Cognition: Module Three Media

Chapter 4

Perception-Based Knowledge Representations 

 

Verbal and Visual Imagery

 

Verbal Imagery: Using words to create a mental picture of something that is an abstract idea. 

Visual Imagery: The ability to mentally see an object or event without the actual stimulus being in view.

    Webpage: technical process of how a visual image is recreated

    

    Research

Roland and Friberg(1985):Their experiments to study brain activation patterns during imagery determined that some of the same areas of the brain are active when people process verbal imagery as when they process visual imagery.To find these results Roland and Friberg had participants mentally rehearse word jingles (verbally) or directions around streets in their neighborhoods (visually) while they measured blood flow in portions of the brain.

 

Dual-code theory:Claims that there are seperate representations for verbal and visual information.

    -Paivio created this theory based on the research showing that human memory is better if it is encoded visually and verbally. It is easier to remember verbal things if you can attach a corresponding visual image to the material. 

Mental Imagery

mental images: mental representations of how things look to us. Visual information needs to be processed, even when it is not present, because this enables us to figure out how things will look even before we see them. Mental imagery is a technique that can be used in sports, studying, overcoming fears and many other things.   

     http://www.imagery-imagination.com/mi-lnks.htm -This website contains definitions, theories, experiments, and many articles relating to mental imagery.

Mnemonics:

Method of Loci-http://www.ababasoft.com/mnemonic/tech03.htm (tips and games to improve memory)

Peg Words- http://www.magicmnemonic.com/pair/pegs_06.html

Homunculus: This is the mythical internal being that sees and hears what a person is imagining.   

     http://www.cs.uta.fi/~jh/homunculus.html

 ~This is how the homunculus appears when it is represented as a person~

                  

Mental Rotation

mental rotation:the process of continuously transforming the orientation of a mental image. This is how we anticipate what images will look like from a different perspective. 

Roger Shepard and his colleagues performed experiments to study functional properties of mental images. in the studies they placed pairs of 3-D objects in front of participants and then asked them if the objects were the same except for orientation (direction). These studies found that people take longer to make their judgments as the angular disparity between the two figures increased. The more degrees the figure needs to be turned, the longer it takes for the participant to do it.  

            
--Many brain-imaging studies have been performed and they show that while performing any task that involves mental rotation the parietal region of the brain is activated. when a person imagines rotating their own hand it also activates the motor cortex.
Georgopoulos, Lurito, Petrides, Schwartz, and Massey(1989)- They had monkeys perform tasks where they moved a handle to a specific angle in response to a stimulus. They found that the greater the angle, the longer it would take the monkeys to move the handle. This suggests that this task involves a mental rotation process.
Image Scanning                                                                                                                                                                         
Brooks:  He performed experiments where the participants scanned a diagram, such as a block letter F. The participants were to start from a certain point and then answer yes if the point was an outside and no if it was an inside. Another version of this experiment was for the participant to scan a sentence and then answer yes if a word was a noun and no if it was not.                                                                                *F
Participants answered in one of 3 ways:1)said yes or no, 2)tapped answer with hand, 3)pointed to Y's or N's on paper.                                   The results for the image tests concluded that it took much longer for participants to answer by pointing. This was not the same when it came to the sentence test. Scanning a visual array conflicts with scanning a mental array. Spatial Visual Tasks make it harder. 
                                                                                                                                  

 

 

 

 

 

 

 

 

 Two Types of Imagery: Spatial and Visual

 

 

 

 

-Spatial Imagery is assesed either tactually and/or auditorially, this assesses the "where" and uses the parietal regions of the brain.

http://www.limsi.fr/RS2005/chm/ps/ps1/   

 This link shows an experiment using spatial imagery and provides two videos.

-Visual Imagery is assesed visually and assesses the "what", using the temporal portions of the brain.

 http://www.mind-body-medicine.com/visual.htm

 This link provides some explanations of visual imagery.

-An experiment done by Levine, Warach, Farah(1985) determined the brain regions used in spatial and visual imagery when analyzing two different participants. The first had damage to the parietal area of the brain, which caused him to have difficulty with spatial imagery but was still capable of visual imagery. The other participant had damage to their temporal lobe and was capable of spatial imagery activities but had difficulty with visual imagery activities.

