chapter 8 Flashcards
semantics
meaning of individual words
- does not specify how we acquired the information
3 semantic memory representations
1) encyclopedic information
2) lexical knowledge
3) conceptual knowledge
what is encyclopedic information (semantic memory representations)
example
- factual info about world
- ex: everest is the worlds tallest mountain
what is lexical knowledge?
(semantic memory representations)
- example
know a lot of words and stuff about them
ex: the word “short” is related to the word “tall”
what is conceptual knowledge?
(semantic memory representations)
example
understanding of categories of objects
ex: a square has 4 equal sides and 4 right angles
2 components of semantic memory
1) category
2) concept
what is a category?
set of objects that belong together
considered to be at least partly equivalent
what is a concept?
mental representation of a category
what is situated cognition approach
make use of information in the immediate environment or situation
- knowledge often depends on the context that surrounds us
3 levels of categorization
basic
subordinate
superordinate
what is the basic level of categorization
example
- learned first
- used most often
- fastest to identify
- moderately specific
- not a prototype
- more likely to produce semantic priming effect
ex: apple, chair, dog, screwdriver
what is the superordinate level of categorization
- most general, higher level categories
ex: fruit, furniture, animal, tool
what is the subordinate level of categorization
example
- more specific than basic (lower level)
- experts tend to use this level
ex: macintosh apple, desk chair, collie, Phillips screwdriver
PET scans looking at semantic memory
examined whether different regions of the brain tend to process different category levels
- brain distinguishes between different levels of categorization (think differently)
what happened during the study by Kosslyn et al.
PET study
- shown pics, had to indicate what level of categorization
what areas of brain were activated shown in PET study for differently levels of categorization (Kosslyn et al)
basic= lateral and medial view of 2 hemispheres
superordinate= prefrontal cortex and some temporal region (process language and associative memory)
subordinate= parietal areas (allocation of attention, spatial orienting- drive more specific search)
example of expertise
- use what level of categorization?
ex: birding
- use subordinate level
how do we decide what something is
analyze shape, colour, texture
2 models of categorization
prototype model
exemplar approach
what is a prototype
item that is best, most typical example of a category, ideal representation of the category
- can change as gain knowledge
what is the prototype model
- mental abstraction
- categorize by comparing to prototype
- categories have graded structure
- typicality effect
- priming task
- prototypes have high family resemblance
Prototype model
- what is mental abstraction?
example
- based on experiences
- not necessarily anything encountered/experiences/seen/lived in in your life, abstraction based on different experiences
ex: draw house- close to same house drawn by everyone
Prototype model
- categorize by comparing to prototype
- decide whether particular item belongs to a category by comparing this item with prototype
Prototype model
- categories have graded structure
begins with most representative/prototypical members, and continues on through non-prototypical members
- all members of a category are not really equal
what is typicality effect
(Prototype model)
example
faster to recognize/respond/more likely to produce, the closer they are to prototype (and more accurate)
- members of a category differ in prototypicality (degree to which they are representative of category)
example: judging whether items belong to category “bird”, people judge robin more quickly than penguin
Priming task (Prototype model)
example
- priming for typical members
- priming inhibited the judgements for non-prototypical colours
example
- picture colour “red”
- how quicker are you to decide the colour when proceeded by XXX VS word “red” (get priming and faster with word “red”)
- priming only happens of colour patch is close to prototype of red
what is semantic priming effect?
people respond faster to an item if it was proceeded by an item with similar meaning
- facilitates peoples responses to prototypes significantly more than it facilitates their responses to non-prototypes
Prototype model
- prototypes have high family resemblance
- no single attribute is shared by all examples of concept, each example has at least one attribute in common with some other example of the concept
(features overlap)
what is the exemplar approach?
example
compare to examples, not to abstraction
- categorize and classify by thinking of examples we have encountered (common/uncommon)
- collection of numerous members of category
ex: platypus (like duck and beaver)
problem with exemplar approach
our semantic memory would quickly become overpopulated with numerous exemplars for numerous categories (more suitable when thinking of category with relatively few members)
part of brain storing prototypes and exemplars
prototypes= LH
exemplars= RH
network models of semantic memory
what is the emphasis
- interconnections among related items
emphasis
- organization of knowledge/understanding
- the meaning of a word depends on the concepts to which it is connected
3 parts to network models of semantic memory
1) spreading activation model
2) adaptive control of thought
3) parallel distributed processing
what is the spreading activation model
Collins and Lotus (study)
- strength represented by distances of lines (short/close= stronger connection)
- nodes (converging ideas/concepts)
- connections (nodes related through connections, unrelated concepts= no direct connections)
- when hear name of concept, the node representing that concept is activated, the activation expands/spreads from that node to other connected nodes
example of spreading activation model
say “apple”
- activate apple node
- spreading activation
- orange node activates
how is strength determined for spreading activation model
measures how closely related are concepts
- distance
- thickness
- number
what is adaptive control of thought?
