This paper describes a new protocol for assessing the phonological systems of two-year-olds with typical development and older children with delays in vocabulary acquisition. The test (Profiles of Early Expressive Phonological Skills (PEEPS), ) differs from currently available assessments in that age of acquisition, based on lexical norms from the MacArthur-Bates Communicative Developmental Inventories, served as the primary criterion for creating a word list. Phonetic and semantic properties of the words were also considered in selecting items for the test. Productions of words using the PEEPS protocol have been gathered from a group of children with typical development and another group with cleft lip and/or palate. By 24 months of age, the children with typical development produced more than 90% of the target words and the children with atypical development produced 73% of the words. Regarding administration, the time needed for administering the protocol decreased with age.

The issues relative to foreign accent continue to puzzle second language researchers, educators, and learners today. Although once thought to be at the root, maturational constraints have fallen short of definitively accounting for the myriad levels and rates of phonological attainment (Bialystok & Miller, 1999, p. 128). This study, a Posttest-only Control Group Design, examined how the pronunciation accuracy of adult, English language learners, as demonstrated by utterance length, was related to two input stimuli: auditory-only input and auditoryorthographic input. Utterance length and input modality were further examined with the added variables of native language, specifically Arabic and Spanish, and second language proficiency as defined by unofficial TOEFL Listening Comprehension and Reading Comprehension section scores. Results from independent t tests indicated a statistically significant difference in utterance length based on input modality (t(192) = -3.285. p = .001), while with the added variable of native language, factorial ANOVA results indicated no statistically significance difference for the population studied. In addition, multiple linear regression analyses examined input modality and second language proficiency as predictors of utterance length accuracy and revealed a statistically significant relationship (R 2 = .108, adjusted R 2 = .089, F(3, 144) = 5.805, p = .001), with 11% of the utterance length variance accounted for by these two factors predictors. Lastly, hierarchical regressions applied to two blocks of factors revealed statistical significance: (a) input modality/native language (R 2 = .069, adjusted R 2 = .048, F(2, 87) = 3.230, p = .044) and ListenComp (R 2 = .101, adjusted R 2 = .070, F(3, 86) = 3.232, p = .026), with ListenComp iv increasing the predictive power by 3%; (b) input modality/native language (R 2 = .069, adjusted R 2 = .048, F(2, 87) = 3.230, p = .044) and ReadComp (R 2 = .112, adjusted R 2 = .081, F(1, 86) = 3.629, p = .016), with ReadComp increasing the predictive power by 4%; and (c) input modality/native language (R 2 = .069, adjusted R 2 = .048, F(2, 87) = 3.230, p = .044) and ListenComp/ReadComp (R 2 = .114, adjusted R 2 = .072, F(2, 85) = 2.129, p = .035), with ListenComp/ReadComp increasing the predictive power by 4%. The implications of this research are that by considering issues relative to input modality and second language proficiency levels especially when teaching new vocabulary to adult second language learners, the potential for improved pronunciation accuracy is maximized. Furthermore, the heightened attention to the role of input modality as a cognitive factor on phonological output in second language teaching and learning may redirect the manner in which target language phonology is approached.

Esta tese investiga a aquisição fonológica variável da nasal velar em coda silábica final (como em living ‘vivendo’ e singing ‘cantando’, por exemplo) por aprendizes brasileiros de inglês como segunda língua. O estudo é motivado pela diferença de status da nasal velar em português (L1, primeira língua) e inglês (L2, segunda língua): enquanto na L1 o segmento é realização fonética resultante de assimilação de ponto de articulação (como em atum, em que a nasal bilabial assimila o traço [+dorsal] da vogal precedente realizando-se como velar), na L2 a nasal velar é fonema (como em sin ‘pecado’ x sing ‘cantar’, realizados com a nasal alveolar e velar, respectivamente). Em algumas comunidades de falantes nativos de inglês, palavras derivadas pelo sufixo {ing} (como read.ing ‘lendo’ e study.ing ‘estudando’, por exemplo) constituem um fenômeno de variação linguística estável, em que os falantes alternam a produção das nasais velar e alveolar no final das palavras. Essa variação é estrutural e socialmente condicionada (LABOV, 1994; 2001). Os dados empíricos deste estudo provêm da