Fork-tailed drongos, glossy black African songbirds with ruby-colored eyes, are the avian kingdom’s masters of deception. They mimic the alarm calls of other species to scare animals away and then swipe their dupes’ dinner. But like the boy who cried wolf, drongos can raise the alarm once too often. Now, scientists have discovered that when one false alarm no longer works, the birds switch to another species’ warning cry, a tactic that usually does the trick.
“The findings are astounding,” says John Marzluff, a wildlife biologist at the University of Washington, Seattle, who was not involved in the work. “Drongos are exceedingly deceptive; their vocabularies are immense; and they match their deception to both the target animal and [its] past response. This level of sophistication is incredible.”
Since 2008, Tom Flower, an evolutionary biologist at the University of Cape Town, has followed drongos in the Kuruman River Reserve in the Kalahari Desert. He’s habituated and banded about 200 of the robin-sized birds, and, using food rewards, has trained individuals to come to him when he calls. After getting its snack, the drongo quickly returns to its natural behavior—catching insects and following other bird species or meerkats—while Flower tags along.
Drongos also keep an eye out for raptors and other predators. When they spot one, they utter metallic alarm cries. Meerkats and pied babblers, a highly social bird, pay attention to the drongos and dash for cover when the drongos raise an alarm—just as they do when one of their own calls out a warning. Studies have shown that having drongos around benefits animals of other species, which don’t have to be as vigilant and can spend more time foraging. But there’s a trade-off: The drongos’ cries aren’t always honest. When a meerkat has caught a fat grub or gecko, a drongo is apt to change from trustworthy sentinel to wily deceiver. In a previous study, Flower showed that the birds get as much as 23% of their daily food by emitting a false alarm and then stealing their victim’s meal. The meerkat that’s just caught a gecko, for instance, is likely to drop it and run for the nearest burrow when it hears a drongo’s alarm—whether true or false.
Drongos also imitate the alarm calls of numerous other species. Altogether, the birds can make as many as 51 different warning cries. Six are those that drongos themselves use to announce the presence of various types of predators; the other 45 are the alarms of other species. And all the species—including the meerkats and babblers—know and pay attention to one another’s warning calls, Flower says. “They’re all eavesdropping on each other. It’s like they speak each other’s language.”
But what benefit do drongos get by imitating the alarm calls of other animals? To answer that question, Flower and his colleagues carried out a series of playback experiments using pied babblers as the target species. The results showed that after hearing a drongo imitate a babbler’s or starling’s alarm cry, the babblers stayed away longer from a foraging area than when they heard the drongo’s warning call. The babblers also ignored alarm calls after they’d heard one type three times in a row. But when the third call was a warning that they’d not previously heard, they took flight. The experiments show that “it pays for drongos to have large alarm repertoires, to use their target’s alarm call, and to vary their calls,” Flower says.
Which is exactly what the birds do. By following 42 marked wild drongos, he and his colleagues observed 151 occasions when the birds repeatedly attempted to steal food from the same victim. In 74 of these attempts, the birds changed the type of alarm they were using. They were most likely to do so, the scientists found, when one type of false call didn’t do the trick. Moreover, when a drongo changed its call, it was more likely to succeed in stealing its target’s food, the team reports online today in Science.
“The drongos are producing their calls tactically. They’re changing their calls in response to the feedback they get from their target,” Flower says. “And that’s how they’re able to overcome the problem of crying wolf too often.”
“The birds’ tactical deception must reflect sophisticated cognitive abilities,” says Karl Berg, an ornithologist at the University of Texas, Brownsville, who was not involved in the work. Indeed, Flower and his colleagues note that some might be inclined to see the birds’ talent as suggestive of something like “theory of mind”—the ability to intuit what others are thinking, a skill that has definitively been found only in humans.
But Dorothy Cheney, a primatologist at the University of Pennsylvania, says that “simpler explanations,” such as associative learning, “are more likely” to explain “what the drongos are thinking when they produce their deceptive calls.” For example, she notes, it’s possible that drongos have learned two behavioral contingencies: “One, targets flee when they hear alarm calls,” and “Two ... switch alarm call type if the previous call failed” to send the victim fleeing. At the very least, the birds seem to have an understanding of cause and effect, notes Flower, who is now working on a new study to try to nail down what goes on in a drongo’s head.