Investigating Statistical Tone Learning by Analyzing Mismatch Negativity in ERP Results

Chen, Jing (2022) Investigating Statistical Tone Learning by Analyzing Mismatch Negativity in ERP Results. Bachelor's Thesis, Artificial Intelligence.

Preview	Text Bachelor_Thesis_Jing.pdf Download (506kB) | Preview
	Text toestemming.pdf Restricted to Registered users only Download (121kB)

Abstract

Statistical learning enables individuals to include regularity in their learning environment, as evidenced from the presence or absence indicated by the Mismatch Negativity (MMN) in Event Related Potential (ERP) data. In the current study, we used statistical learning to analyze 6 Mandarin tone-syllable combinations presented in two multi-featured oddball paradigms to determine the likelihood that non-tonal language speakers were able to learn tones and distinguish between tonal differences. Subsequently, a behavioral test was conducted, wherein the research findings indicated a lack of significant differences in the ERP results between early tone and syllabic learning. The participants were essentially incapable of distinguishing between tone differences during the early phase. Conversely, during the latter phase, we detected the presence of MMN in both syllabic and tone learning. In the context of late learning, participants relied on statistical learning to identify the tone differences. Moreover, the degree of accuracy in perceiving tonal and syllabic differences was above the chance level in the behavioral test. Accordingly, we concluded that non-tonal language speakers were capable of learning tones and distinguishing between tone differences.

Item Type:	Thesis (Bachelor's Thesis)
Supervisor name:	Spenader, J.K. and Tang, M.
Degree programme:	Artificial Intelligence
Thesis type:	Bachelor's Thesis
Language:	English
Date Deposited:	15 Jul 2022 12:18
Last Modified:	15 Jul 2022 12:18
URI:	https://fse.studenttheses.ub.rug.nl/id/eprint/27922

Actions (login required)

View Item