Rock-scissors-paper game on regular small-world networks

TL;DR

This study investigates how the structure of small-world networks influences the spatial rock-scissors-paper game, revealing phase transitions from self-organizing patterns to global oscillations and homogeneous states.

Contribution

It introduces a network-based extension of the spatial rock-scissors-paper game, analyzing the effects of random links on pattern formation and oscillation behavior.

Findings

Limit cycle emergence when random links exceed a threshold

Global oscillations grow and lead to homogeneous states at higher randomness

Monte Carlo results align with dynamical cluster predictions

Abstract

The spatial rock-scissors-paper game (or cyclic Lotka-Volterra system) is extended to study how the spatiotemporal patterns are affected by the constructed backgrounds providing uniform number of neighbors (degree) at each site. On the square lattice this system exhibits a self-organizing pattern with equal concentration of the competing strategies (species). If the quenched background is constructed by substituting random links for the nearest neighbor bonds of a square lattice then a limit cycle occurs when the portion of random links exceeds a threshold value. This transition can also be observed if the standard link is replaced temporarily by a random one with a probability $P$ at each step of iteration. Above a second threshold value of $P$ the amplitude of global oscillation increases with time and finally the system reaches one of the homogeneous (absorbing) states. In this case the results of Monte Carlo simulations are compared with the predictions of the dynamical cluster technique evaluating all the configuration probabilities on one-, two-, four-, and six-site clusters.