Glasbergen-Plas, A. , Gryllia, S. , Pablos-Robles, L. & Doetjes, J.
Leiden University
Maintaining the balance between experimental control and ecological validity is challenging. For our study on the prosody of French wh-in-situ questions, we developed a new paradigm for eliciting context-induced production data in E-Prime. Questions like Tu as preparé quel gateau pour le dessert? [You have prepared which cake for the desert?] can be interpreted as echo questions, which can be used when part of the previous utterance was misheard, or as informal requests for information with broad or narrow focus. We investigated whether the differences between echo questions and information requests with broad and narrow focus are prosodically marked when the preceding context is changed accordingly.
21 participants engaged in a series of short, natural dialogues, reading their text from a screen and hearing their interlocutor through ear phones. All stimuli contained noises, part of which created a context for an echo question. Results suggest that a) our context-induced production paradigm results in very naturalistic data, notwithstanding the bimodal task, and b) all three question types (echo, broad/narrow focus) differ in prosody, with echo questions being characterized by a stronger accent and a longer duration of the wh-word, followed by a comparatively high plateau.