Poster Session 2

Several studies have demonstrated that helping learners recognize, understand and internalize the cognitive mechanisms that govern L2 grammar is particularly effective when L1 and L2 constructions differ in terms of temporal and aspectual features, in the encoding of motion events and illocutionary force intensity (Hijazo-Gascón & Llopiz-García 2019, Tyler & Ortega 2018). Grounding L2 teaching techniques in some core principles of Cognitive Linguistics (CL) is potentially useful to reduce the perceived high degree of inner complexity and the apparent arbitrariness of some grammar rules; moreover, it makes the potential L1-L2 discrepancies explicit and cognizable. In this study, the differential effectiveness of two different pedagogical approaches in teaching L1-Italian students the Russian verb of motion (VoMs) system is tested. The first is a CL-inspired approach, where the teacher used image schemas, bodily representations, and drawings to help the learners explore and understand the underlying logic of VoMs; moreover, real enactment of the meaning of VoMs has been proposed through Total Physical Response techniques (Elliot & Yountchi 2009). The second is a classic PPP approach, that works with mnemonic exercises, drills, and reproduction techniques, and that used classic metalinguistic terms to explain VoMs. Russian VoMs are a good test for examining the differential efficacy of a CL-inspired pedagogy. VoMs are notoriously recognized as one of the most complex topics in Russian grammar, not fully mastered even by highly proficient second language learners (Hasko 2009). Moreover, Russian and Italian follow typologically different lexicalization patterns for referring to motion events, thus making the acquisition of the VoMs system even more difficult for Italian-speaking learners. The two pedagogical interventions are focussed on the five (out of 14) most frequent and prototypical couples of VoMs, as reported in the following table: Monodirctional Pluridirectional Translation ???? (idti) ?????? (xodit’) To go on foot ????? (jexat’) ?????? (jezdit) To go by means of transport ?????? (bežat’) ?????? (begat’) To run ????? (nesti) ?????? (nosit’) To bring (by hand) ????? (vezti) ?????? (vozit’) To bring by using something Participants (n=42) were lower-intermediate L1-Italian students of Russian enrolled at the University of Bologna. All participants had already followed a Russian beginner (A1/A2) course, where they were taught about VoMs through a PPP intervention. The new activities on VoMs that are discussed in this paper were presented as a review of the VoM system. Twenty-three participants were placed in the CL-inspired group, whereas the remaining 19 took part in the PPP group. Participants’ learning of VoMs was tested through a temporized grammatical judgment test, a temporized sentence-picture matching, and a temporized multiple-choice exercise. All tests were manipulated and repeated three times, in a classic pre-, post- and delayed-post test experimental design. Data were analyzed using a mixed-design ANOVA with Treatment as the between-groups variable and Time as the within-group variable. Results show an accrued competence on VoMs of the CL group, that outperformed the PPP group in all the tests. The relevance of these findings for L2 teaching will be discussed. References Elliott E., Yountchi L., (2009). Total physical response and Russian multi- and unidirectional verbs of motion: a case study in acquisition. The Slavic and East European Journal, 53/3. Hasko V., (2009). The locus difficulties in the acquisition of Russian verbs of motion by highly proficient learners. The Slavic and East European Journal, 53/3. Hijazo-Gascón, A., Llopis-García, R., (2019). Applied Cognitive Linguistics and foreign language learning. Introduction to the special issue. IRAL 57/1. Tyler, A., Ortega, L. (2018). Usage-inspired L2 instruction: An emergent, researched pedagogy. In Tyler, A. et al. (eds.): Usage-inspired L2 instruction. Benjamins.

English as an Additional Language (EAL) learners have been shown to struggle academically (Coxhead & Boutorwick, 2018; Murphy & Unthiah, 2015). While research acknowledges the central place for vocabulary in both first and second language contexts (Webb & Nation, 2017), studies suggest that the academic performance of EAL learners is impacted because of their smaller vocabulary size, compared to their L1 counterparts (Henriksen & Danelund, 2015; Webb & Chang, 2012). Attempts to bridge this gap by selecting items EAL learners might require (Ogle et al., 2016) exist, but such approaches tend not to be systematic and it is common for both teachers (Schmitt, 2010) and textbook writers (Harwood, 2014) to disagree as to which words learners should focus on. Accordingly, a corpus-based list designed to meet learner needs is essential (Green & Lambert, 2018). The current study, therefore, attempts to leverage advances in corpus and word list creation (e.g. Gardner & Davies, 2014; Lei & Liu, 2016) for use in EAL contexts. The goal of the research is to develop a large corpus for the purpose of providing the EAL community with a series of lemma-based word lists for use in the international school context. The paper outlines the steps involved in the creation of an EAL specific corpus and subsequent EAL word list, which include: text selection and analysis; an overview of procedures taken to tag part-of-speech; a summary of how the corpus was vetted for problematic tags, scanning noise, and issues with proper nouns/adjectives (Nation, 2016); and how high frequency and academic items were identified from the corpus (following Coxhead, 2000; Green & Lambert, 2018; Greene & Coxhead, 2015). We investigate differences between the newly created word list and existing general as well as academic word lists and outline how EAL teachers can use this word list