Poster abstracts

Please find below the abstracts for the poster session. The numbers refer to the poster board where you can find the presentation. Information for presenters can be found here.

1. Orlanda de Azevedo (Universidade Nova de Lisboa): The flipped classroom in the Portuguese as foreign language class

The innovative character of the Flipped Classroom teaching technique, the potentialities it presents, and the fact that it reconciles previously considered incompatible theories (student-centered theories prescribing interactive activities in the classroom, as opposed to teacher-centered theories advocating explicit teaching methods) make it very appealing. The Flipped Classroom has been implemented in several Universities in the USA, but there is a lack of studies regarding the teaching and learning of Portuguese as a Foreign Language. What are its advantages and disadvantages in this particular subject and context?  I will attempt to answer this question taking into consideration that it is common to find English-speaking students, native speakers of Spanish and heritage speakers of Portuguese in the same class in North American Universities, and exploring the argument that the Flipped Classroom technique could be instrumental in differentiated teaching. In this poster, I will also present a thematic unit for Portuguese Language (L2/FL), created by me according to this method and developed considering the ACTFL’s (American Council on the Teaching of Foreign Languages) Standards for Foreign Language Learning and Guidelines.

2. Mónica Bastos (Camões, I. P. / CIDTFF – Universidade de Aveiro): From “overload” to “support” – Portuguese as heritage Language teachers’ perceptions on the dynamics of a community of practices in changing times

Portuguese is the second most spoken language at home by the students of the Luxembourgish schools (MENJE, 2016) and Portuguese community represents 16% of the residents in the country (STATEC, 2018). Since September 2017, the Camões, IP. offer of Portuguese as Heritage Language (PLH) comprises complementary courses, due to a disagreement which almost led to the end of PLH courses at the second most important town of the country (Esch-sur-Alzette). These courses, based on the plurilingual and intercultural education’s principles (Beacco et al, 2016), count on the expertise of the same teachers of the extinct courses, who embraced the project with commitment, but also with many fears, insecurities and resistance. In order to support these teachers, a community of practices group was created, with weekly meetings and the pedagogical support of the Co-ordination Unit for Teaching Portuguese in Luxembourg. Their work dynamics are organised around four axes: curricular management and development; pedagogical materials production and experimentation; teacher education and supervision; and observation/analysis of the process. In this poster, we will describe the dynamics of these community of practices in more detail and present some preliminary results from teachers’ perceptions about possible implications of these dynamics on their professional development.


  • Beacco, J.-C., Byram, M., Cavalli, M., Coste, D., Cuenat, M. E., Goullier, F., & Panthier, J. (2016). Guide pour le Développement et la Mise en Œuvre de Curriculums pour une Éducation Plurilingue et Interculturelle. Strasbourg: Conseil de l’Europe.

3. Sara D. Beck and Andrea Weber (University of Tübingen): Context and literality in native and non-native idiom processing: Evidence from self-paced reading

The current study investigates how non-native (L2) idiom processing is impacted by context and idiom literality compared to native (L1) processing. Current research has shown that context eases access to figurative meaning, though not necessarily literal meaning in L1 speakers (e.g., Holsinger & Kaiser, 2013).  At the same time, idioms vary greatly in their potential to be interpreted literally (e.g., high-literality: break this ice vs. low-literality: lose one’s cool), and L1 and L2 speakers in their abilities to make use of such linguistic cues and possibly their reliance on literal language (e.g., Cieślicka 2006). In a self-paced reading study, we asked whether proficiency and literality can limit the effects of context on both literal and figurative meaning composition. L1 and L2 participants read sentences containing idioms biasing either a figurative or a literal reading of high- and low-literality idioms and concluded with resolutions either congruent or incongruent with expectations (e.g., [The new schoolboy/the chilly Eskimo] just wanted to break the ice [with his peers/on the lake] …). Analyses show that supporting context improves access to figurative meaning after idiom recognition in both types of idioms but provides new evidence for literal interpretations following supporting contexts in L1 and L2 readers.


  • Cieślicka, A. B. (2006). Literal salience in on-line processing of idiomatic expressions by second language learners. Second Language Research, 22(2), 115–144.
  • Holsinger, E., & Kaiser, E. (2013). Processing (non)compositional expressions: mistakes and recovery. Journal of Experimental Psychology. Learning, Memory, and Cognition, 39(3), 866–878.

