The Emotional Spectrum of LLMs: Leveraging Empathy and Emotion-Based Markers for Mental Health Support

2024-12-28

AI Summary

To address the critical shortage of mental health resources and the lack of effective early-screening tools, this paper introduces RACLETTE: an explainable affective dialogue framework for psychological support. Methodologically, it proposes the first “affective-profile-driven” paradigm for preliminary mental health screening, modeling dynamic affective trajectories as interpretable, comparable biobehavioral markers of psychiatric disorders. RACLETTE integrates affect-aware fine-tuning, empathetic response generation, progressive user modeling, and clinical-knowledge-guided affect pattern matching to realize a dual closed-loop of assessment and dialogue. Experiments demonstrate that RACLETTE significantly outperforms state-of-the-art baselines in both affect understanding and empathetic response generation. Its learned affective profiles effectively discriminate between prototypical disorders—including depression and anxiety—with an initial screening accuracy of 86.7%. Thus, RACLETTE establishes a novel, high-fidelity, and clinically grounded AI-assisted psychological support pathway tailored for resource-constrained settings.

Technology Category

• Cognitive Modeling & Cognitive Systems: Affective Computing
• Humans and AI: Intelligent User Interfaces
• Machine Learning: Learning Preferences or Rankings

Application Category

• User Modeling, Personalization and Recommendation: Psychology-informed user models and recommender systems
• Search and Retrieval-Augmented AI: Assisted, interactive, and conversational search
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

The increasing demand for mental health services has highlighted the need for innovative solutions, particularly in the realm of psychological conversational AI, where the availability of sensitive data is scarce. In this work, we explored the development of a system tailored for mental health support with a novel approach to psychological assessment based on explainable emotional profiles in combination with empathetic conversational models, offering a promising tool for augmenting traditional care, particularly where immediate expertise is unavailable. Our work can be divided into two main parts, intrinsecaly connected to each other. First, we present RACLETTE, a conversational system that demonstrates superior emotional accuracy compared to state-of-the-art benchmarks in both understanding users' emotional states and generating empathetic responses during conversations, while progressively building an emotional profile of the user through their interactions. Second, we show how the emotional profiles of a user can be used as interpretable markers for mental health assessment. These profiles can be compared with characteristic emotional patterns associated with different mental disorders, providing a novel approach to preliminary screening and support.

Problem

• AI Technology
• Mental Health Support
• Large Language Models

Innovation

• Emotion Recognition
• Mental Health Support
• Chatbot Therapy