.png){kind=icon}
.png){kind=icon}
Return
Implementing multimodal learning analytics in authentic settings: A roadmap for ecological impact
.png){kind=icon}
PRE
AI
DOI：10.1016/j.learninstruc.2026.102341.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Multimodal Learning Analytics (MMLA) refers to the use of heterogeneous data sources (e.g., speech, gesture, gaze, posture, physiological signals, digital traces) to model, interpret, and support learning processes as they unfold over time. Research using MMLA is moving from laboratory settings to real-world applications; however, significant challenges remain in implementing these methods in authentic learning environments. This special issue in Learning and Instruction brings together studies that have successfully applied MMLA in various ecological contexts, including classrooms, makerspaces, and high-fidelity simulations. While these contributions show the potential of ecological MMLA, they also highlight that most work remains in the early, descriptive stages, with a limited number of experimental, interventionist, or longitudinal designs capable of establishing causal relationships or sustained impacts in learning and instruction. To support the field's development, this editorial reviews the state-of-the-art, as exemplified by the papers in this special issue, and outlines a roadmap for the next decade, identifying key stages and challenges, to guide MMLA toward broader ecological impact.
Keywords:
Multimodal Learning Analytics
Ecological Validity
Learning Analytics
Authentic Settings
Educational Research
Journal
IF:
4.9
Papers:
319
Citations:
8.7K
Organization
View more organizations