The Organization of Knowledge • Classical concepts are categories that can be readily defined
through defining features. Fuzzy concepts are categories that
Declarative knowledge- Expressed in words and other symbols (i.e., cannot be so easily defined. Their borders are, as their name
"knowing that") implies, fuzzy.
Procedural knowledge- Is how to follow procedural steps for • Classical concepts tend to be inventions that experts have devised
performing actions (i.e., "knowing how") for arbitrarily labeling a class that has associated defining
features. Fuzzy concepts tend to evolve naturally (Smith,1988,
Organization of Declarative Knowledge 1995; see also Brent et al., 1996).
Concepts and Categories
Concepts- An Idea about something that provides a means of Real-World Examples: Using Exemplars
understanding the world. It is the fundamental unit of symbolic Some psychologists suggest that instead of using a single abstract
knowledge, or knowledge of correspondence between symbols and prototype for categorizing a concept, we use multiple, specific
their meaning. exemplars. Exemplars are typical representatives of a category (Ross,
2000; Ross & Spalding, 1994).
Category- One way to organize a large number of concepts is by the
means of categories. A category is a hierarchy of concepts. A category A Synthesis: Combining Feature-Based and Prototype Theories
is a concept with members. • A full theory of categorization can combine both defining and
characteristic features (see also Hampton, 1997; Poitrenaud,
Categories Richard, & Tijus, 2005; Smith,1988; Smith, Sho- ben, & Rips,
Natural Categories- Are groupings that occur naturally in the world, 1974; Wisniewski, 1997, 2000), so that each category has both a
like birds or trees. prototype and a core.
Artifact Categories- Are groupings that are designed or invented by • A core refers to the defining features something must have to be
humans to serve particular purposes or functions, like automobiles and considered an example of a category. The prototype
kitchen appliances. encompasses the characteristic features that tend to be typical of
an example (a bird can fly) but that are not necessary for being
Not all categories are stable, however (Dunbar, 2003; Thagard, considered an example (an ostrich).
2003). Some categories are created to achieve goals in everyday life or
for a specific purpose, for example, "my bestfriends," Theory-Based View of Categorization
"Things one can write on" or "things I need to purchase in the • A departure from feature-based, prototype-based, and exemplar-
supermarket." These categories are called ad hoc categories based views of meaning is a theory-based view of meaning, also
(Barsalou, 1983; Sandberg, Sebastion & Kiran 2012). sometimes called an explanation-based view.
They typically are described not in words but rather in phrases. • A theory-based view of meaning holds that people understand and
categorize concepts in terms of implicit theories, or general ideas
Feature-Based Concepts: A Defining View they have regarding those concepts (Markman, 2003; Mark- man
Feature-Based Concepts et al., 2007).
The classic view of concepts disassembles a concept into a set of
featural components. All those features are then necessary (and Semantic-Network Models- Suggest that knowledge is represented in
sufficient) to define the concept (Katz, 1972; Katz & Fodor, 1963). our minds in the form of concepts that are connected with each other in
This means that each feature is an essential element of the concept. a web-like form.
Together, the features uniquely define the concept; they are defining
features. Collins and Quillian's Network Model- A semantic network is a web
of elements of meaning (nodes) in which the elements are connected
Prototype Theory: A Characteristic View with each other through links (Collins & Quillian, 1969; Maru-paka, lyer,
Prototype Theory & Minai, 2012).
Takes a different approach: grouping things together not by their
defining features but rather by their similarity to an averaged model of Organized knowledge representation takes the form of a
the category. hierarchical tree diagram.
• Nodes are the elements; typically, concept. The connections
Prototypes and Characteristic Features between the nodes are labeled relationships. labeled relationship
• A prototype is an abstract average of all the objects in the (link)
category we previously encountered. That is, we have an average • Within a hierarchy, we can efficiently store Information that applies
representation of the category, and we compare objects to that to all members of a category at the highest possible level in the
average representation (prototype) when making a decision hierarchy. We do not have to repeat the information at all of the
whether or not to place them in a certain category. lower levels in the hierarchy. Therefore, a hierarchical model
• Objects that are prototypical of a category therefore have high provides a high degree of cognitive economy. The system allows
family resemblance. Crucial for prototypes are characteristic for maximally efficient capacity use with a minimum of
features, which describe (characterize or typify) the prototype but redundancy.
are not necessary for it. Characteristic features commonly are
present in typical examples of concepts, but they are not always
present.
