This study employs Cooperative, Behavioral, and Experimental game theory to examine how forest rights are negotiated among tribal communities, government agencies, and civil society organizations in the western Himalayas. It explores how claims over access, governance, and benefit-sharing regarding forest resources are asserted, contested, and mediated in a complex socio-political environment.

Cooperative game theory is used to model alliance-building and bargaining dynamics, shedding light on how stakeholders form coalitions and negotiate distributive outcomes. Behavioral game theory provides insight into how cognitive biases, cultural values, and perceptions of legitimacy influence real-world decision-making. Experimental game theory is applied to simulate institutional scenarios, testing how variations in governance structures and policy design shape patterns of cooperation or conflict.

This research adopts a mixed-methods approach. Qualitative data from in-depth interviews and focus group discussions captures stakeholder motivations, strategic reasoning, and the cultural framing of forest rights. Quantitative data from structured surveys and field-based experiments enables the modeling of negotiation outcomes and the empirical testing of policy interventions.

The findings underscore the importance of integrating traditional knowledge systems with modern development policies. This study emphasizes the need for sustainable and inclusive strategies that protect both the environment and local livelihoods. By offering a nuanced understanding of how strategic interactions shape the outcomes of forest rights debates, this research provides a framework that can be adapted to similar environmental and developmental challenges in other regions.

Background: Psychosis is a severe mental health condition requiring continuous monitoring and timely interventions. Traditional approaches often fail to adapt to real-time fluctuations in a patient’s mental state, limiting their effectiveness. This study introduces a Partially Observable Stochastic Game (POSG) framework to model interactions among key stakeholders, including patients, caregivers, healthcare professionals, and external observers. By incorporating multi-agent decision-making and belief-driven strategies, the model aims to improve psychosis prevention and patient stability. Methods: The proposed POSG model represents the patient’s mental state as healthy, at risk, or psychotic. Stakeholders select actions such as therapy, medication, or passive monitoring based on noisy observations. Bayesian inference updates belief states, ensuring informed decision-making. A reward system quantifies intervention effectiveness, while a Nash equilibrium ensures optimal strategic interactions. Reinforcement learning (RL) refines policies over time, enabling adaptive decision-making. The model integrates dynamic feedback loops, allowing stakeholders to adjust strategies based on observed patient responses, fostering personalized and effective interventions. Results: Simulation results indicate that multi-agent interactions and belief-based strategies enhance patient stability. By dynamically refining strategies, stakeholders reduce unnecessary interventions while improving recovery outcomes. The model demonstrates that coordinated decision-making among multiple stakeholders leads to more stable and predictable intervention success. Conclusions: The POSG framework offers a structured and adaptive approach to psychosis prevention, optimizing intervention effectiveness through multi-agent coordination. The results suggest that AI-driven mental health monitoring can pave the way for more personalized and proactive interventions, reducing the burden of psychosis on individuals and healthcare systems. This study highlights the potential of game-theoretic models in mental health applications, contributing to the development of data-driven intervention techniques.

Understanding the impact of population heterogeneity on the spread of vaccine-preventable diseases is crucial for containment and control. Here, we develop an experimental game model to examine how risks from disease and vaccination shape vaccination decisions in a population with heterogeneous vulnerability. We provide a full theoretical analysis of the game including a formula for Nash equilibrium. We hypothesize that overall vaccination rates will be higher for highly transmissible diseases and amongst the most vulnerable individuals. We then perform a series of experiments to confirm that our theoretical predictions and hypotheses agree with experimental results. Our results show that participants vaccinate strategically inline with Nash equilibrium. Specifically, vaccination rates were higher among more vulnerable individuals than among those less vulnerable. Additionally, we observed minimax behavior in a subset of individuals who consistently chose the secure option (vaccination) regardless of others’ actions. These findings underscore the epidemiological interdependence of vaccination decisions and the need for public health approaches that recognize the different risks and costs faced by vulnerable groups.

Vaccination game theory typically assumes homogeneous populations. In this paper, we develop and solve a vaccination game for an infinite population of agents with non-homogeneous preferences. We assume that agents may vary in how susceptible they are to the disease and how they perceive the cost of the disease and cost of the vaccination, and they may also vary in how they perceive vaccine effectiveness. We encode this heterogeneity by a quantile function describing the distribution of the net relative vaccination cost and give an explicit formula for the Nash equilibrium of the game. We show how this theory can be applied to real-world data and we compare our method to its homogeneous counterpart. We observe that overall, our method outperforms the homogeneous method; the difference in the performance is especially striking when only a small number of survey results are available. Overall, our method is also more robust and provides smaller prediction errors.

