Access Type

Open Access Embargo

Date of Award

January 2019

Degree Type


Degree Name




First Advisor

Noa . Ofen


The ability to remember past events is critical for everyday life and showed robust improvement over development from childhood to adulthood. With advances in noninvasive neuroimaging methods such as functional MRI in recent years, research efforts have been focused on identifying neural correlates underpinning developmental gains in memory performance. In my dissertation work, using a widely-validated subsequent memory paradigm, I aim to characterize functional MRI correlates of memory development. Specifically, I focused my investigation on identifying age differences in the functional patterns of two brain regions critical for memory, the prefrontal cortex and the hippocampus. Focusing on the prefrontal cortex (Chapter 1), I found memory-related activation in inferior frontal gyrus and memory-related deactivation in superior frontal gyrus. Both regions demonstrated developmental effects, but only memory-related deactivation in superior prefrontal cortex mediated the relationship between age and memory performance. The prefrontal cortex showed dynamic developmental effects in its functional connectivity with the medial temporal lobe, including parahippocampal gyrus. Focusing on the hippocampus (Chapter 2), I found that both anterior and posterior hippocampus supported memory formation, with effects that are relatively stable from ages 8 to 25 years. Differential developmental patterns were found for the functional connectivity between hippocampal subregions and prefrontal/visual cortices, suggesting increased functional specialization along the long axis of the hippocampus. Lastly, I tackled critical yet often neglected concerns over the reliability in identifying neural correlates of memory with fMRI (Chapter 3). I estimated the reliability of subsequent memory effects using an independent reliability dataset (n=24, ages 8 to 20 years), with a similar focus on the prefrontal cortex and the hippocampus. Good to excellent test-retest reliability was observed on the group-level contrast, corresponding to group-level analyses with a cross-sectional design. On the individual level, good reliability was observed in cortical regions including the prefrontal cortex and parahippocampal gyrus, but not in the hippocampus. Collectively, through critical evaluation and rigorous analyses, I have made important contributions to the field by providing novel insights into how traditionally-defined “memory-regions” dynamically support memory development on a granular, subregion level. In addition, my work has contributed in establishing the much-needed boundaries to the extent to which fMRI measures can be applied to answer important questions in memory development.