What Do You Mean? Exploring How Humans and AI Interact with Symbols and Meanings in Their Interactions

2025-10-06

AI Summary

Human-AI interaction often suffers from semantic misalignment, as humans dynamically co-construct symbolic meaning through social interaction, whereas AI typically treats symbols as static, compressed representations, neglecting their real-time, context-sensitive emergence in dialogue. Method: Grounded in symbolic interactionism, this study develops an analytical framework for human-AI meaning co-construction and conducts a qualitative experiment with N=37 participants. Contribution/Results: When AI introduced semantically conflicting symbols, 63% of participants actively revised their interpretations—demonstrating that meaning conflict triggers reflexive reconstruction and facilitates shared understanding. The study establishes that shared meaning arises not from pre-established consensus but from the continuous exchange and redefinition of symbols within interaction. It further proposes the first empirically grounded, symbolic interactionist paradigm for human-AI collaborative meaning-making, offering a novel pathway toward robust human-AI semantic alignment.

Technology Category

• Humans and AI: Interaction Techniques and Devices
• Cognitive Modeling & Cognitive Systems: Social Cognition And Interaction
• Intelligent Robots: Human-Robot Interaction

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Economics, Online Markets and Human Computation: Research challenges in human and human-AI computation
• Search and Retrieval-Augmented AI: Assisted, interactive, and conversational search

Abstract

Meaningful human-AI collaboration requires more than processing language, it demands a better understanding of symbols and their constructed meanings. While humans naturally interpret symbols through social interaction, AI systems treat them as patterns with compressed meanings, missing the dynamic meanings that emerge through conversation. Drawing on symbolic interactionism theory, we conducted two studies (N=37) investigated how humans and AI interact with symbols and co-construct their meanings. When AI introduced conflicting meanings and symbols in social contexts, 63% of participants reshaped their definitions. This suggests that conflicts in symbols and meanings prompt reflection and redefinition, allowing both participants and AI to have a better shared understanding of meanings and symbols. This work reveals that shared understanding emerges not from agreement but from the reciprocal exchange and reinterpretation of symbols, suggesting new paradigms for human-AI interaction design.

Problem

• Exploring human-AI symbolic interaction gaps in meaning construction
• Investigating how conflicting symbols trigger meaning redefinition processes
• Developing shared understanding through reciprocal symbol reinterpretation

Innovation

• AI introduces conflicting symbols to prompt reflection
• Humans reshape definitions through reciprocal symbol exchange
• Shared understanding emerges from reinterpretation not agreement