Margarita Kaushanskaya

Background: Bilingual children are frequently exposed to accented speech, yet it remains unclear how accent familiarity affects their ability to learn new words. This study examined Spanish–English bilingual children’s (n = 46) novel word learning from speakers with familiar and unfamiliar accents and investigated individual differences related to learning from accented input. Methods: Children completed an experimental word-learning task in which they learned novel word–object pairings produced by three speakers: a speaker of General American English, a Spanish-accented English speaker (familiar accent), and a Korean-accented English speaker (unfamiliar accent). Individual-differences analyses examined associations between learning outcomes and children’s language skills, length of bilingualism, and characteristics of language input in the home environment. Results: Children showed more difficulty learning novel words from the unfamiliar Korean-accented speaker than from the familiar Spanish-accented speaker and the General American-English speaker. Language skills were associated with learning from the familiar accent but not the unfamiliar accent. Length of bilingualism was positively associated with learning from the unfamiliar accent, whereas greater strength of foreign-accented English in the environment was negatively associated with learning from the native speaker. Conclusions: These findings suggest that accent familiarity facilitates bilingual children’s word learning and that experience-related factors contribute to their ability to accommodate accent variability.

Aims and objectives: Bilingual non-selective language activation is evident in priming experiments, where the use of a syntactic structure in production is more likely if the same structure was used in a previous sentence, even across languages. Here, we assess whether having to translate across two languages affects cross-linguistic priming. Because the need to translate and translation experience involve deeper and more automatized syntactic processing, we hypothesized that translating might increase the strength of cross-linguistic syntactic priming. Methodology: A total of 111 English–Spanish bilinguals were randomly assigned to complete a syntactic priming task in one of four conditions. Participants saw a prime either in active or passive voice in English, Spanish, Codeswitched, or Translation condition, and described a picture in Spanish, their second and non-dominant language. Data and analysis: Logistic mixed-effect models were constructed to analyze the extent to which condition and voice (active/passive) predicted the likelihood of priming. We additionally examined the moderating role of translation experience on both variables. Findings: Overall, results show a similar priming effect across conditions, and only a weak trend for translation experience to enhance priming. Implications: Together, results suggest that cross-linguistic priming is a robust phenomenon that is largely impermeable to translation processing and experience.

Purpose: Speech disfluencies are common in individuals who do not stutter, with estimates suggesting a typical rate of six per 100 words. Factors such as language ability, processing load, planning difficulty, and communication strategy influence disfluency. Recent work has indicated that bilinguals may produce more disfluencies than monolinguals, but the factors underlying disfluency in bilingual children are poorly understood. Method: We investigated the child, lexical, and syntactic factors associated with disfluencies in bilingual children who do not stutter during parent–child interactions. Forty-four Spanish–English bilingual parent–child dyads engaged in a play-based interaction. The children were 4–6 years old ( M = 62.3 months, SD = 6.96; 19 boys, range: 48.0–71.0). Children's language abilities ranged from clinically low (i.e., with a developmental language disorder) to typical. Results: Analyses revealed that children were more disfluent when they produced longer utterances. There was also a tendency for children with lower language skills to produce more disfluencies than children with higher language skills, when producing longer utterances. However, the mean lexical frequency of each utterance, the language of the utterance, and the child's language dominance were not associated with children's disfluency. Conclusions: These findings inform psycholinguistic models of fluent speech production and indicate that in bilingual children who do not stutter, disfluency is largely a reflection of utterance length. Children's overall language ability rather than language dominance was a more important contributor to disfluency in spontaneous speech. These findings have implications for assessment of bilingual children and for considerations of factors that may support or hinder disfluency in bilingual children who do and do not stutter.

Studies of learning indicate that presenting materials from multiple categories together (interleaving them) tends to yield better learning vs. presenting categories separately (blocking them). However, the impact of interleaving on language learning is mixed and has not been tested in the context of dual language learning. Here, we taught monolingual English-speaking adults vocabulary in German or Polish across two experiments, interleaving or blocking the languages. In Experiment 1, we observed a benefit of interleaving. Experiment 2 included a break midway through the task, and participants across conditions performed similarly. Our findings indicate that the interleaving advantage observed in other domains extends to dual language learning and may stem from a spacing effect. These findings have implications not only for theories of second language acquisition and dual language learning, but also for developing pedagogical approaches that optimize learning of multiple languages.

