Mayer's Multimedia Principles – 12 Rules for Better eLearning
Every eLearning screen is a small experiment in how people process words and pictures. Add narration or not? Put text on the diagram or below it? Include a friendly on-screen presenter? Most teams answer these questions by taste or by template. There is a better guide: a long research program that tested these exact choices. What are Mayer's multimedia learning principles, and how do you apply them to real eLearning?
Mayer's multimedia learning principles are twelve evidence-based guidelines, developed by educational psychologist Richard E. Mayer and colleagues, for combining words and pictures so people learn more: they tell you to cut extraneous material, highlight what matters, place related words and visuals together, pace complex content in segments, use narration with graphics and write in a conversational style. The principles come from experiments comparing different versions of the same lesson and measuring what learners could explain and apply afterward. They fall into three groups: reducing wasted processing, managing essential processing and encouraging deeper processing. They work best for novices learning complex material, and Mayer is careful to note the conditions where each effect is weaker.
What Is the Cognitive Theory Behind the Principles?
The multimedia learning principles rest on Mayer's cognitive theory of multimedia learning, set out most fully in his book Multimedia Learning, first published in 2001. The twelve principles here are the classic set from its second edition; the 2020 third edition added a few more, such as the embodiment and immersion principles, so you may see longer lists elsewhere. The theory makes three assumptions.
- Dual channels. People process visual and auditory information through partly separate channels.
- Limited capacity. Each channel can handle only a small amount at a time.
- Active processing. Real learning happens when people select relevant information, organize it into a coherent picture and connect it with what they already know.
From those assumptions come three kinds of processing that compete for limited capacity. Extraneous processing is wasted on poor design. Essential processing is needed to represent the core material. The third kind, which Mayer describes as cognitive processing aimed at making sense of the material, is what builds lasting understanding. This article calls it deeper processing. Each principle targets one of these. A helpful summary of Mayer's 12 principles of multimedia instruction from NUS Medicine shows how they are grouped.
Which Principles Reduce Wasted Processing?
Five principles help learners avoid spending effort on things that don't help them learn.
1. Coherence
People learn better when extraneous words, pictures and sounds are left out. Background music, decorative images, interesting but irrelevant stories and extra detail all compete for attention. In eLearning, this is the principle most often broken. Ask of every element, "Does this help someone learn the objective?"
2. Signaling
People learn better when cues highlight the organization of the essential material. Headings, arrows, highlights and spoken phrases such as "the second step is" help learners see structure. Signaling is cheap and effective, especially in longer explanations.
3. Redundancy
People learn better from graphics and narration than from graphics, narration and identical on-screen text. When narration reads the screen text word for word alongside a diagram, learners split attention between reading and looking. Short on-screen labels or key terms are fine. Full paragraphs duplicated in audio usually are not.
4. Spatial Contiguity
People learn better when related words and pictures are placed near each other. Put labels on the diagram, not in a legend. Put feedback next to the question, not on a separate screen. Small layout fixes like these remove a surprising amount of effort.
5. Temporal Contiguity
People learn better when matching words and pictures are presented at the same time rather than one after the other. If a narrator describes a valve opening, the animation should show the valve opening at that moment, not after the narration ends.
Which Principles Help Manage Complex Material?
Some material is hard by nature. These three principles help learners handle essential complexity without overload.
6. Segmenting
People learn better when a complex lesson is presented in learner-paced segments rather than as a continuous unit. A five-minute animation becomes five short clips with a "continue" button between them. Learner control over the pace gives working memory time to catch up.
7. Pre-training
People learn better when they already know the names and characteristics of the main concepts. Before explaining how a system works, introduce its parts. Learners can then focus on how the parts interact instead of juggling new names at the same time.
8. Modality
People learn better from graphics with spoken narration than from graphics with printed text, especially when the material is complex and fast-paced. Narration lets the eyes stay on the visual while the ears take in the explanation. This is a strong reason to narrate animated diagrams and process walkthroughs, though it applies less when learners need to reread or the text is very short.
Which Principles Encourage Deeper Processing?
The last four principles help learners engage with the material and make sense of it.
9. Multimedia
People learn better from words and pictures than from words alone. This is the foundation of the whole program. The pictures must be relevant, though. A diagram that shows how something works helps. A stock photo of smiling coworkers does not.
10. Personalization
People learn better when words are in a conversational style rather than a formal one. Using "you" and "your" and speaking directly to the learner tends to improve results. It signals a social relationship, which encourages people to try harder to understand.
11. Voice
People learn better when narration is spoken in a friendly human voice rather than a machine voice. In Mayer's experiments, the comparison was between a recorded human speaker and the computer-synthesized speech of the time. Later studies by Scotty Craig and Noah Schroeder found that newer, higher-quality synthetic voices closed much of that gap, so the practical rule is about clarity and warmth, not how the audio was produced. Test any voice with a few learners before you commit to it.
