Effect of Instructor Presence Online

Since I try to focus these SoTL blog posts on current events in higher ed, over the past year, I have been sharing how AI can support teaching and learning. This week I would like to share a tangent to AI, or perhaps an outcome if we use AI to create videos. The article is entitled, “Avatars vs. video presence: Effects of instructor presence on cognitive load in video-based learning” by Nugroho, et al. (2025). 

The authors investigated how different types of instructor presence in video-based learning—no instructor, physical instructor, and avatar instructor—affect cognitive load and learning outcomes. Using eye-tracking technology to measure pupil dilation as a proxy for cognitive load, the study involved undergraduate students from electronic engineering and computer science.

Key findings include:

• No instructor presence minimized distractions; provided only moderate learning gains.

• Physical instructor presence produced the highest extraneous cognitive load due to distracting non-verbal cues, and was negatively correlated with learning gains.

• Avatar presence reduced extraneous load while improving learning outcomes.

Ultimately, avatars emerged as a scalable and efficient instructional design choice for online and asynchronous education, offering the benefits of presence without overloading learners.

These findings align strongly with Cognitive Load Theory (Sweller, 2010), which distinguishes between intrinsic, extraneous, and germane load. The study demonstrates how avatars help minimize extraneous load while enhancing germane load, which supports schema construction and deeper learning. From a social presence perspective (Garrison, Anderson, & Archer, 2000), the study suggests avatars can replicate enough of an instructor’s presence to sustain motivation and engagement, while avoiding the overload introduced by real instructors’ dynamic expressions and gestures.

Faculty can apply these findings in several ways:

1. Use avatar-based tools in online courses: Platforms like Synthesia, Vyond, or built-in LMS avatars can deliver content with controlled gestures and minimal distractions.

2. Simplify video presence: If recording themselves, instructors should avoid unnecessary movements, background distractions, or overly dynamic gestures that may increase extraneous load.

3. Design for guided presence: Even short, avatar-based introductions or summaries in online modules may improve learning gains compared to pure slide-and-voice formats.

4. Balance presence across modalities: Consider alternating between instructor avatars for explanation and slide-based visuals for detail-heavy content to optimize processing.

5. Low-stakes implementation: Instructors could try replacing weekly lecture recaps with a 2–3 minute avatar-based video to see how students respond this term.

Despite its contributions, the study has several limitations:

1. Sample size and demographics: With only 33 students from a single discipline, generalizability across fields and age groups is limited.

2. Single physiological measure: Reliance on pupil dilation alone, while objective, ignores other measures (e.g., EEG, heart rate variability) that could triangulate cognitive load. Emotional states (e.g., stress) may also confound pupil dilation.

3. Simplistic avatar design: The avatar used had limited gestures and static expressions. Future research should compare dynamic avatars or AI-driven adaptive avatars to assess whether richer presence enhances or hinders learning.

4. Controlled lab setting: The study’s tightly controlled conditions may not fully represent real-world online learning environments, where distractions are common.

References

     Garrison, D. R., Anderson, T., & Archer, W. (2000). Critical inquiry in a text-based environment: Computer conferencing in higher education. The Internet and Higher Education, 2(2–3), 87–105.

Nugroho, Y. S., Bourguet, M.-L., Soyel, H., & Mareschal, I. (2025). Avatars vs. video presence: Effects of instructor presence on cognitive load in video-based learning. Electronic Journal of e-Learning, 23(3), 79–91. https://doi.org/10.34190/ejel.23.3.3964

Sweller, J. (2010). Element interactivity and intrinsic, extraneous, and germane cognitive load. Educational Psychology Review, 22(2), 123–138.