Can Lextale-Esp discriminate between groups of highly proficient Catalan–Spanish bilinguals with different language dominances?

Abstract

Researchers have recently introduced various LexTALE-type word recognition tests in order to assess vocabulary size in a second language (L2) mastered by participants. These tests correlate well with other measures of language proficiency in unbalanced bilinguals whose second language ability is well below the level of their native language. In the present study, we investigated whether LexTALE-type tests also discriminate at the high end of the proficiency range. In several regions of Spain, people speak both the regional language (e.g., Catalan or Basque) and Spanish to very high degrees. Still, because of their living circumstances, some consider themselves as either Spanish-dominant or regionallanguage dominant. We showed that these two groups perform differently on the recently published Spanish Lextale-Esp: The Spanish-dominant group had significantly higher scores than the Catalan-dominant group. We also showed that the noncognate words of the test have the highest discrimination power. This indicates that the existing Lextale-Esp can be used to estimate proficiency differences in highly proficient bilinguals with Spanish as an L2, and that a more sensitive test could be built by replacing the cognates.Keywords LEXTALE_Esp . Vocabulary size . Language proficiency . Bilingualism Proficiency has a central role in research on bilingualism (Lemhöfer & Broersma, 2012). Indeed, the second language (L2) proficiency has been demonstrated to affect performance in a variety of experimental tasks and paradigms. Davis et al. (2010), for instance, observed that highly proficient bilinguals show an interlingual cognate priming effect of the same magnitude as a within-language repetition effect (e.g., for an English-Spanish bilingual, there is as much priming for the prime-target pair rico-RICH as for the pair rich-RICH). The same was not true for beginning bilinguals, when the prime was in L2 and the target in the first language (L1). Rossi, Gugler, Friederici, and Hahne (2006) reported that highly proficient late L2 learners showed the same event-related potential responses to syntactic violations in sentences as L1 speakers, but this was not true for low-proficiency L2 learners, who had a qualitatively different response to the violations. Other neuroscientific research has indicated that the pattern of brain activation is modulated by L2 proficiency. Whereas highly proficient L2 speakers activate the same brain areas as L1 speakers during lexico-semantic processing in the L2, less proficient bilinguals show the engagement of additional brain areas when performing tasks in the L2 (see Abutalebi, 2008, and

The BEST Dataset of Language Proficiency

“…Similarly, the correlations between the different tests in each language showed that despite the underlying common aim, the indices provided are complementary and that the additional information provided by each of them is necessary. In line with previous research, some tests correlated quite highly (e.g., Ferré and Brysbaert, 2016), but it is worth noting that none of the correlations were close to ceiling. Furthermore, correlations between self-ratings and some of the objective tests were relatively low, suggesting that self-ratings alone are not an optimal reflection of proficiency (cf.…”

Multilingualism and trait emotional intelligence: an exploratory investigation

“…It has been judged to give a good indication of overall proficiency. Moreover, because results are not at ceiling level for first language users, it can be used for the assessment of highly proficient first and second language users (Ferré & Brysbaert, 2017). Lemhöfer and Broersma (2012)…”

Which words do English non-native speakers know? New supernational levels based on yes/no decision

“…A first limitation is that our lists are based on yes/no word decision. Although this test has been validated several times (Ferré & Brysbaert, 2017;Harrington, & Carey, 2009;Lemhöfer & Broersma, 2012;Meara, & Buxton, 1987;Zhang et al, 2019), it remains true that it does not measure word knowledge in great depth. Participants may know that a letter sequence is an English word, without much knowledge about the meaning of the word, let alone the various senses and unrelated meanings English words may have.…”