The hippocampus (Hc) is a heterogeneous structure composed of cytoarchitectonically distinct regions (subfields): dentate gyrus (DG), Cornu Ammonis (CA) sectors 1–3, and subiculum. Hc subfields are hypothesized to have specialized functions in supporting episodic memory (Bird and Burgess, 2008, Burgess et al., 2002, Ekstrom et al., 2003, Lepage et al., 1998, Maguire et al., 1998, Spaniol et al., 2009, Squire, 1992, Tulving and Markowitsch, 1998) and are implicated in various neurological and psychiatric disorders that vary in the age of onset (i.e., Alzheimer's disease, temporal lobe epilepsy, schizophrenia; for review see de Flores et al., 2015, Geuze et al., 2005). To understand the role of Hc subfields in human cognition and health, it is essential to characterize the age trajectories of volumetric changes throughout the lifespan.

Over the last two decades, numerous studies have investigated Hc subfield volumes in various age groups. These reports primarily rely on cross-sectional study design and investigate age-related differences in regional volumes. A meta-analysis of cross-sectional studies revealed a positive association between age and DG-CA3/4 volumes in childhood and young adulthood (ages 4–33 years) but global negative age differences in all subfield volumes over the adult lifespan (ages 18–94 years; Homayouni et al., 2023a). Correlation between age and Hc subfield volumes shifted from positive to negative depending on the mean age of the study sample spanning from childhood to late adulthood. This pattern of effect sizes is consistent with the notion of a nonlinear age trajectory across the lifespan.

Cross-sectional studies cannot reveal the magnitude of change and individual differences (Molenaar, 2004, Nesselroade, 1991), and longitudinal studies are needed for obtaining valid and reliable estimates on developmental changes. To date, only a handful of longitudinal reports on Hc subfield volume changes in specific age groups are available. Those studies yielded discrepant findings regarding the magnitude and direction of changes across lifespan. In middle childhood (ages 4–8 years), volume increases uniformly across Hc subfields measured after a one-year interval (Canada et al., 2021), yet in children aged 6–10 years and over a two-years interval, subiculum shows differentially greater expansion compared to other regions (Keresztes et al., 2021). Over the span of middle childhood to young adulthood (ages 8–21 years), one longitudinal study with a two-year measurement interval reports a general reduction in Hc subfield volumes across adolescence (Tamnes et al., 2014), whereas another highlights an initial increase in CA1 and subiculum volumes and a linear decrease in DG and CA2–4 volumes (ages 8–26 years, 2 years interval; Tamnes et al., 2018). In older adults, the longitudinal evidence points to shrinkage in all Hc subfield volumes over 4 years in healthy participants of ages 65 years and older (Nadal et al., 2020). With the limited and discrepant longitudinal evidence across childhood and adulthood, the trajectories of Hc subfield volume changes across the lifespan remain unknown.

To clarify the reported discrepancies, we assessed the volume changes of Hc subfields across almost the entire lifespan in a large sample that was measured twice over an approximately two-year interval. To examine potential variations in volume changes across distinct developmental periods, we split the sample into five age groups: childhood (ages 5–10.9 years), adolescence (ages 11–18.9 years), young (ages 19–35.9 years), middle-aged (ages 36–55.9 years), and older adulthood (ages 56–73.9 years). The grouping approach enabled us to estimate changes in Hc subfield volumes at different life stages and discern age-dependency in trajectories of change. We hypothesized that DG-CA3 volume would increase from childhood through young adulthood (ages 5–35 years), while all Hc subfields would exhibit volume shrinkage after middle age and into late adulthood (ages 56–73 years).