The dual coding hypothesis: why words plus pictures help you remember

What the dual coding hypothesis says

The dual coding hypothesis is the idea that information you take in both as words and as images is easier to remember than information stored in only one form. It comes from dual coding theory, developed by the Canadian psychologist Allan Paivio from the 1960s onwards and set out in his 1971 book Imagery and Verbal Processes. For students, it is the theory behind a simple habit: put a picture next to your words, and words next to your pictures.

Paivio proposed that the mind works with two separate but connected systems. A verbal system handles language: words, sentences, the sound and spelling of terms. A non-verbal system handles images and other sensory impressions: what something looks like, its shape, where things sit in relation to each other. Each system has its own units – Paivio called them logogens and imagens – and the two can activate each other.

The key point is that connection. When you read the word 'bicycle', you may also call up an image of one; when you see a bicycle, you can name it. Paivio described three kinds of processing: representational (a word activates its verbal code, a picture its visual code), referential (a word calls up an image, or an image calls up a word) and associative (one word leads to other words, one image to other images).

The evidence behind dual coding theory

Dual coding theory was built to explain findings from memory experiments. One is the concreteness effect: concrete words such as 'apple' or 'bridge' are usually remembered better than abstract ones such as 'justice' or 'theory'. Paivio's explanation was that concrete words are easy to picture, so they are likely to be stored in both systems, while abstract words mostly have only a verbal code. Two codes mean two routes back to the memory.

Another is the picture superiority effect: pictures tend to be remembered better than their names. A picture is easy to name, so it picks up a verbal code as well as a visual one, whereas a word does not automatically call up an image.

Like any influential theory, dual coding has critics. Some researchers argue that words and images are ultimately stored in a more abstract, meaning-based form; others explain the concreteness effect differently, for example by how easily a word brings a context to mind. Later work, including Richard Mayer's cognitive theory of multimedia learning, built on dual coding to study how people learn from text and graphics together. You do not need to settle the debate to use the practical lesson, which holds up well: combining words with meaningful visuals usually helps learning.

The common mix-up: dual coding is not learning styles

Dual coding is often confused with learning styles – the claim that some people are visual learners and others verbal or auditory learners, and that each person learns best when teaching matches their type. These are very different claims.

Dual coding theory is about everyone. It says the same human memory has a verbal and a non-verbal system, and that using both helps people in general. It does not sort students into types. Learning styles, in the version that says matching teaching to a preferred style improves results, have not held up well when tested. A widely cited 2008 review in the journal Psychological Science in the Public Interest found little solid evidence for that matching idea.

So the takeaway is not 'I am a visual learner, so I will only draw'. It is 'whatever I prefer, I will remember this better with a clear explanation in words and a meaningful visual of it'. People do have preferences. They are just not a good reason to leave out one of the two channels.

How to use the dual coding hypothesis when you study

The practical rule is simple: for each important idea, have a version in words and a version as a picture, and make sure they explain each other. Whatever format you use, the visual has to carry meaning. A stock photo of a brain at the top of your neuroscience notes is decoration, not dual coding; interesting but irrelevant extras, sometimes called seductive details, can distract more than they help. These formats work well at university:

• Diagrams of processes and structures: a labelled sketch of a cell, a circuit, a supply-and-demand graph, a flow chart of a legal test. Draw it, then explain it in words next to it.
• Timelines: for history, the development of a theory or the steps of a procedure. Position on the line shows order and spacing in a way a list cannot.
• Mind maps: a central topic with branches for subtopics. Useful for seeing how the parts of a chapter relate, as long as the branches carry real content rather than single vague words.
• Comparison tables and side-by-side sketches: for concepts that are easily confused, such as mitosis and meiosis, or two competing models in economics. The layout on the page makes the contrast visible.
• Infographics and one-page summaries: a whole lecture on one sheet, combining short text, arrows, icons and small diagrams.

Two worked examples

Example one: a fictional biological psychology lecture on the action potential. In words: at rest, the inside of a neuron is negative compared with the outside, at around minus 70 millivolts. When the threshold is reached, sodium channels open and sodium flows in, so the membrane depolarises. Then potassium channels open and potassium flows out, repolarising the membrane; it briefly overshoots into hyperpolarisation before returning to the resting potential. During the refractory period, a new action potential is harder or impossible to trigger.

Dual coded: you draw the classic curve of membrane potential against time and label each phase directly on the curve – threshold, depolarisation, repolarisation, hyperpolarisation – with small arrows showing which ions move in which direction. Then you cover the words and explain the curve out loud, and cover the curve and sketch it from the words. Each version checks the other.

Example two: a history seminar on the Weimar Republic. A list of dates is easy to read and hard to remember, so put them on a timeline instead: the November Revolution in 1918, the Weimar Constitution in 1919, hyperinflation and the Munich putsch in 1923, the stock market crash in 1929, Hitler's appointment as chancellor in January 1933. Spacing the line by years makes the comparatively stable middle years visible at a glance. Add one short sentence under each event about why it mattered, and you have both codes on a single page.

Making dual coding active, not decorative

The biggest gain comes when you produce the second code yourself instead of only looking at someone else's diagram. Drawing a diagram from memory, turning a paragraph into a flow chart or explaining a figure in your own words all force you to retrieve and reorganise what you know. Combine dual coding with testing yourself: sketch the mind map of last week's lecture from memory, then compare it with your notes and add what you missed in a different colour.

A starting point helps when the material is dense. In NotizAI, every note can be turned into a mind map and a one-page infographic alongside the written study sheet, so you have a word version and a visual version of the same lecture to work from. Use them as a model to check against: look at the mind map, close it, redraw it on paper, then compare. A few more habits keep dual coding useful:

• Keep words and pictures close together. Labels belong on the diagram, not in a legend at the bottom of the page.
• Keep visuals simple. Clean lines and a few labels beat a beautiful, overloaded drawing.
• Work in both directions: from words to picture, and from picture back to words.
• Do not try to dual code everything. Save it for processes, structures, sequences and comparisons, where a picture really adds something.