Linking social interactions with parasitism in complete and incomplete networks

抄録

Group living is beneficial for individuals in a group, but also comes with costs. One such cost is the increased possibility of pathogen transmission. Yet, most research that focuses on social transmission in primates has done so with only subsets of the group (e.g. adults or even adult females), which will influence the results and their interpretation. With this in mind, we aimed to test (i) whether social interactions or proximity mediate the spread of intestinal parasites in primate groups, and (ii) whether work that includes only subsets of a group might produce biased results. To test these hypotheses, we investigated the relationship between social network centrality and intestinal parasite infection intensity in a group of Japanese macaques (Macaca fuscata), using both empirical and simulated data. We used social network analysis on data collected over two months on Koshima to relate indices of network centrality to an index of parasite infection intensity (fecal egg counts: FEC). We then ran a series of knock-out simulations to test the effect(s) of missing data on the observed relationship. General linear mixed models showed that social network centrality was positively associated with infection by nodular worm (Oesophagostomum aculeatum) in the complete observed data set but including only subsets of the group (e.g. adult females or random subsets of the group) can yield false negative results; though a juvenile only network retains the positive association between sociality and infection. Results suggest that social interactions or shared proximity can mediate the spread of some intestinal parasites, but researchers that only focus on subsets of their study groups, or where missing data may be an issue, must interpret their results with caution. This work introduces important methodological considerations for research into the dynamics of social transmission, and not just for infectious disease.