Slide 1
REPRESENTATION & MANIPULATION OF KNOWLEDGE IN MEMORY:IMAGES
AND PROPOSITIONSMental Representation of Knowledge2
Approaches:Introspectionist approach where one can ask people to
describe ones own knowledge representations and knowledge
representations processes.Rationalist approach deduce logically how
people represent knowledge.
For centuries philosophers have done the rationalist approach
exactly. From Classic Epistemology the study of the nature,
origins, limits of human knowledge, philosophers have distinguished
between 2 kinds of (2 Types) Knowledge Structures:Declarative
Knowledge refers to the facts that can be stated, such as birth
date, name of friend or how an animal looks.2. Procedural Knowledge
refers to the knowledge of procedures that can be implemented.
Cognitive Psychologists have 2 main sources of empirical data on
knowledge representation:Standard laboratory experiments this is
indirect study of knowledge representations (because we cannot
directly look into the peoples minds). Researchers observe how
people handle various cognitive tasks that require manipulation of
mentally represented knowledge.Neuropsychological studies - 2
methods used here: 1) researchers observe how normal brain responds
to various cognitive tasks involving knowledge representation 2)
they observe the links between various deficits in knowledge
representation and associated pathologies in the brain Some of the
theories researches have proposed to explain HOW WE REPRESENT AND
STORE KNOWLEDGE IN OUR MINDS: Ways Knowledge can be represented in
the mind:As a mental picture we have some mental representations
that resembles pictorial, analogous imagesIn words - we have other
mental representations that are highly symbolic, like wordAbstract
propositions - We have more fundamental propositional
representations that are in a pure abstract mentalese that is
neither verbal nor pictorial Note that there are differences
between images and words when they are used to represent ideas in
the external world. * Cat
* The cat is under the table * under (cat, table)
PICTURES IN OUR MIND Imagery the mental representation of things
that are not currently seen or sensed by the sense organs ( Moulton
& Kosslyn, 2009; Thomas, 2003). 2 Theories how we represent
information in our minds:Dual-Code Theory we use both pictorial and
verbal codes for representing information in our minds (Paivio,
1969, 1971) 1.1. 2 kinds of codes: 1) Analog resemble the object
they are representing; pictorial codes
2) Symbolic Code a form of knowledge representation that has
been chosen arbitrarily to stand for something that does not
perceptually resemble what is being represented; verbal codes.2.
Propositional Theory we do not store mental representations in the
form of images or mere words; our mental representations (or
mentalese) more closely resemble the abstract form of a
proposition. According to this theory we may experience our mental
representations as images but this images are epiphenomena
secondary and derivative phenomena that occur as a result of other
more basic cognitive processes.
a Proposition the meaning underlying a particular relationship
among concepts. Logicians have devised a shorthand means of
expressing propositions called predicate calculus. According to
propositional theory pictorial and verbal information are encoded
and stored as propositions. Then when we retrieved the information
from storage, the propositional representation is retrieved.
Propositional Representations Underlying MeaningsWe may use
propositions to represent any kind of relationship, including
actions, attributes, spatial positions, class membership, or almost
any other conceivable relationship: In the example table next slide
propositions are expressed in a shorthand form (known as predicate
calculus) commonly used to express underlying meaning. This
shorthand is intended only to give some idea of how the underlying
meaning of knowledge might be represented. It is not believed that
this form is literally the form in which meaning is represented in
the mind. In general the shorthand form for representing
propositions is this: [Relationship between elements] [subject
element] [object element]
Type of Relationship Representation in Words Propositional
Imaginal Representation Representation
Actions A mouse bit a cat Bite (action) Mouse (agent of action)
Cat (object)
Attributes Mice are furry (External Surface characteristics)
furry (attribute) mouse (object)
Spatial Positions A cat is under the (vertical higher table.
Position) (table, cat) Class or A cat is an animal (categorical
membership) Category animal (category) Membership cat (member)
Limitations of Mental Images: What are the limits of analogical
representation of images?
