This workshop deals with the cognitive and linguistic effects of the bilingual experience in a number of relatively unexplored domains, such as bilingualism with minority languages, late unbalanced bilingualism, and biliteracy.
Organized by Antonella Sorace and Yulia Rodina
Effects of bilingualism on cognitive performance:
child-adult differences and wider linguistic implications
Mariana Vega-Mendoza, Holly West, Antonella Sorace and Thomas H. Bak, University of Edinburgh
What is the effect of bilingualism on cognitive performance? Different research approaches to this question have taken into account factors such as age of onset of bilingualism, level of proficiency, and patterns of use of the bilinguals’ languages. We addressed this issue focusing on the performance of monolinguals and late non- balanced bilinguals and multilinguals in a non-‐linguistic task of attention.
Data from 51 participants were analyzed from three groups: Monolingual (n=18), Bilingual (n=16) and the Multilingual (n=17). All participants had English as their first language and were resident in Scotland. The participants in the bilingual group had started learning Spanish as a second language after the age of 18. The participants in the multilingual group had learned at least one additional language but not to the same level of proficiency as Spanish. All bilingual and multilingual participants provided self-‐ ratings of their level of proficiency on each of the languages they knew. The non-‐ linguistic, auditory measure of attention consisted of subtests 2, 3 and 5 of the Test of Everyday Attention (TEA) (Robertson et al., 1996), which tested sustained attention, inhibitory control and switching, respectively.
With regards to attentional measures, the only significant difference between monolinguals, bilinguals and multilinguals was found on the subtest 3 (inhibitory control) of the TEA, with bilinguals scoring significantly higher than monolinguals. Likewise, multilinguals scored significantly higher than monolinguals. No significant differences between groups were found for subtests 2 (sustained attention) and 5 (switching). These results are in line with Bak, Everington, Rose & Sorace (submitted), who found that adult bilinguals outperformed monolinguals only on subtest 3 of the TEA, whereas early bilinguals did so for both subtests 3 and 5.
These results reveal that adult bilingualism and multilingualism affect some, but not all, components of executive function: adult second language learning (regardless of number of languages learned) seems to provide an advantage in inhibitory control, but not in switching. This finding confirms recent studies (e.g. Hernandez et al 2013) that restrict the scope of the bilingual advantage to particular task-‐switching processes. There are also potentially important implications for our understanding of patterns of language use extensively attested in late bilinguals, such as the overuse of overt subject pronouns in null-subject languages, since anaphora resolution relies on both linguistic and pragmatic knowledge and on the ability to coordinate the two in real time, as well as to monitor and adapt to contextual changes (Sorace 2011).
Bilingualism, Biliteracy, and Cognition: the case of Gaelic-medium education in Scotland
Fiona O’Hanlon, The University of Edinburgh
This paper explores the attainment of Gaelic‐medium and English‐medium primary and early secondary school pupils using two sources: (i) a survey of Gaelic‐medium and English‐medium pupil attainment commissioned by the Scottish Government in 2006‐07 and (ii) a survey of attainment in Gaelic ‐ medium providing schools conducted in 2009‐10. The surveys collected data on teacher judgements of English‐medium and Gaelic-medium pupil attainment in English, Mathematics and Science and of Gaelic‐medium pupil attainment in Gaelic, assessed against the levels of the Scottish curricular assessment framework.
The present research investigated Gaelic‐medium pupils’ bilingual competencies, and Gaelic‐medium and English‐medium pupils’ comparative attainments in English, Mathematics and Science. For the latter comparisons, the Gaelic‐medium and English‐medium pupil groups were matched for variables that are known to affect attainment (gender, socio‐economic deprivation and local authority of their school), in order to increase the validity of the comparison between the two pupil groups.
The paper will focus on Gaelic-medium pupils’ linguistic attainments across the primary and early secondary school stages, and the comparison of these with English‐medium pupils’ linguistic attainments. The research found Gaelic‐ medium pupils to have literacy levels in Gaelic that were similar to the literacy levels of English‐medium pupils in English. In addition, there was evidence for a Gaelic‐medium pupil advantage in English literacy, as compared with their English‐medium counterparts. The evidence for this advantage was strong in relation to English reading and less strong in relation to English writing. The results will be situated in the context of previous research on the attainments of bilingual pupils in Scotland and internationally.
