Characterizing Learners' Complex Attentional States During Online Multimedia Learning Using Eye-tracking, Egocentric...

Area of Science:

• Educational Technology
• Cognitive Science
• Human-Computer Interaction

Background:

• Online learning requires sustained learner attention, a critical factor for effective knowledge acquisition.
• Traditional eye-tracking measures visual attention but cannot fully capture cognitive engagement or mind-wandering.
• Distinguishing between looking at materials and thinking about them is crucial for understanding learning states.

Purpose of the Study:

• To develop a comprehensive method for characterizing learners' attentional and cognitive states in online learning environments.
• To create a 2x2 matrix of attentional/cognitive states, moving beyond the limitations of eye-tracking alone.
• To generate ground truth data for training machine learning models to identify these complex learning states.

Main Methods:

• Combined eye-tracking with an egocentric camera, webcam, retrospective recall, and mind-wandering probes.
• Collected data from 101 learners interacting with a multimedia physics module.
• Developed a 2x2 matrix to categorize four distinct attentional/cognitive states.

Main Results:

• Successfully captured a nuanced 2x2 matrix of attentional/cognitive states in online learners.
• Demonstrated that eye-tracking alone is insufficient for fully understanding cognitive engagement.
• Generated a rich dataset for further research into learning states and outcomes.

Conclusions:

• A multi-modal approach is necessary to accurately assess learners' attentional and cognitive states during online learning.
• This methodology provides valuable insights for both basic and applied research in educational technology.
• The generated ground truth data will advance machine learning applications for monitoring and supporting learner engagement.