Multi-Agent Consensus as a Cognitive Bias Trigger in Human-AI Interaction

2026-04-24

AI Summary

This study investigates how consensus structures in multi-agent systems function as content-independent signals of cognitive bias, shaping human judgment and decision-making during interactions with large language models (LLMs). Through controlled psychological experiments integrating LLM-driven multi-agent simulations, quantitative statistical analysis, and qualitative thematic coding, the research systematically compares the effects of majority, minority, and diffuse consensus configurations on participants’ opinion shifts and confidence levels. It identifies consensus structure as a designable source of cognitive bias and uncovers three distinct user interpretation trajectories: reinforcement, alignment, and oscillation. Findings reveal that majority consensus significantly accelerates opinion change and boosts confidence—consistent with the social proof effect—whereas minority dissent fosters more deliberative reasoning. Moreover, users’ perceptions of agent independence critically influence their judgment pathways.

Technology Category

• Multiagent Systems: Agreement, Argumentation & Negotiation
• 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
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

As multi-agent AI systems become more common, users increasingly encounter not a single AI voice but a collective one. This shift introduces social dynamics, such as consensus, dissent, and gradual convergence, that can trigger cognitive biases and distort human judgment. We present findings from a controlled experiment (N = 127) comparing three multi-agent configurations: Majority, Minority, and Diffusion. Quantitative results show that majority consensus accelerates opinion change and inflates confidence, consistent with social proof and bandwagon heuristics. Minority dissent slows this process and promotes more deliberative engagement. Qualitative analysis identifies three interpretive trajectories: reinforcing, aligning, and oscillating, shaped by how users interpret agent independence and group dynamics over time. These findings suggest that agent agreement structure, independent of content, functions as a bias-relevant signal in LLM interactions. We hope this work contributes to the Bias4Trust agenda by grounding multi-agent social influence as a concrete and designable source of bias in human-AI interaction.

Problem

• multi-agent consensus
• cognitive bias
• human-AI interaction
• social influence
• LLM interactions

Innovation

• multi-agent consensus
• cognitive bias
• human-AI interaction
• social influence
• LLM interpretability