Derek Muller has done research on the effectiveness of science videos. To summarize his research in brief – when you present only the correct information in a science video without the possible misconceptions that students may have, students learn less (but feel better about the experience) than if you present information in a science video and include the misconceptions.
Of course, we should recognize that this effect probably does not depend on the medium of instruction, only on the nature of the medium. Videos are a unidirectional medium because they present information without the ability of the learner to ask questions. One might guess that any unidirectional medium may have the same effect. So textbooks, lectures, and other unidirectional mediums may suffer from this same deficit; without common misconceptions addressed in these mediums, the learners learn much less than if those misconceptions are addressed.
Dr. Eric Mazur shares essentially the same message – unidirectional instruction (in his case lecture) – has flaws. He relies on peer instruction and student response devices (clickers) to change the nature of the instruction so that it is more bidirectional (from each student’s perspective). The key here is that he has embedded more opportunities for feedback to reduce the chance that students incorporate the new information they are receiving into their existing misconceptions.
Textbooks (another unidirectional instructional tool) rarely present misconceptions and address them. Most students rarely use their textbooks as a learning resource (at least in k – 12), prefering to rely on the bidirectional instruction their teacher (or parent) provides. This means that the vast majority of information presented in a textbook goes unused. There are some changes to the textbook I’d like to see, which would allow for them to be a more bidirectional learning tool.
While it is clear that the medium of instruction influences the type of cognition that occurs, as Marshall McLuhan has pointed out, it should also be clear that different mediums have similarities in how they affect cognition or learning. If we find out that failing to address misconceptions in video instruction results in poor learning of the concepts, we may be able to transfer this finding to other modes of instruction. If that is the case, then we need to look at our instruction carefully, and ask ourselves, how much opportunity do we give students to address their existing models and resolve conflicts between their misconceptions, and the models we suggest?