Deal Me Maybe: The Role of Emotions in Multi-Agent Negotiation

2026-08-07

AI Summary

This study addresses the widespread neglect of emotional factors in large language model (LLM)-based multi-agent negotiation, despite emotion being a critical driver of human bargaining behavior. The authors propose a controlled multi-agent negotiation framework that modulates six emotional states of buyer and seller agents via prompt engineering, systematically evaluating emotional effects across 350 real-world items and two budget conditions. Their experiments—spanning five prominent LLMs and 36 emotion pairings—reveal, for the first time, that emotions exert significant and role-dependent influences on negotiation outcomes: angry buyers rarely reach agreements, happy buyers achieve high success rates but pay substantial premiums, and fearful buyers secure more favorable prices. Emotion is shown to significantly affect agreement rates, concession dynamics, termination behaviors, and price trajectories, highlighting potential risks of deploying emotion-conditioned agents in commercial applications.

Abstract

Negotiation is a demanding social task for LLM agents, requiring strategic reasoning, persuasion, and interpersonal adaptation. Yet existing benchmarks often treat agents as emotionally neutral, overlooking a key driver of human bargaining behavior. We study how prompt-conditioned emotions affect LLM-based price negotiation. In a controlled framework, buyer and seller agents are independently assigned one of six emotional states and negotiate over 350 real consumer products under two budget conditions. Across 36 emotion-pair settings and five widely used LLMs, we find that emotions strongly shape outcomes. Angry buyers almost never reach agreement (0.39% deal rate), while happy buyers agree most often (28.91%), but obtain worse prices than fearful buyers. Emotion effects are role-dependent: buyer emotion mainly drives acceptance and rejection, whereas seller emotion shapes concession dynamics. These effects influence not only language, but also termination behavior and price trajectories, raising concerns for emotion-conditioned agents in commerce.

Problem

• emotion
• multi-agent negotiation
• LLM agents
• bargaining behavior
• price negotiation

Innovation

• emotion-conditioned agents
• multi-agent negotiation
• large language models
• role-dependent emotion effects
• controlled negotiation framework