Cog. Psyc. Chapter 7 Notes: Visual Images 11/7/2012 11:11:00 AM
On Intelligence Tests
Verbal knowledge: measured by vocab q’s or q’s that test
comprehension of written material.
Spatial knowledge: measured by performance of such operations
as mentally folding connected squares into a cube or mentally
rotating an object to see if it matches another object.
o Spatial tests can be difficult.
E.g. of the Dental Admissions Test
The study of Visual imagery was important when first originated, but there
was a gap of time that dismissed it as a whole because of Watson’s novel
Behaviorism which neglected the study of mental processes, and instead
only of behavior. 1960s – it picked up again!
VISUAL IMAGERY AND LEARNING
The use of visual imagery to learn material is a form of elaboration.
Memory for Pictures
Ppl usually better recognizing pictures than words.
Shepard’s experiment: subjects viewed 612 pictures at a self-paced rate and
were later given a recognition-memory test on pairs of pictures. Each pair
had a pic they saw and a new one they hadn’t; they had to identify which
they HAD seen. Almost all succeeded.
Other participants had same task but with words; Theyre accuracy
(87% of subjects) was low but yet equivalent to the Picture group
after a one week delay!
Standing’s experiment: one group viewed 10,000 pics over a 5-day period.
Then given test like Shepard’s. Standing estimated they;d remember 6600.
Memory for details task : The participants could remember pretty well the
details. When it was a category theyd seen vs new category (nearly perfect
accuracy)… When it was a category of same kind.. still pretty well at 75%. Paivio’s Dual-Coding Theory
2 ways a person can elaborate on material in a learning experiment
1) verbal associations
2) visual imagery
argued that the concrete-abstract dimension is the most important
determinant of ease in forming an image. (concrete easy, abstract difficult)
A literal picture though result in best memory, followed by a concrete word,
then abstract word.
The imagery potential of words (ease with which a concept can be imaged)
is usually measured by asking ppl to rate on scale how easy it is to form
image for given word.
The association value of a word = the number of verbal associations
generated for a concept.
Paivio: The imagery potential is better in predicting learning than the
High-imagery words are easier to learn than low-imagery words,
but high-association words are not necessarily easier to learn than
Paivio’s study; 16 pairs of words given to students at UWO. High and low
imagery words. 16/4 of (H-H), (H-L), (L-H), (L-L), and in that order was
H-H had best recall cuz of the finding that interactive pictures
improve recall the best. Students reported what learning strategy
they used; Imagery was highest at H-H and lowest at L-L..
indicating that the reported use of imagery suggests that imagery is
an effective learning strategy.
Paivio’s theory is called a dual-coding theory cuz it proposes TWO
INDEPENDENT MEMORY CODES, either of which can result in recall. Memory is best when items can be rep’d by both verbal and visual memory codes.
Independent in the sense that a person can forget one code without
forgetting the other.
Marshark and Hunt’s Criticism of the DC theory – that is only works in
situations where ppl focus on relational information (information
specifying how concepts are related). Interactive images are at the height of
relational information. Perhaps M and H are right, that relational processing
is necessary to achieve the benefits of concreteness – then dual-coding
theory has a restricted range of application, but even still it is a large range
and is useful in memory.
e.g. The Memory Book by Lorayne and Lucas.
Memory books usually emphasize visual imagery.
Psychology experiments are better for supportive data than memory books,
for improving memory.
How can one learn a foreign word with visual imagery?
ATKINSON and RAUGH:
An associated word can be used to form a link between foreign
word and English translation. This associated word is called a
The keyword method divides the study of a vocabulary word into
two stages. The first stage is to associate the foreign word with an
English keyword, which sounds like some part of the foreign word.
Then form a mental image of the keyword interacting with the
English translation. E.g. Russian word for building (zdanie) is
pronounced like zdawnyeh, with emphasis on the first syllable so
we can use :dawn: as the keyword, and imagine a pink light of
dawn hitting the windows of a building.
A good keyword should:
o 1) sound as much as possible like a part of the foreign word
o 2) be different from the other keywords o 3) easily form an interactive image with the English
Atkinson and Raugh had a study with controls and a group with
keywords to use. Keyword group memorized 72%; Controls 46%;
Impressive also cuz Russian is considerably diff of a pronunciation
Visual imagery is also key in recognizing faces.
E.g. if Mr. Gordon has a large nose, the image might be a garden (sounds
like Gordon) growing out of his nose. In a study, 10 names randomly
assigned to 10 random faces. Visual imagery group got all 10, controls only
EVIDENCE FOR IMAGES IN PERFORMING COGNITIVE TASKS
Pylyshyn’s paper challenged the usefulness of images in psychology.
Argued that it was misleading to think of images as uninterpreted
photographs, analogous to pictures in the head. Rather, supported
the alternative view that an image is much closer to being a
DESCRIPTION of a scene, then a picture of it.
Propositional theory: a theory that all knowledge, including spatial
knowledge can be expressed in semantic-based propositions.
The emphasis on the descriptive characteristics of images, rather
then sensory, is the central theme of a propositional theory.
Kosslyn and Pomeranz summarized five experimental findings that they
thought could be better explained by imagery than by nonsensory images.
(scanning visual images, ….interference). They are all listed below.
Scanning Visual Images
Pictures and images have many shared operations
E.g. visual scanning – a shift of attention across a visual display
or image. o The time it takes to scan b/t 2 objects in an image should be
a function of their distance from each other.
Subjects saw a map of different locations of objects. Then they
were given name of an object, so had to picture it on the map.
Then given name of second object and had to picture a line going
from object A – B.
o Reaction time a linear function of the distance between two
o However, perhaps subjects did not actually mentally scan
their visual images but simply waited longer before pushing
the button as the distance increased between two objects.
o This criticism can be avoided if the outcome of the experiment
CANNOT be predicted
E.g. if shown straight line and spiral, ppl were
unsuccessful in predicting how the different shapes
would influence their scanning time. Cuz they couldn’t
predict the outcome of the experiment, their scanning
times must have been produced by their actually
scanning the different patterns rather than by their
Finke, along with other psychologists though agree that it is highly unlikely
that ppl could perform many spatial tasks w/o using imagery. Finke goes on
to say that some tasks in which imagery is used, there is a not-so-obvious
expected outcome, usually cuz person doing the task for the first time.
Sequential versus Parallel Processing.
Language is sequential (verbal codes) whereas visual images are
not, and thus a visual image makes it possible to match info in
parallel (like perceiving many features of faces simultaneously)
o Describing a face in words depends on order (sequences
basically) of features you wish to list. Parallel representation of spatial info and the sequential
representation of verbal info influence how fast a person can determine
whether a perceived pattern matches one from memory.
If stored visually, match occurs fast; if verbally = more slowly as a
function of how many features to be compared to memory.
Nielsen and Smith –showed students either a schematic face or its verbal
5 features of face, varied in size (s,m,l)
After they studied either the description or the picture for 4 secs,
the stim was removed. Then experimenters pres a test face, and
students had to decide if matched the face or description presented
o Response times show that matching was unaffected by # of
items, but ONLY if previously shown visual pattern.
Implies that when a person can maintain a visual image
of a pattern in STM, a second visual pattern can be
compared with it very quickly.
Maintaining an image in the group that initially saw a
visual avoids a list of separate items by combining the
individual features on the list into a single integrated
Its more difficult to determine if two patterns are the same, if their
orientations are different.
In Shepard and Metzler’s study, it showed that the time required to
decide that two patterns were identical increased linearly with an
increase in the # of degrees they differed in orientation (between 0
degrees and 180 were of all the objects)
o Suggests subjects were rotating a visual image of one of the
forms until it had the same orientation as the other form.
Subjects even said aloud this was how they were deciding if a
match worked or not, and that they could rotate an image
only up to a certain speed without losing its structure. Mental animations are also useful for solving problems
E.g. with the pulley, and determining if it moves clock-or counter
clockwise based on the way the rope moves through it when pulled.
It creates a “causal chain” of events beginning with first pulley,
second pulley, then the last pulley.
o Note: not everyone decides to use mental animations, though
they are useful.
o Hegarty asked students to quickly respond T or F to a
statement beside the diagram. Some (static) statements
didn’t mention movem