12. Image
People do not necessarily learn better when the speaker's image is on screen. A talking head beside a diagram can pull attention away from the content. Use an on-screen presenter when it adds something, such as demonstrating a physical skill, not as decoration.
The 12 Multimedia Learning Principles at a Glance
| Principle | Quick design rule |
|---|---|
| Coherence | Cut anything that doesn't serve the objective |
| Signaling | Highlight structure and key points |
| Redundancy | Don't narrate identical on-screen paragraphs |
| Spatial contiguity | Place words next to the visuals they describe |
| Temporal contiguity | Sync narration with matching visuals |
| Segmenting | Break complex content into learner-paced pieces |
| Pre-training | Teach key terms and parts first |
| Modality | Narrate complex visuals instead of adding text |
| Multimedia | Use relevant pictures with words |
| Personalization | Write conversationally, use "you" |
| Voice | Use a clear, friendly human voice |
| Image | Skip the talking head unless it adds value |
What Does a Principle-by-Principle Review Look Like?
Here is how the principles might reshape one typical screen. Say a module explains how a building's fire alarm system routes a signal from a pull station to the monitoring company.
The original screen shows a stock photo of a fire truck, four bullet points of text, a small system diagram in the corner and a narrator reading the bullets. Background music plays under the narration.
Applying coherence, the photo and music go. Applying the multimedia and spatial contiguity principles, the system diagram becomes the main visual, enlarged, with each component labeled directly. Applying modality and redundancy, the bullets disappear and the narrator explains the path instead, in plain spoken language. Applying temporal contiguity and signaling, each component highlights as the narrator mentions it, and a moving line traces the signal's path. Applying segmenting, the explanation splits into three short steps with a "next" button. Applying pre-training, a previous screen introduces the four main components by name. Applying personalization, the script says "when you pull the station" instead of "upon activation of the manual pull station."
The new screen has fewer elements, less text and the same information. If you pilot both versions, watch for fewer replays and fewer wrong answers on follow-up questions with the new one. That is what the research predicts, and it is easy to check with your own learners.
How Do the Principles Apply to Video?
The principles apply to instructional video just as much as to slide-based eLearning, and video breaks them in its own ways.
Talking-head videos often violate the image principle when the speaker's face takes up the screen while they describe something learners can't see. Show the thing instead. Screen recordings often violate signaling when the cursor moves fast across a busy interface; zooms and highlights fix that. Long videos violate segmenting, so chapter markers or shorter clips help. Captions are a special case: they are essential for accessibility, and learners can turn them on or off, so they are not a redundancy problem in the way that permanent paragraph text on a slide can be.
Do These Principles Support "Visual Learners"?
No, and the distinction matters. The multimedia principle says that almost everyone learns better from relevant words and pictures together. It does not say that some people are "visual learners" who need pictures while others need text.
The idea that matching instruction to a preferred learning style improves results has been tested and has not held up. A 2008 review by Harold Pashler and colleagues found little sound evidence for it. Mayer's principles are about how human cognition works in general, so the same well-designed lesson helps nearly everyone.
When Do the Principles Matter Less?
Mayer describes boundary conditions, situations where an effect is weaker or disappears. A few are especially relevant to workplace training.
- Prior knowledge. Most effects are strongest for novices. Experienced learners can handle more complexity and may find extra support redundant.
- Pacing. Modality and segmenting matter most when content is fast or system-paced. When learners control the pace and can reread, the advantage shrinks.
- Simplicity. For very simple material, design choices have less impact because there is less to overload.
These conditions don't make the principles optional. They help you prioritize. For a dense technical lesson aimed at new hires, follow them closely. For a quick reference refresher aimed at experienced staff, you have more room.
Which Principles Should You Fix First?
If you are reviewing existing courses, I'd start with the principles that are both commonly broken and easy to fix:
- Coherence: remove decorative media and nice-to-know detail.
- Redundancy: stop narrating full paragraphs of on-screen text.
- Spatial contiguity: move labels and feedback next to what they describe.
- Segmenting: split long videos and animations into short, learner-paced pieces.
- Personalization: rewrite formal, passive scripts in a direct, conversational voice.
These five changes often take hours, not weeks, and they usually make a course both shorter and clearer. Conversational, relevant design also respects the people taking the course. That connects closely to adult learning principles, which focus on relevance, experience and respect for learners' time.
Better eLearning Starts with Twelve Small Decisions
Mayer's multimedia learning principles turn decades of experiments into practical rules. Cut what doesn't help, show structure, keep words and pictures together in space and time, break complex material into pieces, teach the parts first, narrate complex visuals and speak to learners like people.
The reasoning behind all twelve is the same: learners have limited capacity, so every design choice either spends it wisely or wastes it.
If you need to convince a stakeholder, run a small comparison. Revise one busy screen using the five fixes above, show both versions to a handful of new hires and ask each person to explain the process back to you. The difference is usually easy to hear, and it makes a stronger case for the principles than any summary of the research.