to support EAL curricula. While this foundational paper is conducted in a Japanese international school setting, we show its appeal to broader EAL settings and discuss policy and practice findings. References Coxhead, A., & Boutorwick, T. (2018). Longitudinal vocabulary development in an EMI International School context: Learners and texts in EAL, Maths, and Science. TESOL Quarterly, 52(3), 588–610. https://doi.org/10.1002/tesq.450 Gardner, D., & Davies, M. (2014). A new academic vocabulary list. Applied Linguistics, 35(3), 305-327. Greene, J. W., & Coxhead, A. J. (2015). Academic vocabulary for middle school students. Baltimore, MA: Paul H. Brookes Publishing. Green, C., & Lambert, J. (2018). Advancing disciplinary literacy through English for academic purposes: Discipline-specific wordlists, collocations and word families for eight secondary subjects. Journal of English for Academic Purposes, 35, 105-115. doi:10.1016/j.jeap.2018.07.004 Lei, L., & Liu, D. (2016). A new medical academic word list: A corpus-based study with enhanced methodology. Journal of English for Academic Purposes, 22, 42-53. Murphy, V. A., & Unthiah, A. (2015). A systematic review of intervention research examining English language and literacy development in children with English as an Additional Language (EAL). Oxford, England: University of Oxford. Nation, P. (2016). Making and using word lists for language learning and testing. Amsterdam, Holland: John Benjamins Publishing Company. Ogle, D., Blachowicz, C., Fisher, P., & Lang, L. (2015). Academic vocabulary in middle and high school: Effective practices across the disciplines. New York, NY: Guilford Publications. Webb, S., & Chang, A. C.-S. (2015). How does prior word knowledge affect vocabulary learning progress in an extensive reading program? Studies in Second Language Acquisition, 37(04), 651–675. https://doi.org/10.1017/s0272263114000606

Recent studies have demonstrated that musical expertise and aptitude promote L2 phonological processing (See Chobert & Besson, 2013; Milovanov & Tervaniemi, 2011 for a review). While rhythmic activities such as clapping have shown positive effects on learning L2 pronunciation (e.g., Zhang, Baills, & Prieto, 2018), the results regarding the potential effects of singing show somewhat mixed results, either positive (e.g., Ludke, Ferreira, & Overy, 2014) or negative (e.g., Nemoto, Wilson, & Perkins, 2016). Yet there is very limited empirical evidence coming from classroom training studies. To our knowledge, Ludke (2018) is the only quasi-experimental classroom study testing the effects of singing on L2 pronunciation, and results revealed no gains in their pronunciation test. Presumably, the negative results in these studies might come from a cognitive taxing effect of using unknown melodies in the singing intervention. Previous studies have reported a 'melody familiarity effect' which shows that training with familiar melodies as opposed to unfamiliar melodies increases L1 word recall and verbal memory in adults (Calvert & Tart, 1993; Tamminen, Rastle, Darby, Lucas, & Williamson, 2017). Thus, our goal is to assess whether using songs in the L2 classroom with familiar melodies can more clearly promote both L2 pronunciation and vocabulary learning. The present study aims at empirically testing whether a 3-session singing intervention (consisting of six 45-min sessions) in the classroom benefits both English pronunciation and vocabulary learning for young lower-intermediate Chinese learners of English. Using a between-subject pre- and posttest design, one hundred Chinese adolescents participated in the study. They were assigned to one of two conditions, singing (singing songs with familiar melodies) and rhythmic speech (rhythmically reading and speaking the same songs). Participants' pronunciation of words and sentences and vocabulary recall will be assessed before and after the training. Five native English speakers will evaluate pronunciation on an accentedness scale from 1 to 9. A set of control measures, including working memory, speech imitation ability, musical and linguistic backgrounds will also be assessed, taking into account individual differences. We expect that participants from the singing group intervention will improve more on both L2 pronunciation and vocabulary than those in the rhythmic-speech group. The statistical analysis for pronunciation assessment was delayed. Both groups improved in the vocabulary test after the training and the singing group improved more significantly than the speech group. References Calvert, S. L., & Tart, M. (1993). Song versus verbal forms for very-long-term, long-term, and short-term verbatim recall. Journal of Applied Developmental Psychology, 14(2), 245-260. Chobert, J., & Besson, M. (2013). Musical expertise and second language learning. Brain Sciences, 3(2), 923-940. Ludke, K. M., Ferreira, F., & Overy, K. (2014). Singing can facilitate foreign language learning. Memory & Cognition, 42(1), 41-52. Ludke, K. M. (2018). Singing and arts activities in support of foreign language learning: an exploratory study. Innovation in language learning and teaching, 12(4), 371-386. Milovanov, R., & Tervaniemi, M. (2011). The interplay between musical and linguistic aptitudes: a review. Frontiers in Psychology, 2, 321. Nemoto, S., Wilson, I., & Perkins, J. (2016). Analysis of the effects on pronunciation of training by using song or native speech. The Journal of the Acoustical Society of America, 140(4), 3343. Tamminen, J., Rastle, K., Darby, J., Lucas, R., & Williamson, V. J. (2017). The impact of music on learning and consolidation of novel words. Memory, 25(1), 107-121. Zhang, Y., Baills, F., & Prieto, P. (2018). Hand-clapping to the rhythm of newly learned words improves L2 pronunciation: Evidence from training Chinese adolescents with French words. Language Teaching Research, 1362168818806531.