fala coletada através de gravações com aprendizes de inglês, que pertecem a dois níveis de proficiência: básico e pré-intermediário. São realizadas duas análises formais: uma que explica a variação sistemática na interlíngua, observável no desempenho dos aprendizes, através do software Goldvarb (SANKOFF, TAGLIAMONTE e SMITH, 2015), e outra que formaliza a organização interna dessa gramática variável, através de algoritmos de aprendizagem vinculados à Teoria da Otimidade Estocástica, o Algoritmo de Aprendizagem Gradual (GLA) (BOERSMA e HAYES, 2001) e o ORTO Ajuste Paramétrico (DORNELLES FILHO, 2014). A Análise de Regra Variável demonstrou que a interlíngua é um sistema linguístico sujeito à variação ordenada como as demais línguas naturais, condicionada por aspectos linguísticos (classe morfológica) e extralinguísticos (nível de proficiência). A variação é verificada pela produção oral dos aprendizes, examinada de oitiva, que alterna as nasais palatal (63,6%) e velar (36,4%) em coda silábica final. A análise estocástica revelou um sistema de interlíngua dominado por restrições de Marcação, cuja variação é decorrente do aumento dos valores de ponto de seleção de restrições de Fidelidade. As restrições diretamente envolvidas na aquisição fonológica variável da nasal velar são AGREEplaceVN#, que exige que a sequência Vogal+Nasal em coda final partilhem ponto de articulação, e IDENTnasal, que requer identidade de traço nasal entre as formas presentes no input e no output. As análises realizadas, que representam a língua dos aprendizes em seus aspectos interno e externo, ou, conforme a terminologia de Chomsky ([1965]1 1975), a competência e o desempenho linguísticos, comprovaram a natureza variável da interlíngua e proporcionaram uma reflexão teórica acerca da possibilidade de diálogo entre o estrutural e o social para a explicação de fenômenos linguísticos variáveis. / This thesis investigates the variable phonological acquisition of the velar nasal in final syllabic coda (such as in living ‘vivendo’ and singing ‘cantando’, for example) by Brazilian learners of English as a second language. The study is motivated by the different status of the velar nasal in Portuguese (L1, first language) and in English (L2, second language): while the segment is a phonetic realization of assimilation of place of articulation in the L1 (‘atum’, where the bilabial nasal assimilates the feature [+dorsal] of the preceding vowel, being realized as velar), the velar nasal is a phoneme in the L2 (sin ‘pecado’ x sing ‘cantar’, realized as the alveolar and velar nasal, respectively). In some native English-speaking communities, words derived from the suffix {ing} (such as read.ing ‘lendo’, study.ing ‘estudando’, for example) are in linguistic stable variation, in which the speakers alternate the production of the alveolar and velar nasal the velar nasal at the end of words. This variation is both structurally and socially conditioned (LABOV, 1994; 2001). The empirical data of this study were collected through recordings with the informants, who belong to two levels of proficiency: basic and pre-intermediate. Two formal analysis are made: one that explains the observable systematic variation in the Interlanguage, through the software Goldvarb (SANKOFF, TAGLIAMONTE and SMITH, 2015) and another that formalizes the internal organization of this variable grammar, through gradual learning algorithms associated to the Stochastic Optimality Theory, the Gradual Learning Algorithm (GLA) (BOERSMA and HAYES, 2001) and the ORTO Ajuste Paramétrico (DORNELLES FILHO, 2014). The Variable Rule Analysis has demonstrated that interlanguage is a linguistic system subject to ordered variation as in all the other natural languages, conditioned by linguistic aspects (morphological class) and extralinguistic ones (level of proficiency). The variation is verified by the oral production of the learners, examined by hearing analysis, which alternates the palatal nasal (63,6%) The stochastic analysis has revealed an Interlanguage system dominated by Markedness constraints, where variation is due to the increase on the values of selection points for Faithfulness constraints. The constraints directly involved in the variable phonological acquisition of the velar nasal are AGREEplaceVN#, which demands the sequence Vowel+Nasal in final coda to share place of articulation, and IDENTnasal, which demands feature identity [+nasal] between input and output forms. The analyses made, which represent the learners’ language in its internal and external aspects, or, according to Chomsky’s terminology ([1965] 1975), the linguistic competence and performance, have proved the variable nature of the Interlanguage and have enabled a theoretical reflection upon the possibility of dialogue between the structural and the social to the explanation of variable linguistic phenomena.

