Comprehension and production of non-canonical word orders in Mandarin-speaking child heritage speakers

Abstract

Across languages, structures with non-canonical word order have been shown to be problematic for both child and adult heritage speakers. To investigate the linguistic and child-level factors that modulate heritage speakers’ difficulties with non-canonical word orders, we examined the comprehension and production of three Mandarin non-canonical structures in 5- to 9-year-old Mandarin-English heritage children and compared them to age-matched Mandarin-speaking monolingual children and adults. Specifically, we examined how linguistic properties, such as linear word order, presence or absence of morphosyntactic cues, and surface structural overlap between languages, as well as child-level factors, such as chronological age and current home language use affect the acquisition of non-canonical structures in heritage children and their monolingual peers. Results showed that although heritage children could use morphosyntactic cues, they did not show monolingual-level sensitivity to passive-related morphology. Additionally, children produced more canonical SVO word order, which is shared between English and Mandarin, and preferred the reverse interpretations of non-canonical structures in comprehension. These responses were taken as evidence for cross-linguistic influence from the majority to the minority language. Finally, although non-canonical structures caused difficulties for child heritage speakers, their performance was modulated by structure and improved with age, over and above heritage language use.

Acquisition of non-canonical word orders in Mandarin Chinese

“…Compared to studies that coded responses based on whether the primed structure was used (e.g., Huttenlocher et al, 2004 ) or directly categorized responses based on their sentence type (e.g., Hao and Chondrogianni, 2021 ), our coding and classification of errors could help us better analyze cases when children did not produce the primed sentence type. Particularly, incomplete sentences and sentences that did not correctly describe the target picture were not trivial cases and an analysis of these errors can provide insights into children’s struggles with the meaning vs. form of different sentence types.…”

“…In our production task, the functional representation was “half established” since the first noun phrase and the verb were exactly the same in paired prime and target sentences. In the future, harder priming tasks [e.g., using different lexical items in the target sentence as in Messenger et al (2012a) and Hao and Chondrogianni (2021) ] should be conducted among older children to better tap into their use of different sentence types. Particularly, performance of 5-year-olds and 6-year-olds is worth further investigation as Messenger et al (2012a) have found that 6-year-old English-speaking children produced reversed passives, but Hao and Chondrogianni (2021) detected adult-like performance in 5-to-9-year-old Mandarin-speaking children.…”

“…We conducted a picture selection task ( Leonard et al, 2013 ; Armon-Lotem et al, 2016 ; Zeng et al, 2016 , among many others) to evaluate Mandarin-speaking children’s comprehension of active sentences with the canonical SVO word order, active sentences with a non-canonical S- ba- O-V word order, and passive sentences with a non-canonical O- bei -S-V word order. We conducted a structural priming task ( Messenger et al, 2012a , b ; Hao and Chondrogianni, 2021 , among many others) to evaluate Mandarin-speaking children’s production of the three types of sentences.…”

Language processing in bilingual children

“…Bilingual and monolingual children's online comprehension has been compared to their production accuracy for various (morpho)syntactic properties (RQ1; e.g., Blom & Vasić, 2011;Chondrogianni & Marinis, 2012;Hao et al, 2023;Marinis & Saddy, 2013; . Outcomes are mixed.…”

Individual differences differentially influence language domains and learning mechanisms

“…This line of research can be further facilitated by the adoption of sophisticated experimental paradigms that can unravel the children's grammatical representations and their potential for learning. Along these lines, Hao (2022) showed that Mandarin-speaking heritage children (with English as ML) were more likely to repeat non-canonical structures when exposed to these structures through priming compared to their age-matched monolingual peers with DLD. If structural priming is taken as a proxy for accessing syntactic representations and implicit learning (Branigan & Pickering, 2017), this study offers first evidence that this ability differs in the two groups despite heritage children receiving less and potentially qualitative different input from their monolingual counterparts.…”