Reorganisation of Cortico-Hippocampal White Matter Pathways in Healthy Ageing: Evidence for Paradoxical Shifts in Structural Connectivity

The hippocampus plays a central role in episodic memory and has been the focus of extensive research over past decades. A substantial body of work has demonstrated that age-related memory decline is linked to changes in how the hippocampus functionally interacts with distributed brain networks. While functional connectivity changes in ageing are well documented, relatively little is known about alterations in the structural connectivity (SC) of the hippocampus, despite its foundational role in supporting communication across neural systems. In this study, we combined high-quality data from the Human Connectome Project and advanced diffusion-weighted imaging (DWI) methods to investigate age-related changes in hippocampal SC. Using a recently developed tractography pipeline that allows greater anatomical specificity than conventional approaches, we systematically compared connectivity patterns between younger (26-30 years) and older (56-60 years) adults. Results revealed reduced hippocampal SC with the entorhinal cortex and medial parietal cortices in older participants, alongside increased SC with anterior temporal areas. This paradoxical pattern suggests that ageing is associated with both vulnerability and reorganisation of hippocampal networks, with increased hippocampal-temporal connectivity potentially reflecting compensatory plasticity in response to reduced posterior medial connection. These findings provide <i>in vivo</i> evidence of cortico-hippocampal structural reorganisation in late middle age, a critical period when pathological processes such as tau deposition are already detectable in cognitively healthy individuals. More broadly, they demonstrate the power of our anatomically refined tractography pipeline as a 'proof of concept' for detecting subtle, regionally specific changes in hippocampal pathway density. This approach holds promise for charting normative ageing trajectories and identifying early biomarkers of vulnerability and compensation in memory-related networks.