Presenter(s): Kayla Walker—Linguistics
Faculty Mentor(s): Melissa Baese-Berk
Session: Prerecorded Poster Presentation
Nonnative-accented speech is more difficult for native listeners to understand than native-accented speech . However, listeners can improve their abilities to understand nonnative- accented speech through exposure and training . The goal of this project is to explore whether exposing native listeners to different sentence types affects listeners’ adaptation to nonnative speech . Listeners will be trained on high predictability sentences (e .g ., “The color of a lemon is yellow”), low predictability sentences (e .g ., “Mom said that it is yellow”), or semantically anomalous sentences (e .g ., “The green week did the page”) . Previous research has demonstrated that semantic predictability impacts speech perception, but its influence on adaptation to nonnative speech is unknown . This research aimed to answer the following questions: Does training with low predictability or anomalous stimuli require listeners to focus more attention on the acoustic-phonetic properties of the accent and thus lead to greater adaptation and generalizable learning? Or will training with high predictability stimuli provide valuable semantic information that will allow listeners to create a better framework for improving perception? Overall, preliminary results indicate that training with low predictability sentences provides listeners with an advantage over training with high predictability sentences . Using higher effort in training with more challenging sentences allows listeners to construct generalizable knowledge about the speech rather than relying on semantic prediction during a transcription task .