PhD project
Branching Onsets instability in phonological acquisition
This project aims to investigate how the CCV (Consonant1+Consonant2+Vowel) branching onsets of Brazilian Portuguese develop in the child phonological system. The main question of this study is: why do children usually modify the CCV syllables in their speech? (as in /'bɾu.ʃa/ > ['bu.ʃɐ] 'witch'; /'bɾa.vo/ > ['bla.vʊ] 'angry'/; /'plã.ta/ > ['pɾã.tɐ] 'plant'). We are investigating factors like i) the articulatory difficulty of CCV; ii) the possible underspecification of the liquid segments in the syllable; and iii) the structural aspects of CCV and its relation to the Maximum Onset Parameter setting. In order to observe the influence of these factors, we gathered experimental data from production, perception and mispronunciation detection tasks. This data is currently been transcribed and tabulated for further analysis.
MA project
Branching Onsets acquisition in Phonology
This research investigates the development of the syllabic forms CCV (Consonant1 + Consonant2 + Vowel) within Brazilian Portuguese child speech (like in 'prato', 'trinta'). The main goal of this work is to observe whether the CCV pattern would be available to the childs Phonology prior to the first correct realizations of CCV syllable. We analyzed longitudinal productions from three children (1;8-5;6 years old) as well as experimental data from 49 children in a repetition task. Three evidence sources were selected: (i) observation of phonotactic patterns and repair strategies imposed to the production of the CCV syllable; (ii) interaction between the branching onset and the palatalization of alveolar implosives; and (iii) comparing the duration of CCV syllables reduced to C1V and target CV syllables. The results in (i) and (ii) point towards the phonological presence of branching onsets even before the first correct realizations of CCV. However, this presence was not observed in part of the younger children, suggesting that onset branching is yet to emerge at those childrens Phonology. This emergency is attributed to the setting of the Maximum Onset Parameter, and the post-parameter setting variability and gradualness observed in the course of acquisition will be investigated in a future project.
BA Projects
The acquisition of phonological rules
Haplology and Coda voicing
The goal of this project is to observe the interaction between the phonological rule of voicing of alveolar fricatives in coda position and the process of syllabic haplology in both adult and child speech (like in 'eu go/s/(to) de bolo' > go[s]_/go[z]_ de bolo). The research aims to determine the order of application of these two phenomena, based on the surface production of the fricative ([z]: haplology>>voicing or [s]: voicing>>haplology). We analyzed 97 naturalistic recordings of four children aged 1;2-4;9 years old, as well as adult speech recordings of 3 female informants with ages between 22 and 33 years. The results indicated a low application of the haplology process in both adult speech and children's productions. The fricative was produced in its voiceless [s] form in the adult form, pointing to an order of application voicing>>haplology; in the children's speech, the fricative codas were omitted when in the context of syllable fall. The results indicate that the production of the fricative coda is affected by the application of syllabic haplology, being avoided by the child.
Palatalization and Coda voicing
The goal of this project is to observe the prosodic course taken by the child when acquiring the rules of fricative coda voicing (casa/s/ brancas > casa[z] brancas) and palatalization of alveolar stops (/t/igre > [tʃ]igre). These rules are prosodically distinct in Brazilian Portuguese: voicing occurs among all the prosodic domains proposed by Nespor & Vogel (1986); but palatalization only occurs within Phonological Words. We questioned whether both rules would be initially applied i) in accordance to the target; ii) in all prosodic domains (like palatalizations beyond the Phonological Word); or iii) in a specific prosodic domain (like voicings only in a specific domain). Longitudinal data from two children aged 2;0-4;09 years old point to an initial compliance to the prosodic hierarchy of the target language: 1) no post-lexical palatalizations were observed; and 2) voicing was correctly applied on all prosodic domains since the beggining of the rule productions.
Phonological awareness and litteracy
In the orthographic system of Portuguese, some graphemes represent more than one phoneme, such as the letter 'c', which stands for the sounds /s/ and /k/; however, some phonemes can also be represented by more than one grapheme, like /z/, which can be written with 'z', 's' or 'x'. This project investigated if any of these multiple representations of the orthographic system would impose greater difficulty to the child in the litteracy process. We analyzed data from notebooks of triplets in their first and second years of Primary School. The results showed that in the 2nd year the triplets presented a greater difficulty in dealing with phonemes with more than one grapheme, whereas in the first year two of the brothers had a greater number of errors involving graphemes with more than one sound value. Since graphemes standing for more than one phoneme have contextual rules guiding their use, but phonemes written with more than one grapheme present a irregular/historically motivated use, these results point that children base the litteracy process into phonologically motivated patterns, seeking for regularities when learning to write their language.