Variation in network evolution through competitive progress via link connection/disconnection

Abstract

Abstract Parameter tuning is a useful method for modelling complex phenomena and replacing one collective pattern with another pattern. However, this approach may not explain how agents in a complex system select an adequate parameter or respond flexibly to a changeable environment. Similar problems are also found in the network evolution where the emergence of characteristic network structures poses a problem. In a previous work, I developed a hawk–dove game on the network where individual players sometimes disconnected links from their partners or connected a link to a new partner in addition to the strategy update. In this game, players coordinated the link connection/disconnection rule by considering the local hierarchy; I found that the coordination of this rule presented the emergence of a scale-free network at one time but not so at another time. Hub nodes may emerge when hawk players are extinct. In this work, I investigated the relationship between the link evolution and the evolution of hub nodes that will be linked to the network structure and found that the network structure could evolve differently according to the competitive progress of the link connection/disconnection.