Chaoxiong Ye^1,2 (), Lijing Guo^1,2, Edward Awh³, Edward Vogel³, Qiang Liu^1,2,4, Weizhen Xie⁵; ¹Anyang Normal University, China, ²University of Jyvaskyla, Finland, ³University of Chicago, ⁴Sichuan Normal University, China, ⁵University of Maryland

Some images, such as faces and scenes, reliably embed themselves in memory, while others are easily forgotten. This phenomenon is often attributed to image memorability, a stimulus-driven property that predicts consistent memory responses across individuals. While this phenomenon is often studied in long-term memory, recent research suggests that image memorability also influences visual working memory (VWM). However, it remains unclear whether this influence stems from facilitated VWM formation or enhanced maintenance that extends VWM storage capacity. To address this, we recorded scalp EEG from 35 participants performing a change-detection task involving images of faces with varying memorability (high vs. low). While maintaining central fixation, participants briefly viewed two faces (100 ms) on one of the visual fields and, after a short delay, were tested for memory accuracy. Consistent with prior findings, high-memorability faces were associated with better behavioral performance. We further analyzed the contralateral delay activity (CDA), an ERP component associated with VWM formation and maintenance. Our results demonstrated that CDA amplitudes at posterior electrode sites were significantly larger for high-memorability stimuli during the early retention period (300–450 ms post-stimulus). However, this difference diminished during the later retention period (850–1000 ms post-stimulus). These findings suggest that high-memorability stimuli enhance early-stage encoding processes, resulting in more robust memory representations without altering the overall storage capacity of VWM. By identifying the specific processing stage where image memorability impacts VWM, this study highlights its role in facilitating VWM formation rather than maintenance.

Acknowledgements: This work was supported by the Research Council of Finland (former Academy of Finland) Academy Research Fellow project (#355369 to Chaoxiong Ye) and Finnish Cultural Foundation (#231373 to Chaoxiong Ye).