4. Serge Bibauw (KU Leuven & UCLouvain), Wim Van den Noortgate (KU Leuven), Thomas François (UCLouvain), and Piet Desmet (KU Leuven): Dialogue-based CALL: A multilevel meta-analysis

Dialogue-based CALL systems allow a learner to practice meaningfully an L2 with an automated agent, through an oral (spoken dialogue systems) or written interface (chatbots) (Bibauw, François, & Desmet, 2015). In order to obtain a better comprehension of their effects on L2 proficiency development, we conducted a multilevel meta-analysis on all the experimental studies measuring an impact of such systems on language learning outcomes (40 publications). Effect sizes for each variable and group under observation were systematically computed (k = 96). By combining all studies into a multilevel linear model, we observed a significant medium effect of dialogue-based CALL on general L2 proficiency development (d = .61). By integrating moderator variables into our statistical model, we are able to provide insights on the relative effectiveness of certain technological and instructional characteristics (spoken vs. written, task-oriented vs. open-ended, form-focused vs. meaning-focused) on different learning outcomes (writing vs. speaking vs. comprehension skills, complexity, accuracy and fluency measures…) and different samples of populations (L2 proficiency, age, context…), as well as to model the effect of treatment duration (number of sessions and time on task) and spacing on these outcomes, to better inform future system and research design.


  • Bibauw, S., François, T., & Desmet, P. (2015). Dialogue-based CALL: an overview of existing research. In F. Helm, L. Bradley, M. Guarda, & S. Thouësny (Eds.), Critical CALL – Proceedings of the 2015 EUROCALL Conference, Padova, Italy (pp. 57–64). Dublin:

5. Elouise Botes (University of Luxembourg), Matthias Stadler (Ludwig-Maximillian University Munich), Juliana Gottschling (University of Luxembourg), and Samuel Greiff (University of Luxembourg): Re-examining foreign language enjoyment and foreign language anxiety through congruence

There has been a recent emergence of positive psychology within studies of motivation in language acquisition. This has led to the development of the construct of Foreign Language Enjoyment (FLE) – a positive state where psychological needs are met in the foreign language classroom (Dewaele & MacIntyre, 2014). FLE is designed as a related, yet independent construct of Foreign Language Anxiety (FLA) and not as two ends of one continuum. This study proposes to examine FLE and FLA through the lens of congruence in order to further the understanding of positive and negative emotions in the foreign language classroom. Adult students learning English in Luxembourg will take part in this empirical study where the effects of FLE and FLA congruence on the outcome variables of Self-Perceived Communication Competence and Willingness to Communicate is examined. Polynomial regression with surface response mapping is used to present a two dimensional space where the effect of the fit between FLE and FLA on Self-Perceived Competence and Willingness to Communicate may be shown to vary nature and magnitude along the lines of congruence and incongruence. This study provides additional insight into the interplay between FLE and FLA in the learning of additional languages.


  • Dewaele, J. M., & MacIntyre, P. D. (2014). The two faces of Janus? Anxiety and enjoyment in the foreign language classroom. Studies in Second Language Learning and Teaching4(2), 237-274.

6. Irene Castellano-Risco (University of Extremadura): Vocabulary learning strategies: Is their selection affected by the L2 languages learnt?

In recent years. the relevance of vocabulary acquisition in the field of Second Language Acquisition (SLA) has grown in importance (Milton, 2009). Some of the research in this field has examined the strategies students use to learn L2 vocabulary in different contexts (Nation, 2001; Schmitt, 1997). However, most studies have focused on the identification of strategies, rather than exploring how their use is determined by learner’s individual factors. This poster aims to explore whether the selection of strategies may be influenced by a number of learners’ individual factors: languages learnt, language learning approach and mother tongue. A total of 138 ninth-year students took part in the study. They were asked to respond a vocabulary learning strategies questionnaire (adapted from Schmitt, 1997). Results seem to point to an expected difference in the selection of the strategies when considering learners’ individual factors. These results have practical applications for second language teachers and materials designers.


  • Milton, J. (2009). Measuring second language vocabulary acquisition. UK: Multilingual Matters.
  • Nation, I. S. P. (2001). Learning vocabulary in another language. Cambridge: Cambridge University Press.
  • Schmitt, N. (1997). Vocabulary learning strategies. In N. Schmitt & M. McCarthy, Vocabulary: Description, acquisition and pedagogy.

