Presentation
Search Abstracts | Symposia | Slide Sessions | Poster Sessions | Lightning Talks
Neural resources underlying prosody-syntax interactions
Poster C51 in Poster Session C, Wednesday, October 25, 10:15 am - 12:00 pm CEST, Espace Vieux-Port
This poster is part of the Sandbox Series.
Giulio Degano1, Narly Golestani1,2; 1University of Geneva, 2University of Vienna
Recent research aims to understand how the brain encodes various levels of linguistic information using naturalistic speech stimuli, allowing more generalizable neuroscientific exploration of brain function (Broderick, 2018; Huth, 2016; Pallier, 2010; Brennan, 2012; Rutten, 2018; Santoro, 2017). Few studies have focused on the spatial distribution or temporal dynamics underlying the interaction between the processing of different levels of language, for example in the processing of higher-level syntactic operations versus that of lower-level features like speech prosody. Linguistic prosody has been shown to play a role in guiding the interpretation and prediction of syntactic structure (Bennett & Elfner, 2019), and it is therefore timely to investigate how the neural processing of these two features interact to influence the predictability and reliability of speech processing. In a recent study, we examined the neural processing of syntax in the temporal domain using MEG data (Degano, 2023). We found that prosodic cues enhance the neural processing of syntax, revealing an intricate interaction between acoustic features and more abstract linguistic information, and their pivotal role in facilitating effective human communication. Given the intrinsic limitation of MEG data in localising relevant brain networks, we are currently extending these questions to an openly available fMRI dataset. Further, we’re interested in seeing the extent to which our findings generalise across languages differing in terms of their typological distance. The current project aims to shed light on the neural localization underlying the prosodic boosting of syntax For this, we are analysing an open source fMRI dataset comprising a total of 112 native speakers of one of three languages: English, Chinese and French speakers (Li, 2022). Each language group listened to excerpts of the audiobook “The little prince” in their respective languages: English (49 participants), Mandarin (35 participants), and French (28 participants). We will first extract prosodic and syntactic information from the language-specific speech stimuli themselves, in the different languages, to a) replicate the previously observed English-language syntax-prosody interaction, b) test for the presence of such a relationship also in French and in Mandarin. Prosodic features will include intonational and rhythmic information obtained through spectral analysis of the speech signal (Suni, 2017), while syntactic features will make use of statistical and linguistic modelling to capture the syntactic information within the stimuli in each language . We will also make use of typological databases (e.g. https://wals.info/) to identify the best local and also more distant syntactic features of interest. We will then analyse the fMRI data using encoding models (Naselaris, 2011) to investigate the brain network that encodes the prosody-syntax interaction, in the fMRI data obtained in the different languages. This dataset offers the unique opportunity to examine the question of the prosody-syntax interaction across three languages that differ in terms of their typological features both at the prosodic and syntactic levels, with English and French being more similar to one another than Mandarin(Dryer, 2013), shedding light on the brain handles these linguistic features in diverse language contexts.
Topic Areas: Syntax and Combinatorial Semantics, Prosody