Lexical representation of variable prosodic patterns in infants

点击翻译按钮获取中文摘要

1. Context of ResearchChildren must learn how to recognise different words in their language. This means that they must remember the sound patterns of the words accurately enough to distinguish similar sounding words. For instance, infants exposed to English must learn that words such as 'pin' and 'bin' refer to different things because of the difference in the first consonant. Thanks to previous research, we have begun to understand when and how this process takes place. For instance, infants' perception of sounds becomes language-specific around 8 to 10 months of age. That is, infants around that age begin to recognise the phonetic distinctions used in their native language. The next step in the development is to learn how to use this knowledge to remember the forms of different words. This seems to happen gradually during the second year. For example, some time between 14 and 17 months, English-speaking infants become able to associate two similar sounding made-up words (such as 'bih' and 'dih') with two different objects. 2. Aims and ObjectivesThe purpose of this research project is to address two particular issues that have not been fully addressed in previous research on this developmental process. First, little is known about when and how infants begin to learn words that are differentiated in terms of what is known as prosodic features: duration, loudness and pitch (highness or lowness) of sounds. For example, in Japanese, 'ame' pronounced with a falling pitch contour means while the same sequence pronounced with a rising contour means . When do Japanese-speaking infants learn how to remember and recognise different words based on such a distinction? Does it happen before they learn how to remember and recognise words based on different consonants or vowels (as in the difference between 'pin' and 'bin' in English)?Second, research so far has not systematically investigated how children learn such detail of word forms when the crucial aspect of the sound pattern changes from one context to another. For instance, the word for in Japanese can also have a flat pitch contour in addition to the rising contour mentioned above, depending on the position of the word in a phrase. How do Japanese-speaking infants deal with this variability? Do they first focus on the form in one context? And how do they learn that 'ame' means if it has a rising OR flat pitch contour, but if it has a falling contour? These questions will be addressed in this project using laboratory experiments in which infants are exposed to recorded sound stimuli (i.e., made-up words) paired with visual presentation of unfamiliar objects. How Japanese infants respond to repeated or changed combinations of the sound and visual materials can indicate the extent to which they can use pitch patterns to differentiate new words under different conditions.3. Potential BenefitsThe reason Japanese is used in this project is that the pitch contrast in the language allows us to tease apart the learning of prosodic contrasts (pitch) from that of segmental contrasts (consonants and vowels) and also systematically observe the effects of variability. However, the outcome of the project will have broader implications for infants' learning of prosodic contrasts in general, including, for example, stress contrasts in English. Furthermore, it will lay a foundation for further research on children's learning of sound variability in word forms, which is a hallmark of speech recognition in human language.This research will make a contribution to the understanding of how very young children begin to associate sounds and meanings in words -- a key step in the process of language development. As such, the results of this research will be appreciated by researchers in many disciplines, most notably linguistics, psychology, cognitive science and education.