Science Watch

Man's best friend has earned his title, palling around with humans for at least 12,000 years. Now, a new study finds that those eons of evolution may have provided dogs with a set of social skills that other animals, including ones that are generally considered "smarter," don't possess.

Dogs, the research finds, are much better than apes at understanding humans' cues to find hidden food. The study, published in the February Journal of Comparative Psychology (Vol. 120, No. 1), is one piece of evidence in a small but growing body of literature examining dogs' special relationship with humans.

The researchers, mostly evolutionary psychologists and anthropologists, hope that the studies will help them understand not only dogs, but also man's early evolutionary history.

"Our main goal is to compare animals with humans," says study co-author and graduate student Juliane Bräuer. "We want to learn about animals, but also what animals can tell us about humans."

Foraging for food

In the study, Bräuer, psychologist Michael Tomasello, PhD, and their colleagues at the Max Planck Institute for Evolutionary Anthropology in Leipzig, Germany, studied 21 dogs and 16 apes. The experimental setup was simple: Bräuer sat behind a hand-concealing screen in front of the animal. Then, she hid a piece of food in one of two opaque plastic cups. Finally, she removed the screen and, as the animal watched, used one of 14 cues to indicate the food's location.

Some of the cues were social or communicative, with Bräuer trying to directly relate information to the animal. For instance, she pointed at the correct container, looked at it continuously and alternated her gaze back and forth between the animal and the container.

Other cues were behavioral: In one, Bräuer opened the correct container. In another, she tried unsuccessfully to reach the correct container, which was placed just out of her grasp.

Finally, Bräuer also provided some causal cues, which required the animals to make logical deductions to find the food. For example, in one trial she rattled the food in the correct cup. In another, she shook the empty cup-allowing the animals to figure out that since the one cup was empty, the other must contain the food.

The researchers found that the apes were able to use the causal cues-even the most difficult one, shaking the empty cup-to locate the food. However, they were unable to figure out any of the communicative clues, even one as simple as pointing.

The dogs failed, though, to use the causal cues-even shaking the correct cup. But they were able to use all of the communicative and behavioral ones at a level much better than chance, accurately sussing out the food's location about 70 percent of the time.

The results weren't unexpected, says Bräuer. Previous research by the group, as well as by University of Louisiana psychologist Daniel Povinelli, PhD, who studies mostly apes, and Hungarian ethologist Adam Miklosi, PhD, who studies mostly dogs, found similar results. But the new study is the first to look at dogs and apes side by side, using identical food finding tasks, and compare their performances. And, Bräuer says, it's the first to test dogs' ability to use causal cues.

The results might align well with previous research, but watching the bewildered apes fail to use the social cues is still striking, says Bräuer.

"It's really strange when you're sitting in front of them and pointing, and they don't understand what to do," she says, "because they understand so much else."

It's possible that one might think that the dogs-all of which had been raised as family pets-had no innate skill at reading human communicative cues, but had simply learned the cues from their owners, notes Bräuer. But a 2003 study by Miklosi, of Eötvös Lorand University in Budapest, suggests otherwise. In the study, published in Current Biology (Vol. 13, No. 9, pages 763-766), he and his colleagues compared dogs with wolves-their closest living relatives-that had been raised by human owners. He found that the dogs still outperformed the human-raised wolves at using communicative cues.

Another study, published in Science in 2002 (Vol. 298, No. 5,598, pages 1,634-1,636) by Max Planck Institute anthropologist Brian Hare, PhD, found the same results. Hare also found in that study that young puppies that had lived their entire lives in kennels, with little human contact, were just as good at reading human social cues as same-age puppies raised as pets.

Evolution and emotion

Researchers are interested in dogs not only to learn more about dogs themselves, according to Hare, but also for what dogs may be able to tell us about human evolution.

One of the mysteries evolution researchers have yet to solve is how humans branched off from other primates to develop language, theory of mind and all of the other things that define humanity.

"Everyone agrees that we're different, but no one knows how it happened," says Hare. "If there were a lot of Neanderthals running around, we might have a better chance at figuring it out."

As it is, though, researchers have mostly been left studying the cognitive capabilities of other primates to figure out how they differ from and overlap with humans.

But if dogs, as this study suggests, share some communicative skills with humans that other primates don't, it might be a case of convergent evolution-two species separately evolving the same skills. And that convergent evolution might help explain the circumstances that led to humans' social and, eventually, cognitive skills.

Some research on dogs, for example, suggests that human social skills might have evolved as an accidental byproduct of genetic selection for low aggression and tame behavior. As evidence for this theory, Hare points to his recent study of a Siberian fox farm. For 50 years, researchers there have been selecting foxes for breeding based only on whether the pups nonaggressively approached a researcher. In the study, published last February in Current Biology (Vol. 15, No. 3, pages 226-230), Hare found that the foxes bred through this method turned out to be as good at reading human communicative cues as dogs.

This suggests, Hare says, that some of humans' communicative abilities-the abilities that might have led to language, theory of mind and other cognitive hallmarks of humanity-also may be a byproduct of evolutionary pressures selecting for certain emotional temperaments.

Such research gives evolution researchers a window into theories impossible to study with apes alone, Bräuer says.

"Cognition is not found only in humans," she says. "Other animals show it too. Apes are interesting because they're our closest relatives, but dogs are interesting because they've been living with us so long."