Title
A tag-mediated N-person Prisoner's Dilemma game on networks with different topologies
Abstract
In tag-based Prisoner's Dilemma games, agents learn tolerance, copy traits from more successful neighbors, cooperate with in-group alters, and otherwise defect. Current research concludes that cooperation can emerge and be sustained, tolerance decreases, and agents become homogeneous. Such emergent societies are vulnerable to invasion by mutants with the same traits as those of cooperators but who defect. However, one question that has yet to be answered is whether these results can vary from one network structure to another. We investigate tag-based parochial cooperation not only on grid but also on local, small-world, and random network. We find that cooperation can be institutionalized at lower benefit-to-cost ratios than the average degree given tag-based identities, which would not otherwise happen. The average tolerance is positively influenced by the benefit-to-cost ratio, the number of identity dimensions, and heterogeneity. None of societies holds evolutionarily stable parochial cooperation since the selection pressure for homogenization is strong in a small population. Nonetheless, there are distinctive patterns in the emergence of cooperation, its diffusion, and invasion by mutants across societies. We propose future directions discussing how persistent heterogeneity and the dynamics of clustering and bridging are critical for the evolutionary stability of parochial cooperation in more tolerant societies.
Year
Venue
Keywords
2009
SpringSim
average tolerance,parochial cooperation,network structure,dilemma game,tag-mediated n-person prisoner,average degree,evolutionarily stable parochial cooperation,benefit-to-cost ratio,different topology,lower benefit-to-cost ratio,tolerance decrease,tag-based prisoner,tag-based identity,social network,prisoner s dilemma
Field
DocType
Citations 
Population,Social network,Random graph,Computer science,Bridging (networking),Microeconomics,Evolutionary stability,Prisoner's dilemma,Network topology,Real-time computing,Artificial intelligence,Dilemma
Conference
0
PageRank 
References 
Authors
0.34
5
1
Name
Order
Citations
PageRank
Jae Woo Kim1289.47