Accessing interlexical homographs: Some limitations of a language-selective access

Abstract

Two experiments are described on the visual processing of interlexical homographs in bilinguals using a priming paradigm. Interlexical homographs are words that have two distinct meanings for one graphemic form common to two languages (e.g., COIN means “corner” in French). English-French bilinguals were biased to read such interlexical homographs in one specific language mode reading. The results of the first experiment suggest that language mode context does not initially block access to the alternative reading of these homographs. The second experiment confirms that the meaning accessed is not affected by language mode context but depends more on the frequency of occurrence of the reading in each language. The results indicate that it is frequency rather than language that determines the order in which lexical entries are accessed. These results suggest that lexical access in bilinguals is not initially language-selective.