Conceptual Knowledge Part 1 - Lecture Notes

Updates and Reminders

• No in-person class on Thursday (4/3) - Recorded lecture instead!
• Exam 2 Update.
• Blackboard assignment due Sunday.

Cognitive Processes

• Attention.
• Short-term / working memory.
• Long-term memory.
• Everyday memory & memory errors.
• Conceptual knowledge.
• Problem-solving and creativity.
• Judgements, decisions, and reasoning.
• Perception.

Today's Roadmap

• Part 1: Introduction to concepts and categories.
• Part 2: Categorization theories.
• Part 3: Levels of categories.

Part 1: Introduction to Concepts and Categories

• Concept: the mental representation of a class or individual that gives it meaning.
• Conceptual knowledge allows us to recognize objects and events and to make inferences about their properties.

Why are Concepts Important?

• To recognize objects and events.
• To make inferences about their properties.
  • What is this item we’ve encountered?
  • What makes it a pumpkin?
  • How do we tell it apart from other things?
  • What do we do with it?
  • Is it safe?
  • Can we eat it?

What is a Category?

• Category: group of objects or entities that have something in common.
  • Defined by the concept.
  • Includes all possible examples of a particular concept.
• Categorization: the process by which things are placed in categories.

Why are Categories Important?

• Allow for generalization across members of a category.
• Allow us to make predictions about new objects based on previous experience with related objects.
• Allow us to communicate our unique experiences to others.
• "Pointers to knowledge" (Yamauchi & Markman, 2000).

Key Questions

• How are different objects, events, or ideas assigned to a particular category?
• How can categories be defined?
• Why do we say that “not all things are created equal”?

Part 2: Categorization Theories

• Categorization theories:
  1. Definitional theories.
  2. Prototype theories.
  3. Exemplar theories.
  4. Theories and rules.

1. Definitional Theories

• We decide whether something is a member of a category by determining whether the object meets the definition of the category.
• Categories have a clear set of defining features.
  • Necessary: an object must have the defining features to be a category member.
  • Sufficient: an object must have all of the defining features to be a category member.
• Prediction: all members of a category should be equally easy to identify.
  • Determine what features the new object has.
  • Compare them to the defining features of each category.
  • Pick the category that has all its defining features satisfied.
• Limitations of this view:
  • Defining features don’t exist for all categories (e.g., chairs, games, furniture).
• Possible solution = family resemblance (Ludwig Wittgenstein, 1953).
  • Category members can resemble each other in different ways.
  • There is no set of characteristics that all category members must share.
  • Allows for some variation within a category.
  • For example: What defines a game?
    • Practiced by children.
    • Engaged in for fun.
    • Has rules.
    • Involves multiple players.
    • Is competitive.

2. Prototype Theories

• Category membership decided by comparing the object to a prototype that represents the category.
• Prototype: a summary or average of all experienced members of a category.
  • Captures the family resemblance of category members - is like a “typical” member.
  • Does not have to be a real exemplar from the category.
  • Objects are categorized by finding their most similar prototype.
• Eleanor Rosch (1975a) Demonstrated that typicality is an important factor in categorization.
  • People consistently rate typical category members as being better category examples.
• Connections between prototypicality and behavior:
  • Mervis and coworkers (1976)
  • Smith and coworkers (1974)
  • Rosch (1975)
  • Solso and MacCarthy (1981)
• Prototypical objects are named first.
• Smith and coworkers (1974) determined how fast people could answer questions about an object’s category.
  • Sentence verification technique: Participants answer “yes” (for true) or “no” (if false) to presented sentences
    • “An apple is a fruit”.
    • “A pomegranate is a fruit”.
  • Results: People were faster to verify sentences that include typical members (an apple is a fruit).
  • Called the typicality effect.
• Eleanor Rosch (1975b) asked whether more prototypical category members are more easily primed than non-prototypical members.
  • Methods:
    • A color name was presented before participants saw two colors.
    • Participants then had to identify whether the two colors were identical or different.
    • Side-by-side colors were paired in three different ways.
  • Results: After priming, participants faster to identify typical colors as being the same compared to atypical colors.
  • Conclusions: Category labels prime typical members more than atypical members.
• Solso and MacCarthy (1981) tested whether we generate prototypes while learning about categories.
  • Methods:
    • Shown a series of faces which were based on a non-presented prototype.
    • Then given a recognition test containing:
      • Old faces.
      • New faces.
      • The prototype.
    • Results: Recognition was highest for the prototypes even though they were never presented during the study phase!
    • Conclusions: We spontaneously create prototypes.
• Posner and Keele (1968) we also generate prototypes while learning about abstract categories.

3. Exemplar Theories

• Categories are not represented by a single prototype; instead, they are represented by individual objects in a category that a person has encountered in the past (called exemplars).
  • Example: sparrows, robins, and bluejays as exemplars for the category “birds”.
• Categorization is based on how similar the object is to all exemplars of a category.
• Benefits: Can explain many of the same phenomenon as prototype theories and are better at dealing with outliers (penguin) and variable categories (games).
• Disadvantages: inefficient due to higher storage requirements.
• We may use both prototype and exemplar approaches.
  • May average exemplars into a prototype early in learning, with some of the exemplar information becoming stronger later in learning!

4. Theories and Rules

• All categorization judgments are not based on similarities
• Gelman and Markman (1986) asked children to rate the similarity between pairs of objects.
  • Children rated the bat and blackbird as being more similar than the flamingo and blackbird.
  • Next, asked “if a flamingo mashes up food to feed its babies and a bat feeds its babies milk, what does a blackbird do?”
  • Children chose “mashes up food”, even though they thought the blackbird was more similar to the bat.
• All categorization judgments are not based on similarities.
  • We also use theories and rules about the natural world.
  • Children in the Gelmen study could use their biological theories to answer questions.
    • For example: Even though blackbirds look more like bats, they work differently.
  • This implies that we can treat some features, or combination of features, as being necessary for group membership.
• Allen and Brooks (1991)
  • Had two groups of participants learn about “builders” and “diggers”.
    • Rule group: builders have at least 2 of
      • Angular bodies.
      • Spots.
      • Long legs.
    • Memory group: memorize the examples.
  • Subjects were given new examples to categorize at test
    • Easy: looked similar to previous examples of the same category.
    • Hard: looked dissimilar to previous examples of the same category, but had the required features.
  • Results
    • Memory condition: wrong answer to hard examples 86% of the time
    • Rule condition: wrong answer hard examples only 45% of the time.
  • Conclusion: Demonstrates we can use different strategies – such as rules – to form categories.

Part 3: Levels of Categories

• Hierarchical organization
  • Larger, more general categories are divided into smaller, more specific categories.
  • Creates a number of levels of categories.
• Is there a level that is more psychologically important than the others?
• Rosch & coworkers (1976a)
  • List all the items you can that would be common to all or most of the objects in the category.
• Rosch & coworkers (1976b)
  • Write down a word that identifies each picture.
    • Guitar [not musical instrument or rock guitar].
    • Fish [not animal or trout].
    • Pants [not clothing or jeans].
• Is there a basic level of that is more psychologically important than the others?
  • Basic level may be psychologically privileged.
• Tanaka & Taylor (1991) study with experts vs. non-experts.
  • Method: Asked bird experts and non-experts to name pictures of objects.
  • Results:
    • Experts responded by providing specific-level category information for the birds (“robin”).
    • Non-experts provided basic-level category information (“bird”).
  • The level that is “special” – that people tend to focus on – is not the same for everyone!