Similarity and Categorizing are not dependent on each other but they complement each other.
How people categorize(Oakes, L, 2003) events & things depends on theories that they have learned. A fully clothed man jumping into a pool will be categorized as "drunk" because a normal diver will take off his suit in tie.
Some researchers say that there is no relation between theory derived categorization and similarity. In fact, conclusions say that children will closely categorize a man to a monkey than a worm due the closer physical features of the monkey and man. Children, even 3 yrs old, will make inferences(Keohane & King,1994) of animal properties using category labels such as dogs eat and move around and man eats and move around , therefore they are alive.
Children described and categorized a set of drawings based on their environmental background. The conclusions from experiments show that categorizations are highly biased by their environmental and educational backgrounds.
Cognition can not be complete without applying similarity rules and categorizing philosophies in the observation of things. Evidences show that categorization and similarity are not based on the exact similar data.
Categorization relies heavily on previous theories and they are more goal driven. Decisions based on similarity is generally not related to choices made under categorization. It seems one is not dependent on the other. Similarity is too flexible to ground categorization (flexible similarity) and categorization is too flexible to be grounded by mere similarity (flexible categorization).
Evidences show that support flexible categorization is refuted by evidences supporting flexible similarity. While categorization is based on theories, goals and knowledge, similarity is also dependent on theories, goals and knowledge. Goodman claims that similarity needs categorization to be understood. Categorization decisions also is depended on the same knowledge and theories that Similarity uses. Therefore, when flexible similarity debates for fancy (sophisticated notion of similarity), flexible categorization debates for a more sensory based similarity.
Raccoon and snake, based on similarity, can be grouped under the subject PETS.
In fact, Barsalou (1983) stated that objects that are similar in some respects are also not similar in other respects. Some similarity comparison criteria are age, expertise, gender, environment, method of presentation and others.
Similarity decreases between objects when they are mixed with other neighbors. An example is when a chicken and falcon ( clearly different from each other) is mixed with a cow and a horse. This new additions will increase the similarity between the chicken and the falcon. A feature affects similarity based on the category used. When asked which country is most similar to Austria, the choice is Sweden when the category is form of government. In the same situation, if the category used this time is Scandinavian, then Sweden is the most similar.
Similarity can be assessed using identification tasks. It is then true that the more similar the two objects are, the longer it will take to say that they are different.
Another similarity situation is when an upright T is commented as more similar to a tilted T than a capitalized L. Similarity then is flexible because