In, on, under: An essay on the modularity of infant spatial competence

According to current standard theory, children acquire a preverbal conceptual organization of spatial relations. In learning spatial re lations words, children are assumed to employ their spatial relational concepts in order to fill in the meaning of the words. The present paper offers an alternative, modular conception. The content, form and manifestation of spatial knowledge are different for the three main cognitive modules, namely perception, action and language. An over view of experimental research on early space perception and action is presented, showing that knowledge in one module cannot account for knowledge in the other. Finally, a longitudinal study of in, on and under use in one child between the ages of 1;9ms and 2;9ms is pre sented. It is argued that the data support a modular view according to which the child learns language specific rules for the lexical distribution of the prepositions. These rules are not based on simple preverbal spatial relations concepts