Emergent Communication for Co-constructed Emotion Between Embodied Agents via Collective Predictive Coding

2026-05-10

AI Summary

This study investigates how embodied agents co-construct emotions through social interaction, with a focus on the emergence of shared emotional categories. Building upon a collective predictive coding framework, the work introduces a Metropolis-Hastings naming game to model emergent communication between dyadic agents, integrating visual, auditory, and simulated interoceptive inputs to enable affective co-construction. It extends collective predictive coding into the domain of social emotion for the first time, demonstrating that communication robustly aligns emotional categories even when agents exhibit systematic differences in interoceptive mechanisms. Notably, this alignment occurs primarily at the symbolic level rather than in perceptual latent representations. The findings reveal that this mechanism significantly enhances the clarity, consistency, and intersubjective alignment of emotional categories across agents, exhibiting category-specific patterns of representational reshaping.

Abstract

According to the theory of constructed emotion, the brain actively forms emotion categories by integrating multimodal bodily signals, and constructs emotional experiences by using these categories to predict and interpret sensory inputs. While research has advanced in modeling individual emotion construction, the social process of co-construction-how a shared understanding of emotions emerges between individuals-remains computationally underexplored. This study investigates this process by modeling emergent communication between two embodied agents using the Metropolis-Hastings Naming Game (MHNG), grounded in the Collective Predictive Coding (CPC) framework. Our experiments, using visual, auditory, and simulated interoceptive inputs, yield two main findings. First, MHNG-based communication significantly improves the alignment, clarity, and inter-agent agreement of the learned emotion categories compared to non-communicative and non-selective baselines, with the alignment effect concentrated at the symbolic layer rather than the perceptual latent representation. Second, even when the two agents have systematically divergent interoceptive dynamics, communication still produces robust categorical alignment, with distinct, category-specific reshaping patterns of each agent's emotion categories-consistent with the constructed-emotion view that interoceptive heterogeneity is constitutive of, rather than an obstacle to, shared emotional meaning. These findings provide computational support for the co-constructionist view of emotion and extend the CPC framework from physical to socially-grounded domains.

Problem

• co-constructed emotion
• emergent communication
• embodied agents
• Collective Predictive Coding
• emotion categories

Innovation

• Collective Predictive Coding
• Emergent Communication
• Constructed Emotion
• Metropolis-Hastings Naming Game
• Interoceptive Heterogeneity