Taxonomy / Convention equilibrium / Giving the agent a norm-Reading module

Giving the agent a norm-Reading module

Rather than let norm-following fall out of ordinary training or ordinary payoffs, each agent is given a dedicated part whose only job is the norm. It works out from watching peers which standard is actually in force — who gets punished, which of several competing institutions the group defers to, what neighbors say in conversation — and feeds that back into how the agent plans and whom it sanctions. Two of these go further and keep the norm stated in words, so the agent can check it and hand it to another agent; the third keeps it in a learned classifier of approved and disapproved behavior, which behaves the same way from outside but cannot be inspected or passed on.

The method, against Convention equilibrium

Scroll the diagram sideways to see all of it.

Concept Analysis: Theoretical Foundations

Each concept is read twice: whether the approach carries it, and whether the approach's own sources claim it. A concept that is absent and was never claimed is a gap in the field rather than a failure of the work, and is marked out of scope.

Convention as Equilibrium

PresentClaimed · delivered

Definition · David Hume (1740); David K. Lewis, Convention (1969)

This concept states that a norm can hold without ever being decreed, as a stable equilibrium of mutual expectation. Each party conforms because the others do, and no party gains by departing from it alone.

Analysis

Equilibrium selection is the stated problem, not an interpretation of it: a coordination game has many equilibria and something has to pick one, and what picks it here is which classification institution the group treats as authoritative. Lewis's convention with the selection mechanism named and built, which is more than the rest of this cell manages.

Conditional Preferences

PresentClaimed · delivered

Definition · Cristina Bicchieri, The Grammar of Society (2006)

This concept states that people follow a norm on two conditions: that they expect others to follow it, and that they believe others expect it of them. The norm therefore holds or shifts with those expectations rather than with anyone's private preference.

Analysis

Bicchieri's structure is what the module implements. An agent conforms because it expects the others to conform and to punish those who do not, and the module's whole function is to align those expectations by settling which institution everyone is reading. Coordinated sanctioning then makes the expectation self-fulfilling, which is the empirical claim the paper tests.

Discovered vs. Designed Order

PartialNot claimed

Definition · F. A. Hayek, Law, Legislation and Liberty (1973)

This claim goes beyond saying that norms can emerge without a designer. It holds that emergent norms are often better, since they absorb local knowledge dispersed across a population that no central rule-writer could collect.

Analysis

The order is designed and only its recognition is discovered. Candidate institutions are placed in the environment by the experimenter; what arises from the agents' own practice is which of them the group defers to. Hayek's distinction has no name for that split, and it is the honest description of what was built: spontaneous recognition of a constructed order.

Concept Analysis: Newly Introduced

Recognition without a rule of recognition

Added

Hart's meta-rule states which rules are valid, and states it in advance. Here the agents work out which institution the group defers to purely from watching each other, with nothing written and nobody declaring it. That is a customary analogue of a rule of recognition arrived at empirically — the function cod|collapse cannot supply, delivered in the cell that is not supposed to have rules at all.

Papers

Normative Modules: A Generative Agent Architecture for Learning Norms that Supports Multi-Agent Cooperation

Atrisha Sarkar et al., May 2024

arXiv:2405.19328MethodBuilt

Equips a generative agent with a normative module that learns, through interaction with peers, which of several candidate institutions a group treats as authoritative — then shows the agent can disregard non-authoritative ones, pick the authoritative one out of several, and reach more stable cooperation than agents without the module.

Emergence of Social Norms in Generative Agent Societies: Principles and Architecture

Siyue Ren et al., Mar 2024

arXiv:2403.08251MethodBuilt

An architecture in four modules that has agents in the Smallville sandbox create norms, hold them in an explicit representation, pass them on through conversation and observation, check them, and act on them in planning.

A learning agent that acquires social norms from public sanctions in decentralized multi-agent settings

Eugene Vinitsky et al., Jun 2021

arXiv:2106.09012MethodBuilt

An agent architecture combining a classifier that sorts observed behavior into approved or disapproved with a motivation to punish in accord with the group, trained in a regime where every agent can see all sanctioning events but learning is otherwise decentralized.