Comparing Semantic Features
Classical and Fuzzy Concepts
, • An alternative theory is that knowledge is organized based on a Representations of How We Do Things: Procedural Knowledge
comparison of semantic features, rather than on a strict hierarchy The "Production" of Procedural Knowledge
of concepts (Smith, Shoben, & Rips, 1974). Although this theory • Procedural knowledge representation is acquired by practicing the
sounds similar to the feature-based theory of categorization, it implementation of a procedure. Once a mental representation of
differs from it in a key way: nondeclarative knowledge is constructed, that knowledge is
• Features of different concepts are compared directly, rather than implicit. As your explicit access to nondeclarative knowledge
serving as the basis for forming a category. decreases, however, your speed and ease of gaining implicit
• A common method for examining semantic networks involves the access to that knowledge increases. Eventually, most
use of word-stem completion. In this task, participants are nondeclarative knowledge can be retrieved for use much more
presented a prime for a very short amount of time and then are quickly than declarative knowledge can be retrieved.
given the first few letters of a word and told to complete the stem • Serial processing - information is handled through a linear
with the first word that comes to mind. The stems could be sequence of operations, one operation at a time.
completed with a semantically related word or any number of • Production - the generation and output of a procedure (Jones &
unrelated words. Normally, participants complete these stems with Ritter, 2003).
a semantically related item. • When the rules are described precisely and all the relevant
• We may broaden our understanding of concepts further if we conditions and actions are noted, a huge number of rules are
consider not only the hierarchical and basic levels of a concept required to perform even a simple task. These rules are organized
(Komatsu, 1992) but also other relational information the concept into a structure of routines (instructions regarding procedures for
contains. Specifically, we may better understand the ways in implementing a task) and subroutines (instructions for
which we derive meanings from concepts by considering their implementing a subtask within a larger task governed by a
relations with other concepts, as well as the relations among routine). Many of these routines and subroutines are iterative,
attributes contained within a concept. meaning that they are repeated many times during the
performance of a task.
Schematic Representations • If you want to complete a particular task or use a skill, you use a
Schema- A schema is a mental framework for organizing knowledge. production system that includes the entire set of rules
It creates a meaningful structure of related concepts. Schemas have (productions) for executing the task or using the skill (Anderson,
several characteristics that ensure wide flexibility in their use 1983, 1993; Gugerty, 2007; Newell & Simon, 1972; Simon, 1999a,
(Rumelhart & Ortony, 1977; Thorndyke, 1984): 1999b).
1. Schemas can include other schemas. • Sometimes, production systems, like computer programs, contain
2. Schemas encompass typical, general facts that can vary slightly bugs. Bugs are flaws in the instructions for the conditions or
from one specific instance to another. for executing the actions.
3. Schemas can vary in their degree of abstraction.
4. Schemas also can include information about relationships Nondeclarative Knowledge
(Komatsu, 1992). Some of this information includes relationships • Encompass a broader range of mental representations than just
among concepts, attributes, attributes in related concepts, procedural knowledge (Squire, 1986; Squire et al., 1990).
concepts and particular contexts, and specific concepts and • In addition to declarative knowledge, the following forms of
general background knowledge. nondeclarative knowledge are mentally represented:
5. Boundary extension - When we remember a certain scene, we • perceptual, motor, and cognitive skills (procedural knowledge)
extend the boundaries of what we have seen and incorporate
• simple associative knowledge (classical and operant conditioning)
information in our memory of details that were actually just outside
• simple no associative knowledge (habituation and sensitization)
the boundaries of what we saw.
• priming (fundamental links within a knowledge network, in which
Scripts the activation of information along a particular mental pathway
facilitates the subsequent retrieval of information along a related
• A script contains information about the particular order in which
pathway or even the same mental pathway; see
things occur. In general, scripts are much less flexible than
schemas. Scripts, however, include default values for the actors,
Two Types of Priming Semantic
the props, the setting, and the sequence of events expected to
Priming- Primed by a meaningful context or by meaningful information.
occur. These values taken together compose an overview of an
event. Such information typically is a word or cue that is meaningfully related
to the target that is used.
• Scripts also may come into play in regard to the ways in which
Repetition Priming- A prior exposure to a word or other stimulus
experts converse with and write for one another. Certainly, experts
primes a subsequent retrieval of that information.
share a jargon-specialized vocabulary commonly used within a
group, such as a profession or a trade.
Integrative Models for Representing Declarative and Non
• Imaging studies reveal that the frontal and parietal lobes are declarative Knowledge Combining
involved in the generation of scripts (Godbout et al., 2004). The
generation of scripts requires a great deal of working memory.
Combining Representations: ACT-R
Further script generation involves the use of both temporal and
An excellent example of a theory that combines forms of mental
spatial information.
representation is the ACT (adaptive control of thought) model of
• Typicality effect - if both typical and atypical actions are provided, knowledge representation and information processing (Borst &
the atypical information will be recalled more readily. This Anderson, 2015; Anderson, 1976, 1993; Anderson et al., 2004).
difference is likely due to the increased effort in processing Anderson intended his model to be so broad in scope that it would offer
required for atypical information as compared with typical an overarching theory regarding the entire architecture of cognition.
information.
through defining features. Fuzzy concepts are categories that
Declarative knowledge- Expressed in words and other symbols (i.e., cannot be so easily defined. Their borders are, as their name
"knowing that") implies, fuzzy.
