Files in this item



application/pdfLi-Hsin_Ning.pdf (5MB)
(no description provided)PDF


Title:Effects of auditory feedback and real-time visual feedback on second language tone learning
Author(s):Ning, Li-Hsin
Director of Research:Shih, Chilin; Loucks, Torrey M.J.
Doctoral Committee Chair(s):Shih, Chilin
Doctoral Committee Member(s):Loucks, Torrey M.J.; Ionin, Tania; Johnson, Aaron; Shosted, Ryan K.
Department / Program:Linguistics
Degree Granting Institution:University of Illinois at Urbana-Champaign
internal model
tone discrimination
L2 tone learning
language experience classification
vocal training experience
Second language (L2)
Abstract:Tone learning is challenging for non-tone speakers. While pitch, like melody, is processed by the right hemisphere for non-tonal language speakers, lexical tone as well as other language aspects (such as phonemes, syntax and semantics) appears to be processed by the left hemisphere for tonal language speakers. Adult second language (L2) learners of Mandarin whose native language is non-tonal may need to acquire left hemisphere dominance of pitch processing for successful tone learning. L2 learners of Mandarin have to acquire new perceptual categories for discriminating and identifying lexical pitch variation along with new sensorimotor skills to produce the rapid tone changes. Language learning, therefore, opens windows for observing changes in behavior and brain functions that impact the learner in multiple ways. I addressed the neural plasticity indirectly by constructing a paradigm that incorporates auditory feedback called pitch-shift paradigm. The pitch-shift paradigm, in which a short and artificial change in pitch is fed back to speakers during vocalization, has been used to investigate how sensory information affects the way we control our speech motor activities. Pitch-shift responses (vocal responses to auditory perturbation) have reflex-like properties and are hard to suppress. So pitch-shift paradigm can be used to understand the stability of internal models for tone and to reveal how internalized pitch representations are built or reshaped. Study 1 examined the language experience effect on pitch-shift responses and non-/linguistic tone discrimination. Discrimination of musical tones was correlated significantly with discrimination of Mandarin tones with the clearest advantage among Mandarin speakers and some advantage for L2 learners. Group differences were found in fundamental frequency (F0) contour shape in response to pitch-shift stimuli. F0 contours of Mandarin speakers were quantitatively least affected by the amplitude and direction of pitch perturbations, suggesting more stable internal tone models. F0 contours of naïve speakers and L2 learners were significantly altered by the perturbation. The findings provide quantitative measurements on how language experience affects tone discrimination and voice F0 control (internal models), and establish the validity of using pitch-shift paradigm to document the success of tone learning. Study 2 included trained vocalists as additional group to investigate the vocal training experience effect on audio-vocal responses and tone discrimination. Mandarin speakers performed significantly better on (adaptive) Mandarin tone discrimination compared to the other three groups. Mandarin speakers also showed more attenuation of pitch shift response amplitude during production of both the sustained vowel (nonlinguistic domain) and Mandarin tones (linguistic domain), especially compared to naïve speakers. The findings suggest Mandarin speakers have more robust pitch control over vocalization and are thus less affected by perturbation in auditory feedback. Trained vocalists also appear to rely more on internal models than naïve speakers in order to regulate voice F0 in the nonlinguistic (sustained vowel) domain, but not in the linguistic domain (Mandarin tone). L2 learners showed only subtle variation in production relative to the other groups. Study 3 explored how enhancing feedback by adding real-time visual feedback to tone production could influence suprasegmental control critical for tones. Results show that both naïve speakers and Mandarin speakers can benefit from the use of real-time visual feedback for stabilizing their voice F0. In this dissertation, I investigated how language experience, vocal training experience, and feedback mode may contribute to successful tone learning by using the pitch-shift paradigm. Native Mandarin speakers demonstrate robust internal models for lexical tones that are evident in perception and production, making language experience an important factor in shaping sensorimotor control. Trained vocalists’ resemblance to Mandarin speakers shows that formal vocal training may speed up tone learning. Furthermore, feedback in any format (auditory or visual) is important for language learning. Motor/feedforward commands for new speech sounds are built and strengthened by repeated practice with the help of auditory feedback. Visual feedback on voice accuracy could also facilitate the construction of language-specific feedforward commands for speech. All in all, the pitch-shift paradigm could be used effectively to tap into speakers’ internal models of lexical tones. Robust internal models built through extensive language experience enable native speakers of Mandarin to maintain steady voice F0 in both linguistic and nonlinguistic domains, when their pitch is altered unexpectedly. Second language learning experience may have changed L2 learners’ internal models because their vocal responses to pitch perturbation were not like naïve speakers’. Although the L2 learners have not acquired native-like internal models of lexical tones, it is expected that extensive exposure to Mandarin will reshape their sensorimotor control.
Issue Date:2014-05-30
Rights Information:Copyright 2014 Li-Hsin Ning
Date Available in IDEALS:2014-05-30
Date Deposited:2014-05

This item appears in the following Collection(s)

Item Statistics