Susan Ajith^1,3 (), Daniel Kaiser^1,2, Lu-Chun Yeh¹; ¹Justus Liebig University Gießen, ²Center for Mind, Brain and Behavior (CMBB), Philipps University Marburg, Justus Liebig University Gießen, and Technical University Darmstadt, ³Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig

When searching for an object, search templates derived from prior knowledge guide attention towards likely targets. Objects with consistent visual appearance, such as basketballs, are likely to activate narrow search templates. In contrast, objects with high variability, like bags (which feature various shapes and colors), are likely to activate broad templates that account for their real-world diversity. To investigate how such narrow or broad search templates affect search for everyday objects, we conducted a visual search study, where participants viewed a word cue followed by a search display that did or did not contain the target. Targets were selected from objects spanning a range of variability in appearances, from small (e.g., basketball or tennis racket) to large (e.g., bag or lamp) variations. To quantify the variability in participants’ mental representation of the objects, participants additionally performed a drawing task, where, before the search experiment, they drew four exemplars for each of the target objects. By assessing dissimilarities among the drawn exemplars, this task allowed us to gauge the variability within each object. Results showed that response times in the search task varied systematically with the variability in the drawings: When participants produced less variable exemplars during drawing (evaluated using a deep neural network model), they were faster during search, while more variable drawings led to slower search. This was true for target-present and target-absent trials. Crucially, differences in search performance linked to template variability were significant even when only the initial trials per object were considered. This suggests that real-world object variability impacted search from the outset, before participants could estimate the variability of the target objects in the experiment. Together, our findings demonstrate that search templates for real-world objects are inherently shaped by the variability of an object in the world.

Acknowledgements: SA is funded by the JLU graduate scholarship. LCY is supported by the MSCA programme (101149060). DK is supported by the DFG (SFB/TRR135,222641018; KA4683/5-1, 518483074, KA4683/6-1, 536053998), “The Adaptive Mind”, and an ERC Starting Grant (PEP, ERC-2022-STG 101076057).