Exploring the format of sensorimotor information in human speech perception

早期言语学习的一个特点是，婴儿一开始对言语的感知能力很好，但直到后来才有了相应的高级言语产生方式。因此，目前还不清楚成人言语知觉和语言产出之间普遍观察到的、高度互动的联系是如何发展起来的。我将在NSERC的支持下探索一种可能性，即关于非言语手势的感觉和运动(感觉运动)信息--如咀嚼、吮吸或呼吸，这些都是出生后不久产生的手势--与婴儿的言语感知相互作用，这是言语感知和生产之间类似成人互动的先兆。我还会问，非言语手势信息是否以及如何继续影响成年人的言语感知。*我提议的研究的长期目标是a)了解关于所有类型的发声运动控制的感觉运动信息(言语或非言语)是否以及如何与言语感知有关，以及b)了解这些信息的格式在整个生命周期中如何变化。在短期内，我将专注于研究言语感知与非言语手势之间的相互作用，如咀嚼、吮吸或呼吸，因为这些是婴儿和成年人都可以产生的声道手势。*婴儿将被调查，因为这是一个言语产生能力的戏剧性发展变化时期，我的研究小组将使用经典工具，如父母问卷，但也使用计算方法(例如，对一天的录音进行自然语言处理)，以及婴儿面部肌肉激活的肌电记录。我的研究小组也将在成年人中探索类似的问题，调查呼吸或咀嚼/吮吸手势的更精确元素--如嘴唇运动的动态特性，或呼吸时气流的方向和速度--如何影响成年人的言语感知。*这项研究对语言学、言语科学、心理学和神经科学的研究具有重要的科学意义，因为人们现在对早期语音发育如何影响或塑造人类语音感知中的学习，以及反之亦然，知之甚少。这项工作也将具有重要的临床应用，因为了解知觉和运动系统之间的相互作用对于区分、诊断和治疗知觉或运动起源的言语病理至关重要。

A hallmark of early speech learning is that infants initially perceive speech well, but do not have correspondingly advanced ways of producing speech until later ages. It thus remains unclear how the commonly observed, and highly interactive links between adult speech perception and production develop. A possibility I will explore with NSERC support is that sensory and motor (sensorimotor) information about non-speech gestures—such as chewing, sucking, or breathing, which are all gestures produced soon after birth—interact with infant speech perception, which is a precursor of adult-like interactions between speech perception and production. I will also ask whether and how non-speech gesture information may continue to inform adult speech perception.******The long-term goals of my proposed research are a) to understand whether and how sensorimotor information about all types of vocal motor control, speech or non-speech, are linked to speech perception, and b) to understand how the format of this information changes throughout the lifespan. In the short-term I will focus on studying the interactions between speech perception and non-speech gestures like chewing, sucking, or breathing, as these are vocal tract gestures that can be produced by both infants and adults.******Infants from 4 to 18 months of age will be investigated, as this is a period of dramatic developmental change in speech production ability, which my research group will measure using classic tools like parental questionnaires, but also using computational methods (e.g., natural language processing of daylong audio recordings), as well as electromyographical recordings of infants' facial muscle activation. My research group will also explore similar questions in adults, investigating how the more precise elements of breathing or chewing /sucking gestures—like the dynamic properties of lip movements, or the direction and velocity of airflow while breathing—affect adult speech perception. ******This research has important scientific implications for the study of linguistics, speech science, psychology, and neuroscience, because so little is now understood about how early vocal development informs or shapes learning in human speech perception, and vice versa. This work will also have important clinical applications, because understanding interactions between perceptual and motor systems for speech is crucial for distinguishing, diagnosing, and treating speech pathologies with either perceptual or motor origins.