Angel Chan

Children's understanding of the agent-patient relations in the transitive construction: Cross-linguistic comparisons between Cantonese, German, and English

Abstract Cantonese-, German-, and English-speaking children aged 2;6, 3,6, and 4,6 acted out transitive sentences containing novel verbs in three conditions: (1) agent and patient were cued redundantly by both word order and animacy; (2) agent and patient were marked only with word order; and (3) agent and patient were cued in conflicting ways with word order and animacy. All three age groups in all three languages comprehended the redundantly cued sentences. When word order was the only cue, English children showed the earliest comprehension at 2;6, then German, and then Cantonese children at 3;6. When the cues conflicted, none of the 2;6 children in any language comprehended in adult-like ways, whereas all of the children at 3;6 and 4;6 preferred word order over animacy (but with some cross-linguistic differences in performance as well). When animacy contrast changed across sentence types, Cantonese children comprehended the sentences differently at all three age levels, German children did so at the two younger ages, and English children only at the youngest age. The findings correspond well with the informativeness of word order in the three languages, suggesting that childrens learning of the syntactic marking of agent-patient relations is strongly influenced by nature of the language they hear around them.

Cross-linguistic influence in simultaneous Cantonese–English bilingual children's comprehension of relative clauses

The current study investigated the role of cross-linguistic influence in Cantonese–English bilingual childrens comprehension of subject- and object-extracted relative clauses (RCs). Twenty simultaneous Cantonese–English bilingual children (Mage = 8;11, SD = 2;6) and 20 vocabulary-matched Cantonese monolingual children (Mage = 6;4, SD = 1;3) completed a test of Cantonese RC comprehension. The bilingual children also completed a test of English RC comprehension. The results showed that, whereas the monolingual children were equally competent on subject and object RCs, the bilingual children performed significantly better on subject RCs. Error analyses suggested that the bilingual children were most often correctly assigning thematic roles in object RCs, but were incorrectly choosing the RC subject as the head referent. This pervasive error was interpreted to be due to the fact that both Cantonese and English have canonical SVO word order, which creates competition with structures that compete with an object RC analysis.

The Cantonese double object construction with bei2 ‘give’ in bilingual children: The role of input

This article examines the role of input in bilingual children’s acquisition of the Cantonese double object construction with bei2 ‘give’ from a usage-based perspective (Tomasello, 2003; Lieven & Tomasello, 2008). In this domain of grammar, bilingual Cantonese—English children use the non-target [bei2-Recipient-Theme] order more frequently and for a more protracted duration than their monolingual counterparts (Yip & Matthews, 2007). We make use of corpus data to analyze the quantitative and qualitative aspects of the input that these bilingual children receive, and discuss three input factors to account for such vulnerability in bilingual children. First, the findings substantiate the idea that a bilingual child hearing fewer hours of Cantonese when compared to his or her monolingual age peers is likely to be exposed to fewer tokens of the target construction. Second, the findings indicate that the mapping between word order and thematic roles is not consistent in the Cantonese input, and as such is not conducive to learning for either bilingual or monolingual children. Third, simultaneous experience with the largely invariant [give-Recipient-Theme] forms from the English input overlaps with and therefore entrenches the developmental errors in Cantonese. These three input factors conspire to make this grammatical domain in Cantonese particularly vulnerable in bilingual children.

Exploring Mental Lexicon in an Efficient and Economic Way: Crowdsourcing Method for Linguistic Experiments

Mental lexicon plays a central role in human language competence and inspires the creation of new lexical resources. The traditional linguistic experiment method which is used to explore mental lexicon has some disadvantages. Crowdsourcing has become a promising method to conduct linguistic experiments which enables us to explore mental lexicon in an efficient and economic way. We focus on the feasibility and quality control issues of conducting Chinese linguistic experiments to collect Chinese word segmentation and semantic transparency data on the international crowdsourcing platforms Amazon Mechanical Turk and Crowdflower. Through this work, a framework for crowdsourcing linguistic experiments is proposed.

A cross-linguistic perspective to the study of dysarthria in Parkinson’s disease

Abstract Cross-linguistic studies aim at determining the similarities and differences in speech production by uncovering linguistic adaptations to specific constraints and environments. In the field of motor speech disorders, such a cross-language approach could be of great interest to understand not only the deficits of speech production that are induced by the pathology, but also the difficulties that are induced by the linguistic constraints specific to the patients’ language. From a more clinical point of view, cross-linguistic studies should specifically focus on the relationship between speech disorders and speech intelligibility. The aim of this opinion article is to identify the currently scarce theoretical and clinical avenues for cross-linguistic studies of dysarthria in Parkinson’s disease, and to establish guidelines that would lead future research in this direction. In turn, the practical and behavioral management of dysarthria in Parkinson’s disease has so far only focused on the ‘universal’ dimensions of speech production and feedback (e.g., treatment of loudness and dysprosody). Such approaches could benefit immensely from proper recommendations that would be more ‘language-driven’ and individually adapted to the patients’ language environment. An additional factor to consider for a better understanding and treatment of dysarthria in PD is the role of adjustment and cultural identity.

Building a Semantic Transparency Dataset of Chinese Nominal Compounds: A Practice of Crowdsourcing Methodology

This paper describes the work which aimed to create a semantic transparency dataset of Chinese nominal compounds (SemTransCNC 1.0) by crowdsourcing methodology. We firstly selected about 1,200 Chinese nominal compounds from a lexicon of modern Chinese and the Sinica Corpus. Then through a series of crowdsourcing experiments conducted on the Crowdflower platform, we successfully collected both overall semantic transparency and constituent semantic transparency data for each of them. According to our evaluation, the data quality is good. This work filled a gap in Chinese language resources and also practiced and explored the crowdsourcing methodology for linguistic experiment and language resource construction.

Create a Manual Chinese Word Segmentation Dataset Using Crowdsourcing Method

The manual Chinese word segmentation dataset WordSegCHC 1.0 which was built by eight crowdsourcing tasks conducted on the Crowdflower platform contains the manual word segmentation data of 152 Chinese sentences whose length ranges from 20 to 46 characters without punctuations. All the sentences received 200 segmentation responses in their corresponding crowdsourcing tasks and the numbers of valid response of them range from 123 to 143 (each sentence was segmented by more than 120 subjects). We also proposed an evaluation method called manual segmentation error rate (MSER) to evaluate the dataset; the MSER of the dataset is proved to be very low which indicates reliable data quality. In this work, we applied the crowdsourcing method to Chinese word segmentation task and the results confirmed again that the crowdsourcing method is a promising tool for linguistic data collection; the framework of crowdsourcing linguistic data collection used in this work can be reused in similar tasks; the resultant dataset filled a gap in Chinese language resources to the best of our knowledge, and it has potential applications in the research of word intuition of Chinese speakers and Chinese language processing.