Chimera states in network-organized public goods games with destructive agents

It is shown that a network-organized metapopulation of cooperators, defectors and destructive agents playing the public goods game with mutations, can collectively reach global synchronization or chimera states. Global synchronization is accompanied by a collective periodic burst of cooperation, whereas chimera states reflect the tendency of the networked metapopulation to be fragmented in clusters of synchronous and incoherent bursts of cooperation. Numerical simulations have shown that the system’s dynamics alternates between these two steady states through a first order transition. Depending on the parameters determining the dynamical and topological properties, chimera states with different numbers of coherent and incoherent clusters are observed. These results present (to my knowledge) the first systematic study of chimera states and their characterization in the context of evolutionary game theory. This provides a valuable insight into the details of their occurrence, extending the relevance of such states to natural and social systems.

The evolution of the three strategies (x: cooperators, y: defectors, z: destructive agents) is described by the replicator-mutator equations
\dot{x} =x (P_x -\bar{P}) + \mu (1 - 3 x)\,\\[5pt]\dot{y} =y (P_y -\bar{P}) + \mu (1 - 3 y) \,\\[5pt]\dot{z} = z (P_z -\bar{P})+ \mu (1 - 3 z)

where \bar{P}=xP_x+yP_y+zP_z is the average payoff and
P_x = r \frac{x}{1-z} \left[ 1 - \frac{1-z^n}{n(1-z)} \right] + \frac{r}{n} \frac{1-z^n}{1-z} - 1 - d \left( \frac{1-z^n}{1-z} - 1\right)\,,\\[5pt] P_y = P_x + 1 - \frac{r}{n} \frac{1-z^n}{1-z}\,,\\[5pt] P_z = 0

This dynamical system has a Hopf bifurcation beyond which self-sustained oscillations emerge. They correspond to periodic bursting oscillations of cooperation.

When the populations playing the public goods game (locally, in each node) are connected with a nonlocal ring networks as described by the equations
\dot{x}_i =x_i (P_{x,i} -\bar{P}_i) + \mu (1 - 3 x_i)+ \frac{\sigma}{2R} \sum_{j=i-R}^{j=i+R}\! (x_j - x_i)\,,\\[5pt]\dot{y}_i = y_i (P_{y,i} -\bar{P}_i)+ \mu (1 - 3 y_i) + \frac{\sigma}{2R} \sum_{j=i-R}^{j=i+R}\! (y_j - y_i)\,,\\[5pt]z_i = 1 - x_i - y_i

then chimera states can emerge, reflecting the tendency of the networked metapopulation to be fragmented in clusters of synchronous and incoherent bursts of cooperation.

 

 

Further reading:

  1. N. E. Kouvaris, R. J. Requejo, J. Hizanidis and A. Diaz-Guilera.
    Chimera states in a network-organized public goods game with destructive agents.
    Chaos 26, 123108 (2016).
    [Chaos][arXiv]
  2. A. Arenas, J. Camacho, J. A. Cuesta, R. J. Requejo.
    The joker effect: Cooperation driven by destructive agents.
    Journal of Theoretical Biology 279, 113-119 (2011).
    [J. Theor. Biol.]