Second language (L2) learners often encounter words that have two or more translations across languages, also referred to as translation-ambiguous words. These words are particularly difficult to process, and thus are remembered more slowly and less accurately than words with only one translation (also known as translation-unambiguous words; e.g., Eddington & Tokowicz, 2013; Tokowicz & Kroll, 2007). Such difficulty has also been reported for cognate words (words that share both form and meaning; e.g., Boada, Sánchez-Casas, Gavilán, García-Albea, & Tokowicz, 2013). Previous research has investigated different training methods attempting to reduce this translation-ambiguity disadvantage for beginning L2 learners. Degani, Tseng, and Tokowicz (2014) found that presenting multiple translations consecutively in the same session improves retention of translation-ambiguous words compared to training translations in separate sessions. The present study investigates the translation-ambiguity effect of cognates in beginning adult L2 learners. Native English speakers were introduced to Dutch translation-ambiguous and translation-unambiguous words that vary in their cognate status (non-cognates, close cognates, and identical cognates). Each English-Dutch word pair was presented on the screen along with the corresponding pronunciation of the Dutch word. Translation-ambiguous words were trained consecutively, that is, each English translation was presented one at a time during the same session (Degani & Tokowicz, 2014). Participants completed a production test in both directions of translation (English-Dutch, Dutch-English). Results suggest that, in contrast to previous research (e.g., Degani & Tokowicz, 2010), English translations that are more similar in meaning were remembered more accurately in the English-Dutch translation task than words with less meaning similarity. This finding is consistent with Bracken, Degani, Eddingon and Tokowicz’s (2017) conclusion that the mapping between the meaning and L2 label is what determines difficulty in L2 vocabulary learning. There was also a larger cognate effect for translation-ambiguous than translation-unambiguous words. References Boada, R., Sánchez-Casas, R., Gavilán, J. M., García-Albea, J. E., & Tokowicz, N. (2013). Effect of multiple translations and cognate status on translation recognition performance of balanced bilinguals. Bilingualism: Language and Cognition, 16, 183-197. Bracken, J., Degani, T., Eddington, C. M., & Tokowicz, N. (2017). Translation semantic variability: How semantic relatedness affects learning of translation-ambiguous words. Bilingualism: Language and Cognition, 20, 783-794. Degani, T., & Tokowicz, N. (2010). Ambiguous words are harder to learn. Bilingualism: Language and Cognition, 13, 299-314. Degani, T., Tseng, A. M., & Tokowicz, N. (2014). Together or apart? Learning of ambiguous words. Bilingualism: Language and Cognition, 17, 749-765. Eddington, C. M., & Tokowicz, N. (2013). Examining English-German translation ambiguity using primed translation recognition. Bilingualism: Language and Cognition, 16, 442-457. Tokowicz, N., & Kroll, J. F. (2007). Number of meanings and concreteness: Consequences of ambiguity within and across languages. Language and Cognitive Processes, 22, 727-779.

The current paper partially replicates an earlier study (Treffers-Daller, Parslow, & Williams, 2018) indicating that lexical diversity (LD) measures can help to discriminate between CEFR levels. We adapt Treffers-Daller et al.’s study to investigate the extent to which vocabulary measures are able to predict second language (L2) writing task (International English Language Testing System (IELTS)) scores. We use the same LD measures as in the earlier paper, along with a variety of vocabulary tools. In addition, the study adopts a recent approach (Clenton, De Jong, Clingwall, & Fraser, 2020) and investigates the multi-faceted construct of participant productive vocabulary knowledge. Our aim, therefore, is to contribute to discussions on how productive vocabulary knowledge and lexical diversity measures can help in the assessment of second language learners’ written work at different proficiency levels. We assess (n = 70) L1 Japanese undergraduate learners of (L2) English (CEFR B2). We adopt three different productive vocabulary tasks: Lex30 (Meara & Fitzpatrick, 2000) (a task based on word association responses); G_Lex (Fitzpatrick & Clenton, 2017) (a gap-fill task); and, the Productive Vocabulary Levels Test (PVLT; Laufer & Nation, 1999) (a sentence completion task). The performance of each of the productive vocabulary tasks has been shown (Clenton et al., 2020) to vary according to the quantity and quality of the productive vocabulary knowledge elicited. Participants responded to (6) different IELTS writing questions. For processing, we maintained a constant text length, and following recent research (Kyle, 2020) we flemmatized all writing samples. Our results reflect Treffers-Daller et al.’s (2018) findings to the extent that basic measures of LD such as TTR (Templin, 1957) explain more variance in writing scores than sophisticated measures such as D (Malvern et al. 2004) or MTLD (McCarthy 2005). A simple count of different words (defined as flemmas) proved to be the best predictor of variance in overall IELTS essay scores. Our data also reveal that higher IELTS task scores tended to reflect a higher quality of productive vocabulary knowledge. We discuss these findings in terms of second language acquisition, with specific implications for pedagogy. References Berger, C., Crossley, S., & Kyle, K. (2017). Using novel word context measures to predict human ratings of lexical proficiency. Educational Technology & Society, 20(2), 201-212. Clenton, J., De Jong, N., Clingwall, D., & Fraser, S. (2020). Investigating the extent to which vocabulary knowledge and skills can predict aspects of fluency for a small group of pre-intermediate Japanese L1 users of English (L2). In: Clenton, J. and Booth, P. (eds.) Vocabulary and the four skills. Routledge. (In Press) Crossley, S. A., & McNamara, D. (2012). Predicting second language writing proficiency: The roles of cohesion and linguistic sophistication. Journal of Research in Reading, 35(2), 115-135. Fitzpatrick, T., & Clenton, J. (2017). Making sense of learner performance on tests of productive vocabulary knowledge. TESOL Quarterly, 51, 844-867. Kyle, K. (2020). Measuring lexical richness. In S.Webb (Ed.). The Routledge handbook of vocabulary studies (pp. 454-476). New York: Routledge. Treffers-Daller, J., Parslow, P. & Williams, S. (2018) Back to basics: how measures of lexical diversity can help discriminate between CEFR levels. Applied Linguistics, 39(3), 302-327.

