The phonetic properties of the speech signal are determined by a variety of linguistic and extra-linguistic factors, among which the speakers' command of the language, their physical and emotional state, the intention of communication, but also the characteristics of the situation in which they evolve. Many studies have investigated the effects of such factors on the segmental and supra-segmental properties of speech sounds. Although some effects have been firmly established (e.g. the effect of stress and cognitive load on fundamental frequency), a large residual variability is typically observed despite the fact that many factors are experimentally controlled. This variability, traditionally attributed to "individual differences", is generally recognized, at best described, but rarely explained. In the framework of a transactional model of the relation between the human subject and her environment, it appears that a significant part of the inter- and intra-individual variability in speech production can be analyzed as the product of the control exercised by each speaker on her phonetic behavior in order to meet the specific constraints, internal and external, imposed by the situation of communication and all its components.
The main object of our research is thus the phonetic behavior in its strategic dimension. We study how phonetic behavior, both in speech production and in speech perception, is controlled by the speaker-listener in order to respond to the various constraints imposed by the situation of communication, and we examine how the associated mental representations are built, updated, or even restructured. In this talk, we will present specific studies exemplifying our work in recent years (in collaboration with our colleagues of UMONS) on different types of situation allowing for strategic phonetic behavior to arise.
Four situations will be considered: (i) that of patients suffering from affections impacting speech motor control (e.g. Parkinson's Disease), with the aim of distinguishing between the effects of the disease, the effects of treatment, and possible individual compensation strategies; (ii) that of foreign-language learners confronted with phonetic phenomena unexploited in their native language (e.g long-lag VOT for French-speaking learners of English); (iii) that of monolingual and bilingual children (2 to 4 years), in the course of acquiring the phonological system of their native language(s); (iv) that of neurotypical adults submitted to several laboratory tasks (vocal disguise, phonetic compliance, perturbation paradigms), in order to investigate their potential for phonetic flexibility as well as their propensity to exploit it when prompted by the necessities of the communication situation.