You are viewing the SNL 2018 Archive Website. For the latest information, see the Current Website.

Poster B53, Thursday, August 16, 3:05 – 4:50 pm, Room 2000AB

The Impact of Semantic Context and Age of Acquisition on Bilingual Speech Perception in Noise: An ERP Study

Kristina Coulter1,4, Annie C. Gilbert2,4, Shanna Kousaie2,4, Shari Baum2,4, Vincent Gracco2,3,4, Denise Klein2,4, Debra Titone2,4, Natalie A. Phillips1,4;1Concordia University, 2McGill University, 3Yale University, 4Centre for Research on Brain, Language, and Music

Speech perception often occurs in suboptimal listening conditions (e.g., background noise or multiple talkers). In these noisy listening conditions, listeners benefit from cues such as semantic context while listening in their native language (L1). However, the extent to which bilinguals can use semantic context while perceiving speech in noise in their second language (L2) is unclear. Therefore, we examined speech perception in noise in bilingual speakers of English and French, with varying levels of L2 proficiency and L2 age of acquisition. Participants included 15 simultaneous bilinguals, 19 early and 16 late sequential bilinguals, matched on L1, age, sex, and verbal fluency in L2. Participants performed an adapted version of the Revised Speech Perception in Noise Task and event-related potentials (ERPs) were analyzed. In this task, participants listened to English and French sentences varying in sentential constraint. High constraint sentences led to a predictable final word whereas low constraint sentences led to an unpredictable final word (e.g., high constraint: “The lion gave an angry roar.”; low constraint: “He is thinking about the roar.”). Sentences were presented in a quiet condition and a noise condition with a 16-talker babble mask. Participants were asked to repeat the final word of each sentence. As expected, participants showed higher accuracy in quiet compared to noise and on high constraint compared to low constraint sentences. While there was no difference in accuracy between L1 and L2 for simultaneous and early sequential bilinguals, late bilinguals were more accurate in L1 compared to L2. In addition, a trending interaction was observed such that simultaneous bilinguals appear to perform with higher accuracy than early and late sequential bilinguals in both L1 and L2 in noise. ERP analyses revealed overall larger N400 amplitudes in L2 compared to L1, suggesting more effortful processing in L2 compared to L1. A larger and later N400 was observed for low constraint sentences compared to high constraint sentences across all groups and in L1 and L2, indicating more effortful processing of low compared to high constraint sentences. Additionally, high constraint sentences elicited larger N400 amplitudes in L2 compared to L1 while low constraint sentences elicited similar N400 amplitudes in L1 and L2. These findings indicate that, although bilinguals benefit from semantic context during speech perception in both their languages, bilinguals appear to benefit from semantic context to a lesser extent in their L2 compared to L1. Additionally, age of second language acquisition does not seem to influence processing of speech in noise as indicated by similar N400 amplitudes and latencies across all groups. Future directions include investigating the role of second language proficiency on bilingual speech perception in noise.

Topic Area: Multilingualism