Abstract:

This dissertation compares brain activation patterns between native speakers of Korean and adult speakers of English who have learned the Korean three-way laryngeal contrast of tense /p' t' k'/ vs. lax /p t k/ vs. aspirated /p h th kh /. The main goal of the study is to investigate whether the Internal Model (Callan et al., 2004) can be extended to second language perception of Korean laryngeal distinctions, as well as to identify the brain regions involved in the learning of these famously difficult phonetic categories. A secondary goal of this dissertation is to test whether the underlying representation of each phonation type as proposed by Ahn and Iverson (2004) is confirmed by the fMRI experiment results reported here. Overall, the results show that there is a significant difference in activation between the native- and the second-language speakers, consistent with the findings of Callan et al. (2004), and thus supportive of the predictions of the Internal Model, as opposed to the Same Area and Same Magnitude hypothesis (SAMH). Interestingly, activities unique to native Korean speakers relative to native English speakers were seen in the cuneus (occipital lobe) and the right middle frontal gyrus (Brodmann Area [BA] 10). As these areas have been reported to be involved with tone or pitch perception (e.g., Zatorre et al. 1992, Wong et al. 2004), the current findings uphold Silva's (2006) claim that, among younger speakers, the laryngeal contrasts of Korean are increasingly distinguished, not by VOT differences, but by their effect on pitch in the following vowel. In order to verify Silva's finding, a subsequent experiment was conducted to establish whether older speakers of Korean with clear VOT distinctions activate the cuneus and BA 10 (areas of the brain associated with pitch) in the same task. The results indicate that they do not, whereas speakers with overlapping VOT distinctions do show activations in these areas, thus corroborating Silva's claim in terms of the incipient tonogensis in Seoul Korean in a way that is consistent with Ahn and Iverson's hypothesis about the underlying representation of each laryngeal type.