Cosmopolitan ancestors: Genetic evidence that prehistoric populations were not small and isolated

Résumé

Did our Late Pleistocene hunter-gatherer ancestors live in isolated family-based bands, their social interactions limited to just a few close relatives or lifelong associates? This widely held view at first glance appears to be consistent with estimates of genetic differentiation (e.g., Wright’s FST) among recent hunter-gatherer groups. These estimates of FST can be used to reverse engineer the likely demographic structure of human groups during the Late Pleistocene because smaller historical group size increases genetic differentiation, while migration among groups has the opposite effect. If we use Wright’s original model, the observed FST values do suggest small and isolated populations. However, ancestral human populations likely deviated from the assumptions of this model in important ways. We use empirical data on hunter-gatherers to calibrate these deviations, including reproductive skew, population crashes, single-propagule recolonization processes, non-random migration, and overlapping generations. Our simulations indicate that the observed FST values are more consistent with the hypothesis that early humans lived in large, cosmopolitan groups.