Intracranial mapping of linguistic structure building during listening speech comprehension

听觉言语理解过程中语言结构构建的颅内映射

阅读:1

Abstract

Understanding how the human brain processes linguistic structures during speech comprehension is essential for understanding the neural mechanisms of language processing. Despite prior efforts to map the cortical organization of hierarchical linguistic structures, a comprehensive characterization with both high spatial and temporal resolution remains limited. To address this issue, our study applied intracranial stereo-electroencephalography (sEEG) recordings to map the neural tracking at distinct linguistic timescales in the human brain in 20 participants. We employed a frequency tagging paradigm to tag syllable-, phrase-, and sentence-level structures at specific frequencies. The findings revealed robust neural tracking in the fronto-temporal regions at different linguistic levels, with the primary auditory cortex (Heschl's gyrus) exhibiting the strongest activity at the syllabic level, and the superior temporal and inferior frontal regions engaged in the building of higher-level phrasal and sentential structures. Importantly, the neural tracking responses to hierarchical linguistic structures were spatially differentiated across different brain regions. Furthermore, high-gamma responses detected by intracranial electrodes within the left language network were observed to be sparsely distributed, predominantly concentrated in specific fronto-temporal sites. These results suggest spatially distributed functional specialization in the brain for tracking different levels of linguistic structure, as reflected in distinct neural oscillatory responses. This study advances our understanding of the functional organization in cortical regions during auditory speech comprehension and may offer insights for aligning brain processes in language models.

特别声明

1、本页面内容包含部分的内容是基于公开信息的合理引用;引用内容仅为补充信息,不代表本站立场。

2、若认为本页面引用内容涉及侵权,请及时与本站联系,我们将第一时间处理。

3、其他媒体/个人如需使用本页面原创内容,需注明“来源:[生知库]”并获得授权;使用引用内容的,需自行联系原作者获得许可。

4、投稿及合作请联系:info@biocloudy.com。