The “Descriptive Norm and Fraud Dynamics” model demonstrates how fraudulent behavior can either proliferate or be contained within non-hierarchical organizations, such as peer networks, through social influence taking the form of a descriptive norm. This model expands on the fraud triangle theory, which posits that an individual must concurrently possess a financial motive, perceive an opportunity, and hold a pro-fraud attitude to engage in fraudulent activities (red agent). In the absence of any of these elements, the individual will act honestly (green agent).

The model explores variations in a descriptive norm mechanism, ranging from local distorted knowledge to global perfect knowledge. In the case of local distorted knowledge, agents primarily rely on information from their first-degree colleagues. This knowledge is often distorted because agents are slow to update their empirical expectations, which are only partially revised after one-to-one interactions. On the other end of the spectrum, local perfect knowledge is achieved by incorporating a secondary source of information into the agents’ decision-making process. Here, accurate information provided by an observer is used to update empirical expectations.

The model shows that the same variation of the descriptive norm mechanism could lead to varying aggregate fraud levels across different fraud categories. Two empirically measured norm sensitivity distributions associated with different fraud categories can be selected into the model to see the different aggregate outcomes.

AGENTS model is an agent-based computational framework designed to explore the socio-ecological and economic dynamics of agricultural production in the Byzantine Negev Highlands, with a focus on viticulture. It integrates historical, environmental, and social factors to simulate settlement sustainability, crop yields, and the impacts of varying climate conditions. The model is built in NetLogo and incorporates GIS-based topographical and hydrological data. Key features include the ability to assess climate impacts on crop profitability and settlement strategies, evaluate economic outputs of ancient vineyards, and simulate agent decision-making processes under diverse scenarios.

The AGENTS model is highly flexible, enabling users to simulate agricultural regimes with any two crops: one cash crop (a crop grown for profit, e.g., grapevines) and one staple crop (a crop grown for subsistence, e.g., wheat). While the default setup models viticulture and wheat cultivation in the Byzantine Negev Highlands, users can adapt the model to different environmental and socio-ecological contexts worldwide—both past and present.

Users can load external files to customize precipitation, evaporation, topography, and labor costs (measured as man-days per 0.1ha, converted to kg of wheat per model patch size area), and can also edit key parameters related to yield calculations. This includes modifying crop-specific yield formulas, soil and runoff indices, and any factors influencing crop performance. The model inherently simulates cash crops grown in floodplain regions and staple crops cultivated along riverbanks, providing a powerful tool to investigate societal resilience and responses to climate stressors across diverse environments.

…

MASTOC-LLM extends the classic Multi-Agent System Tragedy of the Commons (MASTOC) model by replacing hard-coded behavioral rules with autonomous decision-making powered by large language models (LLMs). Three heterogeneous agents manage herds of cows on a shared grassland commons. Each tick, an agent receives a structured prompt describing current resource levels, its own herd size, peer behavior, and — optionally — a rolling memory of recent rounds and messages from neighboring agents. The LLM returns a stocking decision (add, remove, or hold cows) together with a natural-language rationale and, when communication is enabled, a short message to broadcast to peers.

The model is designed to test whether LLM agents spontaneously develop Ostrom-style common-pool resource governance (mutual monitoring, graduated sanctions, graduated rule revision) or instead fall into identifiable failure modes. Preliminary experiments with Claude Haiku 4.5, GPT-5.4-mini, and DeepSeek R1:32b have revealed four recurring collapse patterns — Cooperative Paralysis, Defection Cascade, Overshoot-Panic, and Hybrid Architecture Failure — whose onset timing is sensitive to memory length, inter-agent communication, and the post-training alignment approach of the underlying model.

MASTOC-LLM is intended as a laboratory for generative agent-based modelling (GABM) methodology: it provides a clean, well-understood commons baseline against which LLM behavioral hypotheses can be systematically tested and compared across models, parameter sweeps, and alignment regimes.

The various technologies used inside a Dutch greenhouse interact in combination with an external climate, resulting in an emergent internal climate, which contributes to the final productivity of the greenhouse. This model examines how differing technology development styles affects the overall ability of a community of growers to approach the theoretical maximum yield.

This is a NetLogo 7.0.1 agent-based model of interdependent team productivity under O-ring production logic. The model asks when collective problem-solving capacity (CPS) improves team output and when its effect is constrained by trust, burnout, specialization diversity, weak-link quality, and team-formation rules. Agents are heterogeneous in skill, CPS, trust, burnout, effort, learning rate, openness, specialization, adaptability, and aspiration. Each tick forms temporary teams, computes individual contribution quality, combines contributions through a geometric O-ring production function with an explicit weakest-link term, records output and success, and updates agents through feedback, learning, trust change, burnout, recovery, and turnover. The paper follows the ODD protocol, reports exact implementation equations, and analyzes six BehaviorSpace experiments with 50 stochastic repetitions per condition. Scenario results show that CPS is beneficial but not sufficient: productivity is highest when CPS is combined with trust, low burnout, effective coordination, and a reliable weakest-link floor. Diversity yields only modest gains at low CPS but larger gains when CPS is high. Formation-mode results are especially informative because mixed CPS-skill formation maximizes total output, whereas random formation has the highest binary success rate, showing that output magnitude and threshold success can diverge. The study contributes a transparent computational mechanism linking collective intelligence with O-ring production and identifies boundary conditions under which high-CPS teams can still be limited by weak links and interdependence.