Linguistic Approaches to Bilingualism - Volume 9, Issue 2, 2019

Research has indicated that during sentence processing, French native speakers predominantly rely upon lexico-semantic cues (i.e., animacy) while native speakers of English rely upon syntactic cues (i.e., word order). The present study examined sentence production in L1 French/L2 English and L1 English/L2 French sequential bilinguals. Participants used animate and inanimate entities as sentence subjects while describing motion events represented by static pictures. To test a gradual change in animacy cue weighting in second-language sequential bilinguals with different proficiency levels were included. Sentence production of sequential bilinguals was compared against that of simultaneous bilinguals. The results indicated an overall preference for the use of animate subjects for both languages at all proficiency levels. The effect of animacy was stronger for English L2 than French L2 while it did not differ between languages in simultaneous bilinguals. Evidence for potential change in the animacy-cue weighting was only observed for English L2.

This study investigates effects of selective vulnerability and dominant language transfer in heritage grammar. Mandarin Chinese has a shì…de cleft construction, which, despite its superficial similarities with the it-cleft in English, is subject to additional conditions. Four experimental tasks elicited eighteen adult heritage speakers’ implicit knowledge of the word order and the temporal, telicity and discourse conditions associated with the Chinese cleft. The heritage speakers demonstrated target-like representation of the conditions. Meanwhile, their sensitivity to the telicity and discourse conditions is weaker than that of native speakers in Beijing, suggesting selective vulnerability in the heritage grammar. By comparing the heritage speakers with adult second language learners of Chinese, we concluded that the vulnerability of the heritage grammar in the discourse domain did not result from cross-linguistic influence from English. In different types of Chinese-English bilinguals, the dominant language affects the weaker language in different ways.

When listening to spoken language, bilinguals access words in both of their languages at the same time; this co-activation is often driven by phonological input mapping to candidates in multiple languages during online comprehension. Here, we examined whether cross-linguistic activation could occur covertly when the input does not overtly cue words in the non-target language. When asked in English to click an image of a duck, English-Spanish bilinguals looked more to an image of a shovel than to unrelated distractors, because the Spanish translations of the words duck and shovel (pato and pala, respectively) overlap phonologically in the non-target language. Our results suggest that bilinguals access their unused language, even in the absence of phonologically overlapping input. We conclude that during bilingual speech comprehension, words presented in a single language activate translation equivalents, with further spreading activation to unheard phonological competitors. These findings support highly interactive theories of language processing.

The aim of this experimental study is to examine the development of Greek gender in bilingual English-Greek and German-Greek children. Four gender production tasks were designed, two targeting gender assignment eliciting determiners and two targeting gender agreement eliciting predicate adjectives for real and novel nouns. Participant performance was assessed in relation to whether the ‘other’ language was a gender language or not (English vs. German) along with the role of the bilinguals’ Greek vocabulary knowledge and language input. The results are argued to contribute significantly to disentangling the role of crosslinguistic influence in gender assignment and agreement by bringing together a variety of input measures such as early and current amount of exposure to Greek, the role of area of residence (i.e. whether Greek is the minority or the majority language), the effect of maternal education and the amount of exposure to Greek in a school setting.

Previous studies on second language (L2) vocabulary acquisition with children showed that the use of a picture learning method favours the creation of direct links between the semantic system and new lexical representations at early stages of L2 acquisition (Comesaña et al., 2009). However, recent studies found that this influence seems to vary according to the cognate status of the words being learned (Comesaña et al., 2012), raising the question of how the type of words involved can modulate the lexical-semantic connections between the words of both languages in the bilingual memory. The main goal of the present study was to explore list composition effects in the establishment of L2 word-to-concept connections in Portuguese children by using a picture-based method. Results showed no influence of list composition in the establishment of L2 lexical-semantic connections when cognates have to be learned. Findings are discussed in light of relevant models of bilingual memory.

The experimental literature on the pragmatic abilities of bilinguals is rather sparse. The only study investigating adult second language (L2) learners (Slabakova, 2010) found an increase of pragmatic responses in that population relative to monolinguals. The results of studies on early bilingual children are unclear, some finding a significant increase in pragmatic responses in early bilingual children (preschoolers) relative to monolinguals (Siegal et al., 2007), while another (Antoniou and Katsos, 2017), testing school children, does not. We tested adult French L2 learners of English and Spanish (in their two languages) as well as French monolingual controls in Experiment 1 and Italian-Slovenian early bilingual children (in both languages) and Slovenian monolingual controls in Experiment 2. Our results were similar to those of Antoniou and Katsos (2017) in early bilingual children, but different from those of Siegal et al. (2007). We found no pragmatic bias in adult L2 leaners relative to adult monolinguals.