Second language (L2) learners are expected to gain increased control over a larger linguistic repertoire with advancing proficiency levels. This involves the acquisition and accurate use of more complex syntactic structures and more sophisticated vocabulary (Berman, 2009). Recent years have seen increased efforts to identify reliable indices of syntactic and lexical complexity and their automatic measurement through the use of natural language processing (NLP) tools. This line of research has advanced our understanding of the development of language proficiency and has demonstrated that computational indices can account for a substantial portion of the variance in human ratings of L2 writing quality (Jung et al., 2019). However, the vast majority of the existing studies have focused on a population of university students and have relied on elicited production tasks for research purposes rather than using educationally authentic L2 writing samples (regular school exams). Moreover, much of the prior work has focused on either accuracy or complexity, hindering our understanding of how these aspects interact and their relative impact on L2 writing assessment across school grades. This paper makes an important contribution to the existing research on the development of linguistic complexity and accuracy in L2 writing through a comprehensive analysis of a wide range of traditional indices of syntactic and lexical complexity and recently introduced indices derived from information theory. L2 English writing samples produced by 83 German-speaking learners in curriculum-based exams in upper secondary school (grades 10-12) were collected. Ratings of writing quality and accuracy scores were taken from the teachers’ assessments of the exams. Scores for 32 complexity indices were automatically obtained using CoCoGen (Ströbel et al., 2016). In addition to the writing samples, general demographic information and more specific information on self-rated L2 proficiency and current exposure to L2 English were collected and two standardized tasks to assess receptive vocabulary were administered. Three interrelated research questions were addressed: (1) Which aspects of complexity and accuracy exhibit growth across the school grades? (2) What are the interrelationships between indicators of complexity and accuracy, and how do they relate to learners’ receptive vocabulary knowledge and their self-rated L2 proficiency and exposure, and (3) Can holistic human ratings of L2 writing be predicted by indicators of complexity and accuracy and does the predictive power of different indicators change across school grades. Statistical analysis revealed, first, that growth in linguistic complexity occurred mainly at the lexical level, rather than the syntactic one. Second, significant medium-sized correlations were observed between linguistic complexity and grammatical and lexical accuracy, as well as between these indicators and self-rated L2 proficiency and L2 receptive vocabulary knowledge. Third, indices of accuracy and complexity jointly accounted for between 45% (grade 10) and 81% (grade 12) of the variance in human ratings. The relative importance of the predictors changed across school grades: while human ratings at grade 10 were mainly predicted by indicators of orthographic and grammatical accuracy, ratings at grades 11 and 12 were increasingly determined by indicators of lexical accuracy and complexity. The implications of these findings for theories of L2 assessment and L2 writing development are discussed. Berman, R. A. (2009). Developing linguistic knowledge and language use across adolescence. In E. Hoff & M. Shatz (Eds.), Blackwell Handbook of Language Development (pp. 347–367). Malden, MA: Wiley-Blackwell. Jung, Y., Crossley, S., & McNamara, D. (2019). Predicting Second Language Writing Proficiency in Learner Texts Using Computational Tools. Journal of Asia TEFL, 16(1), 37-52. Ströbel, M., Kerz, E., Wiechmann, D., & Neumann, S. (2016). CoCoGen-Complexity Contour Generator: Automatic Assessment of Linguistic Complexity Using a Sliding-Window Technique. In Proceedings of the Workshop on Computational Linguistics for Linguistic Complexity, 23-31.

