碩士 / 國立政治大學 / 語言學研究所 / 103 / There are four lexical tones and several tone sandhi in Mandarin Chinese. Researchers have been highly interested in how Mandarin-speaking children acquire tonal system; however, compared to the acquisition of four lexical tones, development of tone sandhi are relatively less studied. The aim of this thesis was to investigate the process of acquisition of three different tone sandhi by children aged from 1;6 to 3;0 in Taiwan with longitudinal observation and experiments. Two children were included in longitudinal observation; they were recorded until 27 months old and 23 months old. For the experiments, twelve children were recruited. The videos and sounds were analyzed by PRAAT. Results of this study showed that “pu51” sandhi was acquired earlier than the other two sandhis, “i55” and third tone sandhi. It was also indicated that children acquired the above tone sandhis after 3;0. Furthermore, regarding substitution patterns, they were generally systematic, yet there showed different tendencies individually: in “pu51” sandhi and “i55” sandhi, tone [55] was often used to substitute for the target tone, and in third tone sandhi, the underlying tone was highly used to replace the target tone, which should have turned into tone [35] in the first syllable. As a result, it was revealed that tone [55] and the original tone in the first syllable of tone sandhi played an important role in substitution patterns.

Thesis (M.A.)--Ohio State University. / Includes bibliographical references (leaves 66-68). Available online via OhioLINK's ETD Center

本文主要從句法(Syntax)與音韻(Phonology)的介面關係來探討東勢客語的陽平、陰平及陰去變調現象。本研究以語料庫為根據，觀察東勢客語變調的主要型態，並從句法與音韻的介面來定義變調範域(Tone Sandhi Domain)，包括音步(Foot)、音韻詞組(Phonological Phrase)、語調詞組(Intonational Phrase)等等。本文也發現變調規則必須以循環模式(Cyclic Mode)運作，而且三個變調規則的屬性皆不相同。陽平變調受音步規範、陰平變調較受音韻詞組規範，而陰去變調則不受音韻結構限制。 / This thesis explores the Yangping, Yinping and Yinqu Tone Sandhi Rules (YATSR, YITSR, YQTSR) in Dongshi Hakka from the perspective of the interface between syntax and phonology. According to a Dongshi Hakka Tone Sandhi (DHTS) corpus, this thesis presents the predominant tonal patterns of these rules, and defines the tone sandhi (TS) domain at the interface between syntax and phonology, including foot, phonological phrase, intonational phrase and so on. This thesis concludes that tone sandhi rules must apply in a cyclic mode, and that the three rules are subject to different prosodic structures. YATSR is conditioned by the foot, YITSR is better defined by the phonological phrase, while YQTSR is not structurally-bound.

碩士 / 國立中興大學 / 資訊網路多媒體研究所 / 98 / The task of Taiwanese tone sandhi processing is usually performed with the rule-based approaches. Tone sandhi rules were sorted out by linguists. The rules for Taiwanese are more complex than that for Mandarin. It is not easy for the rules to solve Taiwanese tone sandhi problems to cover all cases. A contradiction may happen when using the rules. Also the rules must be modified by experts. This will be costly and time-consuming. We want to find the sandhi rules by computer learning methods. We constructed a large-scale corpus in our experiments. And then we predict the result of Taiwanese tone sandhi by using Bayesian Network and expectations. We hope to construct a perfect module for processing Taiwanese tone sandhi finally.