- represents knowledge
- propositional network
example of adaptive control of thought (ACT-R)
“Susan gave white cat to maria, who is president of the club”
Propositions (smallest unit of info that can be answered yes/no)
1) susan gave cat to maria
2) cat was white
3) maria is president of the club
what is propositional knowledge
adaptive control of thought
represents in network model relationships between ideas/propositons
what is the propositional network?
adaptive control of thought
- pattern of interconnected propositions
- nodes and links
- propositions are abstract and do not represent a specific set of words
- each of the concepts in proposition can be represented by its own individual network
fMRI on adaptive control of thought
examined how the changes in learning are reflected in selected regions of the cortex and subcortex
what is parallel distributed processing
- cognitive processes can be represented by a model in which activation flows through networks that link together a large number of simple-neuron like units
- this approach emphasizes that we should represent these concepts in terms of networks, rather than specific locations in brain
- parallel (simultaneous activations)
- distributed (activations different locations)
- linked nodes
- spreading activation
- weights
- excitatory/inhibitory
- graceful degradation
what are connection weights (parallel distributed processing)
determine how much activation one unit can pass on to another unit
excitatory/inhibitory of parallel distributed processing
- inhibition: activation of one idea dampens the activation of another
- when a unit reaches a critical level of activation it may affect another unit
what is grace degradation (parallel distributed processing)
- memories face in graceful way (dont just lose everything), lose certain access points but may be able to access memory through other parallel routes
- brains ability to provide partial memory
- explains why brain continues to work somewhat accurately, even when accident/stroke/dementia has destroyed portions of cortex
TOT= tip of tongue effect
what is default assignment (parallel distributed processing)
fill in missing info about particular person/object by making best guess, based on information from other similar people/objects
- draw conclusion about specific member of category
ex: particular engineering student
what is a schema?
generalized knowledge about situations and events
- important in areas of psych and clinical psych
- emphasize how top-down process and bottom-up processing work together
- kind of heuristic (general rule that is typically accurate)
what are scripts
example
- generalized sequence of events in specified order, associated with high familiar activity
- abstraction, prototype of series of events that share underlying similarity
- people recall script more significantly if the script has been clearly identified in advance
ex: going to restaurant
what is a life script
list of events that a person believes would be most important throughout his/her lifetime
- people within culture often share similar life scripts
4 ways schemas influence memories
1) selecting material we want to remember
2) extending the boundaries for visual scenes
3) storing the general meaning of verbal passage
4) forming a single, well-integrated representation in memory
what is comprehension
related to schema
aided by schema activation
- better able to understand world when we have activated appropriate schema
what is memory (related to schema)
schema consistent vs inconsistent
- people highly likely to recall objects consistent (office= desk, chair), less likely to recall objects not consistent (office= wine bottle, picnic basket)
- may not have time to process schema-irrelevent items
- but sometimes show better recall for material that violates our expectations
4 results of memory and schemas that show trends
1) if info describes event and time is limited, people tend to remember info accurately when consistent with schema
2) if info describes minor event, time is limited, people tend not to remember info inconsistent with schema
3) people seldom create completely false memory for lengthy event that did not occur (ex: prof did not dance across room)
4) when info describes major event thats inconsistent with standard schema, people are likely to remember that event (ex: child crashes into Sarah at movies)
what is boundary extension
- recall more that we have seen
- tendency to remember having viewed a greater portion of a scene then was actually shown
- we comprehend a photograph by activating a perceptual schema which features a complete central figure in the photo, also includes a mental representation of visual info that is just outside the boundaries of the photo
example of abstraction & false alarm
ex: ants were on the table. The ants ate the jelly. The jelly was sweet
false alarm: The ants ate the sweet jelly (participants extracted gist)
- remembering items that was not originally presented
what is verbatim memory?