Ce travail entreprend d'évaluer l'évolution de l'acquisition phonologique par des étudiants français des contrastes anglais /ɪ/-/i:/ et /ʊ/-/u:/. Le corpus étudié provient d'enregistrements de conversations spontanées menées avec des étudiants natifs. 12 étudiants, 9 femmes et 3 hommes,ont été suivis lors de 4 sessions espacées chacune d'un intervalle de six mois. L'approche adoptée est résolument quantitative, et agnostique quant aux théories d'acquisition d'une deuxième langue (par exemple Flege 2005, Best 1995,Kuhl 2008). Afin d'estimer les éventuels changements de prononciation, une procédure automatique d'alignement et d'extraction des données acoustiques a été conçue à partir du logiciel PRAAT (Boersma 2001). Dans un premier temps, deux autres logiciels (SPPAS et P2FA, Bigi 2012 et Yuan &Liberman 2008) avaient aligné les transcriptions des enregistrements au phonème près. Plus de 90 000 voyelles ont ainsi été analysées. Les données extraites sont constituées d'informations telles que le nombre de syllabes du mot, de sa transcription acoustique dans le dictionnaire, de la structure syllabique, des phonèmes suivant et précédant la voyelle, de leur lieu et manière d'articulation, de leur appartenance ou non au même mot, mais surtout des relevés formantiques de F0, F1, F2, F3 et F4. Ces relevés formantiques ont été effectués à chaque pourcentage de la durée de la voyelle afin de pouvoir tenir compte des influences des environnements consonantiques sur ces formants. Par ailleurs, des théories telles que le changement spectral inhérent aux voyelles (Nearey & Assmann(1986), Morrison & Nearey (2006), Hillenbrand (2012),Morrison (2012)), ou des méthodes de modélisation du signal telles que la transformation cosinoïdale discrète(Harrington 2010) requièrent que soient relevées les valeurs formantiques des voyelles tout au long de leur durée. Sont successivement étudiées la fiabilité de l'extraction automatique, les distributions statistiques des valeurs formantiques de chaque voyelle et les méthodes de normalisation appropriées aux conversations spontanées. Les différences entre les locuteurs sont ensuite évaluées en analysant tour à tour et après normalisation les changements spectraux, les valeurs formantiques à la moitié de la durée de la voyelle et les transformations cosinoïdales. Les méthodes déployées sont les k plus proches voisins, les analyses discriminantes quadratiques et linéaires, ainsi que les régressions linéaires à effets mixtes. Une conclusion temporaire de ce travail est que l'acquisition du contraste/ɪ/-/i:/ semble plus robuste que celle de /ʊ/-/u:/. / This study undertakes to assess the evolution of the phonological acquisition of the English /ɪ/-/i:/ and /ʊ/-/u:/ contrasts by French students. The corpus is made up of recordings of spontaneous conversations with native speakers. 12 students, 9 females and 3 males, were recorded over 4 sessions in six-month intervals. The approach adopted here is resolutely quantitative, and agnostic with respect to theories of second language acquisition such as Flege's, Best's or Kuhl's. In order to assess the potential changes in pronunciations, an automatic procedure of alignment and extraction has been devised, based on PRAAT (Boersma 2001). Phonemic and word alignments had been carried out with SPPAS (Bigi 2012) and P2FA (Yuan & Liberman 2008) beforehand. More than 90,000 vowels were thus collected and analysed. The extracted data consist of information such as the number of syllables in the word, the transcription of its dictionary pronunciation, the structure of the syllable the vowel appears in, of the preceding and succeeding phonemes, their places and manners of articulation, whether they belong to the same word or not, but also especially of the F0, F1, F2, F3 and F4 formant values. These values were collected at each centile of the duration of the vowel, in order to be able to take into account of the influences of consonantal environments. Besides, theories such as vowel-inherent spectral changes (Nearey & Assmann (1986), Morrison & Nearey (2006), Hillenbrand (2012), Morrison (2012)), and methods of signal modelling such as discrete cosine transforms (Harrington 2010) need formant values all throughout the duration of the vowel. Then the reliability of the automatic procedure, the per-vowel statistical distributions of the formant values, and the normalization methods appropriate to spontaneous speech are studied in turn. Speaker differences are assessed by analysing spectral changes, mid-temporal formant values and discrete cosine transforms with normalized values. The methods resorted to are the k nearest neighbours, linear and quadratic discriminant analyses and linear mixed effects regressions. A temporary conclusion is that the acquisition of the /ɪ/-/i:/ contrast seems more robust than that of the /ʊ/-/u:/ contrast.