Limits To Knowledge Representation in Mental Images: We could
not use mental image to trace the lines to determine which
component shapes were or were not part of a whole figure. Mental
representation of figures are not the same as percepts of these
figures Visual images can be distorted through verbal
informationSemantic (verbal) information tends to distort recall of
visual images in the direction of the meaning of the
imagesPropositional codes may override the imaginal code in some
circumstances Semantic labels clearly influences mental images:
Limitations of Propositional Theory in contrast to the above
discussion, we do not necessarily need a propositional code to
manipulate information, but can manipulate mental images directly.
Manipulation of mental images can be done by combining two distinct
images to form a different mental image altogether. This
manipulation can be thought of as an imaginal Gestalt experience.
In the combined image , the whole of the two combined images
differed from the sum of its two distinct parts. Mental images can
be combined effectively to create mental images of another kind be
it geometrical shapes, objects or letters. Examples: letter H and
XRight triangles letter M
Mental reinterpretation of ambiguous figures involves two
manipulations:Mental realignment of the reference frame this would
involve shift in the positional orientations of the figures on the
mental page or screen on which the image is displayed.Mental
reconstrual reinterpretation of parts of the figureMental
Manipulations of ImagesFunctional-equivalence - refers to
individuals using about the same operations to serve about the same
purposes for their respective domains.Functional-equivalence
hypothesis although visual imagery is not identical to visual
perception, it is functionally equivalent to it. Functional
equivalent things are strongly analogous to each other they can
accomplish the same goals. The functionally-equivalent images are
thus analogous to the physical percepts they represent. This view
essentially suggests that we use images rather than propositions in
knowledge representation for concrete objects that can be pictured
in the mind. Mental images can be rotated just like physical
objects.
Mental rotation involves rotationally transforming an objects
visual mental image .
Image scaling we use mental images the same way we use actual
perceptions
Image scanning images can be scanned in much the same way as
physical percepts can be scanned.
Synthesizing Images and PropositionsBased on studies about
mental rotations, image scaling, and image scanning mental images
appears functionally equivalent to perception in many ways. However
studies involving ambiguous figures and unfamiliar mental
manipulations suggest that there are limits to the analogy between
perception and imagery. ALTERNATIVE SYNTHESIS ALTERNATIVE SYNTHESIS
Johnson-Lairds Mental Models suggests that mental representations
may take any of three forms:Propositions fully abstracted
representations of meaning that are verbally expressibleImages
mental pictureMental Models are knowledge structures that
individuals construct to understand and explain their experiences.
The models are constrained by the individuals implicit theories
about those experiences which can be more or less accurate. The
models depend not on physical or other laws but on your belief on
them.
Hemispheric Lateralization of Information Manipulation Right
hemisphere appears to represent and manipulate visuospatial
knowledge in a manner similar to perception. Left hemisphere
appears to be more proficient in representing and manipulating
verbal and other symbol-based knowledge 2 kinds of imagesVisual
imagery refers to the use of images that represent visual
characteristics such as colors and shapes. 2. Spatial imagery
refers to images that represent spatial features such as depth,
dimensions, distances, and orientations.SPATIAL COGNITION AND
COGNITIVE MAPS Spatial Cognition deals with the acquisition,
organization, and use of knowledge about objects and actions in
two- and three-dimensional space. Cognitive maps are internal
representations of our physical environment, particularly centering
on spatial relationships. Representation and Organization of
Knowledge in memory Organization of Declarative Knowledge Concept
fundamental unit of symbolic knowledge. It is an idea about
something that provides a means of understanding the world.
Category is a group of items into which different objects or
concepts can be placed that belong together because they share some
common features, or because they are all similar to a certain
prototype. 2 distinction of categories: 1) natural categories are
groupings that occur naturally in the world 2) artifact categories
are groupings that are designed or invented by humans to serve
particular purposes or functionsAnother distinction of categories
ad hoc categories are categories specifically created just for the
moment or for specific purposeCategories appears to have a basic
level. Basic level of category a level within a hierarchy that is
preferred to other levels How we decide to put things in a certain
category?A) one theory suggests that we put an object only in one
category if it has several defining features .
