Suppr超能文献

狨猴初级听觉皮层中发声音调的分布式表示。

Distributed representation of vocalization pitch in marmoset primary auditory cortex.

机构信息

Biomedicine Discovery Institute and Department of Physiology, Monash University, Clayton, Victoria, Australia.

Centre of Excellence in Integrative Brain Function, Australian Research Council, Clayton, Victoria, Australia.

出版信息

Eur J Neurosci. 2019 Jan;49(2):179-198. doi: 10.1111/ejn.14204. Epub 2018 Nov 2.

Abstract

The pitch of vocalizations is a key communication feature aiding recognition of individuals and separating sound sources in complex acoustic environments. The neural representation of the pitch of periodic sounds is well defined. However, many natural sounds, like complex vocalizations, contain rich, aperiodic or not strictly periodic frequency content and/or include high-frequency components, but still evoke a strong sense of pitch. Indeed, such sounds are the rule, not the exception but the cortical mechanisms for encoding pitch of such sounds are unknown. We investigated how neurons in the high-frequency representation of primary auditory cortex (A1) of marmosets encoded changes in pitch of four natural vocalizations, two centred around a dominant frequency similar to the neuron's best sensitivity and two around a much lower dominant frequency. Pitch was varied over a fine range that can be used by marmosets to differentiate individuals. The responses of most high-frequency A1 neurons were sensitive to pitch changes in all four vocalizations, with a smaller proportion of the neurons showing pitch-insensitive responses. Classically defined excitatory drive, from the neuron's monaural frequency response area, predicted responses to changes in vocalization pitch in <30% of neurons suggesting most pitch tuning observed is not simple frequency-level response. Moreover, 39% of A1 neurons showed call-invariant tuning of pitch. These results suggest that distributed activity across A1 can represent the pitch of natural sounds over a fine, functionally relevant range, and exhibits pitch tuning for vocalizations within and outside the classical neural tuning area.

摘要

发声的音高是一种关键的通讯特征,有助于识别个体,并在复杂的声学环境中分离声源。周期性声音的音高的神经表示是明确的。然而,许多自然声音,如复杂的发声,包含丰富的、非周期性的或不严格周期性的频率内容和/或包含高频成分,但仍然唤起强烈的音高感。事实上,这样的声音是规则,而不是例外,但编码这种声音的音高的皮质机制是未知的。我们研究了狨猴初级听觉皮层(A1)高频代表神经元如何编码四种自然发声的音高变化,两种以类似于神经元最佳灵敏度的主导频率为中心,两种以主导频率低得多的为中心。音高在狨猴可以用来区分个体的精细范围内变化。大多数高频 A1 神经元对四种发声中的音高变化都很敏感,只有一小部分神经元表现出对音高不敏感的反应。从神经元的单耳频率响应区域来的经典定义的兴奋性驱动,预测了对发声音高变化的反应,在不到 30%的神经元中,表明观察到的大多数音高调谐不是简单的频率水平响应。此外,39%的 A1 神经元表现出对音高的叫声不变调谐。这些结果表明,A1 中的分布式活动可以在精细的、功能相关的范围内代表自然声音的音高,并对经典神经调谐区域内和外的发声表现出音高调谐。

文献AI研究员

20分钟写一篇综述,助力文献阅读效率提升50倍。

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索

文档翻译

学术文献翻译模型,支持多种主流文档格式。

立即体验