Study
African elephants address one another with individually specific name-like calls
Pardo, Michael A.; Fristrup, Kurt; Lolchuragi, David S.; Poole, Joyce H.; Wittemyer, George
Nature Ecology & Evolution, 2024
Hundreds of rumbles from wild elephants in Samburu and Amboseli were analysed with a machine-learning model trained to predict the intended receiver from acoustic structure alone. Playback tests then presented individuals with calls originally addressed to them and calls addressed to others.
- Studied in
- Loxodonta africana
- Sample size
- 469 recorded calls from wild elephants in Kenya, plus 17 playback subjects
The model identified intended receivers better than chance, and elephants responded more strongly and more quickly to calls originally addressed to them.
What it means
The authors' reading, and ours. Where they differ, that difference is the point.
How the authors put it
Elephants use individually specific, learned vocal labels for one another — functionally comparable to names, and not simply imitations of the receiver's own call.
How NatureHQ reads it
The reason this attracted so much attention is the contrast with dolphins and parrots, which address each other by copying the other's signature call. Here the label appears not to be an imitation, which would make it arbitrary — closer to how a human name works. It is recent, from two populations, and the acoustic analysis carries real uncertainty; it is a strong new finding rather than a settled one, and it is exactly the kind of result that should be revisited as replications appear.
- Model accuracy was above chance but far from perfect.
- Small playback sample.
- Two Kenyan populations; the behaviour has not been tested elsewhere.
- Recent, with independent replication still to come.
What this study is used for on NatureHQ
One study can inform several subjects. Here is everywhere this one is cited.
Published by: Nature PortfolioPublished in a Nature Portfolio journal.
doi.org/10.1038/s41559-024-02420-w
This reference was resolved automatically against Crossref and OpenAlex on 2026-08-11.
NatureHQ summarises research in its own words and does not reproduce published text. Reviewed 2026-08-09.