Theta-gamma phase-amplitude coupling during naturalistic memory encoding: a comparison of real-world and 2D experiences

Abstract

Episodic memory allows individuals to recall personal experiences by integrating perceptual, emotional, and contextual information over time. Previous research suggests that episodic memory depends on the coordination of neural activity across different frequency bands, particularly through theta-gamma phase-amplitude coupling (PAC). However, most studies have examined these mechanisms using simplified laboratory tasks, with limited attention to how they operate in more naturalistic contexts. The present study investigated how encoding context influences neural dynamics during episodic memory formation by comparing a real-world performance with a two-dimensional (2D) version of the same narrative. Sixty-two young adults were assigned to either a real-world or 2D condition. Neural activity was recorded using electroencephalography (EEG) during encoding, and PAC between frontal theta (4-8 Hz) and parietal gamma (30-100 Hz) oscillations was computed. Following encoding, participants completed a free verbal recall task. The results showed a significant effect of encoding condition on PAC, with higher theta-gamma coupling observed in the 2D condition compared to the real-world condition. Despite these neural differences, recall performance did not differ significantly between conditions. There was a significant positive correlation between PAC and memory recall performance only observed in the 2D condition. Furthermore, PAC was not consistently related to general cognitive measures, suggesting that it reflects task-specific neural processing rather than stable cognitive ability. These findings show that the neural coordination underlying memory encoding is influenced by the context in which information is experienced, even when similar levels of memory performance are achieved across different conditions. In other words, increased PAC in the 2D condition may reflect greater reliance on internally driven processing when external structure is reduced. Overall, this study highlights the importance of using naturalistic paradigms to better understand how memory operates in real-world settings and suggests that PAC should be interpreted as a flexible, context-dependent neural mechanism rather than a general measure of memory performance.