Abstract:The replicator equation in evolutionary game theory describes the change in a population's behaviors over time given suitable incentives. It arises when individuals make decisions using a simple learning process - imitation. A recent emerging framework builds upon this standard model by incorporating game-environment feedback, in which the population's actions affect a shared environment, and in turn, the changing environment shapes incentives for future behaviors. In this paper, we investigate game-environment feedback when individuals instead use a boundedly rational learning rule known as logit learning. We characterize the resulting system's complete set of fixed points and their local stability properties, and how the level of rationality determines overall environmental outcomes in comparison to imitative learning rules. We identify a large parameter space for which logit learning exhibits a wide range of dynamics as the rationality parameter is increased from low to high. Notably, we identify a bifurcation point at which the system exhibits stable limit cycles. When the population is highly rational, the limit cycle collapses and a tragedy of the commons becomes stable.
What problem does this paper attempt to address?
The paper primarily explores the dynamics of a system under the framework of feedback-evolving games when individuals adopt boundedly rational learning rules. Specifically, the paper investigates the following issues:
- **Background and Motivation**: Traditional feedback-evolving game theory considers individuals changing strategies through simple imitation behavior, assuming that individuals do not possess complex cognitive abilities. However, in the real world, individuals sometimes make rational choices based on maximizing payoffs, and sometimes make suboptimal choices due to noise in decision-making or lack of information. Therefore, the paper attempts to explore how the dynamics of feedback-evolving games change when individuals adopt boundedly rational learning methods.
- **Research Content**:
- The paper proposes a feedback-evolving game model where individuals can adopt two strategies: cooperation (C) or defection (D), and these behaviors affect the state of the shared environment.
- The decision-making process of individuals is described by a learning protocol. This paper considers logit learning, a parametric learning rule controlled by the rationality parameter \(\beta \geq 0\). When \(\beta = 0\), individuals choose actions randomly; as \(\beta\) increases, individuals are more likely to choose actions with higher payoffs.
- The focus of the research is to analyze the behavioral characteristics of the system under different levels of rationality \(\beta\), especially in cases where traditional imitation learning leads to the "tragedy of the commons" (all individuals defect, causing environmental resource collapse), and to examine whether rational learning can stabilize more desirable environmental outcomes.
- **Main Findings**:
- When the level of rationality is low, the system may exhibit the "tragedy of the commons"; as the level of rationality increases, the system may exhibit internal fixed points (i.e., resources maintain a non-zero state) or even stable periodic behavior (limit cycles).
- At a certain threshold of rationality, the system undergoes a Hopf bifurcation, leading to the emergence of stable periodic solutions. When the level of rationality is very high, the system returns to the "tragedy of the commons" state.
- The paper provides a detailed analysis of the system's dynamic properties, including the existence and stability conditions of fixed points, as well as the dynamic behavior under different levels of rationality.
In summary, by introducing the logit learning mechanism, this paper aims to explore the impact of individual bounded rationality on collective behavior in feedback-evolving games, particularly how it affects the sustainable use of environmental resources.