Categories

1
Categories

• My dog sleeping. My dog. All golden retrievers.
  All dogs. All canines. All mammals
• Each of these is a category.
• Categorization is the process of deciding which
  details matter, and which dont, for some purpose.

2
Bruner, Goodnow, Austin (1956)

• Categories
• reduce complexity of environment
• allow us to generalize lessons
• guide choice of response
• make hierarchical knowledge available

3
Two questions about categories

• 1. What is the structure of natural categories
  like?
• This is a question about the world.
• 2. How is information about natural categories
  represented in memory?
• This is a question about your mind.

4
1. The structure of categories in the world

• Is the question below a question about the world,
  or about us?
• Which two are most similar sheep, goats,
  cows?
• To some extent, the structure of natural
  categories is given by the world.
• To some extent, it is impressed upon the world
  by human cognition

5
1. The structure of categories in the world

• Most important work done by Eleanor Rosch.
• Rosch argued that our categorical knowledge is
  organized in a hierarchy
• Superordinate, basic level, subordinate.
• This is a relation of containing.

6
Mammal Dogs Cats Horses Collie
Airedale Persian Siamese Arabian
Superordinate Basic Level Subordinate
7
What is this?
8
The Basic Level

• The basic level falls between superordinate and
  subordinate levels.
• Its the one we use when we name an object.
• Its the one children learn first.
• Things in a basic level category look like other
  things in the same category but not like things
  in other categories. This quality is not true at
  other levels.

9
Furniture Types of chair

• Table
• Chair
• Lamp
• Bookcase
• This is a super-ordinate category

• Dining room table
• Patio table
• Coffee table
• Picnic table
• This is a subordinate category

10
Review Hierarchy according to Rosch

• Things in the world present themselves in a
  hierarchy of levels of categorization
• Basic level, items in a category look like each
  other but not like members of other categories.
• Basic level is first one learned and one used
  spontaneously in naming objects.

11
Roschs second contribution - Typicality

• Rosch argued that some members of a category are
  better than others that is, more typical.
• such members have family resemblance.
• typical members can be verified fastest
  (implying fast access to their repns.)
• typical members are similar to other members,
  unlike non-members of category

12
Things in the world vs. in the head

• So far, weve been talking about things in the
  world. We asked, how do these things assemble
  into categories?
• Now, we turn to the question of things in your
  head.
• What is the nature of your knowledge about
  categories? How are they represented?

13
2. Mental representation of categories

• Three kinds of categories
• Natural (e.g., Mammals)
• Artificial (e.g., Animals that weigh gt 100
  pounds)
• Functional (e.g., Things to bring out of the
  house in case of fire.)
• Well consider only natural categories

14
Mental representation of natural categories

• There are four competing models. Each specifies
  how a decision on category membership is made
  (how do you decide if this is an X?). Note they
  might all be wrong.
• 1. Prototype
• 2. Feature frequency
• 3. Nearest neighbour
• 4. Average distance

15
Prototype models

• A prototype is a typical member of a category
• Prototype theories say that, through experience,
  we create a central example of each category, and
  store that example.
• A prototype captures what is typical of a
  category
• A prototype may exist only in your mind (e.g.,
  not as an actual object in the world).

16
Feature frequency models

• Categorization is based on how many features the
  to-be-classified object shares with each of the
  available categories.
• E.g., a whale shares breathes air and gives
  birth to live young with mammals. It shares
  lives in the ocean and moves by tail and
  flipper action with fish.
• So a whale could be a fish or a mammal
• Models predict confusion about whales

17
Nearest neighbour models

• New object is compared with each exemplar of each
  stored category.
• Difference (on any dimensions) between object
  and each exemplar in each category is computed.
• New object is classified in same category as
  object it is most similar to (smallest
  difference).

18
Category Clothing
19
Average distance models

• Comparison of new object to all stored exemplars
  of categories as in N. Neighbour.
• Object goes in category with smallest average
  distance (exemplars to object).
• Compare with Nearest Neighbour model here, it
  is average distance for the category, not just
  which exemplar is closest, that counts.

20
Review models of representation

• These models reflect 2 very different views of
  category representation
• Prototype model
• what is generally true about something is
  stored and available when needed
• this view emphasizes abstract representations
  (that is, not much detail)

21
Review models of representation

• N.N. and A.D. models
• what is generally true is not stored, but
  computed when needed.
• these views emphasize storage of individual
  experiences with objects rather than storage of
  abstract essences.