Academic journal article Memory & Cognition

Ad Hoc Category Restructuring

Academic journal article Memory & Cognition

Ad Hoc Category Restructuring

Article excerpt

Participants learned to classify seemingly arbitrary words into categories that also corresponded to ad hoc categories (see, e.g., Barsalou, 1983). By adapting experimental mechanisms previously used to study knowledge restructuring in perceptual categorization, we provide a novel account of how experimental and preexperimental knowledge interact. Participants were told of the existence of the ad hoc categories either at the beginning or the end of training. When the ad hoc labels were revealed at the end of training, participants switched from categorization based on experimental learning to categorization based on preexperimental knowledge in some, but not all, circumstances. Important mediators of the extent of that switch were the amount of performance error experienced during prior learning and whether or not prior knowledge was in conflict with experimental learning. We present a computational model of the trade-off between preexperimental knowledge and experimental learning that accounts for the main results.

When people learn new categories, they rely not only on empirical observation of category members, but also on prior expectations and beliefs about the categories. The effects of preexperimental knowledge are varied and range from providing an initial concept representation to informing which features or dimensions are relevant for categorization (see, e.g., Heit, 1997; Murphy & Medin, 1985; Wisniewski, 1995). Preexperimental knowledge may have greatly different effects, depending on when during the course of learning it becomes available. When preexperimental knowledge is available from the outsetfor example, in the form of category labels-it guides learning by providing an explanation for the structure and properties of categories (Kaplan & Murphy, 2000). When preexperimental knowledge is not explicitly provided at the outset, people use the category members presented during experimental learning to determine which preexperimental knowledge, if any, is useful (see, e.g., Heit & Bott, 2000; Heit, Briggs, & Bott, 2004).

To date, research has primarily focused on the integrated use of preexperimental and experimental knowledge during learning of new categories (Heit, 1994; Heit & Bott, 2000; Kaplan & Murphy, 2000; Wisniewski, 1995). Although the integrated use of different types of knowledge undoubtedly occurs in many real-world situations, there may be other instances in which one source of knowledge competes with and potentially replaces another source. Consider, for example, learning to play a new game of cards with a friend. Initially, you might concentrate on memorizing your friend's instructions. But over time, you might realize that this new card game is just like some other familiar game, with a few small changes. In this case, you may abandon the memorized instructions and continue playing by relying on your prior knowledge of the familiar game.

We examined the issue of how people might switch between experimental learning and preexperimental knowledge by conducting experiments with ad hoc categories (Barsalou, 1983)-that is, categories such as "things that you take from a house that is on fire" that dynamically link otherwise disparate items, such as pets, passports, and photo albums. Ad hoc categories allow a set of items to be represented as unrelated or related, depending on whether or not the ad hoc label is available. For example, Barsalou showed that when category labels were withheld, free recall of ad hoc category members did not differ from the recall of random word lists (and recall of both was worse than recall of lists drawn from common categories). It follows that the observation of ad hoc category members in the absence of the ad hoc label is unlikely to permit the application of useful preexperimental knowledge. Instead, when the label is withheld, categorization of ad hoc category members should rely largely on experimental knowledge attained during feedback-based training. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.