Hyenas know when and who to ‘whoop’ at thanks to their built-in caller ID system

Scientists from the University of Nebraska, Lincoln, U.S. have discovered that Hyenas' whoops have specific signals unique to each individual animal.

The researchers determined that hyena whoops have specific characteristics that can be attributed to each individual animal by using machine learning on audio files collected from a field trip, according to a press release published by EurekAlert on Saturday.

"Hyenas don't treat every individual in the clan the same, so if they're deciding whether to show up and help someone, they want to know who they're showing up to help," said Kenna Lehmann, a postdoctoral researcher at the University.

This discovery was made possible by a research trip to the Maasai Mara plains in southwest Kenya. The researchers searched for vocal traits using a method called a "random forest model."

The identities of each recorded hyena were fed to the model, along with an enormous amount of acoustic characteristics gleaned from each whoop during the machine's initial training.

The algorithm then produced decision trees (diagrams/readings) using a randomly selected set of whoops from one hyena. A binary selection from a group of acoustic features was represented by each branch of a tree chosen randomly.

The model predicted a particular whoop's identification depending on which hyena had the most votes from the 500 randomly assembled decision trees.

The model associated a whoop to its hyena 54% of the time

The scientists tested their trained model by asking it to determine which of 13 hyenas made a randomly chosen burst of whoops. They then repeated the test 999 times.

The software correctly associated a whoop bout with its hyena around 54 percent of the time, or almost six times more frequently than would be expected by chance.

This success rate demonstrates that there is enough variation in the whoops of many hyenas and enough consistency within the whoops of a single hyena for the model to tell them apart reasonably.

According to Lehmann, it is reasonable to suppose that the hyenas can distinguish between these differences if the model can.

"There's an understanding that one of the ways to get your message across is to repeat it," said Lehmann.

"Especially if you're in a noisy environment or if you're communicating over long distances."

The random forest model

The random forest model accurately identified a hyena based on just one whoop. It did so only about half as frequently as when it was given three whoops. With more calls, the accuracy increased, even more, reaching a maximum of seven whoops.

"It's like getting a little bit more information (each time)," said Lehmann, who previously studied vocalizations in orcas.

"The first time you hear it, you might notice: Oh, that was definitely a male or a female voice. Then, the next whoop, you might be able to narrow it down further," she added.

Earlier studies have revealed, for instance, that as the wind picks up, penguins repeat their calls more frequently. And other research has discovered evidence that other animal species favor repetition in comparably noisy environments.

'Hyenas and their little interactions and the dramas'

Lehmann and her coworkers were able to determine that "no one had quantified" the degree to which repetition of an animal call might genuinely enhance information transfer.

The researchers spent months waiting and hoping that their observations and recordings would guide future research. They noticed glimmers of the distinctiveness that their analysis of the hyenas' whoops would eventually come to confirm.

"You definitely get to know that different individuals have different personalities or might react a certain way in different situations," said Lehmann.

"So it's always fun to just get to know the hyenas and their little interactions and the dramas that might be going on in their lives."

The team first published its findings in the Proceedings of the Royal Society B journal in July.

Study abstract:

In animal societies, identity signals are common, mediate interactions within groups, and allow individuals to discriminate group-mates from out-group competitors. However, individual recognition becomes increasingly challenging as group size increases and as signals must be transmitted over greater distances. Group vocal signatures may evolve when successful in-group/out-group distinctions are at the crux of fitness-relevant decisions, but group signatures alone are insufficient when differentiated within-group relationships are important for decision-making. Spotted hyenas are social carnivores that live in stable clans of less than 125 individuals composed of multiple unrelated matrilines. Clan members cooperate to defend resources and communal territories from neighbouring clans and other mega carnivores; this collective defence is mediated by long-range (up to 5 km range) recruitment vocalizations, called whoops. Here, we use machine learning to determine that spotted hyena whoops contain individual but not group signatures, and that fundamental frequency features which propagate well are critical for individual discrimination. For effective clan-level cooperation, hyenas face the cognitive challenge of remembering and recognizing individual voices at long range. We show that serial redundancy in whoop bouts increases individual classification accuracy and thus extended call bouts used by hyenas probably evolved to overcome the challenges of communicating individual identity at long distance.