Individual variation in entrenchment of multi-word units
09 / 2011 - 08 / 2015
The proposed project aims to investigate degrees of entrenchment of linguistic processing units, with particular attention to variation, both between units and across speakers. According to usage-based theories of language acquisition and language processing, when linguistic elements are used frequently, they become entrenched as units in speakers' minds. If usage determines what becomes stored, individual differences in entrenchment are likely, since language users have different linguistic experiences. It is not known, however, how large these differences are. Furthermore, it is still an open question which factors determine the degree of entrenchment. While frequency is of great significance, it is improbable that it is the only influential factor. Salience, in particular, is likely to play an important role as well; if a word string is very salient to a speaker, it need not occur very often in order to become entrenched. The proposed project contributes to a better understanding of these underexposed issues. In order to explore variation in degrees of entrenchment of various multi-word combinations and across speakers, quantitative and qualitative data will be gathered by means of corpus research, offline and online tasks. The combination of complementary research methods serves to achieve a more complete picture of the entrenchment of multi-word units and a better understanding of the relative importance of frequency in individuals' storage of units. Moreover, as the focus of this project is on multi-word units, which are on the interface between lexicon and syntax, the research will contribute to the current debate about the distinction between lexicon and grammar. Additionally, the findings will be of value in the fields of first and foreign language acquisition, and in computational linguistics and speech technology areas. Firstly, multi-word units play an important, facilitative role in acquiring a language. Insights into the factors contributing to the entrenchment of multi-word units can advance the effectiveness with which such units are used in educational settings. Secondly, a better understanding of the ways in which degrees of entrenchment can be determined will be of benefit in speech recognition, machine translation, and in unit-based speech synthesis. By providing psycholinguistic evidence regarding the nature of unit status and the effects of context, the proposed project will pave the way for improvements in identifying units and representing its component parts as one chunk.