Scientists say ravens display foresight, a trait thought unique to apes
A series of experiments with ravens finds that the birds display the ability to think ahead and deliberately prepare for future events, adding to the growing body of evidence that intelligence has evolved more than once.
According to Norse mythology, the god Odin has two ravens that fly all over Midgard to gather information. Their names are Huginn and Muninn, the Old Norse words for thought
and memory.
The ancient storytellers who bestowed these names on the birds were onto something: A new study finds that ravens can flexibly plan for events outside their present sensory awareness, a cognitive skill once considered exclusive to humans and other great apes.
This research does more than just reveal that the raven is smarter than we thought, that apes’ intellectual abilities are less unique than we thought, and that a mammalian lineage is a not a prerequisite for complex thinking. It also adds to the growing body of evidence that intelligence has evolved more than once.
It’s pretty amazing in that it works, functionally at least, in a similar way in ravens as in apes, and you don’t see it in many other animals,
says Mathias Osvath, a cognitive zoologist at Lund University in Sweden. Like monkeys – they don’t even pass the baseline studies.
Until fairly recently, comparative psychologists studying the evolution of intelligence focused their attention on mammals – mainly primates and rats – and used birds mostly for associative-learning models that measured responses to stimuli. After all, compared to mammals, birds have small brains relative to their bodies, and bird brains lack a neocortex, which in mammals is thought to be the seat of higher-order thinking such as reasoning, problem-solving, language, and delaying gratification.
But today we know that corvids, a group of about 120 bird species that includes ravens, crows, and jackdaws, possess these abilities, no neocortex required. Corvids can make and use tools, barter, recognize individual faces, and remember who shortchanged them in bartering exchanges. They can understand water displacement and mimic human voices. They can grasp the concept of death and its causes. They can infer when they are being watched, suggesting that they possess a theory of mind. They even engage in play.
Smarter than a toddler?
In a series of experiments on five hand-raised ravens described by Professor Osvath and Can Kabadayi, a doctoral student at Lund, published in the journal Science on Thursday, ravens display the ability to think ahead and deliberately prepare for future events.
In one version of the experiment, Osvath and Mr. Kabadayi trained ravens to use a tool to open a box containing a piece of dog kibble, a popular treat among ravens. An hour after researchers removed the tool and the box, they presented the ravens with the tool on a tray alongside a series of nonfunctional distractor
objects and a smaller food reward. When presented with the box containing the kibble fifteen minutes later, the ravens passed on the smaller reward 86 percent of the time, ignoring the distractors, and picking the correct tool to open the box. When the delay was extended to 17 hours, the ravens picked the right tool 89 percent of the time.
No other animal, aside from apes, is known to be able to pass this test. Even human children under age four typically struggle to get it right. In fact, one of the ravens had to be excluded from further trials after she figured out how to open the reward box using tree bark instead of the tool. She had outsmarted the human experimenters.
In another experiment, where the ravens were trained to use tokens to barter with humans for food rewards, the birds outperformed orangutans, bonobos, and chimpanzees.
I’m pretty impressed by the levels they showed – the levels of success in all experiments,
says Osvath. Every single individual was highly significant in all conditions in all experiments.
How’d they get so smart?
The last common ancestor of humans and birds lived some 320 million years ago, suggesting that these advanced cognitive traits emerged independently in hominids and corvids, using very different brains.
What’s so interesting to me is that birds couldn’t have really adopted the same solution that we adopted in terms of the neuroanatomy,
says Cameron Buckner, a philosopher at the University of Houston who has studied raven cognition. So I think one of the most interesting open questions in the field of animal cognition right now is, how on Earth the birds do it with this smaller brain?
The independent emergence of flexible planning might be an example of convergent evolution, a phenomenon by which an environment selects for strikingly similar adaptations among unrelated lineages. Other examples of convergence include the rise of flight among birds, bats, insects, and pterodactyls, and the evolution of opposable thumbs in primates, pandas, and opossums.
Osvath suggests that corvid cognition could also be an example of a related concept called parallel evolution, in which basic features in related lineages develop into similar traits. Maybe it is something in the common ancestor of mammals and birds that had had some structures that are really good to build complex cognition upon, given certain selective pressures,
says Osvath.
But whether its convergence or parallelism, what sort of environment selects for intelligence? Like many researchers who study the origin of intelligence, Professor Buckner suspects that it has something to do with having a complex social structure. Adolescent ravens live in kind of roving bands without a fixed territory where social alliances can change very quickly,
he says. Who’s your friend and who’s your foe can change very rapidly.
Later in life, ravens form monogamous pair bonds in which they raise young together and defend their territory.
This is pretty similar to the kinds of complex and variable lives that humans lead, where you need to be able to update your beliefs about the world very quickly and think think on the fly about the results of all kinds of different social encounters and their outcomes,
Buckner says.
They’re more social than a dog,
says Osvath. But, he says, We don’t know if they are social like that because they’re clever, or the other way around.