poor word-for-word recall
- remembering the gist
when do you remember word-for-work info
example
- when things matter
- ex: when someone insults you
- takes more effort to encode exact meaning
what is the constructive model of memory
people integrate info from individual sentences in order to construct larger ideas
what is the pragmatic view of memory
- people pay attention to the aspect of a message that is most relevant to their current goals
- people know they usually need to accurately recall the gist (not specific wording)
what are inferences
example
what I read + what i know
ex: “3 turtles rested on a log, and a fish swam beneath them”
- false alarm: :3 turtles rested on FLOATING LOG”
- - made inferences (assumed) based on understanding of world
what is memory integration
- background info encourages to take in new info in schema-consistent fashion
- result: people may remember schema-consistent info, even though not part of original stimulus
- schemas can influence memory when reading ambiguous/unclear material
gender stereotypes
- beliefs and opinions that we associated with females/males, cannot be specified to every individual
implicit association test
based on the principle that people can mentally pair 2 related words together much more easily then they can pair 2 unrelated words
Semantic-memory models focus mainly on:
organized knowledge about the world.
A category is..
a group of objects that belong together
Which of the following is a model to explain how categories are established and identified?
prototype
exemplar
network
You can recognize the name of the state New Hampshire faster if it is preceded by the name Vermont instead of the name Iraq. This effect is called a
priming effect
Semantic priming refers to the finding that accessing the meaning of a word is faster i
it is preceded by another word that is related to it in meaning
Research on prototypes reveals that they: (3)
- judged or supplied as good examples of a category.
- facilitated more by semantic priming than are other examples of a category.
- share many attributes in a family resemblance category.
prototypes are..
- examples of a category.
- judged more rapidly than nonprototypes after semantic priming.
- likely to share common attributes with other members of the category.
According to most network models of semantic memory (e.g., Collins & Loftus, 1975):
presenting the name of a concept activates a node representing that concept.
According to most network models of semantic memory (e.g., Collins & Loftus, 1975):
when a node is activated by presented information, activation spreads to other nodes to which it is connected.
According to the ACT-R model (Anderson and his colleagues, 2004):
the meaning of a sentence is represented by a propositional network.
model that assumes massive interconnections between simple, neuron-like processing units and provides an alternative to semantic-memory models such as ACT-R is called:
parallel distributed processing
a script is
well structured sequence of events in specific order
what does a schema consist of
A schema contains an organized set of generalized information about something that a person knows.
Generalized knowledge about a situation or a type of event (what Bartlett called “an active organization of past reactions or past experiences”) is thought to be stored in a semantic memory structure known as a:
schema
If you are buying tickets at a box-office or eating at a restaurant, you expect certain events to occur in a certain order. This sequence of events is known as:
script
In a typical study about schemas or scripts, people view a common scene, such as a professor’s office or a doctor’s office. When they were subsequently asked to recall objects that they had seen in the office:
they often incorrectly remembered seeing objects that are typically found in offices but were not in the viewed office.
In a study supporting a constructive model of memory, people heard a series of sentences about the same theme (e.g., “The girl next door broke the window on the porch”). This study revealed that:
people integrate related information in order to construct larger ideas
Bartlett’s (1932) classic study, which led to an understanding of the role of schemas in remembering a story, revealed that people:
often added information to a story, remembering inferences that were not actually stated in the story.
The Implicit Association Test (IAT), which has been used to measure attitudes and stereotypes, is based on the principle that people can:
mentally pair related words much more easily and quickly than they can pair unrelated words.
according to network models of semantic memory
function of node
when name of concept is mentioned, node representing concept is activated
attend lecture, psychologust says “memory is created by gathering together and integrating info from variety of sources. After a memory has been formed, we cannot accurately recall the source of each memory.” This speaker would be most likely to endorse ___ view of memoriy
constructive
exemplar approach to concepts suggests that our categories are based on..
less typical items as well as highly typical items
category “furniture” is a
superordinate level category
exemplar approach to semantic memory is especially relevant when..
a category has relatively small number of members
if an important component of your TV set is defective, the TV will not operate, even if all the parts are in good working order, accoring to PDP
memory is different
- memory can often operate even if some of the input is inappropriate
one of the major ideas behind PDP approach to memory is that..
we can show spontaneous generalization, figuring out info about a group of people, based on information about the individual members of that group
important feature of semantic memory is that it allows us to..
it allows us to draw inferences that extend beyond the info supplied in the original stimuls
basic principle that lies at the core of PDP approach is that..
neuron-like units are connected with each other by system of networks
in the PDP approach, the concept called “Spontaneous generalization” means that
we can use individual cases to draw conclusions about some general information
