Many common categories can be objectively defined by a set of rules that define membership in the category. For example, a set of criteria must be met to classify an animal as a mammal (e.g., it must be warm-blooded). Given these criteria, an individual should be able to accurately classify new examples. For other categories, the set of rules that define membership are less clear. Although many people may agree that a particular concerto sounds like a piece by Mozart, it is difficult to create a list of necessary and sufficient conditions that define the category of Mozart concertos other than the fact that they were composed by Mozart. In this case, one gets the sense that classification is occurring based on previous experience with different exemplars of the category. An exemplar-based strategy implies that an individual is classifying new items based on their similarity to previously experienced items in that category.
The fact that some categories can be readily defined by rules and others cannot does not necessarily indicate that two different category learning systems exist. Perhaps people are always using rules to classify, although sometimes those rules are difficult to verbalize and may be imperfect. For example, with enough insight, it seems possible to create a fairly valid set of explicit heuristics that would discriminate Mozart concertos from works by other composers. On the other hand, the existence of explicit rules that define category membership does not automatically mean that people are actually using those rules. People may judge that an animal is a mammal because of its similarity to other mammals rather than by checking whether a list of conditions is satisfied. The fact that people classify a cow as a mammal more readily than a whale demonstrates that the satisfaction of membership conditions is not the only factor influencing classification. Both rule-based and exemplar-based strategies are plausible, but it appears that people are biased to use one strategy over the other depending on the category type and the conditions of learning. Early in training, individuals might use rules to classify items if they are available, or they might try to induce rules if none are given. However, later in training, when they have had extensive experience with examples, they may base their category decisions on their previous experience with these examples. Figure 1 shows examples from two categories that can be classified according to a set of rules or according to similarity to other members of the category.
Neuropsychological data suggest that classification according to rules involves prefrontal cortex. One of the classic tests ofprefrontal cortical dysfunction is the Wisconsin Card Sorting Test, in which the subject must induce a sorting rule (categorize by color, shape, etc.). Patients with prefrontal damage are particularly impaired when they must shift from one category to another. Neuroimaging studies have shown dorsolat-eral prefrontal activity in humans categorizing stimuli
Figure 1 Two categories of stick figures. The figures in category A have at least two of the following three attributes: square body, shaded body, and long legs. The figures in category B have at least two of the following attributes: round body, white body, and short legs. Subjects may be inducing these rules during training, or they may be basing categorization judgments on similarity to previously experienced items in the two categories.
Was this article helpful?
Whenever a doctor informs the parents that their child is suffering with Autism, the first & foremost question that is thrown over him is - How did it happen? How did my child get this disease? Well, there is no definite answer to what are the exact causes of Autism.