-Although, there is both spatial and visual imagery, and one can function without the other, generally imagery includes bot spatial and visual aspects.

http://www.sharpbrains.com/blog/2007/04/20/mental-imagery-and-spatial-rotation-brain-teaser/

 This link shows an activity that requires mental imagery to solve.

 

 Are Visual Images like Visual Perception

 http://www.rybak-et-al.net/vnc.html

 This link describes some aspects of visual perception and provides some images.

http://www.scientificpsychic.com/graphics/  

 This link shows that visual perception is not always trustworthy.

-Finke, Pinker, and Farrah(1989) did an experiment on how well you can imagine images in respect to the actual figures. They found out that you can get to the same conclusions with imagined images or actual figures.  

-Wallace(1984) did an experiment that showed how we percieve the lengths of lines in an image whether it be visual images or visual perception. This experiment was a repeat of a classical experiment know as the Ponzo illusion.

-Chambers and Reisberg(1985) did an experiment to show the differences between mental images and visual perception of real objects with reversable figures. The participates in the study were shown the figure long enough to determine one of the images, then the picture was removed. They were then asked if they could come up with a second image that could have been in the picture, which they were unable to do.

 This reversable picture shows both a rabbits head and a ducks head.

 

-Peterson, Kihlstrom, Rose, and Gilsky(1992) did an experiment to show that with appropriate prompting participants could figure out both images in the figure used in the Chambers and Reisberg experiment. Alhtough, they showed that people prefer an actual figure versus a mental image.

-Kasslyn and Thompson(2000) did an experiment that found that brain-imagining does not always activate the primary visual cortex altough it does activate the the later visual areas and the occipital cortex. The more detailed and vivid an image the more likely there will be activation of the primary visual cortex.

-O'Craven and Kanwisher(2000) did an experiment and determined that the areas used for faces and the areas used for locations are the same whether using an actual figure or a mental image.

 

 Cognitive Maps

 

 This is an example of a cognitive map.

 

http://ralyx.inria.fr/2004/Raweb/siames/uid120.html

 This link provides explanations of cognitive maps.

http://www.pigeon.psy.tufts.edu/asc/schmajuk/default.htm#Hierarch%20Cog%20Map  

 This link provides some examples of different cognitive maps and navigation around them. 

-Cognitive maps are how we use mental imagery to help us remember the ways around our enviroment.

-Route Maps are used by young childern and people in new situations, and indicate specific places such as verbal directions.

-Survey Maps are when you really know landmarks and your way around. Survey maps are more advanced than route maps.

-Thorndyke and Hayes-Roth(1982) did an experiment to determine secertaries abilities to quickly find their ways around a building using route map.

  

 

 Navigation

-Egocentric Representation is how space is represented when we see it, such as in cognitive maps. This cogntive maps can be a route or survey map.

 

 This is an example of a survey map.

 

 

 -Allocentric Representation is a non-specific map because it is not from a certain view point, such as a physical map. Physical maps are different tan cognitive maps because physical maps show something in location to all of the things around it, and cognitive maps do not.

 

 This is an example of a physical map.

 

 -An experiment, done by Maguire, Godian, Johnsrude, Good, Ashbruner, Frackwiak, and Frith(2000), showed that when people navigate their way around an enviroment they show lots of activation in the hippocampus. This was shown in an experiment done with taxi drivers, and as the taxi drivers got more familiar with the streets they had more hippocampal activity.

 

 

 

 This image shows the hippocampus.

 

 

 

 

 

 

 

 Map Distortions


    Our mental maps are usually organized in a hierarchy.  This means we think of a city being part of a state and a state being part of a country, in order.  Same sort of thing as a robin being a bird and a bird being an animal.  Okay, fair enough, so now we've got Stevens and Coupe coming up with the theory that a state or country border can distort how we see a map, making us prone to error.

    To prove their little theory, Stevens and Coupe made some simple maps and tested people on memorizing them.  Turns out, the theory's pretty sound according to the experiment.  Here's an example of the sort of thing they did.