Procedural knowledge- Is how to follow procedural steps for • Classical concepts tend to be inventions that experts have devised
performing actions (i.e., "knowing how") for arbitrarily labeling a class that has associated defining
features. Fuzzy concepts tend to evolve naturally (Smith,1988,
Organization of Declarative Knowledge 1995; see also Brent et al., 1996).
Concepts and Categories
Concepts- An Idea about something that provides a means of Real-World Examples: Using Exemplars
understanding the world. It is the fundamental unit of symbolic Some psychologists suggest that instead of using a single abstract
knowledge, or knowledge of correspondence between symbols and prototype for categorizing a concept, we use multiple, specific
their meaning. exemplars. Exemplars are typical representatives of a category (Ross,
2000; Ross & Spalding, 1994).
Category- One way to organize a large number of concepts is by the
means of categories. A category is a hierarchy of concepts. A category A Synthesis: Combining Feature-Based and Prototype Theories
is a concept with members. • A full theory of categorization can combine both defining and
characteristic features (see also Hampton, 1997; Poitrenaud,
Categories Richard, & Tijus, 2005; Smith,1988; Smith, Sho- ben, & Rips,
Natural Categories- Are groupings that occur naturally in the world, 1974; Wisniewski, 1997, 2000), so that each category has both a
like birds or trees. prototype and a core.
Artifact Categories- Are groupings that are designed or invented by • A core refers to the defining features something must have to be
humans to serve particular purposes or functions, like automobiles and considered an example of a category. The prototype
kitchen appliances. encompasses the characteristic features that tend to be typical of
an example (a bird can fly) but that are not necessary for being
Not all categories are stable, however (Dunbar, 2003; Thagard, considered an example (an ostrich).
2003). Some categories are created to achieve goals in everyday life or
for a specific purpose, for example, "my bestfriends," Theory-Based View of Categorization
"Things one can write on" or "things I need to purchase in the • A departure from feature-based, prototype-based, and exemplar-
supermarket." These categories are called ad hoc categories based views of meaning is a theory-based view of meaning, also
(Barsalou, 1983; Sandberg, Sebastion & Kiran 2012). sometimes called an explanation-based view.
They typically are described not in words but rather in phrases. • A theory-based view of meaning holds that people understand and
categorize concepts in terms of implicit theories, or general ideas
Feature-Based Concepts: A Defining View they have regarding those concepts (Markman, 2003; Mark- man
Feature-Based Concepts et al., 2007).
The classic view of concepts disassembles a concept into a set of
featural components. All those features are then necessary (and Semantic-Network Models- Suggest that knowledge is represented in
sufficient) to define the concept (Katz, 1972; Katz & Fodor, 1963). our minds in the form of concepts that are connected with each other in
This means that each feature is an essential element of the concept. a web-like form.
Together, the features uniquely define the concept; they are defining
features. Collins and Quillian's Network Model- A semantic network is a web
of elements of meaning (nodes) in which the elements are connected
Prototype Theory: A Characteristic View with each other through links (Collins & Quillian, 1969; Maru-paka, lyer,
Prototype Theory & Minai, 2012).
Takes a different approach: grouping things together not by their
defining features but rather by their similarity to an averaged model of Organized knowledge representation takes the form of a
the category. hierarchical tree diagram.
• Nodes are the elements; typically, concept. The connections
Prototypes and Characteristic Features between the nodes are labeled relationships. labeled relationship
• A prototype is an abstract average of all the objects in the (link)
category we previously encountered. That is, we have an average • Within a hierarchy, we can efficiently store Information that applies
representation of the category, and we compare objects to that to all members of a category at the highest possible level in the
average representation (prototype) when making a decision hierarchy. We do not have to repeat the information at all of the
whether or not to place them in a certain category. lower levels in the hierarchy. Therefore, a hierarchical model
• Objects that are prototypical of a category therefore have high provides a high degree of cognitive economy. The system allows
family resemblance. Crucial for prototypes are characteristic for maximally efficient capacity use with a minimum of
features, which describe (characterize or typify) the prototype but redundancy.
are not necessary for it. Characteristic features commonly are
present in typical examples of concepts, but they are not always
present.
