The Effect of High Variability Phonetic Training with Modified Input on English Speakers' Perception of a Non-Native Stop Contrast
Emma Louise Hayter · 2024
The efficacy of the high variability phonetic training (HVPT) method for learning non-native contrasts has been well documented; however, the use of modified input, or acoustically altered stimuli in an HVPT paradigm is relatively underexplored.In this thesis, I discover whether HVPT with modified input helps native English speakers improve their perception of the Korean lenis and aspirated stop contrast.This thesis examines the 'levels approach' from Hayter (2022), which conceptualizes HVPT and cue weighting studies as falling along a continuum according to their uses of variability and acoustic modifications.I test Levels 1, 3, and 4 from Hayter (2022), and hypothesize that participants in Level 4 will have the greatest improvement in their accuracy in the AX and forced-choice identification (FCID) tasks due to the modification of first and second language cues (voice onset time and F0).Furthermore, I propose that participants in Level 3 will have the second-best improvement in their accuracy, as the second language cue (F0) is modified.Level 1 functions as a control group, since participants are listening to naturalistic stimuli.The results were mixed as to the use of modified input in this HVPT experiment.In the AX task, participants' level of modified input had no impact on their improvement by the final session.In the FCID task, the opposite effect occurred: participants in Level 1 improved the most as a result of training, with Level 4 in second.Level 3 participants' accuracy was relatively unchanged by the final training session.These findings suggest that further research is necessary to determine the effectiveness of modified input for this non-native contrast.