Esta tese investiga a aquisição fonológica variável da nasal velar em coda silábica final (como em living ‘vivendo’ e singing ‘cantando’, por exemplo) por aprendizes brasileiros de inglês como segunda língua. O estudo é motivado pela diferença de status da nasal velar em português (L1, primeira língua) e inglês (L2, segunda língua): enquanto na L1 o segmento é realização fonética resultante de assimilação de ponto de articulação (como em atum, em que a nasal bilabial assimila o traço [+dorsal] da vogal precedente realizando-se como velar), na L2 a nasal velar é fonema (como em sin ‘pecado’ x sing ‘cantar’, realizados com a nasal alveolar e velar, respectivamente). Em algumas comunidades de falantes nativos de inglês, palavras derivadas pelo sufixo {ing} (como read.ing ‘lendo’ e study.ing ‘estudando’, por exemplo) constituem um fenômeno de variação linguística estável, em que os falantes alternam a produção das nasais velar e alveolar no final das palavras. Essa variação é estrutural e socialmente condicionada (LABOV, 1994; 2001). Os dados empíricos deste estudo provêm da fala coletada através de gravações com aprendizes de inglês, que pertecem a dois níveis de proficiência: básico e pré-intermediário. São realizadas duas análises formais: uma que explica a variação sistemática na interlíngua, observável no desempenho dos aprendizes, através do software Goldvarb (SANKOFF, TAGLIAMONTE e SMITH, 2015), e outra que formaliza a organização interna dessa gramática variável, através de algoritmos de aprendizagem vinculados à Teoria da Otimidade Estocástica, o Algoritmo de Aprendizagem Gradual (GLA) (BOERSMA e HAYES, 2001) e o ORTO Ajuste Paramétrico (DORNELLES FILHO, 2014). A Análise de Regra Variável demonstrou que a interlíngua é um sistema linguístico sujeito à variação ordenada como as demais línguas naturais, condicionada por aspectos linguísticos (classe morfológica) e extralinguísticos (nível de proficiência). A variação é verificada pela produção oral dos aprendizes, examinada de oitiva, que alterna as nasais palatal (63,6%) e velar (36,4%) em coda silábica final. A análise estocástica revelou um sistema de interlíngua dominado por restrições de Marcação, cuja variação é decorrente do aumento dos valores de ponto de seleção de restrições de Fidelidade. As restrições diretamente envolvidas na aquisição fonológica variável da nasal velar são AGREEplaceVN#, que exige que a sequência Vogal+Nasal em coda final partilhem ponto de articulação, e IDENTnasal, que requer identidade de traço nasal entre as formas presentes no input e no output. As análises realizadas, que representam a língua dos aprendizes em seus aspectos interno e externo, ou, conforme a terminologia de Chomsky ([1965]1 1975), a competência e o desempenho linguísticos, comprovaram a natureza variável da interlíngua e proporcionaram uma reflexão teórica acerca da possibilidade de diálogo entre o estrutural e o social para a explicação de fenômenos linguísticos variáveis. / This thesis investigates the variable phonological acquisition of the velar nasal in final syllabic coda (such as in living ‘vivendo’ and singing ‘cantando’, for example) by Brazilian learners of English as a second language. The study is motivated by the different status of the velar nasal in Portuguese (L1, first language) and in English (L2, second language): while the segment is a phonetic realization of assimilation of place of articulation in the L1 (‘atum’, where the bilabial nasal assimilates the feature [+dorsal] of the preceding vowel, being realized as velar), the velar nasal is a phoneme in the L2 (sin ‘pecado’ x sing ‘cantar’, realized as the alveolar and velar nasal, respectively). In some native English-speaking communities, words derived from the suffix {ing} (such as read.ing ‘lendo’, study.ing ‘estudando’, for example) are in linguistic stable variation, in which the speakers alternate the production of the alveolar and velar nasal the velar nasal at the end of words. This variation is both structurally and socially conditioned (LABOV, 1994; 2001). The empirical data of this study were collected through recordings with the informants, who belong to two levels of proficiency: basic and pre-intermediate. Two formal analysis are made: one that explains the observable systematic variation in the Interlanguage, through the software Goldvarb (SANKOFF, TAGLIAMONTE and SMITH, 2015) and another that formalizes the internal organization of this variable grammar, through gradual learning algorithms associated to the Stochastic Optimality Theory, the Gradual Learning Algorithm (GLA) (BOERSMA and HAYES, 2001) and the ORTO Ajuste Paramétrico (DORNELLES FILHO, 2014). The Variable Rule Analysis has demonstrated that interlanguage is a linguistic system subject to ordered variation as in all the other natural languages, conditioned by linguistic aspects (morphological class) and extralinguistic ones (level of proficiency). The variation is verified by the oral production of the learners, examined by hearing analysis, which alternates the palatal nasal (63,6%) The stochastic analysis has revealed an Interlanguage system dominated by Markedness constraints, where variation is due to the increase on the values of selection points for Faithfulness constraints. The constraints directly involved in the variable phonological acquisition of the velar nasal are AGREEplaceVN#, which demands the sequence Vowel+Nasal in final coda to share place of articulation, and IDENTnasal, which demands feature identity [+nasal] between input and output forms. The analyses made, which represent the learners’ language in its internal and external aspects, or, according to Chomsky’s terminology ([1965] 1975), the linguistic competence and performance, have proved the variable nature of the Interlanguage and have enabled a theoretical reflection upon the possibility of dialogue between the structural and the social to the explanation of variable linguistic phenomena.

