BDPD⁰ — A Computational Laboratory for Generative Agents, Commons Dilemmas and More

Read full text
Authors: Roberto Brunelli
Cited by: 14

Abstract

We present BDPD (Be Different Play Differential), an open-source, dual-mode computational laboratory comprising a configurable multi-agent simulation platform and a card game, The Forest of Humbaba, for the integrated study of common-pool resource dilemmas. Through systematic parameter sweeps we demonstrate four principal findings: a discontinuous fragility threshold (a single aggressive agent suffices to doom the commons under realistic slow-regeneration), the reactive < conservative effect (locally reactive strategies are collectively inferior to unconditionally conservative ones), noise as a selective blunter of aggressive extraction, and the tragedy of the compensator (unilateral conservation subsidising defection rather than preventing it). Preliminary single-model LLM case studies corroborate the safe-betrayal and awareness-without-restraint patterns.

Fields of Study & Subjects

Artificial Intelligence in Games Evolutionary Game Theory and Cooperation Reinforcement Learning in Robotics

Publication details

Journal / source
Zenodo (CERN European Organization for Nuclear Research)
Publication date
13 June 2026
Language
EN
Document type
Preprint