Multimodal phonetic training revisited: Comparing the effectiveness of audio-visual and audio-tactile training on L1 Mandarin speakers’ learning of L2 Italian stops

Previous research has shown positive evidence for multimodal phonetic training in second language (L2) phonetic learning. However, less research has compared the effectiveness of different kinds of training paradigms. This study compares the effectiveness of audio-visual (AV) and audio-tactile (AT) training with a traditional auditory-only (AO) condition on L2 sound learning. Mandarin-speaking learners of Italian were trained to perceive and produce L2 Italian stop voicing contrast through a four-session training in one of the three conditions: the AO condition where they listened and repeated the training words, the AT condition where they touched their throats to sense vocal fold vibration related to prevoicing, and the AV condition where they viewed waveforms and spectrograms of prevoicing. Results showed that multimodal training produced stronger and more persistent improvement than unimodal training in both perception and production. AV training yielded better generalization in perception, whereas AT training enhanced generalization in production. These findings suggest that cross-modal cues strengthen L2 phonetic learning by reducing cognitive load, reinforcing memory, and supporting sensorimotor integration. The study provides empirical support for the complementary roles of visual and tactile modalities in enhancing different aspects of L2 speech learning.

Link to session