Xu Jianzhe¹ (), Jiang Fei¹, You Chenyun¹, Xu Jiyu¹, Zhou Jifan¹, Chen Hui¹, Shen Mowei¹; ¹Zhejiang University

The human mind can effortlessly construct mental representations defined by the intricate web of visual relations among its constituent objects and hold them in working memory for subsequent perceptual processing. To identify an online monitoring method for relation memory, a suitable event-related potential (ERP) index is required. Contralateral delay activity (CDA) stands out as a robust, feature-independent index of the content stored in visual working memory (VWM), and our hypothesis posits its ability to track the memory of visual relations. The current work adapted dynamic relation stimuli from Shen et al. (2021, Experiment 2) to test our hypothesis. Each dynamic relation was demonstrated by the movement of two objects governed by a specific kinematic equation. Up to four dynamic relations were sequentially displayed on either side of the screen, each lasting 1000ms, with a 200ms interstimulus interval (ISI) during the sequential display and a 1000ms ISI preceding the test phase. Participants were instructed to memorize one side of dynamic relations based on a memory cue and determine if the dynamic relation of a specified color pair changed in the test phase. We replicated prior behavioral findings indicating that the average upper limit of relation memory (measured with K-max) is two dynamic relations. The ERP data revealed a different pattern from the behavioral results: CDA progressively increased with the number of displayed relations, instead of reaching a peak after storing the second relation. Furthermore, we observed a positive correlation between the difference in CDA amplitude and individual working memory capacity. We concluded that: (1) the maintenance of dynamic relations induces CDA; (2) the VWM is capable of storing more than two dynamic relations, a result not evident in the behavioral findings, likely due to insufficient memory quality in the high load condition for performing the change-detection task.

Acknowledgements: This research was supported by National Natural Science Foundation of China [Grant 32071044; 32371088; 62337001], Science and Technology Innovation 2030-“Brain Science and Brain-like Research” Major Project (2022ZD0210800).