 

    What you need to keep in mind, then, is that Congruent just has the country border going right down the middle, Incongruent takes the border and stretches it out to try and confuse you, and Homogeneous has no border at all.  People only have trouble with Incongruent maps.  That's Stevens and Coupe on Map Distortion.

Translating from Words to Images

    Franklin and Tversky did an experiment where they used words to describe a participant's surrounding, then they participant had to picture their surroundings cognitively and remember where everything is located relative to them.  They were tested on how fast they responded to a question about an object from their surroundings.  Franklin and Tversky determined that people were quickest making above-below judgments, locating objects either above or below them more easily than something to the left or right of themselves, being slowest making left-right judgments.
   H.A. Taylor and Tversky ran some experiments on making cognitive maps.  Basically, they provided three types of spatial information for participants to convert into a cognitive map.  Route descriptions, survey descriptions, and actual maps.  Route descriptions are just words describing a route you take, like what road you need to turn on.  A survey statement is a bit more detailed, trying to lay out a floor plan, basically, using words to tell you where everything is located relative to each other.  Actual maps, then, are the physical maps using imagery rather than words to show where everything is.  H.A. Taylor and Tversky determined that all three types of spatial information are equally effective.  Participants performed just as well using each type of information to find their way to a destination.

 


Memory for Serial Lists



    Serial-order information is the representation of a sequence of information in a particular order.

    R. Conrad found that when memorizing a serial list of letters, participants experienced acoustic confusion in which they misremembered a letter as a similar sounding letter, such as a V instead of a B.

    Front-line anchoring is the advantage we have in remembering the first elements in a list.  S. Chen, Swartz, and Terrace found that monkeys have the advantage of front-line anchoring as well, but not pigeons.

    Sternberg ran an experiment where he'd give a sequence of numbers and then ask for them to remember what number came after a particular number, called the probe, in the sequence.  So, the sequence is 32956.  What comes after 9?  5 does.  What he found is that it takes longer to answer that than it does to answer what comes after 3.  This seems to indicate that we remember the sequence and search it from start to finish for the probe number we were asked about.



Hierarchical Encoding of Serial-Order Information

   
   Klahr, Chase, and Lovelace did the same sort of thing as Sternberg, but tried a much longer list.  They used the list of all the letters in the Alphabet, which most people have memorized in the form of The Alphabet Song.  So, the sequence is the entire alphabet.  What comes after R?  S does.  The same sort of thing as what Sternberg did, with the same results of it taking longer the farther down the list the letter was.  However, there was one critical difference.  We encode serial-order information hierarchically.  L is part of LMNOP.  B is part of ABCD.  ABCD is part of ABCDEFG and so forth.  Thus, when we are asked what comes after L, most of us will say M much faster than if we were asked what comes after O.  It becomes similar Sternberg's experiment, with the larger sequence being broken down into smaller sequences.  However, what comes after A still gets the quickest response, since it is the earliest small sequence.
    Johnson repeated the above experiment, but used random letters rather than the Alphabet, encouraging a hierarchy by providing spacing, such as DY_JHQ_GW.  It worked just the same, so that's more evidence that front-line anchoring is present in serial-order information more when we encode hierarchically.

 

 

Baddeley's Theory of Verbal Working Memory


    Baddeley proposed that we have two systems of working memory that he calls the visuospatial sketchpad and the phonological loop, which are slave systems of the central executive.  The phonological loop consists of multiple components, including the articulatory loop, which is our so-called inner voice that rehearses verbal information to ourselves.  Another component is the phonological store.  This is the so-called inner ear that listens to the inner voice and stores that information.  The Broca's Area may be related to the articulatory loop, while the parietal-temporal region of the brain may be related to the phonological store, according to Vallar, Di Betta, & Silveri.  There is some controversy on that idea, but the Broca's Area seems to activate when someone is remembering a list of items, while the parietal-temporal region seems to activate when someone is storing verbal information.  See the brain below for locations.