Comparing Semantic Features
Classical and Fuzzy Concepts
, • An alternative theory is that knowledge is organized based on a Representations of How We Do Things: Procedural Knowledge
comparison of semantic features, rather than on a strict hierarchy The "Production" of Procedural Knowledge
of concepts (Smith, Shoben, & Rips, 1974). Although this theory • Procedural knowledge representation is acquired by practicing the
sounds similar to the feature-based theory of categorization, it implementation of a procedure. Once a mental representation of
differs from it in a key way: nondeclarative knowledge is constructed, that knowledge is
• Features of different concepts are compared directly, rather than implicit. As your explicit access to nondeclarative knowledge
serving as the basis for forming a category. decreases, however, your speed and ease of gaining implicit
• A common method for examining semantic networks involves the access to that knowledge increases. Eventually, most
use of word-stem completion. In this task, participants are nondeclarative knowledge can be retrieved for use much more
presented a prime for a very short amount of time and then are quickly than declarative knowledge can be retrieved.
given the first few letters of a word and told to complete the stem • Serial processing - information is handled through a linear
with the first word that comes to mind. The stems could be sequence of operations, one operation at a time.
completed with a semantically related word or any number of • Production - the generation and output of a procedure (Jones &
unrelated words. Normally, participants complete these stems with Ritter, 2003).
a semantically related item. • When the rules are described precisely and all the relevant
• We may broaden our understanding of concepts further if we conditions and actions are noted, a huge number of rules are
consider not only the hierarchical and basic levels of a concept required to perform even a simple task. These rules are organized
(Komatsu, 1992) but also other relational information the concept into a structure of routines (instructions regarding procedures for
contains. Specifically, we may better understand the ways in implementing a task) and subroutines (instructions for
which we derive meanings from concepts by considering their implementing a subtask within a larger task governed by a
relations with other concepts, as well as the relations among routine). Many of these routines and subroutines are iterative,
attributes contained within a concept. meaning that they are repeated many times during the
performance of a task.
Schematic Representations • If you want to complete a particular task or use a skill, you use a
Schema- A schema is a mental framework for organizing knowledge. production system that includes the entire set of rules
It creates a meaningful structure of related concepts. Schemas have (productions) for executing the task or using the skill (Anderson,
several characteristics that ensure wide flexibility in their use 1983, 1993; Gugerty, 2007; Newell & Simon, 1972; Simon, 1999a,
(Rumelhart & Ortony, 1977; Thorndyke, 1984): 1999b).
1. Schemas can include other schemas. • Sometimes, production systems, like computer programs, contain
2. Schemas encompass typical, general facts that can vary slightly bugs. Bugs are flaws in the instructions for the conditions or
from one specific instance to another. for executing the actions.
3. Schemas can vary in their degree of abstraction.
4. Schemas also can include information about relationships Nondeclarative Knowledge
(Komatsu, 1992). Some of this information includes relationships • Encompass a broader range of mental representations than just
among concepts, attributes, attributes in related concepts, procedural knowledge (Squire, 1986; Squire et al., 1990).
concepts and particular contexts, and specific concepts and • In addition to declarative knowledge, the following forms of
general background knowledge. nondeclarative knowledge are mentally represented:
5. Boundary extension - When we remember a certain scene, we • perceptual, motor, and cognitive skills (procedural knowledge)
extend the boundaries of what we have seen and incorporate
• simple associative knowledge (classical and operant conditioning)
information in our memory of details that were actually just outside
• simple no associative knowledge (habituation and sensitization)
the boundaries of what we saw.
• priming (fundamental links within a knowledge network, in which
Scripts the activation of information along a particular mental pathway
facilitates the subsequent retrieval of information along a related
• A script contains information about the particular order in which
pathway or even the same mental pathway; see
things occur. In general, scripts are much less flexible than
schemas. Scripts, however, include default values for the actors,
Two Types of Priming Semantic
the props, the setting, and the sequence of events expected to
Priming- Primed by a meaningful context or by meaningful information.
occur. These values taken together compose an overview of an
event. Such information typically is a word or cue that is meaningfully related
to the target that is used.
• Scripts also may come into play in regard to the ways in which
Repetition Priming- A prior exposure to a word or other stimulus
experts converse with and write for one another. Certainly, experts
primes a subsequent retrieval of that information.
share a jargon-specialized vocabulary commonly used within a
group, such as a profession or a trade.
Integrative Models for Representing Declarative and Non
• Imaging studies reveal that the frontal and parietal lobes are declarative Knowledge Combining
involved in the generation of scripts (Godbout et al., 2004). The
generation of scripts requires a great deal of working memory.
Combining Representations: ACT-R
Further script generation involves the use of both temporal and
An excellent example of a theory that combines forms of mental
spatial information.
representation is the ACT (adaptive control of thought) model of
• Typicality effect - if both typical and atypical actions are provided, knowledge representation and information processing (Borst &
the atypical information will be recalled more readily. This Anderson, 2015; Anderson, 1976, 1993; Anderson et al., 2004).
difference is likely due to the increased effort in processing Anderson intended his model to be so broad in scope that it would offer
required for atypical information as compared with typical an overarching theory regarding the entire architecture of cognition.
information.
