Empirical Evidence

• Eleanor Rosch (1975)

  • Participants were shown lists of words from the same category (e.g., fruit)

  • Rated each example on 7-point scale indicating how "good" a member of the category it was

  • Results: apples & oranges were "best" examples of this category

• Rosch & Mervis (1975)

  • Group 1 provided more prototype ratings for words in several categories (vehicles, vegetables, clothing)

  • Group 2 listed attributes of each item

  • Results: significant correlations between prototype ratings and shared attributes

    • Care shares many attributes with other vehicles

    • Raft shares few attributes with other vehicles

Typicality Effect

• The more features a member possess, the more (proto)
  typical it is

• Some things are better examples of a concept than others

• A songbird is a more typical bird than an ostrich

• High prototypicality: category member closely resembles
  category prototype

  • "Typical member"

  • For category "bird' = robin

• Low prototypicality: category member does not closely resemble category prototype

  • "Non-typical" member

  • For category "bird" = penguin

• Prototypical objects are processed preferentially

• Highly prototypical objects judged more rapidly

  • Sentence verification technique

• Rosch (1973)

  • Participants: children, adults

    • Questions such as

      • Is an apple a fruit? (central member of category)

      • Is a fig a fruit? (peripheral member of category)

  • DV: Response Times

    • Results

      • Adults faster than children overall

      • Faster responses to central than peripheral members

• Smith, Shoben & Rips (1974)

  • Typicality effect: Sentences about (proto)typical instances were verified faster than sentences about less (proto)typical instances

• Mervis, Catlin, & Rosch (1976)

  • Instances rated as highly typical are more often named as examples of the category

    • BIRD category

    • ROBIN is rated as more typical than penguin

  • ROBIN is more likely than penguin to be given as an example of bird

Rosch 1975, The Typicality Effect

• Highly typical influences are used as cognitive reference points

• Presented pairs of stimuli

  • One was a prototype of category (e.g., circle)

  • Other was similar but not prototype (e.g., eclipse)

• People were more likely to say

  • Eclipse is essentially a circle than a circle is essentially an eclipse

Prototype Approach - Limitations

• When the task of categorising things is set by how well it meets a goal - it is no longer about typicality being defined by family resemblance

• Prototypes do not always represent the features of most members of the category

  • For example, Lynch et al. (2000) had participants provide typicality ratings for different trees - instead of trees representing the most common height being identified, the tallest trees were!

    • Maybe representing an 'ideal' tree

• Familiarity may overrule prototype matching

• Abstract concepts

  • Justice, peace, etc

• Ad hoc categories

   

Exemplar Approach

• Concept is represented by multiple examples (rather than a single prototype)

• Examples are actual category members (not abstract averages)

• To categorise, compare the new item to stored examples

• Similar to prototype view

  • Representing a category is not defining it

• Different: representation is not abstract

  • Descriptions of specific examples

• The more similar a specific exemplar is to a known category member, the faster it was be categorised (family resemblance effect)

Benefits of the Exemplar Approach

• Explains peoples' inability to state necessary & sufficient features. How?

  • There are none

• Explains difficulty categorising atypical instances. How?

  • They are too dissimilar to previously stored instances

• Explains Typicality effects. How?

  • Numerous similar stored instances make typical instances easier (faster) to classify

Problems with the Exemplar Approach

• Does not specify…

  • Which instances will / will not be stored as exemplars

  • How different exemplars are called to mind at time of categorisation

  • How are concepts organised?

• Is truth somewhere in the middle?

  • Exemplars may work best for small categories

  • Prototypes may work best for larger categories

• These two approaches may not be mutually exclusive, rather as two ends of an abstraction continuum