Behavioral deviations from subjective expected utility theory, most famously captured by the Allais paradox (Allais, 1953) and the Ellsberg paradox (Ellsberg, 1961), have inspired extensive theoretical and experimental research into risk and ambiguity preferences. While the existing studies (e.g., Gilboa and Schmeidler, 1989; Machina, 2009) analyze these paradoxes independently, little work explores how such heterogeneously biased agents interact in networked strategic environments. Our paper fills this gap by modeling a convergent Markov–network game between Allais-type and Ellsberg-type players, each endowed with fully enriched loss matrices that reflect their distinct probabilistic and ambiguity attitudes.

We define convergent priors as those inducing a spectral radius of < 1 in iterated enriched matrices, ensuring iterative convergence under a matrix-based update rule. Players minimize their losses under these priors in each iteration, converging to an equilibrium where no further updates are feasible. We analyze this convergence under three learning regimes—homophily, heterophily, and type-neutral randomness—each defined via distinct neighborhood learning dynamics. To validate the equilibrium, we construct a risk-neutral measure by transforming losses into payoffs and derive a riskless rate of return representing players' subjective indifference to risk. This applies risk-neutral pricing logic to behavioral matrices, which is novel.

This framework unifies paradox-type decision makers within a networked Markovian environment (stochastic adjacency matrix), extending models of dynamic learning (e.g., Bala and Goyal, 1998; Golub, 2012; Gale and Kariv, 2003) and providing a novel equilibrium characterization for heterogeneous, ambiguity-averse agents in structured interactions.

Keywords: convergent priors; Markov–network game, Allais-type; Ellsberg-type

Introduction: This paper studies human behavioral patterns exhibited in the Iterated Stochastic Prisoner’s Dilemma and trains algorithms that replicate these patterns. We create a training framework that enables learning algorithms to capture human biases and inconsistencies observed during human play. We find that a Reinforcement Learning algorithm (Q-Learner) can emulate how human agents update their beliefs about the future, akin to pessimism and optimism, once the final form of the game is realized.

Methods: We adopt the stochastic setup introduced by Kloosterman (2020), in which human agents play an infinitely repeated game while facing uncertainty regarding the payoff matrix that isused to calculate the reward in each iteration of the dilemma. The defective probability—the probability that the payoff matrix offering lower returns to cooperation is realized—is allowed to vary. To model learning behavior, we implement a randomized, teacherless training framework applied to a Q-Learning algorithm. This approach does not require pre-existing data. Rather, the Q-Learner learns to play via cues from its environment. It is trained on over 230 Prisoner Dilemma strategies.

Results: We find that at the beginning of the game, the trained Q-Learner exhibits a cooperative majority which persists at most values of the defective probability. Flips from a cooperative to a defective majority tend to persist at higher values of the defective probability, although cooperative majorities remain the prevailing outcome. The range of defective probabilities at which the Q-learner flips to a defective majority aligns with the Grim Trigger probability at which humans exhibit the same switch in behavior in Kloosterman's experiments.

Conclusion: The value of defective probability leads players to make optimistic or pessimistic assumptions about the future, impacting their current behavior. Under high learning rates, the Q-Learner can exhibit the same behavior after following randomized training.

Background and Objectives
Traditional evolutionary game theory has relied on bounded rationality. While this approach has led to important insights, it has often neglected the fact that individuals’ decisions are shaped by their perception. We propose a perceptual rationality framework in which the agents are rational and play the Nash equilibrium of their game, however, based on their evolvable perception of the game payoffs and possibly on socially acquired information. We apply this to public goods games to explore how evolving perceptions and social learning drive the emergence of cooperation, diversity, and consistent personality traits.

Methods
In N‑player public goods games (with the group size g), each agent evolves personal cost and benefit perceptions and plays the Nash equilibrium of the game. Equally, an evolvable social trait weights the private and group-averaged perception. We perform evolutionary simulations in both well-mixed and lattice-structured populations and use replicator–mutator dynamics and an evolutionary stability analysis to gain theoretical insights.

Results
Without social learning, cooperation requires the enhancement factor r>g. Introducing social susceptibility fosters the cooperation at a lower r, especially for networks. Even simple games yield power-law distributions of the perceived benefit-to-cost ratios (exponent ≈–2), and social learning amplifies the perceptual and social diversity. Agents with prosocial private perceptions become more social, while antisocial perceptions correlate with individualism—showing consistency in emergent personalities. No pure evolutionarily stable strategy exists in the binary model; instead, polymorphic cyclic dynamics emerge via complementary roles (conditional cooperators, defectors, randomizers). High perceptual/social diversity can hinder the cooperation under strong dilemmas but enhance it when the dilemmas weaken; models with a single evolvable perceptual trait achieve higher cooperation. Stochastic multiplicative processes reproduce macro-ecological laws (species lifespans ~–2, abundance ~–2.5, Taylor’s exponent ~1, 1/f spectral scaling), and evolving sociality shifts these exponents, linking the social structure to biodiversity patterns.

