Cognitive psych exam 3

parallel distributed processing

.a system of handling information in which many steps happen at once (i.e., parallel) and in which various aspects of the problem or task are represented only in a distributed way; proposed method of information processing in connectionist networks

fuzzy boundary

A category defined by a prototype will have a graded membership and a _____________ ______________: - Some items will be very similar to the prototype, and will clearly be a member - Other items will be less similar to the prototype, but may still be similar enough to be classified as a member

distributed

A connectionist representation of learning is a ____________________ process involving thousands of changes across the network

local representation

A representation in which information is encoded in some small number of identifiable nodes. Local representations are sometimes spoken of as "one idea per node" or "one content per location." Often contrasted with distributed representation

rating task

A task in which research participants must evaluate some item or category with reference to some dimension, usually expressing their response in terms of some number. For example, participants might be asked to evaluate birds for how typical they are within the category "birds," using a "1" response to indicate "very typical" and a "7" response to indicate "very atypical." Findings suggest that people perform this task by comparing test items to the prototype and evaluating the similarity/closeness

inference

A theory that involves all members of a category is most likely to lead people to draw an ________________________

center/sentence verification

According to the prototype perspective, in the production task participants will start with the _______________ of the category (e.g., naming birds that more closely resemble the bird prototype) and then work their way outward from there By this logic, birds mentioned early in the production task should be birds that yield fast response times in the __________________ ______________ task- This is because what matters for both tasks is proximity to the prototype; members of categories that are privileged in one task turn out to be privileged in others The responses that are mentioned first in the production task tend to be the basic-level category members in both production and the sentence verification task

sentence verification task

An experimental procedure, used for studying prototype theory, in which participants are given simple sentences (e.g., "Cats are animals") and must respond as quickly as possible whether the sentence is true or false.

slower/connections

Anderson (1974) --> fan effect experiment "The fireman is in the park" "The lawyer is in the park" "The lawyer is in the church" "The doctor is in the bank" If an activated node has two connections, less activation will flow through each connection than if there is just one Prediction that it will take longer to find nodes with more connections E.g., it will take longer to confirm that the lawyer (2 connections) is in the park (2 connections) than to confirm that the doctor is in the bank because the activation is diverted in more than one node for lawyer and park Subjects are _________________ to confirm items where the nodes have more than one connection It takes longer to build up activation when the nodes involved have many connections The Fan Effect --> degree of fan = number of __________________ to a node

sufficient condition

Anything that has this property will be a member, but there may be other members that do not have this property.

represent/link

Associative links don't just tie together bits of knowledge, they also help ___________________ the knowledge (E.g., the concept of knowing "George Washington was an American President" can be represented as an associative link connecting a "Washington" node and a "president" node) The ______________ itself is a constituent of the knowledge

typicality/category

Categorization enables us to apply general knowledge to new cases we encounter- Allows us to make inferences about these new cases How does this use of knowledge proceed? Early research indicated that inferences guided by ____________________ : Study where participants were told a new fact about robins and were also willing to infer the new fact would be true about ducks; However, if they were told a new fact about ducks, they would not infer this to be true about robins People were willing to make inferences from the typical case to the whole ____________________, but not from an atypical case to the category

essential

Category membership depends on decisions about which features are _________________ to the category

basic

Certain members of a category are privileged (e.g., birds that are "birdier") but also certain categories are privileged- in the structure/way they're used Rosch et al. argues there is a "natural" level of categorization, neither too specific nor general, that people tend to use in conversations and reasoning known as ______________-level categorization Easier to decide what members of basic-level categories have in common vs. more inclusive (general/superordinate) categories Children learning to talk often acquire basic-level knowledge before specific subcategories or general superordinate categories

hierarchical

Collins & Quillian (1969) --> proposed idea of concept organization in ________________ networks Gave people sentence verification task It takes longer to confirm that a canary is an animal than that a canary is a bird; Collins & Quillian claim that it takes longer time to respond to question because it takes longer to traverse the hierarchy Hierarchical networks are very useful for storing semantic memories, although there are complications in using them to explain response times

