Computational Model Library

Activation and Foreclosure: An Agent-Based Model of Enforcement and Exit in Codified Moral Communities (1.0.0)

An agent-based model of rule-governed (religious) communities in which networked agents observe one another’s conduct, sanction deviation,
and decide whether to remain. The model separates two structural conditions for durable capture. First, whether an enforcement apparatus
activates: agents punish deviation only where conduct is observable (the code-geometry parameter sigma) and where sanctioning is
institutionally rewarded (pi), a joint threshold in which neither condition alone suffices. Second, whether members’ exit is foreclosed: an
outside-option degradation (delta) that can be imposed exogenously or allowed to drift with enforcement activity. Punishment concentrates in
a small, endogenously recomputed enforcer cadre, but that concentration is produced by an institutional privilege bundle (a punishment
monopoly above an authority threshold, cost and backlash discounts, capital accumulation, and patronage) rather than by fixed differences
among agents. Implemented in Python with Mesa on a Barabasi-Albert network; supports parameter sweeps over sigma, pi, and delta, privilege
ablations, and a threat-shock schedule. Each run writes a step-level metrics file and an agent-level summary. The model is documented in
ODD-protocol order and its calibrated parameter set is frozen for bit-for-bit reproduction.

Release Notes

Initial public release accompanying the JASSS submission. Frozen v2.7 model
(all headline results) and the v3.3 endogenous-exit-capacity extension, with the
calibrated parameter set (data/BEST_PARAMS_frozen.json), reproduction scripts,
an ODD model description, and the four committed headline sweep summaries. The
full raw sweep tree is in the version-control repository.

Associated Publications

Activation and Foreclosure: An Agent-Based Model of Enforcement and Exit in Codified Moral Communities 1.0.0

An agent-based model of rule-governed (religious) communities in which networked agents observe one another’s conduct, sanction deviation,
and decide whether to remain. The model separates two structural conditions for durable capture. First, whether an enforcement apparatus
activates: agents punish deviation only where conduct is observable (the code-geometry parameter sigma) and where sanctioning is
institutionally rewarded (pi), a joint threshold in which neither condition alone suffices. Second, whether members’ exit is foreclosed: an
outside-option degradation (delta) that can be imposed exogenously or allowed to drift with enforcement activity. Punishment concentrates in
a small, endogenously recomputed enforcer cadre, but that concentration is produced by an institutional privilege bundle (a punishment
monopoly above an authority threshold, cost and backlash discounts, capital accumulation, and patronage) rather than by fixed differences
among agents. Implemented in Python with Mesa on a Barabasi-Albert network; supports parameter sweeps over sigma, pi, and delta, privilege
ablations, and a threat-shock schedule. Each run writes a step-level metrics file and an agent-level summary. The model is documented in
ODD-protocol order and its calibrated parameter set is frozen for bit-for-bit reproduction.

Release Notes

Initial public release accompanying the JASSS submission. Frozen v2.7 model
(all headline results) and the v3.3 endogenous-exit-capacity extension, with the
calibrated parameter set (data/BEST_PARAMS_frozen.json), reproduction scripts,
an ODD model description, and the four committed headline sweep summaries. The
full raw sweep tree is in the version-control repository.

Version Submitter First published Last modified Status
1.0.0 Kiran Boggavarapu Wed Aug 19 20:02:23 2026 Wed Aug 19 20:02:25 2026 Published

Discussion

This website uses cookies and Google Analytics to help us track user engagement and improve our site. If you'd like to know more information about what data we collect and why, please see our data privacy policy. If you continue to use this site, you consent to our use of cookies.
Accept