Emotional Cognitive Modeling Framework with Desire-Driven Objective Optimization for LLM-empowered Agent in Social Simulation

2025-10-15

AI Summary

Current LLM-based agents exhibit dual deficiencies in affective cognition: (1) an inability to model bounded rationality, and (2) a lack of empirically grounded mechanisms for integrating affect with decision-making. To address these gaps, we propose a desire-driven affective cognitive framework that unifies affective state evolution, desire generation, goal optimization, and cognitive transfer—thereby enabling dynamic coupling between emotion and decision processes. The framework embeds empirically validated affect models within a multi-agent social simulation environment, enhancing affective consistency and behavioral authenticity of virtual agents. Experimental results demonstrate significantly higher ecological validity and superior human behavioral fit compared to baseline approaches. Notably, this work constitutes the first implementation of a psychologically grounded, affect-driven decision architecture in LLM agents—establishing a principled foundation for affectively intelligent agent design.

Technology Category

• Cognitive Modeling & Cognitive Systems: Affective Computing
• Multiagent Systems: Agent-Based Simulation and Emergent Behavior
• 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
• Search and Retrieval-Augmented AI: Agentic search
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

The advent of large language models (LLMs) has enabled agents to represent virtual humans in societal simulations, facilitating diverse interactions within complex social systems. However, existing LLM-based agents exhibit severe limitations in affective cognition: They fail to simulate the bounded rationality essential for bridging virtual and real-world services; They lack empirically validated integration mechanisms embedding emotions within agent decision architectures. This paper constructs an emotional cognition framework incorporating desire generation and objective management, designed to achieve emotion alignment between LLM-based agents and humans, modeling the complete decision-making process of LLM-based agents, encompassing state evolution, desire generation, objective optimization, decision generation, and action execution. This study implements the proposed framework within our proprietary multi-agent interaction environment. Experimental results demonstrate that agents governed by our framework not only exhibit behaviors congruent with their emotional states but also, in comparative assessments against other agent types, demonstrate superior ecological validity and generate decision outcomes that significantly more closely approximate human behavioral patterns.

Problem

• Modeling bounded rationality in LLM agents
• Integrating emotions into agent decision architectures
• Achieving emotion alignment between agents and humans

Innovation

• Emotional cognition framework with desire generation
• Objective optimization for human-like decision modeling
• Multi-agent environment validation of ecological validity