Neural correlates of recognition memory obtained after encoding under conventional laboratory, virtual, and real-life...

Traditional psychological laboratory research provides fundamental insights into the functioning of human recognition memory. However, concerns have been raised whether recognizing simplified two-dimensional (2D) laboratory stimuli accurately reflects real-world retrieval processes of their three-dimensional (3D) physical counterparts. To explore potential differences in mnemonic processing, we conducted a comparative study using a conventional recognition memory task after encoding objects on a 2D screen (PC) or in a physical replica of the experimental setup. Additionally, the encoding environment was recreated in virtual reality (VR) to further bridge the realism gap between physical reality and screen presentations. Within all conditions, behavioral results were complemented by memory-sensitive indices in the EEG (i.e., ERPs and oscillatory brain activity). Our results revealed overall high memory performance. Notably, the EEG indices indicated that different subprocesses contributed to successful recognition depending on the degree of realism. Specifically, while a parietal ERP old/new effect was observed across conditions, only the PC condition yielded a significant mid-frontal ERP old/new effect. Furthermore, the lower theta band response reliably marked successful recognition in all conditions, while the upper theta band response indicated more effortless recognition in the VR condition compared to the PC condition, with physical reality lying in between. Importantly, comparable alpha band responses across conditions suggest that these differences cannot be solely attributed to attentional resource allocation. Concludingly, this study provides evidence for modality-specific mnemonic processing characteristics as a function of realism. It advocates for the application of VR in psychological research to efficiently investigate lifelike cognitive processing.