A comparison of sleep-based and retrieval-mediated memory consolidation using sigma-band activity

Sleep and retrieval training are reported to promote memory consolidation and are thought to have similar underlying brain-based mechanisms; however, these mechanisms have yet to be directly compared side-by-side, across equivalent timescales. The current study aimed to see if we could replicate sleep spindles enhancement of weakly encoded memories, across both sleep-based and retrieval-mediated memory consolidation, using EEG sigma power ([~]12-16 Hz) as a proxy for sleep spindles (i.e., sigma-band activity). Thirty subjects (27F, 18-34, M=22.17) participated in four separate sessions where they learnt different sets of 104 object-word pairs. Subjects were then tested on their recognition accuracy of the pairs before and after one of four 120-min memory interventions where EEG was recorded: retrieval training (i.e., cued recall practice), restudy (i.e., pair re-exposure), a nap opportunity, or a wakeful rest. Our results did not replicate an enhancement of weakly encoded memories, moderated by either sleep spindles or sigma power. Instead, posterior regions of sigma power negatively related to subsequent memory outcomes. We also detected a prioritisation to enhance memory outcomes for strongly encoded memories and greater memory outcomes following retrieval training and restudy compared to sleep and wake interventions. These results shed new light upon the role of sigma-band activity to predict subsequent memory outcomes and inform how future studies should measure encoding strength and sigma power. Importantly, our study provides a methodological approach to comparing sleep-based and retrieval-mediated memory consolidation that should be explored across different memory paradigms in future studies.