Session 4D

Incidental learning from L2 reading is said to be a slow process because of the massive amounts of input from graded readers that are required (Webb & Nation, 2017). In an attempt to maximize learning from reading, research in SLA has been conducted on the use of several instructional strategies such as input enhancement, assisted reading, repeated reading or narrow reading. However, most of these strategies were developed with teenagers and adult learners in mind and few studies document their impact on children (Author et al., 2019). One additional challenge in L2 reading with young learners is the fact that graded readers for children are very short so repeated exposures to new language items are less likely to occur and therefore chances for incidental learning to happen low (Uchiara, Webb & Yanagisawa, 2019). In the present study an instructional procedure was developed (referred to here a ‘Multiple exposures’ ME) to be used with children’s graded readers that ensures multiple exposures in the context of meaning-focused instruction. The present classroom-based study compares the ‘Multiple exposures’ procedure with traditional focus on forms instruction and documents its impact on L2 vocabulary, grammar and pronunciation. To this end, 44 Catalan/Spanish bilingual children (aged 11) learners of English (L2) participated in the present study. Participants belonged to two intact classes in a semi-private school in Catalonia (Spain). One class followed ME procedure (n= 23) and the other one followed a traditional form-focused (FF) approach (n= 21). Both groups of students were asked to read the same two graded readers, but the ME group was exposed to the content of the story four times and through multiple formats (storytelling, collective and individual reading-while-listening, and jigsaw reading), whereas the FF group read the books once and carried out a number of focus on forms exercises that were developed for them to learn the target vocabulary and grammar (simple past forms). Both groups spent two sessions (2 hours) per book, so the treatment lasted four sessions (4 hours). The tests were the same for the pre-, post- and delayed post-test. While the pre-test took place the week before the treatment, the post-test was administered the week after the treatment, and a delayed post-test was administered three weeks after the post-test (the latter included only the vocabulary test). The paired-samples T-tests showed that the participants in the FF group improved vocabulary and grammar significantly from the pre- to the post-test, whereas participants in the ME group improved significantly vocabulary and pronunciation. However, the ANCOVA test showed that the ME and FF groups were statistically significant only in terms of grammar, which favored the FF group. References Author et al. (2019). Uchiara T., Webb S., & Yanagisawa A., (2019). The Effects of Repetition on Incidental Vocabulary Learning: A Meta-Analysis of Correlational Studies. Language Learning, 69(3), 559-599. Webb S. & Nation P. ( 2017). How vocabulary is learnt. Oxford University Press.

In real-world communication, second language (L2) is commonly used for the purpose of information transfer. In educational contexts, for instance, students often exchange information on what they have read (e.g., book chapters, articles) to enhance their content learning. In response to the demands for developing L2 oral skills needed for successful information transfer, scholars have investigated integrated speaking tasks such as reading-to-speaking tasks (e.g., Iwashita, Brown, McNamara, O’Hagan, & De Jong, 2008). Recent advances in multi-modal text presentation and processing have also highlighted the facilitative role of multi-modal input in reading comprehension. These beneficial effects are primarily due to the enhancement of phonological representations of linguistic items in multimodal presentation conditions (Liu & Todd, 2014). However, little is known about the extent to which such facilitative effects of multi-modal input can be transferred into students’ oral summaries of reading texts. Using reading-to-speaking tasks, the present study investigated the effects of multi-modal presentation on the subsequent oral text summary in terms of the lexical overlaps with the source texts and utterance fluency. The multi-modal condition of reading in the study was operationalized by read-aloud assistance (RAA) in which participants read the source text while simultaneously listening to its oral recording (Košak-Babuder, Kormos, Ratajczak, & Pižorn, 2019). A total of 88 Japanese learners of English completed reading-to-speaking tasks under two conditions: (a) reading without RAA (–RAA; written-only) and (b) reading with RAA (+RAA; written and aural). Their speech data were evaluated by lexical overlap indices which indicate the ratio of single-words and multiword sequences characteristic of the source texts (for the detailed description, see Crossley, Kyle, & Dascalu, 2019). Moreover, a set of fluency measures was also employed to capture three major dimensions of utterance fluency—speed, breakdown, and repair fluency (Bosker, Pinget, Quené, Sanders, & De Jong, 2013; Tavakoli & Skehan, 2005). A series of Wilcoxon signed-rank tests revealed that participants’ speech under the +RAA condition was characterized by higher articulation rate (speed fluency; z = 1.997, p = .046, d = .436) and fewer false starts (repair fluency; z = 2.057, p = .040, d = .449) with the higher ratio of the single-word overlap (z = 2.745, p = .006, d = .613) than under the –RAA condition. In addition, correlational analyses showed that the use of multiword keywords was supportively associated with both articulation rate and final-clause pause ratio under both conditions. However, we found the supportive role of multiword keywords use in the frequency of mid-clause pauses and self-repetitions only under the +RAA condition. These findings suggest that the enhanced phonological representations of linguistic items in the multi-modal presentation condition may lead to efficient speech processing as well as the successful retrieval of lexical items characteristic of the source texts. Moreover, the stronger links between the use of multiword keywords and utterance fluency under the +RAA condition may indicate that multi-modal input might help L2 learners to strengthen the association of individual lemmas embedded in multiword sequences during text comprehension. Selected references: Crossley, S. A., Kyle, K., & Dascalu, M. (2019). The Tool for the Automatic Analysis of Cohesion 2.0: Integrating semantic similarity and text overlap. Behavior Research Methods, 51(1), 14–27. Košak-Babuder, M., Kormos, J., Ratajczak, M., & Pižorn, K. (2019). The effect of read-aloud assistance on the text comprehension of dyslexic and non-dyslexic English language learners. Language Testing, 36(1), 51–75. Liu, Y. T., & Todd, A. G. (2014). Dual-modality input in repeated reading for foreign language learners with different learning styles. Foreign Language Annals, 47(4), 684–706.

Previous work on literacy development reveals that certain cognitive and linguistic variables have either a direct or indirect role in reading comprehension. While much work has focused on reading comprehension in monolingual children (e.g. Babayi?it & Stainthorp, 2011), the underlying processes of reading in bilingual populations have received somewhat limited attention (e.g. Verhoeven & Van Leeuwe, 2012). On the basis of longitudinal data from the same children in three successive years, the aim of this study is to examine Turkish literacy skills of Turkish monolingual and Kurdish-Turkish sequential bilingual children, a language pair which has not been investigated in the previous literature. Data collection in this study started at the kindergarten level and will continue until the end of Grade 2. Here, we present some portions of the preliminary findings from 123 kindergarten (63 girls, 60 boys, mean age= 67.8 months) and Grade 1 students , residing in a city in the Eastern Region of Turkey, where the majority of the residents speak Kurdish as a mother tongue. We specifically investigate how Turkish monolinguals’ (N=62) and Kurdish-Turkish bilinguals’(N=61) phonological awareness (PA), rapid automatized naming (RAN), verbal memory, listening comprehension and vocabulary knowledge measured in the kindergarten predict their Turkish reading comprehension in Grade 1. The participants were individually tested at three times: at the beginning (time 1) and at the end of the kindergarten (time 2), and also at the end of Grade 1 (time 3). Measures of PA, RAN, verbal memory, vocabulary knowledge and listening comprehension were administered in the kindergarten and at Grade 1. Also, reading comprehension and word reading tasks were added at Grade 1. The preliminary descriptive findings revealed a development in PA, RAN, vocabulary knowledge and listening comprehension performances across groups from time 1 to time 3. As for inferential statistics, a 2(groups:monolingual and bilingual) x 5(measures: PA, RAN, verbal memory, listening comprehension, vocabulary knowledge) x 3(times: kindergarten 1, kindergarten 2, Grade 1) mixed ANOVA was conducted in order to find out possible performance differences among monolingual and bilingual children across the times. Despite the performance development in both groups, the monolingual