7. Semantic LTM

Semantic memory

_______________ - Our permanent memory store of general world knowledge. All about MEANING

Loftus & Palmer

_________________- Showed people car accidents and asked:

How fast were the cars going when they ____ each other?

• Follow-up question: Was there broken glass?

  • Smashed: 30% yes

  • Contacted 10% yes

Network model of semantic memory

The Collins & Quillian Model

• Network

• Node

• Pathways

Network

The Collins & Quillian Model/Network model of semantic memory

_________________ - An interrelated set of concepts/ body of knowledge

Node

The Collins & Quillian Model/Network model of semantic memory

_______________ - A point of location in the network representing a single concept

Pathways

The Collins & Quillian Model/Network model of semantic memory

________________ - Labeled directional associations between concepts

Spreading activation

_________________ - The mental activity of accessing and retrieving information from the network

• Takes passive concepts and activates them

• Activation then spreads to related nodes

  • e.g., activation of the “planet” node would also spread to the “earth” node or the “moon” node

Sample of the Semantic Network

A) The word “robin” is activated

Propositions

________________ - Express a relationship between two concepts

• Examples:

  • A robin has wings

  • An apple is a fruit

Pathways + propositions

__________ + _________ - Pathways connect two nodes together to form propositions

• “ISA” pathways express category membership

  • E.g., a robin is a bird

• Property pathways express properties that concepts possess

  • E.g., A robin has the property of wings

Intersection search

_____________ - Two concepts can be activated, each with spreading activation to related nodes(spreading activation) — the two spreads of activation eventually collide—and intersection—which lets you answer: True, a robin is an animal.

• True of False: A robin is an animal?

• Activation lights the robin node, and then spreads to its neighbors

  • `E.g., bird, blue eggs, red breast

Semantic relatedness

________________ - The distance between two nodes in a network is determined by ______

• Concepts close in meaning/highly related (e.g., doctor and nurse) are stored close together in memory

• Unrelated concepts (e.g., doctor and truck) are stored far away

Smith’s Feature Comparison Model

______________ - Semantic memory is a collection of lists

• Feature Lists

Feature lists

Smith’s Feature Comparison Model

_______________ - Contain semantic features (simple, one element characteristics) of each concept stored in memory

…have a much simpler structure than network models

• They are more parsimonious (Occam's razor) — easiest explanation is prolly the right one

Defining features

Feature lists

_______________ - Features absolutely essential to the concept

• Robins must be physical objects, have red breasts and feathers

Defining features appear at the top of each feature list

Characteristic features

Feature lists

_____________ - Features that are common but not essential to the meaning of a concept

• E.g., A robin perches in trees

Characteristic features appear at the bottom of each feature list

Feature comparison

________________ - The major process of information retrieval in the feature list model is this

Mind makes heuristic comparisons

• True or false: A robin is a bird?

To answer, first access each feature list from memory

1. Stage 1: Comparison is fast and involves a global comparison of how much the features in each list overlap

Second, compare each list for common features (feature overlap

2. Stage 2: Comparison is slow, occurring only when the lists have an intermediate amount of overlap

   1. Stage 2 involves only the defining features of each list

Problems w/ feature comparison

Feature lists → Feature comparisons

_______________- No objective way to define a feature as a defining or characteristic feature

• Does not account for fuzzy boundaries

  • Many items in a category do not share a defining feature

    • E.g., are bookends pieces of furniture?

  • How many features of a bird can you lose and still have a bird?

Limited feature lists

• Why would the robin list contain every important concept about robins except that a robin is a bird?

Empirical tests of the network model

_______________ - Used the Sentence Verification Task

• True or False, an x is a y?

  • E.g., True or false, a robin is a bird?

• Response Time is measured

Collins & Quillian (1969) Key Prediction:

• Two concepts that are closer in the network should take less time to verify than two that are farther apart.

Collins & Quillian

Empirical tests of the network model

_________________ (1969) - proposed semantic memory formed hierarchical associations between different categories of knowledge.

Good initial results, but the results of other research indicated problems with the idea

Semantic relatedness and concepts

Semanticity

________________ - Concepts that are more highly interrelated are retrieved faster— organized around meaning

• Ex: Name the 12 months of the year

• Now, name them in alphabetical order

Typicality effects

Typicality

_______________ - Some members of a category are more typical – more representative – than others.

More typical members of a category are judged faster than are less typical ones.

