Session 4C

It has been suggested that bilingualism has a positive effect on cognition by enhancing the inhibition and attentional mechanisms. These effects could benefit the ageing brain and delay the onset of symptoms of degenerative illness (i.e. cognitive-reserve hypothesis), such as dementia. This investigation aimed to study the bilingual effect on multiple sclerosis (MS), a known neurodegenerative disease which commonly impacts patients' cognitive, neuropsychological and physical abilities. Although variation among symptoms is usual in MS disease, common cognitive deficiencies are observed in executive control (EC) tasks such as attention, inhibition and shifting. Given that studies suggest that performance on EC tasks is enhanced in bilinguals, we aimed: to study whether being bilingual has an effect on the inhibitory control and attention abilities in MS patients. Following Costa et al.'s (2009) methodology, two groups of bilingual (matched by L2 proficiency) and monolingual young adults, one with MS and a second healthy control (matched by age and education) were tested. MS groups were balanced by cognitive impairment and controlled by MS characteristics. Participants performed two flankers with different attentional demands: one with high demands, and a second one with low demands. Participants' indexes on inhibition (measured by the difference in reaction times (RT) and accuracy in responding to congruent and incongruent trials in the flanker task) and attention (measured by: i. the overall accuracy and RTs in the high-attention demand and low-attention demand tasks; and ii. the difference in RTs between high-attention demand and low-attention demand tasks, collapsed across congruent and incongruent trials) were analysed. Results showed that both bilingual MS patients and bilingual healthy controls had similar inhibitory control and attention abilities. Monolingual MS patients and monolingual healthy controls showed similar inhibitory control. However, monolingual MS patients exhibited significantly worse attention abilities than the other groups. We propose that the similar behaviour between bilingual groups could indicate that bilingualism might counteract cognitive deficits related to MS, especially with respect to attention. On the other hand, results in monolingual MS patients seem related to underlying deficits in attention and possibly switching, executive control abilities commonly impaired in MS patients from an early stage. Our findings provide for the first time some preliminary evidence for the cognitive reserve hypothesis in bilingual MS patients.

Earlier studies have found clear associations between the amount of language exposure bilingual children receive in their two languages and their development in these languages. However, these studies are typically restricted to the home context (Hoff et al., 2012). In many countries, children attend daycare from a young age onward, and many of them are exposed to a different language at daycare than at home. Only few studies to date have examined the effects of bilingual or second language exposure at daycare, however, and, in the available studies, home exposure has typically not been taken into account (Bergström et al., 2014; Buyl & Housen, 2014). In this study, we investigate how exposure at home and at daycare relates to bilingual development, to see if exposure at daycare has an additional effect on children's language proficiency above and beyond the effect of exposure at home. Participants were 193 children aged two or three years who attended daycare centers in the Netherlands. Children formed a heterogeneous group, as they attended Dutch-only or Dutch-English daycare centers, and learned Dutch, English, or other languages (or combinations thereof) at home. All children were administered English and Dutch language tests at two waves, which were approximately nine months apart. Specifically, at each data wave, children completed Dutch and English tests assessing receptive vocabulary (PPVT, Dunn & Dunn, 2007), active vocabulary, and sentence comprehension (CELF-Preschool-2, Wiigg, Second, & Semel, 2006). Weighted estimates of relative language exposure to Dutch and English at home and at daycare were derived from a detailed parental questionnaire. A confirmatory factor analysis showed that children's scores on the language tasks represented two latent factors at each wave, representing Dutch and English proficiency. The results of a structural equation model showed that Dutch and English exposure at home positively predicted Dutch and English proficiency, respectively. Dutch exposure at home negatively predicted English proficiency at wave 2. Importantly, the results also showed that Dutch language exposure at daycare explained additional variance in Dutch proficiency, and English language exposure at daycare explained additional variance in English proficiency, at least at wave 1. Our results support earlier work showing within-language relationships between language exposure at home and young bilingual children's language development (Hoff et al., 2012). In addition, they show that language exposure at daycare is positively associated with children's language development, even after exposure at home is controlled. Given that our sample contained a considerable number of children who received exposure to their second language at daycare only, these results provide initial evidence that second language exposure at daycare supports language learning, at least at toddler age. Bergström, K., Klatte, M., Steinbrink, C., & Lachmann, T. (2016). First and second language acquisition in German children attending a kindergarten immersion program: A combined longitudinal and cross-sectional study. Language Learning, 66, 386-418. Buyl, A., & Housen, A. (2014). Factors, processes and outcomes of early immersion education in the Francophone community in Belgium. International Journal of Bilingual Education and Bilingualism, 17, 178-196. Dunn, L.M., & Dunn, D.M. (2007). Peabody Picture Vocabulary Test. Fourth Edition. Pearson. Hoff, E., Core, C., Place, S., Rumiche, R., Se?or, M., & Parra, M. (2012). Dual language exposure and early bilingual development. Journal of Child Language, 39, 1-27. Wigg, E.H., Secord, W. A., & Semel, E. (2012). CELF Preschool-2: Clinical Evaluation of Language Fundamentals Preschool. San Antonio: Pearson.

