It is a tough time to be human. After centuries of congratulating ourselves on being unique, the qualities we once thought only we possessed are being found, one by one, in other animals. No longer are we alone in making tools, engaging in complex social interactions or adhering to a moral code. Now, a recent study published in the Proceedings of the National Academy of Sciences has taken our fall from grace one step further: we aren't even the best at telling ourselves apart.
As dealing with threats from predators is energetically costly, it is important that only those animals that really pose a threat are met with the appropriate fight-or-flight response. However, the threat posed by humans is difficult to gauge, as not all members of our species are dangerous. A team of researchers based in the UK and Kenya, working on groups of free-ranging African elephants, therefore wondered: can animals correctly classify different groups of humans based on the level of threat they pose to them?
They decided to focus on acoustic signals, as these can be used to distinguish predator from prey from long distances. African elephants most frequently meet with two groups of human predators: Maasai pastoralists, who occasionally attack elephants, and Kamba farmers, who generally do not. The team decided to test whether elephants can distinguish between these two groups and the different languages that they speak by playing back recordings of adult men saying, ‘Look, look over there, a group of elephants is coming’, in the two languages and assessing the behavioural response of different groups of elephants.
They found that elephants are significantly more likely to display defensive behaviours (such as investigative smelling and defensive bunching) in response to recordings of Maasai men than to recordings of Kamba men. This suggests that, amazingly, elephants can pick up differences in sound that are the result of differences in the language that is being spoken. However, if elephants should indeed be capable of accurately assessing threat level, they ought to be able to distinguish men from women, as Maasai women never attack elephants. The authors therefore repeated their experiment with sound clips of adult Maasai men and adult Maasai women.
The elephants again were more likely to be defensive in response to the clips of Maasai men. Could it be that this difference in behaviour is simply a response to differences in tone and pitch of the male voices? To test this possibility, the team resynthesised the audio clips to mimic sex differences, changing them in such a way as to make the male clips – to the human ear – indistinguishable from female clips, and vice versa. Incredibly, the elephants still responded defensively to the clips that were originally from Maasai men, but mostly ignored the female-to-male resynthesised clips. The altered clips therefore still contained cues that allowed the elephants to assess whether they were in danger or not.
This study is the first to show that elephants are capable of accurately distinguishing between different subgroups within a species of predator on the basis of acoustic cues, and meeting them with the appropriate behavioural response. Moreover, it shows that elephants are capable of detecting the gender of humans more accurately than we are. So much for our prime position in the animal kingdom, then.