B) Another approach is we compare an object with an averaged
representation (a prototype) to decide whether it fits into a
category. C) Still another approach is that people can categorize
objects based on their own theories about those objectsDiscussion
A) Feature Based Categories the classic view of categories break
down a concept into a set of featural components. Each of these
features are necessary to define the category which means that each
feature is an important element of the category. Together the
features uniquely define the category and are called defining
features. Problems of feature-based categories: Some categories do
not readily lend themselves to featural analysis. Ex. Game
Violation of those defining features does not seem to change the
category we use to define them. Feature based categories has some
attractive features but it does not give a complete account of
categories. Discussion B) Prototype theory the characteristic view
of categories; grouping things not by their defining features but
by their similarity to an averaged model of the category.Prototype
is an abstract average of all the objects in the category we have
encountered before Characteristic features describe (characterize
or typify) the prototype. Some psychologists suggest that instead
of using a single abstract prototype for categorizing a concept, we
use multiple specific prototype called exemplars. Exemplars are
typical representative of a category Some psychologists make a
synthesis of the feature-based and prototype-based theories;
combining both defining and characteristic features so that each
category has both a prototype and a core. Core refers to the
defining features something must have to be considered an example
of a category. Prototype encompasses the characteristic features
that tend to be typical of an example but not necessary to be
considered to be an example Discussion C) Theory-based
categorization also called an explanation-based view. Theory-based
view of meaning holds that people understand and categorize
concepts in terms of implicit theories, or general ideas they have
regarding those concepts. How do people use their theories for
categorization? Ex: What make someone a good sport In the
componential view you would try to isolate features of a good
sport. In the prototype view you would try to find characteristic
features of a good sport. In the exemplar view you might try to
find some good examples you have known in your life. In the
theory-based view you would use your experience to construct an
explanation for what make someone a good sport.SEMANTIC NETWORK
MODELS Semantic network models suggest that knowledge is
represented in our minds in the form of concepts that are connected
with each other in a web-like form. Semantic network is a web of
elements of meaning (nodes) that are connected with each other
through links. The elements are nodes. They are typically concepts.
The connections between nodes are labeled relationships. a b R In a
simple semantic network nodes serves as junctures representing
concepts linked by labeled relationships : a basic network
structure showing that relationship R links the nodes a and b.
Hierarchical Structure of Semantic Network A semantic network has a
hierarchical structure. The concepts (represented through the
nodes) are connected by means of relationships Example of
hierarchical Structure of Semantic Network: GENOGRAM
Example of a Genogram:
SCHEMATIC REPRESENTATIONS Schemas mental frameworks for
representing and organizing knowledge that encompasses an array of
interrelated concepts in a meaningful organization. Example:
Scripts a particular kind of schema. It contains information
about the particular order in which things occur . In general
scripts are much less flexible than schemas. Ex:
Representations of How we Do Things: Procedural Knowledge
Processing Procedural Knowledge: Serial processing information is
handled through a linear sequence of operations, one operation at a
time. Production there are rules governing production o example 1)
if-then rule, or 2) production system input-output Bugs flaws in
production
Non declarative Knowledge: Perceptual, Motor and Cognitive
skills ( procedural knowledge) Simple associative knowledge
(classical and operant conditioning) Simple non-associative
knowledge (habituation and sensitization) Priming fundamental links
within a knowledge network , in which the activation of information
along a particular mental pathway facilitates the subsequent
retrieval of information along a related pathway or even the same
mental pathway. Integrative Models of Representing Declarative and
Non-Declarative Knowledge Combining Representations ACT R Adaptive
control of thought representation
3 stages of acquisition of procedural knowledge Cognitive think
of explicit rules for implementing the procedureAssociative
practice using explicit rules extensively usually in highly
consistent mannerAutonomous we use these rules automatically and
implicitly without thinking about them. We show a high degree of
integration and coordination , as well as speed and accuracy
Example: