Representation and Planning of Prosody – RAPP
Representation And Planning of Prosody
Prosody refers to the global timing, rhythm and pitch of speech. It has generally a linguistic function (e.g., marking the contrast between questions and statements) and a paralinguistic one (e.g., conveying attitude and emotion). The RAPP research project aims at investigating prosodic planning in French, by determining how far ahead speakers plan the prosodic structure. It will be the first study addressing such an issue, by determining whether (1) the size of the planning unit is flexibly adapted by the speakers depending on the pragmatic situation; (2) prosodic information can be used to predict the pragmatic value of an utterance as soon as it becomes available; and (3) differences in speaker-specific behavior can be explained by variations in cognitive abilities. These questions will be examined from an interdisciplinary perspective, combining knowledge and experimental methods from phonetics, phonology, pragmatics, psycholinguistics and speech pathology. Possible cognitive constraints will be investigated by looking at the effects of working memory capacity in healthy adults and in individuals with cognitive disorders, and in particular, with patients affected by multiple sclerosis (MS). Our results will provide evidence for/against the recent claim in psycholinguistics that incremental planning is under strategic control, and on the role of prosody in models of language production. En particular, our research will contribute to the theoretical debates concerning the planning of speech prodiction, the phonological representation of intonation and the construction of tune meaning. Finally, our research program will contribute to highlight the relationship between cognitive-emotional skills and speech in MS, thus supporting the development and application of assessment and remediation strategies for this pathology.
We will run several experiments including different types of neuropsychological and linguistic tasks in order to evaluate the possible relationship between cognitive capacities and production/perception of prosody. The neuropsychological tasks are chosen among standard tests in pathology, notably to assess cognitive impairement in MS. Cognitive measures were restricted to memory and speed of processing in this work as we have in mind specific hypotheses to be tested concerning cognitive processes underling speech production planning.Linguistic tasks will include behavioral (acoustic recordings of spontaneous and laboratory speech) and psychophysiological (eye-tracking, electrodermal activity) data.
• Constitution of a multi-parametric database for Multiple Sclerosis, which is unique in the world. Data will be made available via the Speech and Language Data Repository after the publication of the related papers.
• Improvement/introduction of sets of standards for collecting, annotating, and quantifying dialogues in pathological speech.
• New collaborations with regional hospitals and with research laboratories in France, US, Australia.
• The project was a springboard for a new funded project with cultural actors at the regional level on the expression of emotions in movies and in research.
Our results will have a strong impact on the literature on speech planning (especially concerning flexibility), prosody (for debates concerning tune representation and meaning) and pathology (for the impact of cognitive impairement on speech and communication). In the future, neuroimaging studies will be used to explore whether our findings can be interpreted in a directional, causal manner.
New scientific perspectives have been open to me after these studies. Since October 2019, I will be responsible of a new project on emotions which is funded by A*MIDEX (with a RAPP partner, M. Champagne Lavau). A PhD student will work with me on the psychophysiological effects of prosody in clinical interactions. Because of the interest in emotional prosody developed during the ANR project, I was recently contacted for external consultance services.
The project allowed the submission of 3 papers to major journals, 1 chapter, 4 invited talks, 2 papers in conference proceedings, and 13 communications in national and international conferences.3 papers are in preparation and 3 more papers will be prepared by 2020. They will be submitted to major journals. 7students at AMU worked on related topics for their masters. 5 students in Classical Philology from Italy worked on related topics during their Erasmus Plus stage at LPL. Two of them moved towards prosody and clinical phonetics and got a PhD grant. Since Oct 2019, a new PhD student will start working with C. Petrone on the role of prosody in clinical contexts.
The RAPP project is the first one at investigating prosodic planning in French by determining whether (1) the size of the planning unit is flexible adapted by the speakers depending on the pragmatic situation; (2) prenuclear information can be used as soon as it become available to predict the pragmatic value of an utterance; and (3) speaker-specific differences can be explained by variations in cognitive abilities. The first two questions will be investigated through (semi-)spontaneous acoustic corpora and eye-tracking experiments. The existence of possible cognitive constraints will be determined by comparing data from healthy adults with individuals with working memory disorders (multiple sclerosis). Our results will provide evidence for/against the recent claim that incremental planning is under strategic control, and on the role of prosody in models of language production. The results will also have implications for theories of intonational phonology and meaning.
Madame Caterina Petrone (CNRS/Délégation Provence et Corse/Laboratoire Parole et Langage)
The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.
CNRS DR12/LPL CNRS/Délégation Provence et Corse/Laboratoire Parole et Langage
Help of the ANR 189,939 euros
Beginning and duration of the scientific project: September 2014 - 48 Months