Alladi Ramakrishnan Hall

Does the bounded rationality of agents help to establish conditional cooperation?

Balaraja Battu

Centre of Behavioural and Cognitive Sciences (CBCS), University of Allahabad, Allahabad

Suppose, you are in a new city and somebody asked for a help, do you help a random met agent, if there is no chance of meeting that agent again and no interaction history of that agent is available to you, but you only know how generous the population was (all you know is what was the cooperation level of the city). However, helping somebody cost you a little, but provide a greater benefit to the receiver. Clearly, unless somebody returns your help, helping a random met agent is not a rational decision. A rational agent could do is take the help and never return the help. If everybody behaves rationally, will ever the society can establish cooperation? Does the population consist of heterogeneous conditional agents, who help if and only if the number of donations in the past more than the agents' conditional cooperative criterion (ideal conditional rule) ever establish cooperation? It turns out that when agents are not operating with a strict conditional rule and occasionally not imitating successful agents' social behavior, the population consists of heterogeneous conditional agents can establish high levels of cooperation. It seems social systems require certain levels of flexibility, not the strict rules, to function cooperatively.