Notes: Wermeskerken et al. Effects of Instructor Presence in Video Modelling Examples on Attention and Learning

van Wermeskerken, M., Ravensbergen, S., & van Gog, T. (2018). Effects of instructor presence in video modeling examples on attention and learning. Computers in Human Behavior, 89, 430-438. https://doi.org/10.1016/j.chb.2017.11.038

Does including a video that includes a view of the presenter add (through instructor presence) or detract (through distracting the viewer by forcing them to split attention)? This paper takes a look at a specific design pattern - an instructor presenting some content in a presentation - and uses eye tracking to observe where a viewer’s attention is focused during the video, and tries to assess the impact on learning (using pre/post test scores as a proxy). Is the a deeper connection through social response stronger than the “split attention effect” where viewers are automatically drawn to look at a human face rather than the content in a presentation? Does working memory¹ get messed up by having the instructor visible (but separate from the content)?

My Questions:

• Should we include a picture-in-picture or split-view that shows the instructor during a presentation?
• Should we separate a presentation into an introductory bit that shows the instructor (yay instructor presence) but not during the presentation itself?
• Does the lightboard^{2 3} mitigate this? Instructor presence during a presentation, without distracting the viewer? Or is that just moving the distraction closer to the content so it’s not as severe?
• The split-attention effect - does it actually hinder learning, or is immediate recall of processes affected differently from longer-term understanding of concepts?
• Does this study generalize to non-process learning through video, or could instructor presence help in other topics?

Notes:

They looked at 2 versions of a video, one “Instructor Visible”, one “No Instructor”.

3 hypotheses:

1. viewers of “Instructor Visible” version would spend significant time looking at the instructor rather than the content of the presentation.
2. viewers of “Instructor Visible” version would spend less time focused on the relevant content displayed, taking longer to focus attention on relevant content when it is displayed.
3. presence of instructor in “Instructor Visible” version would hinder learning.

They used a video on calculating probability. The study involved:

1. pre-test of 5 questions of different types
2. watch one of 2 formats of the video (“Instructor Visible” or “No Instructor”). eye-tracking observation while students watched the video
3. post-test of 10 questions

Viewers of the IV version demonstrated split-attention, as documented by the eye-tracking system.

BUT - there was no significant difference in either retention or test performance, so maybe it doesn’t affect learning?

Despite the substantial amount of attention that was paid to the instructor’s face, at the expense of attention paid to the task overall and to the task parts the instructor was referring to, learning out- comes (i.e., retention and transfer scores) were not significantly affected by instructor presence (H3); even though performance was slightly higher (about 10% on both retention and transfer test) when no instructor was visible, this difference was not statistically reliable. The exploration of the association between viewing behavior and learning outcomes in the instructor visible condition (Q2) did not reveal any significant association.

and

Thus, the fact that the instructor attracts learners’ attention, can be used to guide learners’ attention when the instructor employs social cues such as gaze and/or gesture cues, or when the instructor manipulates objects

Which really sounds like something like Lightboard could be a useful method of integrating instructor presence and social cues without distracting attention from the displayed content… Which, obviously, but data!