Human social networks might not be limited by 'Dunbar's number' — but Dunbar disagrees

By Nick Gallagher
May 4, 2021
A long-standing mathematical limit on the size of social circles is being challenged. (Pexels/Jeffrey Czum)

A long-standing mathematical limit on the size of social circles is being challenged. (Pexels/Jeffrey Czum)

A team of Swedish researchers has found no basis for the existence of Dunbar's number, a hypothesized average of the number of individuals who can practically fit within a functioning social circle, often cited as 150 people. 

Their findings, published Wednesday in Biology Letters, suggest that human social groups aren't strictly bounded by cognitive ability, countering a theory first developed by British evolutionary psychologist Robin Dunbar. The team says they benefited from an updated dataset as well as advanced analytical techniques that were not yet available when the theory was first proposed during the early 1990s. Their results suggest it may not always be beneficial to reorganize corporate or social structures to accommodate strict group sizes. The study also broadens scientists' understanding of encephalization — the relationship between animals' body and brain sizes and their corresponding levels of cognition. 

Dunbar's number has become influential both in scientific circles and among the general public. One of Dunbar's key studies on the number, published in 1998, has received more than 3,000 citations, according to Google Scholar. Meanwhile, numerous articles in major magazines and newspapers have been written on the subject, as the concept has captured the popular imagination. 

"The scientific system is quite slow," Johan Lind, an ethologist and an associate professor at Stockholm University and a co-author of the study, told Fastinform. "When ideas are criticized, we as scientists, as well as the general public, [find it] difficult to reject ideas, even though there are good reasons."

Dunbar had originally compared the size of primates' neocortical regions — thought to play a significant role in social interaction — with the number of members of their social groups. He then used a mathematical model to adapt those findings to the human brain. The findings helped explain why primates have unusually large brains compared with other animals, suggesting that the added brain volume in those species is primarily used to maintain complex social systems. Dunbar has extended his findings in a May 4 Biological Reviews paper that places a particular emphasis on four distinct subgroups of the multilevel social dynamics seen in some primates. 

But the Swedish researchers found that when they reconstructed the study, they could land on no single number that connected neocortex size and social group size. "The variation is enormous," Lind explained. One model that the team developed, for instance, showed that the number of people the brain is equipped to socialize with could fall anywhere between four and 508, a range so broad and imprecise that they think it is not useful in any practical context.

Responding to the Biology Letters study in an interview with Fastinform, Dunbar dismissed the Swedish researchers' analytical methods as well as their understanding of the complexity of primate social circles. "Despite nearly a decade and a half of publications on this topic, they seem not to be aware of the scaled structure of human and primate social groups," he said. "They seem not to have understood, or even close to understood, what primate and human social networks are like or about."

The British evolutionary psychologist also criticized the Swedish team's choice to analyze the brain sizes of multiple species of primates, rather than apes alone. "You will get underestimates for the predicted value if you do the analysis on primates as a whole," he said.

But, in response to Dunbar, the Swedish researchers contend that variations between the brains of apes and humans make a direct correlation impossible to pinpoint. "Research points to a distinct difference between humans and other animals in how information is stored and processed, so any regression line based on ape data would incorrectly predict human mental capacities," Lind said.

The Swedish team isn't the first to question Dunbar's number, but most prior studies regarding the limits of primate social cohesion have relied on the same limited data set, which has sometimes led to unreliable, self-contradictory results, according to the authors. 

The findings suggest that, rather than being strictly circumscribed by the limits of our brains, our capacity to interact with large numbers of humans may be augmented by our own social motivations or preferences, which may, in some cases, overcome brain limitations. This has led to a "cumulative cultural evolution resulting in marvels such as Stockholm, symphonies and science," the researchers reported in the study.

The new results also suggest that, despite our social capabilities, humans interpret social cues in much the same way as any other information. "We can remember thousands of digits of pi. Or if you're a zoologist, you can remember 6,000 different beetle names or characteristics," Lind said. "And when it comes to storing and processing information, there is nothing special about social stimuli." 

Dunbar has applied his ape research in a variety of human contexts, with one 2016 study, for instance, considering whether humans possess different tiers of friendship by measuring rates of phone calls made to their contacts. In another study, Dunbar used his number to better understand the dynamics of church congregations. He has also found that the sizes of hunter-gatherer groups match his concept.

Dunbar's theory was intended to explain, through an evolutionary lens, why primates developed larger brains and greater cognitive capacities as compared with other animals. But even before conducting the study, some scientists had intuitive doubts about the rigor of Dunbar's findings. For instance, prior studies had shown that environmental adaptations, including primates' foraging ability as well as their capacity to avoid predators, may play a more significant role in determining the size of their social groups than the size of their neocortical regions alone. 

"Certain human group sizes are good for certain purposes, and other group sizes are good for other purposes," Patrik Lindenfors, a co-author on the study and a researcher at the Institute for Futures Studies, told Fastinform. "Trying to specify one group size for all purposes is mistaken. Humans are very adaptable."

Because of its cultural influence, Dunbar's number has played a role in shaping the fabric of some corporate and governmental systems. Organizations such as the Swedish Tax Authority, for instance, reformatted their office spaces to ensure that employees were separated into 150-person pods. Facebook's former chief product officer seems to have referred to the number while giving a talk about office culture at the 2016 Aspen Ideas Festival, explaining that "weird stuff starts to happen" once companies exceed 150 employees. This new research calls those policies and assertions into question.

"I think we should of course be informed by science when we make decisions," Lind said. "But when it comes to societal questions or organizational questions, I think there are other moral issues that also should be given heavier weight."

The study "Dunbar's number deconstructed," published May 5 in Biology Letters, was authored by Andreas Wartel and Johan Lind, Stockholm University; and Patrik Lindenfors, Institute for Futures Studies and Stockholm University.

Saving
We use cookies to improve your experience on our site and to show you relevant advertising.