Huixin Song¹ (), Xinran Chen¹, Beibei Cai¹, Yingtao Fu¹, Mowei Shen¹, Hui Chen¹; ¹Zhejiang University

Attentional templates maintained in either working memory (WM) or long-term memory (LTM) can guide visual search. However, a definitive conclusion remains elusive regarding the attentional guidance efficiency of WM and LTM templates, attributed to diverse experimental designs and measures. This study systematically compared the attentional guidance of WM and LTM templates across various contexts (independent/competitive) and template loads, employing a unified and canonical measurement - search slope. In the first section, the comparison was made in an independent context using a blocked design, preventing interference between WM and LTM templates. Results consistently indicated similar search efficiency and sensitivity to load manipulations for both WM and LTM templates (Experiments 1 and 2). Moreover, we ruled out the alternative explanation that the observed comparable search efficiency resulted from retrieving LTM templates into the WM system during search (Experiment 3). Instead, it indicated that LTM templates directly guided search with an efficiency equivalent to WM templates. The second section introduced a competitive context with a mixed design, incorporating interference or competition between WM and LTM templates. Findings revealed that compared to LTM templates, WM templates exhibited superior search efficiency and were less impaired by increased template loads, suggesting WM templates hold an advantage in the competitive context (Experiment 4). Furthermore, it was demonstrated that this advantage is a result of an unequal prioritization between WM and LTM templates in the competitive context (Experiments 5 and 6). Overall, these results suggest that attentional templates maintained in WM and LTM are fundamentally the same in their ability to guide search, but their search efficiency may be influenced by their relative priorities in contexts involving interference or competition.

Acknowledgements: This work was supported by grants from Science and Technology Innovation 2030-“Brain Science and Brain-like Research” Major Project (2022ZD0210800), National Natural Science Foundation of China (32171046, 32071044, 32200844)