redundant/associative links

Collins and Quillian (1969)- proposed a hierarchical conceptual network model based on research of associative links (e.g., proposed that the memory system avoids ________________ storage of connections between "Cats have hearts" and between "Dogs have hearts," and so on for all other animals. Instead "have hearts" is stored as a property of all animals. To confirm that cats have hearts, therefore, you must traverse two links; from "cats" to "animals," and from "animals" to "have hearts") Using sentence verification task, participants were asked to hit a "true" or "false" button in response to sentences they were presented with (E.g.," "a robin is a bird," "a cat has claws," "a cat is a bird") Prediction that sentences with concepts that were more directly linked by an association would be responded to faster because presumably, this would mean it takes less time for activation to spread through the network; Slower responses to sentences requiring a "two-step" connection longer response times associated with number of _____________________ ______________ that must be traversed in memory (e.g., "a canary is a canary" requires no links to be traversed but "a canary can fly" requires traversing between two links: from canary to bird, and from bird to fly)

both/experience

Combination of exemplar and prototype proposals is feasible because in any given circumstance we may be relying on ___________ simultaneously; can be used at the same time Ratings of category membership in both exemplar and prototype theories can change with __________________

complex/experience

Compared to nonexperts, experts in a particular field will have more _____________ explanatory theories; Experts will have acquired much _____________________ and hence will have more explanatory theories in order to cover the large amount of information they have accumulated

locally/parallel

Computers modeling PDP (parallel distributed processing) networks are more successful at problem solving than computers using other locally represented models; Processing is across the network rather than ______________ represented (where each individual node is associated with meaning) Processing happens in __________________ rather than serially

needs

Conceptual knowledge has many elements: A prototype, exemplars, a theory, and representations of perceptual properties and actions associated with the concept Which of these elements that we choose to focus on depend on our ________________ at the moment

associative/propositional

Different kinds of ____________________ links between nodes can represent different kinds of relationships We can use labels to represent different kinds of links E.g., "Louise is a pig" vs. "Louise has a pig" These different kinds of links allow us to represent more complex combinations; we can have other types of links as well We can build even more complex representations by using __________________ nodes

anomia

Different types of concepts (i.e., living vs. non-living things) represented in different brain areas- Confirmed by observations of people with ________________- a disorder, often arising from specific forms of brain damage, in which the person loses the ability to name certain objects (e.g., does a whale have legs?) Often, this problem is specific to certain categories; some patients lose ability to name living but not non-living things, or vise versa

proposition/role

Each ellipse (or node) within a propositional network identifies a single ___________________ Associations between ellipses (nodes) connect each ellipse to ideas that are constituents of the proposition, and the associations are labeled to specify the constituent's ________________ within that proposition --> Enables the distinction between "Dogs chase cats" and "Cats chase dogs"

explanatory theories

Each of us has our own theories about the concepts we hold- a web of beliefs- that provide a crucial knowledge base that we rely on in thinking about an object, event, or category; and they enable us to understand new facts we might encounter about the object or category

standard

Exemplar-based reasoning is similar to prototype view in many ways: - Categorize objects by comparing them to a mentally represented "standard" The difference between the exemplar view and the prototype view lies in what the "_____________" for the category is is: - Prototype theory standard --> prototype (average representing entire category) - Exemplar theory standard --> whatever example of the category comes to mind (different examples may come to mind on different occasions)

variability

Exemplars are useful because they may provide information that is lost from the prototype- including information about ______________________ within the category People routinely "tune" their concepts to match contextual circumstances (E.g., gifts for a student vs. gifts for a faculty member or, considering Chinese birds vs. American birds)

