Memory:Schema Theory, A European Solution, Generalization hierarchies

<< Memory:Prototype, Rosch and Colleagues, Experiments of Stephen Read

Memory:Superset Schemas, Part hierarchy, Slots Have More Schemas >>

Cognitive Psychology PSY 504

Lesson 36

Memory

Representation of knowledge (Continued)

Schema Theory

Schema:

Schema is a Greek word which means frame.

A general knowledge structure that provides a framework for organizing clusters of knowledge.

When representing the knowledge we have about various categories, it is useful to be able to

encode the information that certain features are typical of a category while others are not.

Schemas are organized sets of facts. Subjects use schemas to infer that certain unobserved and

unmentioned elements must be present. It seems that schemas are a major mechanism for

elaborating material during study, and are also the major mechanism for reconstructing memories

at test.

Schema Theory refers to a collection of models presuming that we encode such knowledge

clusters into memory and use them to comprehend and store our experiences.

A European Solution

While Psychologists in US were busy developing Stimulus-Response Theories, Bartlett in

England and Piaget in Switzerland were arguing that behavior is influenced by large units of

knowledge organized into schemas.

Bartlett's Schema Theory

Bartlett (1932) conducted an experiment for getting evidence for the role of schemas in memory.

He used a story called "The War of the Ghosts". It has been used in research on many

subsequent occasions and is still a popular research item today. Bartlett was interested in how

the subjects would remember a story that fit in so poorly with their cultural schemas. He had

subjects recall the story after various delays. Bartlett's subjects showed clear distortions in their

memory for the story and these distortions appeared to grow with time.

Subjects were distorting the story to fit with their own cultural stereotypes. So he found when

subjects read a story that does not fit with their own schemas, they will exhibit a powerful

tendency to distort the story to make it fit.

According to the Bartlett, Schema is an active organization of past experiences in which the mind

abstracts a general cognitive structure to represent many particular instances of those

experiences. It means all past experience are organized actively in mind. Then we absorb new

experiences in our schemas.

A schema provides a knowledge structure for interpreting and encoding aspects of a particular

experience. At his time, Bartlett was not taken seriously.

The Rise of Schema Theory

In 1975 a number of prominent scientists argued that schemata/schemas are needed to organize

knowledge in artificial intelligence, cognitive psychology, linguistics, and motor performance. A

number of psychologists from different fields gave importance on schemas. Artificial intelligence

focused on schemas because schemas make knowledge that helps the computer in information.

Knowledge structures include all parts of information, for example a bird; it includes beak, flying

quality, legs, hair etc. Linguistic psychologist also gave importance to the schemas because

schemas help in organizing the linguistic knowledge.

Cognitive psychology was working on atomistic little bit idea at that time then they realized the

importance of schemas.

Bartlett's major assumptions were adopted and developed further into a modern schema theory.

106

Cognitive Psychology PSY 504

Modern Schema Theory

Minsky (1975):

Minsky started the modern schemas theory again, for representing knowledge in Artificial

Intelligence Programs. He made many computer programs in which schemas were used. He said

mega knowledge structures are very important for computer programs.

Rumelhart (1980):

He said schemas are building blocks of cognition. Like atoms are building blocks of element.

Schema is very rich, brave and complex building block of memory.

Schemas provide skeleton structure to be filled out with detailed properties of a particular

instance. For example a bird schema of bird includes instances like feathers, beak, or other bird

features.

House: A Schema example

The basic insight is that concepts like house are defined by a configuration of features, and each

of these features involves specifying a value the object has on some attribute. The schema

representation is the way to capture this basic insight. Schemas represent the structure of an

object according to a slot structure, where slots specify values that the object has on various

attributes. So we have the following partial schema representation of a house:

For instance, our knowledge of what a house is like.

Slots are Values.

House

Superset schema: building a special slot in each schema is its superset schema. In

House, it is Building. Building has walls, roof, it is built on the ground etc

Parts: Rooms (living, bed, kitchen, etc.)

Materials: wood, bricks, stone, cement

Function: human dwelling

Shape: rectilinear, triangular

Size: 100-10,000 sqft

Exemplars: images of various houses

Generally these slots are present in all things. Each pair of a slot and a value specifies a typical

feature. The fact that houses are typically built of materials like wood and brick does not exclude

such possibilities as cardboard.

If the computer program has all these information then it can match the information with your new

information.

Generalization hierarchies

Supersets schemas are basically hierarchies' that e saw with semantic networks. In this case of

schemas, they are sometimes called generalization hierarchies. These hierarchies provide a lot of

information about an object.

Part hierarchy

Schemas have another type of hierarchy, called part hierarchy. Thus, Parts of houses, such as

walls and rooms, have their own schema definitions. Stored with schemas for walls and rooms we

would find that these have windows and ceilings. Thus, using the part relationships, we would be

able to infer that houses have windows and ceilings.

Schemas are designed to facilitate making inferences about the concepts. If we know something

a house, we can use the schema definition to infer that it is probably made of wood or brick, and it

107

Cognitive Psychology PSY 504

has walls, windows, and the like. However the inferential processes for schemas must be able to

deal with exceptions. It is also necessary to understand the constrains between slots of

schemas.

108

Table of Contents: