Objects and places: Geometric and syntactic representations in early lexical learning

Three experiments investigated differences in the geometric properties of objects and places underlying children's knowledge of count nouns and prepositions. Three-year-olds, 5-year-olds and adults were shown a novel object being placed in a single location on a box, and were told either "This is a corp" (noun condition) or "This is a corp my box" (preposition condition). They then were asked what new shapes and what new positions were acceptable instances of the novel term. Both children and adults could use the syntactic contexts to make appropriate semantic inferences, assuming that the noun referred to the object itself and the preposition to the object's location. Furthermore, there were striking differences in the geometric properties considered relevant to the two new words. In the noun condition, children and adults attended to the particular shape of the object, extending the new word to objects of the same shape as the standard, regardless of its position. In the preposition condition, they either completely disregarded object shape and accepted a wide variety of positions, or preserved just one component of object shape-the located object's principal linear axis-and accepted positions preserving the object's orientation relative to the box. In both conditions, there was sharpening of attention to shape over age. The results are discussed in terms of distinct geometric representations underlying the two domains of objects and places, and the role these representations play in guiding early word learning.