Presentation
Search Abstracts | Symposia | Slide Sessions | Poster Sessions | Lightning Talks
Prosodic cue weighting by French-English bilinguals when assigning lexical stress in English: Preliminary ERP and behavioural results
There is a Poster PDF for this presentation, but you must be a current member or registered to attend SNL 2023 to view it. Please go to your Account Home page to register.
Poster C39 in Poster Session C, Wednesday, October 25, 10:15 am - 12:00 pm CEST, Espace Vieux-Port
Annie C. Gilbert1,2, Claire T. Honda1,2, Shari R. Baum1,2; 1McGill University, Montréal, Canada, 2Centre for Research on Brain, Language, and Music, Montréal, Canada
Learning to adapt one’s prosodic processing to a second or non-dominant language can be difficult because learners need to assign new meaning to acoustic cues already used in their native language (L1). This process is even more difficult if the second language (L2) uses different prosodic constructs than the L1. For instance, while fundamental frequency (F0) rises are associated with lexical stress in English, they are associated with phrase boundaries in French. Lexical stress and phrase boundaries both provide important information about how to segment the speech stream; thus, the misinterpretation of an F0 cue in an L2 (e.g., as indicative of a phrase boundary instead of lexical stress) could lead to segmentation errors, hindering speech processing. In the present study, we paired a behavioral and an event-related potential (ERP) task to examine how French-English bilinguals with varied language experience assign weight to F0 and syllabic duration cues when locating English lexical stress. The stimuli consisted of bisyllabic homonyms with different stress patterns (trochaic nouns, iambic verbs). The homonyms were recorded as both nouns and verbs by a native speaker of English and edited to create stimuli with conflicting lexical stress cues. For instance, syllabic duration could suggest a trochaic pattern while the F0 suggested an iambic pattern. In the behavioural task, participants were presented with original and manipulated versions of the homonyms and were asked to report which syllable was accented. In the ERP task, participants listened to original and manipulated recordings of one homonym in an oddball paradigm blocked by word category (e.g., standard = original noun, deviants = F0 edited noun, duration edited noun, and original verb, and vice versa; Näätänen, et al., 2004) while watching a silent movie. The recorded ERPs were pre-processed according to field standards (Luck, 2005), and difference waves between deviant and standard conditions were computed across language groups (English-L1 and French-L1). To date, twenty English-French bilinguals have participated in both tasks (10 English-L1, 10 French-L1). Interestingly, preliminary analyses of the behavioural and ERP results suggest different patterns of effects, with the behavioural task showing that prosodic modulations (F0 or syllabic duration) affected stress assignment in both homonym categories (noun or verb) in a similar manner across listeners, irrespective of their L1. In contrast, visual inspection of topographic representations of the ERP difference waves suggests that French-L1 listeners reacted differently from English-L1 listeners when hearing stimuli with conflicting prosodic cues, particularly during the verb block. Specifically, English-L1 listeners showed MMN-like negativities to two types of deviant stimuli (original noun and F0 edited verb) while the same deviants elicited positivities in the traditional P300 time-window in French-L1 listeners. Taken together, these preliminary results suggest that although listeners’ preattentive processing of prosodic cues might be conditioned by the cues’ weight in the L1, it does not prevent L2 listeners from adapting their interpretation of the perceived acoustic cues to the specifics of their L2.
Topic Areas: Prosody, Speech Perception