Endocrine sources of within- and between-individual variability in large-scale brain networks

Dense Investigation of Variability in Affect (DIVA)

• OpenNeuro (WIP)
• DIVA on GitHub

The rise of large neuroimaging datasets and multi-dataset mega-analyses brings the power to study interindividual differences in brain structure and function on a heretofore unseen scale. However, unknown and poorly characterized intra-individual variability continues to undermine the detection of robust brain-behavior associations and, ultimately, our understanding of the brain on the whole. Women’s and reproductive health underlie variability in more than half of the population, but have long been overlooked in the study of both inter- and intra-individual differences in the brain. To this end, the Dense Investigation of Variability in Affect (DIVA) Study was designed to study intra-individual variability in the brain and behavior across the menstrual cycle in a small cohort of premenopausal female participants. The DIVA Study acquired weekly actigraphy, self-report, biospecimen, and both functional and structural magnetic resonance imaging data with concurrent peripheral physiological recordings. These data facilitate the study of several common sources of variability in the brain and behavior: the menstrual cycle and ovarian hormones, sleep, stress, exercise, and exogenous sources of hemodynamic variability.

Manuscripts

Presentations

A dense investigation of functional connectome variability, hormonal contraceptives, and the menstrual cycle

The brain’s critical role in the endocrine system is one of bidirectional influence and communication. While ovarian hormones, estradiol (E2) and progesterone (P4), fluctuate throughout the menstrual cycle, with important roles in uterine function, they and their metabolites cross the blood-brain barrier and have central nervous system (CNS) targets, as well. Perhaps due to this CNS action, changes in both brain structure and function across the menstrual cycle have been associated with fluctuating E2 and P4 levels, which is further complicated by hormonal contraceptive (HC) use. Neuroimaging studies of such associations are marked by experimental design shortcomings, with too few within-subject time points to accurately characterize E2 and P4 levels across the menstrual cycle, which varies as much between individuals as it does within. Here, we use predictive connectomics to assess E2, P4, and HC associations with functional connectivity (FC) across the brain, laying groundwork for future neuroendocrine research.

Manuscripts