Emergence and decline of scientific paradigms in a dynamic complex network

We study the idea spreading process by extending a recent model proposed by Bornholdt et al. [Phys. Rev. Lett. 106, 058701 (2011)] to the population on a two-dimensional square area either in the presence of static topological randomness or in which agents are allowed to freely move. We find that the static topological randomness induced by the random distribution of agents in the square area disfavors the common pattern of scientific paradigms, which is characterized by quick rise and slow decline of an arbitrary dominant idea. However, the common pattern of scientific paradigms is recovered when the movement of agents is enabled. In addition, we find that, when the moving speed of agents is low, the average lifetime of a dominant idea displays an optimal behavior. In contrast, the evolution of the model is no more sensitive to the movement speed once it is high enough.