language-concepts and meanings Flashcards
(30 cards)
lecture 4 learning objectives
- Understand prototype, exemplar and theories approaches to concepts
- Be able to describe converging sources of evidence showing typicality effects
- Be able to describe potential problems with prototype and exemplar models
- Be able to describe evidence for a “concepts as theories” approach
-Understand how prototype, exemplar and theories approaches might combine
what are concepts
-mental representations of categories
-all words have an underlying concept
-not all concepts are labelled by a word eg ‘brown dog’
-
one way we conceptualise
dictionary definition approach
-dog
a dog is a creature that is
-mammalian
-has 4 legs
-barks
-wags its tail
this doesn’t fit well with the way people behave when we ask them to reason about concepts and categories
-when we get people in the lab nd aks them to think about concepts and categories they don’t behave as if this is the way they organise their information
dictionary definition approach problem
Ludwig Wittgenstein (1953)
-‘philosophers have tried for thousands of years to define terms such as ‘virtue’ and ‘knowledge’
-and there is always exceptions eg defining game
-an activity most often practiced by children
-engaged in for fun
-has certain rules
-involves multiple people
-is in some ways competitive
-is played during periods of leisure
-its difficult to create a definition for any given concept, always exceptions that don’t meet the rule
eg poker cant be played by children, games dont always involve multiple people, c
explain family resemblance in words
how does this look in psychology?
-Wittgenstein proposed that members of a category have a family resemblance to each other
-each category has ‘characteristic features’
-these are features that most category members have
-the more features an object has, the more likely it is to be in the category
how does this look in psychology
-prototype theory (Eleanor Rosch and colleagues , mid 1970s onwards)
what is prototype theory
prototype theory (eleanor Rosch and colleagues, mid 1970s onwards)
-when we think about categories, we have a prototype of that category and that’s how we make decisions about members of that category by reference to this prototype
-a prototype = the ‘average’ member of a category
prototypes in practice
-simple category like birds, your prototype bird is going to be your average of all the birds you’ve ever encountered.
-when you encounter a new object and trying to decide whether or not thats a bird, you can compare it to your prototype and if its close enough you can categorise it as a bird
implications of prototypes
-fuzzyboundaries
-graded membership
fuzzy boundaries - no strict dividing line between different concepts,
-you just have prototypes and things can be closer or further away from these individual prototypes , but there’s no point at which you can see if something is definitely part of a category
graded membership- (not all category members are equal)some members of the category are going to be better members of that category in that they’re closer to the prototype
prototype theory- evidence
sentence verification task
-consistent in labs when you get people to reason about categories
-you get participants in the lab and present them with sentences
,pressing one button if their true and button if its false
-you measure their reaction time (how quickly can they make these descion)
-eg sentences robin is a bird, penguin is a bird, camel is a bird
-interested if their faster to respond to robin then penguin
-they are faster in responding to robin, because robin is closer to the prototype of that category
What is an example of the Typicality effect?
For example, a robin is more often judged to be a typical member of the category bird than is a chicken.
prototype theory evidence
production tasks
eg list 5 vegetables
-people tend to be quite consistent in what they do in these tasks
-items closest to the prototype are likely to be named earliest
-the assumption is that their memory search starts with the prototype and works outwards
eg prototype veg is carrot and prototype fruit is apple
prototype theory evidence
picture identification
-participants see a series of pictures and press ‘yes’ if its a dog and ‘no’ if its not
-aa with sentence verification, responses to certain dogs are faster than others
-eg faster to say border collie is a dog than chihuahua
prototype theory evidence
rosch 1975 typicality studies
rosch
‘we all know that some birds are ‘birdier’ than others , some dogs are ‘doggier’ than others, and so on. I’m going to present you with a list of birds and dogs, and i want you to rate each one on the basis of how ‘birdy’ or ‘doggy’ it is’
-items farther from the prototype are rated as less ‘birdy’ / ‘doggy’
-people reason explicity as if they have a sense of what the prototype is.