children outperformed the bilingual children except for PA performance that was in favor of the latter group, especially at the beginning of the kindergarten. However, with formal instruction, the monolingual children caught their bilingual counterparts in PA tasks performance. On the other hand, as in line with the literature (e.g. August & Shanahan, 2008), the monolingual children statistically did better than bilingual children especially in vocabulary and listening comprehension measures. Also, several hierarchical regression models were run to see the predicting power of PA, RAN,verbal memory, vocabulary knowledge, listening comprehension and word reading in reading comphrehension. The results also revealed that listening comprehension (41%), vocabulary (16%) and word reading (10%) were the best predictors of Grade 1 reading comprehension performance of the monolingual children. On the other hand, PA, RAN and verbal memory had an indirect role in reading comprehension, which was through these groups’ word reading performance. As for the bilingual children, however, the situation was slightly different. PA (40%) and word reading (12%) explained the most variance in reading comprehension. These results were discussed in line with simple view of reading (Gough & Tunmer, 1986), which proposes reader’s decoding and listening comprehension skills as the driving forces in reading comprehension performance. Selected References: August, D., & Shanahan, T. (2008). Developing literacy in second-language learners: Report of the national literacy panel on language-minority children and youth. New York: Routledge. Gough, P. B., & Tunmer, W. E. (1986). Decoding, reading, and reading disability. Remedial and Special Education, 7, 6–10.

Vocabulary’s relation to reading proficiency is frequently cited as a justification for the assessment of L2 written receptive vocabulary knowledge. However, to date there has been relatively little research regarding which modalities of vocabulary knowledge have the strongest correlations to reading proficiency, and observed differences in correlations have often been statistically non-significant (e.g., Laufer & Aviad-Levitzky, 2017). The present research employs a bootstrapping approach to reach a clearer understanding of relationships between various modalities of vocabulary knowledge to reading proficiency. Test-takers (N = 103) answered 1,000 vocabulary test items spanning the 3rd 1,000 most frequent English words in the New General Service List corpus (Browne, Culligan, & Phillips, 2013). Items were answered under four modalities: Yes/No checklists, form-recall, meaning-recall, and meaning-recognition. These pools of test items were then sampled with replacement to create thousands of simulated tests ranging in length from 5 to 200 items and the results were correlated to Test of English for International Communication (TOEIC®) Reading scores. For all examined test lengths, meaning-recall vocabulary tests had the highest average correlations to reading proficiency, followed by form-recall meaning recognition and Yes/No tests, respectively. On average, meaning recall tests of 40 items had correlations to reading proficiency approximately .09 higher than comparable meaning recognition tests and approximately .12 higher than comparable Yes/No tests, with large effect sizes of 3.62 and 4.90, respectively. Although mean differences remained comparable, for 100 item tests, effect sizes for these differences rose to 5.28 and 8.56 respectively. Given the very large effect sizes, the differences in correlations do not appear to be negligible, as few recognition tests could approach the predictive power or associative strength of otherwise equivalent meaning recall tests. Accounting for the time required to take each test did not alter the rank order of predictive power. For meaning-recognition, form-recall and meaning-recall tests, peaks in correlation occurred after approximately 30 minutes of test-taking time, suggesting that testing learners’ vocabulary knowledge for longer lengths of time may provide diminishing returns if the tests are used as an indicator of reading proficiency. The results indicated that tests of vocabulary recall are stronger predictors of reading proficiency than tests of vocabulary recognition, despite the theoretically closer relationship of vocabulary recognition to reading. Browne, C., Culligan, B., & Phillips, J. (2013). The new general service list. Retrieved from www.newgeneralservicelist.org. Laufer, B., & Aviad–Levitzky, T. A. M. I. (2017). What type of vocabulary knowledge predicts reading comprehension: Word meaning recall or word meaning recognition?. The Modern Language Journal, 101(4), 729-741.