Wikipedia

User:DMR24/sandbox

Eleanor Rosch
Born
Eleanor Heider

1938
Alma materHarvard University
Known forPrototype Theory
Family resemblance
Scientific career
FieldsPsychology
InstitutionsUniversity of California, Berkeley

Eleanor Rosch (née Heider; born 1938[1]) is an American cognitive psychologist. She is known for her innovate work in both the fields of psychology and cognitive science, with a focus on categorization, concepts and linguistics. Rosch is perhaps best known for her Prototype Theory, which states that items in a category are assessed in relation to a category prototype that is found to be more typical or central of that category[2].

Rosch holds a Ph.D. from Harvard University and is currently a professor of psychology at the University of California, Berkeley[1].

Biography

edit

Eleanor Rosch was born in 1938 as Eleanor Heider. She received her Ph.D. in 1969 from Harvard University [1]. Her more recent research interests have grown to include cross-cultural, religious and Eastern psychology. Rosch currently resides in the United States where she is a professor of psychology at the University of California, Berkeley. She is also UC Berkeley's Professor of the Graduate School, where she teaches a course on Buddhist Psychology [3].

Research and Theory

edit

Concepts and Categories

edit

A great deal of Rosch's research is dedicated towards the study of concepts and how they are categorized in people's minds. Rosch believes that categorization is one of the basic functions of all living creatures[4]. The world is a highly categorized place, and being able to categorize novel concepts allows one to learn from experience[5]. Concepts are mental representations of real world ideas, experiences, events etc. They are images in one's mind that represent these ideas and serve as a link between one's internal mind and the external world. Being able to categorize novel concepts allows one to compare present situations to similar ones from the past[6]. In doing so, one can recognize novel stimuli and deduce how to act towards it.

Graded Structure

edit

The classical view of categorization states that categories are exact, meaning that a concept is either a member of a category, or is not[5]. In order to be a member of a category, a concept must possess specific features that are deemed necessary for membership, and sufficient for acceptance into the category [6].

In 1973, Rosch found that real world categories were not organized as classical logic had led on. Using colours in her experiments, she showed that people judge certain shades of colours (e.g., fire engine red versus red hair) to be better examples of that category (i.e. “red”). If classical logic were true, it would be impossible to have “better” examples of a colour category because colours have no necessary attributes or definite boundaries between them. Rosch’s research gave light to a new view of categories. Members of categories with graded structure have varying degrees of membership in the category [6]. This accounts for Rosch’s finding that certain concepts are better examples of a category.

Principles of Organization

edit

According to Rosch, categories are formed around 2 principles: cognitive economy and perceived world structure. These principles have implications for the level of abstraction of categories and for their internal structure once formed [7].

Cognitive Economy
edit

Cognitive Economy (CE) assumes that categories will form to provide maximum information while using the least cognitive effort. The investment of cognitive effort in the formation of categories will be proportionate to the advantage one will gain from that category [8]. According to Rosch, when one makes categories in their minds they will attempt to make them as informative as possible with as little expenditure of mental energy as possible[7].

Perceived World Structure
edit

Perceived World Structure (PWS) views the world as having a correlational structure. World objects contain a highly correlated structure, and individuals will use categories to deduce similarities and draw correlations between concepts and pieces of information [7]. When PWS is combined with cognitive economy, one is left with a system that not only organizes information logically, but also does so in its simplest form.

Levels of Organization

edit
 
Figure 1: Rosch's 2 Category-Specific Levels of Organization

Rosch's work focuses on concepts that are organized according to two levels of organization within categorization: vertical and horizontal.

Vertical Level
edit
 
Figure 2: Example of a 3 Level Hierarchy

Categories are arranged along a vertical level of inclusiveness. This divides categories into hierarchies where each subsequent layer provides increasing specificity about the concept. General hierarchies are organized into 3 layers: superordinate, basic level, and subordinate (see figure 2).

Superordinate categories are those which are placed at the top of the hierarchy and are the most general of the levels. At the bottom of the hierarchy is the subordinate level. Categories at the subordinate level are highly specific and membership is exclusive. Between these two levels is the basic level. It is at this level where most research, including Rosch's, has been focused.

The basic level of categories minimizes between category similarity and maximizes within category similarity[4]. From the basic level, concepts are discriminated further at the subordinate level, and are abstracted at the superordinate level[6]. Research has shown that the basic level contains many properties unique from the other levels. It is the most general level at which:

  • A person interacts with category members using similar motor actions
  • Similar overall shapes are used for category members
  • The entire category can be reflected by a mental image

Objects are also recognized at the basic level more rapidly than members of categories at other levels (i.e. faster to recognize a Granny Smith as an apple, than as a fruit). Basic level categories are also acquired first during language acquisition, as children learn terms such as "apple" before learning concepts like "Granny Smith". Research by Rosch in 1976 showed that children at this young age were faster at solving simple sorting problems at the basic level, before solving the problems at the subordinate level[6].

