Localizing Language Cognition

Language is a quintessential, and arguably unique, human capacity. Unlike basic sensory processes, or even learning, memory, and decision-making, animal models do not exist.

Traditionally, brain mapping of language function relied on post-mortem studies of language deficits and brain lesions, or later on invasive intracranial recording and stimulation in patients. Modern, non-invasive brain imaging has made it possible to map language function in healthy individuals as well as patients, and to study the neural dynamics of natural language in action.

Using dense-array EEG (dEEG) and source localization, the EGI Science Team is applying this capability to 1) study and accelerate foreign language acquisition, and 2) develop a non-invasive protocol to map language function and hemispheric lateralization for presurgical planning in patients suffering from, for example, epilepsy or tumors.

Foreign language training

With funding support from the Office of Naval Research, we implemented a neurocognitive-based approach to foreign language training called Automatization in Communicative Contexts of Essential Speech Segments (ACCESS; Gatbonton and Segalowitz, 1988, 2005). The ACCESS approach provides intensive practice of target language phrases within authentic, communicative interactions in order to develop language fluency, flexibility, and rich embedded memory traces for transfer to real-life situations. Our source-localized dEEG results provided the first direct neural evidence of brain engagement early in training designed to exercise natural language processing networks of real communication.

EEG patterns during semantic processing

Figure 1. Grand-average source-localized EEG brain activation patterns during semantic processing (~435 ms after word onset). Brain signature associated with access to word meaning (highlighted red box) is observed after two days of training with a traditional method, but just a single day of training with ACCESS.

Learners trained with ACCESS demonstrated a brain signature (Figure 1) associated with processing the meanings of words (i.e., engagement of posterior inferior-temporal cortex) after a single training session, in contrast to learners trained with a more traditional approach who required two days of training to show this same signature (Poulsen et al., 2011a, 2011b). Interestingly, a second brain signature (i.e., engagement of temporal pole) associated with integrating words within context was demonstrated only by the ACCESS learners, consistent with the rich, contextualized practice provided early in ACCESS training.

Language mapping

In assessing neurosurgical candidates for resection of epileptogenic brain tissue or tumor, regions associated with essential brain functions, including language, must be identified and preserved. Safer, non-invasive alternatives to the risky Wada test for determining language lateralization and intracranial EEG for regional language mapping are highly desirable.

The EGI Science Team, as part of its focus on epilepsy diagnosis and treatment, is developing a dEEG protocol for language mapping. Scalp-recorded dEEG during a verb generation task (VGT) (Petersen et al., 1988; 1989), in which individuals generate a verb associated with a concrete noun presented auditorily or pictorially, was source localized to identify the brain regions engaged during language reception and production.

Map of ERP response and source localization of language processing

Figure 2. Left: Topographical map of grand average event-related potential at the centroparietal peak of the late positive complex (LPC) for the auditory version of the verb generation task. Right: Inverse source localization shows distinct left lateralized activity in network of regions associated with language processing.

Preliminary results, based on the averaged event related potential (ERP) to stimulus onset (Figure 2), show reliable lateralization assessment and localization of activity to brain regions generally consistent with invasive, intracranial EEG studies (Edwards et al., 2010), and other more costly and less portable non-invasive modalities, including functional magnetic resonance imaging (fMRI; Rowan et al., 2004), and magnetoencephalography (MEG; Doesburg et al., 2012; Pang et al., 2011). Upcoming analyses will examine refined source localization using individual head models and tractography constraints, frequency-domain source localization, oscillatory network dynamics across identified language regions of interest, and development of metrics for individual-level reliability and uncertainty assessment, which is essential for patient evaluation. We have also begun collaborations with epilepsy centers in Seattle and Shanghai who will implement the dEEG VGT protocol with patients, and cross-validate the results with fMRI and icEEG in the same patients.

References

Doesburg, S. M., Vinette, S. A., Cheung, M. J., and Pang, E. W. (2012). Theta-modulated gamma-band synchronization among activated regions during a verb generation task. Front Psychol, 3, 195.

Gatbonton, E., and Segalowitz, N. (2005). Rethinking communicative language teaching: a focus on access to fluency. Canadian Modern Language Review, 61, 325-353.

Gatbonton, E., and Segalowitz, N. (1988). Creative Automatization: Principles for Promoting Fluency within a Communicative Framework. TESOL Quarterly, 22(3), 473-492.

Pang, E. W., Wang, F., Malone, M., Kadis, D. S., and Donner, E. J. (2011). Localization of Broca's area using verb generation tasks in the MEG: validation against fMRI. Neurosci Lett, 490(3), 215-219.

Petersen, S. E., Fox, P. T., Posner, M. I., Mintun, M., and Raichle, M. E. (1988). Positron emission tomographic studies of the cortical anatomy of single-word processing. Nature, 331(6157), 585-589.

Petersen, S. E., Fox, P. T., Posner, M. I., Mintun, M., and Raichle, M. E. (1989). Positron emission tomographic studies of the processing of singe words. J Cogn Neurosci, 1(2), 153-170.

Poulsen, C., Luu, P., Davey, C., Tucker, D., and Nelson, J. (2011a). From Sound to Meaning: Changes in EEG Source-Localized Brain Activity with Foreign-Language Training. In D.D. Schmorrow and C.M. Fidopiastis (Eds.): FAC 2011, HCII 2011, LNAI 6780, pp. 203–211. Springer-Verlag: Berlin.

Poulsen, C., Luu, P., Davey, C., Tucker, D., and Nelson, J. (2011b). From Sound to Meaning: Changes in EEG Source-Localized Brain Activity with Foreign-Language Training. Poster presented at the Eighteenth Annual Meeting of the Cognitive Neuroscience Society, New York, NY.

Rowan, A., Liegeois, F., Vargha-Khadem, F., Gadian, D., Connelly, A., and Baldeweg, T. (2004). Cortical lateralization during verb generation: a combined ERP and fMRI study. Neuroimage, 22(2), 665-675.