Elephants imitate each other when they produce deep rumbling sounds
African savannah elephants use a variety of calls to communicate, and now machine learning has shown that they tune their deepest sounds to match those of other nearby elephants
Elephants appear to harmonise their vocal rumbles when they join together to form choruses, tuning their voices to sound more alike as the chorus unfolds.
The findings come from a machine learning study, which also reveals that this convergence is particularly likely to occur during specific kinds of choruses – such as when elephants reconnect with one another or organise group activities – and that the calls themselves may contain information about what the animals are doing.
The findings could offer new insights into how elephants communicate and navigate life within their complex societies, says Sara Keen at Earth Species Project, a US non-profit organisation that uses artificial intelligence and machine learning to attempt to decode non-human animal communication.
Advertisement
“It’s sort of like what humans do in conversation, where you almost start to adopt the speaking patterns or accents of whoever you’re interacting with,” says Keen.
Elephants communicate using a rich repertoire of body language, but also a wide range of sounds. These include their iconic trumpet call, and also throatier roars. The most common call is the rumble – so low-pitched it is often undetectable by humans – which can travel up to 10 kilometres, transmitting long-distance information that can potentially include the identity of the caller and what’s happening at the time.
Joyce Poole at Elephant Voices, a non-profit organisation based in Washington DC, that promotes the study and protection of elephants, says scientists have long been intrigued by the way elephants will rumble together in choruses. Studying those complex sounds, however, has proved extremely difficult because it is hard to pick out any given rumbling elephant in the chorus – what researchers call “the cocktail party problem”.
With significant developments in machine learning, however, Poole, Keen and their colleagues have found that computers are up to the task. They used a collection of 902 recordings of rumbles collected across four decades, emitted by wild female African savannah elephants (Loxodonta africana) from dozens of family groups in Kenya’s Amboseli National Park. The calls were grouped according to six distinct social situations: staying in contact while separated; preparing to move; coordinating what to do next; comforting or reassuring calves; greeting close family members; and participating in large family reunions.
Machine analyses revealed that elephant choruses carry multiple layers of information at once, and vary in a way that suggests they encode information about what a family is doing together and what happens next, says Keen. Tracking rumbles of individual elephants in the same chorus, the data showed that in some cases, the calls became more acoustically alike as the chorus progressed. Specifically, this convergence occurred when elephants were staying in contact while separated, coordinating what to do next, and greeting after a separation. This suggests that vocal convergence may play a role in helping elephant families build a plan of action, says Keen.
Earlier research that Poole and her colleagues published in 2005 showed that elephants can learn to imitate each other’s voices when they are calling more generally: the new work shows that this imitation involves the rumbles, too.
Combined with other communication tools, including body language, the rumbling calls seem to help facilitate complex social interactions. “They appear able to negotiate, agree and disagree – and make a plan of action,” says Poole.
The findings not only shed light on elephant communication and cognition, but they could also eventually aid conservation by helping researchers track their behaviour and movements remotely, the team says.
Because rumbles allow for long-distance communication, elephants’ ability to unite through their voices is an exciting reflection of their advanced cognition and evolution in their environment, says Andrew McLean, a member of the International Union for the Conservation of Nature’s Asian Elephant Specialist Group. “Evolving and elaborating this for social cohesion, group survival and foraging efficiency is very adaptive,” he says. “This is a fascinating finding – although it’s not really surprising, given how intensely social elephants are.”
Scientific Reports DOI: 10.1038/s41598-026-68525-3
What's Your Reaction?
