Dual coding theory (DCT), proposed by Allan Paivio in 1971, holds that human cognition operates through two distinct but interconnected coding systems: a verbal system that processes and represents linguistic information, and a nonverbal (imagery) system that processes and represents perceptual and spatial information. Information encoded in both systems — dual coded — is better remembered than information encoded in only one system, because it has twice as many retrieval routes.
The Two Systems
The verbal system processes language in sequential, propositional units called logogens. The imagery system processes sensory-perceptual information in holistic, analog units called imagens. The two systems are functionally independent (they can operate in parallel) but interconnected (verbal information can evoke images, and images can be described in words). Concrete words like "elephant" easily activate both systems, while abstract words like "justice" primarily activate the verbal system.
Abstract words (single coded): verbal only → weaker memory
Pictures: imagery + verbal label → best memory ("picture superiority effect")
The Concreteness Effect
One of the strongest predictions of DCT is the concreteness effect: concrete, imageable words (e.g., "apple," "mountain") are remembered better than abstract words (e.g., "truth," "liberty") because they are represented in both coding systems. This effect is one of the most robust findings in verbal learning research, with effect sizes typically large. The concreteness effect interacts with other variables: it is larger in free recall than in recognition, and it is enhanced by imagery instructions.
Pictures are remembered much better than words — the picture superiority effect. A single brief exposure to a picture can produce recognition accuracy above 90% even after thousands of intervening items (Standing, 1973, showed 90% recognition for 10,000 pictures). DCT explains this because pictures are automatically dual coded: they directly activate the imagery system and are spontaneously labeled verbally. The picture superiority effect has important implications for education and communication — visual representations enhance learning.
Applications to Education
DCT has influenced instructional design through the multimedia learning principle: learning is enhanced when information is presented in both verbal and visual formats. Richard Mayer's cognitive theory of multimedia learning draws directly on DCT, proposing that meaningful learning occurs when learners build connections between verbal and visual representations. However, the benefits of dual coding depend on the relationship between the verbal and visual information — adding irrelevant images can actually impair learning by increasing extraneous cognitive load.
Critiques and Alternatives
DCT has been challenged by propositional theories (Pylyshyn, Anderson) that argue for a single, abstract representational format (propositions) underlying both verbal and imagery processes. The imagery debate — whether mental images are analog (picture-like) representations or redescribed in propositional format — was one of the most contentious debates in cognitive science. Neuroimaging evidence now largely supports the existence of distinct verbal and spatial/imagery systems, with visual imagery activating many of the same brain areas as actual visual perception.