A Role Recognition Model Based on Students' Social-Behavioural-Cognitive-Emotional Features during Collaborative Learning
Saved in:
| Title: | A Role Recognition Model Based on Students' Social-Behavioural-Cognitive-Emotional Features during Collaborative Learning |
|---|---|
| Language: | English |
| Authors: | Cixiao Wang, Jianjun Xiao (ORCID |
| Source: | Interactive Learning Environments. 2025 33(4):3203-3222. |
| Availability: | Routledge. Available from: Taylor & Francis, Ltd. 530 Walnut Street Suite 850, Philadelphia, PA 19106. Tel: 800-354-1420; Tel: 215-625-8900; Fax: 215-207-0050; Web site: http://www.tandf.co.uk/journals |
| Peer Reviewed: | Y |
| Page Count: | 20 |
| Publication Date: | 2025 |
| Document Type: | Journal Articles Reports - Research |
| Education Level: | Elementary Education Grade 5 Intermediate Grades Middle Schools |
| Descriptors: | Elementary School Students, Grade 5, Student Behavior, Affective Behavior, Role Theory, Role, Social Influences, Cognitive Processes, Cooperative Learning, Group Dynamics, Group Behavior, Foreign Countries, Identification, Automation |
| Geographic Terms: | China (Beijing) |
| DOI: | 10.1080/10494820.2024.2442706 |
| ISSN: | 1049-4820 1744-5191 |
| Abstract: | Role recognition is critical for labor division and risk identification in group coordination during collaborative learning. Students' social, cognitive, behavioural, and emotional performance during collaboration are essential dimensions for role recognition. However, most studies classify roles using single dimensions, such as cognitive (knowledge construction level) or social (social network status), neglecting behavioral and emotional indicators. This study develops a multidimensional role recognition model integrating social, cognitive, behavioural, and emotional features to automatically detect students' roles (coordinator, inquirer, assistant, marginal) during collaboration. Results show that the multidimensional model outperforms single-dimensional models, with ensemble classifiers (e.g., random forest, XGBoost) outperforming single classifiers (e.g., support vector machine, decision tree). Additionally, an interpretable framework is proposed for global and local explanations of the model. Globally, social, cognitive, emotional, and behavioural factors influence role recognition, with eight key features identified for each role. Common features, such as investment and overall responsivity, influence all roles, while others vary in their impact. Locally, the framework supports personalized interventions, such as tailored collaborative scripts. These findings offer valuable insights for researchers and practitioners, enabling early identification of roles associated with academic risks and the design of targeted instructional strategies. |
| Abstractor: | As Provided |
| Entry Date: | 2026 |
| Accession Number: | EJ1508450 |
| Database: | ERIC |
| Abstract: | Role recognition is critical for labor division and risk identification in group coordination during collaborative learning. Students' social, cognitive, behavioural, and emotional performance during collaboration are essential dimensions for role recognition. However, most studies classify roles using single dimensions, such as cognitive (knowledge construction level) or social (social network status), neglecting behavioral and emotional indicators. This study develops a multidimensional role recognition model integrating social, cognitive, behavioural, and emotional features to automatically detect students' roles (coordinator, inquirer, assistant, marginal) during collaboration. Results show that the multidimensional model outperforms single-dimensional models, with ensemble classifiers (e.g., random forest, XGBoost) outperforming single classifiers (e.g., support vector machine, decision tree). Additionally, an interpretable framework is proposed for global and local explanations of the model. Globally, social, cognitive, emotional, and behavioural factors influence role recognition, with eight key features identified for each role. Common features, such as investment and overall responsivity, influence all roles, while others vary in their impact. Locally, the framework supports personalized interventions, such as tailored collaborative scripts. These findings offer valuable insights for researchers and practitioners, enabling early identification of roles associated with academic risks and the design of targeted instructional strategies. |
|---|---|
| ISSN: | 1049-4820 1744-5191 |
| DOI: | 10.1080/10494820.2024.2442706 |
