Dual coding theory
Also known as: Dual coding
Information encoded as both words and images is remembered better.
What it means
Dual coding theory proposes that cognition runs on two distinct but interconnected systems — a verbal system for language and a nonverbal imagery system for pictures and sensory information — and that material represented in both codes is more memorable than material in only one. Concrete words and pictures are advantaged because they readily evoke both a verbal label and a mental image, providing two independent retrieval routes. The theory explains the picture-superiority effect and the concreteness effect, and it grounds the instructional value of pairing diagrams with text. It has been refined and challenged by purely propositional and embodied accounts of representation. It remains a workhorse framework in memory and multimedia learning.
How the two systems connect
Paivio's model is not just two parallel stores but a specific wiring between them. The verbal system holds units he called logogens; the imagery system holds imagens. Three kinds of link do the work. Representational connections tie an incoming word or picture to its own code. Referential connections cross the divide, so a word can evoke its image and a picture its name. Associative connections run within each system, linking word to word and image to image. Memory improves under dual coding because encoding a concrete item lays down two traces that are functionally independent: if one is lost or inaccessible at retrieval, the other can still be found. The advantage is additive, not a single stronger trace, which is why it grows with the chance to form both codes.
What the evidence shows
The empirical spine is robust. Recognition memory for pictures is famously high: after a single brief exposure, accuracy stays well above chance across thousands of items, far outpacing words. In free recall, concrete words reliably beat abstract ones, and pictures beat their own labels, the pictorial superiority effect Nelson, Reed and Walling documented in 1976. These effects replicate widely and anchor much of instructional design. What is contested is the mechanism, not the phenomenon. Rival explanations attribute picture superiority to greater perceptual or conceptual distinctiveness rather than to a second code, and some recall data fit a depth-of-processing account as well as a dual-coding one. The pattern is real and useful; the claim that two separate codes cause it is one interpretation among several that the same data support.
The propositional challenge
The sharpest attack came from Zenon Pylyshyn, who argued in 1973 that mental images are not picture-like entities at all but abstract, propositional descriptions, redescribed in a single format shared with language. On this common-coding view the imagery system is theoretical surplus: one code, more parsimonious, and a better fit with the computational picture of mind. The concreteness effect drew a parallel rival in context-availability theory, which holds that concrete words are easier because they summon richer verbal context, not a distinct visual store. Neuroimaging has not settled it cleanly. Studies of concrete-word processing find both left-hemisphere verbal-associative activity and right parietal engagement consistent with imagery, suggesting the two accounts each capture part of the truth rather than one winning outright. Dual coding survives as a framework, not a proven architecture.
Using it in practice
The durable lesson is design, not decoration: pair words with a relevant image and let learners build both codes. This underpins Mayer's cognitive theory of multimedia learning and its modality principle, where a diagram plus spoken narration beats the same diagram with on-screen text. But the effect has failure modes that matter for anyone applying it. Adding pictures helps only when they carry information; decorative or redundant images raise extraneous load and can hurt. Presenting the same content simultaneously as narration and identical text produces a redundancy cost. And splitting attention between a diagram and a distant caption undoes the gain. The practical rule is to integrate meaningful verbal and visual information in space and time, so the two channels reinforce rather than compete for a limited working memory.
Examples
A vocabulary word learned with a vivid illustration is recalled better than the same word learned from text alone.
Airline safety cards pair every instruction with a picture, so passengers store a verbal rule and a mental image at once — two separate routes back to it when it is suddenly needed.
Flat-pack furniture steps that show a diagram beside the sentence are followed more reliably than a page of prose, because each step lands in words and in pictures together.
A medication label that shows a sun-and-moon pictogram beside 'take one in the morning and one at night' gives a patient both a picture and a phrase to recall.
In a pitch, saying 'revenue doubled' while showing a bar that visibly doubles lets an investor store the claim as a sentence and a shape, each recoverable on its own.
First described in Allan Paivio (1971).
Key references
- Jessen, F., Heun, R., Erb, M., Granath, D. O., Klose, U., Papassotiropoulos, A., & Grodd, W. (2000). The concreteness effect: Evidence for dual coding and context availability. Brain and Language, 74(1), 103-112. doi.org/10.1006/brln.2000.2340
- Clark, J. M., & Paivio, A. (1991). Dual coding theory and education. Educational Psychology Review, 3(3), 149-210. doi.org/10.1007/BF01320076
- Paivio, A. (1991). Dual coding theory: Retrospect and current status. Canadian Journal of Psychology, 45(3), 255-287. doi.org/10.1037/h0084295
- Nelson, D. L., Reed, V. S., & Walling, J. R. (1976). Pictorial superiority effect. Journal of Experimental Psychology: Human Learning and Memory, 2(5), 523-528. doi.org/10.1037/0278-7393.2.5.523
- Pylyshyn, Z. W. (1973). What the mind's eye tells the mind's brain: A critique of mental imagery. Psychological Bulletin, 80(1), 1-24. doi.org/10.1037/h0034650