2026-10-06T14:10:15Z https://api.figshare.com/v2/oai

oai:figshare.com:article/33368110 2026-09-23T01:33:20Z category_29425 category_25648 category_30391 category_29338 portal_886 item_type_3 month_year_09_2026

Supporting data for "Music-to-Language Transfer in English as a Second Language Learners: A Scoping Review of Models and an Empirical Study on Speech Perception and Production" Cheuk Lam Chau (19020368) Linguistics not elsewhere classified Musicology and ethnomusicology Cognitive and computational psychology not elsewhere classified English as a second language Speech perception final lengthening boundary pauses Speech production prosody MFCCs Grounded in Meredith’s (1993) typology of theory construction which distinguishes models (descriptive, non-explanatory, simplified depictions of reality), frameworks (explanatory structures with organised variables and relationships), theories (coherent systems for prediction and explanation of phenomena), and hypotheses (empirically testable statements), this thesis evaluates the mechanisms of music-to-language transfer among English as a second language (ESL) learners through a scoping review of models and an empirical study on speech perception and production, specifically investigating the Precise Auditory Timing Hypothesis (PATH; Tierney & Kraus, 2014).The scoping review addresses models representing music-to-language transfer among ESL learners in linguistically distant contexts, mapping represented populations and empirical evidence that support these models. Models of transfer are classified into three pathways—cognitive, perceptual, and affective—informing frameworks such as the OPERA (Overlap, Precision, Emotion, Repetition, Attention) Hypothesis (Patel, 2011, 2012; 2014) and beyond.Guided by the predominantly perceptual model of transfer, the cross-sectional study evaluates PATH. A total of 72 native Cantonese-speaking ESL participants with above-average English proficiency in Hong Kong were recruited, with 36 categorised into a musician group (seven years or above of formal instrumental music training) and 36 into a non-musician group (less than two years of formal instrumental music training). Individual differences in auditory working memory and gender were controlled as covariates.For final lengthening perception, binomial generalised linear mixed-effects models and pairwise contrasts revealed that musicians exhibited significantly higher discrimination accuracy than non-musicians across monosyllabic and disyllabic phrasal structures, whereas group differences disappeared under higher structural complexity. Working memory showed an independent covariate association. For boundary pause discrimination, grouped quasibinomial logistic regression revealed no significant main effect of musicianship but identified a significant positive relationship between auditory working memory and accuracy.In standardised speech production, multivariate analyses revealed no global variation in prosodic or spectral profiles by musical status. However, targeted follow-up models identified a localised, statistically significant main effect of musicianship on a static Mel-frequency cepstral coefficient (MFCC).These findings provide converging evidence for PATH, indicating that long-term music training enhances temporal perception precision in speech. Interestingly, the music-to-language transfer effect may be more measurable in spectral features rather than in prosodic features of speech. The localised effect of musicianship on static MFCC highlights its potential as an acoustic marker. Future research may explore the utilisation of MFCCs as diagnostic features in speech pathology, offering a measurable index of training-related adaptations and therapeutic progress. 2026-09-23T01:33:20Z Dataset Dataset 10.25442/hku.33368110.v1 https://figshare.com/articles/dataset/Supporting_data_for_Music-to-Language_Transfer_in_English_as_a_Second_Language_Learners_A_Scoping_Review_of_Models_and_an_Empirical_Study_on_Speech_Perception_and_Production_/33368110 CC BY-NC 4.0 Embargoed