The sleep benefit in episodic memory: Investigating underlying mechanisms

Berres, Sabrina

PDF
SabrinaBerres_Dissertation.pdf - Published
Download (2MB)

URN:	urn:nbn:de:bsz:180-madoc-647068
Document Type:	Doctoral dissertation
Year of publication:	2023
Place of publication:	Mannheim
University:	Universität Mannheim
Evaluator:	Bröder, Arndt
Date of oral examination:	19 June 2023
Publication language:	English
Institution:	School of Social Sciences > Kognitive Psychologie (Seniorprofessur) (Erdfelder 2019-)
Subject:	150 Psychology
Keywords (English):	episodic memory , meta-analysis , multinomial processing tree modeling , psychology , sleep
Abstract:	Sleep not only benefits health and well-being but is also considered a prerequisite for proper episodic memory functioning. Even though almost a century of research has documented a positive effect of sleep in episodic memory, our knowledge about the underlying mechanisms is still limited. The overall goal of the present thesis is to improve our understanding of the mechanisms underlying the sleep benefit by introducing an integrative framework that incorporates prominent theories from neuroscience and psychology. Because a quantification of the sleep benefit and a systematic investigation of potential moderator variables has been missing, I addressed these points in a comprehensive, up-to-date meta-analysis. The results revealed a robust effect of moderate size. Furthermore, sleep benefits in episodic memory are significantly moderated by retrieval procedures, the definition of the dependent variable, and sleep study designs. On the basis of the meta-analytic results, I developed an integrative framework that provides a broader and more comprehensive explanation of the sleep benefit than earlier theories. Essentially, it combines active systems consolidation, opportunistic consolidation, and interference accounts (see Article 1; Berres & Erdfelder, 2021). Using multinomial processing tree (MPT) models, the two additional original research articles of this thesis focus on testing core assumptions of the theories incorporated in the integrative framework. This methodological approach allows a more thorough investigation of hypothesized underlying mechanisms by providing uncontaminated measures and disentangling the mechanisms’ joint contribution to the sleep benefit. According to the active systems consolidation account, information salience induced by encoding strength and test expectation should moderate sleep benefits in episodic memory storage. However, previous research of Howe (1970) and McGeoch (1929) suggests that encoding strength also affects interference-based sleep benefits in memory retrieval. In line with the integrative framework, the results showed that sleep improves both episodic memory storage and retrieval. Moreover, whereas encoding strength moderates sleep benefits in retrieval rather than in storage, test expectation appears to have no effect on neither storage nor retrieval (see Article 2; Berres & Erdfelder, 2023). The active systems consolidation account also predicts that sleep improves source memory for retention intervals of up to 12 hr (Inostroza & Born, 2013; Klinzing et al., 2019). However, previous empirical evidence for sleep benefits in source memory is quite mixed. One of the possible reasons for the conflicting results is that the source memory measures used likely confound item memory, source memory, and guessing. Therefore, I reexamined the sleep-strengthens-source-memory hypothesis with a multinomial modeling approach. Whereas findings for sleep benefits in item memory were mixed, results for source memory were in line with the active systems consolidation account. Specifically, sleep within a 12-hr retention interval improved source memory for spatial positions (see Article 3; Berres et al., 2023). Hence, the three articles included in this thesis indicate that both active systems consolidation and interference accounts are required to fully explain the pattern of results. Moreover, the meta-analytical results provide guidance to increase the replicability of sleep-induced episodic memory benefits. Finally, I also demonstrated the power of multinomial modeling in investigating underlying mechanisms of the sleep benefit. In sum, this thesis improves our understanding of the underlying mechanisms and contributes to the theoretical progress in sleep and memory research.