Accessing lexical ambiguity in different types of sentential contexts

Abstract

Access of ambiguous words was investigated in three cross-modal experiments using the same ambiguous items and the same task (lexical decision), but different types of sentential contexts and semantic relations between an ambiguity and its paired target word. The results demonstrate that in a context biasing its dominant meaning, an ambiguity can be accessed selectively or exhaustively depending upon whether the context places sufficient constraints on the semantic features of its contextually congruent meaning. Different semantic relations between the ambiguity and the subsequent target word, namely, association and feature-denoting, did not alter this pattern of results. It is argued that the dominant meaning of an ambiguous word can be selectively activated—and the cross-modal technique reflects this activation—provided that the ambiguity occurs in a sufficiently constraining sentential context.