This research project aims to explore how listeners' language experience and speaker accents affect the processing of speech disfluencies. It will involve monolingual and bilingual participants listening to both native and non-native speakers of their respective languages. The study will employ a visual world paradigm, tracking participants' eye movements as they view images related to the speech. The investigation will assess whether listeners adapt their anticipatory processing based on cues like the speaker's accent. Notably, previous studies suggest that listeners tend to anticipate semantically related repairs over phonological ones, but bilinguals, especially those with lower proficiency, may exhibit different patterns, influenced by their language background and the speaker's accent.
This research delves into language control in proficient bilinguals, focusing on how speaker identity influences language switching. Recent studies reveal that bilinguals can proactively choose the language based on their conversational partner's identity. The study employs two experiments with varying levels of detail in cues: one with coarse cues (Experiment 1) and another with fine-grained cues (Experiment 2). By monitoring event-related potentials (ERPs), which offer precise temporal insights, the research aims to unravel when and how speaker identity impacts switch costs. Past experiments demonstrate processing costs for language switches compared to non-switches, although certain cues like subtle phonological shifts can mitigate these costs in comprehension. This project explores the dynamics of language control, shedding light on the temporal and cognitive aspects of bilingual language switching.
Recent research has supported the nonselective access hypothesis, suggesting that bilinguals automatically activate multiple languages during language processing. A significant study by Thierry and Wu (2007) indicated that bilinguals unconsciously translated words into their first language (L1) while processing their second language (L2), affecting the N400 response. However, independent replication of this influential finding is lacking. Recent studies have also questioned the universality of nonselective access, particularly in single-language settings. To address these issues, a high-powered multi-lab replication effort aims to provide robust empirical support for discussions on bilingual language control, addressing the need for further verification of this phenomenon.