ISSN 1671-3710
CN 11-4766/R
主办:中国科学院心理研究所
出版:科学出版社

心理科学进展 ›› 2024, Vol. 32 ›› Issue (3): 499-513.doi: 10.3724/SP.J.1042.2024.00499

• • 上一篇    下一篇

嗓音模仿认知神经加工的多阶段模型

胡砚冰1, 蒋晓鸣1,2()   

  1. 1上海外国语大学语言研究院
    2上海外国语大学语言科学与多语智能应用重点实验室, 上海 201620
  • 收稿日期:2023-08-04 出版日期:2024-03-15 发布日期:2024-01-19
  • 通讯作者: 蒋晓鸣, E-mail: xiaoming.jiang@shisu.edu.cn
  • 基金资助:
    国家自然科学基金委面上项目(31971037);上海市教育发展基金会和上海市教育委员会“曙光计划”项目(20SG31);上海市自然科学基金面上项目(22ZR1460200)

Multi-stage model of neurocognitive processing for vocal imitation

HU Yanbing1, JIANG Xiaoming1,2()   

  1. 1Institute of Linguistics, Shanghai International Studies University, Shanghai 201620, China
    2Key Laboratory of Language Science and Multilingual Artificial Intelligence, Shanghai International Studies University, Shanghai 201620, China
  • Received:2023-08-04 Online:2024-03-15 Published:2024-01-19

摘要:

嗓音模仿是言语交流中关键的认知过程, 即对话一方(说话人)将感知到的另一方(目标说话人)的嗓音信号映射为自己的发声器官运动表征, 达到发声器官运动表征复制和目标说话人嗓音再现的目的。成像结果表明, 嗓音模仿的认知加工涉及颞上回到左侧额下回, 再到发声相关初级运动皮层的神经网络, 并且基底神经节在该网络中发挥协调作用。嗓音辨别能力、嗓音信号至发声运动表征的映射能力以及发声器官肌肉的控制能力的个体差异都会影响嗓音模仿的认知加工。未来研究应该考虑将嗓音模仿与发声障碍以及侵入电极技术结合起来, 旨在共同揭示脑与行为的因果机制, 并进一步应用于言语的终身发展、认知可塑性以及言语预期领域。

关键词: 嗓音模仿, 嗓音再现, 发声器官运动复制, 模仿神经网络, 个体差异

Abstract:

This article provides a comprehensive evaluation of the neurocognitive mechanisms underlying vocal imitation, approached from the speaker's perspective through a multi-stage model. Initially, we dissect this process by analyzing stages in two different models of vocal imitation, illuminating the unique aspects of this cognitive function. Subsequent sections revisit neuroimaging studies to demystify the neural networks underpinning vocal imitation, clarifying specific neural mechanisms which drive cognitive processing in this domain. The third part of the discussion hones in on individual capabilities like voice recognition ability, the skill to map vocal perception onto vocal organ motor commands, and the control over vocal organs. These elements are crucial for comprehending how various factors contribute to the process of vocal imitation.

Firstly, we have clearly addressed the task irrelevant nature of vocal imitation. in a shadowing paradigm, neither the shadowing condition nor the tasks in stimulus-response consistency paradigms require speakers to imitate the voice of the target speaker. These paradigms provide measures of the cognitive processing of imitation through behavioral response indicators, suggesting that vocal imitation can work spontaneously. Under the imitation condition of the shadowing paradigm, speakers are asked to mimic the voice of the target speaker, revealing that vocal imitation can also be deliberately targeted at specific sounds. The spontaneity involved in vocal imitation differs between these paradigms: in stimulus-response consistency, it primarily manifests as automated reactions based on sensorimotor coupling, almost an automatic response triggered by stimuli. In contrast, the spontaneity in the shadowing paradigm is more about reproducing acoustic characteristics without explicit intent, meaning that speakers can accurately replicate the acoustic features of the target speaker even without a clear intention to imitate.

Secondly, we have deepened our understanding of the neural pathways integral to the perception and vocal motor activities that support vocal imitation. This exploration encompasses key neural structures such as the superior temporal gyrus, arcuate fasciculus, inferior frontal gyrus, laryngeal motor cortex, and basal ganglia, all of which are crucial in the selection, coordination, and execution of motor sequences.

Thirdly, the study further emphasizes the critical role of multimodal information in vocal imitation. It highlights how metrics, including the reproduction of acoustic features and the replication of vocal organ movements, are essential in delineating the intricate cognitive processes at play. This comprehensive examination not only sheds light on the complex neural mechanics underlying vocal imitation but also underscores the synergy between various sensory inputs and motor outputs in this sophisticated cognitive function.

In conclusion, this article unravels the cognitive neural foundations of vocal imitation, discussing its definition, behavioral evidence, neural mechanisms, and the role of individual differences. It is proposed that successful vocal imitation relies on the replication of acoustic features and the corresponding movements of the vocal organs. Critically, three primary stages are involved in vocal imitation—perception, perceptual mapping to output, and actual vocal output—significantly impact the behavioral outcomes of this process. An in-depth analysis of these stages enhances our understanding of the complexities of vocal imitation and illustrates the collaboration of various cognitive models and neural foundations in this phenomenon. Future research should integrate vocal imitation studies with vocal disorders and intracranial electrode technology, aiming to decipher the causal links between brain functions and behaviors, and apply these insights to lifelong speech development, cognitive plasticity, and predictive processing in language communication.

Key words: vocal imitation, vocal reenactment, vocal motor movements copy, imitation neural network, individual differences

中图分类号: