PAVE: A Cognitive Architecture for Legitimate Violation in Generative Agent Societies

TL;DR

This paper introduces PAVE, a four-module cognitive architecture enabling generative agents to reason about and execute legitimate rule violations in cooperative scenarios, enhancing interpretability and plausibility.

Contribution

The paper presents PAVE, a novel architecture with explicit legitimacy assessment and scoped violations, implemented in Voville for evaluating rule-breaking behaviors in LLM-based agents.

Findings

PAVE agents satisfy properties of legitimate violation, authority deference, bounded scope, and recovery.

PAVE agents are more structured, interpretable, and rated more plausible than vanilla models.

Ablation of the legitimacy gate causes vanilla-like failures.

Abstract

Generative agents based on large language models reproduce believable human behavior in cooperative settings, but how they should reason in situations where rule-breaking may be required, such as fire evacuation or authority-supervised emergency, remains poorly characterized. We propose PAVE (Perception, Assessment, Verdict, Emulation), a novel four-module cognitive architecture that addresses this gap end to end: (i) Perception extracts a structured context with explicit authority distance, peer behaviors, and severity-tagged situational cues; (ii) Assessment scores the context along five scalars including an explicit legitimacy judgment that checks necessity, proportionality, and absence of alternatives; (iii) Verdict decides to comply or violate under a hard legitimacy gate, with a per-agent threshold elicited from the persona; (iv) Emulation enacts the verdict and scopes the violation to the rule the trigger justifies. We instantiate PAVE in Voville, a tile-based traffic environment forked from Smallville, and evaluate across three scenarios, four LLM backbones, and a focused ablation. PAVE agents satisfy four properties simultaneously: legitimate violation (only when a trigger justifies it), authority deference (officer instructions override even high legitimacy), bounded scope (violations confined to the targeted rule), and recovery (baseline restored once the trigger ends). PAVE agents make more structured and interpretable decisions than vanilla across all four properties, and human evaluators rate them as more plausible. Ablating the legitimacy gate reproduces vanilla-like failures. We release Voville, the PAVE prompts and code, and the evaluation pipeline.