Do bilinguals have an advantage for learning additional languages? Previous work suggests that they do for lexical (Bartolotti & Marian, 2012) and phonological acquisition (Antoniou et al., 2015) but relatively little research has addressed whether bilinguals have an advantage for grammar learning (Cox, 2017). When addressing effects of bilingualism for acquisition, one can consider both the categorical distinction about being bilingual or not (Bialystok, 2009) as well as the degree of bilingualism (e.g., by considering proficiency and use). Previous research has tended to use categorical distinctions. For example, Grey at al. (2018) compared bilinguals and monolinguals and did not find differences in grammatical development on L2/L3 behavioral measures. However, results from such research might differ if bilingualism is examined on a continuous level. In the current study, we examine whether bilingualism contributes to the early stages of grammar learning using categorical and graded measures of bilingualism. Forty-one native English speakers (bilinguals, N=25; monolinguals, N=16) from a midwestern U.S. university completed incidental training on a semi-artificial language with English words and simple and complex German syntax (Rebuschat, 2008). L2 grammar learning was assessed with a grammaticality judgment task (GJT) that included source attributions (i.e., guess, intuition, memory, or rule). Additional measures of awareness during learning were collected via a debriefing questionnaire to classify learners as aware or unaware of the grammatical regularities. Finally, degree of bilingualism was examined via a detailed language history questionnaire. Preliminary analysis show that, overall, learners performed above chance on simple items with no difference found between groups. However, when degree of bilingualism was considered, results showed that the percentage of non-English use was negatively correlated with learning of the complex rule (r = - .337, p = .031). In regard to whether awareness contributed to the development of different types of grammatical knowledge, preliminary analysis showed that bilinguals performed above chance when attributing their responses to source attributions that are hypothesized to reflect implicit and explicit knowledge across all L2 grammatical rules only when reporting minimal or no awareness of the grammatical regularities that they had been exposed to during training. Monolinguals, on the other hand, showed no relationship between awareness and L2 learning. Thus, it appears that bilingualism may play a role in learning the grammar of an L2, but that the effect it plays may be driven by the degree and/or type of bilingual experience rather than whether a person is bilingual or not, which highlights the importance of viewing bilingualism as a continuous rather than as a dichotomous variable. In addition, our preliminary results suggest that the bilingual experience may confer certain benefits to learners for acquiring the grammar of an L2 incidentally, in order words, when learners are exposed to an L2 under more naturalistic learning conditions. References Antoniou, M., Liang, E., Ettlinger, M., & Wong, P. C. (2015). The bilingual advantage in phonetic learning. Bilingualism: Language and Cognition, 18(4), 683-695. Bartolotti, J., Marian, V., Schroeder, S. R., & Shook, A. (2011). Bilingualism and inhibitory control influence statistical learning of novel word forms. Frontiers in Psychology, 2, 324. Cox, J. G. (2017). Explicit instruction, bilingualism, and the older adult learner. Studies in Second Language Acquisition, 39(1), 29-58. Grey, S., Sanz, C., Morgan-Short, K., & Ullman, M. T. (2018). Bilingual and monolingual adults learning an additional language: ERPs reveal differences in syntactic processing. Bilingualism: Language and Cognition, 21(5), 970-994. doi:10.1017/S1366728917000426. Rebuschat, P., & Williams, J. (2009). Implicit learning of word order. In Proceedings of the Annual Meeting of the Cognitive Science Society (Vol. 31, No. 31).

The sources of crosslinguistic influence (CLI) in third language (L3) acquisition are hotly debated. At issue is whether the influence of the previously acquired languages can be only facilitative, and whether there can be influence from one or both languages (cf. Rothman et al. 2019, Westergaard 2019). Another interesting question is whether such influence into the L3 works similarly within different parts of the grammar: morphology, syntax, semantics. In this study, we look at CLI with an array of linguistic properties that are facilitative in one or the other of the previously acquired languages. We cross source of CLI with linguistic module (morphology, syntax, syntax-semantics). The trilingual group comprises Norwegian-Russian bilinguals learning L3 English. We also include two groups of Russian and Norwegian learners of L2 English of the same age (11-12-year-olds) and comparable proficiency in English. The prediction is that the L2-learner groups will be better at the constructions where their native language offers facilitation, compared to those without facilitation. For the L3-learner group, we expect cumulative influence from both of their languages (see Westergaard et al. 2017). To test this prediction, we created an acceptability judgment task with six grammatical/ungrammatical sentence pairs in each condition, presented written and aurally. The conditions were genericity, agreement and V2 (RUS=ENG?NOR), definiteness, copula and Pronoun-Verb word order (NOR=ENG?RUS). Below we present data from 68 L2-learners (34 Russian L1 and 34 Norwegian L1) as well as 17 Norwegian-Russian L3 learners, closely matched by lexical proficiency. More participants have been tested and the data analysis will be completed by the conference. As the mixed-model analysis revealed, trilinguals are significantly better at rejecting ungrammatical items in three conditions: definiteness (2L1>Rus, Nor>Rus), V2 (2L1>Now, Rus>Nor), and agreement (2L1>Rus, Nor>Rus) (Fig1). This indicates that trilinguals benefit from knowing a language that, through CLI, can improve performance in L3 properties. This suggests that both L1 and L2 are active and contribute to grammar building in L3 acquisition. The final results will be discussed in light of existing L3 models, e.g., the Typological Primacy Model (Rothman 2015), the Cumulative enhancement model (Flynn et al. 2010) and the Linguistic Proximity Model (Westergaard et al. 2017).