Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Native language experience shapes the behavioral profile and neural plasticity of non-native tone learning
Poster Session E, Friday, October 2, 11:00 am - 1:00 pm, Wangari Maathai
This poster is part of the Sandbox Series.
Yuqin Shu1, Ran Tao1, Tyler Perrachione2, Gang Peng1; 1The Hong Kong Polytechnic University, 2Boston University
While native language background is known to shape the perception of non-native speech, the neural mechanisms underlying the acquisition of lexical tones in individuals with different linguistic experiences remain debated. This study investigated how native language experience (tonal vs. non-tonal) influences the behavioral profile and neural plasticity associated with learning Cantonese tones. We recruited 31 native speakers of a tonal language (Mandarin; Females: 13; age: 21.13 ± 2.87 years) and 30 native speakers of non-tonal languages (European languages; Females: 11; age: 21.6 ± 2.03 years), all with no prior knowledge of Cantonese. Participants underwent a four-day perceptual training program to learn to distinguish Cantonese words based on tonal contrasts. The training utilized a visual-feedback-based perceptual paradigm, where participants were presented simultaneously with numbers indicating tone categories, symbols representing abstract pitch contour information, English translations for word meanings, and traditional Chinese characters. Participants completed three tasks of increasing cognitive demand (passive listening, silent repetition, and word identification) during fMRI scanning both before and after training. Behavioral performance was also assessed out-of-scanner using identical word identification tasks. Behavioral results showed significant post-training improvements in the accuracy of identifying Cantonese words through tonal contrast in both groups, with the tonal group achieving better overall performance. However, the patterns of error for the six Cantonese tones were distinct across groups: while the non-tonal group frequently confused various level tones and struggled to distinguish low-level from low-falling tones, the tonal group primarily confused low- and mid-level tones. Relative weight analysis of acoustic cues revealed that the tonal group utilized both pitch height and pitch contour dimensions. In contrast, the non-tonal group relied heavily on height-related cues (e.g., mean F0, onset/offset F0) and duration, with limited sensitivity to pitch contours. Preliminary univariate fMRI analysis (post- vs. pre-training) revealed both shared and group-specific patterns of neural plasticity. Both groups recruited a dorsal frontoparietal network, including the precentral gyrus and superior parietal lobule (SPL), as well as lateral occipital regions which likely reflect the visual feedback component of the training paradigm. Crucially, the two groups exhibited distinct lateralization patterns in the functional recruitment of the inferior frontal gyrus (IFG) and supramarginal gyrus (SMG). The non-tonal group showed right-lateralized recruitment of the IFG (pars opercularis and pars triangularis) and right insular cortex. In contrast, the tonal group demonstrated significant activation in the left IFG and bilateral SMG. Additionally, extensive activation in the precuneus was observed specifically in the tonal group. These preliminary findings suggest that while training can improve non-native tone perception across different language backgrounds, the underlying cognitive strategies and neural pathways may be fundamentally different.
Topic Areas: Speech Perception, Language Development/Acquisition