Skip to main content

Brain functional and anatomical correlates of variability in the degree of success in the learning of the L2 phonemes

Final Report Summary - BILBRAIN (Brain functional and anatomical correlates of variability in the degree of success in the learning of the L2 phonemes)

Economic and cultural globalization has turned the learning of foreign languages into a socioeconomic need. For that reason, the learning of at least one foreign language is a compulsory subject in the academic curriculums of most countries in Europe and around the world. One of the greatest difficulties for the learners of second languages (L2) is to accurately produce and perceive the speech sounds (phonemes) of the new language and, despite educational efforts of foreign language programs, few individuals manage to achieve high proficiency levels in these skills. The present study aims to investigate the functional and anatomical brain correlates of variability in the degree of success in the learning of the L2 phonemes.
Our research question is addressed by investigating (i) functional, (ii) morphological, and (iii) functional connectivity differences between good and poor perceivers of the L2 by means of magnetic resonance imaging (MRI). Individual differences in the learning of the L2 sounds are investigated for all the levels of the phonological processing hierarchy (i.e. processes that cover auditory-sensory analysis, speech analysis, and attention) involved in producing and perceiving native phonemes.
To find the good and the poor L2 perceivers we collected data from more than 2,000 persons by means of an online questionnaire about language history and relevant factors. One-hundred twenty-one persons meet the inclusion criteria for the study: they were healthy Spanish (L1)-Catalan (L2) bilinguals who learned the L2 after the age of 4. The 121 bilinguals were tested in three behavioural tasks designed to evaluate their ability to perceive the Catalan /e/-/ε/ vowel contrast that is very difficult for Spanish native speakers to discriminate. Based on the performance of 20 native Catalan speakers, a cut-off point was calculated to establish “native performance”. Out of the initial sample of 121 bilinguals, 27 performed below natives’ performance in the three tasks and were considered poor perceivers and 16 performed as natives in all three tasks and were categorized as good perceivers.
Poor and good perceivers were asked to participate in an MRI study. Currently, 15 poor and 10 good perceivers agreed to participate in the MRI study. We are currently recruiting more participants to include 16 participants in each group. The MRI study consists of three sessions. In each MRI session different tasks are performed by the participants.
Study 1. Functional localizer of brain regions involved in phonological processing. With two auditory fMRI tasks we functionally map the brain areas that sustain the processes involved in phonological processing (Figure1). Participants are presented with sentences, sequences of syllables created by scrambling the syllables of the sentences, and the sentences spectrally rotated (that are perceived as noise). For all participants together, we localize the auditory cortices and the auditory thalamus when listening to all auditory stimuli. Related to the processing of speech and phonology (sentences – rotated sentences and syllables – rotated sentences), we localize Broca’s area and bilateral regions in the superior/middle temporal gyrus. Additionally, we functionally define areas related to working memory by asking participants to perform a 2-back and a 0-back task. We localize executive areas at the frontal pole, dorsolateral prefrontal cortex, lateral and medial posterior parietal cortex and the cingulate. All these cortical and subcortical regions will be the regions of interest for the analysis in studies 2 and 3.
Study 2. Are individual differences in the learning of the L2 phonology explained by differences in how the speech-processing system modulates general cognitive processes? It has been previously shown that the sensitivity to detect phonological changes is diminished in poor perceivers of the L2 sounds (Díaz et al., 2008), as assessed with the brain potential MMN. The differences between good and poor learners of the L2 phonology has been claimed to be caused by differences in the activity of frontal regions (based on the spatial location of the ERP differences in the scalp). The function behind the frontal component of the MMN has been claimed to be the re-allocation of attentional resources in the event of unexpected changes in the auditory scenario (Deouell, 2007; Shalgi & Deouell, 2007). The frontal contribution to the learning of speech sounds has also been shown in a training study that explored the brain functional differences between successful and unsuccessful learners of a non-native phoneme contrast (Golestani & Zatorre, 2004). Interestingly, white matter volume of the right insula correlates with the degree of mastery of the L2 sounds in bilinguals (Sebastián-Gallés et al., 2012). Study 2 aims to investigate the interactions between speech-specific areas and memory areas during speech changes, as compared to non-linguistic auditory changes. Participants are presented in the scanner with sequences of different syllables and with sequences of a repeated syllable. The comparison of the activity related to these two types of blocks will shed light on the differences between good and poor perceivers in detecting phonological changes. Participants are also presented with spectrally rotated syllables (perceived as non-linguistic sounds) in blocks, in half of the blocks the rotated syllables change and in the other half the same rotated syllable is presented. The comparison of these two blocks allows investigating whether differences between the groups are speech-specific. We expect that poor and good L2 perceivers will differ in the function/anatomy/functional connectivity of the frontal regions for speech stimuli only.
Study 3. Are individual differences in the learning of the L2 phonology explained by differences in how the speech-processing system modulates auditory sensory structures? Individual differences in the mastery of the native language are hardly reported because the processing of native speech is very robust. However, imaging studies on native listeners have showed that the activity of the auditory sensory thalamus, the medial geniculate body, (i) is more active when processing phonological changes, as compared to voice changes, on the same speech material, and, crucially, (ii) correlates with speech recognition and phonological measures (Díaz, Hintz, Kiebel, & von Kriegstein, 2012; vonKriegstein, Patterson, & Griffiths, 2008). The sensory thalamus is dynamically tuned by specific feedback from cortical areas to optimize demanding sensory processing (Díaz et al., 2012; Krupa, Ghazanfar, & Nicolelis, 1999; von Kriegstein et al., 2008). Study 3 aims to investigate the interactions between speech-specific cortical areas and the auditory thalamus during phonological processing, as compared to voice processing on the same speech stimulus. Participants are presented with sequences of syllables said by different voices. For half of the sequences participants have to push a button whenever a syllable is different than the previous one and in the other half of the sequences participants have to press a button whenever the voice is different than the previous one. We expect that poor and good L2 perceivers will differ in the function/anatomy/functional connectivity of the auditory thalamus (defined by the functional localizer, study 1) for phonological processing as compared to voice processing.
The results of this research are not only relevant because they will enhance our knowledge about how the brain processes and learns languages but also because they have the potential to help in the development of efficient language teaching programs.
Díaz, B, Baus, C, Escera, C, Costa, A, & Sebastián-Gallés, N (2008). Proc Natl Acad Sci USA, 105, 16083// Díaz, B, Hintz, F, Kiebel, S, & von Kriegstein, K (2012). Proc Natl Acad Sci USA, 109, 13841//Deouell, L (2007) J Psychophysiol, 21, 188//Golestani, N, & Zatorre, R (2004). NeuroImage, 21, 494//Krupa, D, Ghazanfar, A, & Nicolelis, M (1999). Proc Natl Acad Sci USA, 96, 8200 Sebastián-Gallés, N, et al. (2012). J Neurolinguist, 25, 150//Shalgi, S, & Deouell, L (2007). Neuropsychologia, 45, 1878//von Kriegstein, K, Patterson, R, & Griffiths, T (2008). Curr Biol, 18, 1855
Contact details: Martin, Eva (Head of Research Services), Tel.: +34935422078, info.recerca@upf.edu UNIVERSITAT POMPEU FABRA, SPAIN