Neural and psychological correlates of phonological categories

理解语言的结果是将语音信号沿着听觉通路非常有效地解码成长期记忆中的联想结构。声音和意义之间的路径由不同类型的表征所中介。这些编码梯度信息中的一些来自语音输入(语音表示)，但有些是离散的和抽象的，例如语音也必须映射到其上。在这个项目中，我们将试图通过以下方式破译这些表征的性质：(1)通过脑磁图的早期神经反应来监测它们的激活；(2)通过脑磁图和功能磁共振来评估服务于这些表征的大脑底物。主要集中在两个方面：音位范畴和音节范畴。音素和/或(次)音节表征将被探讨：(A)它们与语音输入中的信息的适合性，涉及语段和相邻语音语境的不足；(B)它们对组合的听觉和视觉语音信息的敏感度；(C)通过训练和学习评估的表征系统的可塑性和灵活性；以及(D)在音素和音节类别中处理和表征现有和新学习的表征所涉及的神经基础。

Understanding language results from a very efficient decoding of the speech signal along the auditory pathway to associative structures in long-term memory. The path between sound and meaning is mediated by different types of representation. Some of these code gradient information from speech input (phonetic representations), but some are discrete and abstract, such as phonological categories, onto which speech also must be mapped. In this project, we will try to decipher the nature of these representations by (1) monitoring their activation by means of early neural responses with MEG and by (2) assessing the brain substrate subserving such representations with MEG and fMRI. Two main representations are focused upon: phonemic and syllabic categories. Phonemic and/or (sub)syllabic representations will be probed for (a) their fit with information in the speech input, in relation to underspecification of segments and of adjacent speech context; (b) their sensitivity to combined auditory and visual speech information; (c) the plasticity and flexibility of the representational system as assessed by training and learning, and (d) the neural substrate involved in processing and representation of existing and newly learned tokens within phonemic and syllabic categories.