Gossip and reputation in everyday life

Abstract

Gossip-a sender communicating to a receiver about an absent third party-is hypothesized to impact reputation formation, partner selection, and cooperation. Laboratory experiments have found that people gossip about others' cooperativeness and that they use gossip to condition their cooperation. Here, we move beyond the laboratory and test several predictions from theories of indirect reciprocity and reputation-based partner selection about the content of everyday gossip and how people use it to update the reputation of others in their social network. In a Dutch community sample (N = 309), we sampled daily events in which people either sent or received gossip about a target over 10 days (n_gossip = 5284). Gossip senders frequently shared information about targets' cooperativeness and did so in ways that minimize potential retaliation from targets. Receivers overwhelmingly believed gossip to be true and updated their evaluation of targets based on gossip. In turn, a positive shift in the evaluation of a target was associated with higher intentions to help them in future interactions, and with lower intentions to avoid them in the future. Thus, gossip is used in daily life to impact and update reputations in a way that enables partner selection and indirect reciprocity. This article is part of the theme issue 'The language of cooperation: reputation and honest signalling'.

Keywords: cooperation; experience sampling; gossip; indirect reciprocity; partner selection; reputation.

Figure 1.

The content of gossip in daily life. Note. Distribution of the ratings of gossip content, comparing gossip relevant to a norm violation and with gossip irrelevant to a norm violation, for valence (1 = extremely negative, 7 = extremely positive); trustworthiness (1 = untrustworthy/dishonest, 7 = trustworthy/honest), warmth (1 = cold/disagreeable, 7 = warm/agreeable); competence (1 = ignorant/incompetent, 7 = knowledgeable/competent); and dominance (1 = weak/submissive, 7 = powerful/dominant); in all dimensions 4 = neutral. The box limits show the 25th percentile, the median and the 75th percentile, respectively. The whiskers extend to 1.5 times the interquartile range [70].

Figure 2.

Relationship between the parties involved in gossip. Note. The three lines with arrows depict the relationship value between the participant and the gossip partner (sender and receiver combined), the participant and the gossip target, and the participant's perception of the relationship value between the partner and target for (a) gossip relevant to a norm violation and (b) gossip irrelevant to a norm violation. Darker colours represent higher relationship values. *significant difference (p < 0.001) between the corresponding values for gossip relevant to a norm violation and gossip irrelevant to a norm violation.

Figure 3.

Change in relationship value assigned to the target mediates the effect of gossip content on behavioural intentions. Note. Three mediation models show the indirect effect of the gossip content describing the target's trustworthiness on intentions to (a) help, (b) avoid, and (c) confront the target through a change in relationship value assigned to the target from intake to experience. We also controlled for the interaction between time since intake and relationship value at intake, which was not significant (see the electronic supplementary material). **p < 0.01; ***p < 0.001.