۱- Introduction

۲- Methods

۳- Results

۴- Discussion

References

Abstract

The human brain has been uniquely equipped with the remarkable ability to acquire more than one language, as in bilingual individuals. Previous neuroimaging studies have indicated that learning a second language (L2) induced neuroplasticity at the macrostructural level. In this study, using the quantitative MRI (qMRI) combined with functional MRI (fMRI) techniques, we quantified the microstructural properties and tested whether second language learning modulates the microstructure in the bilingual brain. We found significant microstructural variations related to age of acquisition of second language in the left inferior frontal region and the left fusiform gyrus that are crucial for resolving lexical competition of bilinguals’ two languages. Early second language acquisition contributes to enhance cortical development at the microstructural level.

Introduction

One of the key characteristics of the bilingual brain is that when processing the target language, bilinguals need to successfully monitor and resolve lexical interference from the non-target language that competes for representation and selection (Crinion et al., 2006; Green, Crinion, & Price, 2006; Hernandez, Li, & MacWhinney, 2005; Kovelman, Baker, & Petitto, 2008; Price, Green, & von Studnitz, 1999; Tan et al., 2011; Thierry & Wu, 2007; Xu, Baldauf, Chang, Desimone, & Tan, 2017). This has been argued to lead to cognitive advantages on executive tasks due to bilingualism (Bialystok, Craik, & Luk, 2008; Bialystok, Craik, Klein, & Viswanathan, 2004; Birke Hansen et al., 2016; Colzato et al., 2008; Costa, Hernández, & Sebastián-Gallés, 2008; Gold, Kim, Johnson, Kryscio, & Smith, 2013; Perani et al., 2017; Prior & MacWhinney, 2010). Past neuroimaging studies have demonstrated that learning a second language (L2) induced neuroplasticity at the macrostructural level, as indexed by gray matter density (Grogan, Green, Ali, Crinion, & Price, 2009; Mechelli et al., 2004), white matter integrity (Elmer, Hänggi, Meyer, & Jäncke, 2011; Hamalainen, Sairanen, Leminen, & Lehtonen, 2017; Kuhl et al., 2016; Pliatsikas, Moschopoulou, & Saddy, 2015) and cortical thickness and volume (Klein, Mok, Chen, & Watkins, 2014; Li, Legault, & Litcofsky, 2014). Moreover, significant functional and structural imaging data points to the neural correlates of both L2 age of acquisition (AoA) and L2 proficiency. Early evidence suggests that childhood bilingualism may lead to distinct neural representations for L1 vs. L2, as compared with adulthood bilingualism (Kim, Relkin, Lee, & Hirsch, 1997). Later studies found out that proficiency, instead of AoA, may be the more important factor for determining the patterns of activation in L1 vs. L2 (Chee, Hon, Lee, & Soon, 2001). It is unclear, however, whether effects due to AoA and proficiency can be separated or isolated, as age and proficiency are often confounded or conflated (Kim et al., 1997; Hernandez, 2013). The neuroimaging measures used in previous studies, however, are qualitative because they are derived from uncalibrated T1-weighted images, which are sensitive to multiple features of tissue organization and microstructure (Mezer et al., 2013). To quantitatively evaluate microstructural properties in vivo, we employed the qMRI technique to compute the brain macromolecular tissue volume (MTV) and quantitative T1, which linearly contributes to iron and myelin concentrations (Stüber et al., 2014).