Esta tese investiga a aquisição fonológica variável da nasal velar em coda silábica final (como em living ‘vivendo’ e singing ‘cantando’, por exemplo) por aprendizes brasileiros de inglês como segunda língua. O estudo é motivado pela diferença de status da nasal velar em português (L1, primeira língua) e inglês (L2, segunda língua): enquanto na L1 o segmento é realização fonética resultante de assimilação de ponto de articulação (como em atum, em que a nasal bilabial assimila o traço [+dorsal] da vogal precedente realizando-se como velar), na L2 a nasal velar é fonema (como em sin ‘pecado’ x sing ‘cantar’, realizados com a nasal alveolar e velar, respectivamente). Em algumas comunidades de falantes nativos de inglês, palavras derivadas pelo sufixo {ing} (como read.ing ‘lendo’ e study.ing ‘estudando’, por exemplo) constituem um fenômeno de variação linguística estável, em que os falantes alternam a produção das nasais velar e alveolar no final das palavras. Essa variação é estrutural e socialmente condicionada (LABOV, 1994; 2001). Os dados empíricos deste estudo provêm da fala coletada através de gravações com aprendizes de inglês, que pertecem a dois níveis de proficiência: básico e pré-intermediário. São realizadas duas análises formais: uma que explica a variação sistemática na interlíngua, observável no desempenho dos aprendizes, através do software Goldvarb (SANKOFF, TAGLIAMONTE e SMITH, 2015), e outra que formaliza a organização interna dessa gramática variável, através de algoritmos de aprendizagem vinculados à Teoria da Otimidade Estocástica, o Algoritmo de Aprendizagem Gradual (GLA) (BOERSMA e HAYES, 2001) e o ORTO Ajuste Paramétrico (DORNELLES FILHO, 2014). A Análise de Regra Variável demonstrou que a interlíngua é um sistema linguístico sujeito à variação ordenada como as demais línguas naturais, condicionada por aspectos linguísticos (classe morfológica) e extralinguísticos (nível de proficiência). A variação é verificada pela produção oral dos aprendizes, examinada de oitiva, que alterna as nasais palatal (63,6%) e velar (36,4%) em coda silábica final. A análise estocástica revelou um sistema de interlíngua dominado por restrições de Marcação, cuja variação é decorrente do aumento dos valores de ponto de seleção de restrições de Fidelidade. As restrições diretamente envolvidas na aquisição fonológica variável da nasal velar são AGREEplaceVN#, que exige que a sequência Vogal+Nasal em coda final partilhem ponto de articulação, e IDENTnasal, que requer identidade de traço nasal entre as formas presentes no input e no output. As análises realizadas, que representam a língua dos aprendizes em seus aspectos interno e externo, ou, conforme a terminologia de Chomsky ([1965]1 1975), a competência e o desempenho linguísticos, comprovaram a natureza variável da interlíngua e proporcionaram uma reflexão teórica acerca da possibilidade de diálogo entre o estrutural e o social para a explicação de fenômenos linguísticos variáveis. / This thesis investigates the variable phonological acquisition of the velar nasal in final syllabic coda (such as in living ‘vivendo’ and singing ‘cantando’, for example) by Brazilian learners of English as a second language. The study is motivated by the different status of the velar nasal in Portuguese (L1, first language) and in English (L2, second language): while the segment is a phonetic realization of assimilation of place of articulation in the L1 (‘atum’, where the bilabial nasal assimilates the feature [+dorsal] of the preceding vowel, being realized as velar), the velar nasal is a phoneme in the L2 (sin ‘pecado’ x sing ‘cantar’, realized as the alveolar and velar nasal, respectively). In some native English-speaking communities, words derived from the suffix {ing} (such as read.ing ‘lendo’, study.ing ‘estudando’, for example) are in linguistic stable variation, in which the speakers alternate the production of the alveolar and velar nasal the velar nasal at the end of words. This variation is both structurally and socially conditioned (LABOV, 1994; 2001). The empirical data of this study were collected through recordings with the informants, who belong to two levels of proficiency: basic and pre-intermediate. Two formal analysis are made: one that explains the observable systematic variation in the Interlanguage, through the software Goldvarb (SANKOFF, TAGLIAMONTE and SMITH, 2015) and another that formalizes the internal organization of this variable grammar, through gradual learning algorithms associated to the Stochastic Optimality Theory, the Gradual Learning Algorithm (GLA) (BOERSMA and HAYES, 2001) and the ORTO Ajuste Paramétrico (DORNELLES FILHO, 2014). The Variable Rule Analysis has demonstrated that interlanguage is a linguistic system subject to ordered variation as in all the other natural languages, conditioned by linguistic aspects (morphological class) and extralinguistic ones (level of proficiency). The variation is verified by the oral production of the learners, examined by hearing analysis, which alternates the palatal nasal (63,6%) The stochastic analysis has revealed an Interlanguage system dominated by Markedness constraints, where variation is due to the increase on the values of selection points for Faithfulness constraints. The constraints directly involved in the variable phonological acquisition of the velar nasal are AGREEplaceVN#, which demands the sequence Vowel+Nasal in final coda to share place of articulation, and IDENTnasal, which demands feature identity [+nasal] between input and output forms. The analyses made, which represent the learners’ language in its internal and external aspects, or, according to Chomsky’s terminology ([1965] 1975), the linguistic competence and performance, have proved the variable nature of the Interlanguage and have enabled a theoretical reflection upon the possibility of dialogue between the structural and the social to the explanation of variable linguistic phenomena.

