Psst — Wanna Know Why Gossip Has Evolved in Every Human Society?

A few years ago, researchers attached recording devices to a few hundred Americans and sampled their conversations over several days. They found that these volunteers spent, on average, 52 minutes per day having conversations about people who were not there. In other words, gossiping.

Gossip is a ubiquitous feature of human communication, explains Michele Gelfand, a professor of organizational behavior at Stanford Graduate School of Business. Everyone seems to do it no matter where or when they live. “You can go back to the history and see that it was prevalent in Mesopotamia and Greece. Anthropologists talk about it being at the center of life in hunter-gatherer societies,” Gelfand says. Yet from an evolutionary standpoint, gossip is puzzling. “It’s unclear why gossiping, which requires considerable time and energy on a person’s part as well as giving away potentially very valuable information, evolved as an adaptive strategy at all,” she says. “What’s in it for gossipers?”

To explore the social evolution of gossip, Gelfand assembled a multidisciplinary team that was led by Xinyue Pan, a doctoral student at the University of Maryland, and also included Vincent Hsiao, a postdoctoral research fellow at the U.S. Naval Research Laboratory, and Dana S. Nau, a professor emeritus of computer science at UMD. In a recent article published in PNAS, they conclude that gossipers have an evolutionary edge not only because they spread useful information about people’s reputations but also because they encourage people to behave less selfishly.

“Gossip is helpful because it disseminates information about people’s reputations, and that can help recipients of this information connect with more cooperative people,” Gelfand says. “As more people are thinking about others’ reputations, they’re getting concerned about their own reputations too, and they don’t want to be the subject of future gossip. So that deters them from acting selfishly.”

That may sound counterintuitive. Many people, particularly Americans, assume that gossip is usually disparaging or undermines social cohesion. This misconception, Gelfand says, has “crowded out our understanding of the positive nature of gossip” and its role in the evolution of cooperation. Her team did find, however, that cooperation thrived only when gossipers’ information was accurate.

Gelfand’s and her colleagues observed this playing out in an evolutionary game-theoretical model that they built to simulate how people gossip. The model was populated by “agents” who utilized one of six strategies that determined how they would exchange gossip with neighboring agents. After each interaction, the agents would decide whether to continue with their original gossip strategy or adopt their neighbor’s strategy. “We included many different strategies in this model,” Gelfand says. “The strategy with the highest payoff is more likely to be learned or emulated.”

How Virtual Gossipers Survive

After running thousands of iterations of their model, the researchers found that virtual gossipers not only survived but flourished: Across all the simulations, 90% of agents eventually became gossipers. Gelfand and her coauthors concluded that gossipers proliferated because they served the double function of “reputation dissemination” and “selfishness deterrence.” Individuals acted more cooperatively if they knew that information about their behavior was being shared with others. In turn, they behaved less selfishly around known gossipers, which gave gossipers an evolutionary advantage.

In general, the agents employed strategies in which they used gossip to protect themselves or take advantage of others. By the end of the simulation, nearly 60% of all agents had adopted an “exploitive” strategy, while another 18% took an “opportunistic” strategy. However, the evolution of gossip tempered these common approaches as agents became more sensitive to what others thought of them and more willing to work together, ultimately promoting high cooperation rates.

This dynamic was illustrated by opportunistic agents. These reputation-conscious individuals, Gelfand explains, “are very strategic: When they learn that another agent is a gossiper, then they cooperate with them, but otherwise they don’t.” The opportunists formed a mutually beneficial relationship with gossipers: The gossipers provided a check on the opportunists’ desire to burnish their reputations, while the opportunists boosted the gossipers’ power to deter selfish behavior. In this way, she says, “opportunistic agents and gossipers co-evolve — they need each other.”

“Opportunistic agents kind of get a bad rap. They’re seen as kind of sneaky,” Gelfand says. “But in fact, they’re actually helping a lot in the population. If there’s not a lot of opportunists, there won’t be many gossipers and vice versa.”

This finding, Gelfand notes, “would be very difficult to prove in a large-scale, long-term experiment.” The game-theoretical model made it possible to observe an entire population as it changed over several generations. It also made it possible to dial up or down parameters such as how often the agents interacted and how much information they shared. “We can manipulate mobility; we can manipulate other aspects of network structure,” Gelfand says. “That’s really difficult to do in experiments with real people.”

Gelfand notes that collecting more data in the field could refine the theory introduced in this study and let it be examined across the globe. Yet the evolution of virtual gossipers suggests some takeaways that real people might apply to their communication strategies. “If you’re a gossiper, then you’ll have a reputation as a gossiper and people will know you’re a gossiper,” Gelfand says. “This probably will make them particularly opportunistic to cooperate more with you. And this can be an advantage — so long as you also have a reputation for being accurate about what you spread, of course.”