Gossip and Ostracism Promote Cooperation --Gossip With Ostracism ◆-G0ssip ●Basi 80- 5 70 65 8 60 ● 0 5 、 ● 50 0 ● 45 40 Round 1 Round 2 Round 3 Round 4 Round 5 Round 6 b 18 17 .--0 16 15 ● 14 13 12 11 10 Round1 Round 2 Round 3 Round 4 Round 5 Round6 (a)group eamings and (b)individual eamings in each round as a function research has shown that gossip can deter selfishness known defectors.By removing defectors from thei 01 einberg et al,2012;Piazza Additionally.the threat of ostracism facilitated by ings in showing that gossip can also foster cooperation effectively deterred defection,as evidenced by our find by facilitating partner selection.When given the opportu- ing that even in Round 1,the gossip-with-ostracisr game featur the hig levels of contribution.Finally select partners.Specifically,participants chose to interact increased their contributions substantially,because exclu with others known to be cooperative while excluding sion compelled them to conform to the more cooperative 3,2014 Gossip and Ostracism Promote Cooperation 7 research has shown that gossip can deter selfishness (Beersma & van Kleef, 2011; Feinberg et al., 2012; Piazza & Bering, 2008) and facilitate indirect reciprocity (Sommerfeld et al., 2007), the present research goes beyond these find￾ings in showing that gossip can also foster cooperation by facilitating partner selection. When given the opportu￾nity, participants readily spread reputational information about other participants; recipients of this gossip, in turn, used the information to form reputation judgments and select partners. Specifically, participants chose to interact with others known to be cooperative while excluding known defectors. By removing defectors from their groups, more cooperative individuals could more freely invest in the public good without fear of exploitation. Additionally, the threat of ostracism facilitated by gossip effectively deterred defection, as evidenced by our find￾ing that even in Round 1, the gossip-with-ostracism game featured the highest levels of contribution. Finally, ostracism influenced the behavior of defectors. When ostracized individuals returned from exclusion, they increased their contributions substantially, because exclu￾sion compelled them to conform to the more cooperative Average Group Earnings 40 80 75 70 65 60 55 50 45 Round 1 Round 2 Round 3 Round 4 Round 5 Round 6 Average Individual Earnings 10 18 17 16 15 14 13 12 11 Round 1 Round 2 Round 3 Round 4 Round 5 Round 6 a b Basic Gossip Gossip With Ostracism Fig. 3. Average (a) group earnings and (b) individual earnings in each round as a function of type of game. Downloaded from pss.sagepub.com by Cai Xing on February 13, 2014