Submitted by Aline Batista (alinehb.ufpel@gmail.com) on 2016-08-08T15:59:41Z No. of bitstreams: 2 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Dissertação Paula Penteado de David.pdf: 1660108 bytes, checksum: 9e31ea77aed66458dddff26f630b597e (MD5) / Approved for entry into archive by Aline Batista (alinehb.ufpel@gmail.com) on 2016-08-08T17:18:54Z (GMT) No. of bitstreams: 2 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Dissertação Paula Penteado de David.pdf: 1660108 bytes, checksum: 9e31ea77aed66458dddff26f630b597e (MD5) / Made available in DSpace on 2016-08-08T17:19:05Z (GMT). No. of bitstreams: 2 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Dissertação Paula Penteado de David.pdf: 1660108 bytes, checksum: 9e31ea77aed66458dddff26f630b597e (MD5) Previous issue date: 2015-04-29 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / Esta dissertação analisa os processos de sândi externo presentes em um corpus de falantes nativos e residentes do Uruguai, em um contexto de entrevistas. O método utilizado para a coleta de dados baseia-se em Labov (1972), que afirma ser válido obter dados de programas de rádio e de televisão, principalmente no caso de entrevistas, palestras ou debates. Este estudo tem por finalidade apresentar os processos de sândi externo ? ditongação, degeminação e elisão ?, bem como sua recorrência nos dados do espanhol uruguaio. Atém-se à descrição e análise dos mencionados processos, a partir do corpus de espanhol (LM), com base em Nespor & Vogel (1986) e estudos relacionados ao sândi externo na língua espanhola (Contreras, 1966; Hualde, 1991). A partir da análise realizada, constatou-se que o processo de sândi externo mais recorrente foi o de degeminação, seguido do processo de elisão e, por último, com um número pouco expressivo de ocorrências, a ditongação. / This essay analyzes the external sandhi processes present in a corpus of reporters natives and residing in Uruguay, in a context of interviews. The method utilized for data collection is based in Labov (1972), which states that is valid to obtain data from radio and television shows, mainly in the case of interviews, lectures or debates. This paper has the objective of presenting the external sandhi processes ?diphthongization, degemination and elision ?, as well as its recurrence in Uruguayan Spanish data. This paper sticks to the description and analisys of the mentioned processes, on the Spanish corpus (LM), based on the prosodic theory of Nespor & Vogel (1986) and studies related to external sandhi in Spanish language (Contreras,1966; Hualde, 1991). From the analysis realized, one has concluded that the external sandhi process with the greatest number of occurrences in the sample was the degemination, followed by the elision process and, at last, with an unexpressive number of occurrences, the diphtongization.

碩士 / 國立政治大學 / 語言學研究所 / 104 / This thesis aims at examining Mandarin tone sandhi in Mandarin-English code-mixing. In Mandarin, when two L tones are adjacent, the first L is changed to LH. Hsiao and Lin (1999) have shown that a Southern Min L can trigger Mandarin tone sandhi in a code-mixed environment. Cheng (1968) also observes that the English weakest stress can trigger a preceding Mandarin L to change. This study builds a database that contains 400 tokens of Mandarin-English code-mixing data by inserting Englsih words in Mandarin sentences and Mandarin L words must occur before English words, whose first syllable must be unstressed. We discussed how tone sandhi domain is related to prosody and syntax in two perspectives: Prosody phonily and OT theory. Three research questions are investigated. 1) What is the tone sandhi domain of Mandarin-English code-mixing? 2) How is tone sandhi domain related to prosody and syntax? 3) What tonal and prosodic constraints are used in Mandarin-English code-mixing? What are their rankings? We concluded that first of all, the tone sandhi domain of Mandarin-English code-mixing is JF, which contains at least one Mandarin syllable and the following English syllables. Secondly, the foot parsing process is related to the directions of syntax. Third, tonal constraints are: Ident-R >> OCP-L >> IO-Ident ; and prosodic constraints are: ParseSyll >> *MonoF, Align(Ft, IC)R >> FtBin.

