An Exploration of Internal States in Collaborative Problem Solving

2025-07-02

AI Summary

This study investigates the dynamic evolution of individual emotional states during collaborative problem solving (CPS). Employing a mixed-methods approach, it integrates video-cued retrospective self-reports, concurrent think-aloud protocols, and natural language processing to conduct multidimensional sentiment analysis on post-task linguistic data—encompassing emotion lexicon identification, extraction of unigram/bigram affective features, and semantic similarity modeling. The work systematically uncovers language-based patterns underlying emotional fluctuations in CPS for the first time, quantifying the intrinsic coupling among cognitive, social, and affective processes. Its key contribution is a novel, natural-language-based framework for dynamic emotion analysis that overcomes limitations of conventional Likert-scale instruments. This framework provides a reproducible methodological foundation and empirical evidence for investigating implicit affective dynamics in collaborative contexts. (149 words)

Technology Category

• Cognitive Modeling & Cognitive Systems: Affective Computing
• Natural Language Processing: Sentiment Analysis, Stylistic Analysis, and Argument Mining
• Application Domains: Humanities & Computational Social Science

Application Category

• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data
• User Modeling, Personalization and Recommendation: Practical large-scale studies of user experience
• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding

Abstract

Collaborative problem solving (CPS) is a complex cognitive, social, and emotional process that is increasingly prevalent in educational and professional settings. This study investigates the emotional states of individuals during CPS using a mixed-methods approach. Teams of four first completed a novel CPS task. Immediately after, each individual was placed in an isolated room where they reviewed the video of their group performing the task and self-reported their internal experiences throughout the task. We performed a linguistic analysis of these internal monologues, providing insights into the range of emotions individuals experience during CPS. Our analysis showed distinct patterns in language use, including characteristic unigrams and bigrams, key words and phrases, emotion labels, and semantic similarity between emotion-related words.

Problem

• Investigates emotional states during collaborative problem solving
• Analyzes linguistic patterns in self-reported internal monologues
• Explores emotion-related language use in CPS teams

Innovation

• Mixed-methods approach for emotional analysis
• Video review for self-reported internal experiences
• Linguistic analysis of emotion-related language patterns