We recently explored whether our eye movements during online learning could reveal how well we’re understanding the material. With my colleagues, I ran two studies on eye-tracking in educational videos, and we found some pretty exciting insights into how students’ visual behavior can relate to their learning outcomes. Let me walk you through what we discovered and why it matters for the future of online education.

What’s the Issue?

In our first study, we wanted to see if webcam-based eye tracking could predict how well a student would do on comprehension tests after watching educational videos. Eye movement synchronicity—how similar a student’s eye movements are to those of the average viewer—has been shown to predict test performance in controlled settings. But what happens if we use regular webcams, like in most real-world online learning environments, where the data quality isn’t perfect?

The results were mixed. It was tough to find a reliable link between synchronicity and learning outcomes due to the variability in webcam quality—different resolutions and frame rates made it difficult to track attentional behavior consistently. Still, we saw that having an on-screen instructor seemed to influence students’ eye movement patterns, though we’ll need more research to really understand that effect.

Distance to the Pointer: A Better Predictor

In the second study, we decided to look at something different: the distance between where a student was looking and the position of a digital laser pointer used by an instructor. We wanted to know if students who closely followed the pointer performed better in post-video quizzes.

Here, the results were more promising. We found that the average distance between a student’s gaze and the laser pointer was a strong predictor of how well they did on the quiz, regardless of whether an on-screen instructor was present. In other words, students who tracked the laser pointer more closely understood the content better. This was true even when an instructor was visible and potentially distracting. This finding opens up some exciting possibilities for making online education more adaptive: systems that can tell when a student isn’t following key visual cues could provide real-time feedback to help them refocus.

What Does This Mean for E-Learning?

These two studies show how complex it is to use webcam-based eye tracking for educational purposes. While the first study highlighted the challenges of using webcam quality to measure synchronicity, the second study’s findings about gaze-to-pointer distance point to a new direction. If we focus on how well students visually engage with important elements of a video, we might be able to create learning environments that adapt to each student in real time.

Eye tracking for understanding and predicting learning outcomes is still in its early days, but these insights are a step toward making online education more effective. Moving forward, we need to refine the technology, improve reliability, and create smart algorithms that can tailor feedback to individual learning behaviors.

Why Does It Matter?

Eye tracking gives us a fascinating window into the future of online learning, where systems might soon be able to “see” what students are focusing on and give personalized feedback right away. While there are challenges, particularly with getting consistent webcam data, the positive results with gaze-to-pointer tracking show real potential for making e-learning more interactive and efficient.

These studies pave the way for new research into how we can better monitor and optimize attention in digital learning settings. By understanding where students are looking, we can better understand what—and how—they’re learning.