Horizontal Level
edit

The horizontal level of organization consists of categorical segmentation at the same vertical level. This level separates concepts such as "Apple" from terms like "Pear" and other fruits. The mind organizes categories according to prototypes, which maximizes their distinctiveness [7]. Prototypes contain attributes that are most representative of items within their own category and least representative of items outside of their category [8].

Prototype Theory

edit
 
The robin is judged to be a more prototypical bird than a penguin

Prototype Theory was developed by Rosch in the 1970s [6]. It was formed from her notion that natural categories have a graded structure, whereby concepts can have varying levels and degrees of membership in their category. As a result, some members are judged to be more central or prototypical of their category [5]. Rosch first used the term "prototype" in her 1973 study entitled "Natural Categories", where she defined a prototype as a perceptually salient stimuli [9]. In the present day, it is generally agreed upon that prototypes are defined as the central tendency of a category [2].

Due to the graded structure of categories, not all examples are seen as being equally good representations. For example, research has shown that in the category "bird", robins are judged to be better examples of the category than penguins [10]. Therefore, robins serve as a better prototype for "birds" than would penguins. This finding has been replicated in numerous other studies that used varying instances of categories. Chairs and tables are viewed as more prototypical of the category "furniture" than concepts such as "lamp" [2]. Again, chair and table would be viewed as better prototypes of that category.

Rosch theorizes that many of the world's natural categories inherently develop to contain an internal structure revolving around a prototypical example. Prototypes are seen as the best examples of that category and serve as a reference point to which non-prototypical examples are judged [11]. Furthermore, novel stimuli are compared to the category's prototype during categorization. If that novel stimuli is similar enough to the prototype, it is then deemed to be part of that category and is granted membership. Prototypes are also learned first before other concepts in the category. In addition, categories containing a central prototype are easier to learn and remember [9].

Rosch's development of this theory marked a drastic departure away from the classical view that had dominated cognitive science. Categories were no longer seen as having definite boundaries or equivalent members. Now membership could vary due to a concept's differing similarity with a prototype, allowing for more diversity among concepts in categories.

Family Resemblance

edit

One major factor in prototype formation is the concept of family resemblance [12]. The idea of family resemblance was first made popular by philosopher Ludwig Wittgenstein. In his book "Philosophical Investigations" he explains how even though there is no one feature that all games have in common, as all games are vastly different, humans have no trouble deciding whether something is a game or is not. Taking this notion further, he related this idea to humans. Looking at a family, its members may have some similar features (i.e., height, body structure, colouring etc) but there is no one single feature that every family member has in common. Furthermore, there is no defining feature that each person must have to be considered a member of that family. It is this observation that developed into the idea of family resemblance[13].

Rosch explains family resemblance in terms of categories, as she describes how there is no one distinct feature that all concepts in a category have. Instead, each concept has elements in common with other members, but there is no single element that must be common to all members [14]. The more elements a concept has in common with other category members, the more it is viewed as a representative member of that category and thus a good example[12]. Consequently, exemplars with the greatest family resemblance will share little (if any) elements or attributes with members of differing categories.

This dissimilarity from other categories, and high resemblance to a concept's own family, has been found by Rosch to be highly correlated to representativeness ratings for superordinate and basic level concepts. No effect was shown for subordinate concepts as they do not differ much from each other compared to concepts at higher levels (i.e. granny smith apple vs. red delicious apple). In respect to categorization, membership is therefore most explicit for examples that have family resemblance and are highly representative of their category. Furthermore, categorization of novel stimuli will be most accurate when they are compared to these examples [4].

Linguistics

edit
Dani Tribe of New Guinea
edit
 
Member of the Dani tribe of New Guinea

Rosch conducted one of the most well-known studies of linguistics in the 1970s. Rosch (at the time known as Heider) argued against the perspective of linguist Benjamin Lee Whorf. According to Whorf, the speakers of a language perceive reality differently as a result of their languages coding for domains in differing ways. This later became known as the Sapir-Whorf Hypothesis or linguistic relativity, and is simplified by the notion that language determines thought [15]. Rosch, however, insisted that there was one domain that was universal to all languages: the division of colour space [16].

To further inquire her intuition, Rosch conducted an investigation of colour categories between American English speakers and a stone-age population in New Guinea. Simply referred to as the Dani, this population in the Dugum Dani only has two basic colour terms:

  • Mili = dark
  • Mola = light

These two terms divided the Dani's colour space on the basis of brightness. This contrasts with the English language where colours are divided on the basis of hue [17]. In her experiment, Rosch presented both sets of speakers with coloured chips according to the Munsell Colour System. Participants were then asked which chip was the best example of "X" where "X" was a basic colour term of the participants language [15]. If Whorf was correct, then Dani speakers would confuse the coloured chips, as their language only enables them to see a limited number of colours due to their two colour terms. The results of the study showed that the Dani were able to differentiate among the colours, and that the way in which they structure the colour space in their memory is little influenced by language [17]. Furthermore, both the English and Dani speakers remembered the colour terms in similar ways, even though they have differing colour vocabularies. In light of only having two colour terms, Rosch found that it was easier for the Dani to not only learn, but also recognize, colours that were the best examples of their category.

