Perception of phrasal prosody in the acquisition of European Portuguese

• Autor(a) principal: Severino, Cátia Sofia
• Data de Publicação: 2016
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10451/26300
• topic: Língua portuguesa - Aquisição; Língua portuguesa - Prosódia (Linguística); Língua portuguesa - Entoação (Linguística); Teses de doutoramento - 2016; Domínio/Área Científica::Humanidades::Línguas e Literaturas

A central issue in language acquisition is the segmentation of speech into linguistic units and structures. This thesis examines the role played by phrasal prosody in speech segmentation in the acquisition of European Portuguese, both in the processing of globally ambiguous sentences by 4 and 5 year old children and in early word segmentation by 12 month-old infants. Past studies have shown that phrasal prosody is used by adults in ambiguity resolution, for example to disambiguate syntactically ambiguous sentences involving a low or high attachment interpretation of a given phrase (e.g, Hide the rabbit with a cloth). In a first exploratory experiment, and given previous unclear findings in the literature on European Portuguese, we investigated whether prosodic phrasing might guide speech chunking and interpretation of these globally ambiguous sentences by adult listeners. In an eye-tracking experiment, which also included a pointing task, we found that EP adult speakers were not able to use phrasal prosody to disambiguate the structures tested. Both the results from eye gaze and the pointing task indicated the presence of a high attachment preference in the language, regardless of phrasal prosody. These findings required a better understanding of adult interpretation of these utterances before a productive study could be conducted with young children. Building on the lessons learned from this exploratory study, we conducted two new experiments examining young children (and adults) abilities to use prosody, in a different sort of globally ambiguous utterances where differences in phrasal prosody were triggered by the syntaxprosody interface and part of the common, default prosody of the sentences (i.e., in compound word versus list reading structures, like ‘guarda-chuva e pato,’ umbrella and duck vs. ‘guarda, chuva e pato’, guard, rain and duck). An eye-tracking paradigm (along the lines of De Carvalho, Dautriche, & Christophe, 2016a) was used to monitor the use of phrasal prosody, namely the contrast between a Prosodic Word boundary (PW) in the compound word interpretation and an Intonational Phrase boundary (IP) in the list interpretation, during auditory sentence processing. An offline pointing task was also included. Results have shown a clear developmental trend in the use of phrasal prosody to guide sentence interpretation, from a general inability at age 4 to a still developing ability at age 5, when local prosodic cues were still not enough and the support of distal cues was necessary to achieve disambiguation, unlike for adults. While the previous experiments investigated the ability to use prosody to constrain lexical and syntactic analysis, thus looking into the combination of lexical, syntactic and prosodic knowledge at a young age, in a final set of experiments, we asked whether phrasal prosody is exploited to chunk the speech signal into words by infants, in the absence of prior lexical knowledge. Using a modified version of the visual habituation paradigm (Altvater-Mackensen & Mani, 2013), we tested 12-month-olds use of phrasal prosody in early word segmentation beyond the utterance edge factor, by examining the effects of two prosodic boundaries in utterance internal position, namely the IP boundary (in the absence of pause) and the PW boundary. Our findings showed that early segmentation abilities are constrained by phrasal prosody, since they crucially depended on the location of the target word in the prosodic structure of the utterance. Implications of the findings in this thesis were discussed in the context of prosodic differences across languages, taking advantage of the atypical combination of prosodic properties that characterizes EP.