Association patterns of female gorillas

Abstract

Social interactions ultimately impact health and fitness in gregarious mammals. However, research focusing on the strength of affiliative interactions has primarily been conducted on female philopatric species. Gorillas provide an interesting counterpoint to previous research as females emigrate multiple times throughout their lives. We compare female-female association strength, duration and consistency in wild mountain (Gorilla beringei beringei) and western gorillas (Gorilla gorilla gorilla). Additionally, we examine whether the alpha male influences female association strength and if these associations are an artefact of both females concurrently in spatial proximity of the alpha male. In this between-species comparison, female gorillas had differentiated association patterns that were consistent on average for 2 years. The alpha males did not influence female association strength, with associations being similar in his presence or absence. Finally, we found more variability in association patterns among mountain gorillas with higher average association scores and higher proportion of 'preferred associates' than western gorillas. The rare dispersal pattern in the Gorilla genus may lead to greater flexibility in female association patterns than in species exhibiting female philopatry and strong kinship bonds. This may echo ancestral human society and provide new evidence to help us understand the evolution of modern human society. This article is part of the theme issue 'Cooperation among women: evolutionary and cross-cultural perspectives'.

Keywords: association patterns; cross-species; female dispersal; social network analysis; social relationship.

Figure 1.

The relationship between female dyadic association scores (y-axis) and gorilla group (x-axis: western gorilla = ATA (N = 75; purple) and mountain gorillas = BIT (N = 30), KYA (N = 310) and ORU (N = 46); blue). The dots represent the distribution of the raw data. See table 2 for the full results of the output of model 1 (N = 461 dyads). (Online version in colour.)

Figure 2.

Showing the relationship for female association score between a dyad's current association score (y-axis) and their previous association score the year before (x-axis). Results are from the output of model 2 (N = 385 dyads); for full model results, see table 3. Darker dots are due to data points overlapping.

Figure 3.

Showing the relationship for whether the dyad partner is the individual's top partner (y-axis: predicted probability of being the top partner) and whether the dyad partner was the individual's top partner one-year earlier (x-axis: no and yes). Results are from the output of model 3 (N = 798 dyads); for full model results see table 4.