prototype theory - evidence
-tasks that involve thinking bout categories
participants asked to make up sentences about a category eg birds
tendency to respond like
-;i saw two birds in a tree’
‘i like to feed the birds in the park’ but not something like ‘the bird was resting on an iceberg’
experimenter rewrites the sentence, substituting the name with a specific member of the category
‘i saw two robins in a tree’
‘i like to feed penguins in a park’
-then ask a new group of participants to rate the new sentences for silliness/implausibility
-people seem to think about the prototype when they generate the sentences, once you replace it with another example that isnt a prototype it does work
-this is all typicality effect(there are typical responses)
problems for prototype theory
-concepts can be very complex
-works well with dogs, birds and animals ,colours but not certain concepts
Australian aboriginal language of Dyirbal (Lakoff,1987)
-a book called ‘Balan’ - women, fire, and dangerous things- its called this because there’s a language called Dyirbal and has this categories for things, that are really important because they tell you how to use those words within a sentence
- so if a word falls into a certain category, you have the adjust the rest of the sentence to fit that
-this isn’t unusual in language eg French has dramatical gender which can change the sentence
- Balan is a category that primarily includes women fire and dangerous things ‘but it also includes birds that are not dangerous as well as exceptional animals, such as the platypus,bandicoot and echidna,’ p5
peoples prototypes seem to change with context
-eg European animals verses African animals
-if they had a prototype animal it shouldnt matter what the context is, but it does so suggests categories are not as simplistic
exemplar theory/ instance theory
-exampler theory says when we are reasoning about categories what we do is we loo k at our cumulative experience of individual instances of that category and we compare new objects against that
-if your asked to make decsion a piece of furniture eg is this a chair, apparently you compare the chair to your prototype of chairs
-another way we might do this is go through our memory of all the chairs weve encountered and if the object matches one of those then we say yes- this is an example of examplar theory
exemplar theory vs prototype theory
-definitions
exemplar based reasoning
-we categorise objects by comparing them to examples of a category in memory
prototype theory
- we categorise objects by comparing them to a mentally represented ‘standard’
exemplar theory and typicality effects
exemplar theory is fine with typicality effects
-‘a robin is a bird’
-our memory search will be very fast- robins are often encountered
-‘a penguin is a bird’
-our memory search will be somewhat slower-penguins are a less common occurrence
exemplar theory
-why people preserve information about variability within a category when the perform tasks
rips and collins (1993)
rips and collins (1993)- people seem to be sensitive to the nature of the category
study done in the 90s-wanted to know whether people were sensitive to the nature of a category when they performed a task
-they gave the example of a pizza and a ruler
-average pizza and ruler back then were both 12 inches long (but pizzas are much more variable than rulers, you get all different sizes of pizzas)
-they were interested if people reason differently due to this (in prototype theory this should not matter)
-asked participants whether a new 19” object was likely to be a pizza or a ruler
-prototype theory says you should get an even 50-50 split of responses
-but participants usually say ‘pizza’
exemplar theory preserve information about this kind of variability
prototype vs exemplar
which one is better?
-likely we reason using both
consistent with exemplars
-people seem to be able to adjust their categories
-eg if asked to make judgements about European versus African animals
-consistent with exemplars - we bring different exemplars to mind in different situations
-exemplar theories overall seem to provide a slightly better account of peoples performance
-but prototypes are a good quick measure,
prototypes and exemplars-developmental perspective
what happens when a child first encounters eg a dog
eg kids can figure out dogs (even from weird cartoons) and know whats a dog and whats not regardless if its hard to tell and regardless of the variability in dogs
-whats likely going on is that when they initially encounter these objects they probably use exemplar based reasoning
-so when they first learn what a dog is they have one example to go off of and the first few times all the can do is add new members to the category as they encounter more dogs
-but the more they see them they start to develop a prototype for the category so they start to average them together
-movement from exemplar to prototype reasoning seems plausible
-this is consistent with adults in labs
problems with typicality
keil(1986) study
keil (1986) study with children as participants
-they asked children to reason about different categories -ask them things like could you turn a toaster into a coffee pot
-kids often acknowledged this would be possible
then they would be asked could you turn a skunk into a racoon?
-they understood you could paint and teach a skunk to be like a racoon
-but they wouldn’t accept that it would be a racoon
-resistant to the idea a species could change
-argued that children have theories about how world works
-info like something that’s mechanical you can change its function
-but animals you cant change their species
problems with typicality
sea lampreys versus whales (McCloskey and glucksberg 1978)
-particpiants judge whales to be more typical of the concept ‘fish’ than sea lampreys
-but they still respond that sea lampreys are fish and whales are not
-people have an awareness of typicality but that’s not how they reason about these animals.
-suggests they have an additional layer of information that they use
prototype and exemplars account for much of how we reason about concepts…but..
but other beliefs and knowledge must also contribute to our reasoning about categories
concept as theories
murphy and medin 1985
clean vs unclean animals bible
-what makes these categories sensible?
-idea that concepts are a theory about the way a wordl works
example they used etc
-clean versus unclean animals in the bible
-clean animals include gazelles,frogs,most fish,grasshoppers, and some locusts
-unclean animals include camels,ostriches,crocodiles,mice,sharks, and eels
what makes these categories sensible or coherent?
a theory or explanatory framework
-theory that governs if its clean or unclean relating to :
-relating to habitat,biological structure and form of locomation
