Neural Processing of Complex Natural Sounds

复杂自然声音的神经处理

• 批准号: 6585149
• 负责人: THANE E FREMOUW
• 金额: $ 4.62万
• 依托单位: UNIVERSITY OF CALIFORNIA BERKELEY
• 依托单位国家: 美国
• 财政年份: 2002
• 资助国家: 美国
• 起止时间: 2002-09-13 至 2005-09-30
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/6585149
• 关键词: auditory stimulus; learning; neural information processing; neurons; postdoctoral investigator; songbirds; sound; sound perception; vocalization

DESCRIPTION (provided by applicant): As humans, we use auditory information in many ways. Perhaps most importantly, the production and comprehension of speech is one of the fundamental ways in which we communicate and interact with one another. Songbirds also use vocalizations to interact with one another, and song learning is an attractive and widely used model for studying the basic neuronal mechanisms involved in the production and comprehension of complex vocalizations such as speech. To fully understanding how the song system (a group of interconnected nuclei that are specialized for song perception and production) works, one must understand the auditory input to the song system. However, relatively little is known about auditory processing, particularly processing of complex sounds such as song, in the nuclei that link the auditory system with the song system. I propose to examine the neural processing of complex synthetic and real song in two such nuclei: Field L and cHV. The first aim is to characterize the response properties of neurons in these two areas using conventional methods and by estimating their spectral temporal receptive fields (STRFs). Results from the two methods will be compared. The second aim is to determine if neurons in these two areas are tuned to the species-specific song and, if so, to what specific features. The third aim is to build a framework for understanding how data obtained from these two areas in anesthetized birds can be linked with that obtained in awake birds.

描述（由申请人提供）：作为人类，我们以多种方式使用听觉信息。也许最重要的是，言语的产生和理解是我们相互交流和互动的基本方式之一。鸣禽还利用发声相互交流，而鸣叫学习是一种有吸引力且广泛使用的模型，用于研究复杂发声（例如语音）的产生和理解所涉及的基本神经元机制。为了充分理解歌曲系统（一组专门用于歌曲感知和产生的相互连接的核）的工作原理，我们必须了解歌曲系统的听觉输入。然而，人们对听觉处理，尤其是连接听觉系统与歌曲系统的细胞核中复杂声音（例如歌曲）的处理知之甚少。我建议在两个这样的核中检查复杂的合成和真实歌曲的神经处理：Field L 和 cHV。第一个目标是使用传统方法并通过估计它们的谱时间感受野（STRF）来表征这两个区域中神经元的响应特性。将比较两种方法的结果。第二个目标是确定这两个区域的神经元是否适应物种特定的歌曲，如果是，则适应哪些特定特征。第三个目标是建立一个框架，以了解如何将麻醉鸟类这两个区域获得的数据与清醒鸟类获得的数据联系起来。