With about 90% of all characters in a Chinese dictionary belonging to the semantic-phonetic compound category, Chinese orthography is really more phonetic than logographic. Previous studies have shown that regularity and consistency in the phonetic radical facilitate lexical access of phonetic compound characters. These findings are in line with the literature on lexical access of alphabetic languages, suggesting that phonology plays a common role in the process of visual word recognition across orthographies. The contradictory orthographic neighbourhood density effects found in Chinese and English studies, however, challenge the universal applicability of current models of lexical access. This paper reports an empirical study which investigates regularity, consistency and orthographic neighbourhood density effects on the reading and naming of traditional Chinese phonetic compound characters based on Cantonese phonology. Results showed that by manipulating regularity and consistency at the body rime level, a facilitatory orthographic neighbourhood density effect could be found in lexical decision but not naming. The implication is that regularity and consistency at the level of rime (in addition to the syllable level) is functional in Chinese reading. It also suggests that the body rime might have a general role in lexical access across languages. These findings are interpreted within the connectionist and dual-route models of lexical access. / published_or_final_version / Linguistics / Master / Master of Philosophy

No description available.

No description available.

No description available.

No description available.

The Simple View of Reading (SVR; Gough & Tunmer, 1986) provides a conceptual framework for describing the processes involved when readers comprehend text and strong evidence for the SVR comes from factor analytic studies showing dissociation between decoding and comprehension skills. The aim of the present study was to investigate if pre-decoding and comprehension components exist in Canadian English-speaking pre-readers (n = 36) with the use of Principal Components Analysis and to create a parent literacy questionnaire that contains parent literacy practices that may predict children's pre-decoding and comprehension skills. All children were administered a battery of pre-reading measures and parents completed a literacy survey. Principal Components Analysis demonstrated that listening comprehension and pre-decoding measures loaded as distinct components. The findings provide support for the SVR framework. No clear patterns were identified between parent literacy practices and children's pre-reading skills. Further work is needed with a larger and more representative sample.

No description available.

This thesis presents two novel methods for isolated word speech recognition based on sub-word components. A digital neural network is the fundamental processing strategy in both methods. The first design is based on the 'Separate Segmentation & Labelling' (SS&L) approach. The spectral data of the input utterance is first segmented into phoneme-like units which are then time normalised by linear time normalisation. The neural network labels the time-normalised phoneme-like segments 78.36% recognition accuracy is achieved for the phoneme-like unit. In the second design, no time normalisation is required. After segmentation, recognition is performed by classifying the data in a window as it is slid one frame at a time, from the start to the end of of each phoneme-like segment in the utterance. 73.97% recognition accuracy for the phoneme-like unit is achieved in this application. The parameters of the neural net have been optimised for maximum recognition performance. A segmentation strategy using the sum of the difference in filterbank channel energy over successive spectra produced 80.27% correct segmentation of isolated utterances into phoneme-like units. A linguistic processor based on that of Kashyap & Mittal [84] enables 93.11% and 93.49% word recognition accuracy to be achieved for the SS&L and 'Sliding Window' recognisers respectively. The linguistic processor has been redesigned to make it portable so that it can be easily applied to any phoneme based isolated word speech recogniser.

No description available.

Photocopy of typescript. / Thesis (Ph. D.)--University of Hawaii at Manoa, 1978. / Bibliography: leaves [155]-160. / Microfiche. / x, 160 leaves ill