A prototype is the best or most central member of a category. An object can be described in terms of prototypicality, which refers to the degree to which it is a good example of a category. For example, baseball is a more prototypical sport than is billiards or bullfighting, and an automobile is a more prototypical vehicle than is a sled or skateboard.
The idea that category members differ in how well they fit their category is an important component of what is known as the natural view of categories, which emerged in the 1950s with the publication of Ludwig Wittgenstein’s Philosophical Investigations. Radically transforming how categories were understood, the natural view replaced the classical view, a perspective originating from Aristotle’s thinking about categories that had been the accepted belief for two millennia.
According to the classical view, a category, like a formal set, has specific defining characteristics that make the determination of category membership unambiguously clear. Objects that possess all the defining characteristics are category members and objects that do not are nonmembers. Having an absolute criterion for category membership implies that there is no gradation among category members. All objects that meet the standard for inclusion are equivalently good category members. The classical view also assumes that categories are arbitrary, as expressed in Benjamin Lee Whorf’s writing on language and thought, which portrays categorization as a linguistic community’s agreement about how to organize its otherwise chaotic reality. In this view, a category is merely sociolinguistic convention, without any inherent order or constraint in which attributes cluster together to define it.
The Natural View and Categorization
Despite its longevity, the classical view ultimately gave way because the natural view better describes how people actually categorize objects. The natural view recognizes that most categories are not defined by a set of specific properties that are true of all category members. Instead, category members are linked by family resemblance, a group of related characteristics that category members will likely, but not necessarily, possess. For example, a number of things are typically found on vehicles, such as wheels and a motor. However, none of these typical attributes are found on all vehicles, and there is no essential characteristic that an object must possess to be categorized as a vehicle.
Family resemblance implies that category members may not be equivalent. When individual category members possess some but not all of the category’s common features, an object with more of these common features will be considered a better example of the category than one that has fewer. Natural categories have an internal structure, with the prototype, or best example of the category, at the center and less prototypical objects radiating away from it. Although the category’s center is clear, its boundaries are fuzzy. There is no definite point at which one can say the category ends. People will agree about the status of most objects. Things like cars and bicycles are clearly vehicles, but coffeepots and neckties are obviously not. However, at the margins of a category, there will be objects whose status is unclear. People will disagree about whether things like a wheelbarrow, an elevator, or a pair of skates can be considered a vehicle. According to the natural view, no absolute boundary divides the things that are vehicles from those that are not.
The natural view rejects the idea that categories are arbitrary. It contends that a category’s common attributes are things that naturally belong together. For example, it is not merely chance that attributes like feathers, beaks, laying eggs, and the capacity to fly are characteristic of the category “bird.” They form a meaningful category because these things naturally occur together. Creatures that possess any one of these attributes are also very likely to have the others.
Substantial empirical evidence indicates that the natural view provides a more accurate account of categorization than does the classical view, much of which was obtained by Eleanor Rosch and her collaborators in the 1970s. For example, research participants uniformly find it to be an easy and reasonable task to rate whether an object is a better or worse example of a category. There is remarkable consensus in their ratings, and their level of agreement is usually greatest for the most typical category members. Consistent with the idea of family resemblance, objects that possess more of a category’s common attributes are judged to be more prototypical. When listing members of a category, people generate the highly prototypical examples first, and less prototypical examples come later, if they are produced at all. People recognize category membership more quickly for highly prototypical objects than for less prototypical objects. Similarity ratings between high and low prototypicality objects are asymmetrical, with less prototypical objects being seen as more similar to highly prototypical objects than vice versa; for example, people more strongly endorse the statement “a sled is similar to a car” than “a car is similar to a sled.” Interestingly, prototypicality effects also extend to categories with specific defining characteristics, such as geometric shapes or even numbers. People reliably judge 4 to be a “better” example of an even number than 104, though they equally satisfy the formal definition of an even number.
Categorization involves making generalizations that are essential for people to organize and make sense of the information they encounter. However, that people readily generate highly prototypical examples when thinking about categories can be problematic for social judgments. Many members of a social group will not share all the characteristics of the most prototypical member. Failure to recognize this may contribute to pervasive errors in social judgment, such as overestimating the degree to which group members possess certain characteristics, underestimating the variability among group members, and focusing on category membership at the exclusion of relevant information such as base rates.
- Brown, R. (1976). Reference: In memorial tribute to Eric Lenneberg. Cognition, 4, 125-153.
- Brown, R. (1986). Language and thought. In Social psychology: The second edition (Chap. 13, pp. 467-494). New York: Free Press.
- Khorsroshahi, F. (1989). Penguins don’t care, but women do: A social identity analysis of a Whorfian problem. Language in Society, 18, 505-525.
- Lakoff, G. (1987). Women, fire, and dangerous things: What categories reveal about the mind. Chicago: University of Chicago Press.
- Mervis, C. B., & Rosch, E. (1981). Categorization of natural objects. Annual Review of Psychology, 32, 89-115.
- Whorf, B. L. (1956). Language, thought and reality: Selected writings of Benjamin Lee Whorf (J. B. Carroll, Ed.). Cambridge: MIT Press.