Submitted by Cristiane Chim (cristiane.chim@ucpel.edu.br) on 2018-07-16T12:23:03Z No. of bitstreams: 1 Camila Motta Avila.pdf: 5908666 bytes, checksum: fec1aaf83edeb30b980911d427091061 (MD5) / Made available in DSpace on 2018-07-16T12:23:03Z (GMT). No. of bitstreams: 1 Camila Motta Avila.pdf: 5908666 bytes, checksum: fec1aaf83edeb30b980911d427091061 (MD5) Previous issue date: 2017-12-21 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES# / #2075167498588264571# / #600 / This Master’s Dissertation had as its main goal analyzing and formalizing how Brazilian learners of English language (L2) and Americans identify and discriminate the voicing patterns for plosive consonants under artificial manipulation in English monossilabic words. Americans normally follow the acoustic cue Voice Onset Time (VOT) to differentiate voiced plosives from voiceless ones. The VOT is measured in milisseconds and can be classified into three different categories (LISKER & ABRAMSON, 1964): negative VOT (for /b/, /d/ and /g/ in Portuguese); short-lag VOT (for /p/, /t/, /k/ in Portuguese and /b/, /d/, /g/ in English) and long-lag or positive VOT (for [ph], [th], [kh] in English). Previous studies (ALVES, BARATZ e MOTTA, 2012; SCHWARTZHAUPT, ALVES &FONTES, 2015; ALVES & MOTTA, 2014; MOTTA, 2014; ALVES & ZIMMER, 2015) demonstrated that the manipulation of VOT can result into different types of perceptual behaviour, depending on the group analyzed: Americans tend to follow this cue to attribute voicing patterns to consonants, while Brazilians tend to follow the voicing patters stablished for their L1. In order to analyze the perception in L2, this Dissertation was based on Speech Learning Model (SLM) – (FLEGE, 1995) and Perceptual Assimilation Model – L2 (PAM-L2) – (BEST e TYLER, 2007) studies. To investigate how perception was processed in Americans, explanations in Kent and Read (2015), Gussenhoven and Jacobs (2004), Matzenauer (2015) and Boersma, Escudero and Hayes (2003) were found. The method consisted of two perceptual tests: (a) Identification Test, (b) Discrimination Test. Both tests were designed with artificially manipulated stimuli. The types (pee, pit, tick, tip, kit, kill) were CVC (where V was a high vowel /ɪ/ or /i/). Each one of them was multiply manipulated, generating five tokens from one original type. This manipulation was gradually performed, in 25% layers, aiming to achieve the zero artificial VOT pattern. The Identification Test was designed with 60 target words and 12 distractors, resulting in 72 tokens per participant. The Discrimination Test was designed with 36 target trials and 9 catch-trials, in a total of 45 trials per participant. To summarize, results show that Americans can be considered to be more perceptually sensitive to the multiple VOT manipulation than Brazilians, who normally tend to follow their L1 way of attributing voicing patterns to plosives, regardless their proficiency level in English (basic, intermediate or advanced). In addition, different phonetic tresholds were found in each analyzed group and in each place of articulation observed (bilabial, alveolar, velar). Finally, this study tries to offer a formalization for perceptual grammars in each group, based on Bi-directional Phonology OT Model (BiPhon). It is expected that this study, its detailed data description and theoretical observations can contribute to and makefurther academic studies possible. / Esta Dissertação de Mestrado teve como principal objetivo analisar e formalizar de que forma brasileiros aprendizes de inglês (L2) e americanos identificam e discriminam o vozeamento das consoantes plosivas iniciais da língua inglesa sob múltiplas manipulações em palavras monossilábicas. Na língua inglesa, para a diferenciação de vozeamento, tem-se como principal pista acústica a aspiração, que pode ser medida a partir dos valores de Voice Onset Time (VOT). O VOT é medido em milissegundos e pode ser classificado em 3 diferentes padrões (LISKER & ABRAMSON, 1964): VOT NEGATIVO (que se realiza foneticamente em português em /b/, /d/ e /g/); VOT ZERO, (que se realiza como /p/, /t/ e /k/ no português e /b/, /d/ e /g/ no inglês) e o VOT POSITIVO (encontrado em [ph], [th] e [kh] no inglês). Estudos anteriores (ALVES, BARATZ e MOTTA, 2012; SCHWARTZHAUPT, ALVES &FONTES, 2015; ALVES & MOTTA, 2014; MOTTA, 2014; ALVES & ZIMMER, 2015) demonstraram que a manipulação da pista acústica VOT resulta em comportamentos perceptuais diferenciados de acordo com a natureza do participante: americanos tendem a seguir a pista acústica em questão ao atribuir ou não vozeamento para as consoantes plosivas, ao passo que os brasileiros continuam a seguir os padrões de vozeamento previstos para a sua L1. Para dar conta da percepção em inglês como L2, este trabalho fundamentou-se, basicamente, nos modelos perceptuais Speech Learning Model (SLM) – (FLEGE, 1995) e Perceptual Assimilation Model – L2 (PAM-L2) – (BEST e TYLER, 2007). Para analisar a percepção dos participantes americanos, buscaram-se explicações em Kent e Read (2015), Gussenhoven e Jacobs (2004), Matzenauer (2015) e