Experiment 1: Full Body
In this experiment, they split the participants into three groups who watched videos with the same content and audio, but different presentation styles. One group (screenshot on the left) saw the presenter talk through an already-drawn diagram explaining the content. The middle screenshot is of the presenter drawing the diagram as she talks through it. The right side screenshot shows the presenter pointing at an already drawn presentation.
Results: Seeing the diagram being drawn improved learning for people with low prior knowledge. This highlights how novice learners have different needs than learners with more expertise. This is called the expertise reversal effect and has been researched extensively. It's one reason we need to consider our audience when designing instruction.
Experiment 2: Hand Only
This time they compared audio explanation of an already drawn diagram (no hand or body) with a version that showed the diagram being drawn with the hand.
Results: The group who saw the hand drawing the diagram performed better. In this case, prior knowledge did not matter.
Experiment 3: No Hand
At this point, the question is whether seeing the hand or seeing the diagram being drawn caused the improvement in experiment 2. Researchers then used an iPad app to record a diagram being drawn and one already drawn (both with audio), no hand in either condition.
Results: No difference was recorded, leading them to conclude the difference was due to the hand. This was surprising to me because it seems clearly better to see the diagram being drawn instead of simply staring at a stagnant drawing while listening to a lecture. This was a very short presentation (around 2 minutes), so I'm curious what would happen if it was longer. Prior knowledge did not matter.
Experiment 4: Body vs Hand
The last experiment compared being able to see the hand with being able to see the body in the recording.
Results: The group who saw the hand performed better. Fiorella and Mayer speculated that the hand provides enough of a visual cue to stimulate a social connection with the instructor, which helps motivate the learner (social agency theory) whereas seeing more of the instructor can be a distraction.
I presented on this at a conference yesterday, and a participant asked if the size could have made a difference, since the version with the hand is more zoomed in. That's a good point; the researchers didn't address it.
I wish they would have compared drawing with the hand to drawing without the hand on the iPad app so there was a true hand/no hand experiment and lengthen it to about 7 minutes instead of 2 so it's more realistic. Experiment 2 compared hand to no hand, but included the variable of being drawn or already drawn.
The point, though, seems to be that the hand provides a valuable social cue. This is good news for instructors who like to write on paper under a document camera and record it to make a video. This is a low tech solution that can be done in an empty classroom. Writing on a whiteboard as they did in these screenshots is a good option, but I think a little more difficult because you need to stay in a particular spot or have someone record it for you. In addition, you might need to wear a mic to get clear audio since the video camera wouldn't be as close as a document camera.
Fiorella, L., & Mayer, R. E. (2015). Effects of observing the instructor draw diagrams on learning from multimedia messages. Journal of Educational Psychology. Advance online publication. http://dx.doi.org/10.1037/edu0000065