Abstract
The dynamical mechanisms that can stabilize the coexistence of species (or strategies) are of substantial interest for the maintenance of biodiversity and in sociobehavioural dynamics. We investigate the mean extinction time in the coevolutionary dynamics of three cyclically invading strategies for different evolutionary processes on various classes of complex networks, including random graphs, scale-free and small world networks. We find that scale-free and random graphs lead to a strong stabilization of coexistence both for the Moran process and the Local Update process. The stabilization is of an order of magnitude stronger compared to a lattice topology, and is mainly caused by the degree heterogeneity of the graph. However, evolutionary processes on graphs can be defined in many variants, and we show that in a process using effective payoffs the effect of the network topology can be completely reversed. Thus, stabilization of coexistence depends on both network geometry and underlying evolutionary process.
Original language | English |
---|---|
Publication status | Published - 15.03.2010 |