OPTICAL SPECIFICATIONS OF PHONOLOGY

第三个项目考察了语音行为的显著灵活性，这种灵活性允许语音行为通过各种光学信号进行传输，从而扩展了我们最初的主张，即阅读是建立在从印刷品中提取语音形式的基础上的，语音结构是手势的。说话的面孔、印刷的文字、图片和手势都是提供语音信息的光学信号，但它们在揭示语音信息的完整性和直接性方面有所不同。项目I-II主要关注声学语音信号，而项目III则从观察语音的发音手势也对动态面部运动产生影响开始，其直接但不完全地反映了焦束运动。我们的第一个具体目标是研究语音感知器如何从这些动态的面部模式中提取语音形式。相比之下，印刷文字提供了相对完整的语音信息，但它是通过一种任意的、习得的代码间接传达的。虽然项目IV-VI探索了阅读本身的语音基础，但我们的第二个具体目标是评估从印刷品中提取的语音形式是否与语音形式相同，以及两者是否都是手势。图片通过语义联想而不是正字法编码提供了（尽管是间接和不精确的）对单词作为语音形式的访问。他们的不足之处不同于面孔。我们的第三个具体目标是比较通过图片，印刷，面部和语音的语音访问。如果有一个共同的音韵，个人的表现差异应该共同变化的语音和这些不同的光学信号的感知。我们的四个具体目标探讨了在访问语音形式从语音，印刷品和面孔的性能的共变。手语使用一种非常不同的光学信号，提供关于符号的语音结构的直接和完整的信息。然而没有一种书面语被广泛使用，因此符号内识字不能与其语音结构相互作用。我们的第五个具体目标测试如何签名提取语音信息的标志，在任务类似于那些用于口语。研究结果将告诉我们声学语音信号之外的语音行为的多个实例。这些知识具有重要的理论和临床意义。

Project III examines the remarkable flexibility of phonological behavior that allows its transmission by various optical signals, extending our primary claims that reading is founded on extracting phonological form from print and that phonological structure is gestural. Talking faces, printed words, pictures, and signaling hands are all optical signals that provide access to phonological information, but they differ in the completeness and directness with which they reveal it. Whereas Projects I-II focus primarily on acoustic speech signals, Project III begins with the observation that the articulatory gestures of speech also have consequences in dynamic facial movements, which directly but incompletely reflect focal tract movements. Our first specific aim examines how speech perceivers extract phonological form from those dynamic facial patterns. Printed words, by contrast, provide relatively complete phonological information, but it is conveyed indirectly via an arbitrary, learned code. While Projects IV-VI probe the phonological basis for reading per se, our second specific aim here is to evaluate whether the phonological form extracted from print is in common with that of speech and whether both are gestural. Pictures offer access (albeit indirect and imprecise) to words as phonological forms via semantic association rather than orthographic encoding. Their inadequacies differ from those of faces. Our third specific aim compares phonological access via picture, printing, face and speech. If there is a common phonology, individual performance differences should co-vary across perception of speech and these various optical signals. Our four specific aim explores co-variation in performance in accessing phonological forms from speech, print and faces. Sign languages use a very different optical signal offering direct and complete information about the phonological structure of signs. Yet no written are in wide use, so within-sign literacy cannot interact with its phonological structure. Our fifth specific aim tests how signers extract phonological information from signs, in tasks similar to those used for spoken languages. The results will inform us about the multiple instantiations of phonological behavior beyond the acoustic speech signal. Such knowledge of central theoretical and clinical significance.