7. Xiaobin Chen, Detmar Meurers (University of Tübingen), and Patrick Rebuschat (Lancaster University and University of Tübingen): Promoting the development of L2 complexity

Linguistic complexity analysis has been applied to analyze second language (L2) learning input and learner productions. This research has mainly focused on the separate analysis of either input or production, rather than unifying the two spaces with the complexity construct. Thus it is difficult for language educators to make informed decisions on how to choose individualized input of appropriate complexity levels for learners at different developmental stages, although it is widely acknowledged in second language acquisition literature that such input is indispensable for L2 acquisition. Therefore, intervention studies on the effects of complex input on L2 proficiency development are on demand. Building on previous findings on the relationship between the complexity of learning input and learner production (Chen & Meurers, in press), the purpose of this study is to investigate how input of different challenge levels with regard to the linguistic complexity of the learners’ production would result in the development of their L2 proficiency. A dedicated experimental system that implements the Complex Input Primed Writing (CIPW) task has been developed by the authors. It is capable of automatically selecting appropriate texts at four challenge levels (zero, low, medium, and high) based on the complexity of the learners’ L2 production for continuation writing tasks. We are going to use the challenge levels as treatment conditions and recruit participants for the CIPW tasks. Results of the experiment will help shed light on the relationship between L2 input and production from the perspective of linguistic complexity. It will also offer empirical evidence on the effects of complex input on L2 complexity development, which contributes fundamental insights into language learning and has immediate practical relevance for Intelligent Computer Assisted Language (ICALL) system design.

8. Isabel Margarida Duarte (Universidade do Porto): Representations, perceptions and pedagogical practices of EPE teachers

Recent research confirms the influence of teachers’ beliefs on pedagogical practice (Nespor 1987, Pajares 1992, Calderhead 1996, Fang 1996, Woods, 1996, MacDonald et al., 2001, Andrade, Canha, Martins & Pinho, 2006). Therefore, any intervention in the area of professional development of teachers and innovation of pedagogical practices should start from a study on the professional knowledge and action of teachers. The objective of the present study is to describe the beliefs and practices of the teachers of the EPE network related to Portuguese Heritage Language teaching, in several contexts. For this we will apply two instruments: a questionnaire survey and a semi-structured interview. The questionnaire will be applied to all teachers and coordinates of the network, the interview to a smaller number, being carried out twice with the same teachers (one at the beginning, the other at the end of the school year) and will allow to discuss the representations of these professionals regarding PLH, besides the perceptions about students’ difficulties and pedagogical practices that are better or worse in each context. The data obtained by questionnaire will be submitted to quantitative analysis, the interviews object of content analysis. The results will form a basis for deciding on interventions necessary for pedagogical innovation, teacher training, and construction of didactic material.


  • Andrade, A. I., Canha, M. B., Martins, F. & Pinho, A. S. (2006). As línguas e as suas representações: um estudo sobre experiências de formação de professores. In R. Bizarro & F. Braga (orgs.), Formação de professores de línguas estrangeiras: reflexões, estudos e experiências (pp. 179-191). Porto: Porto Editora.

9. Lurdes Gonçalves (Camões, I.P. & CIDTFF University of Aveiro), Ana Isabel Andrade (University of Aveiro), and Francesco Arcidiacono (HEP-BEJUNE, Switzerland): Heritage Language Teaching: Designing professional development

 Portuguese teachers still have little preparation for teaching Portuguese as Heritage Language (PHL) and only recently special attention has been dedicated to this issue (Cálderón et al, 2013; Melo-Pfeifer, Grosso, 2016; Gonçalves, 2017). Helping teachers in accomplishing the complex task of meeting students’ diverse profiles and expectations implies designing professional development offers (PDO) based on their needs, concerns and specific professional knowledge (PK), which teachers have been constructing in a hybrid professional development (PD) space including PDO in the country of origin, as well as in the host country. This study intends to share a 3-phased PD strategy (preparatory, launching, co-designing) designed and supervised by a pedagogical coordinator, which has been being implemented since 2014 with PHL teachers in Switzerland.  The findings point out to the difference between being active or receptive participant in PDO, and also to the positive impact of active participation on the motivation to invest in PD. This study helps to understand how PDO, which explore and enact the formative potential of the constructed PK and experience, can promote individual and collective professional learning, fostering sustainable change and transformative practice. This study also brings forward the need to assist teachers in the middle of their careers.


  • Calderón, R, Fibbi, R, & Truong, J. (2013). Situation professionnelle et besoins en matière de formation continue des enseignant-e-s des cours de langue et culture d’origine. Neuchâtel: Université de Neuchâtel.
  • Gonçalves, M.L.S. (2017). Tackling sustainability: first steps into co-designing teacher professional development. APPLES – Journal of Applied Language Studies. 11(3), 81-89.
  • Melo-Pfeifer, S., & Grosso, M.J. (2016). Didática do Português Língua de Herança. Lisboa: Lidel, Edições Técnicas Lda.

