灵长类大脑如何加工声音信息：颅内记录、fMRI 与比较研究的新图景

Voices are among the most socially informative sounds in our auditory landscape, but only recently has a coherent picture of the neural machinery supporting their perception begun to emerge. New evidence from intracranial recordings, fMRI, and comparative primate research reveals specialized voice patches and neural pathways for rapid parallel extraction of identity, emotion, and social cues. These findings point to an evolutionarily conserved system with striking parallels to face processing. Emerging computational models further show how the brain may transform complex acoustic signals into abstract, low-dimensional representations. These advances raise key outstanding questions about the organisation, causal role, and computational principles of voice-selective circuits across primates.

中文翻译还没有生成，先显示原文。