Acoustic variation and group level convergence of gelada, Theropithecus gelada, contact calls

The acoustic structure of nonhuman primate vocalizations can vary substantially within a call type and may converge between social partners. Examining which social partners share call structure can inform our understanding of the function of vocal learning and communication in nonhuman animals. We assessed vocal convergence of female gelada contact grunts within multiple levels of gelada society: small female-philopatric reproductive units that maintain close proximity and larger bands of units that preferentially associate while foraging. We also measured the extent to which grunt acoustic structure varied by caller identity and behavioural state at the time of the call, as well as genetic relatedness, when assessing acoustic similarity between females' grunts. The acoustic structure of female gelada grunts differed between individuals as well as between foraging, travelling and socializing behavioural states. Female gelada grunts were more similar between pairs living in the same band than between pairs living in different bands. This effect was not found for pairs living in the same unit compared to pairs living in different units within the same band. As genetic relatedness did not predict similarity in grunt acoustic structure, we propose that vocal convergence within bands is the result of vocal learning. Vocal convergence at the higher level of gelada society suggests this vocal learning is selective and not simply the result of auditory exposure to conspecific calls. Vocal convergence in geladas may function to maintain spatial cohesion, which is a more critical challenge for bands than for units.