Electrophysiological Correlates of Encoding Indicate Altered Processing but Comparable Subsequent Memory Effects Across Virtual Reality and Desktop-Based Conditions.

Successful episodic memory formation emerges from the interplay of perceptual, attentional, contextual, and elaborative processes. The relative contribution of these processes to encoding may vary with the conditions under which information is encountered. In particular, different presentation modalities, like virtual reality (VR) and desktop-based presentations (PC), can convey the same semantic identity of an item while differing in perceptual, spatial, and contextual information. Although VR- and PC-encoded memories have been observed to differ in retrieval processes, it remains unclear whether these differences arise during encoding. The present study addressed this question by comparing successful object encoding across presentation modes. Participants incidentally encoded everyday objects in VR and on a PC, and successful encoding was assessed by relating encoding-related EEG activity to later memory. Subsequently remembered objects elicited reliable ERP subsequent memory effects, linked to semantic, contextual, and elaborative encoding processes, as well as an enhanced theta-band response, indicating contextual integration. However, these effects were not modulated by presentation mode, consistent with comparable memory performance across conditions. In contrast, presentation mode affected encoding activity irrespective of subsequent memory success: VR elicited a more negative early ERP and stronger theta-band activity at frontal sensors than PC-based presentation, potentially reflecting altered object evaluation and enhanced visuospatial integration. These modality-dependent differences could not be attributed primarily to early visual or attentional processing, as they were not accompanied by differences in P1 or alpha-band responses. Thus, VR altered general object processing during encoding, whereas the neural processes predicting later recognition were comparable across presentation modes.