Structural and functional substrates of associative memory load in aging - PROJECT SUMMARY
Healthy older adults and individuals with Alzheimer’s disease and related dementias show the most severe
memory deficits for associative memory (AM), in which multiple elements must be remembered as one
cohesive event (e.g., items on a grocery list). Neuroimaging research has separately related age-related
declines in AM to increased neural activity in the hippocampus and lower integrity of white matter tracts directly
emanating from the hippocampus (e.g., fornix). However, very few aging studies have assessed interactions
between these functional and structural neural substrates and none have done so in relation to AM. In addition,
because most prior AM research has used paradigms that involve remembering at most only pairs of stimuli,
the effect of aging on more demanding associations remains unknown. Neurocognitive aging theories predict
that older adults cannot recruit sufficient neural resources under more demanding cognitive loads, leading to
lower neural activity and poorer cognitive performance than younger adults. Whereas this theory has mainly
been tested with working memory paradigms and in relation to neural activity, it follows that a similar
mechanism may underlie effects of AM load, and that the behavioral and neural responses may be mediated
by brain structure. To address these limitations, we propose to characterize age differences in AM and the
functional and structural neural substrates of AM as a function of Associative Load. Forty younger and 40 older
healthy adults will perform an AM task with word pairs (low load) and triplets (high load) during functional
magnetic resonance imaging (fMRI) acquisition followed by a multicompartment diffusion imaging (MDI)
sequence. We will assess age group differences in hippocampal activity during low and high load conditions
using fMRI (Specific Aim 1) and in fornix integrity using MDI (Specific Aim 2), and how these neural substrates
individually relate to AM performance. To test our novel hypothesis that the neural response to AM load is
constrained by white matter structural integrity, we further propose to explore the moderating effects of age
group and AM load on the direct relationship between hippocampal activity and fornix integrity (Specific Aim 3).
In addition to advancing our understanding of AM deficits in healthy aging and the contribution of the fornix and
hippocampus to memory performance, this proposal will support the applicant’s training aims of learning
additional theoretical (e.g., cognitive neuroscience of aging), statistical (e.g., univariate and multivariate
approaches), and methodological (e.g., advanced diffusion imaging) skills. Ultimately, this work will identify
behavioral and neural markers of memory declines in healthy aging, thereby informing future work using these
markers to detect older adults at-risk for Alzheimer’s disease and related dementias. By focusing on complex
structure-function interactions, this work will also inform the field as it moves toward large-scale, multimodal
imaging datasets.
