Does Education Slow Cognitive Aging?

Post by Natalia Ladyka-Wojcik

The takeaway

Previous studies examining the link between education and cognitive decline in aging have yielded mixed results, often relying on small or single-country samples. In this large, multi-national cohort study, researchers found that higher education was associated with better memory performance and greater brain volume, but it did not protect against age-related neurodegeneration.

What's the science?

The relationship between higher education and cognitive function in aging remains a subject of debate. Although a substantial body of evidence has identified education as a major protective factor against age-related dementia in later life, the underlying mechanisms of this are unclear. Two prominent theories – the brain maintenance and cognitive reserve accounts – suggest that education can slow or postpone age-related cognitive decline. However, emerging longitudinal data challenge this view, showing that more educated individuals do not necessarily experience reduced cognitive decline over time. Instead, an alternative hypothesis posits that higher education provides an early-life cognitive advantage that persists into old age, without altering the trajectory of decline. This week in Nature Medicine, Fjell and colleagues examined a large, multi-national longitudinal dataset of memory performance and brain imaging to test whether education offers protection against cognitive aging.

How did they do it?

Addressing the relationship between education and cognitive aging requires large, diverse, and longitudinal datasets with sufficient statistical power. To test competing theories, the authors analyzed longitudinal memory scores in 170,795 participants over the age of 50, along with over 15,000 brain MRI scans from 6,472 participants across 33 Western countries. These data were drawn from large, population-based sources, including the Survey of Health, Ageing and Retirement in Europe (SHARE), which provided repeated measures of verbal episodic memory – a form of memory for specific events in time and space that is particularly sensitive to aging. The researchers also examined neuroimaging markers of cognitive decline, including intracranial brain volume and volume of memory-related brain regions such as the hippocampus and thalamus. To broaden the generalizability of their findings beyond WEIRD (Western, Educated, Industrialized, Rich, and Democratic) populations, they replicated their memory findings in an independent cohort from China, India, Mexico, and South Africa.

What did they find?

The researchers found that memory scores declined with age, consistent with expected age-related declines in episodic memory. Specifically, across the datasets analyzed, they observed a general pattern of higher memory scores among individuals with more education at all ages. Importantly, however, they found no evidence that higher education reduced memory decline or influenced repeated measures over time. To assess whether these results were specific to verbal memory, the authors extended their analysis to include tests of mathematical ability and temporospatial orientation within the SHARE cohort. For brain-based markers of aging, higher education was associated with greater intracranial volume and slightly larger volumes in memory-sensitive regions. However, the rate of decline in these brain regions was similar regardless of education level.

What's the impact?

This study is the first to show, using large-scale longitudinal data, that the commonly held view of education as a protective factor against cognitive aging lacks strong support. Instead, individuals with more years of formal education tend to begin adulthood with higher cognitive functioning, but they do not experience slower cognitive decline as they age.

Access the original scientific publication here.

Can The Menstrual Cycle Impact Memory Processing?

Post by Rebecca Glisson

The takeaway

Hormones such as estradiol, the primary female reproductive hormone, can alter the number of inputs a brain cell receives from within the hippocampus, the memory center of the brain. When estradiol levels are highest during an estrous cycle (the mouse menstrual cycle), these brain cells have the most connections to other cells within the hippocampus, suggesting that higher estradiol is related to better learning and memory function.

What's the science?

Cells in the brain communicate with each other when a synapse (the end of one neuron) sends signals to a dendritic spine (the beginning of another neuron). The more dendritic spines there are on one neuron, the more inputs it receives and the more communication it has in turn with other neurons. This week in Neuron, Wolcott and colleagues explored whether the fluctuations of estradiol during the estrous cycle in female mice affect the density of dendritic spines within the hippocampus.

How did they do it?

To study the spine density of neurons in the hippocampus, the authors used a specialized microscope in combination with two-photon imaging, or used fluorescence to look specifically at the dendritic spines of cells. They first measured the concentration of estradiol in females in order to track each stage of the estrous cycle, which lasts 4-5 days in mice and has four different stages, similar to the human menstrual cycle. Estradiol is most concentrated just before ovulation in both humans and mice, which occurs during the proestrus stage in mice. In order to measure changes in the dendrites of neurons, the authors implanted a permanent glass microperiscope in the brain, since the hippocampus is found deep within the brain and is typically difficult to look at in live mice. They used mice that were genetically bred to have fluorescent cells in this area of the brain so that they could track changes in the dendrites of these cells. When new dendritic spines formed, the authors also tracked whether these spines were lost quickly afterwards or remained as a new, lasting part of the cell.