Teacher identity development is a complex and multidimensional process affected by both internal and external factors (Beauchamp & Thomas, 2009; Farrell, 2017). Acknowledging the impact teacher education programs have on teacher identity formation (Clarà, 2015), this presentation puts forward an inquiry into the role a reflective identity-oriented approach can play in molding critically reflective language teachers. Aiming to examine how the identities of language teacher educators and candidates intercept and how explicit focus on reflection and teacher identity formation can foster reflective teachers, the project employed the Reflective Teaching Model (Weinburgh, Hart, & Carriere, 2007; Weinburgh, Smith, & Clark, 2008). Two teacher educators (TEs) adopted reflection and teacher identity as a pedagogy, inviting eighteen teacher candidates (TCs) to reflect on their own teacher identity formation through a series of structured reflective assignments that culminated in teacher identity texts posted online. The TEs used their own identity texts as models and engaged in a written mediation with the TCs. The data comprised a reflective blog and responses written jointly by two TEs; reflective blogs written by 18 TCs; TEs’ responses to TCs’ blog entries; and teacher identity texts written by TCs. The following research questions guided the study: 1) What are the foci of reflection in TEs’ blog entries? 2) What are the foci of reflection in TCs’ blog entries? 3) What themes emerge from TEs’ responses to TCs’ blog entries? 4) What are the foci of reflection in TCs’ finished identity statements? 5) In what ways do TEs’ and TCs’ reflection themes intercept? The data were analyzed using the procedure of content reduction (Miles & Humberman, 1994) to identify central themes through the process of coding. Themes from previous studies (Farrell, 1999; Liakopoulou, 2012; Tajeddin & Aghababazadeh, 2018; Yang, 2009) were applied deductively, but additional themes were derived inductively from the data. The findings suggest that TEs and TCs identities intercept and influence one another in complex ways which should be considered when designing TC courses. Insights into TCs’ identity and ways to optimize the way teacher-training programs are construed are discussed. References Beauchamp, C., & Thomas, L. (2009). Understanding teacher identity: An overview of issues in the literature and implications for teacher education, Cambridge Journal of Education, 39(2), 175-189. Clarà, M. (2015). What is reflection? Looking for clarity in an ambiguous notion. Journal of Teacher Education, 66(3), 261-271. Farrell, T.S.C. (1999). Reflective practice in an EFL teacher development group. System, 27(2), 157-172. Farrell, T.S.C. (2017). Research on reflective practice in TESOL. New York, NY: Routledge. Liakopoulou, M. (2012). The role of field experience in the preparation of reflective teachers. Australian Journal of Teacher Education, 37(6), 42-54. Miles, M. B., & Huberman, A. M. (1994). Qualitative data analysis. Newbury Park, CA: Sage. Tajeddin, Z., & Aghababazadeh, Y. (2018). Blog-mediated reflection for professional development: Exploring themes and criticality of L2 teachers’ reflective practice. TESOL Canada Journal, 35(2), 26-50. Weinburghh, M.H., Hart, L., & Carriere, J. (2007). A new perspective on integrating math and science: The Decatur elementary math/science project. In D. F. Berlin & A. L. White (Eds.), Global issues, challenges, and opportunities to advance science and mathematics education (pp. 21–38). Columbus: OH: International Consortium for Research in Science and Mathematics Education. Weinburghh, M., Smith, K., & Clark, J. (2008). Using the reflective teaching model in a year-long professional development: A case study of a second year urban elementary teacher. Electronic Journal of Science Education, 12(2), 1-20. Yang, S.-H. (2009). Using blogs to enhance critical reflection and community of practice. Educational Technology and Society, 12(2), 11-21.

The current study presents a fully-controlled comparison between human-human and human-computer phonetic alignment. Phonetic alignment in interaction is relevant for L2 acquisition since it is one of the ways in which L2 speakers can achieve a more native-like pronunciation. RQ1 investigates whether speakers align differently to a human confederate than to a computer even with the exact same speech input. We can, thus, determine if the results from studies using computer-based approaches are generalizable to real-life human-human interactions. We further investigate whether phonetic alignment is triggered by the presence of (RQ2) exposure to a native pronunciation of a difficult English L2 sound contrast, /æ-?/, and (RQ3) feedback on the learner’s pronunciation of the contrast. Methodology: Participants played a cooperative puzzle-solving game with a human or computer interlocutor. On each trial both players had to exchange information on their screen. Crucially we used the same pre-recorded speech in the computer and the human setting, combining the spontaneity of the conversation with phonetic control. The Ventriloquist Paradigm [1], used in the human-human setting, allowed for the implementation of prerecorded speech unbeknownst to participants, who believed they were having a live conversation with person sitting in front of them. In the computer setting, participants believed the speech was coming from a smart computer, in the style of the Wizard of Oz [2]. Participants: Ninety-six female Dutch speakers participated in this study. Half of the participants were assigned to the group who interacted with a human confederate and the other half to the group who interacted with the computer interlocutor. Design: During the interaction, exposure to the native pronunciation of the L2 contrast (presence/absence) and implicit negative visual feedback on the learner’s pronunciation (presence/absence) were fully crossed in the two settings. The presence of exposure consisted of incorporating 12 tokens of each vowel in the prerecorded speech of the interlocutor, whereas the absence was implemented by carefully avoiding them throughout the interaction. In the feedback condition, visual feedback always implied that the interlocutor interpreted participants’ /æ/ was interpreted as /?/. Results: Participants’ pronunciation of words including /æ/ and /?/ was recorded immediately before and after the interaction while they performed a sentence completion task out loud. The acoustic analyses on the formants of the two vowels showed that participants aligned significantly differently to the two interlocutors, even when they were exposed to the same speech input (RQ1). In the human-human setting, both exposure to the contrast (RQ2) and feedback on the learner’s pronunciation (RQ3) alone triggered a more native-like pronunciation of the contrast. This effect disappeared when exposure and feedback were combined. In the human-computer setting, to the contrary, exposure and feedback alone did not trigger alignment; however, the two factors combined elicited more native-like pronunciation in the posttest. Discussion: The results show that speakers aligned differently to humans than to computers. Whether speakers thought they were interacting with a computer or a human affected the effect of exposure to the native pronunciation of the contrast and implicit negative feedback on phonetic alignment. Therefore, this study provides evidence that the conclusions from computer-based studies are not always generalizable to human-human interaction. [1] E. Felker, A. Troncoso-Ruiz, M. Ernestus, M. Broersma, The ventriloquist paradigm: Studying speech processing in conversation with experimental control over phonetic input, The Journal of the Acoustical Society of America. 144 (2018) EL304–EL309. https://doi.org/10.1121/1.5063809. [2] L. Riek, Wizard of Oz Studies in HRI: A Systematic Review and New Reporting Guidelines, Journal of Human-Robot Interaction. (2012) 119–136. https://doi.org/10.5898/JHRI.1.1.Riek.

