Strong ties formation, composition and processes at play during the developmental period of juvenile vervet monkeys

Abstract

Understanding the development of social relationships can provide insights into the processes by which social network structures emerge and vary across species. Here we extend a previous analysis (Vilette et al., 2022, Animal Behaviour, 194, 205–223) that tested Kohn's (2019, Animal Behaviour, 154, 1–6) model of social relationship formation in three groups of wild juvenile vervet monkeys, Chlorocebus pygerythrus. This previous analysis showed that, although developmental patterns did not conform to the exploration, pruning and consolidation phases identified by Kohn (2019), juveniles formed a core of strong social ties (a social ‘bubble’) across development. Here, we use a novel approach to objectively extract strong ties and ask whether Kohn's (2019) phases are more appropriately evaluated with reference to the subset of strong ties that constitute social bubbles. We investigate social bubble formation and composition, as well as the role of maternal behaviour in the development of these bubbles. As before, spatial and grooming social bubbles did not develop according to Kohn's (2019) framework. On the one hand, spatial bubbles were composed mainly of juveniles and showed increased association rates for family members during the annual birth season. Juvenile grooming bubbles, on the other hand, were stable over time and restricted solely to their mothers. Resulting from this, we found that juveniles did not simply groom their closest spatial associates but distributed their grooming towards specific partners. Finally, we found that a mother's grooming ties and her offspring's grooming weak ties remained mostly different as juveniles developed. This finding supports a previous analysis on our population (Jarrett et al., 2018, Proceedings of the Royal Society B: Biological Sciences, 285(1876), Article 20172668), where juveniles had to develop connections with nonmaternal associates in order to replicate the overall grooming network. These results indicate that the structure and composition of social bubbles in our sample reflect both the behaviour under consideration (grooming or spatial proximity) and group demography.