Richard Mayer's multimedia principles come from decades of experiments on how people learn from words and pictures together. The short version: working memory is small, and every screen either spends it on the content or wastes it on everything else. This checklist turns the twelve principles into things you can look for on a real slide, and what to do when you find them.
The principles fall into three groups. The first five cut out extraneous processing, the effort learners spend on things that aren't the lesson. The next three help learners manage essential processing when the material itself is hard. The last four encourage generative processing, the effort of actually making sense of it.
Cut the noise: reduce extraneous processing
1. Coherence
People learn better when extra material is left out. Interesting-but-irrelevant stories, decorative images, background music, and "fun facts" all compete with the content.
Look for: stock photos that don't explain anything, anecdotes that don't serve an objective, music under narration. Fix: cut it, or replace a decorative image with one that shows the idea.
2. Signaling
People learn better when cues highlight what matters and how it's organized: headings, outlines, arrows, bolding the key term, a narrator saying "there are three steps".
Look for: a dense diagram with no callouts, a screen where everything has the same visual weight. Fix: add headings, number the steps, point at the part the narration is describing.
3. Redundancy
People learn better from graphics and narration than from graphics, narration, and the same words printed on screen. Reading and listening to identical text at once splits attention.
Look for: narration that reads a paragraph of on-screen text word for word over a graphic. Fix: keep a short key phrase on screen and let the narration carry the explanation.
This is not an argument against captions. Captions and transcripts are an accessibility requirement (see the WCAG checklist). The principle is about designing screens that duplicate the narration in blocks of text, not about removing access. Give learners captions they can turn on.
4. Spatial contiguity
People learn better when words sit next to the part of the picture they describe, not in a legend or on the next screen.
Look for: a numbered diagram with the explanations in a list below it. Fix: put the labels on the diagram.
5. Temporal contiguity
People learn better when narration and the matching animation happen at the same time, not one after the other.
Look for: a video that plays, then a narrator explaining what just happened. Fix: sync the explanation to the moment it describes.
Manage hard material: essential processing
6. Segmenting
People learn better when a complex lesson is broken into learner-paced chunks instead of one continuous stream.
Look for: a long video or a screen that scrolls forever, with no pause points. Fix: split it into short segments with a "continue" the learner controls.
7. Pre-training
People learn better when they already know the names and characteristics of the key concepts before the main lesson.
Look for: a process explanation that introduces six new terms mid-stream. Fix: introduce the parts first, briefly, then show how they work together.
8. Modality
People learn better from graphics with spoken words than from graphics with printed words, especially when the pace is fast and the material is complex.
Look for: an animated process explained entirely in on-screen text. Fix: narrate it, and keep text for labels and key terms. (Again, with captions available.)
Help them make sense of it: generative processing
9. Multimedia
People learn better from words and pictures than from words alone. The picture has to do explanatory work, not decorate.
Look for: a text-only explanation of a process, a layout, or a relationship. Fix: add a diagram, a labeled screenshot, or a simple chart that shows it.
10. Personalization
People learn better from a conversational style than a formal one. "You" and "your" instead of "the learner" and "one".
Look for: "Employees shall ensure that data is classified appropriately." Fix: "Before you save a file, check which of the four labels it needs."
11. Voice
People learned better from a natural human voice than from the machine voices used in the original studies. Text-to-speech has improved a lot since then, so treat this as: use the most natural, friendly voice you can, and never one that's hard to listen to.
12. Image
Adding a static image of the narrator to the screen doesn't reliably improve learning. A talking head earns its place when it adds something, like modeling a skill, not by default.
The one-screen checklist
- Everything on the screen serves an objective. Decoration is gone.
- The most important thing on the screen is visually the most prominent.
- On-screen text doesn't duplicate the narration in full, and captions are available.
- Labels sit on the graphic they describe.
- Narration and animation happen together.
- Complex content is chunked, and the learner controls the pace.
- Key terms are introduced before they're used.
- Processes and relationships are shown, not only described.
- The tone talks to the learner, not about them.
Checking a whole course
Going screen by screen with this list is thorough and slow. Pedagrade reviews a whole course against Mayer's principles along with the rest of the learning science and accessibility, in about a minute, and tells you which screens have the problems and how to fix each one.
See what it finds in your course.
Paste a course or upload the file. In about a minute you get a scored report with every finding cited and the fix written for you. Your first audit is free.
Start your free audit