Conceptual Knowledge I: Categories in the Mind

Question 1

Mental Representation and how is semantic knowledge represented?
– questions

Answer 1

What is the structure of memory, or relationship between memories?

How are objects grouped together in the mind?

Question 2

Categorization - Philosophers have noted the difficulty of defining categories?

Answer 2

Things that are challenging to define:
Family
Games (Wittgenstein, 1953)
Love

Question 3

Does the mind use category definitions? + Family resemblance

Answer 3

Probably not
- Not all members of common categories share the defining features
- It is computationally intense to match objects on many feature dimensions

Wittgenstein proposed family resemblance:
- Members of a category resemble each other in some way

Question 4

Categorization based on Prototypes?

Answer 4

• A prototype is the ‘typical’ member
• Prototypes is formed by averaging together the commonly experienced members of a category
• Therefore, objects within a category each have a distance from the average: typicality

Question 5

Typicality Study (Rosch, 1975)?

Answer 5

Participants told a category, then shown 50 members of the category
- Asked to rate how well each member represents the category

Bird? 1-10
Sparrow 10
Penguin 2

Results support prototype approach because members vary in distance from an ideal

Question 6

Family Resemblance study (Rosch & Mervis, 1975)

Participants asked to list defining characteristics of several objects, which belonged to the same category
- E.g., Furniture: Chair, sofa, mirror, telephone

Answer 6

• Objects close to each other (chair & sofa) shared many characteristics.
• Less overlap for others (mirror & telephone)

Results support relationship between family resemblance andprototypicality

Question 7

Categorization Speed Study? Will be on exam!
Participants asked to respond yes/no to category statements:
– “An apple is a fruit”
– “A pomegranate is a fruit”

DV: response time

Answer 7

Result: highly prototypical objects judged faster
-> Typicality effect!

Question 8

Prototypes and Priming?

Answer 8

• Recall repetition priming
• Category priming also facilitates perceptual speed

Question 9

Prototypes and Priming (Rosch, 1975)

Participants were shown two colours, and were asked to respond if they match each other or not (same/different judgement task)

But before each trial, participants were primed with a category colour (e.g., “green”)

IV:
a) Same + good example of prime
b) Same + poor example of prime

DV: Response time

Answer 9

Results: Participants responded faster when colours were good example of the prime

Results suggest prime contained some information that could help speed perception
- But was more helpful when stimuli were more prototypical of the prime category

Question 10

Definitions vs. Prototypes?

Answer 10

These experiments showed:
- Category membership is not black and white
- Members vary in how well they fit
- Prototypicality helps processing speed, above definitions

Prototype approach is a good way to understand mental representation of objects/categories

Question 11

Exemplar approach?

Answer 11

• While prototypes are average examples of members
• Instead of comparing objects to a prototype, perhaps the mind compares objects to a set of common exemplars

Question 12

Advantages of exemplar approach?

Answer 12

• Can more easily account for categories with loose members
• Can more easily deal with categories that are hard to average together (i.e, games)

Both Prototype and Exemplar approaches are useful

Question 13

Categories of Categories - Hierarchical organization?

Answer 13

• Any object can fit into several categories
• Categories vary from specific to general

Global (Superordinate)
Basic
Specific (Subordinate)

Question 14

Is there a basic category level?

Participants asked to list as many features as possible of different categories
– Furniture (general)
– Table (hypothetically basic)
– Kitchen table (specific)

DV: number of features listed

Answer 14

Global (Furniture) - Lose a lot of information, 3 common features

Basic (Table) - 9 common features

Specific (Kitchen table) - Gain just a little information, 10.3 common features

Rosch concluded that basic categories (i.e., “table”) are psychologically special because they maximize information and generality

Question 15

Naming things Study?

Participants asked to name a variety of objects

Answer 15

Results: “basic” labels more common
– “guitar,” not “instrument”; “pants,” not “clothing”

Question 16

Priming Study:

Participants primed with a category label

Then shown an object

Task: identify whether object belongs to category or not

DV: response time

Answer 16

Result: response time faster when prime was “basic” category, not too general or too specific

Basic-level categories facilitate processing better!

Question 17

Expert Knowledge and Categorization?

Answer 17

Experts tend to use specific categories as their “basic” categories

Experts have a richer category structure
- More fluent perception within area of expertise
- Richer experience?