Abstract
The present study consists of an empirical test of the psychological reality of the formal room theory proposed by Rumelhart (1974) for the characterization of appropriate answers to where‐questions. This theory defines an answer‐generation algorithm which specifies the level (room) on which an answer is given as a function of the category of the target questioned about and its distance from the conversants. The answers generated are vertical in the sense that they consist of the placement of the target in a room which contains it (e.g., “The Empire State Building is in New York City “). Three experiments conducted on normal adults in both Israel and the U.S. showed that the answers people give to where‐questions do, in fact, vary with the category of the target and its distance. However, systematic deviations from the predictions of the room theory were noted. Consequently, an extended room theory is proposed which includes amendments having to do both with conceptual representation and with processing. The extended theory provides for the generation of not only vertical answers but also of horizontal relational (“North of Israel”) and featural ones (“It is hot there”). In addition, the study includes a structural‐linguistic typology of the variety of expressions employed in the answering of where‐questions, a consideration of the interaction between the formal‐topological room theory and the factors of knowledge and intention, and a discussion of the place of the present proposal within a general pragmatic framework of cognition.

This publication has 2 references indexed in Scilit: