Fifteenth international online seminar "Artificial Societies and Information Technologies"

Gossip, Reputation, Norms, and Cooperation
Friday, October 15, 2021
Károly Takács

On October 15, 2021, a meeting of the international online seminar "Artificial Societies and Information Technologies" was held, dedicated to the report Károly Takács  "Gossip, Reputation, Norms, and Cooperation".

Brief annotation:

Problems of cooperation are frequent among living organisms, but they are difficult to solve. Human often find a solution following social norms. Social norms dictate appropriate behavior and judgement on others in response to their previous actions and reputation. Using agent-based models, we show that the so-called leading eight norms can maintain high level of cooperation also with local information and local strategy updates. Four of the leading eight norms that do not reward justified defection offer better chances for cooperation with quick evolution, reputation with noise, larger networks, and when unconditional defectors enter the population. Further problems arise, however, because subjective evaluations are more nuanced than just judging others as good or bad. We show that the incremental adjustment of subjective reputation scores after receiving reputation scores of others in gossip is not sufficient to establish cooperation. By contrast, our simulations demonstrate that gossip that also includes perspective taking about possible strategies of others can resolve the problem of cooperation also considering non-binary reputations.

Video of this seminar on our YouTube channel