It is largely unknown how language learners manage to differentiate the vowels of the language they are exposed to. The main focus of this project is on two major tasks that any first language learner masters successfully at a very early age, and the project also examines how the typically less successful second language learners cope with one of these tasks.
The first task consists of solving the perceptual constancy problem: The same vowel category, e.g., /o/ in the Danish word to, varies greatly in its acoustic realizations, depending on whether men, women, or children pronounce this vowel, whether it is spoken rapidly or slowly, and it also depends on which consonant occurs before and/or after the vowel. Computers have a very hard time discovering that all the different realizations belong to one category, but infants are highly successful at this even before their first birthday. Using a well-established technique for studying achieve perceptual abilities, the project examines how infants aged 7-12 months achieve perceptual constancy for vowel categories.
The same technique will be used for experiments aimed at understanding the second task, which consists of structuring the so-called vowel space. Each language has a particular way of making use of the human potential for forming vowels, and the successful language learner somehow figures out how the universal vowel space is exploited in a particular language. The project contains a series of experiments examining how infants go about this task, and it also examines how the ability to perform this task develops over the life span. In particular, the project examines how pre-school children and adults structure the vowel space and how they differentiate non-native vowel contrasts.
The project will increase and deepen our knowledge of the very beginnings of first language acquisition. In particular, it will provide further insights into the astonishing perceptual abilities of young infants. These abilities provide the basis for the acquisition of all other aspects of language. The project will also result in a better understanding of age-related changes in speech perception, and of some of the mechanisms that non-native listeners use when they try to differentiate foreign language vowel contrasts. The results of this part of the project are likely to improve attempts to teach the perception and production of foreign language speech sounds.
Back to main page Back to research page