Processing dynamics of grammatical and phonological encoding in language production

What is the processing relationship between grammatical and phonological encoding? The Standard Model of language production assumes that grammatical and phonological encoding take place at two separate levels of processing. The primary evidence that supports this assumption comes from the analyses of word and sound exchange errors. This evidence suggests that word exchanges, assumed to arise during word ordering, are grammatically but not phonologically constrained, while sound exchanges, assumed to arise during sound encoding, are phonologically but not grammatically constrained. In Chapter 1, the role of this evidence is called into question by three observations from the literature: (1) Word exchanges are influenced by phonological information; (2) Word ordering is influenced by phonological information; (3) Constraints on the form of the determiner in Romance languages reveal a close relationship between word ordering and phonological processes. These observations motivate the Response Ordering Model in which it is assumed that grammatical and phonological encoding take place within a single processing level.; In Chapter 2, the predictions of the two models are investigated within the double naming paradigm, in which two word utterances are elicited. Two experimental factors are manipulated: (1) The syntactic information relevant for ordering words in the utterance is either present or absent; (2) The phonological onset of the words in the utterance is either related or unrelated. The Standard Model predicts additive effects of these two factors, while the Response Ordering Model predicts interactive effects of these two factors. Contrary to the predictions of the Standard Model, the results reveal phonological inhibition in the absence of syntactic information, and phonological facilitation in the presence of syntactic information. In Chapter 3, it is attempted to provide further evidence for a role of word ordering in determining the outcome of the phonological effect. Finally, in Chapter 4, predictions of a modified version of the Standard Model that can account for the results of Chapter 2 are tested. The results of the experiments in this thesis are discussed in the context of a model of language production that assumes a single level of processing for grammatical and phonological encoding.