Chapter 7: Imagery
Memory and Imagery
Paivio’s Dual-Coding Theory
Paivio defines imagery as the ease with which something elicits a mental image, where the term ‘mental
image’ refers to experiences such as a mental picture or sound. Paivio’s approach is called the dual-
coding theory, which is the theory that verbal and non-verbal systems are alternative ways of
representing events. The relationship between the two systems is (fig 7.1):
(1) Incoming information can be either verbal or non-verbal.
(2) After being picked up by the sensory systems, information can be represented in either the
verbal or the non-verbal system.
(3) The units that comprise the verbal system are logogens, which are units that contain
information underlying our use of a word. Logogens operate sequentially: words are not all
present at once, they come one after the other.
(4) The units that make up our non-verbal system are called imagens, which contain information
that generates mental images. Imagens operate synchronously: the parts that they contain are
simultaneously available for inspection, which means that a variety of mental images can be
generated from imagens.
(5) The two systems have referential connections, which allow information in one system can give
rise to a process in the other system. In other words, a verbal description can elicit a mental
image, and a mental image can elicit a verbal description.
Concrete words (e.g., ‘table’) easily elicit a mental image, whereas abstract words (e.g., ‘purpose’) do
not. Concreteness refers to the degree to which a word refers to concrete objects, persons, places, or
things that can be heard, felt, smelled, or tasted. The notion that ideas have their origin in concrete
sensory experience has a long tradition in the history of Western thought.
In order to measure concreteness and imagery, participants rate their values on a 7-point scale. Low
imagery and high concreteness (and vice versa) are highly correlated, which led Paivio to argue that
imagery and concreteness measure two aspects of the same process.
There are some words, often referring to emotions, which are not concrete but still elicit vivid mental
imagery (e.g., ‘pain’, ‘love’). This demonstrates that in addition to external sources of imagery, there are
also internal, emotional sources of imagery.
Research Related to Dual-Coding Theory
In one of Paivio’s first experiments on the role of imagery in learning, he used a paired-associate
learning task (in which participants learned stimulus-response word pairs, then were presented with the
stimulus word and tested on recall rate of the response word) and divided the participants into four groups. Group 1 learned pairs in which both words were concrete, group 2 learned pairs in which the
first word was concrete and the second was abstract, group 3 learned pairs in which the first word was
abstract and the second was concrete, and group 4 learned pairs in which both words were abstract.
Results: The concrete/concrete group had the best results, and the abstract/abstract group had the
worst results. In addition, the biggest difference was between concrete and abstract stimuli (i.e., when
the first word was concrete the results were good, and when the first word was abstract the results
The explanation of these results, according to the dual-coding theory, is that concrete words can be
coded by either the verbal or non-verbal system, whereas abstract words can only be coded by the
verbal system due to their low imageability. It is easier to recall concrete words because they have the
advantage of having both codes available to memory.
Dual-Coding Theory and the Brain
Paivio argued that the verbal system is dependent on the left hemisphere of the brain, which controls
speech, and the non-verbal system is dependent on the right hemisphere. 2According to the dual-
coding theory, the more concrete a word is, the more imageable it is, and the more activity there should
be in the right hemisphere. This hypothesis has been challenged by fMRI research (Fiebach & Friedrici)
which showed that concrete words don’t elicit greater activity in the right hemisphere than abstract
words. Fiebach & Fredrici’s experiments showed that abstract and concrete words elicit different
patterns of activity in the left hemisphere, but concrete words don’t elicit greater activity in the right
Imagery and Mnemonics
Mnemonic techniques are procedures used to aid memory, and have been used since ancient times. A
second-century document called Ad Herennium described the method of loci that can be used to
remember a list many items in order. The method of loci involves learning places in a building (the loci)
and then associating an image for each item with one of the loci. Bizarre imagery facilitates recall.
