The Development of Phonetic Perception

我的研究项目是在语音片段（元音和辅音）的水平上考察言语感知的发展。 广泛的目标是充分描绘感知偏见，婴儿显示在回应讲话，解释如何以及何时这些感知偏见形成的发展，并了解他们在口语习得的作用。本文提出的研究重点是两个公认的感知偏见。在研究1和2中，我们考察了焦点元音偏向。这种偏见表明我们对发音运动塑造物理语音信号的方式非常敏感，并且在成人和6到12个月大的婴儿中有很好的记录。 在研究1中，我们将测量4个年龄组（新生儿，2个月，4个月和6个月）的焦点元音偏差（使用听觉元音），并使用LENA（语言环境分析）记录技术收集婴儿语言输入的信息。 在研究2中，我们将使用相同的元音，并在两个额外的条件下跟踪焦点元音偏见的早期发展：1）当婴儿可以看到和听到正在产生的元音时，以及2）当婴儿只能看到正在产生的元音时。 研究1和2的数据将建立焦点元音偏差的发展路径，澄清形成这种偏差的因素，并扩大我们对婴儿如何整合不同感觉方式的语音信息的知识。** 在研究3和4中，我们调查了婴儿说话者的偏见，这是我们最近的突破性发现，即婴儿（4 - 6个月）更喜欢听具有婴儿说话者声音特性的元音声音，而不是成年人发出的元音。 在研究3中，我们将测量4个较低年龄组（新生儿，2个月，4个月和6个月）的婴儿说话者偏差，并使用LENA技术收集婴儿语言输入的信息。 研究结果将揭示这种偏见是什么时候形成的，并将澄清婴儿导向的语言和婴儿自己的发声在形成这种偏见的作用。 在研究4中，我们将探讨是否特定的婴儿反应更有效地诱导婴儿的讲话比成人的语音声音。当婴儿听元音时，我们会记录他们的发声和嘴部运动，并测量他们的心率（指示注意力）和面部肌肉（指示微笑）的变化。 我们还将使用一种新的、对婴儿友好的超声波记录界面来记录婴儿的舌头运动。 婴儿语言是激发特定婴儿行为的有力刺激，这一证据具有重要的理论和实践意义。这一结果表明，婴儿语言的获得具有非常广泛和有意义的影响-影响语言发展的接受，表达和动机方面。新的超声界面也将产生超越目前工作的重大影响，使我们能够探索早期发音发育的未知方面。

My research program examines the development of speech perception on the level of phonetic segments (vowels and consonants). The broad objective is to fully portray perceptual biases that infants display in response to speech, explain how and when these perceptual biases form in development, and understand their role in the acquisition of spoken language. The research proposed here focuses on two well-established perceptual biases.******In Studies 1 and 2, we investigate the focal vowel bias. This bias demonstrates our exquisite human sensitivity to the way that articulatory movements shape the physical speech signal, and is well documented in adults and 6- to 12-month-olds. In Study 1, we will measure the focal vowel bias (using auditory vowels) in 4 age groups (newborns, 2, 4 & 6 months) and collect information on infant language input using LENA (Language Environment Analysis) recording technology. In Study 2, we will use the same vowels and track the early development of the focal vowel bias in 2 additional conditions: 1) when babies can both see and hear the vowels being produced and 2) when babies can only see the vowels being produced. The data from studies 1 and 2 will establish the developmental path of the focal vowel bias, clarify the factors that shape this bias, and expand our knowledge on how infants integrate speech information across different sensory modalities. ******In Studies 3 and 4 we investigate the infant talker bias, which is our recent, ground breaking discovery that pre-babbling infants (4 - 6 months) prefer to listen to vowel sounds that have the vocal properties of an infant talker over vowels produced by an adult. In study 3, we will measure the infant talker bias in 4 younger age groups (newborns, 2, 4 & 6 months) and collect information on infant language input using LENA technology. The findings will reveal when this bias is formed and will clarify the role of infant-directed speech and the infant's own vocalizations in shaping this bias. In study 4, we will examine whether specific infant responses are induced more effectively by listening to infant speech than to adult speech sounds. While pre-babbling infants listen to vowels sounds, we will record their vocalizations and mouth movements and also measure changes in their heart rate (to index attention) and facial muscles (to index smiling). We will also record infant tongue movements using a new, infant-friendly ultrasound recording interface. Evidence that infant speech is a potent stimulus for provoking specific infant behaviors would have significant theoretical and practical implications. This outcome would indicate that access to infant speech has a very broad and meaningful impact - influencing receptive, expressive, and motivational aspects of speech development. The new ultrasound interface will also have a significant impact beyond the present work, by allowing us to explore uncharted aspects of early articulatory development.*********