Human Evolution: Is Dunbar's Number Really Up?

Researchers challenge, not for the first time, psychologist Robin Dunbar's insistence that the human brain will only allow for about 150 friendships. Behind that debate: How did our brains get so big?

Robin Dunbar/Wikimedia Commons

Back in 1993, the American anthropologist Leslie Aiello and the British psychologist Robin Dunbar published a paper in the journal Current Anthropology that has greatly influenced thinking about human brain evolution ever since. The title already says a lot: “Neocortex Size, Group Size, and the Evolution of Language.” (Although the original paper is behind a paywall, Dunbar published an accessible version of the paper the same year.)

Aiello and Dunbar crunched a bunch of numbers from the hominid fossil record and observations of living apes, and found that the larger the average group size of a species, the larger its brain. This was particularly true, they found, of the animal’s neocortex, the outer layers of the brain where most of the serious thinking goes on. The pair concluded that the correlation among group size, brain size, and neocortex size held pretty tightly throughout the human evolutionary lineage, from australopithecines more than three million years ago to modern-day humans.

Scientists who specialize in the evolution of the human brain know this paper very well, along with the wellspring of research it generated over the nearly 30 subsequent years. But many in the general public may know about one key conclusion of the paper: The average group size of today’s Homo sapiens, Aiello and Dunbar concluded, was about 150 people. This figure, ever since known as “Dunbar’s number” (possibly because Dunbar is generally more talkative and effusive than the more reserved Aiello, a statement I base on knowing them both personally), has really stuck in the public imagination, and shows up in lots of discussions of human group dynamics, ranging from village life to the internet and Facebook groups.

But Dunbar’s number has repeatedly come under attack from other researchers, who have taken issue not only with the number itself—which perhaps is somewhat less important in the scheme of things—but the so-called “social brain hypothesis” of the evolution of human intelligence, which is what the Aiello-Dunbar paper was actually all about (more on that in a minute.)

Most recently, a trio of scientists from Stockholm University, publishing in the journal Biology Letters (open access), claimed to “deconstruct” Dunbar’s number and find it lacking. Using some fancy statistics, including the very trendy Bayesian methods that lend prestige and gravitas to many of today’s scientific papers, the Swedish group concluded that the real number could be as high as 500; more importantly, the researchers concluded that it was not possible to designate an upper limit to human group size using Dunbar’s assumptions.

The news made it to the New York Times, which published an article on the controversy by Jenny Gross entitled “Can You Have More Than 150 Friends?”

I read this piece with some despair, for at least two reasons. First, as far as I am aware (but happy to be corrected in the comments section) no one has really defined “friendship” in a rigorous way, such that a specific number—whether Dunbar’s or anyone else’s—would be meaningful in a scientific context. We all may think we know the difference between a friend and an acquaintance, perhaps based on the kind of intimate conversations we can have with them or other factors. But in times of trouble, as we know, we find out who our friends really are, and sometimes they turn out to be people we thought were just acquaintances (I have had that experience many times, and I will bet you have too.)

Not only that, but in the Internet Age, we seem to be able to form trusting relationships with people we have never met in person and may never meet in person. As a reporter, I have anonymous sources who have put a huge amount of trust in me to safeguard their identities, risking their jobs and even their careers by telling me things that are, in essence, dangerous for them to know.

Moreover, I certainly do not have 150 friends with whom I can talk about very personal things; perhaps 10-15 at most. (And sadly, as I grow older, some of those very close friends have died and not been replaced.)

The other reason I despair is that the real point of the Aiello-Dunbar paper, and all the research since, gets lost in this more limited discussion about human group size. In the Times article, Dunbar is quoted as saying that the Swedish analysis is “bonkers, absolutely bonkers,” and that their statistical analysis was flawed. I’m sure Dunbar went into more detail with the reporter than that, but if so, it got edited out. In a piece the following day in The Conversation, Dunbar defended his number and went into just a tiny bit more detail about why his own statistical analysis was superior. But again, Dunbar said very little about the social brain hypothesis, which his research is really all about, and which seeks to provide an understanding of why the modern human brain got so big.

Dunbar did, however, briefly discuss the importance of “touch-based bonding” to human relationships, and the importance of hugging and other forms of physical touch. This made me wonder if my Facebook friends really were friends, especially those I have never touched; and while I took heart that I could now hug my fully vaccinated friends as the pandemic winds down, I also wondered how many insincere hugs I had given to acquaintances and been given in return.

But I digress, which is exactly what I am accusing others of doing. The real issue is why the human brain got so big, and it is on this topic that most human evolution researchers have focused their minds and their talents.

Why are human brains so ridiculously big?

“What would you do with a brain if you had one?” Dorothy’s question to the Scarecrow in “The Wizard of Oz” elicited one of the movie’s most delightful songs, in which her straw-filled friend assured Dorothy that, among other things, he would “think of things I’d never thunk before.”

But the Scarecrow seemed to do quite well without one, thus avoiding the high energy costs of fueling and cooling a human brain—which, with an average volume of about 1,400 cubic centimeters, is humongous relative to our body size (especially the neocortex.)

The basic principle of the social brain hypothesis is that the evolution of the human brain was driven by our increasingly complex social relationships. In this view, we required greater neural processing power so that we could keep track of who was doing what to whom.

Our expanded brains were useful for other things, of course, such as innovations in tool use and food gathering. Most researchers, including Dunbar, agree that these hypotheses are not mutually exclusive. But whatever the reasons for the very large human noggin, there is a lot of explaining to do. That’s because big brains actually have a lot going against them.

