The relationship between resting‐state amplitude fluctuations and memory‐related deactivations of the default mode network in young and older adults

Abstract

The default mode network (DMN) typically exhibits deactivations during demanding tasks compared to periods of relative rest. In functional magnetic resonance imaging (fMRI) studies of episodic memory encoding, increased activity in DMN regions even predicts later forgetting in young healthy adults. This association is attenuated in older adults and, in some instances, increased DMN activity even predicts remembering rather than forgetting. It is yet unclear whether this phenomenon is due to a compensatory mechanism, such as self‐referential or schema‐dependent encoding, or whether it reflects overall reduced DMN activity modulation in older age. We approached this question by systematically comparing DMN activity during successful encoding and tonic, task‐independent, DMN activity at rest in a sample of 106 young (18–35 years) and 111 older (60–80 years) healthy participants. Using voxel‐wise multimodal analyses, we assessed the age‐dependent relationship between DMN resting‐state amplitude (mean percent amplitude of fluctuation, mPerAF) and DMN fMRI signals related to successful memory encoding, as well as their modulation by age‐related hippocampal volume loss, while controlling for regional grey matter volume. Older adults showed lower resting‐state DMN amplitudes and lower task‐related deactivations. However, a negative relationship between resting‐state mPerAF and subsequent memory effect within the precuneus was observed only in young, but not older adults. Hippocampal volumes showed no relationship with the DMN subsequent memory effect or mPerAF. Lastly, older adults with higher mPerAF in the DMN at rest tend to show higher memory performance, pointing towards the importance of a maintained ability to modulate DMN activity in old age.

Single-value fMRI scores reflect both stage and risk across the Alzheimer’s disease continuum

“…As such, they mirror previously described activation differences between healthy older adults and individuals with MCI or AD 31–34,24 . Qualitatively, these patterns are similar to memory-related fMRI activation differences between healthy young and older adults 25,29,51 . One explanation for the observed pattern would thus be that the progressive deterioration of memory-related brain activity across the AD risk spectrum might reflect accelerated neurocognitive aging.…”

“…In addition to the opposite directions, the SAME score thus also considers novelty- and memory-related deactivations, characterized by, e.g., negative BOLD responses in the DMN 24,51 . For more information on the calculation and interpretation of the scores, see the original descriptions (cf.…”

Age-related differences in fMRI subsequent memory effects are directly linked to local grey matter volume differences

“…An inverse relationship between GMV and SMV was found in the precuneus and medial PFC (Table 1). This may be best explained by the fact that these regions show negative subsequent memory effects (Kizilirmak et al, 2023), such that the correlation with absolute BOLD amplitudes would be positive in these regions. Other studies failed to find an association between GMV and fMRI contrasts (Pur et al, 2019), which may be due to the use of ROI-wise GMV means, pointing out the importance of assessing the regional variation in a more fine-grained, e.g., voxel-wise manner.…”

“…The design matrix is shown in Figure 1D. The model represents an adaptation of model 2 from Kizilirmak et al (2023):…”

“…Direct conclusions on age-related differences in the functional recruitment of brain regions can only be drawn when considering potential links to local differences in brain structure. Here, building on previous analyses of the same data set (Kizilirmak et al, 2023), we assessed the direct, voxel-wise, relationship between functional magnetic resonance imaging (fMRI) correlates of successful episodic memory encoding (subsequent memory effects, SME) and grey matter volume (GMV) in a large sample of young and older healthy adults using multimodal neuroimaging.…”