The effect of orthographic knowledge on the segmentation of speech

As literate English speakers, we are accustomed to the idea that words are made up of individual consonant and vowel sounds, called segments, that these sounds group into larger units called syllables, and that syllables naturally break down into intermediate units that include the rhyme (rime). But are elements like the segment, syllable, and rime universal, i.e., appropriate for the description of all languages? There is experimental evidence that speakers of some languages (e.g., Mandarin or Taiwanese Chinese) may not segment words into units smaller than the whole syllable, while in other languages (e.g., Korean and Japanese) units called the body or the mora may supplant the rime. However, the native speakers tested so far in all four of the language groups mentioned (English, Chinese, Japanese, and Korean) were all relatively well educated, literate, and often even bilingual. Thus they were all exposed to the writing systems of their own and/or their second language, which might have predisposed them to perform the way they did. Since knowledge of the writing systems has not been controlled in previous studies, in the present research we will test speakers of these languages who have not been subjected to the influence of L1 spelling. Such speakers include preliterate children, adult nonliterates, and bilinguals with “split literacy” (e.g., second generation immigrants in Canada and the US, who have learned to speak Japanese, Korean, or a Chinese language natively, but who are literate only in English).