Thesis (M.S.)--Rochester Institute of Technology, 2006. / Typescript. Includes bibliographical references (leaves 71-73).

Le présent mémoire vise à décrire et définir l’effet de transfert musique-parole, qui signifie d’une part les bienfaits transférés entre les deux domaines similaires, et d’autre part les difficultés causées à un domaine, dues à un déficit dans un autre domaine, et à discuter de ses origines à différents niveaux : du niveau acoustique au niveau cérébral, en s’intéressant aux résultats comportementaux et neuroscientifiques des études disponibles. L’objectif principal est de récapituler les manifestations de cet effet de transfert et d’en découvrir l’origine. Dans un premier temps, nous présenterons les phénomènes observés résultant de l’interaction musique-parole, et nous nous intéressons particulièrement aux effets de transfert de la musique à la parole, en les catégorisant en trois aspects principaux : l’amorçage musical à court terme, l’entraînement musical à long terme, et les bienfaits apportés par la musique aux personnes neurotypiques et à celles atteintes des troubles langagiers. Nous ferons des hypothèses sur l’origine de cet apparent effet de transfert, compte tenu de la relation complexe entre la musique et la parole. Ensuite, en passant en revue le débat sur la relation entre musique et parole, et en présentant des hypothèses sur l’association et la dissociation des deux domaines, nous tenterons d’en discuter l’origine. Finalement, nous proposerons quelques remarques concernant cet effet de transfert, en discutant des questions et hypothèses posées au début. / The present dissertation aims to describe and define the music-to-speech transfer effect, which means, on the one hand, benefits transferred between the two similar domains, and, on the other hand, difficulties caused to one domain due to a deficit in another domain, and to discuss its origins at various levels: from the acoustic to the cerebral level, reviewing the behavioural and neuroscientific results of available studies. The main objective is to summarize the manifestations of this transfer effect and to discover its origins. First, we will present the observed phenomena resulting from the interaction of music and speech, with a particular focus on the transfer effects of music to speech, categorising them into three main aspects: short-term musical priming, long-term musical training, and the benefits of music for neurotypicals and those with language disorders. We will hypothesize about the origins of this apparent transfer effect by considering the complex relationship between music and speech. Then, by systematically reviewing the debate on the relationship between music and speech, and by presenting the hypotheses of association and dissociation of the two domains, we will attempt to discuss its origins. Finally, we will propose some remarks concerning this effect of the transfer, discussing the questions and hypotheses posed at the beginning.

Children with Autism Spectrum Disorders demonstrate deficits in speech and language, with the most outstanding speech impairments being in comprehension, semantics, prosody, and pragmatics. Perception and production of music and speech in children with ASD appear to follow the same principles of Gestalt pattern perceptual organization. In addition, common neuroanatomical structures and similar patterns of cortical activation mediate the perception and production of speech and music. Therefore, the purpose of this study was to explore how the perception of musical stimuli would impact the perception and production of speech and language in children with ASD. The study examined the effect of developmental speech-language training through music on the speech production of children with ASD. The participants were 50 children with ASD, age range 3 to 5 years, who had previously been evaluated on standard tests of language and level of functioning. The children completed the pre-test, six sessions of training, and the post-test. The pre-and post-tests consisted of the Verbal Production Evaluation Scale (VPES) and measured each participant's verbal production including semantics, phonology, pragmatics, and prosody, of 36 target words. Eighteen participants completed music training, in which they watched a music video containing six songs and pictures of the 36 target words. Another group of eighteen participants completed speech training, in which they watched a speech video containing six stories and pictures of target words. Fourteen participants were randomly assigned to a no-training condition. Results of the study showed that participants in both music and speech training significantly increased their scores on the VPES from the pre-test to the post-test. Both music and speech training were effective for enhancing participants' speech production including semantics, phonology, pragmatics, and prosody. Participants who received music training made greater progress on speech production than participants who received the speech training; however, the difference was not statistically significant. Results of the study also indicated that the level of speech production was influenced by the level of functioning in children with ASD. An interaction between level of functioning and training conditions on speech production approached significance. The results indicate that both high and low functioning participants improved their speech production after receiving either music or speech training; however, low functioning participants showed a greater improvement in speech production after the music training than after the speech training. Collectively, music training was more effective for speech production in low functioning children with ASD than was speech training. The study suggests that the superior performance in speech production in children with ASD who received music training might be generated from music stimuli which were organized by the Gestalt laws of pattern perception. In conclusion, children with ASD appear to perceive important linguistic information (i.e., target words) embedded in music stimuli, and can verbally produce the words as functional speech. These results provide evidence for the use of music as an effective way to enhance speech production in children with ASD.

Dans cette étude, nous analysons dans quelle mesure l’exploitation d’un chant populaire, pourrait aider les élèves francophones dans la lecture de textes en allemand. L’exercice musical peut-il contribuer à faciliter la prononciation des voyelles de la langue allemande et ainsi permettre de remédier aux nombreuses difficultés mentionnées dans la littérature et repérées en classe ? L’expérimentation décrite ici a été menée dans une classe bilingue de Strasbourg. Vingt élèves d’une même classe ont été répartis en deux groupes. Le premier groupe avait pour objectif l’apprentissage de la chanson « La Belle au Bois Dormant » tandis que le deuxième s’est focalisé sur la lecture et la compréhension du même texte, sans usage du chant. Les sujets ont été enregistrés pendant la lecture du texte en amont (test T1) et en aval (test T2) de l’expérimentation. Nous avons également examiné l’impact d’un support multimédia élaboré en 2008. Cet outil est supposé pouvoir constituer une aide à l’apprentissage de la lecture par le biais d’exercices adaptés et d’une version chantée dite « karaoké ». La spécificité des résultats des tests T1 et T2 dans chacun des groupes (éléments musicaux, accents d’intensité, premiers et seconds formants) ainsi que la comparaison de deux séquences incluant /a/ d’une part et /a:/ d’autre part, confortent les hypothèses issues de la psycholinguistique, des neurosciences, de la sociolinguistique et de la pédagogie ainsi que d’expériences antérieures : la musique a effectivement un impact sur la qualité vocalique d’une langue accentuelle comme l’allemand. / In this research, we try to analyse how a popular lullaby could be helpful for 7 years-old French pupils while reading German texts. Could musical entertainment enhance the abilities to pronounce more successfully the vowels of German and also to avoid many of current difficulties described by many authors and observed at school?The experimentation we describe here was achieved in a bilingual class in Strasbourg. Twenty learners were divided into 2 groups. The 1st group had to learn the song about the story of “The Sleeping Beauty”, whereas the other group had to understand and read the same text, but without singing it. The subjects were recorded while reading the text of the song before the experimentation (test T1) and a few days after the end of the experimentation (test T2). We also examined the role of multimedia material developed in 2008 and containing a karaoke version of the song and exercises meant to facilitate learning to read. The specificity of T1 and T2 in each group (musical features, accentuation, formant frequencies) and the comparison of two similar sequences containing /a/ on the one hand and /a:/ on the other hand, lead us to the conclusion that, as predicted from theories in psycholinguistics, neurosciences, sociolinguistics, pedagogy and previous experiences, music definitely has an effect upon the quality of vowel pronunciation of a specific stress-timed language like German (L2).