CATEGORIZATION

1
CATEGORIZATION

• Classical view (Aristotle) categories are
  defined by a set of necessary and sufficient
  conditions
• example BACHELOR
• each of these four features (human, unmarried,
  adult, male) is necessary
• Together, the four features are sufficient.

2

• Problem difficult to identify the necessary and
  sufficient features
• Ludwig Wittgenstein used as an example the
  category GAME.
• Another problem no basis for distinguishing
  among the members of a category - no levels of
  membership

3

• Categories should have clear and well-defined
  boundaries
• experimental observation suggests that categories
  have a graded structure with fuzzy boundaries
• some members of a category are better members
  than others

4

• Returning to our example
• BACHELOR
• 35 year-old unmarried man?
• 21 year-old unmarried man?
• 17 year-old unmarried man?
• Is a divorced man a bachelor? A priest? The Pope?

5

• Eleanor Rosch - proponent for the view that
  categories have a fuzzy and graded structure
• investigated how people use categories
• she asked people to simply rate on a 7-point
  scale how well a word fits a category

6

• Example category VEGETABLE
• PEA and CARROT were rated as very good exemplars
  (1.07 and 1.13)
• ONION (2.52) and YAM (3.31) moderately good
  exemplars
• PEANUT (5.36) and RICE (5.59) poor exemplars

7

• it takes a shorter amount of time to decide that
  POTATO is a VEGETABLE than to decide that OLIVE
  is a VEGETABLE
• This finding is known as the Typicality Effect
• in contradiction with the view that category
  judgments are based solely on a fixed set of
  necessary and sufficient features

8

• Two main theories of categorization
• prototype theory
• a mental representation of the average or most
  typical member of a category
• prototype does not exist, abstract notion
• exemplar theory
• individual entities representative of a category

9
EXPERIMENTAL PROCEDURE

• TASK is this word a member of the given
  category?
• 64 trials
• 8 with each of 8 categories
• FURNITURE, FRUIT, WEAPON, VEHICLE, TOOL, BIRD,
  SPORT, CLOTHING

10

• On half of the trials the exemplars will be
  category members, on the other half not.
• On half of the YES trials the exemplar will be a
  highly typical member
• (BIRD-ROBIN).
• On the other YES trials the exemplar will be
  moderate or low in typicality
• (FURNITURE-BENCH, BIRD-PENGUIN).

11

• The experimental trials are presented in a
  different random order for each student.
• The task is to decide whether the probe is a
  member of the category.
• If the answer is incorrect, feedback will be
  given
• respond as fast as possible, but also try to
  avoid making errors.

12
Design

• IV typicality (high, low)
• DV1 RT
• DV2 accuracy

13
DISCUSSION

• Compare means of RT and accuracy
• benefits enables useful and intelligent
  inferences
• costs stereotyping

14
How This Experiment Relates to the Real World

• Categorization is fundamental to almost
  everything we do.
• How do children learn categories?
• Are categories real or in the mind of the
  categorizer?
• Stereotypes illustrate the costs of categorization