Effects of behavioral patterns and network topology structures on Parrondo’s paradox

Abstract

A multi-agent Parrondo’s model based on complex networks is used in the current study. For Parrondo’s game A, the individual interaction can be categorized into five types of behavioral patterns: the Matthew effect, harmony, cooperation, poor-competition-rich-cooperation and a random mode. The parameter space of Parrondo’s paradox pertaining to each behavioral pattern, and the gradual change of the parameter space from a two-dimensional lattice to a random network and from a random network to a scale-free network was analyzed. The simulation results suggest that the size of the region of the parameter space that elicits Parrondo’s paradox is positively correlated with the heterogeneity of the degree distribution of the network. For two distinct sets of probability parameters, the microcosmic reasons underlying the occurrence of the paradox under the scale-free network are elaborated. Common interaction mechanisms of the asymmetric structure of game B, behavioral patterns and network topology are also revealed.

Predator Dormancy is a Stable Adaptive Strategy due to Parrondo's Paradox

“…So why does this counter‐intuitive result occur? Just as in other manifestations of Parrondo's paradox, one of the “losing” strategies is capital‐dependent (in the case of economic games) or density‐dependent (in the case of ecological games), while the other strategy is not . In particular, the growth of the active form of the dormitive predator is dependent on the density of the prey, which itself depends on the density of the active predator.…”

Paradoxical Survival: Examining the Parrondo Effect across Biology

“…One such model was developed by Ye et al, who studied a multiagent social game in which participants trade capital between themselves in accordance to a multitude of behavioral modes, encompassing the Matthew effect, cooperation, poor‐competition–rich‐cooperation, and random patterns. The social game was constructed as a linear combination of a zero‐sum first game and a losing second game, but winning outcomes were found to be possible across all types of behavioral patterns.…”

Paradoxical Simulations to Enhance Education in Mathematics

“…Details on this variant and the related capabilities of the simulation suite can be found in Section A of the Supplementary Information. The multi-agent, graph-theoretic nature of this variant makes it suitable as a higher-order thinking question for students, and had also enabled close connections with the analysis and control of networked systems, encompassing spatio-temporal noise suppression and topological effects on information propagation [65]- [70].…”