Exposure to multimodal input from L2 captioned videos can promote the update of phono-lexical representations, preventing the fossilization of inaccurate forms [1]. One main reason is that on-screen text offers visual support where utterances are already segmented into words and remain available longer than in running speech [2]. Enhancing words in captions may direct learners' attention to specific target features, and combining enhancement with L2 instruction for extended periods may promote deeper processing and awareness of these features [3].
Since words in captions tend to be fixated around 200 ms before the corresponding auditory onset (AO) [4], we hypothesized that highlighting a word around 300 to 500 ms before AO would help learners recognize its written form, attend to its auditory form and test the accuracy of pre-existent representations. In an eye-tracking study (N=58), we found that target word enhancement in the captions of video clips reduced skipping probability, and that synchronized enhancement promoted closer audio-visual synchrony than unsynchronized enhancement. Participants also did an auditory lexical decision task requiring the rejection of inaccurately pronounced forms of the words enhanced in the clips. Synchronizing enhancement 500 ms before AO produced the largest reaction time gains from pre- to post-test.
We then designed a longitudinal study with L1 Spanish/Catalan high school learners of English, targeting the past -ed phonological realization of high frequency verbs. Each of the 5 sessions starts with the viewing of an enhanced video clip, followed by an audiovisual (AV) activity requiring manipulation of multimodal input, such as dubbing or captioning [5]. These activities, traditionally used to train translators, have recently been introduced in the classroom and may promote language learning in the context of form-focused instruction. However, the study of pronunciation gains with AV activities has been mostly qualitative and limited to university students [6]. In our intervention, the learners complete one AV activity per session, then watch again the enhanced video clip to get feedback on their perception or production of the target forms. Finally, they carry out an awareness-raising activity to consolidate the knowledge acquired. Phono-lexical update is monitored with pre-, post- and delayed post-tests targeting the auditory forms in the intervention as well as other untrained forms. Another group of learners carries out the same intervention without textual enhancement, and a control group only does the pre- and post-tests.
References[1] Darcy, I. (2018). Powerful and Effective Pronunciation Instruction: How Can We Achieve It? CATESOL Journal, 30(1), 13–45.
[2] Charles, T., & Trenkic, D. (2015). The Effect of Bi-Modal Input Presentation on Second Language Listening: The Focus on Speech Segmentation. In Y. Gambier, A. Caimi, & C. Mariotti (Eds.), Subtitles and language learning (pp. 173-197). Bern: Peter Lang.
[3] Leow, R. (2009). Input Enhancement and L2 Grammatical Development: What the Research Reveals. In: J. Watzinger-Tharp, & S. Katz (Eds.), Conceptions of L2 grammar: Theoretical approaches and their application in the L2 classroom (pp. 16–34). Boston, MA: Heinle.
[4] Wisniewska, N., & Mora, J. C. (2018). Pronunciation Learning through Captioned Videos. In J. Levis (ed.), Proceedings of the 9th Annual Pronunciation in Second Language Learning and Teaching Conference, pp. 204-215. Ames: Iowa State University.
[5] Sokoli, S. (2018). Exploring the possibilities of interactive audiovisual activities for language learning. Special Issue of Translation and Translanguaging in Multilingual Contexts. Audiovisual Translation in Applied Linguistics: Beyond Case Studies, 4(1), 77–100.
[6] Chiu, Y. (2012). Can film dubbing projects facilitate EFL learners' acquisition of English pronunciation? British Journal of Educational Technology, 43(1), 24-27.