What did they find?

Female mice had the highest density of dendritic spines when they were in the proestrus cycle, when estradiol concentrations were highest. This suggests that estradiol affects the structure of neurons and that higher concentrations of estradiol lead to more connections between cells in the hippocampus. The authors also found that, while many of the new spines that were added in the proestrus stage were shortly lost afterwards, a portion of these new spines remained present on the cells as a permanent addition. This demonstrated that estradiol can permanently change the structure of and connectivity between cells in this part of the brain.

What's the impact?

This study is the first to show that the estrous cycle in mice is related to the structure and function of hippocampal neurons. It suggests that learning and memory processing within the brain are influenced by the changes in hormone levels during the estrous cycle. However, females are not necessarily more variable than males due to their estrous cycles. For example, hormones also change throughout the day based on the sleep/wake circadian cycle, which occurs in both males and females. Studies like this can help us better understand how our learning and memory functions can change based on our hormones.

Access the original scientific publication here.

A Dietary Pattern That Could Prevent Dementia

Post by Amanda Engstrom

The takeaway

Despite the association between a healthy diet and reduced risk of dementia, precise dietary patterns for dementia prevention have not been well defined. Using machine learning, the authors identify a dietary pattern that is associated with reduced dementia risk. 

What's the science?

Dementia, a class of disease characterized by cognitive decline, currently has no effective treatment, making prevention a major focus of research. Dietary factors have been suggested to impact disease, with healthy eating being positively associated with reduced risk of dementia. However, previous studies have lacked the timescale and scope to properly establish dietary patterns that prevent dementia. This week in Nature Human Behavior, Chen and colleagues utilize machine learning to identify patterns linked with reduced dementia risk and examine how this diet can support overall brain health.

How did they do it?

The authors analyzed UK Biobank data from 185,012 participants with 24-hour diet recall and after 10 years, 1,987 developed some sort of dementia (referred to as all-cause dementia, ACD). The authors conducted a food-wide association study and determined which foods were statistically associated with ACD incidence. Utilizing a machine learning approach (LightGBM), the food groups associated with dementia were ranked by importance in predicting dementia risk, and used to develop the MODERN (Machine learning-assisted Optimizing Dietary intERvention against demeNtia risk) diet score. The MODERN diet was compared against other current diets, validated in multiple study cohorts, and evaluated for its associations with other health-related outcomes. Finally, the authors investigated the underlying biological mechanism using multimodal neuroimaging, metabolomics, inflammation biomarkers, and proteomics. 

What did they find?

Of the 34 food groups analyzed, 25 were individually associated with ACD. Interestingly, many of them were not linearly associated, highlighting the importance of the amount consumed. The authors applied their machine learning approach to identify the optimal combination for dementia prevention. Termed the MODERN diet, it’s made up of eight key foods most strongly linked to lower risk (such as green leafy vegetables, citrus fruits, eggs, and poultry) and one linked to increased risk (sweetened beverages). The MODERN diet was validated in multiple independent cohorts, and each time performed better at predicting dementia outcome compared to previously established diets. Additionally, the MODERN diet was significantly associated with predicting other mental and behavioral disorders. Using brain magnetic resonance imaging, from a subset of participants, the MODERN diet score was associated with larger mean thickness of multiple brain structures, suggesting a protective role in maintaining brain health. Finally, the authors identified significant changes in metabolites based on the MODERN diet score, as well as decreased markers of inflammation and dementia related plasma proteins. 

What's the impact?

This study is the first to combine a food-wide longitudinal analysis with machine learning to develop a new dietary pattern, the MODERN diet, to predict dementia risk. The MODERN diet is associated with better brain health via metabolism-inflammation pathways. This dietary pattern can inform primary dementia prevention and be tested in future randomized controlled trials. 

Access the original scientific publication here.