category structure/typicality

Experiments that Support the Idea that Concepts are Represented by Prototypes: 1) production task (list members of a category; items closer to the prototype typically named earlier on) 2) typicality ratings ("On a scale of 1-7, rate how typical each of these birds is of the category 'bird'") 3) picture identification (In a task where participants are presented with pictures of different dog breeds and asked whether or not each is a dog, people respond "yes" more to German shepherds and collies more quickly than to chihuahuas and dachshunds) 4) sentence verification (true or false: a robin is a bird vs. true or false: a penguin is a bird --> Responses faster for robin than for penguin) 5) induction of unfamiliar properties ("A duck has an omentum, does a robin?" vs. "A robin has an omentum, does a duck?"- More likely to say yes to the second than the first) More willing to generalize from a typical member (robin) than from an atypical member (ducks) to the whole category 6) make up sentences- (e.g., about birds- "birds fly south for the winter" then replace "birds" with "penguins" or "robins") The sentences about birds are more likely to make sense when typical birds are substituted than when atypical birds are substituted Results from all of these different types of experiments support the same sort of ________________ _______________ --> The category members that get the highest _________________ ratings and that are named first also produce the fastest sentence verification, etc.

graded/family

For most common categories, it is difficult to find defining features that work for all members: - Some members seem to fit better than others (______________ membership) - It may be difficult to draw a boundary between members and nonmembers (fuzzy boundary) - Categories often exhibit a "_________________ resemblance" structure; Many features shared across members; No feature shared by all members.

deep/beliefs/understandings/implicit theory

How are category judgments made when they don't rely on typicality? Frank Keil (1986)- Study where preschool students asked what makes something a "coffeepot" a "raccoon" etc., then if a toaster could turn into a coffeepot/a raccoon could turn into a skunk findings hint that: 1) People reason differently about naturally occurring items like raccoons than they do about manufactured items like coffeepots; Naturally occurring items (like lemons, raccoons) are defined in ways that refer to _______________ properties rather than surface level appearances 2) Claims about an objects deep properties depend on a web of other _______________ that are "tuned" in each case to the category being considered (E.g., having raccoon parents means we can assume the offspring is a raccoon; having doctor parents does not mean the offspring will be a doctor) 3) Our _____________________ of categories matter- Some categories (raccoons, dollar bills) seem to be defined by something other than just resemblance to prototype or exemplars; There seem to be some "essential properties" (raccoon genes, being printed by the government) that are necessary to be a member of that category; To know what properties are essential for a particular category, we need a theory about what defines that category and how it relates to other category; Part of the mental representation of a category can be an ________________ ___________________

exemplars

How can we represent categories with family resemblance structure and fuzzy boundaries? - Prototypes - Collection of ____________________

potential/activation

How do PDP models detect patterns? Knowledge refers to a ____________________ rather than to a state; knowing something in a network corresponds to how the _______________________ will flow if there is activation; Once connections between patterns are in place, activation of either pattern will lead to activation of the other (E.g., knowledge that George Washington was a president- pattern of activation of nodes that represent "President" will elicit pattern of activation of nodes that represent "Washington" and vise versa)

emotional

How do prototypes differ from stereotypes? Prototypes are a summary of experiences- An "average" of a specific concept Stereotypes are often acquired through social channels that shape our ideas about a specific concept; Often include an ________________/evaluative dimension

shared/essential

How do we decide if two objects/concepts resemble each other? More properties shared = greater resemblance However, resemblance cannot simply be judged by counting shared properties- Resemblance does depend on ___________________ properties, but it depends on whether the objects share important, _____________________ properties

beliefs

How do we decide which features to ignore vs. consider when assessing similarity? Depends on our ____________________ about the concept in question

beliefs/knowledge

How we think about, use, and construct concepts are shaped by a web of _______________ and background __________________

agent/object/relation

In a propositional network, each node represents an entire proposition; The node is connected to other nodes by labeled links Each proposition is generally connected to three nodes: 1) ____________ link- the one performing the action (dog) 2) ___________ - the thing being acted on (cat) 3) _____________ - the relationship between the agent and the object (chase) In this case, the propositional node represents "The dog chases the cat." Propositional nodes can be combined with "isa" and "hasa" links There can also be links for general categories (DOG) and specific instances (MY DOG) There can be other types of links, such as Time and Location--> These allow the representation of specific events or episodes

activation

In distributed representations, Meaning is attached to patterns of nodes firing, rather than interpreting what one single node represents We can only learn what's being represented by looking at many nodes simultaneously to find out what pattern of ____________________ exists across the entire network

