A new study led by the University of Pennsylvania's Damon Centola provides a scientific explanation for how social conventions can emerge suddenly, seemingly out of nowhere, with no external forces driving their creation. (Annenberg School for Communication, University of Pennsylvania)

Two to three decades ago, parents couldn't get enough of the names “Jessica” and “Jennifer” for their little newborn girls. They had been mainstays on the Social Security Administration's top 10 most popular female baby names for roughly 25 years.

“Jessica” had peaked a decade or so after “Jennifer,” but in the mid-2000s, both saw a shift. These once-ubiquitous names had reached some kind of tipping point, and quickly plummeted down the list. Suddenly “Jessica” and “Jennifer” had become uncool in the eyes of new parents, who preferred names like “Isabella” and “Madison.”

Trends like baby names ebb and flow, seemingly without any rhyme or reason. The same goes with slang words and fashion — the term of endearment “boo” has become “bae,” and jeans nowadays are skinnier than ever. But how does a style or word rise up from utter nothingness to becoming the latest hot thing?

While some believed a central institution or figure had to be behind a skyrocketing trend — say, Kim Kardashian or Vogue magazine — researchers have discovered through a new Web-based experiment that doesn't have to be the case. In fact, the study suggests that populations can come to a consensus about what's cool and what's not in a rapid, yet utterly spontaneous way.


Researchers found that the structure of a social network could determine whether a population came to consensus on one idea. (Annenberg School for Communication/University of Pennsylvania)

“Things take over very quickly, and they also sometimes have a short half-life,” said philosopher Brian Skryms of the University of California at Irvine, who was not involved in the study. “Think about teenage slang: How do you say something is good? Is 'awesome' in now? It changes, and if you use the wrong term, then you're out of it.”

The idea of spontaneous emergence, has been around for decades. However, this study marks the first time it has been observed experimentally in a population of almost 100. The results were published online Monday in the Proceedings of the National Academy of Sciences.

“Spontaneous emergence is like the existence of an invisible hand,” said study author and physicist Andrea Baronchelli of City University London. “Everybody is just trying to agree with their social circle, and spontaneously there will be the emergence of a single consensus.”

Take “spam,” for instance. Now it means junk e-mail, but in the past, its only association was with the canned meat. The likely explanation for the word first being used for e-mail originates in a Monty Python sketch.

“They go to a restaurant, and on the menu, spam is everywhere: Egg, bacon, and spam. Spam, spam, egg, and spam. So it is something annoying that is everywhere,” he said. “In the beginning, there were many other names like 'junk e-mail,' but somehow without anyone telling us that we should agree, we use 'spam' as the way that we should refer to email that we don't want to receive.”

Then a large institution like Google used “spam” for their e-mail client Gmail, which further reinforced the term. But at that point, we as a population had already decided on “spam.” In other words, it had become a social convention.

“Conventions are the fundamental bricks of our social lives — from the way we greet each other, like shaking hands, or how we dress, like a tie,” said Baronchelli. “If you think about it, a tie is a ridiculous piece of clothing.”

To experimentally test for how conventions arise, Baronchelli and co-author Damon Centola of the University of Pennsylvania had online participants play a Web-based game, “The Name Game,” against one another. First, a photo of a human face appears on the screen, and the game asks Player 1 to come up with a name for that face. Simultaneously, Player 2 does the same, and if the names match up, both players get a reward. If not, each received a penalty and saw what the other person put for a name.


To understand how social norms arise, Damon Centola and Andrea Baronchelli invented a Web-based game, which recruited participants from around the world using online advertisements. In each round of the "Name Game," participants were paired, shown a photograph of a human face and asked to give it a name. If both players provided the same name, they won a small amount of money. If they failed, they lost a small amount of money and saw their partner's name suggestion. Depending on the structure of interactions, the players either failed to come to consensus, or all agreed on the same name within a handful of rounds. (Annenberg School for Communication/University of Pennsylvania)

This went on for several rounds with the same photo, and each time players were switched up randomly to face another person in the game. Baronchelli and Centola found that how people are connected will determine whether these spontaneous trends will go global or stay confined within smaller local pockets.

When participants were arranged geographically and restricted to play only against their nearest few neighbors, they would quickly agree on names within their region. But for the game as a whole, there would be several names competing for dominance with no global consensus.

“We see this regionalism in the U.S., where people use certain words in the South vs. the East vs. the West,” said Centola. He suggests this type of geographic network can explain the “soda”/”pop”/”coke” phenomenon and other regional divisions in lingo.

Then the researchers rearranged the participants to be able to play against anyone else in the game, meant to be an equivalent of the Internet. Initially, the game was chaotic, with new names popping up every round. But somewhere after 10 to 12 rounds, a curious shift occurred: Everyone started to use the same name.

“In the very first few rounds, it was just utter failure and no one could match,” Centola said. “But then you had a small, minor option that all of a sudden became explosively used by strangers — it just caught on like wildfire, and within a few rounds, was universally adopted.”

Oddly enough, the names that ended up winning everyone over would be those that weren't all that popular in the other networks. In other words, trends themselves can actually be pretty meaningless. But some names simply give way to others in a more or less random way.

“If you have everybody bumping into everybody else, then they really have to communicate with all sorts of people,” said Skryms. “Sooner or later, one language gains the ascendancy, and then everybody piles onto it and learns it.”

According to the theoretical model, the study results should generalize up to 100 million people.

Of course, some conventions are dictated by a central authority, like driving on the right vs. left side of the road. But the fact that others are spontaneously built up in a collective fashion can be an empowering feeling. For instance, Centola wonders whether one day it would be possible to control the viral spread of a certain idea, in order to shift people's beliefs on pressing issues like vaccinations, gun control, and global warming.

“It's nice work, and I'm happy to see it, ” said Skryms. “By using the internet, they were able to do experiments on much larger networks than people were able to do in the laboratory.”

Kim is a freelance science journalist based in Philadelphia.

Read more:

Why this college student is eating bugs for 30 days

The science behind Brian Williams’s mortifying memory flub

Will Canada become an assisted ‘suicide tourism’ destination?

UK voted yes. Here’s how one baby can have three people’s DNA