Eleanor Rosch, Feels That This Example Illustrates

Eleanor Rosch, feels that this example illustrates her perspectives on human classification. Rosch believes that human classification should be based on "the result of psychological principles of categorization." (189) The five main points that she discusses in this chapter are:

The two general principles that are proposed to underlie categorization systems

The way these principles result in a basic level of categorization, as reported from previous research

Relating the principles of categorization to the formation of prototypes, particularly clarifying operational concepts of prototypicality

The two problematical issues of the relation of context to basic level objects and prototypes, and the assumptions about the nature of attributes of real objects underlying the claim of structure.

Attempts to analyze the attributes, functions, and contexts of objects as props in culturally defined events.

Essentially, this is a discussion of the concept that categorizations created by humans have little or nothing to do with real-world or tactile objects. Categorizations are made through the linguistics and semantics of human language as it functions within the human mind.

The first principle of forming categories is that the function of the system is to provide the most information possible with the least cognitive effort. The second principle relates to the structure of the information being such that the world can be perceived as having structure. Cognitive Economy explains that we wish to get the most information from categories, and that things within the same category will be perceived as similar, and also different from things outside of that category. Perceived World Structure explains that the perceived world is not an unstructured total set of equiprobable co-occurring attributes. This means that our perceptions of attributes do not occur uniformly, certain attributes co-occur more often, such as wings co-occurring with feathers more than with fur, or something being good to sit on is more likely to co-occur with other attributes that chairs have, rather than with attributes that cats have. Though these are not absolutes and there are exceptions.

Category systems can also be conceived as having a vertical dimension, which would identify the level of inclusiveness of the category, for example how broad the category is. The vertical dimension would explain the difference between the classifications of collie, dog, mammal, animal, and living thing. Those things which are in the least inclusive category, in this case the specific breed of dog, are s till included in all of the more inclusive categories above it, but as the classification becomes more broad, more and more other things are included, such as other breeds (ie, dobermans, pitbulls, rottweilers) in the "dog" classification, and other animals in the "mammal" classification (such as cats, horses, cows, pigs, and monkeys.) Along the horizontal dimension would be the segmentation of the classifications that have the same level of inclusiveness, such as dog, cat, car, bus, chair, and so forth. On the same vertical level would be pitbull, siamese, miata, recliner, and so on, as this indicates how inclusive each segment of classification is. Taxonomy is a system by which categories are related to one another by means of class inclusion; every category of taxonomy is included in another category of taxonomy, except the highest level category, and the "level of abstraction" within a taxonomy refers to the level of inclusiveness.

Research was conducted to show the co-occurrence of attributes in the most common taxonomies of biological and man-made objects in our culture. The hypothesis was that basic level objects are the most inclusive level of classification at which objects have numbers of attributes in common. The nine taxonomies tested were tree, bird, fish, fruit, musical instrument, tool, clothing, furniture, and vehicle. These categories were chosen because they contain the most commonly used nouns in English. A set of words were taken from these taxonomies and the subjects of the experiment were asked to list all of the attributes he or she could think of. Subjects could list a great deal of attributes for the broader categories with a higher level of abstraction, but very few for the lower levels of abstraction, because the higher levels of abstraction had more things included in that category. Similarity in shapes, or the appearance of objects, in the class, was tested using outline-drawings of objects, and the similarity between these drawings were found to be highest when objects were from the same basic-level category, such as two cars or two dogs, and lower from the same superordinate category, such as a car and a motorcycle, or a cat and a dog. More specific subordinate categories, such as two sports cars, or two pitbulls. Basic levels, again such as car or dog, are the most inclusive categories at which the objects could be identified by outline shape. It is easy to identify the shape of a dog, but difficult to distinguish whether that basic shape is a beagle or a bloodhound. Likewise, the shape of a car may be a sports car or a sedan. Imagery, perception, development, and language are all elements which are related to the basic-level category of taxonomies.