Semantic Similarity to Known Second Language Words Impacts Learning of New Meanings

Second language (L2) learners need to continually learn new L2 words as well as additional meanings of previously learned L2 words. The present study investigated the influence of semantic similarity on the growth curve of learning of artificially paired new meanings of previously known L2 words in Chinese–English bilinguals. The results of a translation recognition task showed that related meanings are learned faster and more accurately than unrelated meanings. The advantage of learning related new meaning persisted and increased for a week after learning the new meanings. These results suggest that semantic similarities impact the learning of new meanings for known L2 words, and that the shared features between previously known and new meanings of a word facilitate the process of incorporating the related new meaning into the lexical semantic network. Our results are discussed under the framework of the connectionist model.