Kassey Chang

Kassey Chang

Language Technology & Computational Linguistics

Projects

Using Neighborhood-based Clustering to Categorize Visual Mental Imagery and Perceptual Domains

Do both visual mental imagery (VMI) and perception exhibit domain- specific clustering? If yes, how similarly or differently? Using an English adaptation of the French Enhanced Imagery and Perception battery (eBIP), we aim to uncover the domain-specific nature of VMI and perception. By comparing performance scores, we seek to provide insights into their potential shared and distinct cognitive processes through dimension reduction and cluster analysis.

A Comparative Study of Sentence-Final Particles Acquisition in Monolingual, Bilingual, and Trilingual Cantonese-Speaking Children: A Corpus-Driven Approach

Sentence-final particles (SFPs) play a crucial role in expressing speakers’ attitudes and intentions in Cantonese, making them integral to the language’s pragmatics. Despite their importance, the acquisition of SFPs by Cantonese-speaking children has not been extensively studied, particularly in multilingual contexts. This study compares SFP frequencies and usage patterns across monolingual, bilingual (Cantonese-English), and trilingual (Cantonese-English-Mandarin) children to identify early versus late-acquired SFPs and determine influential factors such as language exposure, dominance, and cross-linguistic effects. The findings reveal that assertive particles are among the first acquired SFPs across all language backgrounds. However, bilingual children do not acquire SFPs as well as their trilingual counterparts, showing a slower acquisition rate and the non-acquisition of certain particles. This difference can be attributed to the lack of direct SFP cognates between Cantonese and English and the limited cross-linguistic facilitation from English. In contrast, the presence of SFPs in Mandarin, despite the lack of direct cognates, appears to provide some facilitative effects for trilingual children acquiring Cantonese. These results demonstrate the importance of considering cross-language influence and typological distance when examining SFP acquisition in multilingual contexts. The absence of direct SFP cognates between Cantonese and English, as well as the structural differences between the two languages, may hinder the acquisition of certain particles in bilingual children. The outcomes of this research enhance our understanding of SFP acquisition in Cantonese and contribute to the broader knowledge of language development in multilingual settings. By providing quantitative insights into the developmental progression of SFP acquisition across different language backgrounds and examining the role of language distance in shaping acquisition patterns, this study sets a foundation for future research on the acquisition of pragmatic markers in multilingual contexts and informs pedagogical strategies to support Cantonese language learners.

The Effect of Spatial Devices in L1 on the Construal of Spatial Relationships: A Corpus-based Study

Cross-linguistic patterns in the human conceptualization of space have been one of the cognitive-linguistic topics frequently discussed by scholars. Despite the presence of universal semantic primes, their syntactic classes and encoding spatial scenes vary from one another (Talmy, 2000). As a primary cognitive and perceptual category, space is regularly encoded by speakers in daily communication. A successful acquisition of spatial relations in the target language (TL) could determine whether the learners have achieved their language goals (Becker & Carroll, 1997). Although various factors could influence the acquisition, the interaction between speakers’ prioritized spatial knowledge, possibly shaped by their native languages (L1), and the TL is crucial in investigating the construal of spatial relationships.

Scale-free and Small-world Networks in Word Associations

The notable theoretical linguist Noam Chomsky (1957) described language as a set of infinite combinations constructed from a finite set of linguistic units. However, these combinations are not formed randomly or unsystematically due to the lexico-syntactical rules and collocations in real-life language use. Along with the rise of quantitative methodology in linguistics, more statistical evidence has proven the mutual information in the co-occurrence of words. This report aims to present a review of the article The small world of human language (Canchon & Solé, 2001) and how the findings in physics interact with the disciplinary knowledge in linguistics.

Focus System in Yorùbá: The Particle ni and its Syntax-Semantics Interface

The Yoruba language (Yorùbá), belonging to the Benue-Congo subfamily in the Niger-Congo family, is spoken by over 20 million people, primarily in Nigeria, Benin, and Togo. With its rich morpho-syntactic and phonological structures, Yoruba offers a vast array of linguistic phenomena for investigation, including serial verb construction, noun classes, and the tonal system. This paper aims to explore the focus construction in Yoruba, specifically focusing on the role of the focus particle ni in its syntax and semantics. To achieve this goal, I will delve into several key questions, including:

Vowel Adaptation in English Loanwords Borrowed into Cantonese: A Data-driven Approach

Throughout history, English loanwords have constituted a substantial component of the Cantonese lexicon, with the earliest recorded instances dating to 1828. A majority of these loanwords exhibit phonetically adapted syllables; for instance, /bʌs/ “bus” transforms into /pa1 si2/ in Cantonese. Certain loanwords meld a single phonetically adapted syllable with native Cantonese morphemes, such as /bɪr/ “beer” evolving into /pɛ1 tsɐu2/ (with tsɐu2 signifying “wine”). Predominantly, loanwords function as nouns spanning a range of semantic categories, including food and beverages, leisure, and the workplace (Wong, 2012). However, the central question remains: is the pattern of vowel adaptation in English loanwords entering Cantonese predictable?

Correctness and Fluency across Speech Styles: A Study on the Speech Errors made by Cantonese Speakers

In the previous decades, speech production and processing have been well-established linguistic topics. Much has been done to map the evidence elicited in speech errors and the mental processing during discourse production. Along with the intensive growth of corpus linguistics, the slips in raw spoken data have been a gateway to studying language performance and actual language use. Speech errors among children and patients with language disorders have played a significant role in investigating early language acquisition and clinical linguistics.

The Semantics of the Cantonese Sentence-Final Particle lo1

Sentence-final particles (SFPs), or utterance particles, are essential in colloquial conversation usage in many Asian languages, namely Chinese, Japanese, Thai and Vietnamese. They are necessary for self-expression, indicating speakers’ attitudes towards addressees or situations, assumptions, intentions, or emotions (Wierzbicka, 2003; Leung, 2016). As one of the languages with the most significant number of SFPs, Cantonese SFPs have been richly investigated, especially from a formal syntactic perspective. Yet, scholars have diverse views on them; some consider they have ‘no semantic content’ (Luke, 1990) or ‘subjective’ and ‘illogical’ (Wierzbicka, 1986). Another problem arose when most studies had not considered the disparity between the meaning of the SFP and the attaching utterance. The present study intends to carry out a semantic analysis of a frequently used Cantonese SFPs, lo1 (high-level tone), using the Natural Semantic Metalanguage (NSM) approach.

Endangered Language in Macao: Review on Patuá Macaense

Language users might tend to take their languages for granted, especially their native languages. Not being aware of the linguistic phenomena could lead to adverse consequences to the language’s survival, i.e. language loss/language death, particularly those minority languages with fewer and fewer speakers left. Thus, the United Nations Educational, Scientific and Cultural Organization (UNESCO) has been attempting to examine the ethnolinguistic vitality of world languages and has identified some endangered languages (Moseley, 2010). Little people in Hong Kong know about the existence of a critically endangered language right next to them - the Macanese Patois (Portuguese: Patuá Macaense; Cantonese: 澳門土生葡語) spoken in Macao.