The oversized Homo sapiens brain let us take over the planet, build cities, send space probes to Mars, and do all the other marvelous things we humans are so proud of. But none of these accomplishments makes us better at reproducing our species, and in terms of evolution, that’s all that really matters. It’s not so obvious why Darwinian natural selection should have favored the brain’s dramatic expansion given the huge costs. Although the human brain is only about 2 percent of total body weight, it siphons off about 20 percent of our total calorie intake; this overall percentage varies little whether we are engaged in hard mental tasks or just zoning out.

If having a large brain were all that advantageous, it seems that every animal would have one. And yet most species have been content, evolutionarily speaking, with relatively small ones. The brain of our closest cousin, the chimpanzee, is less than a third the volume of ours, even though chimps weigh almost as much as humans.

One possibility, Aiello, Dunbar, and others have suggested, is that as human ancestors moved out of the forests occupied by other apes and into open savannahs in Africa, millions of years ago, they banded together to avoid predators such as hyenas and other large carnivores. That greater social proximity, they suggested, may have encouraged the evolution of language, leading to even more brain expansion.

Coming up with direct evidence for such hypotheses is tricky because researchers have to establish that enlarged brains and big social groups led to oversized reproductive success. Unfortunately, neither language nor romance leave traces in the fossil record. But back in 2003, Joan Silk, an anthropologist now at Arizona State University, showed that the offspring of female baboons who have stronger social bonds with their peers survive longer. Subsequent work with primates, and even horses, has shown similar reproductive advantages to sticking together.

Other brains, other ideas

As stimulating as the social brain hypothesis has been in generating literally thousands of research papers, many scientists are troubled by its focus on just one factor, human sociality, to the prejudice of other elements. Back in the 1980s, Richard Byrne, a cognitive neuroscientist in the United Kingdom, teamed up with psychologist Andrew Whiten to formulate what they called the “Machiavellian intelligence” hypothesis. Byrne and Whiten argued that apes and humans evolved larger brains “to solve challenging food acquisition problems better than monkeys,” as Byrne put it to me some years ago for an article in Science (paywall.) These challenges, Byrne added, led to brains better equipped to understand cause and effect—necessary for the development of tool use and understanding what other apes were thinking. (This mind-reading talent is key to what many researchers call Theory of Mind, and what the Max Planck Society primatologist Michael Tomasello has called “shared intentionality.”)

A similar emphasis on a broader range of factors behind brain evolution has been argued by University of Zurich primatologist Carel van Schaik, who, together with his former student Judith Burkart, developed what they call the “cultural intelligence hypothesis” about a decade ago. They contended that the social brain hypothesis, with its emphasis on social relationships and group complexity, did a poor job in predicting the brain size of some primates. Van Schaik and Burkart, along with other researchers more recently, have argued that equal billing has to be given to factors such as behavioral flexibility and their ability to learn from others (social learning.)

Will the minds meet over human brain evolution?

There’s nothing to solve a scientific argument like some empirical evidence, or at least one would hope (I will leave the question of how rational scientists really are to a later post.) To me, one of the most interesting attempts to test these hypotheses was published in 2017, by a research group led by Simon Reader of McGill University in Montreal and Kevin Laland the University of St. Andrews in the UK. Publishing in the Proceedings of the National Academy of Sciences (paywall, sadly), together with St. Andrews researchers Sally Street and Ana Navarrete, the research team used fancy Bayesian statistics and a large data set of primate brain measurements to explore the relationship between social learning, brain size, social complexity, and lifespan. I think I will let the researchers speak for themselves, by reproducing their entire abstract—not out of laziness, after this long post, but to provide readers with the maximum of information short of reading the paper themselves.

Abstract

Explanations for primate brain expansion and the evolution of human cognition and culture remain contentious despite extensive research. While multiple comparative analyses have investigated variation in brain size across primate species, very few have addressed why primates vary in how much they use social learning. Here, we evaluate the hypothesis that the enhanced reliance on socially transmitted behavior observed in some primates has coevolved with enlarged brains, complex sociality, and extended lifespans. Using recently developed phylogenetic comparative methods we show that, across primate species, a measure of social learning proclivity increases with absolute and relative brain volume, longevity (specifically reproductive lifespan), and social group size, correcting for research effort. We also confirm relationships of absolute and relative brain volume with longevity (both juvenile period and reproductive lifespan) and social group size, although longevity is generally the stronger predictor. Relationships between social learning, brain volume, and longevity remain when controlling for maternal investment and are therefore not simply explained as a by-product of the generally slower life history expected for larger brained species. Our findings suggest that both brain expansion and high reliance on culturally transmitted behavior coevolved with sociality and extended lifespan in primates. This coevolution is consistent with the hypothesis that the evolution of large brains, sociality, and long lifespans has promoted reliance on culture, with reliance on culture in turn driving further increases in brain volume, cognitive abilities, and lifespans in some primate lineages.

The take home lesson, besides the obvious conclusion that human evolution is complicated, is that a number of factors have blended together to give us our admirably large and complex brains, but that culture and lifespan—a factor not previously emphasized in much of human evolution research—are playing the key roles.

Does this mean that everyone is a little bit right, including Dunbar? Most likely, if they would only admit it.

Another take home lesson: Perhaps it’s just as well that the Scarecrow never got the brain the wizard promised him (I’m assuming you have seen the movie, but otherwise, spoiler alert.) If he had, he might have spent the rest of his days—like many researchers do now—trying to figure out where it really came from.