- Home
- e-Journals
- Journal of Second Language Studies
- Previous Issues
- Volume 1, Issue, 2018
Journal of Second Language Studies - Volume 1, Issue 1, 2018
Volume 1, Issue 1, 2018
-
Grammatical gender processing in L2 speakers of Spanish
Author(s): Lauren Halberstadt, Jorge R. Valdés Kroff and Paola E. Dussiaspp.: 5–30 (26)More LessRecent findings indicate that native speakers (L1) use grammatical gender marking on articles to facilitate the processing of upcoming nouns (e.g., Lew-Williams & Fernald, 2007 ; Dussias, Valdés Kroff, Guzzardo Tamargo, & Gerfen, 2013 ). Conversely, adult second language (L2) learners for whom grammatical gender is absent in their first language appear to need near-native proficiency to behave like native speakers ( Dussias et al., 2013 ; Hopp, 2013 ). The question addressed here is whether sensitivity to grammatical gender in L2 learners of Spanish is modulated by the cognate status of nouns due to their heightened parallel orthographic, phonological, morpho-syntactic and semantic activation. Additionally, the role of transparent and non-transparent word-final gender marking cues was examined because past studies have shown that native speakers of Spanish are sensitive to differences in gender transparency ( Caffarra, Janssen, & Barber, 2014 ). Participants were English learners of Spanish and Spanish monolingual speakers. Data were collected using the visual world paradigm. Participants saw 2-picture visual scenes in which objects either matched in gender (same-gender trials) or mismatched (different-gender trials). Targets were embedded in the preamble Encuentra el/la ___ ‘Find the ___’. The monolingual group displayed an anticipatory effect on different gender trials, replicating past studies that show that native speakers use grammatical gender information encoded in prenominal modifiers predictively. The learners were able to use gender information on the articles to facilitate processing, but only when the nouns had gender endings that were transparent. Cognate status did not confer an advantage during grammatical gender processing.
-
Suprasegmental information cues morphological anticipation during L1/L2 lexical access
Author(s): Nuria Sagarra and Joseph V. Casillaspp.: 31–59 (29)More LessWe use visual-world eye-tracking and gating methods to investigate whether Spanish monolinguals and English late learners of Spanish use prosodic cues (lexical stress) to anticipate morphological information (suffixes) during spoken word recognition, and if they do, whether L2 proficiency and working memory (WM) mediate their anticipatory abilities. Our findings show that the monolinguals used prosodic information to predict word endings in both tasks, regardless of first-syllable stress (stressed, unstressed) and structure (CV, CVC). In contrast, the beginning learners did not use prosodic information to anticipate word suffixes in any task or condition. Importantly, the advanced learners mirrored the monolinguals, except in words with first-syllable CV structure, but were slower than the monolinguals. Finally, WM was not associated with anticipatory eye movements, though results were inconclusive for offline processing. Taken together, the present study shows that suprasegmental information facilitates morphological anticipation during spoken word recognition, and that adult learners can gain anticipatory processing patterns qualitatively, but not quantitatively, similar to monolinguals.
-
Possibility of semantic involvement in the L1-L2 congruency effect in the processing of L2 collocations
Author(s): Junko Yamashitapp.: 60–78 (19)More LessA growing body of second language (L2) research has identified the congruency effect in the processing of L2 collocations (L2 collocations that have word-for-word translation equivalents in learners’ first language [L1] are processed more quickly and accurately than those that have no such forms in L1). However, the locus of this effect has yet to be fully understood. This study explored the possibility of semantic involvement in this phenomenon by categorizing congruent and incongruent items used in the past studies according to the semantic transparency of collocations. Results showed a clear dominance of transparent items in the congruent category and that of opaque items in the incongruent category, suggesting the possibility of semantic involvement in the congruency effect. However, considerations of other factors observed in the pertinent studies led to the conclusion that there is something more in the congruency effect beyond semantic transparency.
