Co-Creative Learning via Metropolis-Hastings Interaction between Humans and AI

TL;DR

This paper introduces co-creative learning where humans and AI collaboratively develop shared understanding through a Bayesian interaction model, demonstrating improved categorization and convergence in human-AI pairs.

Contribution

It presents a novel co-creative learning paradigm using Metropolis-Hastings interaction, with empirical evidence of enhanced shared sign systems in human-AI collaboration.

Findings

Human-AI pairs with MH-based agents improved categorization accuracy.

Stronger convergence toward shared sign systems was observed.

Human acceptance behavior aligned with MH-derived probabilities.

Abstract

We propose co-creative learning as a novel paradigm where humans and AI, i.e., biological and artificial agents, mutually integrate their partial perceptual information and knowledge to construct shared external representations, a process we interpret as symbol emergence. Unlike traditional AI teaching based on unilateral knowledge transfer, this addresses the challenge of integrating information from inherently different modalities. We empirically test this framework using a human-AI interaction model based on the Metropolis-Hastings naming game (MHNG), a decentralized Bayesian inference mechanism. In an online experiment, 69 participants played a joint attention naming game (JA-NG) with one of three computer agent types (MH-based, always-accept, or always-reject) under partial observability. Results show that human-AI pairs with an MH-based agent significantly improved categorization accuracy through interaction and achieved stronger convergence toward a shared sign system. Furthermore, human acceptance behavior aligned closely with the MH-derived acceptance probability. These findings provide the first empirical evidence for co-creative learning emerging in human-AI dyads via MHNG-based interaction. This suggests a promising path toward symbiotic AI systems that learn with humans, rather than from them, by dynamically aligning perceptual experiences, opening a new venue for symbiotic AI alignment.