We’ve all sat through this presentation before. A slide appears, dense with bullet points. The presenter turns to the screen and reads them out, word for word. You try to both read and listen and end up doing neither.
Naturally, by the third slide you have stopped absorbing anything at all. It even has a name: death by PowerPoint.
Here’s the frustrating part. That slide was almost certainly made by someone trying to help you learn. They added the text so you wouldn’t miss anything. They read it aloud to reinforce it. Every instinct they followed was reasonable, and every one of them made the learning worse.
This is the problem that multimedia design quietly runs into. Our intuitions about how to combine words and pictures are often wrong. Good intentions can produce cluttered, exhausting content that the brain cannot process.
Fortunately, help is at hand. Over more than 30 years and 200-plus experiments, the psychologist Richard Mayer worked out what actually helps people learn from words and images, and turned it into a set of clear, tested principles.
Get them right, and your content will teach effectively. Get them wrong, and you’ll produce yet another slide nobody remembers. This article is about the difference.
What is the Cognitive Theory of Multimedia Learning?
The cognitive theory of multimedia learning, developed by Richard Mayer, explains how people learn from words and pictures combined. Its central claim is deceptively simple. Mayer calls it the multimedia principle:
People learn better from words and pictures than from words alone.
A diagram with clear explanations beats either on its own. The theory rests on three things we already know about how the brain handles information, and each one is covered in depth elsewhere in this series.
- Two Channels: The brain processes visual and verbal information through separate channels, a picture through one, spoken words through another. This is the basis of dual coding, and it’s why pairing a visual with a verbal explanation works. You’re using two channels rather than overloading one.
- Limited Capacity: Each channel can only handle so much information at once. Overload either one and learning stalls. This is the heart of cognitive load theory. Good multimedia design is largely the art of not exceeding that limit.
- Active Processing: Learning is not passive consumption. The brain has to select what matters, organise it into a coherent structure, and connect it to what it already knows. Content has to be designed to support that work, not bury it.
From these three foundations, Mayer and his colleagues derived a set of design principles, tested across more than 200 experiments. They are not style preferences or rules of thumb. They’re evidence-based findings about what helps the brain turn a presentation into true understanding.
What Are Mayer’s 15 Principles of Multimedia Learning?
So, what are these 15 evidence-based principles for designing multimedia? Here’s the complete set.
| Principle | What It Means |
|---|---|
| Coherence | Learning improves when unnecessary material, extra detail, decoration, and background music are removed. |
| Signalling | Cues that highlight the key information and structure help learners focus. |
| Redundancy | Graphics with narration beat graphics with narration plus identical on-screen text. |
| Spatial contiguity | Related words and pictures work best placed close together, not far apart. |
| Temporal contiguity | Related words and pictures work best presented at the same time, not one after another. |
| Segmenting | Break content into learner-paced chunks rather than one continuous stream. |
| Pre-training | Teach the key terms and concepts before the main lesson. |
| Modality | Graphics with spoken narration beat graphics with on-screen text. |
| Multimedia | People learn better from words and pictures than from words alone. |
| Personalisation | Conversational language works better than formal, impersonal wording. |
| Voice | A friendly human voice works better than a machine voice. |
| Image | Adding a static image of the speaker to the screen does not necessarily help. |
| Embodiment | An on-screen presenter who uses natural gestures and movement aids learning. |
| Immersion | Learners do not necessarily learn better in 3D virtual reality than from a good 2D presentation. |
| Generative activity | Prompting learners to summarise, explain, or organise as they go deepens learning. |
The Three Jobs of Good Multimedia Design
Mayer’s 15 principles are easier to hold onto if you group them by what they are for. Each is doing one of three jobs: cutting out what wastes mental effort, managing the effort the material genuinely demands, and encouraging the effort that builds understanding.
1. Cut the Clutter
Because each channel has limited capacity, anything that consumes it without aiding learning is a tax on the learner. Here are some intuitive ways to cut that waste:
- Signalling: Highlight what matters. Use headings, arrows, bold key terms, or a highlighted region of a diagram to guide attention to the important parts. That way your learners won’t burn effort working out where to look.
- Contiguity: Keep related things together. Put a label beside the part of the diagram it describes rather than in a key below it, and sync narration to the visual it explains rather than showing one and then the other. Splitting them makes learners hold one thing in mind while hunting for the other.
The most powerful clutter-cutting principles, though, are also the most counterintuitive. We’ll cover those in the next section.
2. Manage the Load
Some effort is unavoidable, especially when the material itself is demanding. However, there are always steps we can take to manage this load. You can start by following these principles:
- Segmenting: Break content into learner-paced chunks. A long video split into short segments, each advanced by the learner, lets the brain process one piece before the next arrives instead of being swept along.
- Pre-training: Teach the key terms and parts first. If learners already know what the components are called before you explain how they work together, they’ll have spare capacity to follow the explanation.
- Modality: Where you can, narrate a visual rather than captioning it. A diagram with spoken explanation spreads the load across both channels. On the other hand, a diagram with dense on-screen text forces the picture and the words through the visual channel all at once.
3. Drive Engagement
This last group encourages the active processing that turns exposure into understanding. The key is to step away from passive consumption and towards doing something with the material.
- Personalisation & Voice: Use a conversational tone and a warm, human voice rather than formal, robotic delivery. People work harder to understand someone who seems to be speaking to them.
- Generative Activity: Prompt learners to do something with the material as they go. Ask them to explain it, summarise it, sketch it. This builds the generation effect into your multimedia. Producing always beats receiving.
