A Multimodal Real-Time Feedback Platform Based on Spoken Interactions for Remote Active Learning Support.

Area of Science:

• Educational Technology
• Learning Analytics
• Human-Computer Interaction

Background:

• Technology's potential in education remains largely untapped, particularly for improving collaborative learning efficiency.
• Remote learning environments pose challenges for monitoring and facilitating group interactions effectively.

Purpose of the Study:

• To introduce NAIRA, a Multimodal Learning Analytics platform designed for real-time feedback in collaborative learning.
• To qualitatively assess teacher interventions using NAIRA's visualizations during remote group activities.
• To quantitatively analyze differences in student verbal interactions with and without NAIRA support.

Main Methods:

• A case study involving 24 undergraduates engaged in a remote Jigsaw learning activity.
• Utilized NAIRA's real-time visualizations of student verbal interactions for teacher-led interventions.
• Compared quantitative metrics of spoken interactions between NAIRA-supported and unsupported activity stages.

Main Results:

• NAIRA enabled effective monitoring and facilitation of a supervised remote learning activity, even with students in separate virtual rooms and cameras off.
• A more homogeneous distribution of speaking time was observed during the NAIRA-supported stage compared to the unmonitored stage.
• Teacher interventions were qualitatively described based on the use of NAIRA's visualizations.

Conclusions:

• NAIRA is a viable tool for enhancing the efficiency and equity of remote collaborative learning activities.
• Real-time multimodal learning analytics can empower educators to provide timely and precise support in virtual learning environments.
• The platform facilitates more balanced participation in group work, addressing a key challenge in online education.