Ravens Are Smart Enough to Understand Other Birds' Social Interactions

In other words, ravens are really smart.

Apr 22 2014, 5:12pm
Ravens kickin' it. Image: Shutterstock

My mom, and perhaps yours too, always said that you are who you walk with. It was generally a warning against hanging out with the 7th graders who smoked cigarette butts off the ground, but in answering the question of how we humans got our giant brains, it's an axiom that seems to ring true. 

The social brain hypothesis (SBH) posits the truly fascination notion that our brains evolved to become highly complex organs, not simply because we need the brain power to solve physical problems, but because we need our smarts to navigate our highly complex social relationships.

If I may offer a highly simplistic example: While it's not always easy to fit a round peg into a round hole, it can be far more taxing to fit yourself into whatever social gathering you're attending, what with navigating the crowd, assessing other people's interactions to sort out cool folks and shitbirds, and then trying to make, you know, actual friends.

If you rather I refer to an actual scientist's explanation, I'll let Robin Dunbar explain, as written in a 1998 article in Evolutionary Anthropology:

The consensus view has traditionally been that brains evolved to process information of ecological relevance. This view, however, ignores an important consideration: Brains are exceedingly expensive both to evolve and to maintain... Claims that primate ecological strategies involve more complex problem-solving are plausible when applied to the behaviors of particular species, such as termite-extraction by chimpanzees and nut-cracking by Cebus monkeys, but fail to explain why all primates, including those that are conventional folivores, require larger brains than those of all other mammals.

An alternative hypothesis offered during the late 1980s was that primates’ large brains reflect the computational demands of the complex social systems that characterize the order. Prima facie, this suggestion seems plausible: There is ample evidence that primate social systems are more complex than those of other species.

Dunbar focuses on primates, as does much of the discussion of the SBH. But there are a whole lot of other animals out there that are either social, smart, or both, all of which may be able to provide insight into how our social systems may have helped our brains evolve.

A study published today in Nature Communications offers a new perspective on the SBH, and it doesn't even involve mammals. Instead, the University of Vienna team led by Jorg Massen focused on ravens, which are known to be highly intelligent, tool-using creatures. But if the SBH were a factor in corvids' intelligence, they, like primates, would have to be able to understand social cues.

"Recently, the SBH has been extended to birds and used to explain the apparent case of convergent evolution of intelligence in apes and corvids," the authors write, citing a 2003 paper in Animal Behaviour. "However, evidence that birds have an understanding of social dynamics similar to that of mammals is still scarce."

Simple sociality abounds in the animal kingdom, from the more or less drone-like nature of ants to the comparatively complex vigilance behavior in gulls. But much of that is stimulus-based: The big boss elephant seal gets to be king, not because everyone can just tell he's in charge, but because he fought to prove his dominance.

A better understanding of high-level social nature is based on the ability to assess third-party relationships. If someone directly interacts with you, that's one thing, but being able to understand the interactions of two other people—like figuring out if that cute human across the room is interested in the other human he or she is talking to, for example—takes far more intelligence.

The test site was divided into the two ravens groups' areas (A and C above) as well as a separated test area which featured the playback speaker. Image: Massen et. al

This test of third-party relationship assessment is exactly what the Vienna team posed for their ravens. Two groups of 8 sub-adults (5:3 female/male in one, 4:4 in the other) were kept separated in an aviary for nine months, although they were allowed full visual and auditory access to each other. After being fully habituated to their experimental setup, which included a center room in which ravens could hear the other groups, but not touch, researchers played audio of two ravens interacting for a lone raven. The groups could either be from the raven's own group or the other group.

In essence, each raven was played a short "conversation" from a pair of individuals. The conversations consisted of two types of vocalizations, self-aggrandizing displays (SAD) and submissive calls, which, when played in combination, can simulate "a mild conflict with clear outcome," as the authors write.

To test if the ravens could really distinguish between other ravens' relationships, the authors played some conversations that were expected—a raven known to be dominant over another winning the argument—and some that were unexpected. Should the ravens actually be paying attention to who was talking, you'd expect a difference in reaction to the "normal" outcome versus the unexpected.

While not all permutations of the study produced significant differences, a number of combinations of interactions did. For the in-group tests, birds of the same sex responded with markedly more activity to unexpected interactions. If I may anthropomorphize a bit, this means Joe the raven was far more animated when hearing Bill the nerd verbally assert his dominance over Frank the tough guy than the other way around—as you might expect, if a bird was smart enough to assess the social relationships of other birds. 

As Nature Communications put it, "two ravens nurturing their good relationship by preening each other." Image: Jorg Massen

Similar differences in response were found for females as whole in in-group tests—males seemed to be less reactive—while for out-group tests, males were the only sex to show a significant difference. Part of this variation may be due to how each sex perceives its own competition or risk of following lower in a hierarchy, and part of it may be due to how ravens assess and respond to potentially threatening social brouhahas.

"Social bonds and pairbond-like friendships are highly important in corvids and the other avian species of interest, creating a system of dependent ranks," the authors write. "This is especially true for female ravens, which substantially gain in rank by bonding to males. The crucial role of males in raven society for gaining and maintaining status might be the reason why only males responded to simulated rank reversals in the outgroup condition of our experiment."

Part of the variation may also be due to the smaller sample size of birds, and the authors note that more study is needed. Regardless, what seems clear is that ravens can and do assess and respond to others' social interactions. 

"Ravens seem to understand third-party rank relations," the authors write. "As they do so, both of individuals within their own group as well as of individuals in a neighboring group, we suggest that ravens are capable of forming representations of others’ relationships that are entirely based on observation of other’s interactions."

This isn't direct evidence that ravens' intelligence is based on a need to understand complex social interactions. However, it is a positive result for the application of SBH in corvids. That the birds can understand and respond to third-party interactions suggests their cognition of social interactions is quite high, which would be expected if SBH was a factor in their intelligence. In other words, ravens are really smart.