natural

In general, people tend to assume more stability and homogeneity when reasoning about _______________ kinds

prototype/faster

In the sentence verification task, Responses were slower for sentences like "a penguin is a bird" than for sentences like "a robin is a bird" These results can be explained by a ____________________ perspective Participants chose their response by comparing the thing mentioned (e.g., penguin, robin) to their prototype for that category (e.g., bird) The closer the thing mentioned was to the prototype, the _______________ participants responded (penguins cannot fly, so they may seem less similar to a bird prototype than a robin) The slowest responses in a standard sentence verification task are the ones that require the reader to trace a path from one basic category to a different superordinate category (E.g., "A bat is a bird")

beliefs/interact

Inferences are also guided by preexisting sets of ____________ We can think about how we know certain concepts __________________ with each other (E.g., people willing to believe that if a gazelle's blood contains a certain enzyme, that lion's blood contains the same enzyme too (lions eat gazelles which gives a feasible explanation of why this would occur) but they do not believe it the other way around (lion's blood --> gazelles blood)

typicality/independent

Judgments of category membership depend on judgments of ___________________ Some results show no linkage at all between judgments of category membership and judgments of typicality Armstrong, Gleitman and Gleitman (1983)- gave participants instruction: "We all know that some numbers are even-er than others. What I want you to do is rate each of the numbers on this list for how good an example it is for the category 'even number.'" Participants then given a list of numbers and asked to rate "how even" each number was; Participants could judge category membership and typicality with equal amounts of ease, but these judgments were _______________ of each other (E.g., judged 4 as a more typical even number than 7,534, even with awareness that both numbers fall into the category of "even") Clearly there is some basis for judging category membership that is separate from assessment of typicality People are generally consistent with each other in which odd numbers they consider more odd and which even numbers they consider more even People tend to rate single-digit numbers as most odd/even and are very consistent at doing so

prototypes/exemplars/beliefs

Knowledge contains several parts: 1) _________________ for most concepts 2) Set of remembered ___________________ used to predict judgments about relevant categories 3) People have individual ________________ about each concept they hold- these beliefs reflect the person's understanding of cause-and-effect relationships --> These beliefs are woven into the broader network that manages to store all the information in memory The network influences how items are categorized/how objects in the world are reasoned about Even the simplest concepts require a multifaceted representation in the mind

exemplars

Malleability of concepts is easy to understand when considering effects of __________________- Different settings trigger different memories of circumstances and thus bring different exemplars to mind

goal/cause and effect/relational

Many concepts can be characterized in terms of features Other concepts involve ______________ derived categories (e.g., "diet foods" or "exercise equipment")- Understanding of concepts like these depends on understanding of the goal itself (e.g., losing weight) and some ______________-___________-____________ reasoning about how a particular concept (e.g., certain diet foods) may help to achieve this goal Similarly, __________________ categories (e.g., "rivalry," "hunting") and event categories ("visits," "dates," "shopping trips") involve a web of beliefs about how various elements (e.g., predator and prey in hunting; the shopper and the store in shopping trips) are related to each other

parallel/patterns

Many theorists argue the model of Parallel Distributed Processing (PDP) makes sense: - Biologically, the brain relies on _______________ processing, with ongoing activity occurring simultaneously - Brain also known to use a "divide and conquer" strategy- complex tasks broken down into small components, with separate brain areas working on each component -PDP models have great capacity for detecting __________________ in the input they receive, despite a range in variations of how the pattern is implemented; Able to generalize what they have "learned" from past patterns of activation to new, never-before-seen variations on the pattern

prototype/strength/nodes

More recent data reveal several complications (contradictions to Collins and Quillian findings): 1) Sentence verifications are known to be faster if a sentence involves examples that more closely fit a __________________- This was not accounted for in Collins and Quillian example 2) Also, proposed principle of non-redundancy doesn't always hold- Redundant connections are variable in associative ______________________ (E.g., although penguins and peacocks are both birds, participants respond more quickly to sentences like "peacocks have feathers" than "penguins have feathers" because feathers are a more notable characteristic of peacocks so they are more easily associated) Even though it is informationally redundant, a strong association between "peacocks" and "have feathers" is likely to be established Despite the complications, it is often predictable how fast someone will access knowledge by counting the number of _______________ people will have to navigate in order to generate their response --> Associative links play a pivotal role in knowledge representation

necessary condition

Must have this property to be a member, but this property by itself may not be enough to make it a category member.

