INVESTIGADORES
GARCIA adolfo Martin
congresos y reuniones científicas
Título:
Scalp and intracranial connectivity measures during bilingual word translation
Autor/es:
GARCÍA, A. M., MIKULAN, E., ADOLFI, F., SANTILLI, M., GONZALEZ VILAS, M., RIVERA-REI, Á., GARCÍA, M. DEL C., SILVA, W., VAUCHERET, E., CIRAOLO, C. E IBÁÑEZ, A
Lugar:
San Miguel de Tucumán
Reunión:
Congreso; XV Reunión Nacional y IV Encuentro Internacional de la Asociación Argentina de Ciencias del Comportamiento; 2015
Institución organizadora:
Asociación Argentina de Ciencias del Comportamiento
Resumen:
Introduction: Translation processes have been widely studied through behavioral paradigms. Yet, the neural basis of this bilingual skill remains largely unknown. Previous research indicates that these processes (i) predominantly engage the left hemisphere, (ii) always generate distinct activity in Broca?s area, and (iii) elicit wider activation patterns in forward translation (FT) than backward translation (BT). However, little is known about the neurocognitive mechanisms that are differentially engaged by BT and FT in the language-non-dominant hemisphere (LNDH). Objective: To address this issue, we examined word translation processes using scalp and intracranial electroencephalographic (EEG) recordings. Here we report preliminary multilevel evidence, focusing on the role of the LNDH in both tasks. Method: Ten professional translators and a proficient bilingual participated in a scalp and an intracranial EEG study, respectively. They all spoke Spanish as a native language (L1) and English as a non-native language (L2). They performed four noun processing tasks: L1 reading, L2 reading, BT, and FT. Within and across languages, stimuli were matched for frequency, length, concreteness, and cognate status. To assess functional connectivity in the first study, a weighted phase lag index was calculated for each pair of electrodes in five canonical EEG bands (0-4; 4-8; 8-13; 13-30; 30-40) and thresholded to retain only the 10% of the strongest connections. In the second study, intracranial EEG data were acquired over 105 channels from semi-rigid multilead electrodes with a diameter of 0.8 mm and five, ten, or fifteen 2-mm-wide contact leads placed 1.5 mm apart. Entropy-based estimations of the degree of joint nonrandom fluctuations between two signals were performed for each pair of electrodes in the beta band through the weighted symbolic mutual information measure. Results: In the scalp EEG study, subtraction of L2 reading from BT and L1 reading from FT revealed differential connectivity patterns for each translation direction. Notably, beta frequency patterns showed greater information exchange in LNDH temporo-occipital networks for BT, and in bilateral fronto-temporal networks for FT. Whereas the former areas subserve primary and otherwise ballistic processes, such as visual perception and lexical access, the latter have been implicated in bilingual executive control. Results from the intracranial EEG study replicated this broad pattern. Specifically, we observed marked differences in intra- and inter-lobe activity between BT and FT in the LNDH. Information sharing among anterior temporal, frontal, and prefrontal regions (including the medial frontal and the orbitofrontal cortices) was significantly more intense for FT than BT. Furthermore, in FT, significant connectivity patterns were mostly concentrated in those frontal areas. Conversely, in BT, they were more widely and less densely distributed throughout frontal, temporal, and parietal regions, notably converging in a posterior node located in the superior portion of the precuneus. Discussion: These results suggest that, with single-word units, BT depends distinctively on the integration of low-level information via automatic processes, while FT would be characterized by greater information exchange among areas supporting cognitive control mechanisms. This pattern aligns with previous reports of increased retrieval and selection efforts for FT than BT. Instead, LNDH involvement in BT would be characterized by a greater interplay of functions, most likely associated with perceptual, conceptual, and even pragmatic processes. More generally, this study illustrates the benefits of functional connectivity approaches. In particular, the above global brain interaction differences between translation directions may have been masked by the more isolated evoked responses of previous studies with highly proficient bilinguals. Taken together, our data sheds new light on the functional dynamics of translation processes. [Research partially supported by CONICET, CONICYT/FONDECYT Regular (1130920), COLCIENCIAS (1115-545-31374 and 1115-569-33858), FONCyT-PICT 2012-0412, FONCyT-PICT 2012-1309, and the INECO Foundation.]