Study: Gossip Trumps Truth

Jeanna Bryner

Date: 15 October 2007 Time: 01:00 PM ET

Participants relied on gossip about others, even when it contradicted their own direct observations.
CREDIT: Dreamstime.com

People are influenced by gossip about others, even when it contradicts what they see with their own eyes, suggests a new study.

Past research has found that gossip—those juicy tidbits of supposed fact we share about a third party—serves many purposes, including strengthening social ties, spreading social norms and helping others avoid double-crossers and other risky partners.

Hearsay can be the most reliable source of information about situations with which you have no experience. But when you hear gossip that's incongruent with a person or incident you are familiar with, you'd be smart to throw that chitchat out the window in favor of your own direct knowledge, right?

The new study, published this week online in the Proceedings of the National Academy of Sciences, reveals individuals sometimes place so much stock in gossip that they accept it as true even if their own observations and experiences suggest otherwise.

"Gossip has a strong manipulative potential that could be used by cheaters to change the reputation of others or even change their own," lead author Ralf Sommerfeld of the Max-Planck-Institute for Evolutionary Biology and his colleagues write. "This finding suggests that humans are used to basing their decisions on gossip, rumors or other spoken information."

Gossip game

Sommerfeld and his colleagues examined how gossip transmitted information and how it affected another person's behavior.

In the study, 126 undergraduate biology students played a computer-based game in which each student was paired up with another student (via their computers) and had to decide whether to give a certain amount of their starting money to the partner. By dishing out 1.25 Euros, the receiver got 2 Euros, so being on the receiving end was a must. The assumption was that in later rounds, your generosity would be rewarded with generosity toward you.

Over a series of rounds, students switched their partners and received that partner's track record—how many times the person had given money and not given money. Students were more likely to give money to cooperative partners who had previously given money to others.

Then, they had to write a snippet of gossip about the other players they had virtually-interacted with. Sommerfeld noted some gossip examples: "He's a generous player" or "He's a Scrooge, watch out."

No surprise: Players who read a positive comment about another individual, having no knowledge of that person's past generosity record, were more likely to hand over cash to that individual. The opposite was true for negative gossip, where players held tight to their money.

Golden gossip

In another set of rounds, it got more interesting: Players received information on each partner's track record (how often they said "yes" and "no" to doling out money) as well as the gossip blurb.

Without any added gossip information, students cooperated 62 percent of the time. That number increased to 75 percent when students had positive gossip in addition to the partner's track record. Even in instances where the partner had a track record of no giving, positive gossip won out and the other individual handed over money to their partner.

The weirder outcome is that negative gossip decreased cooperation to just 50 percent, regardless of the players' track records.

"If people would act rationally, they would only base their decisions on what they really see because they know exactly the past behavior of these people," Sommerfeld told LiveScience. "But they were still influenced by this gossip."

Gossip also showed this persuasive power in light of any information marring the reputation of the actual gossip monger. For instance, participants acted on gossip even when a blurb (also considered gossip) described the actual source as a "nasty miser" or other uncooperative description.

The scientists suggest the added information might be an overload for participants, or perhaps people don't link cooperative behavior with gossip honesty.