Examining Spanish Counseling with MIDAS: a Motivational Interviewing Dataset in Spanish

2025-02-12

AI Summary

This study investigates how linguistic and cultural factors influence counselor behavior in Motivational Interviewing (MI), specifically examining cross-linguistic differences between Spanish and English contexts. Method: We introduce MIDAS—the first manually annotated Spanish-language MI video transcription dataset—featuring fine-grained annotations of counselor reflections and questions. Using MIDAS, we systematically compare monolingual (Spanish and English BERT) versus multilingual BERT models on behavioral coding tasks. Contribution/Results: We demonstrate significant language-specific patterns in counselor utterances; monolingual models substantially outperform multilingual BERT in reflection and question identification (F1 = 0.82). MIDAS establishes the first benchmark for NLP research on non-English MI, addressing a critical gap in cross-lingual psychotherapy analysis. The findings provide empirical support for language-specific modeling in automated MI process evaluation and advance equitable, culturally responsive computational behavioral health tools.

Technology Category

• Natural Language Processing: Language Grounding & Multi-modal NLP
• Machine Learning: Large Multimodal Models (LMMs)
• Cognitive Modeling & Cognitive Systems: Adaptive Behavior

Application Category

• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data
• User Modeling, Personalization and Recommendation: Studies of user behavior, including longitudinal effects of personalized systems

Abstract

Cultural and language factors significantly influence counseling, but Natural Language Processing research has not yet examined whether the findings of conversational analysis for counseling conducted in English apply to other languages. This paper presents a first step towards this direction. We introduce MIDAS (Motivational Interviewing Dataset in Spanish), a counseling dataset created from public video sources that contains expert annotations for counseling reflections and questions. Using this dataset, we explore language-based differences in counselor behavior in English and Spanish and develop classifiers in monolingual and multilingual settings, demonstrating its applications in counselor behavioral coding tasks.

Problem

• Analyze Spanish counseling language
• Develop multilingual counseling classifiers
• Examine cultural counseling differences

Innovation

• MIDAS dataset introduction
• Cross-language counseling analysis
• Multilingual behavior classifiers development