Dual Coding: Why Words Need Pictures
Dual coding says learning is stronger when words and visuals work together. Here is the research — and how Amistio thinks about explanation design.
Some ideas are awkward when they arrive as words alone. A grammar pattern, a feedback loop, a geometry proof, a data pipeline, a melody shape — you can describe each in sentences, but the learner often needs to see the relationships at the same time.
Dual coding gives a name to that intuition. Allan Paivio's theory argued that cognition can represent information verbally and nonverbally, and that learning can benefit when those representations reinforce each other instead of competing for attention.
What dual coding actually means
The point is not that every learner has a fixed visual or verbal style. The stronger claim is simpler: when a concept has both linguistic and spatial, visual, or relational structure, a well-matched visual can give memory another useful route. Clark and Paivio's review connected dual coding to education by emphasising coordinated verbal and imagery-based representations rather than one-size-fits-all learning styles.
Richard Mayer's multimedia-learning work adds an important constraint. Words and visuals help when they are aligned and reduce the work of building a mental model. They hurt when they add clutter, split attention, or ask the learner to reconcile mismatched signals.
How to use it without making a prettier mess
- Use diagrams for relationships: timelines, loops, hierarchies, maps, contrasts, and cause-and-effect chains.
- Keep labels close to the part of the diagram they explain, so attention does not bounce around the page.
- Remove decorative detail that does not carry meaning; it spends working memory without teaching.
- Ask the learner to explain the picture back in words, then redraw it from memory later.
The strongest use of dual coding is active. A learner should not only look at a diagram; they should translate between diagram and words. That translation is where understanding becomes visible.
How Amistio applies the principle
Amistio's design target is not a chat transcript with occasional decoration. When Ami teaches something with structure, the product should be able to surface a compact representation: a timeline for a history argument, a flow for an algorithm, a contrast table for grammar, or a sketch of the mental model behind a math step.
The same idea carries into homework. A good assignment can ask the learner to produce both forms: explain the process in words, then draw or outline the structure. If the two disagree, the mismatch reveals exactly where feedback is needed.
Sources
Every claim above is grounded in peer-reviewed research. Follow the links to the original papers.
- 1. Paivio (1991). Dual Coding Theory: Retrospect and Current Status. Canadian Journal of Psychology, 45(3), 255–287.https://doi.org/10.1037/h0084295
- 2. Clark & Paivio (1991). Dual Coding Theory and Education. Educational Psychology Review, 3, 149–210.https://doi.org/10.1007/BF01320076
- 3. Mayer & Moreno (2003). Nine Ways to Reduce Cognitive Load in Multimedia Learning. Educational Psychologist, 38(1), 43–52.https://doi.org/10.1207/S15326985EP3801_6