Problem 4 - Conceptual Knowledge

Question 1

Category

Answer 1

• set of objects that can be treated as equivalent in some way.
• they share many properties and are informative

Question 2

Concepts

Answer 2

• mental representations formed of categories
• core of intelligent behavior
• allows to extend what you have learned about a limited number of objects to a potentially infinite set of entities.

Question 3

Nature of categories

Answer 3

• well defined = definitions about what is in and out of the category.
• provides necessary feature for category membership
• the features must be jointly sufficient for membership
• limitation: clear cut boundaries while the world is not.

Question 4

Fuzzy categories

Answer 4

• categories have unclear boundaries that can shift over time.

Question 5

Fuzzy categories: borderline members

Answer 5

• outliers to categories or items

Question 6

Fuzzy categories: typicality

Answer 6

• some items are more typical than others.
• most important variable in predicting how people interact with categories.
• category prototype: most typical category member. Other items are compared to this.
• influences of typicality on cognition: judgement, speed, memory, accessibility, understanding and repeatability.
• source: family resemblance theory (features frequent in other categories enhances typicality) and frequency.

Question 7

Category hierarchies:

Answer 7

• most concrete categories are nested inside larger, abstract categories.
• the basic level: not too small, not too big but just right and from a neutral situation = easy to learn. Experitise, quality of life and upbringing influences this.
• explanation: category members are similar to one another but they are different from members of other categories.

Question 8

Prototype theory

Answer 8

• general description learned that applies to the category as a whole.
• weighted features by frequency
• typical members have high weight = easier to match

Question 9

Exemplar theory

Answer 9

• theory denies the general description of a category
• claims that concepts are remembered examples.
• classification happens through comparison of previously seen items.
• close similarity has a large effect on classification.

Question 10

Modern ideology

Answer 10

• concepts are represented through multiple cognitive systems: both general descriptions and exemplars.

Question 11

Semantic networks

Answer 11

• knowledge is stored in the form of associative networks = concepts are represented by nearby nodes corresponds to related concepts.

Question 12

Semantic networks: category verification task

Answer 12

• used to determine how we access categorical knowledge.
• participants as to verify or deny simple statements
• accuracy not important
• speed is important = connections and proximity of the features.
• the process where the activation of one nodes spreads to other, related nodes

Question 13

Semantic networks: feature verification tasks

Answer 13

• used to asses how the features of categories are stored and accessed.
• participants asked to verify or deny simple statements
• accuracy not important
• speed is important = connections and proximity of the features.
• the process where the activation of one nodes spreads to other, related nodes

Question 14

Semantic networks: spreading activation model (collins and loftus)

Answer 14

• Assumptions about representation of knowledge and aout nodes being lined in an associative network.
• the strength of activation decreases as a function of time, distance and number of concepts activated (more concepts, the less activation any once concepts receives).
• the activation taht reaches any concept node is summed up = after threshold, concept is activated
• semantic priming: tendency for the processing of one stimulus to enhance/speed up the processing of another related stimulus.

Question 15

Functions of concepts

Answer 15

• building blocks of thought
• sever as mental shorthand that allows for quick and efficient understanding.
• going beyond the present and making predictions.
• infer knowledge not explicitly related
• support new learning
• important for communication

Question 16

Categories as a concept:

Answer 16

1. Natural kind: they define themselves. They share characteristics and are labeled after discovery.
2. Artifacts: includes objects/conventions designed by humans to serve particular function. They dont share the same basic features.
3. Ad hoc: formed in the service of the same goal. They do not share any characteristics.

Question 17

Similarity based categorization: classical view

Answer 17

• items are classified into category if they have certain feature = both necessary and sufficient
• limitations: difficult to specify, cannot explain typicality, boundaries are too clear cut.

Question 18

Similarity based categorization: prototype approach

Answer 18

• specific features of the category that members are likely to have.
• evaluated and classified based on ressemblance to other members
• high family ressemblance = typical member
• abstract through repeated experience = most representative, quickly and easily accessed, distance between labels and prototype is short.
• solves the rigidity, typicality and fuzzy boundary problems.
• limitations: categories are more complex and sensitive to context.

Question 19

Similarity based categorization: exemplar approach

Answer 19

• represent categories in terms of examples
• extreme version: no abstraction or generalization process = every single encounter
• typicality effect: no problem here (more likely to retrieve items that been encoded frequently).
• biasing effect: context is no problem (content activates certain examples due to priming retrieval).
• sensitive to correlations: typical = high correlation, less typical = low correlation.
• limitation: abstract representations, economy (not every example is stored).

Question 20

Similarity based categorization: knowledge approach

Answer 20

• connecting concept to existing knowledge
• psychological essentialism: believe categories have underlying properties that are found only in that category.
• decisions made through assumptions
• signs of essentialism: in or out of the category, resistance to change, essence is passed on to progeny.
• limitation: no statement about typicality, too difficult/broad, boundaries too clear-cut.

Question 21

Conceptual knowledge using picture drawing in semantic dementia: abstract

Answer 21

• 6 patients with semantic dementia asked to produce drawings of concrete concepts from dictation
• drawing were characterized by a loss of distinctive features
• artifact domain = feature loss resulted in box-like representations.
• living domain = distinctive features lost and tendency to include incorrect features that resulted in more familiar and prototypical representations of bigger concept.

Question 22

Conceptual knowledge using picture drawing in semantic dementia: introduction

Answer 22

• semantic memory = knowledge about concepts, facts and words/meanings
• semantic demantia happens late in life
• aim: investigate the structure and internal representation of visual conceptual knowledge

Question 23

Conceptual knowledge using picture drawing in semantic dementia: methods

Question 24

Conceptual knowledge using picture drawing in semantic dementia: results

Answer 24

• less target features drawn
• either less features or included more features that were typical in neighboring categories

Question 25

Conceptual knowledge using picture drawing in semantic dementia: discussion

Answer 25

• supports prototype theory: hierarchy of things
• against exemplar approach: it should be more likely that the distinctive features will be kept.