The functional relevance of vocal learning in elephants

In the last decade clear evidence has accumulated that elephants are capable of vocal learning. The term vocal learning refers specifically to vocal production learning in which signallers modify the structure of their vocalizations in response to auditory experience, i.e. by vocal imitation. Vocal learning is especially rare among mammals. Humans are versatile vocal learners, but otherwise vocal learning is not well developed in primates. Among non-primate mammals, only whales, dolphins, seals and bats have been previously shown to be capable of vocal learning. A study that colleagues and me published in 2005 provided two examples of vocal imitation in African elephants. In 2012, we published a report about a speech-imitating Asian elephant. Both papers document vocal imitation produced by captive elephants living in socially abnormal conditions. No information is available on the function of vocal learning within the natural communication system of either species. Elephants are highly social mammals where individuals possess different levels of association. Related females maintain close bonds throughout their lives forming family groups, where vocalizations have an important role for social coherence and coordination. A general idea on the function of vocal learning in social mammals is that vocalizations are an indicator of group membership. Some social group-living species (toothed whales, bats and humans) possess vocal dialects or accents (that arise through call convergence) shared by social affiliates. The overarching goal of this project is to explore the functional relevance of vocal learning in wild African elephants. I specifically suggest that vocal production learning in elephants facilitates vocal convergence leading to vocal dialects as a mechanism for creating and reinforcing social bonds among affiliated individuals of matrilineal family groups. The proposed research aims to combine classic bioacoustics work and sophisticated experimental designs. We will conduct acoustic recordings, playback experiments and will apply an acoustic camera array to visualize sound emission in order to define the calling individuals during complex bonding and chorusing behaviour. In these intense bonding situations, calls might particularly converge to emphasize social bonds. Vocal production learning is a critical skill for human communication, but its evolutionary origins are poorly understood, and evidence for production learning in non-human primates is limited at best. Elephants represent the first terrestrial mammalian species for which comparative studies of vocal production learning with humans are possible. Understanding vocal learning in elephants might therefore provide particularly important comparative data towards understanding the evolutionary origin of vocal learning in our own species.

Elephants are known for their complex vocalization system with a lot of variation, and for being able to imitate sounds, where a vocalization is modified based on auditory model. They are considered vocal learners. This is special, because besides humans, only song birds and parrots, whales and dolphins, seals, bats and elephants proved capable of vocal learning. But the term vocal learning refers to two main forms of learning that may influence vocalizations and communication systems. Despite of vocal production learning, it also includes vocal usage learning. Vocal usage learning appears when an individual learns to produce an existing signal in a novel context. It may be relevant for young animals learning how and when to use vocalizations, as well as other age-related periods requiring linking a new context with an existing vocalization. Surprisingly, nothing was known about usage learning in elephant communication. Usage learning can be experimentally demonstrated by training animals to vocalize in an arbitrary (cue-triggered) context. We showed that 13 adult African savannah elephants can vocalize in response to verbal cues, reliably producing social call types such as the low-frequency rumble, trumpets and snorts as well as atypical sounds using various mechanisms, thus displaying compound vocal control. The combined ability of vocal production and usage learning highlights the value of elephants as a study species for specifically addressing the behavioural and ecological relevance of vocal learning. We further showed that African elephants apply unusual and individualistic sound production mechanisms to generate atypical, idiosyncratic sounds. These sounds were produced by manipulating non-phonatory structures, e.g., applying an ingressive airflow at the trunk tip to emit extraordinarily high-frequency sounds or repeatedly contract superficial muscles at the trunk base to generate lower-frequency pulsated sounds. Intriguingly, each individual establishes its own distinctive sound producing strategy (e.g., contracting different muscle bundles). The production of these sounds on cue is encouraged via positive reinforcement training. This suggests that social feedback and reinforcement can facilitate vocal creativity and learning behaviour in elephants. Social interactions and positive feedback are also crucial for early speech learning in human infants. Increasing knowledge on sound production plasticity in elephants-long-living, highly social mammals-is crucial in the effort to better understand their communicative and vocal learning ability and its function in wild elephant populations. These and other discoveries from this project, which was severely disrupted by the corona pandemic, still significantly contributed to increase our understanding of the vocal learning skills of elephants and on their vocal abilities in general.