Purpose: In the current study, we examined the alignment of language choice of bilingual parent–child dyads in play-based interactions. Method: Forty-four bilingual Spanish–English parent–child dyads participated in a 10-min naturalistic free-play interaction to determine whether bilingual children and their parents respond to each other in the same language(s) across conversational turns and whether children's language ability and children's and parents' language dominance affect language alignment. Children's language ability was indexed by the Bilingual English–Spanish Assessment. Logistic regression was used to test the effects of children's language ability and children's and parents' language dominance on the alignment of language choice. Results: Results revealed that children and parents largely aligned their language choice and that children's and parents' language dominance, but not children's language ability, influenced alignment. Patterns of alignment differed between children and parents. Children aligned to their dominant language, and this was true for both English- and Spanish-dominant children. In contrast, English-dominant parents aligned equally to both languages, whereas Spanish-dominant parents aligned significantly more to Spanish. Conclusion: Together, these findings suggest that bilinguals' alignment of language choice is deeply sensitive to language dominance effects in both children and adults but that parents may also choose their language strategically in conversations with their children.

Children born preterm are at an elevated risk of language delays compared to children born full-term. However, there is a dearth of research investigating language outcomes in premature children exposed to more than one language. There is also limited empirical evidence linking the quantity and quality of parent input and language outcomes in premature children and the strength of these relationships in bilingual contexts remains unknown. Therefore, the current study examined language skills, parent input, conversational turns, and their associations at 16 months to 36 months in monolingual and bilingual children born preterm. Nine English-speaking monolingual parent–child dyads, and nine Spanish–English bilingual parent–child dyads participated in parent–child interactions that occurred over time in play-based contexts. Results revealed that preterm monolingual and bilingual children exhibited similar language abilities at all time points assessed. While both monolingual and bilingual dyads engaged in a comparable number of conversational turns at 16 m, monolingual mothers produced more words than bilingual mothers during play. Significant associations were observed between children’s vocabulary skills and their ability to combine words within and across most time points. Notably, relationships between parental input, conversational turns, and language skills were limited to a significant association between conversational turns at 16 m and vocabulary skills at 24 m. Together, findings indicate that bilingual children born preterm acquire language on the same timeline as monolingual children born preterm. While it is crucial that the current work be replicated in larger samples of children born preterm, the current work is the first to compare relationships between children’s language outcomes and the quantity and quality of parental input in monolingual and bilingual contexts.

Purpose: Traditional approaches to studying bilingual language development through bilingual–monolingual comparisons are deeply flawed. They are also insufficient as the evidence base for informing advice to bilingual parents regarding the optimal bilingual exposure strategy and for supporting the formulation of bilingual intervention approaches. The purpose of this review article is to provide an overview of empirical studies that have queried the question of how different types of dual-language input shape learning and language outcomes in bilingual children. Method: We rely on tightly controlled experimental studies of word learning in Spanish–English bilingual children, where we contrast children's learning in dual-language conditions (defined as distributed exposure and code-switched input) to a single-language condition in a within-subjects design. Results: Word-learning studies querying the role of distributed exposure indicated that distribution of exposures across Spanish and English reduced children's performance in comparison to English-only exposure. However, this effect was rooted in the abrupt switch from Spanish to English rather than distributed exposure itself. In contrast, an experiment designed to test the role of code-switched context on children's word learning revealed that code-switched context where switches resembled naturalistic code-switching behaviors enhanced learning in Spanish–English bilingual children. Notably, across different studies, children with weaker language skills (developmental language disorder) were no more affected by dual-language input than children with typical language skills. Conclusions: Together, experimental studies of word learning indicate that bilingual children can effectively learn from dual-language input but that different ways of combining languages in the input to bilingual children can have distinct consequences for learning. Ultimately, word-learning experiments, beyond answering critical questions regarding bilingual learning, can serve as an effective bridge between laboratory-based work and intervention studies whose goal it is to discover the optimal way of combining languages in the input to bilingual children with communication impairments. Presentation Video: https://doi.org/10.23641/asha.23929515