Spatial navigation is a complex cognitive skill and multisensory process with large individual and sex differences, and deficits in navigational abilities are apparent in older adults (ages 65+ years). When in the aging process these changes first emerge is unclear. Further, little is known about whether endocrine aging during the midlife transition to menopause (ages ~45-55 years) impacts navigation behavior and the neural circuitry that processes spatial information. Given accumulating evidence of the overlapping neural circuitries that are sensitive to cognitive aging, enriched with sex hormone receptors, and support spatial navigation abilities, this represents a critical gap in our knowledge for how the brain’s navigational system changes with age.

In this dissertation, the three main lines of inquiry that are addressed are: (1) whether spatial navigation abilities and strategies change early in the aging process; (2) whether endocrine changes that occur during the menopausal transition are associated with spatial navigation abilities and strategies; and (3) whether endocrine changes during endocrine aging alter the brain’s navigation circuitry. To execute this, three virtual reality paradigms testing distinct aspects of spatial navigation were used in conjunction with structural magnetic resonance imaging. The results provide evidence that: (1) age-related changes in some aspects of navigation emerge as early as midlife; (2) spatial strategy is influenced by endocrine aging in women; and (3) age-related changes during endocrine aging impact the neural circuitry that supports spatial navigation performance.

By studying the brain’s navigational system from a women’s health perspective, this dissertation work provides novel insight into the biological factors that influence spatial navigation. These findings will inform future studies investigating the utility of spatial navigation as a behavioral marker for the early detection of individuals at risk for neurodegenerative disease. Women make up about two-thirds of patients living with Alzheimer’s disease, suggesting that sex-specific factors are crucial to understanding disease risk.

Since its inception, the field of neuroendocrinology has provided evidence for a tightly coupled relationship between the nervous and endocrine systems. In rodents and nonhuman primates, estrogen and progesterone’s impact on the brain is evident across a range of spatiotemporal scales. Yet, the influence of sex hormones on the structural and functional architecture of the human brain is largely unknown. The body of work presented in my dissertation aims to advance our understanding of the human brain through the lens of neuroendocrinology by examining how distinct hormonal transition periods shape brain morphology and function. In Study 1, I present findings from my keystone ‘28andMe’ precision imaging experiment, in which a participant underwent brain imaging and venipuncture every 24 hours over 30 consecutive days across a complete menstrual cycle and again, one year later, while on an oral hormonal contraceptive regimen. The results from this study reveal the rhythmic nature in which brain networks reorganize across the cycle, with transient increases in estradiol enhancing global efficiency of several large-scale networks. In Study 2, this approach was expanded by conducting the first precision imaging experiment on pregnancy, in which a primiparous woman underwent 26 MRI scans and venipuncture beginning 3 weeks pre-conception through two years postpartum. Pronounced decreases in gray matter volume and cortical thickness paired with increases in white matter microstructure were evident across the brain, with few regions untouched by the transition to motherhood. Finally, in Study 3, I present findings from our Midlife Hormones and Cognition Study, where a highly characterized sample of 85 midlife women (ages 43–60) in various states of ovarian decline underwent MRI scanning and venipuncture to establish how endocrine aging influences whole-brain intrinsic network organization. Results suggest that the frequency of menopause-related vasomotor and neurological symptoms exacerbate network connectivity decline in postmenopausal women, especially among higher-order cognitive networks. Together, these studies provide novel insight into the spatiotemporal extent of human brain–hormone relationships over the lifespan, a severely understudied area in cognitive neuroscience with significant implications for women’s health.

The prefrontal cortex (PFC) is exquisitely sensitive to its neurochemical environment. Minor fluctuations in cortical dopamine (DA) can profoundly alter working memory (WM), a PFC-dependent cognitive function that supports an array of essential human behaviors, from problem-solving to fluid intelligence. Dopamine's action in the PFC follows an inverted U-shaped curve, where an optimal DA level is necessary for maximal function and both insufficient and excessive DA activity impairs PFC processes. In animals, estrogen has been shown to increase dopaminergic activity, yet this relationship has not been demonstrated in humans. This suggests that working memory performance might be affected by estrogen's rhythmic changes throughout the menstrual cycle, and that baseline DA levels will influence the direction of estrogen's effect.

In a series of cognitive genomic, neuroendocrine studies in healthy young women, we examined estrogen's impact on the performance of DA-dependent tasks as a function of COMT Val158Met genotype and COMT enzyme activity (indices of baseline DA). The results demonstrate that estrogen status impacts working memory function and, crucially, that the direction of the effect depends on an individual's COMT genotype and, at a finer scale, COMT enzyme activity, demonstrating a dependence on baseline DA. At a neural level, functional MRI revealed that cortical dopamine (shaped by a balance of genetic and hormonal factors) is associated with a broadly `efficient' pattern of sustained activity (that which occurs across WM blocks), and a selective, event-related enhancement of activity during episodes of high interference (e.g. lures), when the demand for cognitive control is greatest. Furthermore, the extent to which an individual enhances PFC activation during the demanding lure trials is predictive of their performance.

Next, we used a visual selective attention paradigm to probe the effects of estrogen and COMT genotype on top-down, goal-directed modulation of neural activity in visual association cortices (VAC). We used a recently established metric of goal-directed `enhancement' and `suppression' that is sensitive to identifying group differences in VAC modulation. Scene-selective regions of interest (bilateral PPA) showed robust suppression and enhancement effects at the group level, which were dependent on task goals, but further analyses revealed an important difference between low and high estrogen groups. While both groups successfully enhanced PPA activity during the Remember Scenes condition above a perceptual baseline, only the high estrogen subjects were able to appropriately attenuate the processing of task-irrelevant scenes in the Ignore Scenes condition. This effect of estrogen on distracter filtering parallels the suppression deficit observed in older adults, and young adults when attentional resources are taxed.

Furthermore, when attentional resources were imposed upon (during a dual-task condition in which two stimuli from different object categories must be attended to and maintained over a delay) low estrogen subjects succumbed to an `enhancement deficit', which has been shown to occur in young adults when attentional resources are limited. High estrogen subjects, however, were resilient to the high load condition. Thus, even when attentional/working memory resources were taxed, if estrogen levels were high women showed no evidence of strained top-down, goal-directed processing. When estrogen levels dropped (during the beginning of the cycle) the enhancement deficit emerged.

Multivariate functional connectivity data assessing coherence between frontal control regions and visual association cortices revealed an estrogen*genotype interaction. Subjects with naturally reduced prefrontal DA (val/val genotype) showed greater top-down coherence when estrogen levels were high versus low; but subjects with naturally elevated prefrontal DA (met/met genotype) showed the opposite pattern, with the most robust coherence when estrogen levels were low. These data parallel the interaction observed in the N-back task, which both follow the theoretical inverted-U shaped DA model.

In humans there has been a strong effort to understand the effects of estrogen on cognition, but the data have been inconsistent. This study establishes that taking baseline DA into account is pivotal to detecting the direction of estrogen's effect on working memory. The results carry direct ramifications for women's health, as the response to DA medications (e.g. Ritalin for attention-deficit disorder and l-DOPA for Parkinson's disease) may differ between men and women, and within women in different endocrine states. A man and woman's milieu differ; until we understand how we cannot fully understand neural processes as they unfold in the healthy state, less still in the diseased state.