Conclusions
Perceptual rationality unifies rational decision-making with evolvable perception and social learning, explaining cooperation, power-law diversity, consistent personalities, and macroecological patterns in a single framework.

We study a population of players who are coupled up to play a repeated prisoner's dilemma, with the possibility of making errors in the implementation of their strategies. The partnerships may be broken due to external factors (exogenous separation) or because one of the partners decides to leave (endogenous separation). Players who separate and go solo are randomly matched with another single player and keep on playing the game with their new partner.

Strategies in this setting map the histories of play within the partnership to one of the following actions: Cooperate, Defect or leave. We focus on memory-1 strategies, which are strategies in which players condition their actions on the outcome of their last interaction with their partner. Some examples are Tit-For-Tat, Win–Stay–Lose–Shift, Grim and AllD. With the option to leave, there are 162 memory-1 strategies, since one must decide on their initial action in a new partnership (Cooperate or Defect) and what to do (Cooperate, Defect or leave) after each of the four possible outcomes in the game. Consequently, there are 2 x 3⁴ = 162 strategies.

We consider an evolutionary setting where players occasionally revise their strategies. When revising them, players preferentially switch to strategies that are performing well, and they may also try new ones (experimentation).

We study the levels of cooperation and the success of each strategy, and we compare models with and without the option to leave. With low probabilities of an exogenous breakup, cooperation is significantly greater and more robust with the option to leave than without it. Furthermore, classical strategies that support cooperation in the repeated prisoner's dilemma without the option to leave do not fare well when players are allowed to leave their partner: with moderate probabilities of error and experimentation, classical strategies—which do not include leaving—are basically wiped out in the population.

Game theory is widely present in various fields of life and can be divided into complete and incomplete information games. Incomplete information games are difficult to solve due to information uncertainty. Although artificial intelligence has made remarkable achievements in the field of complete information games, its progress in incomplete information games remains limited. Mahjong, as a typical incomplete information game, has interestedmany researchers; however, current Mahjong AI still faces numerous challenges and struggles to effectively handle information uncertainty and complex decision-making problems.

This paper proposes a novel neural network, MJ-Net, which integrates ResNet-CBAM, LSTM, and attention mechanisms to construct an opponent model for Mahjong games. Based on MJ-Net, a residual tile prediction model (for the available tiles in the current field) and a fan-type prediction model (for the possible information on the opponent's hand) are developed. By perceiving changes in the game state and sequences of opponent actions, these models dynamically reassess tile-drawing probabilities and opponents' potential hand information to optimize expected decision values.

Additionally, leveraging domain knowledge and opponent modeling, a defense model is constructed to capture opponents' strategic information, enabling the prediction of their possible actions and fan types. This allows for dynamic adjustment of strategies to reduce risks and increase returns. Finally, a hybrid decision-making framework is established by integrating the game tree with the opponent model and the defense model derived from it. This composite framework optimizes the game tree's search strategy, path evaluation, and pruning, thereby improving the overall decision quality and achieving a balance between offense and defense.

Experimental results demonstrate that MJ-Net performs effectively in hidden information prediction, significantly improving the accuracy of residual tile and fan-type predictions. The decision-making system built upon MJ-Net enhances winning rates and strengthens defensive capabilities in Mahjong games, achieving a balance between defense and offense.

Introduction: Canonical search-and-matching models determine wages by imposing fixed Nash weights that are interpreted as bargaining power and treated as exogeneous. I propose a tractable model in which workers and firms may keep prospecting for alternative matches while they negotiate. The key mechanism is straightforward: a party exits the current match only if its previous offer was rejected and it subsequently meets a new potential partner. I impose this rule in a benchmark set-up and later endogenise it in the full model. This feature, first imposed and later endogenised, generates surplus shares that move with market tightness and, to my knowledge, is the first bargaining model in matching markets that does so.

Methods: The model embeds Rubinstein bargaining in a matching market with tightness θ, defined as the ratio of vacancies over unemployed. Workers find jobs at rate p(θ) (strictly increasing), while firms fill vacancies at rate q(θ) (strictly decreasing). In the benchmark variant the firm proposes first and a proposer does not search within the same period; these restrictions deliver closed-form solutions that make the mechanism transparent. All assumptions are relaxed in the full game.

Results: The ratio of worker surplus over firm surplus is given by

β(θ)= (1−q(θ))/(2−q(θ))​

so the worker’s surplus share grows monotonically in θ. Intuitively, a tighter market raises the probability the worker can answer with a counter-offer before being deserted, increasing the bargaining power. When move order and exit decisions are endogenised, a unique subgame-perfect equilibrium survives and the positive sign of ∂β/∂θ >0 persists under reasonable parameter values.

Conclusions: By linking surplus division directly to market tightness, the protocol results in a tractable, game-theoretic foundation for variable and asymmetric bargaining power inside matching markets and opens new ground for analysing wage dynamics and other two-sided negotiations.