Mikayla N. Krech¹ (krech036@umn.edu), Roger W. Remington¹, Vanessa G. Lee¹; ¹University of Minnesota

Visual search is faster on displays that occasionally repeat (contextual cueing, CC). Traditionally, CC is attributed to learning the spatial context of a search array independent of item identities. However, research using real-world objects showed that CC emerged only when both item identities and locations were repeated (Makovski, 2016). Building on these findings, we test whether identity repetition is necessary when (1) the stimuli do not perturb the perceived global configuration via changes in elongation or axis orientation, and (2) the task does not require identity-based search. Participants searched for a target among distractor items over 24 trial blocks. Across blocks, displays were either fully repeated (All-Repeat), repeated in locations but not identities (Location-Repeat), or entirely new (New-Display). Experiments 1 and 2 tested Chinese readers using Chinese characters—square-like stimuli with relatively homogeneous elongation and axis orientation. Participants performed either an identity search (finding an animal character; Experiment 1) or a non-identity search (finding a faint dot superimposed on a character; Experiment 2). Experiment 3 tested English speakers using visually variable common objects in the identity-irrelevant dot-search task. We examined whether CC depended on identity repetition when identities were neither task-relevant nor visually distinctive. In Experiment 1, CC emerged only in the All-Repeat condition; Location-Repeat yielded no CC, suggesting that in identity search tasks, identity and location were jointly learned even with visually homogeneous stimuli. In the dot-search task of Experiment 2, Location-Repeat alone was sufficient to produce CC, indicating task modulation. In Experiment 3, however, Location-Repeat failed to produce CC when stimuli were visually distinctive, even though identity was task-irrelevant. These findings demonstrate that identity information is incorporated into spatial contextual cueing except when it is both visually nondistinctive and irrelevant to the search task.

Acknowledgements: Supported in part by the McKnight Foundation and a University of Minnesota Graduate Student Research Award.