This NetLogo model is an implementation of the mostly verbal (and graphic) model in Jarret Walker’s Human Transit: How Clearer Thinking about Public Transit Can Enrich Our Communities and Our Lives (2011). Walker’s discussion is in the chapter “Connections or Complexity?”. See especially figure 12-2, which is on page 151.

In “Connections or Complexity?”, Walker frames the matter as involving a choice between two conflicting goals. The first goal is to minimize connections, the need to make transfers, in a transit system. People naturally prefer direct routes. The second goal is to minimize complexity. Why? Well, read the chapter, but as a general proposition we want to avoid unnecessary complexity with its attendant operating characteristics (confusing route plans in the case of transit) and management and maintenance challenges. With complexity general comes degraded robustness and resilience.

How do we, how can we, choose between these conflicting goals? The grand suggestion here is that we only choose indirectly, implicitly. In the present example of connections versus complexity we model various alternatives and compare them on measures of performance (MoP) other than complexity or connections per se. The suggestion is that connections and complexity are indicators of, heuristics for, other MoPs that are more fundamental, such as cost, robustness, energy use, etc., and it is these that we at bottom care most about. (Alternatively, and not inconsistently, we can view connections and complexity as two of many MoPs, with the larger issue to be resolve in light of many MoPs, including but not limited to complexity and connections.) We employ modeling to get a handle on these MoPs. Typically, there will be several, taking us thus to a multiple criteria decision making (MCDM) situation. That’s the big picture.

Our aim is to show effects of group living when only low-level cognition is assumed, such as pattern recognition needed for normal functioning, without assuming individuals have knowledge about others around them or warn them actively.
The model is of a group of vigilant foragers staying within a patch, under attack by a predator. The foragers use attentional scanning for predator detection, and flee after detection. This fleeing action constitutes a visual cue to danger, and can be received non-attentionally by others if it occurs within their limited visual field. The focus of this model is on the effectiveness of this non-attentional visual information reception.
A blind angle obstructing cue reception caused by behaviour can exist in front, morphology causes a blind angle in the back. These limitations are represented by two visual field shapes. The scan for predators is all-around, with distance-dependent detection; reception of flight cues is limited by visual field shape.
Initial parameters for instance: group sizes, movement, vision characteristics for predator detection and for cue reception. Captures (failure), number of times the information reached all individuals at the same time (All-fled, success), and several other effects of the visual settings are recorded.

MERCURY aims to represent and explore two descriptive models of the functioning of the Roman trade system that aim to explain the observed strong differences in the wideness of distributions of Roman tableware.

What is stable: the large but coordinated change during a diffusion or the small but constant and uncoordinated changes during a dynamic equilibrium? This agent-based model of a diffusion creates output that reveal insights for system stability.

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.

A hybrid predator-prey model of fish and plankton that switches dynamically between ABM and SD representations. It contains 6 related structural designs of the same model.

This model simulates diffusion curves and it allows to test how social influence, network structure and consumer heterogeneity affect their spreads and their speeds.

HousingABM_Japan is a NetLogo agent-based model of the residential market of Tokyo’s 23 wards. It evaluates whether a single parameter configuration can jointly reproduce key features of prices, rents, yields, and market turnover across distinct market regimes from 2001 to 2025, with particular attention to demand- and supply-side trend-following during the 2021–2025 price surge.

The model builds on the Bank of England housing-market lineage (Baptista et al. 2016; Carro et al. 2023) and introduces four extensions: (1) dynamic linkages between the sale and rental markets through vacancy, rents, and yields; (2) heterogeneous demand-side trend-following; (3) supply-side trend-following through construction-cost trend anchoring and a momentum-dependent dynamic premium; and (4) housing-equity borrowing that converts unrealized equity into additional borrowing capacity.

Twenty parameters are calibrated using 2001–2015 data and held fixed for post-calibration evaluation over 2016–2020 and 2021–2025, while annual exogenous inputs follow observed historical paths. The model reproduces the shift from moderate price growth to the 2021–2025 surge, as well as rent acceleration, surge-period yield compression, and persistently low market turnover, although it understates the intermediate acceleration of 2016–2020.

…

Negotiation Lab 1.0 is an agent-based model of peace negotiations that explores how the parties’ readiness — their motivation and optimism to engage in talks — evolves dynamically throughout the negotiation process. The model reconceptualizes readiness as an adaptive state variable that is continuously updated through feedback from negotiation outcomes, rather than a static precondition assessed at the onset of talks.
The model simulates two parties negotiating a multi-issue agenda. In each round, parties allocate effort to the current sub-issue; outcomes depend on their joint effort and a stochastic component representing external factors. Results feed back into each party’s readiness, shaping subsequent engagement. The negotiation ends either when all agenda items are resolved (agreement) or when a party’s readiness falls below a critical threshold (breakdown).
Key parameters include the initial readiness of each party, agenda structure (balanced, hard, easy, red, or random), type of negotiation (from highly cooperative to highly competitive), and each party’s effort strategy (always high, always low, random, or pseudo tit-for-tat). The model shows that while initial readiness is associated with negotiation outcomes, it is neither necessary nor sufficient to determine them: process variables — the type of interaction, agenda design, and adaptive effort strategies — exert comparatively larger effects on outcomes. Identical initial conditions can produce widely divergent trajectories, illustrating path dependence and sensitivity to feedback dynamics.
The model is implemented in NetLogo 7.0 and is documented using the ODD+D protocol. It is associated with the paper “Beyond Initial Conditions: How Adaptive Readiness Shapes Peace Negotiation Outcomes” (Arévalo, under review).

Several taxonomies for empirical validation have been published. Our model integrates different methods to calibrate an innovation diffusion model, ranging from simple randomized input validation to complex calibration with the use of microdata.