Li Gong^1,2 (), Penglan Liu², Dan Nie², Chaoxiong Ye^1,2; ¹School of Education, Anyang Normal University, China, ²Department of Psychology, University of Jyvaskyla, Finland

Visual working memory (VWM) is a core cognitive function in humans, with memory representations flexibly stored in distinct representational states (online and offline) depending on task demands. Among these, the offline storage state plays a crucial role in stably maintaining memory representations and supporting the online storage state. Previous research has demonstrated that an individual's sleep quality can predict performance in VWM tasks. However, these studies have primarily focused on the online storage state of VWM. It remains unclear whether sleep quality affects the offline storage state, and whether its influence differs between the two storage states. To address thess questions, the present study employed a mixed experimental design, combining the Pittsburgh Sleep Quality Index (PSQI) with a sequential presentation retrieval paradigm to examine the effects of sleep quality on different storage states of VWM. The results revealed significant main effects of both sleep quality and memory storage state on VWM accuracy. Moreover, a significant interaction between sleep quality and memory storage state was observed. Simple effect analyses indicated that, in the offline storage state, participants with high sleep quality exhibited significantly higher accuracy than those with low sleep quality. These findings suggest that sleep quality differentially affects the distinct storage states of VWM. Specifically, when individuals experience poor sleep quality, the offline storage state appears to be more adversely impacted than the online storage state. This study provides direct evidence supporting the hypothesis of a functional dissociation between the online and offline states of VWM, and further elucidates the intrinsic relationship between sleep quality and VWM. The findings underscore the critical role of sleep in cognitive functioning and contribute to a deeper understanding of this domain.

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).