Since Dörnyei’s (2009) seminal article, the L2 motivational self system (L2MSS) has been one of the most popular theoretical foundations in L2 motivation research for more than 10 years. The L2MSS aims to reconsider social psychological approach to L2 motivation, and the notion of self was introduced from mainstream psychology. The central concepts of the L2MSS are developed based on the learners’ self image, and there are three components: the ideal L2 self, the ought-to L2 self, and the L2 learning experience. Drawing on the concept, the impacts of the components of the L2MSS on language learners’ efforts, attitudes, and learning outcomes have been investigated in the past decade. Particularly, quantitative approaches using questionnaires were commonly conducted in many different research contexts. Among those studies, one of the most widely used questionnaires is the one developed for Taguchi, Magid, and Papi’s (2009) which aimed to validate the concept of the L2MSS in the Japanese, Iranian, and Chinese contexts. This study aims to re-evaluate Taguchi et al.’s (2009) questionnaire by synthesising Cronbach’s alpha reliability coefficients of the elements. Studies were searched on 19 international journals in the field of SLA (e.g., Studies in Second Language Acquisition), language education (e.g., Language Learning, Language Teaching Research, System) and applied linguistics (e.g., Applied Linguistics, Modern Language Journal). Inclusion criteria for the article pool were the studies using Taguchi et al.’s (2009) questionnaire items in quantitative investigations, and reporting sample size and Cronbach’s alpha reliability coefficients for the ideal L2 self, the ought-to L2 self, the intended effort (or criterion measurement), and the L2 learning experience (or attitudes toward L2 learning). Subsequently, information required for the synthesis of Cronbach’s alpha coefficients was recorded in a spreadsheet. Items recorded in the spreadsheet were: (1) article information (e.g., author, publication year), (2) Cronbach’s alpha coefficients, (3) sample size, and (4) number of items in each construct. As a result of the literature search and screening, 30 studies involving more than 20,000 participants were included in the article pool. Obtained Cronbach’s alpha coefficients for each construct were synthesised using a transformation method suggested in Bonett (2002, 2010). As a result of a preliminary data analysis, the aggregated Cronbach’s alphas for ideal self, ought-to self, intended effort, and L2 learning experience are .875 [CI: .872–.878], .774 [CI: .769–.779], .810 [CI: .806–.815] and .865 [CI: .862–.868], respectively. When comparing the values for each construct, Cronbach’s alpha coefficient for ought-to L2 self was relatively low. However, according to Plonsky and Derrick’s (2016) synthesis of reliability coefficients in SLA research, the minimally acceptable value of the internal reliability of the instrument in L2 research should be .74. Therefore, based on the overall value of Cronbach’s alpha coefficients, Taguchi et al.’s (2009) scales measuring central components of the L2MSS maintain a sufficient level of internal consistency. Further analysis will be carried out to investigate the differences in the values mediated by moderator variables, such as target language, age of learners, and the context of learning. Dörnyei, Z. (2009). The L2 motivational self system. In Z. Dörnyei & E. Ushioda (Eds.), Motivation, language identity and the L2 selves (pp. 9–42). Bristol, England: Multilingual Matters. Plonsky, L., & Derrick, D. J. (2016). A Meta-Analysis of reliability coefficients in second language research. The Modern Language Journal, 100, 538–553 Taguchi, T., Magid, M., & Papi, M. (2009). The L2 motivational self system among Japanese, Chinese and Iranian learners of English: A comparative study. In Z. Dörnyei & E. Ushioda (Eds.), Motivation, language identity and the L2 selves (pp. 66–97). Bristol, England: Multilingual Matters.

In this paper we will report on an intervention study that focuses on teaching English academic reading at Dutch vwo, the highest tier of Dutch secondary education, which prepares students for university. Despite the relatively high proficiency in their second language (L2) English, many Dutch university students struggle with reading academic texts (Wierda-Boer, 2008). The specific features of scientific texts such as nominalization, intertextuality, writer’s stance and modal verbs are often new to students and limits their understanding (Charles & Pecorari, 2016). In addition, many pre-academic high-schools do not provide instruction on how to read critically, a crucial strategy at university that allows students to evaluate an authors’ message (Westhoff, 2012). An extra concern is the recent PISA-report that demonstrates that Dutch students’ reading skills are dropping in comparison with other countries, which illustrates that reading deserves serious attention (NRO, 2019). Without the knowledge and skills to apply successful reading strategies, students might not obtain optimal study results once they are at university, which potentially affects students’ self-efficacy and may trigger demotivation (McNamara, 2007). It is therefore important for students to learn how to read academic texts before entering university. The focus of our contribution is to evaluate the outcomes of a lesson series targeting academic reading (e.g. reading progress, meta-cognitive reading strategies and self-efficacy) by means of quantitative data-analysis methods (questionnaires and summaries). The academic course was implemented at pre-university level in the final year of Dutch secondary schools. Twelve classes (N= ca. 25 each) at eight schools participated in the study. Over the course of two months they received a series of ten lessons of reading instruction integrated in their regular program. Metacognitive reading strategies and critical reading were at the core of the course, with a strong focus on annotation (Porter-O’Donnell, 2004). Students were introduced to the strategy through inductive learning, after