Adorable Ducklings Have Abstract Thoughts

Within hours of hatching, these baby birds can learn concepts like “same” or “different”—and they do so faster than human infants.

Nguyen Huy Kham / Reuters

In 1690, English philosopher John Locke claimed that “brutes abstract not.” By “brutes,” he meant our fellow animals. By “abstract,” he meant pulling out general concepts from specific examples. While humans might look at chalk, snow, and milk, and conceive of a property called whiteness, other animals would not. “The having of general ideas is that which puts a perfect distinction betwixt man and brutes,” Locke said. “They are the best of them tied up within those narrow bounds, and have not (as I think) the faculty to enlarge them.”

Mm-hmm. Try telling that to Alex Kacelnik’s ducklings.

Within hours of hatching, ducklings and many other young birds rapidly learn the traits of their mothers. This process, known as imprinting, is one of the fastest and most reliable forms of learning in nature. Typically, it ensures that the vulnerable youngsters follow the right individuals around. But ducklings can also mistakenly imprint on birds of the wrong species, humans, or even inanimate moving objects like bouncing balls.

“In a way, imprinting appears to be simple,” says Kacelnik, who studies animal behavior at the University of Oxford. “But it’s extremely complex because mum is an extremely complex collection of properties. So what is it that a young animal stores in its brain to recognize the identity of its mother?”

A devotee of Locke would argue that the ducklings are just picking up simple traits—perhaps a smell, sound, color, or shape. But Kacelnik and his student Antone Martinho III showed that they can do more. The duo presented newborn ducklings with pairs of objects that were either identical or different in shape or color. And they found that the birds could learn these traits. They weren’t imprinting on a specific shape or color, but on the concepts of “same” or “different.” They were looking beyond the individual objects to think about how they are related. In short, “they were abstracting properties,” says Kacelnik.

“I’m very impressed with the results, and the fact that after so many studies on imprinting, nothing like this has been done before,” says Nathan Emery from Queen Mary University of London, who studies bird behavior and was not involved in the research. “I think this has got the potential to revolutionize our field and will be discussed for years to come.”

This is far from the first blow to the idea that abstract thought is a human-only skill. Researchers have shown that other primates, including chimps and monkeys, can discriminate collections of the same items from sets of different ones—a skill that has been described as “the keel and backbone of thought and reasoning.” Rats can tell same from different too, as apparently can pigeons, parrots, crows, and even bees.

But in almost all of these experiments, the animals were extensively trained. They saw many combinations of objects, which were variously paired with rewards, before being tested. Martinho and Kacelnik did nothing of the kind. They hatched mallard eggs in the dark and, within an hour, ushered the newborn ducklings straight into an experiment.

Put yourself in the ducklings’ shoes. You leave the dark and enter a white arena with a pair of hanging objects—a red cone and a red cylinder—revolving around the centre. You study them for 25 minutes before being herded back into the dark waiting room. When you reenter the arena, there are now two pairs of objects. In one corner,  two pyramids revolve around each other. In the other, a cube goes round a cylinder. Neither pair exactly matches the one you saw before. But the cube and cylinder have a familiar property—they’re different from each other. So you waddle over to them.

Martinho and Kacelnik did this experiment with 76 ducklings. At first, they saw a pair of objects that were same or different, in either shape or color. Then, they gave them a choice between two fresh pairs—one same, the other different.

Around 68 percent of the young birds headed towards the relation that they had seen before. If they had imprinted on one identical pair of objects, they were more likely to prefer a second identical pair over a different one. But if they imprinted on mismatched objects, they preferred other mismatched pairs over identical ones. “Their brains are accumulating a catalog of properties that could apply to other objects,” says Kacelnik.

As Edward Wasserman from the University of Iowa notes in a related commentary, the ducklings “needed to see those relations exemplified by only a single pair of stimuli.” In that sense, they actually outperform human babies. Seven-month-old infants can also pick up concepts of same or different without any supervision, but only after seeing at least four pairs of objects. The ducklings did it in one.

But Jennifer Vonk from Oakland University argues that the ducklings might just have been treating the two objects as one, rather than considering the relations between them. “They may have simply seen one color versus two colors rather than a relationship of sameness or difference,” she says.

Kacelnik says that’s unlikely because the objects were constantly moving and ducklings were seeing them from ever-changing perspectives. But “we do need to investigate in greater detail exactly what’s going on,” he says. “It’s not a finished story.”

Does this mean that ducklings are highly intelligent? Probably not. It might be that we’ve overestimated how important abstract concepts are to intelligence. “We make so many assumptions about which cognitive abilities are more or less complex than others, without actually knowing much about the underlying mechanisms,” says Emery. “We now have to question whether relational concepts should be still be considered a form of complex cognition if they can be programmed into very young brains.”

Perhaps there are different levels of abstract concepts, from simple ones that young birds can quickly learn after limited experience, to complex ones that adult birds can cope with. And perhaps birds pick up these concepts in different ways from other animals.

“Both humans and birds can generalize these abstract relations quickly, but are they using the same mechanisms to do so, or different mechanisms that lead to the same ability to learn relations?” adds Alissa Ferry from the International School for Advanced Studies. “We know there are certain situations that can help or hinder humans from learning abstract relations and it is an open question if we find the same patterns in birds as well.”

These are questions that Kacelnik will pursue. “There’s been a transformation in the last few decades of what we think animals are capable of,” he says. “We think we’re starting to understand better what it is to be a bird.”