The Influence of syllabic structure on computational processes : an electrophysiological and behavioural approach

In language learning two mechanisms are of critical importance, namely, rule learning and statistical learning. Additionally, an important linguistic unit, the syllable, has been proposed to be the unit of speech segmentation and speech production. The present dissertation explores the influence of syllabic structure on rule learning and statistical learning mechanisms. First, I explored the interference of syllabic structure changes over adjacent repetition- based rules and statistical word segmentation in a series of behavioural experiments. Then, I explored the ERP signatures of a syllabic structure change over adjacent and non-adjacent repetition- based rules. Overall, results show that the learnability of abstract adjacent or non-adjacent repetition-based rules and statistical learning are not interfered by a change in syllabic structure. Our results also show that the extraction of regularities over syllables was easier to perform than over vowels, attesting the pre-eminent role of the syllable in speech processing. The electrophysiological responses to syllabic structure changes were readily detected a few hundred milliseconds after the presentation of the stimulus, manifesting the automatic perceptual nature of its detection.