Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.599108
Title: Morphology in the mental lexicon : frequency, productivity and derivation
Author: Ford, M. A.
Awarding Body: University of Cambridge
Current Institution: University of Cambridge
Date of Award: 2004
Availability of Full Text:
Full text unavailable from EThOS.
Please contact the current institution’s library for further details.
Abstract:
This thesis examines a critical issue in psycholinguistics, of whether the fundamental unit of representation in the mental lexicon is the morpheme. Are complex words, such as darkness, processed as whole forms or are they decomposed into their component morphemes dark and -ness? Research on this topic has provided preliminary evidence for decomposition by showing that response times to polymorphemic words such as darkness are influenced by summed frequency of the other words in the language containing the root morpheme (e.g. darken, darkly, darkish, etc). This thesis uses multivariate regression techniques to explore in greater depth than previously, both in terms of numbers of items and numbers of variables, the role of morphology in lexical access. It explores not only whether words are influenced by the summed frequency of the morpheme but also whether whole word frequency is important. It also asks whether for singular nouns the word-form frequency (the frequency of the form itself) or the lemma frequency (the summed frequency of all inflectional variants) is the better predictor of response times. Additionally, the role of meaning relationships between words sharing a morpheme and the role of the productivity of the suffix (the likelihood that a new form will be coined with a suffix) are investigated. Results showed that for singular nouns in standard lexical decision, the frequency of the form itself was a better predictor of response times than the lemma frequency. The semantic relationships between items and the complex forms in which they are root morphemes were also shown to be predictive of response times. However, when sampled at an earlier time point in processing, using lexical decision with a response deadline, lemma frequency was the better predictor and semantic effects were not significant. Response times to derived words were influenced both by the frequency of the root morpheme and the frequency of the form itself. The productivity of the suffix was also important. Productive items showed significantly stronger effects of morphemic frequency and of the semantic relatedness between the derived form and its stem. Distributed connectionist modelling captured important aspects of the behavioural data indicating that some aspects of morphology may be an emergent property of regularities in the mapping between form and meaning in morphologically complex forms.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.599108  DOI: Not available
Share: