1304944
| Question | Answer |
| How do people demonstrate categorization? | Bruner - conceptual grouping via behaviour - eg petting dogs Different behaviour = think differently about different concepts Ross & Murphy - sorting tasks Eggs = "Breakfast" or "Dairy" |
| Cognitive reasons for concepts? | Makes it easier to remember stuff Semantic classification for recognition Units of semantic memory - facts; r'ships between concepts Enable inferences - reasoning Lexical concepts - 'dog' represents what we believe 'dog' to mean |
| Criticisms of the Classical view of concepts? | 1) Typicality: how closely item fits category - eg Robin more typical bird than penguin Rosch - Ps verified sentence "Robin is a bird" quicker than "Penguin is a bird" 2) Borderline cases: McCloskey & Glucksberg - Ps rated items inconsistently @ different times eg bookends as furniture 3) Intransitivity: If A is in B & B is in C, then A is in C Hampton - Ps judgements not fit w/ rule eg car seats 4) Lack of defining characteristics: all members share common properties? eg Games? (card, board, Olympic) Wittgenstein ('53) - overlapping set of family resemblances 5) People vary in knowledge of taxonomies |
| Main points of Prototype theory of concepts? | Rosch '75 >Prototype = most typical exemplar of category >Item = member if match enough properties/similarity to prototype >NO necessary condition >A property can have many values (eg apple colour) >Highly typical instances match highly weighted properties >Categories have rich internal structure |
| Criticisms of Prototype Theory? | 1) Meaning of Typicality Effects: Armstrong et al - existence of typicality effects aren't conclusive evidence that membership determined by similarity to prototype. Suggest Dual Process Model i) Concept Core, ii) Set of identification procedures 2) Context Sensitivity: Typicality effects vary with context eg Medin & Shoban - wooden & metal spoons. Metal more typical than wooden; small more than large; BUT large wooden more than small wooden!! 3) Complex concepts: combinations of concepts (eg red car). Difficult to explain how to combine prototypes. Different concepts combine eg stone lion 4) no agreement on what is prototypical 5) cant explain acquisition of non-referential words (eg hot-dog) |
| What are the main points for the Common-sense theory of concepts ('Theory' theory) | >Issues with similarity based approaches (classical & prototype) - deeper analysis req'd >Explanation based >Other factors affect accuracy re prediction of category membership >Problems defining similarity - what properties relevant? >Items can share infinite things in common (eg Plumb and Lawnmower - Murphy & Medin '85) >Categorisation draws on background knowledge/beliefs >ad-hoc categories |
| What is the evidence for the Common-sense theory? | Medin & Murphy ('85)- categorisation = part of process to explain shared behaviour. May not be awareness of theories apart from novel or borderline items Rips - Pizza & US quarter; 3rd object more like pizza but more similar to quarter = dissociation between category & similarity ie underlying theory not just similarity based Kiel - developmental aspects. Discovery: Hybrid animal - zebra w/ horse insides 4yrs - Zebra (appearance) 7yrs - Horse (lineage) Transformation: Raccoon > operation > skunk 4yrs - Skunk (appearance) 7yrs - Raccoon (lineage) "The characteristic - defining shift" w/ age |
| What are the main criticisms of the Common-sense theory? | 1) Just because it suggests deeper knowledge & not just similarity - doesn't make it more correct 2) "Theory" poorly defined; "common-sense" not compatible with science theory. More like "knowledge" - Murphy 3) Does not explain complex concepts 4) May just replace 'similarity' with 'theory' Both wooly terms! |
| Main points of the Psychological Essentialism approach | >Kripke '72, '80 & Putnam '75: Things have an essence, whether we can define their nature or not - eg children may have belief that m/f essentially different w/o knowing what that difference entails ('Placeholder' Medin & Ortony) >Assumes deeper principles >Objects categorised according to unobservable shared "essential" properties - makes category members what they are >Superficial properties likely known, may not be able to define essential properties >View that certain categories have an underlying reality or true nature that cant be observed directly but that gives an object its identity, and is responsible for other similarities that category members share |
| Evidence for Psychological Essentialism approach | >Innate Potential: Important evidence for essentialism is belief that properties are fixed at birth, ie an organism displays innate potential Paradigm - test children's beliefs re innate properties eg animal switch from its bio parents to new ones (kangaroo & goat) Pre-school kids - it will have pouch & hop ie innate kangaroo properties >Underlying structure: Gelman & Wellman ('91) - young children = insides more import re id of category Is a dog still a dog if you take away its: Insides> 'no' Outsides> 'yes' >Inductive Potential: Capacity to extend knowledge to novel instances eg inferring a newly encountered mushroom is poisonous on the basis of past encounters w/ other poisonous mushrooms. Gelman & Markman - leaf, bug & leaf insect Young children - words signify category membership even when properties not obvious Older children - other criteria |
