on an independent survey in which participants rated the extent to which pairs of objects depicted action relations, we found a memory advantage for contextually-related relative to unrelated items regardless of active/passive contextual associations. Taken together, our results suggest that during an extremely brief visual glance, contextually associated stimuli benefit from reduced object-to-object competition. Consequently, stimuli are 'bound' within a unified representation, allowing enhanced encoding and memory of their gist as well as of their perceptual visual details.
Meeting abstract presented at VSS 2014