Reactivation and consolidation of memory traces during post-encoding rest across the adult lifespan

Abstract

Episodic memory is a critical cognitive function that enables the encoding, storage, and retrieval of new information. Memory consolidation, a key stage of episodic memory, stabilizes this newly encoded information into long-lasting brain "storage." Studies using fMRI to investigate post-encoding awake rest holds promise to shed light on early, immediate consolidation mechanisms. Here, we review fMRI studies during episodic memory to document common methods to investigate post-encoding consolidation, such as multivoxel pattern analysis (MVPA) and functional connectivity, and the current state of the science in both healthy younger and older adults. In young adults, post-encoding reactivation of stimuli-specific neural patterns in the hippocampus and its connectivity with cortical and subcortical areas (e.g., visual-temporal cortex, medial prefrontal, and medial parietal cortex) correlate with subsequent memory performance. Conversely, studies in older adults highlight the importance of large-scale brain networks during post-encoding rest, particularly the default mode network (DMN). Alterations in connectivity between the DMN and task-positive networks may help older adults maintain episodic memory. Furthermore, non-invasive brain stimulation techniques can enhance these post-encoding consolidation processes and improve memory performance in both younger and older adults. Notably, a lack of studies has investigated post-encoding memory consolidation in neurodegenerative disorders. This review underscores the importance of understanding how post-encoding neural reactivation and connectivity evolve with age to partially explain age-related declines in episodic memory performance and how such declines can be restored.

Figure 1

Overview of episodic memory paradigms and the two common post‐encoding consolidation analysis approaches. (A) Overview of the experimental structures where subjects undergo MRI scanning, including baseline resting state scan, encoding scans while the participants perform memory encoding tasks, immediate rest scans, and memory tests. (B) Analysis of hippocampal multivoxel activation patterns during encoding and matching these patterns with the post‐encoding rest patterns using MVPA. (C) An approach for assessing consolidation‐related dynamics of hippocampus connectivity from baseline to post‐encoding rest. FC, functional connectivity; MRI, magnetic resonance imaging; MVPA, multivoxel pattern analysis.

Figure 2

Representational similarity analysis of encoding and post‐encoding rest (unpublished data). (A) Visualization of beta‐weighted activation patterns across all voxels within the left hippocampus associated with 16 visual stimuli. Activation patterns for each stimulus were extracted and voxel‐wise correlations were computed between encoding and post‐encoding rest periods. (B) The left panel shows the correlation matrix derived from the beta‐weighted activation patterns and neural activity patterns during immediate post‐encoding rest. The middle panel highlights potential neural replay by applying a threshold of mean + 1.5 standard deviations. The right panel depicts the sum of potential replay events, which may be further analyzed to investigate the relationship between neural replay and memory performance.