Learning the Hindi dental-retroflex contrast through sound-to-meaning associations and sound discrimination tasks.

• Main Author: Teo, Gerielle Ern Ai.
• Other Authors: Fabrizio Galli
• Format: Final Year Project
• Language: English
• Published: 2013
• Subjects: DRNTU::Humanities::Linguistics::Psycholinguistics
• Online Access: http://hdl.handle.net/10356/52194
• Institution: Nanyang Technological University

Studies have shown that perceiving contrasts which do not occur in one’s native language is difficult. Native English speakers, for example, have trouble discriminating between Hindi dental and retroflex consonants as both tend to be heard as instances of English alveolar stops. This study investigated native English speakers’ learning of the Hindi dental-retroflex contrast through a sound-to-meaning training paradigm involving word identification, comparing pre- and post-training performance using a same-different (AX) discrimination task. It also examined if increased exposure to non-native sounds without feedback could improve discrimination, by observing the progress of subjects who underwent multiple AX tasks before training. The role of interstimulus interval (ISI) was also considered, given past research which has shown its effect on working memory. The sound-to-meaning training paradigm was found effective for improving discrimination of not only the Hindi dental-retroflex contrast, but also of sounds already contrasted in subjects’ native language, suggesting the role that incorporation of meaning plays in creating long-term memory representations of sounds. Undergoing multiple pre-training AX tasks without feedback seemed to benefit word identification by the end of training, although it did not improve discrimination before, after and at the start of training. The effect of ISI was only significant before training, with discrimination being better when ISI was shorter.