To identify something is to group it with like objects called by the same name; that is, to place it in a category. This process begins in our earliest days of acquiring language: the toddler who learns that his spherical rubber toy is a “ball” will spontaneously use that same word for striped beach balls and fuzzy tennis balls. Categorization allows us to make assumptions about an object at hand based on experience with related ones; without this capacity we would be first-time visitors to our own planet with every new experience, unable to understand our surroundings. Hence categorization is essential to the organization of semantic memory, that is, our long-term representation of meaningful non-episodic information.

How does categorization take place? Semantic content knowledge (e.g. appearance, specific features, and functions) and ability to process that knowledge (e.g. select appropriate features and make comparisons) are both necessary. Consider this scenario: someone describes a 45-year-old female traffic cop who strictly disciplines her grandchild as “not what one usually thinks of as a grandmother.” Presumably the speaker is referring to characteristics commonly associated with grandmothers, e.g. being elderly, sedentary, sweet-natured, and eager to spoil their grandchildren. Of course, the speaker also knows that what qualifies someone as a grandmother is being a mother of a parent, age, appearance, and lifestyle notwithstanding. This scenario illustrates two primary categorization processes: similarity-based and rule-based (Smith et al., 1998).