• “A robin is a bird” is verified faster than is “a chicken is a bird”

• Models:

  • The Feature Comparison Model was built to explain typicality effects

  • Adding semantic relatedness to the network model allows it to explain typicality effects

Cognitive economy

_______________ - The incorrect notion from the original network model that redundant information is NOT stored in semantic memory

Redundant Information is stored in memory

• E.g., Breathes appears under the animal node and under the robin node, under the human node, etc.

The Revised Network Model

Schema

______________ - mental framework or body of knowledge about some topic

• Example: Bartlett’s “War of the Ghosts” study

  • When a child is young, they may develop a schema for a dog. They know a dog walks on four legs, is hairy, and has a tail.

Categorization

________________ - Combining entities (information, objects, people, events, etc.) into meaningful units, or categories, is critical for semantic memory

• Organization

• This help us quickly make sense of the world

• Categorization can have drawbacks

  • E.g., Stereotypes

Classic view of categorization

Categorization

_________________ - People create and use categories based on a system of rules

• If something satisfies a set of rules, then it is a member of the category

  • Necessary features: if not present, not a member of category

  • Sufficient features: Nothing more is needed to satisfy category membership

• Arranged according to scientific taxonomies

Graded membership

Characteristics of human categories

_______________ - Categories are loose and fuzzy

• Some members of categories are “better” members  than others

• •E.g., eagle is a “better” bird than ostrich

Central tendency

Characteristics of human categories

______________ - there is some mental core or center to the the category where the best members are found

• Typicality

Typicality

Characteristics of human categories → central tendency

_____________ - the degree to which items are viewed as typical, central members of a category

Prototype theory

Prototypes and exemplars

_______________- the central or core instance of a category

• The prototypical bird may not even exist in the real world

Compare new entities to established prototypes —Allows for graded membership

• Limitations:

  • No information about variability of members

  • No information about category size

Exemplar theory

Prototypes + exemplars

_______________ - People mentally take into account each experience, instance or example, of the encounters they have had with members of that category

• We store exemplars in our memory

  • Judge new information by its resemblance to the stored exemplars

Explanation-based categorization

Categotization

_______________ - Semantic categories are theories of the world we create to explain why things are the way they are.

• Ad hoc categories

• Psychological essentialism

Ad Hoc categories

Explanation-based categorization

____________ - Categories created as needed, often spontaneously

Psychological essentialism

Explanation-based categorization

_______________ - Members of a category are treated as if they have the same underlying property or essence

Basic Principles of Semantic Priming

Priming in semantic memory

______________

1. The priming process takes time

2. The activation of primed concepts is smaller the more removed concepts are from the origin

   1. Nodes are activated for a short amount of time

3. The priming effect is reduced across time

   1. After activation, they return to baseline

Priming

____________ - an exposure of info which can influence the processing of info

• Stimulus presented first in the hopes of influencing some later process

Target

Priming in semantic memory

_____________ - the stimulus that follows the prime

Facilitation

Priming in semantic memory

_____________ - Prime decreases processing time needed for the target

Inhibition

Priming in semantic memory

_____________ - Prime increases processing time needed for the target.

Implicit priming

_______________ from Marcel (1980) - trying to see how info we aren’t consciously aware of can influence our processing of info

• Lexical Decision Task: “yes/no” this is/isn’t a word

• Presented primes followed by a “mask” (scrambled visual patterns)

  • Participants were unaware of the prime

• Primes facilitated lexical decisions even though Ps were unaware of the prime

• Conclusion: Priming is an implicit process (partly

Say: “child” -> then toddler -> the toddler shows up easier in your mind because it was primed

Context effects

______________ - Simpson (1981): Ambiguity and priming

• Target word: Duke

• “The vampire was disguised as a handsome count.”

Context effects are semantic priming effects (based on meaning)

Facilitated priming, similar to the Marcel study. Marcel study just accounts for visual stiff because of scrambled visual patterns

Lexical memory

______________ - The mental lexicon or dictionary where word knowledge is stored

• Distinct from conceptual knowledge

Anomia

______________ - deficit in word finding

Kay & Ellis (1987)

• PDP models can be “lesioned” to mimic the effects of anomia

• Can have an idea in mind but not be able to find it

Category specific deficit

_______________ - A disruption where a person loses access to one semantic category of words but not others

• E.g., Patient J.B.R.

  • Had problems naming living things, but no problems naming non-living things

What caused this dissociation?

• Warrington & Shallice (1984): Disruption in sensory versus functional knowledge