This is really neat. It's a shame that they didn't take a look at this through an anthropological lens (even in their conclusion), because "Dunbar's number" [0]

This is really neat. It's a shame that they didn't take a look at this through an anthropological lens (even in their conclusion), because "Dunbar's number" [0] would have given some perspective on the limits of small groups, where a community of 30-50 seems to be a well-established grouping size (and ~100-250 seems to be the limit for a "small" group). For that matter, this seems relevant to the anthro theory that communities fall into one of four types of organization (families/bands, tribes/clans, chiefdoms, and states) [1]. In this context, it looks like the authors have identified the emergence of these groupings of organization, especially through their metric of "efficiency" and how bands and tribes differ. 0: http://en.wikipedia.org/wiki/Dunbar's_number http://en.wikipedia.org/wiki/Dunbar's_number 1: http://en.wikipedia.org/wiki/Sociopolitical_typology http://en.wikipedia.org/wiki/Sociopolitical_typology