| Criticisms for Psychological Essentialism approach | 1) Influence of context & perspective: Malt - Ps more influenced by source of water eg tears ≠ water Suggests essentialism not work - context & perspective affect consistency & contradiction of category Ps response not just determined by molecular structure (essence) but source, function etc 2) Expert opinion: Malt - more Ps ask expert re borderline natural item than for artifact (eg soap) Suggests natural category may involve psych essentialism as Ps recognise info experts bring 3) Indirect evidence Some evidence needs inferences (eg insides = essence) 4) Failure to explain complex concepts 5) Nature of experiments with children! |
| What are the main flaws of each approach? | Classical - Necessary & sufficient conditions only work for few categories Prototype - Doesn't explain context sensitivity or complex concepts 'Theory' theory - Imprecise, no definitions, doesn't explain complex concepts Psych Essentialism - Mixed evidence, much doesn't map to 'essences' |
| What type of scenario could each approach be used for? | Classical - for definitions Prototype - rapid categorisation, fuzzy matching Theory based - considered judgements, explanation for category membership Psych Essentialism - integrate scientific knowledge, expert view |
| Are all categorisers the same? | Different approach could depend on knowledge/skill Medin - trees. Taxonomists, maintenance workers, landscapers all asked to categorise different trees. All used different approach Lynch et al - Typicality ratings vary between expert & novice |
| Prototype theory - levels of category structure? | Vertical organisation: Superordinate (Animal) Basic (Dog) Subordinate (Collie) |
| Prototype theory (Rosch) - typicality weighting? | Instances that differ in typicality are assumed to differ in terms of the weighting of values on which they match the concept eg difference in typicality between an apple and a fig is reflected in a difference in the weighting for an apple and a fig |
| Evidence for prototype theory (Rosch) | Explains over-extension (hard to define concepts) Explains typicality Handles irregular items (eg traditional window & in a computer system) |
| Main points of the classical view? | >Features are primitives >All category members equally good - necessary condition >All concepts represented in this way - sufficient condition >Everybody represents concepts in the same way >Hierarchical - super-ordinate<>sub-ordinate >Inheritance - Sub-ordinate inherit super-ordinate properties & have additional features >Concepts divide the world into distinct classes & boundaries between classes well defined & rigid |
| "Ideal is typical" - Burnett et al '05 Main points? | Expert fisherman - card sorting task Typicality ratings of freshwater fish Typical = desirability Different cultural groups, differing ideas re fish: Native Americans (NA); Majority Culture (MC) Results: Typicality based on desirability>centrality (supported) Main effect of cultural group - NA gave higher overall ratings (supported) Culture x fish type interaction - NA higher ratings to sturgeon, trout (supported) |
| Psychological Essentialism in Children - Gelman '04 (extra paper) | >Certain categories (eg lion or female) have underlying reality - cannot be directly observed >Psych essentialism is early cognitive bias >Young children look beyond the obvious: *Learning words *Generalising knowledge of new category members *Reasoning about the insides of things *Beliefs re nature v nurture *Constructing causal explanations >Unlikely language is source of psych essentialism but may play indirect role >Argues against view of children as 'concrete thinkers' (eg Piaget) - claims children have early tendency to search for hidden, non-obvious features |
| Evidence for the Classical View? | >Empirical investigations (e.g. Hull, 1920; Bruner et al., 1956) - people categorized instances according to whether they possessed the necessary and sufficient conditions of the category >Sporadically defended (e.g. Sutcliffe, 1993) |
| Classical View - main points? | >Roots in philosophical writings of Aristotle (Sutcliffe, 1993), >Things belong to categories as possess certain properties in common >If something is a member, it must possess the properties common to the category’s members >If something possesses the properties common to a category’s members, then it too must be a member of the category >Possession of the common properties is necessary for category membership >Possession of the common properties is sufficient for category membership. >Classical view is that concepts provide definitions of their corresponding category |
Want to create your own Flashcards for free with GoConqr? Learn more.