We asked whether listeners can use phonological information to generate predictions about upcoming words by capitalizing on two key properties of Mandarin Chinese, a tonal language with a rich classifier system. First, a noun must be preceded by a classifier when it is modified by a numeral (e.g., ‘san ben shu’ = three-CL-books) and different classifiers are associated with nouns with different semantic-conceptual properties. Further, a syllable may undergo tone change depending on the tone of the syllable that follows (tone sandhi). We adapted a classic visual-world eye-tracking paradigm to examine two different tone sandhis that can occur within a noun phrase, one general to all tone-3 syllables in the language (third tone sandhi) while the other is more specific (‘yi’ sandhi). Crucially, we found initial evidence that listeners can use the tone of the numeral to predict the upcoming classifier and noun when the tone sandhi rule involved is general, but not when it is lexically specific.
Hello this is Wing-Yee. I'll be online during the poster session on Friday (12-2pm EDT). Please send me an email with your preferred way of contact at wingyee.chow@ucl.ac.uk and I'll try to find you online! Alternatively if you'd like to leave your questions on OSF (please use the chat button on the top-right corner) I'll try to get back to you asap!