Boersma, Escudero e Hayes (2003). O método aplicado neste estudo consistiu em dois testes de percepção: (a) Teste de Identificação, (b) Teste de Discriminação. Ambos foram montados com estímulos de palavras monossilábicas da língua inglesa (CVC), sendo V /ɪ/ ou /i/. Cada type (pee, pit, tick, tip, kit, kill) passou por manipulação múltipla da pista acústica VOT, gerando 5 tokens cada um. Tal manipulação se deu de forma gradual em camadas de 25%, objetivando-se alcançar o padrão de VOT zero artificial. O design do Teste de Identificação contou com 60 palavras-alvo e 12 distratoras, somando um total de 72 tokens por participante. O design do Teste de Discriminação, que teve por objetivo contrastar diferentes camadas de VOT entre si, contou com 36 tríades do tipo AxB e 9 catch-trials, somando 45 tríades por participante. Sucintamente, os resultados apresentados demonstraram que os participantes americanos são mais sensíveis ao corte da pista acústica VOT do que os participantes brasileiros, independentemente do nível de proficiência em língua inglesa (básico, intermediário e avançado). Além disso, limiares fonéticos distintos foram encontrados em cada um dos grupos analisados e em cada ponto de articulação observado (bilabial, alveolar, velar). Finalmente, propôs-se uma formalização da percepção de ambos os grupos para cada consoante analisada baseando-se no Modelo de OT Bidirecional (BiPhon), proposto por Boersma (2009) e Boersma e Hamman (2011). Espera-se, com este estudo, que a descrição detalhada dos dados e a reflexão teórica possam contribuir à Academia e possibilitar estudos futuros.

This study investigates the patterns in nonword repetition performance of children with a developmental language disorder and bilingual children. It has been shown by previous research that both children with developmental language disorder and bilingual children tend to perform poorly in nonword repetition tasks. As these tasks are one of the tools often used for diagnosing markers of DLD in young children, diagnosing bilinguals with DLD proves to be difficult, since both of the groups exhibit a poor performance. An analysis of the patterns found in NWR performance of bilingual children and children with DLD might shed more light onto the issue. The study focuses on analysing the performance in a widely used assessment task - The Children's Test of Nonword Repetition. Three samples of data were analysed. The first sample of data consisted of monolingual English-speaking children diagnosed with a developmental language disorder. The second sample of data consisted of Czech-English bilingual children from international schools in Prague who started acquiring English at the time of birth, i.e. simultaneous bilinguals. The third and final sample consisted of Czech- English bilingual children from international schools in Prague who started acquiring English after one year of age, i.e. sequential bilinguals....

Les langues ont de nombreux types de dépendances, certaines concernant des éléments adjacents et d'autres concernant des éléments non adjacents. Au cours des dernières décennies, de nombreuses études ont montré comment les capacités précoces générales des enfants pour traiter le langage se transforment en capacités spécialisées pour la langue qu'ils acquièrent. Ces études ont montré que pendant la deuxième moitié de leur première année de vie, les enfants deviennent sensibles aux propriétés prosodiques, phonétiques et phonotactiques de leur langue maternelle concernant les éléments adjacents. Cependant, aucune étude n'avait mis en évidence la sensibilité des enfants à des dépendances phonologiques non-adjacentes, qui sont un élément clé dans les langues humaines. Par conséquent, la présente thèse a examiné si les enfants sont capables de détecter, d'apprendre et d’utiliser des dépendances phonotactiques non-adjacentes. Le biais Labial-Coronal, correspondant à la prévalence des structures commençant par une consonne labiale suivie d'une consonne coronale (LC, comme bateau), par rapport au pattern inverse Coronal-Labial (CL, comme tabac), a été utilisé pour explorer la sensibilité des nourrissons aux dépendances phonologiques non-adjacentes. Nos résultats établissent qu’à 10 mois les enfants de familles francophones sont sensibles aux dépendances phonologiques non-adjacentes (partie expérimentale 1.1). De plus, nous avons exploré le niveau auquel s’effectuent ces acquisitions. En effet, des analyses de fréquence sur le lexique du français ont montré que le biais LC est clairement présent pour les séquences de plosives et de nasales, mais pas pour les fricatives. Les résultats d'une série d'expériences suggèrent que le pattern de préférences des enfants n’est pas guidé par l'ensemble des fréquences cumulées dans le lexique, ou des fréquences de paires individuelles, mais par des classes de consonnes définies par le mode d'articulation (partie expérimentale 1.2). En outre, nous avons cherché à savoir si l’émergence du biais LC était liés à des contraintes de type maturationnel ou bien par l'exposition à l’input linguistique. Pour