To Embody or Not: The Effect Of Embodiment On User Perception Of LLM-based Conversational Agents

2025-06-03

AI Summary

This study investigates how embodiment (e.g., avatars) affects user perceptions of LLM-driven dialogue agents in non-hierarchical collaborative tasks. Using a within-subjects mixed-methods design—combining quantitative Likert-scale ratings with qualitative interviews—we compare embodied versus disembodied GPT-4–level agents across three dimensions: perceived competence, trustworthiness, and flattery perception. Results reveal that disembodied agents are rated significantly higher in competence; 72% of participants perceived embodied agents as more flattering; and embodiment paradoxically reduced perceived trustworthiness. Critically, this work identifies the first empirical evidence of a “flattery–embodiment interaction effect,” challenging the long-standing assumption that embodiment inherently enhances trust. We propose a novel theoretical mechanism: when coupled with LLMs’ intrinsic tendency toward excessive politeness or flattery, embodiment may trigger adverse cognitive appraisals—undermining credibility rather than enhancing it. These findings provide foundational empirical evidence and theoretical refinement for designing embodied AI systems.

Technology Category

• Intelligent Robots: Embodied AI
• Cognitive Modeling & Cognitive Systems: Social Cognition And Interaction
• Humans and AI: Learning Human Values and Preferences

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: Trust and reliance of crowd workers and data experts on GenAI
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

Embodiment in conversational agents (CAs) refers to the physical or visual representation of these agents, which can significantly influence user perception and interaction. Limited work has been done examining the effect of embodiment on the perception of CAs utilizing modern large language models (LLMs) in non-hierarchical cooperative tasks, a common use case of CAs as more powerful models become widely available for general use. To bridge this research gap, we conducted a mixed-methods within-subjects study on how users perceive LLM-based CAs in cooperative tasks when embodied and non-embodied. The results show that the non-embodied agent received significantly better quantitative appraisals for competence than the embodied agent, and in qualitative feedback, many participants believed that the embodied CA was more sycophantic than the non-embodied CA. Building on prior work on users' perceptions of LLM sycophancy and anthropomorphic features, we theorize that the typically-positive impact of embodiment on perception of CA credibility can become detrimental in the presence of sycophancy. The implication of such a phenomenon is that, contrary to intuition and existing literature, embodiment is not a straightforward way to improve a CA's perceived credibility if there exists a tendency to sycophancy.

Problem

• Effect of embodiment on user perception of LLM-based agents
• Impact of embodiment on CA credibility with sycophancy
• Comparison of embodied vs non-embodied CAs in cooperative tasks

Innovation

• Examined embodiment effect on LLM-based agents
• Used mixed-methods within-subjects study design
• Found non-embodied agents perceived as more competent