which they saw a clip in which a teacher modelled the strategy (Schunk, 2003) By means of gradual release of responsibility students practiced the strategy before they discussed their process in groups (McNamara, 2007). At the end of the course students were expected to individually apply the strategies enabling them to critically process a scientific text. The pre-post test design (Repeated Measures ANOVA) allowed us to measure the effectiveness of the reading intervention in in comparison with results from a control group of students. Before and after the lesson series students filled out questionnaires measuring metacognitive reading strategies, motivation and self-efficacy (McNamara, 2007). In the same lessons students also completed a summary to an academic text to measure reading comprehension. The discussion will highlight how the academic reading course may have supported students’ academic reading skills development in their L2 English. Charles, M., & Pecorari, D. (2016). Introducing English for academic purposes (Routledge introductions to English for specific purposes). London: Routledge, Taylor & Francis Group. McNamara, D.S. (2007). Reading comprehension strategies: Theories, interventions and technologies. Mahway, NJ: Lawrence Erlbaum Associates. NRO. (2019, December 3). Nederlandse 15-jarigen lezen minder goed en hebben minder plezier in lezen, maar presteren nog even goed in de exacte vakken. Retrieved December 16, 2019, from https://www.nro.nl/pisa-2018-uitkomsten/ Porter-O’Donnell, C. (2004). Beyond the yellow highlighter: Teaching annotation skills to improve reading comprehension. English Journal, 93(5), 82--89. Schunk, D.H. (2003). Self-efficacy for reading and writing: Influence of modeling, goal setting, and self-evaluation. Reading and Writing Quarterly, 19(2), 159-172. Available from http://www.tandf.co/journals/ Westhoff, G. (2012). Mesten en meten in leesvaardigheidstraining: Leesvaardigheidstraining en examentraining zijn twee verschillende dingen. Levende Talen Magazine, 4(99), 17–20 Wierda-Boer, H. (2008). Aansluiting Engels. Een onderzoek naar de aansluitingsproblematiek van het vwo-vak Engels met de universiteit, 1–11.

Numerous studies have confirmed that output is the crucial part of second language learning, in which noticing of form has been captured in the form of language-related episodes (LREs) (Swain & Lapkin, 1998). Although output in oral and written modalities has distinctive impact on language learning, major focus has been on the oral modality especially in TBLT research, and there is little research which contrasts these modalities (Gilabert, Manchón, & Vasylets, 2016). Until now, it has been shown that written tasks produce more LREs, thus may lead to more language reflection and learning (Garcia Mayo & Azkarai, 2016; Niu, 2009). As for the relationship between LREs and the learning of a specific form through a task, studies found that positive learning outcome is associated with more engaged LREs (Kuiken & Vedder, 2002; Storch, 2008); however, there is a paucity of research on the possible relationship of task modality, quality of LREs, and the learning of a specific form. To address this gap, the current study aims to investigate the impact of oral and oral + written interactive output tasks operationalized as a picture and word-cued sentence reconstruction task on the development of passive voice of 16 to 17-year-old EFL students in Japan. 11 pairs of students collaboratively reconstructed 21 sentences orally, 17 of which required the use of passive voice and the rest being distracters, while 19 pairs employed writing to mediate their oral collaborative discussion, lasting 30 minutes for both groups. The tasks were contextualized in the existing theme-based (e.g., the space elevator) English class and student interaction was voice recorded and transcribed. The transcription data was first analyzed in terms of orientation (e.g., lexis-, form-, or meaning-based LRE), resolution, and engagement (e.g., elaborate or limited) of LREs to see what kind of learning opportunities the two task conditions afforded. In addition, their gains in passive form knowledge was measured by means of pre-, post- and delayed error correction tests. Preliminary results show that both task modalities elicited learners’ focus on passive form and various other forms as well, but the oral task afforded more actual use of passive form within a set timeframe, possibly due to its fast pace nature of production. The results of error correction tests showed that both groups significantly developed their passive form knowledge from pre- to post-/delayed post-tests, and notably the oral task group outperformed the oral + written group in the post-/delayed post-tests. The results will be discussed in reference to different qualities of LREs generated, and pedagogical implications as to the use of oral /written tasks to learn a specific form will be presented. García Mayo, M. P. & Azkarai, A. (2016). EFL task-based interaction: Does task modality impact on language-related episodes? In Peer Interaction and Second Language Learning (pp. 241-266). John Benjamins. Gilabert, R., Manchón, R., & Vasylets, O. (2016). Mode in Theoretical and Empirical TBLT Research: Advancing Research Agendas. Annual Review of Applied Linguistics, 36, 117–135. https://doi.org/10.1017/s0267190515000112 Kuiken, F., & Vedder, I. (2002). The effect of interaction in acquiring the grammar of a second language. International Journal of Educational Research, 37(3–4), 343–358. https://doi.org/10.1016/S0883-0355(03)00009-0 Niu, R. (2009). Effect of task-inherent production modes on EFL learners’ focus on form. Language Awareness, 18(3–4), 384–402. https://doi.org/10.1080/09658410903197256 Storch, N. (2008). Metatalk in a pair work activity: Level of engagement and implications for language development. Language Awareness, 17(2), 95–114. https://doi.org/10.2167/la431.0 Swain, M., & Lapkin, S. (1998). Interaction and Second Language Learning?: Two Adolescent French Immersion Students Working Together. The Modern Language Journal, 82(iii), 320–337.