Introduction: The chronic stress, measured by the hypercortisolism, has been associated with cognitive decline, but cognitive impairment, low education level and the collection time of the salivary cortisol might modify this association. Objectives: To investigate the association between the performance in cognitive tests and salivary cortisol levels among elderly with medium to high education level and no cognitive impairment. Methods: 244 elderly (60+ years-old) from Belo Horizonte, Brazil, were invited, with at least four years of schooling with no cognitive impairment defined by the Mini-Mental cutoff points of 18 for elderly between four and seven years of schooling and 26 points for those with eight or more years. Exclusion criteria were: a) severe sight, hearing or speech dysfunction as it could affect the apprehension and performance of cognitive tests; b) regular use of oral corticosteroids, benzodiazepines, cholinesterase inhibitors or psychotropic drugs. The cognitive tests were: semantic and phonemic verbal fluency, Word List learning, recall and recognition from the Consortium to Establish a Registry for Alzheimers Disease and the Trail Making test B. Participants were trained to collect the salivary cortisol at home, two days before the interview, at 11p.m., using a saliva device (Salivette®). The data collected also included: socio-demographic characteristics, lifestyle habits, medications, comorbidities, mental disorders, insomnia and stressful life events. Results: 196 elderly participated with an average age of 68.9 (SD= 6.4) years, 89.8% women, and 12.2 (SD=4) years of schooling. The number of corrected words in the recall test, in multiple linear regression, increased among women (=1.273, 95%CI=2.121, 4.259) and increasing years of schooling (=0.075, 95%CI=0.151, 3.844) and decreased with increasing cortisol levels (= -0.301, 95%CI= -5.940, -0.008), ageing (=-0.036, 95%CI=-7.644, 0.004), decreasing monthly family income (from =-0.893; 95%CI=-1.529, -0.256 to =-1.272, 95%CI=-2.064, -0.480), use of dual-action antidepressants (=-1.350, 95%CI=-2.491, -0.210) and common mental disorders (=-0,505, 95%CI=-1.036, -0.026). The ability to accomplish the Trail Making test B, evaluated by multivariate logistic regression, was directly associated with higher levels of cortisol (OR= 1.46, 95%CI=1.01, 2.13), increasing years of schooling (OR=1.16, 95%CI=1.07, 1.27) and higher scores of subjective social status in the society (OR= 1.26, 95%CI=1.04, 1.55). It was also inversely associated with ageing (OR=0.94, 95%CI=0.90, 0.99). Discussion: The association found between higher levels of salivary cortisol and the decrease of the memory performance is compatible with the results of the majority prospective elderly studies without cognitive impairment and similar schooling level. In the Trail Making test B, the capacity to accomplish the test was associated with the increase of cortisol, possibly because of the improvement of the elderly attention. Conclusion: Higher cortisol levels increased the capacity to accomplish the TMB test, but decreased the performance on the memory test. / Introdução: O estresse crônico, medido pelo hipercortisolismo, tem sido associado à piora cognitiva. Porém, a baixa escolaridade, a presença de déficit cognitivo e o horário de coleta do cortisol podem modificar esta relação. Objetivos: Avaliar a correlação entre o desempenho em testes cognitivos e os níveis de cortisol salivar, em idosos com nível mediano a alto de escolaridade e sem déficit cognitivo. Metodologia: Em 2012, foram convidados 244 idosos ( 60 anos), em Belo Horizonte, Brasil, com pelo menos quatro anos de escolaridade, sem déficit cognitivo, definido pela pontuação no Minimental, de maior ou igual a 18 para os idosos com escolaridade entre quatro e sete anos, e maior ou igual a 26 para aqueles com escolaridade maior ou igual a oito anos. Os critérios de exclusão foram: a) perda visual, auditiva ou de linguagem grave, que pudessem dificultar a compreensão e o desempenho nos testes cognitivos; b) uso regular de corticoide oral, benzodiazepínicos, inibidores da colinesterase e psicotrópicos. Os testes cognitivos foram: fluência verbal semântica e fonêmica, Lista de Palavras (fixação, evocação e reconhecimento) do Consortium to Establish a Registry for Alzheimers Disease e teste de Trilhas B. Os participantes foram treinados para realizar a coleta salivar do cortisol, no domicílio, dois dias antes da entrevista, às 23h, utilizando o Kit Salivette® (Sarstedt). Outros dados coletados foram: características sociodemográficas, hábitos de vida, medicamentos, comorbidades, transtornos mentais comuns, insônia