Rosch's findings in this study proved influential beyond issues of colour representation. Not only did they give evidence that opposed Whorf's concept of linguistic relativity, but they also supported Rosch's Prototype Theory. Rosch's field work with the Dani people of New Guinea extends to all cultures, showing that there are commonalities to be drawn in people around the world [15].

Criticisms

edit

Although Rosch's work is highly praised for its insight into the fields of categorization, concepts, and linguistics, there are still some who believe her work is of concern.

Researchers Daniel N. Osherson and Edward E. Smith believe that an acceptable theory of concepts should be able to account for compound concepts. Furthermore, it should be able to answer how concepts are combined and give insight into the mechanisms of combination. They criticize Rosch's prototype theory as they believe it fails to contain information about how simple concepts are combined to form larger more complex ones. Instead, they suggest that Zadeh's (1965) fuzzy-set logic is the way in which prototype theory can account for the combination of concepts [18].

Rosch has also been faulted by psychologist Ulric Neisser, as he states her theories only deal with a specific type of concept. This type of concept is one which categorizes everyday events and objects from the environment. Consequently, Neisser states that one is left with a specific set of characteristics of "Roschian" categories:

  • No decisive or defining features
  • Presence of "prototypical" or "best" examples
  • A hierarchical organization of categories
  • A basic level with special characteristics and qualities

As a result, basic level concepts are determined largely based on appearance. Neisser, however, argues that appearance is not everything. He states how even though young children may first classify new concepts based on similarity to a prototype's appearance, they soon learn that even two objects that look alike may behave differently. Furthermore, there are some concepts that should not be compared to a prototype- mainly the concept of "intelligence" which is based on fact, not appearance. Thus, according to Neisser, Prototype Theory is superficial in its emphasis on physicality, and is thus flawed [19]

References

edit
  1. ^ a b c Eleanor Rosch. (1995). Retrieved March 8, 2013, from http://www.sis.pitt.edu/~mbsclass/hall_of_fame/rosch.html
  2. ^ a b c Edwards, D. (1991). Categories are for talking. Theory Psychology, 1(4), 515-542.
  3. ^ Eleanor H. Rosch. (2013). Retrieved March 8, 2013, from http://psychology.berkeley.edu/people/eleanor-h-rosch
  4. ^ a b c Mervis, C.B., & Rosch, E. (1981). Categorization of natural objects. Annual Review of Psychology, 32, 89-115.
  5. ^ a b c Rosch, E. (1999). Reclaiming concepts. Journal of Conscious Studies, 6(11-12), 61-77.
  6. ^ a b c d e f Gabora, L., Rosch, E., & Aerts, D. (2008). Toward an ecological theory of concepts. Ecological Psychology, 20(1), 84-116.
  7. ^ a b c d Rosch, E. (2002). Principles of categorization. Foundations of Cognitive Psychology, 251-270.
  8. ^ a b Garbarini, F., & Adenzato, M. (2004) At the root of embodied cognition: Cognitive science meets neurophysiology. Brain and Cognition, 56, 100-106
  9. ^ a b Rosch, E. (1973). Natural categories. Cognitive Psychology, 4(3), 328-350.
  10. ^ Kamp, H., & Partee, B. (1995). Prototype Theory and compositionality. Cognition, 57(2), 129-191.
  11. ^ Rosch, E. (1975). Cognitive reference points. Cognitive Psychology, 7(4), 532-547.
  12. ^ a b Rosch, E., & Mervis, C.B. (1975). Family resemblances: Studies in the internal structure of categories. Cognitive Psychology, 7(4), 573-605.
  13. ^ Wittgenstein, L. Philosophical Investigations. Oxford: Basil Blackwell, 1953.
  14. ^ Rosch, E., Simpson, C., & Miller, R.S. (1976). Structural bases of typicality effects. Journal of Experimental Psychology: Human Perception and Performance, 2(4), 491-502.
  15. ^ a b c Heider, E. R. (1972). Universals in color naming and memory. Journal of Experimental Psychology, 93, 10-20.
  16. ^ Roberson, D., Davidoff, J., & Davis, I. (2000). Color categories are not universal: Replications and new evidence from a stone-age culture. Journal of Experimental Psychology, 129 (3), 369-398.
  17. ^ a b Heider, E.R., & Oliver, D.C. (1972). The structure of the color space in naming and memory for two languages. Cognitive Psychology, 3, 337-354.
  18. ^ Osherson, D.N. & Smith, E.E. (1981). On the adequacy of prototype theory as a theory of concepts. Cognition, 11, 237-262.
  19. ^ Neisser, U. (1979). The concept of intelligence. Intelligence, 3(3), 217-227.
Retrieved from "https://en.wikipedia.org/w/index.php?title=User:DMR24/sandbox&oldid=548757673"