CHAP 9: CONCEPTS & GENERIC KNOWLEDGE
we need a way of identifying concepts that highlights what the various members of a category have in
common. while simultaneously allowing exceptions to whatever rule we propose.
one way to achieve this is by keeping the content of our definitions but being more flexible in our use of
the probabilistic phrasing of a sentence preserves what is good about definitions - the fact that they do
name sensible, relevant features.
but this phrasing also allows some degree of uncertainty.
Wittgenstein's notion was that members of a category have a family resemblance to each other.
family resemblance: notion that members of a category resemble each other. Family resemblance
relies on some number of features being shared but may shift from one subset of the category to
imagine the "ideal" for each family - someone who has all of the family's features; this person may
each member of the family has at least some features in common with this idea, and therefore some
features in common with other family members
proposed that thousands of categories we can think about (e.g. dog, game, furniture) work in the
we can identify "characteristic features" for each category - features that most category members
PROTOTYPES AND TYPICALITY EFFECTS
one way to think about definitions is that they set the "boundaries" for a category
prototype theory: the claim that mental categories are represented by means of a single "best example",
or prototype, identifying the "center" of the category.
may represent the 'ideal' for the category
all judgments about a dog for example, are made with reference to this ideal.
categorization; deciding whether something is a dog or not would involve a comparison between a
test case and the prototype represented in your memory.
if there is no similarity between them, then the creature before you is probably not in the category,
if there is some similarity, then you identify it in the 'dog' category.
a prototype will be an average of the various category members you have encountered
o e.g. the prototype dog will be the average colour of the dogs you've seen, average size of
dogs you've seen, etc.
the prototype will serve as a benchmark, an anchor for our conceptual knowledge
FUZZY BOUNDARIES AND GRADED MEMBERSHIP
what it means to "know" a concept is simply to have some mental representation of the concept's
since the category is characterized by its center (the prototype) and not by its boundaries, there's no way
we can say whether something is inside of the category or outside.
each category will have a fuzzy boundary with no clear specification of category membership and non-
categories that depend on a prototype have graded membership: the idea that some members of a
category are "better" members and are more firmly in the category than other members.
e.g. some dogs may be more "doggier" than others TESTING THE PROTOTYPE NOTION
in a sentence verification task, the speed of response varies from item to item within a category.
e.g. response times are longer for sentences like "a penguin is a bird than for "a robin is a bird"
this is because participants make these judgements by comparing the test item to their
prototype for that category.
when there is similarity between the test case and the prototype, participants can make their
in a production task (a task where a person is asked to name as many example as possible), according to
a prototype view, they will do this production task by first locating their bird or dog prototype in memory
and then ask themselves what resembles this prototype.
they will start with the center of the category (the prototype) and work their way outward from
what matters in both tasks is proximity to the prototype
over and over, in category after category, members of a category that are "privileged" on one task, turn out
also to be privileged on other tasks.
various tasks converge in the sense that each task yields the same answer.
category members mentioned early in production task are also "privileged" in a picture-identification
task, yielding faster responses.
the same category members are privileged in a rating task.
when people think about a category they are thinking about the prototype for that category
people will come up with appropriate statements for that prototype
Rosch argued that there is a "natural" level of categorization, neither too specific nor too general, that we tend to
use in our conversations and reasoning.
basic-level categorization: usually represented in our language via a single word, while more specific
categories are identified only via a phrase.
e.g. "chair" is basic level, where as "lawn chair" is not so basic
if asked to describe an object, we are likely spontaneously to use the basic-level term
if asked to explain what members of a category have in common with one another, we have an easy time
with basic-level categories, but some difficulty with more-encompassing categories.
the importance of basic-level categories also show up in our memory errors.
if a story contains general terms, they're normally misremembered as being more specific than they
the errors tend to "revise" the story in the direction of basic-level categorization
our conceptual knowledge is represented via a prototype and that we categorize by making comparisons to
ANALOGIES FROM REMEMBERED EXEMPLARS
in some case, categorization can draw on knowledge about specific category members rather than on more
general info about the overall category.
exemplar-based reasoning: reasoning that draws on knowledge about specific category members,
rather than drawing on more-general info about the overall category.
it can be supported by memories of a specific category
you categorize objects by comparing them to a mentally represented "standard".
for prototype, the standard IS the prototype
for exemplar, the standard is provided by whatever example of the category comes to mind
the judgment is the same: assess similarities between the candidate object and this standard EXPLAINING TYPICALITY DATA WITH AN EXEMPLAR MODEL
exemplar-based can also explain the graded-membership pattern
people make their judgements by comparing the pictures to specific memories of an object, mentally
if a picture shows a co