Bi/Multi-lingualism

Sunday, October 02nd,   2011 [18:00 - 20:00]

PS_3.105 - The bilingual and monolingual differences in comprehension processing: An fMRI study

Román, P. E. ^{1, 2} , Rodríguez-Pujadas, A. ² , Ventura-Campos, N. ² , Sanjuán, A. ² , González, J. ² & Ávila, C. ²

1 Max Planck Institute. Leipzig, Germany
2 University Jaume I. Castellón de la Plana, Spain

Kovelman et al., (2008) have addressed the neural signature of bilingualism. They considered the left inferior frontal cortex (LIFC, Broca’s area) to be the best candidate. Wartenburger et al., (2003), in the same direction, found that early bilinguals showed higher activation in similar areas for L2 grammatical judgment rather than for semantic, but not late bilinguals. This can be interpreted in terms of the declarative/procedural model of language (Ullman, 2004) where grammatical rules are dependent on implicit knowledge sub served by Broca’s area and basal ganglia. Late L2 acquisition might not rely on the same structures as it would be acquired explicitly. We explore if grammatical and semantic processing in bilinguals differ in L1 too. An fMRI study was conducted where bilinguals and monolinguals performed a grammatical and semantic judgment task. We observed a more extensive activation of the bilingual brain in both judgments. More interestingly, higher activation of LIFC for L1 grammatical jugdments than semantic was found in bilinguals and specially in less proficient. These results are in agreement with those observed for L2 processing, suggesting that bilinguals recruit a more extensive network than monolinguals in L1 too. However, our data question the interpretation based on the declarative/procedimental model.

PS_3.107 - Grammatical gender effect in English!

Paolieri, D. ¹ , Morales, L. ¹ , Dussias, P. ² , Cubelli, R. ³ & Bajo, M. T. ¹

1 University of Granada, Spain
2 Pennsylvania State University, USA
3 University of Trento, Italy

Recently the importance of grammatical gender in monolingual (Cubelli et al., 2005) and bilingual (Paolieri et al., 2010) production has been observed. In this study we explored whether the grammatical gender of the native language affects the production of words in a second language where the grammatical gender system is absent. Twenty-four Spanish-English bilinguals were instructed to name pictures in English during a picture-word interference task, producing the bare noun. Words distractors were presented in Spanish and half of the
nouns were gender congruent with the Spanish translation of the target while the other half were gender incongruent. The results showed slower English naming times in the L1 gender-congruent pairs relative to the L1 gender-incongruent pairs. This interference effect confirms that grammatical gender selection is crucial in languages with a complex morphological structure, like Spanish. Moreover, it suggests that in L2 naming task the grammatical gender of L1 is always active and can affect lexical selection also in a language, like English, where the grammatical gender is absent.

PS_3.108 - Are eye-fixations in cognate processing dependent on entropy?

Mulder, K. ^{1, 2} , Dijkstra, T. ¹ & Schreuder, R. . ¹

1 Donders Institute for Brain, Cognition and Behaviour, Radboud University, Nijmegen, The Netherlands
2 International Max Planck Research School for Language Sciences, Nijmegen, The Netherlands

Decades of research have brought more insight into how bilinguals recognize cognates presented in isolation. However, little is known about what bilinguals actually look at when they process cognates. Cognates are translation equivalents with form overlap across languages, such as WINTER, which is spelled in the same way in Dutch and German (and English). Cognates provide form-ambiguous input in a word identification task, because they could belong to both the target and non-target language. In the case of non-identical cognates, such as GENERATIE (Dutch) and GENERATION (German, English), language-specific orthographic cues solve this ambiguity. We reasoned that if bilinguals do indeed use such cues during word identification, this might be reflected in their eye-fixations, which would be longer on the position at which the two cognates differ. In a Dutch (L2) lexical decision task, German-Dutch bilinguals were presented with German-Dutch identical and non-identical cognates, and with Dutch non-cognate control words while their eye-movements were monitored. The Dutch orthography of the non-identical cognates differed at either the beginning (e.g., CILINDER/ZILINDER) or end (MYSTERIE/MYSTERIUM) of the word from its German equivalent. As a separate test, pseudowords were included that differed at the beginning or end from existing cognates or Dutch non-cognate words.

PS_3.109 - On the effects of a brief L2 immersion on executive control

Baus, C. ^{1, 2} , Costa, A. ^{1, 3} & Carreiras, M. ^{2, 4, 5}

1 UNIVERSITAT POMPEU FABRA
2 BCBL
3 INSTITUCIO CATALANA DE RECERCA I ESTUDIS AVANÇATS (ICREA)
4 IKERBASQUE
5 UNIVERSIDAD DEL PAIS VASCO

There is ample evidence showing that bilingualism has an influence over the cognitive control abilities: bilinguals outperform monolinguals in different attentional tasks requiring conflict resolution. The link between bilingualism and cognitive control stems on the continuous engagement of the control mechanisms to solve cross-language competition during production. In the present experiment we explore the extent to which this bilingual advantage can be observed for low-proficient bilinguals that for a brief time of L2 immersion will use both their languages. The changes in the magnitude of the conflict effect across the immersion period were explored in two tasks (the Numerical-Stroop and the ANT) by comparing the beginning and the end of the immersion. As a control, a group of Spanish monolinguals was tested. The results revealed no differences between the groups in the ANT task due to the immersion experience. In contrast, the conflict effect in the Numerical-Stroop task reduced significantly for the immersed group but not for the monolingual group. These results suggest that the intensive practice in managing two languages in the L2 immersion context uniquely enhances those cognitive processes related to the inhibition of the more automatic response (L1) to successfully respond to the less automatic one (L2).

PS_3.110 - Lexicality effect and stress assignment in bilingual children reading Italian as a second language

Bellocchi, S. ¹ , Contento, S. . ¹ , Ceccarelli, I. ¹ & Burani, C. ²

1 Department of Psychology. University of Bologna. Bologna, Italy
2 Institute for Cognitive Sciences and Technologies (ISTC-CNR). Rome, Italy

Contrary to the claim that in transparent orthographies word reading is accomplished mainly by the nonlexical route, Italian developing readers show lexicality and frequency effects and seem to be sensitive to the distributional properties of the language. How do bilingual children with different age of L2 (Italian) first exposure and vocabulary size read L2 words and pseudowords? Two reading aloud experiments investigated lexicality effect and stress assignment in fourth and fifth grade bilinguals and monolinguals. Naming latencies and pronunciation accuracy were analyzed. In Experiment 1, lexicality effect (words read better than pseudowords) and differences between groups (bilinguals and monolinguals) emerged. In Experiment 2, word frequency effect emerged. Moreover, late bilinguals, who are also characterized by lower L2 vocabulary size, were less accurate than early bilinguals and monolinguals in assigning the less dominant stress. Similarly to monolinguals, lexical information seems to be employed in reading Italian as a second language. Furthermore, bilingual readers are sensitive to the distributional properties of the language. Stress assignment pattern seems to be affected by the characteristics of second language learners: results are discussed with respect to L2 vocabulary size and age of L2 first exposure.