Children's use of mutual exclusivity to constrain the meanings of words

Abstract

For children to acquire vocabulary as rapidly as they do, they must be able to eliminate many potential meanings of words. One way children may do this is to assume category terms are mutually exclusive. Thus, if a child already knows a label for an object, a new label for that object should be rejected. Six studies with 3-year-olds tested this hypothesis. Study 1 demonstrated that children reject a second label for an object, treating it, instead, as a label for a novel object. In the remaining studies, this simple novel label-novel object strategy was precluded. If the only object present is familiar, children cannot map a novel term to a novel object. Instead they must analyze the object for some other attribute to label. In Studies 2–6, children were taught either a new part term, e.g., trachea, or a new substance term, e.g., pewter, by showing them an object and saying, “This is a trachea” or (“It is pewter”). For unfamiliar objects, children tended to interpret the term as a label for the object itself. For familiar objects, they tended instead to interpret it as a part or substance term. Thus, mutual exclusivity motivates children to learn terms for attributes, substances, and parts as well as for objects themselves.