Scholarlink

<p>Immersion represents a critical characteristic of the gaming experience and serves as a key indicator for evaluating learners’ enjoyment in Digital Game-based Learning (DGBL). The scientific and systematic assessment of immersion levels is an important fundamental aspect of DGBL research. However, most existing assessment tools are based on self-reported questionnaires, which are not well-suited for real-time or embedded assessment of immersion. To address this limitation, we propose an immersion assessment scaffold based on behavioral analysis in DGBL(DGLIAS), designed for the quantitative and qualitative assessment of learners’ immersion. Subsequently, behavioral and self-reported data were collected from 174 participants at various checkpoints within the same game-based learning environment. By applying the Multi-Layer Perceptron (MLP) algorithm, we learned the weights of 50 indicators in the DGLIAS measurement framework using machine learning. The validation results demonstrated strong model performance, with an average classification accuracy of 95.66% (ACC = 0.9566), a mean squared error of 29.54 (MSE = 29.54), and a coefficient of determination of 0.8734 (R² = 0.8734). These findings confirm that DGLIAS can effectively assess learners’ current immersion in real time through behavioral analysis. In summary, this study contributes valuable theoretical and practical insights into immersion assessment within the context of DGBL.</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

An immersion assessment scaffold for digital game-based learning environments through behavioral analysis

  • 期刊论文

  • 发表日期：

    2025年10月24日

查看全文
  • Yong-Dai Miao,
  • Jie Zhang,
  • Hai Zhang

摘要

Immersion represents a critical characteristic of the gaming experience and serves as a key indicator for evaluating learners’ enjoyment in Digital Game-based Learning (DGBL). The scientific and systematic assessment of immersion levels is an important fundamental aspect of DGBL research. However, most existing assessment tools are based on self-reported questionnaires, which are not well-suited for real-time or embedded assessment of immersion. To address this limitation, we propose an immersion assessment scaffold based on behavioral analysis in DGBL(DGLIAS), designed for the quantitative and qualitative assessment of learners’ immersion. Subsequently, behavioral and self-reported data were collected from 174 participants at various checkpoints within the same game-based learning environment. By applying the Multi-Layer Perceptron (MLP) algorithm, we learned the weights of 50 indicators in the DGLIAS measurement framework using machine learning. The validation results demonstrated strong model performance, with an average classification accuracy of 95.66% (ACC = 0.9566), a mean squared error of 29.54 (MSE = 29.54), and a coefficient of determination of 0.8734 (R² = 0.8734). These findings confirm that DGLIAS can effectively assess learners’ current immersion in real time through behavioral analysis. In summary, this study contributes valuable theoretical and practical insights into immersion assessment within the context of DGBL.