Biard, N., Cojean, S., and Jamet, E. (2018) Effects of segmentation and pacing on procedural learning by video. Computers in Human Behavior, 89, 411-417. https://doi.org/10.1016/j.chb.2017.12.002
In this study, Biard, Cojean, and Jamet (2018) explore how “the pace of presentation of an instructional video affects the learning of professional skills” (413). Particularly, they are interested in the effect of system-directed or learner-directed pauses in instructional videos designed to teach procedural skills.
The authors justify their study by accurately pointing out that videos are becoming more frequently used in education. Particularly, videos are often used to teach procedural skills: “learning to perform a series of actions to achieve a particular goal” (411). According to the authors, some studies show that instructional videos are more effective than static images, text, or drawings, while other studies show the opposite because video information is “transient, and more cognitively costly processing is thus required to extract the relevant items and hold them in memory” (412). A continuous flow of information, like that encountered in an instructional video, imposes a heavy cognitive load on the learner. The most current recommendation for this problem found in the literature is to present “learner-paced segments” (412). If learners have control over the video, the theory goes, they can halt this flow of information and lessen the cognitive burden.
The authors hypothesize that “the pace of presentation of an instructional video affects the learning of professional skills” (413). Particularly, they expect that novices have difficulty figuring out where they should pause in order to memorize the content. They explain that, “Students spontaneously make little use of a pause button, as they do not know when to halt the video, so learner-paced pausing can only improve procedural learning when it is combined with system-paced segmentation” (413). Having a pause button does not mean students (especially novices) will actually use it and thus gain the cognitive load benefit, and so system directed pauses will be more useful.
To test this hypothesis, 68 occupational therapy students viewed a video on a clinical procedure that consisted of 7 main steps and additional sub-steps. The students were split into 3 groups: (1) the noninteractive video group was shown a video they could not pause, (2) the interactive video group was shown a video they could pause, and (3) the segmented interactive video group could pause their video, but the video also had a system-directed pause after each step. The authors recorded the number, timing, and length of the pauses during student viewings, and had students complete a recall and procedural learning test after watching the video.
The duration of total pauses in segmented interactive video group were 10x longer than spontaneous use of pause button in group two. All three groups had the same performance on the recall tests, but the interactive segmented video group performed the best on the procedural test. Why is this the case? The authors postulate that it is in part due to Mayer’s segmentation principle (the idea that segmenting or halting the continuous flow of information can help prevent cognitive overload), but also because splitting the video after each of the 7 tasks helps students to build a “relevant mental model” of the task. In procedural knowledge (one that is step-by-step, in other words), a “well-structured mental representation of the procedure is needed to succeed” (415).
My primary critique of this article relates to the authors’ claims that novices in particular do not know when to pause videos, and so it is even more important to have system-directed pauses. Their entire sample (68 students) consisted of “novice” learners. I think that in order to make this claim, they needed to include “expert” learners. Would an expert really pause a video any more frequently than a novice? The assumption is that it is the novice’s lack of metacognition that prevents them from pausing the video, but both procedural novices and experts can have poor metacognitive skills. Perhaps, too, the lack of leaner-directed pauses could be less about metacognition and more about engagement — a student may be more inclined to “power through” a boring video as compared to pausing and engaging with an interesting one. In other words, I am not sure that the claim that novices in particular need system-directed pauses was adequately supported by the experiment design.
Overall, I appreciate that this article combines both theory and practical applications. The literature review is extensive, and the authors take care to build on the work of previous research, particularly through their reference of Mayer’s multimedia theories. Similarly, when designing procedural learning videos, designers and instructors can practically apply the findings of this article. As a designer, faculty often push back against “short” videos, as they are used to much longer face-to-face class sessions and would prefer to simply put their 50-minute lecture into their online course. Articles such as this one are beneficial because they offer research paired with a recommendation: videos that can be paused by the learners are good, but segmented videos or videos with system-directed pauses are even better — particularly for procedural learning.