e eventos de vida estressantes. Resultados: 196 idosos participaram, com média de idade 68,9 (DP=6,4 anos), 89,8% do sexo feminino e 12,2 (DP=4) anos de escolaridade. O número de palavras no teste de evocação, na regressão linear múltipla, aumentou no sexo feminino (=1,273; IC95%=2,121; 4,259) e com os anos de estudo (=0,075; IC95%=0,151; 3,844) e diminuiu com o aumento dos níveis de cortisol (= -0,301; IC95%=-5,940; -0,008), o envelhecimento (=-0,036; IC95%=-7,644; 0,004), a diminuição da renda familiar mensal (de =-0,893; IC95%=-1,529; -0,256 a =-1,272; IC95%=-2,064; -0,480), o uso de antidepressivos de dupla ação (=-1,350; IC95%=-2,491; -0,210) e transtornos mentais comuns (=-0,505; IC95%=-1,036; -0,026). A capacidade de realizar o teste de Trilhas B, na regressão logística múltipla, foi associada diretamente ao aumento do cortisol (OR= 1,46; IC95%=1,01; 2,13), anos de estudo (OR=1,16; IC95%=1,07; 1,27) e à maiores pontuações na posição social subjetiva na sociedade (OR= 1,26; IC95%=1,04; 1,55). Foi, também, associada inversamente ao envelhecimento (OR=0,94; IC95%=0,90; 0,99). Discussão: A associação encontrada entre maior nível de cortisol salivar e a diminuição do desempenho da memória é compatível com os resultados observados na maioria dos estudos longitudinais em idosos sem déficit cognitivo e nível educacional similar. No teste de Trilhas B, a capacidade de completar o teste foi associada ao aumento do cortisol, possivelmente pela melhora da atenção nesses idosos. Conclusão: Níveis maiores de cortisol aumentaram a capacidade de realizar o teste de Trilhas B, mas reduziram o desempenho no teste de memória.

Introduction: Voice disorder affects several occupational groups. Some studies have found higher risk of vocal symptom reports among teachers when compared to other occupational groups as well as to the general population. Studies focusing on issues related to the process of the teaching profession, their effects on teachers health and the possible associations with their working conditions are recent. This research combines a study of the profile of teachers on leave due to voice problems with an inquiry on the referred vocal morbidity. Objectives: To learn about the profile of teachers under definitive functional readaptation due to voice disorders, to determine prevalence and factors associated to voice disorders in day-shift public primary school teachers and the association with factors related to the teaching profession. Materials and Methods: Between January/1996 and May/2004, a descriptive study of secondary data was developed referring to the socio-demographic-occupational profile of teachers in readaptation due to voice disorder registered at the Gerência de Saúde do Servidor e Perícia Médica GSPM ((Public Servant Health Management and Medical Examination), at the Gerência de Planejamento Escolar GPE (School Planning Management) and at Gerência de Pessoal GP (Personnel Management Department) of the Municipal City Hall of Belo Horizonte. At the same time, a transversal study was carried out on a random sample of schools in the citys nine regions between May/2004 and July/2005. The eligible schools had at least 80% of the questionnaires responded. 2,103 day-shift female teachers of 83 primary schools took part in the study. Data collection was performed by means of a self-applied questionnaire containing questions on socio-demographic issues, general and mental health (GHQ-12), working environment and organization and aspects related to voice. The analysis of factors associated to the dependent variable (absent, possible and probable voice disorder) was carried out in three steps using multinomial logistic regression, having as reference category the absence of voice disorder. Results: The number of subjects readapted for voice disorder was 164 teachers and the female sex was 98% of the sample. Nodules in the vocal folds were responsible for 32.5% of the cases. Epidemiological inquiry showed 33% of female teachers with no vocal symptom, 15% with probable voice disorder and 52% with possible voice disorder. Probable voice disorder was shown to be associated to poor working conditions, to aspects related to general health and to jeopardy in the professional performance. Conclusion: The results shown in both articles point out the magnitude of voice disorder among female teachers unable to perform their teaching duties as well as among those who are active in the classroom. The obtained data indicate the complexity of the teachers vocal disease and the need to implement new investigations aiming at a better understanding of the associated factors in order to favor