cela, nous avons tout d’abord testé l'émergence du biais LC dans une population présentant des différences de maturation, à savoir des enfants nés prématurément (± 3 mois avant terme), puis comparé leurs performances à un groupe d‘enfants nés à terme appariés en âge de maturation, et à un groupe de nourrissons nés à terme appariés en âge chronologique. Nos résultats indiquent qu’à 10 mois les enfants prématurés ont un pattern qui ressemble plus au pattern des enfants nés à terme âgés de 10 mois (même âge d'écoute) qu’à celui des enfants nés à terme âgés de 7 mois (même âge de maturation ; partie expérimentale 1.3). Deuxièmement, nous avons testé une population apprenant une langue où le biais LC n’est pas aussi clairement présent dans le lexique : le japonais. Les résultats de cette série d'expériences n’a montré aucune préférence pour les structures LC ou CL chez les enfants japonais (partie expérimentale 1.4). Pris ensemble, ces résultats suggèrent que le biais LC peut être attribué à l'exposition à l'input linguistique et pas seulement à des contraintes maturationnelles. Enfin, nous avons exploré si, et quand, les acquisitions phonologiques apprises au cours de la première année de la vie influencent le début du développement lexical au niveau de la segmentation et de l’apprentissage des mots. Nos résultats montrent que les mots avec la structure phonotactique LC, plus fréquente, sont segmentés (partie expérimentale 2.1) et appris (partie expérimentale 2.2) à un âge plus précoce que les mots avec la structure phonotactique CL moins fréquente. Ces résultats suggèrent que les connaissances phonotactiques préalablement acquises peuvent influencer l'acquisition lexicale, même quand il s'agit d'une dépendance non-adjacente. / Languages instantiate many different kinds of dependencies, some holding between adjacent elements and others holding between non-adjacent elements. During the past decades, many studies have shown how infant initial language-general abilities change into abilities that are attuned to the language they are acquiring. These studies have shown that during the second half of their first year of life, infants became sensitive to the prosodic, phonetic and phonotactic properties of their mother tongue holding between adjacent elements. However, at the present time, no study has established sensitivity to nonadjacent phonological dependencies, which are a key feature in human languages. Therefore, the present dissertation investigates whether infants are able to detect, learn and use non-adjacent phonotactic dependencies. The Labial-Coronal bias, corresponding to the prevalence of structures starting with a labial consonant followed by a coronal consonant (LC, i.e. bat), over the opposite pattern (CL, i.e. tab) was used to explore infants sensitivity to non-adjacent phonological dependencies. Our results establish that by 10 months of age French-learning infants are sensitive to non-adjacent phonological dependencies (experimental part 1.1). In addition, we explored the level of generalization of these acquisitions. Frequency analyses on the French lexicon showed that the LC bias is clearly present for plosive and nasal sequences but not for fricatives. The results of a series of experiments suggest that infants preference patterns are not guided by overall cumulative frequencies in the lexicon, or frequencies of individual pairs, but by consonant classes defined by manner of articulation (experimental part 1.2). Furthermore, we explored whether the LC bias was trigger by maturational constrains or by the exposure to the input. To do so, we tested the emergence of the LC bias firstly in a population having maturational differences, that is infants born prematurely (± 3 months before term) and compared their performance to a group of full-term infants matched in maturational age, and a group of full-term infants matched in chronological age. Our results indicate that the preterm 10-month-old pattern resembles much more that of the full-term 10-month-olds (same listening age) than that of the full-term 7-month-olds (same maturational age; experimental part 1.3). Secondly we tested a population learning a language with no LC bias in its lexicon, that is Japanese-learning infants. The results of these set of experiments failed to show any preference for either LC or CL structures in Japanese-learning infants (experimental part 1.4). Taken together these results suggest that the LC bias is triggered by the exposure to the linguistic input and not only to maturational constrains. Finally, we explored whether, and if so when, phonological acquisitions during the first year of life constrain early lexical development at the level of word segmentation and word learning. Our results show that words with frequent phonotactic structures are segmented (experimental part 2.1) and learned (experimental part 2.2) at an earlier age than words with a less frequent phonotactic structure. These results suggest that prior phonotactic knowledge can constrain later lexical acquisition even when it involves a non-adjacent dependency.