7. Representation and Organization of Knowledge

Overview of Knowledge Types

• Declarative Knowledge (Knowing that)

  • Expressed through words and symbols.

  • Examples include personal history and math.

• Procedural Knowledge (Knowing how)

  • Involves the steps for performing actions.

  • Examples include driving a car and writing.

Organization of Declarative Knowledge

• Concept: Symbolic knowledge used to understand the world.

  • Often represented as single words.

• Category: A group of similar objects or concepts.

Concepts & Categories

• Natural Categories: Naturally occurring (e.g., birds, trees).

• Artifact Categories: Human-designed groups (e.g., kitchen appliances).

• Basic Level of Specificity: An optimal level of detail for categorization, maximizing distinguishing features.

Feature-Based Categories

• Involves essential elements called defining features.

  • If one important characteristic is missing, something does not belong in that category.

• Advantages: Systematic organization.

• Disadvantages: Difficult to apply universally (e.g., defining 'game').

Prototype Theory

• Prototypes: Abstract averages of previous examples within a category.

• Characteristic Features: Features that describe a prototype but are not essential.

• Classical Concepts: Easily defined through defining features (e.g., bachelor).

• Fuzzy Concepts: Not easily defined (e.g., game), evolve around prototypes.

A Synthesis of Theories

• Combines feature-based and prototype theories to create a core category model.

Theory-Based View

• Understanding categories through implicit theories and explanations, recognizing complexity beyond simple definitions.

Semantic Network Models

• Knowledge represented as concepts interconnected in a web-like structure (nodes and links).

• Collins and Quillian’s Network Model: Hierarchical network allowing for efficient retrieval.

  • Example: Relationships among animals like canaries, birds, and fish.

Schematic Representations

• Schemas: Mental frameworks that organize knowledge, creating meaningful structures.

  • Typically include general, variable facts.

  • May encompass other schemas (e.g., animals categorized as cows, dogs, sheep).

• Scripts: Define sequences of events in specific contexts (e.g., steps in ordering at a coffee shop).

Characteristics of Schemas

• Include relationships between concepts and their attributes (e.g., characteristics of an elephant).

• Capture causal relationships and can foster stereotypes.

Procedural Knowledge

• Acquired through practice, knowledge becomes implicit and fast to retrieve compared to declarative knowledge.

• Production Systems: Set of rules defining a procedure or skill, structured in an iterative manner.

Nondeclarative Knowledge

• Encompasses perceptual, motor, and cognitive skills, as well as simple associative knowledge.

• Relates to priming and basic learning processes like classical conditioning.

Integrative Models of Knowledge

• ACT-R (Adaptive Control of Thought): Combines declarative and procedural knowledge in an information processing framework.

  • Models interactions between nodes representing concepts and rules for execution.

• Parallel Distributed Processing (PDP): Represents knowledge through patterns of activation across networks, mimicking human cognitive flexibility.

Conclusions on Knowledge Representation

• Knowledge organization can take many forms and is influenced by the type of knowledge (declarative vs. procedural).

• Understanding these frameworks aids in grasping how humans categorize and process information, enhancing our approach to learning and memory.