Chapter 8

Visual/mental imagery is a mental representation of a non-present object or event.
Mnemonic techniques to improve memory (the first 3 use visual imagery).
1. Method of Loci - use mental imagery to picture some route or place with which you are very familiar, and as you mentally move through that environment you see the information or items that you need to remember. Principles to improve method of Loci (Box 8-1, p. 275).
2. Interacting Images - interacting images are the most memorable.
3. Pegword method - learn a rhyme, then associate one to-be-remembered piece of information with each line of the rhyme using imagery (preferably interactive images).
4. A technique that involves recoding the material to be remembered in some way. So, that extra words or sentences are added to mediate (or connect) the to be remembered material to information already in your memory (e.g., taking the first letter of each word that you need to remember and making a word out of them, such as "HOMES" being used to remember the names of the Great Lakes).
5. Categorizing and organizing material to improve recall.
Three theories regarding visual imagery (should study the experiments for/against each theory)
1. dual-coding hypothesis postulates that information in our long term memory can be stored or coded in two different ways: (a) as verbal information - containing information about an item's abstract, linguistic meaning; or (b) as visual imagery - containing mental pictures of some sort that represent what the item looks like. In addition, imagery involves a richer more complex coding of information (than verbal coding). Paivio found that the more concrete the word, the richer the mental/visual image will be, and the more elaborated the internal code and the better your memory will be.
2. relational-organizational hypothesis sees imagery itself as not being more complex or more elaborative than verbal coding, but instead imagery is able to form (or produce) more associations or connections between the items to be recalled and other information already in memory (if it is properly used, such as using interacting images). This creates more hooks to help us recall the to be remembered information.
Symbolic-distance effect - individuals are faster to respond when the two objects they are asked to compare differ greatly in some way (such as size).
3. functional-equivalency hypothesis suggests that imagery and perception are very similar to each other (in other words, imagery and perception are functionally equivalent).
Mental rotation studies. Shepard & Metzler (1971) with the unfamiliar 3-D objects and their mirror images that could be rotated either in the picture plane (2-D) or three dimensionally (3-D). The larger the angular disparity of the two objects the longer it took the participants to respond. Presumably this is because the participants had to take a longer time to rotate their visual/mental images.
Cooper & Shepard (1973, 1975) examined mental rotation of recognizable objects - alphabet letters and drawings of hands. Similar findings, but if enough time was left between the arrow cue and the presentation of the rotated object, then participants' reaction time was not longer for the objects that had to be rotated more.
Evidence points to participants mentally rotating the entire polygons/entire figures, thus supporting functional-equivalency.
Imaginal scanning (Kosslyn's studies) - the greater the distance between the starting point and ending point of the scan, the longer the reaction time of the participants.
Cognitive maps - 2 kinds of cognitive maps (route & survey maps).
Individuals tend to expand familiar places on their cognitive maps - so, cognitive maps are distorted and this does not support functional-equivalency hypothesis. Individuals also make other distortions in their cognitive maps - organizing items more and lining things up more than they actually are in reality.
Ambiguous duck/rabbit figure study - demonstrates that beliefs/expectations can influence and distort mental/visual imagery.
Mental imagery can be detrimental. Study of three-term series problems (two premises followed by a conclusion). Participants were faster as solving the problems that did not easily give rise to visualizations, so visualizing the problem interfered with the solution.
Nature of Mental Imagery
Finke's (1989) 5 Principles of Visual Imagery:
• mental/visual imagery is often implicitly encoded (and so it can be used to retrieve info. about the
physical properties of objects); e.g., how many cabinets are there in your kitchen, undoubtedly
you have never explicitly encoded this info., yet you were able to visualize your kitchen and
count the cabinets.
• visual/mental imagery is functionally equivalent to perception (the same areas of the brain are
activated when one perceives visual stimuli & when one creates a mental image; i.e., the visual
cortex or occipital lobe is activated during both processes); e.g., Farah (1985) found that mental
imagery could prime visual pathways so that participants were more accurate in perceiving a real
stimulus after they imagined the stimulus previously.
• mental imagery and physical stimuli are spatially equivalent (i.e., the spatial arrangement of the
elements of a mental image are arranged as they would be in actual physical space); evidence for
this is found in the earlier scanning studies (spatial relationships preserved in the mental images).
Kosslyn et al. (1978) found that both congenitally blind and sighted people took longer to scan
greater distances in their mental images. So, mental imagery seems to have spatial properties
(even if it is not necessarily visual imagery - as in the congenitally blind).
• mental images and physical stimuli are transformationally equivalent (i.e., they both exhibit
corresponding dynamic characteristics and are governed by the same laws of motion); evidence
for this is found in the earlier mental rotation studies (mental rotation being continuous & has to
move through intermediate orientations).
• the structure of mental images corresponds to that of actual perceived objects (i.e., mental
images are coherent, well organized, & can be reorganized). So, the larger an object, or the more
complicated the object, the longer it takes to look at it and the longer it takes to mentally create
and look over it. Kosslyn et al. (1983) found that the more complex the object the longer it took
participants to form an image of it. In addition, complicated descriptions of a geometric stimulus
lead participants to take longer in forming an image.
3 Critiques of Mental Imagery Research and Theory
• Results from mental imagery studies may be the result of tacit knowledge, demand characteristics
and/or experimenter effects. Pylyshyn (1981) points out that participants in scanning experiments
know that two points that are far apart should take longer to scan between, so they let their knowledge
guide their responses (instead of actually scanning a mental image). The way the experiment is set up
may suggest to participants how they should respond (scanning distances may suggest how long it
should take). Also, experimenters may unintentionally give subtle cues to participants about how to act
in an experiment (e.g., Intons-Peterson, 1983, demonstrated this).
• Mental images and pictures actually differ in a number of ways (so, the picture metaphor is
inappropriate). First, one can physically look at a picture without knowing ahead of time what it will be,
whereas with mental/visual images one must first know what the image is before one can examine it in
more detail. Second, mental images are organized more meaningfully than pictures (i.e., the meaningful
parts disappear first/are disrupted). Third, images are more easily distorted by the viewer's interpretation
(i.e., pictures are more objective). So, mental images are less accurate and tend to be distorted more
easily by a person's subjective view.
• There is no visual code nor a verbal code (per se), instead there is one storage code that is propositional
in nature. Propositions are a means of specifying relationships between different concepts (e.g.,
propositions as in the ACT* model). Propositions can be arranged in networks, so that two very related
ideas/objects, etc. joined by sharing a number of propositions. Pylyshyn (1973) argues that imagery
experiments can be explained by participants using propositional representations (instead of by
mental/visual representations). According to propositional theory, if two ideas/things have high association strength with each other, then they share more propositions together in a network and so they should be closely connected to each other and reaction time to the two being presented together should be faster. Kosslyn (1976) tested this and found that when participants did not use mental imagery when verifying a statement, such as "a cat has claws," then their reaction time was faster - this finding supported the conclusion that participants were using propositional representations. However, Kosslyn (1976) also found evidence that participants were using mental/visual imagery - when participants used mental imagery to do the task they were faster to verify the visually larger parts of the images (rather than the visually smaller parts that had higher propositional association strengths with each other), such as "a cat has a head." Propositional theory cannot explain this second finding in the Kosslyn (1976) study.
Neuropsychological Findings - Studies examine the pattern of blood flow in the brain, it is assumed that active areas in the brain require more blood flow.
• Roland & Friberg (1985) had participants engage in 3 tasks (mental arithmetic, memory scanning of auditory info., & visual/mental imagery) - different areas of the brain were active for these different kinds of tasks, furthermore, the occipital lobe (visual cortex) was activated for the mental imagery task.
The finding that mental images activate similar areas in the brain as visual processing has been replicated over and over.
• Kosslyn et al. (1995) further reported that when individuals imagined mental images of different sizes (small, medium, or large) different areas of the visual cortex (occipital lobe) were activated.
• Zatorre et al. (1996) expanded this consistency between imagery & perception to audition. They found that participants who either saw two words and heard the music or saw two words and imagined the music both had activation in their auditory cortexes (temporal lobes).
• In reference to the earlier debate on demand characteristics, Farah (1985) points out that brain imaging studies are not susceptible to demand characteristics.
• In addition, Kosslyn et al. (1995) argues that neuropsychological findings go against propositional theory. Instead, these studies support the conclusion that mental images are functionally equivalent to pictures, etc. because they involve the same areas of the brain.