improvements in the environmental and organizational working conditions. / A disfonia é um distúrbio de voz que atinge vários grupos ocupacionais, entre eles os professores. Alguns estudos encontraram um aumento do risco de relato de sintoma vocal entre os professores quando comparado com outros grupos ocupacionais, e também, quando comparado com a população geral. São recentes os estudos que examinam as questões relativas ao processo de trabalho docente e os efeitos sobre a saúde, e às relações inerentes às condições que esse trabalho ocorre. Esta pesquisa combina um estudo do perfil de professores afastados do trabalho por problemas de voz a um inquérito de morbidade vocal. Objetivos: Conhecer o perfil dos professores em readaptação funcional definitiva por disfonia, determinar a prevalência e fatores associados à disfonia em professoras do ensino público fundamental diurno e a associação com fatores relacionados ao trabalho docente. Materiais e Métodos: Na fase exploratória, foi realizado estudo descritivo de dados secundários, entre janeiro/1996 e maio/2004, referentes ao perfil sócio-demográfico-ocupacional de professores em readaptação por disfonia, registrados na Gerência de Saúde do Servidor e Perícia Médica (GSPM), na Gerência de Planejamento Escolar (GPE) e na Gerência de Pessoal (GP) da Prefeitura Municipal de Belo Horizonte. Paralelamente, foi realizado estudo transversal em amostra aleatória das escolas de nove regionais do município entre maio/2004 e julho/2005. As escolas elegíveis tiveram no mínimo 80% dos questionários respondidos. Participaram do estudo 2.103 professoras do ensino fundamental diurno de 83 escolas. A coleta de dados foi feita por meio de questionário auto-aplicado contendo questões sociodemográficas, de saúde geral e mental (GHQ-12), ambiente e organização do trabalho e aspectos relacionados à voz. A análise dos fatores associados à variável dependente (disfonia ausente, possível e provável) foi realizada em três etapas utilizando-se a regressão logística multinomial, tendo como categoria de referência a ausência de disfonia. Resultados: O número de readaptados por disfonia foi de 164 professores e o sexo feminino compõe 98% da amostra. Os nódulos de pregas vocais foram responsáveis por 32,5% dos casos. O inquérito transversal evidenciou 33% das professoras sem sintomas vocais, 15% com disfonia provável e 52% foram consideradas com disfonia possível. A disfonia provável mostrou-se associada à precárias condições de trabalho, aspectos relacionados à saúde geral e comprometimento no desempenho profissional. Conclusão: Os resultados apresentados nos dois artigos colocam em evidência a importância da disfonia tanto entre as professoras incapacitadas de desempenhar a docência, quanto entre àquelas que se encontram atuantes em sala de aula. Os dados obtidos indicam a complexidade do adoecimento vocal em professoras e a necessidade de implementar novas investigações visando compreender melhor os fatores associados para subsidiar melhorias nas condições ambientais e organizacionais de trabalho.

Introduction and objective: The development of automated hematology analyzers made possible the automatic measurement of various parameters of platelet volume, such as mean platelet volume (MPV), platelet distribution width (PDW) and platelet large cell ratio (P-LCR). Since these parameters had become available in the 70s, many studies to define its utility and significance had been published. Over the past two years, there was an increasing number of literatures that correlates the platelet parameters to cardiovascular and cerebrovascular diseases. The standardization of preanalytical and analytical factors and determination of reference intervals is essential for the clinical use of these parameters. Thus, the aim of this study was to determine the reference range of parameters of platelet volume in participants from the Longitudinal Study of Adult Health (ELSA-MG). Methods: 173 subjects were selected among ELSA-MG according to established criteria based on the recommendations of the Clinical and Laboratory Standards Institute (CLSI). All procedures were standardized based on these recommendations: analysis of equivalence or reciprocal of equipments used and hematological interfering preanalytical factors (storage time and used anticoagulant) prior to data collection. The values distribution of all parameters were analyzed and the limits and reference intervals were defined using non parametric test. Results: The analysis showed that the results from different equipments used in the study were equivalent, the use of ethylenediaminetetraacetic acid (EDTA) as anticoagulant showed