Cultural variation in voting patterns reflects the landscape of US genetic diversity: A test of the cultural niche hypothesis

Abstract

Humans are unique in the animal world in the extent to which we have the potential to affect our own biological evolution through culturally mediated behavioural variation. However, there is scant clear evidence of the phenomenon. Here, I estimate genetic diversity for every county in the US by merging demographic data from the 2010 US Census with genomic reference populations and show that genetic diversity is predicted by cultural variation as reflected by the 2016 Presidential Election. Remarkably, the 2016 Election alone explains 18% of US county-level variation in genetic diversity, with diversity decreasing as Trump support increases. When state level variation is included in the model, 71% of variation is explained. Within states, genetic segregation increases with cultural segregation. I then tested whether the observed genetic patterns might be explained by culture acting on race and ethnicity and found that genetic diversity tends to increase with racial and ethnic diversity, and that the 2016 Election is an even better predictor of the US landscape of ethnic diversity. Finally, I measured patterns of ethnic assortative mating nationwide from the Census data, and found that ethnic assortment is primarily explained by the proportion of minorities in each county, and that the most diverse counties are the most assortative. Overall, regional variation in cultural tolerance appears to be structuring biological diversity on a massive scale. Also, assortative mating is maintaining ethnic and genetic diversity within the most diverse and tolerant areas. Culture is a primary driver of biological evolution across the US.