Fig. 5: Social and group dynamic processes in connection with the social network.
a, b Distance and movement similarity of harems, averaged over five 5-min observation sessions, for different kinship relationships between their stallions: for full sibling (n = 9 harem pairs), maternal (n = 4) and paternal half-sibling (n = 40), and more distantly related stallions (n = 411). c, d Harems between which female transfer occurred in the 2 years prior (c) or following (d) the movement observations were closer (and had more similar movement) than harems without female transfer (nprior,transfer = 27,nprior,no transfer = 438,nfollowing,transfer = 24, nfollowing,no transfer = 441 harem pairs). Boxes range from the 25th to 75th percentile, while central marks denote medians, whiskers extend to extreme data points not considered outliers (marked in red). e Closeness centrality of harems in the network against the harem’s age in years. Node size denotes the harem size, while node colour the stallion’s experience of harem keeping (i.e., how long the stallion kept a harem in years). f Average distance of the closest bachelor male to harems against the number of adult female members in the harem. Node size denotes the harem size, while node colour the closeness centrality of harems.
