Given that blended and online courses are growing, creating instructional videos is in great demand as this format of instruction allows the integration of visual and verbal information for learning. In an upcoming special issue in the Journal of Computers in Human Behaviour, “Developments and Trends in Learning with Instructional Video”, Fiorella and Mayer (2018) summarised the findings of the studies in the issue and provided recommendations for future work:
What worked?
Segmenting the video: Biard, Cojean, and Jamet compared three formats of an instructional video about a medical prodedure: i) continuous video, ii) learner could control pause and play the video, and ii) system pauses the video at different segments and learner could decide when to continue. Students who viewed the video in the condition in which the system pauses the video performed better than students in the other two conditions.
Mixing perspective using different camera viewpoints: Boucheix, Gauthier, Fontaine, and Jaffeux compared viewing the video from the first person persepective (over the actor’s shoulders), the third-person perspective (camera facing the actor), and a mix of first and third person perspectives. Results showed that students who viewed the mixed perspective video performed better. However, the effect of perspective on performance depended on the type of task that students needed to perform.
What did not work?
Matching gender of the learner to the instructor: Hoogerheide, van Wemeskerken, van Nassau, and van Gog found that although learners perceived the instructor as more similar to them when the gender matched, there was no significant influence on learning.
Making instructor visible: Van Wemeskerken, Revensbergen, and van Gog compared a video with a visible throughout the video and a video without a visible instructor. Using eye-tracking data, they found that students looked at the instructor’s face for about 30% of the time. However, there were no significant differences in learning outcomes between the two groups.
Practice without feedback: van der Meij, Rensink, and van der Meij compared performance of students who watched a video only, did a practice before watching the video, and did a practice after watching a video. There were no significant differences in learning outcomes across the three conditions. One possible explanation for the lack of finding was the absence of feedback after practice.
Inserting pauses: Merkt, Ballman, Felfeli, and Schwan explored the use of pauses to segment the videos. The learners in this study were given videos that paused at certain time points and continues after three seconds whereas in Biard, Cojean, and Jamet’s study, learners choose when to continue with the video. The authors did not find any significant differences between videos with and without pauses nor between the videos that paused at meaningful or non-meaningful points.
Gender and spatial ability: Wong, Castro-Alonso, Ayres, and Paas examined the relationship among gender, spatial ability, instructional format (static images and dynamic animation), and learning outcomes. Results of the three experiments in the study was mixed and measure of spatial ability was inconsistent. More research is needed to examine how spatial ability relates to performance on specific learning tasks.
Fiorella, L., & Mayer, R. E. (2018). What works and doesn't work with instructional video. In press.