Gerald Ajam

Media and multimedia for learning

This is the most settled of the domains, and the most widely ignored. We have known for decades how to combine words and pictures without overloading a reader. Most educational media still does the opposite, usually in the name of engagement.

Status Established

Published August 2026Revised September 2026

All work

Why it matters

Attention is finite and roughly fixed. Every element in a piece of media either contributes to the explanation or competes with it, and the competing ones do not politely stand aside. Decorative additions made in good faith — a background track, an energetic transition, a stock image chosen because the slide looked bare — take capacity away from the thing they were meant to support.

The cost is invisible in production and obvious in comprehension, which is exactly the wrong way round for anyone trying to make a case for restraint.

What I have come to believe

Cut anything that is not doing explanatory work, including material that is accurate and interesting. Interesting-but-tangential detail is the most expensive kind, because it is memorable enough to crowd out the point it was decorating. This is the hardest edit to make and the one that most improves the result.

Put labels where the thing is, rather than in a legend the reader has to hold in memory while looking somewhere else. Splitting related information across a gap converts an explanation into a matching exercise.

Do not narrate text that is already on screen. Reading and listening to the same words at the same time is worse than either alone, which is counterintuitive enough that it gets designed in repeatedly. Narration paired with a visual works; narration paired with a transcript of itself does not.

And signal the structure. Headings, ordering, and consistent visual grammar tell a reader what kind of thing they are looking at and what to do with it. Most media that feels hard to follow is not conceptually difficult — it is unsignposted.

How I approach it

I treat every piece of media as making a claim about what matters, whether or not it intends to. Size, position, colour, and motion all assert importance. When those assertions disagree with the actual argument, the reader believes the design, not the text.

So the first question is what the argument is, and the second is whether the visual hierarchy is telling the same story. If a diagram cannot be drawn, the framework behind it is usually not finished.

In short

The position in five claims

The argument above, cut down to what travels.

  • 01

    Everything on screen competes

    An element that is not explaining is taking attention from the one that is.

  • 02

    Cut the interesting tangent

    Accurate, engaging detail that is off the point is the most expensive kind. It is memorable enough to crowd the point out.

  • 03

    Label in place

    Put words where the thing is. Splitting them across a gap turns an explanation into a matching exercise.

  • 04

    Narrate the picture, never the text

    Reading and hearing the same words at once is worse than either alone.

  • 05

    Hierarchy is a claim

    Size, position, colour and motion assert importance. When they disagree with the argument, readers believe the design.

In practice

The questions I ask

Before designing anything in this domain, and again when something is not working.

  1. What is the argument, in one sentence?
  2. Which element is the most prominent, and is it the most important?
  3. What here is decoration, and what would the explanation lose without it?
  4. Where does the reader have to hold something in memory while looking somewhere else?
  5. Is any narration repeating text that is already on screen?
  6. Could someone tell what kind of thing this is, and what to do with it, from its structure alone?

Most educational media is not too difficult. It is too busy, and the fix is to take things away.

Unsettled

What I would still argue about

The honest list. These are open in my own head, not rhetorical.

  • How far the classic findings generalise to learners who choose their own pace and route, rather than sitting through a fixed presentation.
  • Whether some redundancy is worth its cost for learners with weaker language proficiency, where the theory says remove it and practice keeps suggesting otherwise.
  • Where legitimate motivational design ends and the seductive kind begins, given that the same techniques serve both.
Singapore