The Japanese mental lexicon : The lexical retrieval and representation of two-kanji compound words from a morphological perspective

要旨

iii at the word formation processes that underlie two-kanji compound words, which are primarily morphological. Compounding is the combining of abstract linguistic elements—morphemes—to form larger word units, with the representation of these larger word units being a product of combining the constituent representations. The discussion of two-kanji compound word morphology also provides background for the experiments presented in this thesis. Part 2 of the thesis is psycholinguistic in nature, focusing primarily on the internal representation. After arguing in Part 1 that kanji function as a morphographic writing system, Part 2 investigates the implications of this for the organization of the Japanese mental lexicon. Specifically, a lemma unit model version (Joyce, 1999, in press) of the multilevel interactive-activation framework is proposed as a model for the Japanese mental lexicon. The major feature of the lemma unit model is the incorporation of lemma units that mediate the links between access representations and semantic units. This model is examined through a series of constituent-morpheme priming experiments, which investigated the lexical retrieval and representation of two-kanji compound words within the Japanese mental lexicon from a morphological perspective by controlling for the word-formation principle of compound word stimuli as an experimental condition. As background to the conducted experiments, important tasks and effects are discussed in a brief review of visual word recognition research, which focuses on the two central issues for the mental lexicon—lexical retrieval and lexical representation. A short review of models of lexical representation, which all allow for some form of morphological involvement in the lexical retrieval of

主題