-
Effects of second language proficiency and working memory span on novel language learning
Author(s): Kaitlyn L. Zavaleta and Janet L. Nicolpp.: 79–105 (27)More LessIn this paper we examine the effects of prior language learning experience and working memory capacity on learning a novel language. Participants with a range of proficiency in a second language were trained on a third language. They were presented with 20 Turkish words (to test word-learning) and their plural counterparts (to test rule-learning). After training, participants were asked to judge whether a given label correctly matched the pictured object (comprehension) and to verbally produce the correct label for a given object (production). The comprehension test took place immediately after training and again a week later. The production test was administered after the second comprehension test. We found that performance on the initial comprehension test was predicted by working memory span but not L2 proficiency; L2 proficiency predicted learning performance only for the delayed tests. This suggests that the two variables – L2 Proficiency and Working Memory – play different roles in L3 learning.
-
Early and late learners decompose inflected nouns, but can they tell which ones are inflected correctly?
Author(s): Kira Gor, Anna Chrabaszcz and Svetlana Cookpp.: 106–140 (35)More LessAn auditory lexical decision task tests morphological decomposition and sensitivity to violations in inflection in late second language learners, early learners (heritage speakers), and native speakers of Russian. Two datasets compared reaction times and error rates to real Russian inflected nouns and nonce nouns. Two parameters of real nouns were manipulated: case (the nominative, or the oblique case), and inflection (overt or zero). Nonce nouns had (a) real stems and inflections combined in an illegal way (lemoning), and (b) inflected nonce stems (lemosing). Results suggest that heritage and late learners process inflectional morphology; however, their processing of inflected words is unreliable: they are willing to accept words with incongruent inflections. While no major differences were found in the processing patterns of early and late learners, a developmental trajectory was observed in both groups of learners: their sensitivity to violations in inflection improved with proficiency.
-
The effects of transliterations, thematic organization, and working memory on adult L2 vocabulary learning
Author(s): Alison M. Tseng, Martin C. Doppelt and Natasha Tokowiczpp.: 141–165 (25)More LessWe examine how L2 vocabulary learning is affected by the information provided to the learner during training, organization of the to-be-learned vocabulary, and working memory capacity of the learner. Native English speakers were taught Arabic vocabulary in seven sessions, during which they heard L1 (English)-L2 (Arabic) translation pairs. Training was manipulated between participants by crossing the presence vs. absence of a transliteration and thematic vs. random organization of vocabulary. Session, working memory capacity, transliteration condition, and organization condition interacted in English-Arabic translation accuracy. Participants with lower working memory capacity performed best in the transliteration-thematic organization condition, whereas participants with higher working memory performed best in the transliteration-random organization and no transliteration-thematic organization conditions. Translation RT and free recall were not related to working memory, and were best in the transliteration conditions. Results suggest that adult L2 vocabulary learning is aided by exploiting well-established spelling-sound connections to improve L2 lexical representations.
-
The use of lexical and sublexical cues in speech segmentation by second language learners of English
Author(s): Candise Yue Lin and Min Wangpp.: 166–198 (33)More LessThis study examined the use of lexical and sublexical cues in speech segmentation by Mandarin L2 learners of English, focusing on two types of lexical cue, lexical knowledge and semantic relatedness, and three coda (sublexical) cues, /n, s, ŋ/ due to their varying phonotactic probabilities in Mandarin and English. Thirty-five native English speakers and 30 L2 learners participated in two experiments. Experiment 1 showed that learners were able to use lexicality as a cue to segment L2 speech. The lexicality effect significantly interacted with L2 proficiency. Experiment 2 showed that learners did not use semantic cues to the same extent as native listeners did. All participants experienced more difficulty with word boundary identification preceded by /s/. This difficulty may stem from weak allophonic cues of /s/ in English. L2 learners with better proficiency may be better at recognizing familiar words from continuous speech, thus more efficiently utilizing the lexicality cue.
-
Associative networks from L2 words in early and late Vietnamese-English bilinguals
Author(s): Juliet Huynh and Naoko Witzelpp.: 199–230 (32)More LessThis study investigates whether highly-proficient Vietnamese-English bilinguals can activate associations in a native-like way from studying L2 words, whether these L2 associations interact with L1 words, and whether this depends on when they started learning L2 English. The results suggest that early bilinguals have native-like L2 English associative networks and these networks are interconnected with L1 Vietnamese words. Late bilinguals, in contrast, seem to indicate that their L2 English associative networks might be activated less automatically, and they are not so strongly connected with L1 words. This pattern of results are discussed in terms of how age of L2 exposure might affect the development of L2 associative networks.
Most Read This Month
