In 2002, clinical psychologist György Gergely and colleagues published a paper in Nature entitled Rational imitation in preverbal infants. The study suggested that 14-month-old infants do not simply mimic; they make rational choices about what to imitate depending on the situation. A recent paper in the journal Current Biology shows that selective imitation is not limited to humans—dogs do it too.

In Gergely’s study, a demonstrator turned on a touch-activated lamp with her head. If the demonstrator's hands were free when she turned on the lamp, the infant-observer usually chose to imitate the action and used his own head to turn on the light. If, however, the demonstrator’s hands were occupied when she turned on the lamp, the infant was more likely to use his hands instead.

Gergely, the head of the developmental research department at the Hungarian Academy of Sciences, and colleagues interpreted these findings to mean that the infants were making a rational calculation: “If infants noticed that the demonstrator declined to use her hands despite the fact that they were free, they may have inferred that the head action must offer some advantage in turning on the light. They therefore used the same action themselves in the same situation,” the study says.

Friederike Range, a post-doc in the department for neurobiology and cognition research at the University of Vienna, swapped the lamp for a kibble-dispenser and babies for doggies, but Range says the results were nearly identical.

Range trained Guinness, her female Border collie, to pull a wooden rod with her paws. When tugged, the rod dispensed a dog treat. Dogs prefer to use their mouth to pull, Range says. When the observer-dogs watched the demonstrator, they imitated the same way the infants imitated in the lamp experiment. On average, if the demonstrator-dog had a ball in her mouth during the pulling, the imitator dogs used their mouths to tug; if the demonstrator’s mouth was free, the imitators used their paws. This suggests, the study says, that when the imitators saw a free mouth combined with paw use, they deduced that there was some special advantage to using their paws.

“It is a very analogous study," Range says, referencing Gergely’s infant study. "The interesting thing is that we get just exactly the same results with dogs as with the children, which is intriguing, but it does not really tell us anything about the cognitive processes behind it.”

This type of circumstance-driven imitation has not been documented in chimps, or any other animal besides humans. Range says this may have to do with canine domestication: “Through domestication, dogs were bred and educated through their ontogeny to pay attention to humans. Humans might ask them to do some really stupid things that don’t really make any sense,” says Range, adding, “but if they do it, they get a reward.”

One difficulty Range encountered in performing the experiments was finding suitable subjects. All the dogs in the study were well-trained—as rescue dogs, agility dogs, even dancing dogs. (There was no discrimination on the basis of species or sex.) “These dogs are usually trained to try things by themselves," Range explains. "It wouldn’t help to have a dog who is very obedient and does exactly what the owner tells him to do.”

Instead, the dogs are supposed to be able to learn something by themselves. To make sure meddling owners were not skewing the results, Range and her colleagues blindfolded a percentage of the dogs' owners so they wouldn't influence the dogs during the experiment. The results were the same.

Although the dogs behaved remarkably similarly to the infants, Range says this study should not be taken as evidence that dogs are somehow smarter than previously thought; Range points out that 14-month-old infants aren't particularly high-level thinkers either.

In fact, while Gergely dubbed the behavior “rational imitation” in infants, Range and her colleagues prefer the term “selective imitation" for the dogs. "We deliberately did not want to use the word rational for dogs because we don’t believe there is anything rational in here,” Range says. “We’re not sure what’s going on but it’s probably very low-level cognitive processes.”

-Flora Lichtman

What did you think of the story? Send us some feedback.