Chapter 8

 

Introduction to Problem Solving

http://www.virtualsalt.com/crebook3.htm

http://www.sis.pitt.edu/~is1042dm/class-notes/08.pdf

http://www.umcs.maine.edu/~larry/latour/ECAI/paper-sent/node3.html


This article discusses context effects on problem solving 

 http://www.nbu.bg/cogs/personal/kokinov/ctxtef_p.pdf

 

Kohler's Research on the Mentality of Apes

http://www.pigeon.psy.tufts.edu/psych26/kohler.htm

        


 

                       Chica on the jumping stick

 

toolmonkeys.gif (16243 bytes)

 

http://www.kirjasto.sci.fi/kohler.htm

http://www.nap.edu/readingroom/books/biomems/wkohler.html

 

An excerpt from Edward Thorndike's Animal Intelligence: An experimental study of the associative process in animals (1898)

http://www.pigeon.psy.tufts.edu/psych26/thorn.htm

http://psychclassics.yorku.ca/Thorndike/Animal/

http://www.coe.uh.edu/courses/cuin6373/idhistory/thorndike_extra.html

 

Artificial Intelligence Example of a “toy” problem Problem space

http://www.cse.msstate.edu/~hansen/classes/AIspring04/slides/bruteforcesearch.pdf

 

Eight-tile Puzzle

http://www.comp.dit.ie/rlawlor/Prob_Solv/puzzles/Eight%20Puzzle.pdf

http://www.zib.de/reinefeld/bib/93ijcai.pdf

 

Analogy

http://www.nbu.bg/cogs/personal/kokinov/COG501/Analogy-making.html

 

Operator Selection

 

 

 Backup avoidance is the tendency in problem solving to avoid operators that take one back to a state already visited.  The book uses the example of an eight-tile puzzle in which the people show great reluctance to take a step back, even if it is necessary in solving the problem.

 

Eight Tile Puzzle

 

Rush Hour 

 

 

 Playable Races: Race of Hobbits    Difference Reduction

~The tendency in problem solving to select operators that eliminate a difference between the current state and the goal.

 ~The book gave the example of a chicken walking directly towards a fence to reach food on the other side, as opposed to walking around it.  The chicken then could not move forward and refused to backtrack, which made the problem impossible to solve.

 ~Difference reduction is also referred to as "hill climbing."  Image going straight up a hill/mountain and trying to reach the highest point.  Obstacles might keep one from reaching the highest point if they follow a direct path up the mountain.

  

Hobbits and Orcs Problem/Solution

 

The Three Container Problem/Solution

 

 

 

Means End Analysis- The creation of a new goal (end) to enable a problem solving operator (means) to apply in achieving the old goal.

General Problem Solver (GPS)- A problem-solving simulation program created by Newell and Simon that embodies means-end analysis.

 

 

Example of Means-End Analysis

 

Zebra puzzle

Cereal Mystery Chart

Logic Problems Page


 Tower of Hanoi

 

Problem Solving Methods - Textually explains the Tower of Hanoi, and gives more information about the three types of general problem solving

 

Problem-Solving Strategies - Breaks down the problem solving strategies into everyday terms

 

 

Differences in Selection of Operators - Helps explain what an operator is, and helps to differentiate between the different methods  

 

 

 

 

 

 

 

 

  Functional Fixedness -


    Refers to people's tendency to represent objects as serving conventional problem solving functions and thus failing to see them as

    serving novel functions.


   People don't see beyond an objects given function. All objects can be used for novel or various other purposes.




 


 

 Dunker's classic experiment

 

 

 

 

  Set Effects -

 

 Becoming biased to prefer certain methods when solving a problem. People continue to use a specific solution that has worked in

  prior experiences even though a simpler solution is possible.


The Luchins water jar problem


This experiment best illustrates the Einstellung effect, or mechanization of thought. Individuals get accoustomed to using a specific
solution, that when presented with problem that doesn't fit into said solution, get stuck and unable to continue.

Generally, set effects occur when some knowledge structures become more available than others.

Best advice to dissipate set effects....TAKE A BREAK!!


Incubation Effects

  Related to set effects; during intial attempts to solve a problem, people set themselves to think about about the problem in certain
 ways and use certain specific knowledge from past experiences (Set Effects). If the individual cannot solve the problem because
 their methodology proves useless, going away from the problem allows dissipation of useless knowledge structures. This, in turn, allows
 for a fresh approach...


INSIGHT     =
                                                          =  NO!


      Insight is not a magical moment when everything falls into place and the solution is finally clear.
Experiment summaries