Abstract
Visual search is often facilitated when the search display occasionally repeats, even though participants are unaware of the repetition. A prominent explanation of this memory-based search, known as contextual cueing (Chun & Jiang,1998),isthat it is a form of top-down, associative learning of display configurations and target location. However, recent findings emphasizing the importance of local context near the target (Brady & Chun, 2007; Olson & Chun, 2002) give rise to the possibility that low-level local-repetition priming may account for the effect. This study distinguishes these alternatives by testing whether search is guided towards a target's expected location, even when the target is relocated to another location. This manipulation allows us to directly test the associative nature of contextual cueing. After participants searched for a T among Ls in displays that repeated 24 times, they completed a transfer session where the target was relocated. In Experiment 1, the target moved to a previously empty location positioned near (2.16°) or far away (4.32°) from the trained target location. Search in the near-relocation condition was faster than search in novel displays, suggesting that the contextual guidance is not spatially precise. However, the contextual cueing benefit was eliminated in the far-relocation condition. Experiment 2 used a similar design, except that the target swapped locations with a near or far distractor during the transfer session. Contextual cueing was abolished when the target swapped locations with a near distractor, and was reversed into a contextual cost (slower RT than the new condition) in the far-swap condition. Together, these findings show reduced contextual cueing when targets move away from their expected locations. We conclude that target predictability is a key factor in memory-based attentional guidance, supporting a top-down, associative learning account.
This study was supported by NIH MH071788.