人类语言学习的神经机制：婴儿音素学习关键期的脑功能影像追踪研究

人类具有高效的语言学习能力。在出生后一年内，婴儿就已经自主地发展出针对母语音素系统的特异性识别能力。而音素系统作为各种语言中连续语音信息的有效切分的关键线索，为婴儿后续的词汇和句法的学习奠定了重要基础。但是，目前我们对婴儿语言学习的神经机制仍知之甚少。基于申请人前期对语言学习脑机制的追踪影像研究经验（0-10岁），本项目拟采用婴儿语音学习和脑功能影像相结合的纵向追踪实验范式，考察婴儿母语（汉语）音素学习关键期的神经机制。具体而言，项目将聚焦于母语音素学习的关键期（6-12月龄）的前后时间节点（3-6月龄；15-18月龄），通过表征相似性的分析技术来考察在两个时间点上, 婴儿大脑中加工母语和非母语音素的神经机制的异同，揭示婴儿的母语音素知识习得的脑机制。项目的实施，将有助于我们进一步理解人类发育早期高效的语言学习能力的神经机制，同时也能为促进早期语言发展的教育实践工作提供理论基础。

Humans are effective language learners. Within the first year of life, infants have developed the implicit ability to recognize the phonemes in their mother tongue, which serve as a universal cue to segment continuous speech streams, paving the road for the subsequent learning of vocabulary and grammar. So far, little is known about the neural mechanisms underlying the infant language learning process. Based on the applicant’s previous experience on the longitudinal imaging investigation on language acquisition in the 0-10-year-old population, the current project proposes to study the neural trajectory underlying the phonetic learning process in infants through an infant longitudinal imaging approach. Specifically, this project aims to characterize the neural mechanisms underlying native and non-native phoneme recognition at two time points, one before (3-6 months old) and one after (15-18 months old) the critical learning period of the Mandarin phonetic system (6-12 months old). Combing the high spatial resolution of the functional MRI data and the representational similarity analysis technique, we will be able to identify and compare the neural representation of the phoneme systems of different languages at the two time points, unraveling the neural mechanisms underlying phonetic learning in infants. The proposed research has the potential to enable our understanding of the neural foundation of effective language acquisition unique to human kind, and also inform educational practices regarding facilitating language learning abilities at an early stage.