How Constructed Environments Shape Visual Search Performance

Poster Presentation 36.433: Sunday, May 17, 2026, 2:45 – 6:45 pm, Pavilion
Session: Visual Search: Features, scenes, real-world stimuli

Yoshiyuki Ueda1, Kikuno Iizuka1, Ryusuke Nakai1; 1Kyoto University

Visual search studies have traditionally investigated how the cognitive system operates within a given environment, implicitly assuming that the observers play a passive role with respect to their surroundings. However, in real-world contexts, environments are not simply given; rather, they are often actively constructed by people. In this study, we investigate the relationships between environment construction and subsequent visual search processes within a single experimental framework. To address this issue, we developed a constructing environment–visual search task. In the constructing environment task, participants arranged twelve visual objects in ways they believed would make searching easier. Under the same instruction, each participant created four types of search displays. Afterward, participants performed a visual search task in which the target appeared in displays constructed by (i) themselves (self-constructed condition), (ii) another person (other-constructed condition), or (iii) randomly (random condition). Displays in the self- and other-constructed conditions were presented repeatedly to examine learning efficiency (i.e., the contextual cueing effect). The results showed that reaction times decreased more strongly in both the self-constructed and other-constructed conditions than in the random condition, although no difference was observed between the self-constructed and other-constructed conditions. Moreover, accuracy was significantly higher in the other-constructed condition than in the self-constructed and random conditions. These results suggest that constructing environments may interfere with efficient visual processing or introduce memory-related confusion, whereas environments constructed by others may facilitate more straightforward processing. Preliminary findings from a brain-activation study further support that self-constructed displays elicit broader activation, particularly occipital regions.

Acknowledgements: JSPS Kakenhi #25K00900