Grouped this way, the principles stop being a list to memorise and become a set of questions to ask of any content: what can I cut, how do I keep the essential load manageable, and how do I get the learner to do something with the information?
The Principles People Get Wrong
Most of Mayer’s principles confirm what good instinct already suggests. Two do the opposite. They’re worth dwelling on, because they are where well-meaning design most often goes wrong.
1. Redundancy: Stop Reading Your Slides Aloud
The instinct is that showing text and narrating the same words at the same time reinforces the message. It does the reverse. When on-screen text duplicates what the voice is saying, the brain tries to read and listen in parallel. Reconciling the two activities clogs the very channel you’re trying to help.
This is death by PowerPoint in a single principle. The packed slide read out word for word feels like a conscientious choice, yet learners take away less from it than from a near-empty slide with a spoken explanation.
2. Coherence: Less is More
It’s often hard to resist adding an additional flourish to your content to make it more engaging. This can include:
- Background music
- Decorative images
- Irrelevant anecdotes
- Interesting but inessential facts
The problem is that these seductive details compete for the same limited capacity as the actual lesson. Mayer’s research found that stripping these out consistently improved learning. The material people enjoy most is not the material they learn most from.
Of course, this may be an uncomfortable finding for anyone who has ever added a stock photo or a music bed to their content to “liven things up”.
What unites both these principles is a single, counterintuitive truth. In multimedia, addition usually subtracts. The urge to add more, be it more text, more visuals, more atmosphere, is the urge to help. But this urge spends capacity the learner needs for understanding.
Often the most effective thing a designer can do is take things away.
Do Mayer’s Principles Actually Work?
It’s fair to ask whether a set of findings from psychology labs holds up in the real world. On the whole, they do. In fact, Mayer’s principles are among the most thoroughly tested ideas in all of learning research, and independent reviews have borne them out.
One large synthesis pooled 29 reviews covering more than a thousand studies and 78,000 learners. It found that most of the design principles produced significant, positive effects on learning, with the biggest gains coming from signalling and from keeping related elements together.
But “it works” is not the full story. These principles do come with a condition. They help most for novices, but they can fade or even reverse for experts. This is known as the expertise reversal effect.
The explanations, cues, and hand-holding that rescue a beginner can become clutter for someone who already knows the material. A 2025 meta-analysis of Mayer’s work confirmed this pattern. The principles are real, but their strength depends on the learner and the medium.
For workplace learning, this is the difference between using the principles well and using them blindly. For instance, the redundancy principle says cut the on-screen text for a novice, but an expert skimming a refresher might genuinely prefer it.
Likewise, signalling helps someone new to a topic and can patronise someone fluent in it. The rule is not “apply all 15 principles always”. It’s “design for the learner in front of you”. This means acknowledging their level of prior knowledge above all else.
Mayer’s Principles in the Age of AI
Something has changed since Mayer began this work in 1989. Training content is no longer made only by people. AI now generates slides, videos, voiceovers, and even on-screen avatars in seconds. It’s fast becoming the default way of producing learning content at scale.
That makes these principles more important, not less. A human designer applies Mayer’s rules slowly, one slide at a time. An AI tool applies whatever rules it has been given to thousands of assets at once. This means it can build good design into content at a scale no team could match.
Of course, it can also do the opposite and mass-produce the very mistakes this article warns against.
The early evidence is encouraging. AI-generated content works when it follows the principles. A 2025 study comparing AI-generated and human-made instructional videos found that an AI voice paired with an AI avatar actually reduced extraneous cognitive load, and did no harm to motivation. This eases a common worry about machine-made training.
As making content gets automated, the value shifts to knowing what good content looks like, so you can direct the tools, judge their output, and catch the mistakes they make at speed. Mayer’s principles are exactly that yardstick.
Putting the Principles to Work
None of this is useful unless it changes what you build. Here’s the whole set as a working checklist. Keep this in mind whether you’re making a slide deck, an eLearning module, or a video:
- Add a meaningful visual to support words, not a decorative one.
- Narrate visuals rather than captioning them.
- Never put text on screen and read it aloud word for word.
- Cut background music, decorative images, and interesting-but-irrelevant details.
- Signal what matters with headings, highlighting, and cues.
- Break content into short, learner-paced chunks.
- Teach the key terms before the main explanation.
- Keep related words and pictures close together and in sync.
- Write and speak in a warm conversational voice.
- Prompt learners to explain, summarise, or apply as they go.
You don’t have to memorise every principle. If you hold on to the three jobs behind them (cutting wasted effort, managing the effort material demands, and prompting the effort that builds understanding), you can work out the rest.
Final Words
Everything in this article comes back to a single shift in thinking. Good multimedia design is not about adding more. It’s about respecting the limits of the brain that has to process it. The mind has two channels, each of which gets easily overwhelmed. Content either works with that reality or fights it.
This is what neurogogy means in practice: designing learning around how the brain actually takes in information, rather than around what looks impressive or feels thorough.
A brain that processes words and images through separate, limited channels needs content that pairs them well, cuts the clutter, manages the load, and prompts real thinking. It doesn’t need a slide crammed with everything the designer could think to include.
Remember, the goal isn’t to impress your learners. It’s to help them learn. Design for how the brain actually takes in information, and you’ll dazzle them less and teach them more.
Ready to build learning that lasts?
- Book a demo to see how the Impact Suite turns the science of multimedia into training that sticks.
- Explore the Impact Suite and the neuroscience behind every feature.
- Download the Learning Theories guidebook for the full picture of the science that makes learning stick.
