An fMRI Study of Bilingual Sentence Comprehension and Workload

Abstract

To examine the relation between the cortical substrates that support the comprehension of one's native language and those that support a second language, fMRI measures of cortical activation were taken as native Japanese participants, who had acquired moderate fluency in English, listened to auditory sentences in Japanese and English. In addition, to examine the impact of processing difficulty within a language, sentence difficulty was manipulated by including affirmative (easy) and negative (hard) sentences. The volume of activation was greater for English in most of the cortical regions, suggesting that more cognitive effort was required to process English. Also, a high percentage of the voxels that were activated for the Japanese condition were also activated for the English condition, with as much overlap between Japanese and English as between the processing of affirmative and negative sentences within Japanese. Negative sentences elicited greater activation than affirmative sentences primarily for English, indicating that the structural difficulty of negation has a larger impact on cortical activation if it occurs in the context of the second language, which may serve as another source of difficulty. These results suggest that a shared network of cortical regions supports the processing of both a first and a second language, such that the second language requires more computation and activity from the network.

Foreign language affects the contribution of intentions and outcomes to moral judgment

“…Neuroimaging studies suggest that foreign language comprehension requires more cognitive resources than native language comprehension (Hasegawa, Carpenter, & Just, 2002), perhaps because foreign language users "must devote more cognitive resources to lower level processes, such as word identification, semantic access, and syntactic processing" (Miller & Keenan, 2011, p. 874). Critically, studies suggest that cognitive depletion prompts a reduced consideration of intentions versus outcomes in moral evaluations (e.g., Buon, Jacop, Loissel, & Dupoux, 2012).…”

Mapping of verbal working memory in nonfluent Chinese–English bilinguals with functional MRI

“…It should be noted, however, that some of the empirical evidence for neural dissociation for late-acquisition bilinguals might have been confounded by language proficiency (Perani et al, 1998), time of exposure , or task difficulty (Chee et al, 2001). To disentangle these confounding factors (especially task difficulty), the effect of workload needs to be taken into consideration (Hasegawa et al, 2002), a point to which we will return in the next section.…”

“…On the one hand, research has shown that brain lesion causes equal deficit of working memory capacity in bilinguals' two languages, which suggests common or largely overlapping neural networks for WM processing in L1 and L2 (van Lieshout et al, 1990;Warren et al, 2000). In addition, several functional imaging studies of bilinguals have found common activation for the two languages in the abovementioned three components of the working memory system: the executive system (Hernandez et al, 2000(Hernandez et al, , 2001, the speech/rehearsal system (Hasegawa et al, 2002;Kim et al, 1997;Klein, 2003;Klein et al, 1995Klein et al, , 1999Rodriguez-Fornells et al, 2002), and the retention/storage system (Hasegawa et al, 2002).…”

“…As Hasegawa et al (2002) pointed out, workload is a major potential confounding variable for studies of nonfluent bilinguals. Many behavioral studies of bilinguals' working memory have consistently demonstrated that participants would experience a greater difficulty in L2 tasks than in L1 tasks Kemper, 1993, 1994;Cheung et al, 2000;Chincotta and Underwood, 1998;Hoosain, 1979;Stigler et al, 1986;Gathercole, 1999, 2001;Thorn et al, 2002).…”

“…In this case, a subtraction between the activation maps for L1 and L2 might be inadequate to discover the real pattern of contrast. An alternative method suggested by Hasegawa et al (2002) is to quantify the volume of activation for each task as well as the proportion of overlapping activation within each ROIs. This method can provide a quantitative description of the neural dissociation and integration between L1 and L2, as well as an examination of the effect of workload.…”