This dissertation research will test whether second language learners have shared or separate mental representations of speech sounds that occur in both of their languages. The researchers will recruit English-German and German-English second language learners from the U.S. and Germany. The participants will listen to English speech in which all instances of one particular sound ("f" or "s") have unusual pronunciations. Listeners are predicted to adjust their mental representations of the relevant English speech sound categories, as has been demonstrated for monolingual listeners. Modified sound categories can be revealed through a categorization task in which listeners categorize ambiguous sounds that range on a continuum from "s-" to "f-like". The listeners will categorize sounds on a continuum in English as well as in German. The researchers predict that English-German and German-English second language learners will show adjusted sound categories not only in English, but also in German. This prediction is based on the hypothesis that speech sounds common to two languages, such as "f" or "s" in English and German, have shared or closely interconnected representations in second language learners.
The participants will be grouped into two proficiency levels in their second language (high/near-native vs. low/intermediate) depending on how they perform in a second language perception test. Listeners are predicted to differ in their adjustments of the speech sound categories depending on their proficiency level in their second language, and depending on whether English or German is their native language.
Most language users are not monolingual, but are familiar with more than one language. This project will contribute to the understanding of the malleability and connectedness of linguistic representations of speech sounds in listeners who know more than one language. The knowledge gained from this study will advance the scientific understanding of second language learning, speech perception and bilingualism.