hierarchies

Networks of nodes can be organized into ________________ Each node has its own properties, and also those from the nodes above; Prevents duplication/redundancy When a new member of a category is added, it automatically inherits properties from above

exemplar/correlated

One advantage of exemplar models over prototype models: Exemplar models can preserve information about the correlation of different features within a category Medin et al. (1982) showed that subjects tend to use information about correlated features - Their subjects first studied a set of examples to learn a category. Then they had to decide whether new instances belonged in the category or not. Made up disease known as "burlosis"- designated by a series of symptoms (swollen eyelids, splotches on ears, discolored gums, nosebleed) - Not every symptom shared by all patients, not every patients had all symptoms -Each symptom occurs equally often across patients (6 times) - The last two symptoms (discolored gums and nosebleed) are perfectly correlated --> If someone has discolored gums, they also had nosebleed; If they do not have discolored gums, they did not have nosebleed - After subjects learned the category of patients with burlosis, subjects received new pair of patients--> Both patients have the same number of symptoms, but in the first patient, the last two symptoms appeared together, while in the second patient they are not. None of the earlier patients had a pattern like the second patient Since earlier patients had symptomatic patterns like the first, people were more likely to pick the first new patient The experiment by Medin et al. suggests that subjects rely on _______________ representations for this newly learned category; _________________ features are predicted by the exemplar theory, but not by the prototype theory Other evidence suggests that "experts" who have extensive experience with a specific category may rely more on a prototype

naturally/manufactured/beliefs

People tend to think of the properties of ______________________ occurring objects as stable people's conceptualizations of ____________________ artifacts are malleable and constantly subject to our own constructions People will reason differently about natural kinds and artifacts because they have different ___________________ about why categories of either sort are as they are (E.g., the finding that children agree toasters could be turned into coffee pots (artifact) but skunks could not be turned into raccoons (natural))

prototype/exemplar

Problems associated with ________________ and ________________ accounts of categorization: - Participants know that the definition of an even number is absolute, but they still give even numbers different ratings of evenness - An orange that has been flattened, painted with brown and red stripes, and covered with coconut flakes is still identified as an orange - A counterfeit $20 bill can look identical to a real $20 bill but is not considered money

propositions/relationships

Propositional networks share many claims with those of other theorized networks: - Nodes are connected by associative links; some links are stronger than others - Strength of these links depends on how frequently/recently it has been used - Once a node is activated, the process of spreading activation causes nearby nodes to be activated as well Propositional network models are distinctive in their attempt to represent knowledge in terms of ____________________: - They are more advantageous than other categorization models because they can better differentiate between different types of ____________________ - Easy to differentiate between different kinds of relationships between concepts because they are made up of propositions rather than just concepts

resemblance/essential/beliefs

Prototype and exemplar use depends on judgments of ____________________ Judgments of resemblance depend on being able to focus on ________________ features (rather than trivial ones) Decisions about what is essential vary by category, and particularly depend on existing _______________

typical

Prototype- and exemplar-based theories both help to explain the fact that some items appear to be more _________________ of a category than others A typical item can be a particular exemplar (or set of similar exemplars) or a prototype (an item corresponding to the ideal)

center/prototype

Prototypes refer to the ideal item in a category, and thus define the center and not the borders of the category. According to Rosch's Prototype Theory of category membership, the center (or ideal) item in a category is defined, rather than the borders of the category Since the prototype defines the ______________ of a category, if people can still decide if an item belongs to a category even when it is different than the _________________, then it suggests that they must have been using some other process to make decisions about categories

graded membership

Rather than relying on a single prototype, we can also represent a category by storing many or all known exemplars of the category When a new item is encountered, it is compared against all members --> Of course, this increases the memory requirement Exemplar theories of concepts can also explain __________________ ________________ --> The more exemplars that a stimulus matches, the better it fits into a category