Literacy development in global Sápmi
Kirk Sullivan, Eva Lindgren, Asbjørg Westum,Hanna Outakoski
Umeå universitet, Institutionen för språkstudier
This study reaches over the national borders of Sweden, Norway and Finland and studies literacy in North Sami, English and the national majority language of each country. This on‐going research project focuses on the literacy development and the writing processes of children and young adults, who in their daily lives and at school encounter situations where three languages are present, and where language contact, code switching and language domains are part of the shared literacy landscape. Using computer keystroke logging that allows writing process reply in real time around 900 texts have been collected. This allows us to study the fluency in writing, writing and spelling difficulties, forming of ideas during the writing process, communication of ideas and opinions, and the finished text from many different linguistic and cognitive perspectives. Extensive questionnaires and interviews that have been gathered from the pupils, the parents, language teachers, principals and language officials shed light on the prerequisites that steer, support or constrain written language development and language education in different languages in Sápmi. The different research methodological and theoretical aspects of the study, the collected data and conducted case studies as well as the national language policies are being brought together and analyzed using Hornberger’s the Continua of Biliteracy framework. This paper focuses on the research methodological aspects of conducting a to date comparative study of the development of literacy among bilingual children and young in Sápmi.
Language and memory measures in Greek-German biliterate children
Ianthi Tsimpli and Christiane Bongartz
University of Reading/Aristotle University of Thessaloniki and University of Cologne
Early bilingualism including simultaneous bilinguals reflects to a large degree the home language situation particularly when bilingual children are tested during preschool ages. Literacy development in older children is equally, if not more, influential than the home language choice. We tested 70 8- 12 year old Greek-German bilingual children in Greece and Germany. The children are literate in both languages although their literacy is higher in German than in Greek especially those children living in Germany. The children were screened for verbal (lexical) abilities in order to establish level of proficiency in each language as well as dominance. Testing included verbal and non-verbal working memory tasks (Alloway 2007) as well as narrative retelling which was analysed in terms of length, syntactic complexity and use of referential forms. We particularly looked for possible correlations between working memory measures on one hand and narrative length as well as syntactic complexity on the other. Between‐group differences were identified with respect to the weaker language (Greek) in terms of lexical abilities and number of simple clauses produced. Syntactic complexity did not distinguish between dominant vs. weaker language in any of the bilingual groups. Children with more ‘balanced’ bilingualism in terms of lexical abilities in each language showed a significant correlation with non‐verbal working memory scores. We interpret this finding as indicative of a potential advantage of bilingualism in these older bilingual children when literacy levels are similar in the two languages.
Norwegian-Russian Bilinguals in North-Norway
Yulia Rodina and Marit Westergaard
Department of Language and linguistics/CASTL
University of Tromsø – The Arctic University of Norway
There is an increasing number of bilingual Norwegian-Russian children growing up in North Norway, due to considerable immigration from Russia over the last two decades. This language combination is so far virtually unstudied in the literature on bilingualism. In this paper we report on an experimental study, focusing on the acquisition of grammatical gender in Norwegian and Russian by two groups of bilingual children living in Norway, those growing up with one and those who have two Russian-speaking parents.
This is especially interesting since the two gender systems share certain properties but are also quite different in some ways. Both languages have a three-gender system (masculine, feminine and neuter), but the main difference between them is the transparency of gender assignment: While Russian gender is to a large extent predicable based on a few morphophonological rules, Norwegian gender is relatively opaque. We thus investigate whether the transparency of gender assignment in
Russian can have a facilitating effect on the acquisition of gender by bilingual children, who typically receive less input in this (minority) language than their monolingual peers.
Our results show that bilingual children with two Russian-speaking parents are different quantitatively and qualitatively from bilingual children with one Russianspeaking parent and monolinguals. The differences between the two groups of bilingual children suggest that the amount of parental input is crucial for the acquisition of gender in the minority language, especially with nouns that have ambiguous gender cues. These results also suggest that bilingual children with only one parent speaking the minority language are in a vulnerable situation linguistically and need the most language support in order to become true bilinguals.