time-dependent increase in the values of the parameters of platelet volume. The distribution of results was a normal or Gaussian distribution, and reference interval obtained was: MPV 8,9 a 12,2 fL; PDW 9,4 a 16,0; P-LCR 15,8 a 41,6 Conclusions: Analysis of the tests showed that equipments could be used interchangeably, and the time for the analysis of specimens collected in EDTA were standardized. The reference intervals for the parameters of platelet volume were defined, and can be used to follow-up ELSA-MG participants to assess possible correlation with subsequent atherothrombotic events. Key words: Platelets, parameters of platelet volume, reference intervals / Introdução e objetivo: O desenvolvimento de analisadores hematológicos automatizados tornou possível a mensuração de vários parâmetros de volume plaquetario, que determinam, além da contagem plaquetária, parâmetros como volume plaquetário médio (VPM), coeficiente de variação do volume plaquetário(PDW) e porcentagem de macroplaquetas (P-LCR). A presença de trabalhos na literatura que associam o uso de parâmetros de volume plaquetário com estados protrombóticos inicia-se na década de 70, cresce na literatura mundial, especialmente nas publicações nos últimos 2 anos, relacionando os parâmetros plaquetários com doenças cardiovasculares e cerebrovasculares. A padronização dos fatores pré-analíticos, analíticos e a determinação de Intervalos de referência é fundamental para o uso clínico destes parâmetros. Assim, o objetivo do estudo foi determinar o intervalo de referência dos parâmetros de volume plaquetário em participantes do Estudo Longitudinal de Saúde do Adulto (ELSA-MG). Métodos: Os 173 indivíduos referência foram selecionados entre os participantes do ELSA-MG, segundo critérios estabelecidos, baseados nas recomendações da Clinical and Laboratory Standards Institute (CLSI). Toda a padronização dos procedimentos baseou-se nestas mesmas recomendações: Análise de equivalência ou comutatividade dos equipamentos hematológicos utilizados e padronização dos interferentes pré-analíticos (tempo de armazenamento e anticoagulante usado), que precederam a coleta dos dados. A distribuição dos valores encontrados foi analisada, com definição dos limites e intervalos de referência, usando teste não paramétrico. Resultados: A análise da comutatividade mostrou que os diferentes equipamentos utilizados no estudo são equivalentes; o uso do ácido etilenodiaminotetracético (EDTA) como anticoagulante mostrou aumento tempodependente nos valores dos parâmetros de volume plaquetário; todos os parâmetros apresentaram distribuição normal ou gaussiana. Os intervalos de referência, considerando os percentis 2,5 e 97,5 obtidos foram VPM: 8,9 a 12,2 fL; PDW: 9,4 a 16,0; P-LCR: 15,8 a 41,6. Conclusões: Os equipamentos puderam ser usados indistintamente para a realização das medições, sendo necessária a padronização do tempo para análise da amostra colhida em EDTA. Os intervalos de referência para os parâmetros de volume plaquetário determinados poderão ser utilizados no acompanhamento longitudinal dessa população para avaliar possível correlação com eventos aterotrombóticos.

碩士 / 國立政治大學 / 語言學研究所 / 101 / This thesis investigates the tone sandhi in Liujia Raoping Hakka and Ningdu Tiantou Hakka under the framework of Optimality Theory. The internal structure of the tone reveals the motivation and the mechanism of tone sandhi. In terms of the tone sandhi in Liujia Raoping Hakka, the universal tonal markedness tendency could be regarded as the motivation triggering the positional tone sandhi in the left syllable. Moreover, the morpheme-specific phonology is adapted to account for the tone sandhi of historical merged tones, Yangping and Yangqu. Yangping and Yangqu are both high level tones; however, they display different tone sandhi patterns. Yangping displays the contextual tone sandhi whereas Yangqu displays the positional tone sandhi. Local conjunction constraints are posited to restrict markedness constraints to specific contexts in order to account for the mechanism of contextual tone sandhi. On the other hand, in terms of the tone sandhi in Ningdu Tiantou Hakka, the tone sandhi is construction sensitive and takes place in both syllables. The indexed constraint approach is adopted to explain the tone sandhi in different constructions. In addition, the preservation of tone is argued to be the typological difference according to different rankings of the positional faithfulness constraints. Finally, the tonal markedness tendency motivates the positional tone sandhi and the conjoined constraints are posited to govern the contextual tone sandhi in this dialect.