activation pattern

Reliance on distributed representation has many negative implications for how a connectionist network functions: -A network using distributed representations must use processes that are similarly distributed, so that one widespread __________________ ________________ can evoke a different (but equally widespread) pattern - Additionally, the steps occurring to bring this about must all occur simultaneously (in parallel) with each other so that one entire representation can smoothly trigger the next This pattern is referred to as Parallel Distributed Processing (PDP)- a system of handling information in which many steps happen at once (i.e., parallel) and in which various aspects of the problem or task are represented only in a distributed way

spreading activation/less

Retrieving knowledge from the network relies on processes of _____________________ ________________- The farther the activation has to spread, the more time it takes Leads to the idea that closely related ideas take ___________ time to retrieve than knowledge about more distant ideas

retina/activation

Simple example of a Distributed Representation: Representation of color at one location in the _______________--> Thousands of possible colors represented by three different types of cones The color that is perceived depends on the pattern across the three cone types Networks with distributed representations are used in many artificial intelligence systems A learning algorithm can adjust the connections between nodes to store new information in the network After the adjustments, different concepts will be represented as different patterns of _____________ across the network

grounded cognition

Some theorists argue for a position of "embodied" or "________________ _________________" - States that the body's sensory and action systems play an essential role in all cognitive processes, thus, it is inevitable that our concepts will include representations of perceptual properties and motor sequences associated with each concept

fan effect

The phenomenon whereby retrieval time to retrieve a particular fact about a concept increases as more facts are known about that concept; it takes longer to build up activation when the nodes involved have many connections

graded membership/production

The proposal that mental categories have a ______________ ______________- was tested in a series of experiments: 1) sentence verification task 2) ______________ task 3) rating task

propositions

The smallest units of knowledge that can either be true or false; Often expressed via simple sentences, but this is merely a convenience; other modes of representation are available such as a structure of nodes and linkages (example: "Children like candy" --> "children" or "candy" on its own are not qualified)

implicit theory/heuristic

To know what properties are essential for a particular category, we need a theory about what defines that category and how it relates to other category. Part of the mental representation of a category can be an _________________ __________________. Prototypes and exemplars are very useful for quickly identifying and categorizing things--> We can use them as a ________________ for categorization Our implicit theories can tell us when it is appropriate for us to use prototypes or exemplars, and when it is not - They can also tells us which properties are most important to consider when comparing against a prototype or exemplar

concepts/theme

We also draw on theories in thinking about new possibilities for a category- Allow us to decide what is practical and what is not Theories affect how quickly we learn new ______________________: - People learn a new category more easily if the features are coherent - People learn a new category more easily if they are given a ____________________ to explain category membership - People use theories to help them stretch concepts to encompass new, atypical examples People still rely on typicality heuristics even once they have established theories about a concept

beliefs/resemblance

When we are not guided by typicality in defining category membership, we focus on attributes that are believed to be essential for each category; Judgments about what is essential depend on ___________________ Prototype and exemplar views both depend on judgments of ___________________ (either to a prototype or some remembered instance)

perceptual/functional/interact

Why does the brain separate the concepts of living vs. non-living things in this way? One proposal emphasizes idea that different types of information are essential for different concepts : - Recognition of living things may depend on ______________ properties (especially visual properties) - Recognition of nonliving things may depend on _________________ properties Brain scans show that sensory and motor areas are activated when people are thinking about certain concepts (e.g., Thinking about the concept "kick" activates brain areas that control movement of the legs vs. Thinking about concept "rainbow" activates brain areas ordinarily involved in color vision) - Suggests that conceptual knowledge is intertwined with knowledge about what particular objects look like (or sound like, or feel like) and also with knowledge about how one might ___________________ with the object/concept

explanatory

_____________________ theories: - Provide a crucial knowledge base we rely on to think about a particular category - Serve the same function as a scientist's theory, though they are less precise (e.g., revealing cause-and-effect relationships); These cause-and-effect understandings may still be inaccurate even if they seem to make sense to us - Can increase the ease with which we learn a new category

basic-level categorization

a level of categorization hypothesized as the "natural" and most informative level, neither too specific nor too general; People tend to use basic-level terms (such as "chair" rather than the more general "furniture" or the specific "armchair") in their ordinary conversation and in their reasoning

distributed representations

a mode of representing ideas or contents in which there is no one node (or specific group of nodes) representing the content and no place where the content is stored; Instead, the content is represented via a pattern of simultaneous activity across many nodes within a network. The same nodes will also participate in other patterns, so those nodes will also be a part of other distributed representations

production task

an experimental procedure used in studying concepts in which the participant is asked to name as many examples (e.g., as many birds) as possible; According to prototype perspective, participants will accomplish this by first locating a prototype in memory for the target category and then asking themselves what else resembles or fits into this prototype

Ludwig Wittgenstein

argued that simple terms we all use everyday actually don't have a definition; proposed that members of a category have family resemblance- first to propose that we use category membership in a probabilistic way

categories/anomia

different brain sites support different ___________________ brain damage often causes ________________- an inability to name familiar objects; specific loss depends on where exactly the brain damage has occurred patients with damage in the brain's temporal pole had difficulty naming persons (only 59.8% correct) but were easily able to name animals and tools patients with damage in the inferotemporal region had difficulty naming persons and animals, but did somewhat better naming tools patients with damage in the lateral occipital region had difficulty naming tools but did reasonably well naming animals and persons

inferences

explanatory theories allow us to make _______________________ about how certain situations will unfold- or allow us to make attributions about why events unfold in a certain way; More complex; allow us to think about cause and effect (E.g., person jumping into a pool fully clothed might fit our theory of "drunkenness")

local

in propositional networks, individual ideas are represented with ________________ representation

adjustments/PDP (parallel distributed processing)

learning involves ______________________ of the connections among nodes, so that after learning occurs, activation will flow in a way that can represent the newly gained knowledge Learning involves the adjustment of Connection weights- the strength of the individual connections among nodes; In the ________________ model, learning requires the adjustment of many connection weights

Graded membership

objects closer to the prototype are "better" members of the category than objects farther from the prototype; membership in a category depends on resemblance to a prototype, and resemblance is a matter of degree

connectionist networks

proposed systems of knowledge representation that rely on distributed representations, and therefore require parallel distributed processing to operate on the elements of a representation (the opposite approach to propositional networks

similarity

prototype representation: - Single exemplar selected as prototype--> May be real exemplar, or combination of features from different exemplars - Membership in category determined by ________________ to prototype - Higher similarity = better member of category

exemplar-based reasoning

reasoning that draws on knowledge about specific category members, or exemplars, rather than drawing on more general information about the overall category; we categorize object based on their similarity to previously stored instances; used in cases in which categorization relies on knowledge about specific category members

necessary/sufficient

representing a category by defining features: example: triangle -Closed figure -Made of straight line segments -3 sides and 3 vertices each feature is ___________________ to be defined as a triangle Together these features are ___________________ to define the category. some instances where things may have defining features for a category, but don't seem to be particularly good members of the category (e.g., "bachelor" example)

propositional network

set of interconnected concepts and relationships in which long-term knowledge is held

Prototype theory

the claim that mental categories are represented by means of a single "best example," or prototype, identifying the "center" of the category; In this view, decisions about category membership, and inferences about the category, are made with reference to this best example; Often, the prototype is the average of the examples of that category that the person has actually encountered- Involves fuzzy boundaries, graded membership, and a perceived inequality of category members

typicality

the degree to which a particular object or situation or event is typical for its kind; Plays a large role in people's thinking- more-typical category members are privileged in many ways __________________ is to be expected if category knowledge is reliant on prototypes and exemplars

family resemblance

the idea that members of a category (e.g., all dogs, all games) resemble one another; In genera, relies on some number of features being shared by any group of category members, even though these features may not be shared by all members of the category

storage/inferences

why are categories useful? -simplify memory ______________ -allows ____________________ about new objects based on old members of same category