The coherence principle states that when designing instruction using multimedia, extraneous audio, graphics, and text have not led to improved learning outcomes. According to research in cognitive science, there are three principles to remember when thinking about the coherence principle:

  • People have separate channels for processing visual/pictorial material and audio/verbal material. This is referred to as dual channels.
  • People can actively process only a few pieces of information in each channel at one time. This is referred to as limited capacity.
  • Learning occurs when people engage in appropriate cognitive processing during learning. This is referred to as active processing. (Clark & Mayer, 2016, p. 35)

     Clark and Mayer examined two studies related to extraneous audio. The first study looked at the effect of background music embedded within animated narration. The results showed that background music substantially hurt learning.  (Clark & Mayer, 2016) The second study involved students reading text in a quiet environment versus an environment that included irrelevent conversational background speech. Recall of text was significantly better among those reading in the quiet environment. (Clark & Mayer, 2016) The reasoning behind this is that extraneous sounds compete for limited cognitive resources in the auditory channel.

     The same fight for limited cognitive resources applies to graphics in multimedia as well. According to Harp and Mayer (1998), irrelevant pictures can interfere with learning due to their ability to distract, disrupt, and seduce the learner away from the most important content. Below is an example from a recent instructional coaching meeting I attended that illustrates how the flower images do not support any content on the page and seduce the viewer in another direction. As well, it is extremely text heavy. A visual step-by-step graphic would have worked much better while communicating the same information. The text was read to us as well by the presenters so there was no need for all the text on the slide.


     However, a fifth grader I worked with applied the use of a basic graphic with audio support well. This is a great use of dual channels related to the modality principle. As well, the visual applies aspects of the contiguity principle by adding support text right next to the visual it is referencing. Finally, the slide supports the coherence principle by showing exactly and not much more than was intended for learning to occur.


     The third consideration when designing multimedia instruction has to do with eliminating extraneous text for interest, elaboration, or technical depth. Clark and Mayers (2016) recommend providing learners with concise summaries instead of providing the learning material with additional complementary text. Due to the high cognitive load that is naturally experienced in technical learning, more text is not often helpful. It prevents the learner from constructing a visual representation in their head about what they are learning.

     It is easy to conclude that the coherence principle results in boring learning material. However, even before the computer, John Dewey understood that “When things have to be made interesting, it is because interest itself is wanting.” Therefore, a learner can experience enjoyment from their sense making process of material that is already inherently engaging.

     In conflict with this view, the arousal theory predicts that students will learn more from multimedia presentations that contain engaging sounds, visuals, and/or text compared to presentations without them. The belief here is that learning needs to emotional reach the learner in order for learning to occur and some sound, graphics and text allow for the emotional engagement to happen.

     I generally agree with the coherence principle, but I am also interested in the engagement factors related to learning. Research conducted by Park, Flowerday and Brunken ( 2015) examined the cognitive and affective effects of seductive details in multimedia learning. They believe that there has been an overemphasis on studies related to cognition in multimedia and more research needs to be done to examine affective processes. The key question for consideration of designers of multimedia instruction in their mind is: Do learners have enough cognitive resources free for using motivating bits and pieces of information?

     Personally, I tend to favor a balanced approach to all my instruction with a concern for the learners in front of me first. There are times when I need to use sensational tactics to launch a unit or conclude a unit to put a spark in my learners or add closure to a unit. Most of the time, I try to balance my use of text, graphics, and audio or a combination of them. The area I have to work hardest at remembering is not to include font or graphics that might look beautiful, but could prove to be distracting to learners. In closing, engagement in the learning process is extremely important. Educators should always remember that the content should be engaging in and of itself first and then consider how multimedia effects might support that content.

     An area of interest that was not pursued by Clark and Mayer has to do with the spatial abilities of learners. Koroghlanian, C., & Klein, J. D. (2004) discovered that students who tested high in spatial ability performed better on practice items within multimedia instruction compared to those who scored low in spatial ability. This suggests that how information is organized on a web page, termed “screen real estate”, can positively or negatively affect the learner as well.



Clark, R. C., & Mayer, R. E. (2016). E-learning and the science of instruction, 4th edition.       John Wiley and Sons: Hoboken, New Jersey.

Harp, S.F., & Mayer, R.E. (1998). How seductive details do their damage: A theory of cognitive interest in science learning. Journal of Educational Psychology, 90, 414-434.

Koroghlanian, C., & Klein, J. D. (2004). The effect of audio and animation in multimedia instruction. Journal of Educational Multimedia and Hypermedia, 13(1), 23-46.

Park, B., Flowerday, T., & Brünken, R. (March 01, 2015). Cognitive and affective effects of seductive details in multimedia learning. Computers in Human Behavior, 44, 267-278.