Psychologists have shown that aspects of this technique really do aid memory. A central part of this
mnemonic technique is the process of relating images to particular locations. Asch found that items
interrelated to form units are easier to related than items that are not. The effectiveness of imagery as a
mnemonic may be partly due to its ability to organize disparate items into meaningful units. A mental
image also makes information distinctive, which makes it easier to recall.
Imagery and Distinctiveness
People remember bizarre items better when they occur along with common items. This is consistent
with the von Restorff effect, which holds that if one item in a set is different from the others, it will be
more likely to be recalled. A bizarre item is not distinctive in a list of bizarre items, so it will not be easier
Humour and Distinctiveness
Similar to the effect of bizarreness, humorous items are easier to recall only when they are surrounded
by non-humorous items. However, humorous items are more memorable than weird items, suggesting that the effect of bizarreness may be due to the fact that people often find bizarre items funny. Humour
by itself may be a strong memory aid, especially when humorous material is contrasted with neutral.
The Problem of Distinctiveness
If distinctiveness makes an item easier to remember, then why do we forget when we put something
special in a special place that others are unlikely to discover? Winograd & Soloway discovered that no
matter how memorable we think that a location will be, we will remember it less well if it is an unlikely
location than we will if it is a likely location. While distinctiveness is effective for remembering individual
items, it is not so effective for remembering the association between items.
The special places strategy is similar to the strategy that is often used to create passwords. Brown et al.
suggest using easily remembered (and therefore easily discoverable) passwords for most situations that
do not require security, and only creating distinctive passwords when necessary. You should test
yourself on a new password a few times over the following several days, and you should keep a written
record of your passwords in a secure, but not special, location.
The special places strategy and creating unique passwords involves a failure in metamemory (our beliefs
about how memory works). It is very important to have accurate beliefs about how memory works.
Synaesthesia and Eidetic Imagery
Synaesthesia is the power of a stimulus appropriate to one sense (e.g., a sound) to arouse an experience
appropriate to another sense (e.g., a colour). People who routinely have synesthetic experiences in
everyday life are called synaesthetes. The most common experience is chromaesthesia, coloured
hearing. The cue that elicits a synesthetic response is called an inducer, and the response is called the
concurrent. Synaesthesia runs in families and occurs more often in women than in men. It is possible
that synaesthesia improves memory.
Theories of Synaesthesia
The traditional explanation of synaesthesia is that it reveals the underlying unity of our senses, which
are believed to have evolved out of a single primordial sense. Building on this, Maurer believed that
newborns’ senses are not well differentiated, but have transient connections that diminish with time
through apoptosis (programmed neuron death). He advanced the idea that perhaps adult synaesthesia
is the failure of proper apoptosis.
Failure to weed out inter-sensory connections can’t be the whole story, however, because synaesthetic
responses can be elicited by concepts in addition to percepts experienced by our sensory organs. So the
experiences of synaesthetes, at least in some cases, could be mediated by neural connections that exist
in normal adult brains.
Strong and Weak Synaesthesia
Strong synaesthesia indicates the ‘classic’ instances, involving an inducer in one sensory modality and a
concurrent in another sensory modality. Weak synaesthesia indicates the experience of cross-modal
effects, which is the ability to appreciate that the sensations of one modality can be similar to those in another modality. Most of us are weak synaesthetes, since these cross-modal associations may develop
over childhood from experience with percepts and language. The fact that one sense can represent
information from another sense facilitates the use of figurative language such as metaphor.
Icon: the initial, brief representation of the information contained in a visual stimulus (a
snapshot of information).
Eidetic imagery: images projected onto the external world that persists for a minute or more
even after a stimulus, such as a picture, is removed.
o Icons diminish quickly but eidetic imagery can persist for a few minutes.
o Similar to synaesthesia as both are cognitive dedifferentiations: perceptual processes
that typically function independently are fused instead.
o Features of Eidetic Images:
the image is perceived as being located “out there” and