Scholarlink

<p>With the growing adoption of Computer-Supported Collaborative Learning (CSCL), learners’ regulatory abilities have become crucial for enhancing learning quality and collaborative efficiency. Group awareness, an information mechanism integrating behavioral, cognitive, and social data and feedback, helps learners perceive their own and others’ collaborative states, thereby supporting individual and group regulation. However, research on the impact of group awareness on regulatory learning remains limited. This study developed three AI-based group awareness tools (behavioral, cognitive, and social) to enable real-time group awareness during collaborative discussions in CSCL. A quasi-experimental design was conducted in an authentic teaching context with three experimental groups and one control group. Multiple data sources, including log records, discussion texts, learning outcomes, and questionnaires, were analyzed using one-way ANOVA and Ordered Network Analysis (ONA) to evaluate regulatory abilities, learning performance, and regulation pathways. Results indicated that all three tools improved regulatory abilities, though the differences were not statistically significant. The behavioral group performed better in formative assessments, while the cognitive and social groups excelled in summative assessments; the control group performed the weakest. Methodologically, ONA combines node temporal information, directionality, and hierarchical structure to reveal the evolution of interactive behaviors and collaborative relationships, overcoming the limitations of static network analysis. ONA identified distinct regulatory patterns: the control group followed a loose “monitoring-planning” sequence, the behavioral group formed a closed loop of “planning-monitoring-adjustment-re-planning”, the cognitive group adopted a post-regulation pattern of “adjustment-monitoring”, and the social group showed a dynamic collaborative chain of “planning-monitoring-adjustment”. These findings provide theoretical and practical implications for designing intelligent, personalized regulation support strategies and instructional interventions.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

The impact of group awareness on CSCL regulated learning process: Insights from an ordered network analysis (ONA)

  • 期刊论文

  • 发表日期：

    2026年3月18日

查看全文
  • Qian Guo,
  • Wei Zhao,
  • Qiang Jiang,
  • Xiaoqing Xu

摘要

With the growing adoption of Computer-Supported Collaborative Learning (CSCL), learners’ regulatory abilities have become crucial for enhancing learning quality and collaborative efficiency. Group awareness, an information mechanism integrating behavioral, cognitive, and social data and feedback, helps learners perceive their own and others’ collaborative states, thereby supporting individual and group regulation. However, research on the impact of group awareness on regulatory learning remains limited. This study developed three AI-based group awareness tools (behavioral, cognitive, and social) to enable real-time group awareness during collaborative discussions in CSCL. A quasi-experimental design was conducted in an authentic teaching context with three experimental groups and one control group. Multiple data sources, including log records, discussion texts, learning outcomes, and questionnaires, were analyzed using one-way ANOVA and Ordered Network Analysis (ONA) to evaluate regulatory abilities, learning performance, and regulation pathways. Results indicated that all three tools improved regulatory abilities, though the differences were not statistically significant. The behavioral group performed better in formative assessments, while the cognitive and social groups excelled in summative assessments; the control group performed the weakest. Methodologically, ONA combines node temporal information, directionality, and hierarchical structure to reveal the evolution of interactive behaviors and collaborative relationships, overcoming the limitations of static network analysis. ONA identified distinct regulatory patterns: the control group followed a loose “monitoring-planning” sequence, the behavioral group formed a closed loop of “planning-monitoring-adjustment-re-planning”, the cognitive group adopted a post-regulation pattern of “adjustment-monitoring”, and the social group showed a dynamic collaborative chain of “planning-monitoring-adjustment”. These findings provide theoretical and practical implications for designing intelligent, personalized regulation support strategies and instructional interventions.