10. Heiko Holz, Benedikt Beuttler (University of Tübingen), Katharina Brandelik, Jochen Brandelik (Tübinger Institut für Lerntherapie GmbH), and Detmar Meurers (University of Tübingen): Prosodiya – A mobile game for German dyslexic primary-school children. Preliminary results of an RCT

Developmental dyslexia is one of the most frequent learning disorders and affects approximately 4-10 % of the German population. An impaired perception of prosodic features is a strong predictor for developmental dyslexia (Goswami et al., 2013). One of these features is syllable stress. The performance in detecting stress highly correlates with reading and writing skills (Brandelik, 2014) of German primary-school children. One explanation is thought to be found in the association between stress and German orthographic markers – vowel length markers generally occur in stressed syllables. We propose a mobile serious game called “Prosodiya” (Holz et al., 2017) that aims at improving reading and spelling performance of German dyslexic children. The focus of Prosodiya is primarily on spelling acquisition by training the awareness of linguistic features related to syllable stress and linking these features to orthographic regularities of German orthography. We present the preliminary results of a randomized controlled field trial with a waiting control group design in which 129 German 2nd to 4th graders with and without reading and/or spelling deficits trained over the course of 9 weeks with Prosodiya on a tablet. Besides the evaluation of the game’s efficacy, we also investigate its game and user experience.


  • Brandelik, K. (2014). Sprachrhythmische Fähigkeiten im Schriftspracherwerb. Südwestdeutscher Verlag für Hochschulschriften.
  • Goswami, U., Mead, N., Fosker, T., Huss, M., Barnes, L., & Leong, V. (2013). Impaired perception of syllable stress in children with dyslexia: A longitudinal study. Journal of Memory and Language, 69(1), 1–17.
  • Holz, H., Brandelik, K., Brandelik, J., Beuttler, B., Kirsch, A., Heller, J., & Meurers, D. (2017). Prosodiya – A Mobile Game for German Dyslexic Children. In J. Dias, P. A. Santos, & R. C. Veltkamp (Eds.), Games and Learning Alliance: 6th International Conference, GALA 2017, Lisbon, Portugal, December 5–7, 2017, Proceedings (pp. 73–82).

11. Bronson Hui (Michigan State University): Processing variability in novel word learning

 Building upon Solovyeva & DeKeyser (2017), this study aimed to capture the entire trajectory of change in processing variability in both intentional and incidental word learning. This study consisted of two parts: an intentional word learning experiment and a re-analysis of published eye-tracking data from an incidental vocabulary learning study (i.e., Elgort et al., 2017). In the word learning experiment, native English speakers studied 16 Swahili-English word pairs before performing ten testing blocks of animacy judgment tasks on the Swahili words. Results replicated initially more variable processing of the Swahili word meanings. At the same time, processing stability developed after the 6th block, which was then consistent with automatization. Taken together, an inverted U-shaped development was observed. In the second part of the study, I also investigated processing variability in Elgort et al.’s eye-tracking data with individual participants’ first fixations and gaze durations on the target words at each of the first 12 occurrences in natural reading. Results showed that processing variability did not change significantly as participants gained familiarity of the orthographic forms. Discussion addressed the importance to investigate the process of learning, and the signature of processing variability across different stages of vocabulary acquisition in both learning paradigms.

12.  Kathy MinHye Kim (Michigan State University): The interface of explicit and implicit knowledge: A longitudinal study

One of the central issues in second language acquisition (SLA) is the relationship between explicit and implicit knowledge; the so-called, the interface issue. While scholars have taken theoretical stances, empirical studies adducing evidence for these hypotheses are surprisingly limited (e.g., Suzuki & DeKeyser, 2017). Evidentially, lack of consensus on valid knowledge measures and practical challenges in longitudinal research limit extended research. In an attempt to bridge theoretical and empirical gaps in SLA, I investigate the interface of explicit and implicit knowledge by contributing to the field in three aspects: first, validating a battery of explicit and implicit knowledge measures currently employed in our field (e.g., Godfroid, Kim, Hui, & Isbell, 2018); second, with the validity ensured, conducting a one-year longitudinal study to directly observe the long-term attainment of different types of knowledge and their interface; and finally, exploring whether the amount and types of input mediate the developmental trajectory of the changes. A longitudinal structural equation model (LSEM) will be constructed to empirically test the interface issue and mobile app will be used for learners to systematically record the amount and types (e.g., explicit or implicit) of second language (L2) input in a naturalistic L2 setting.


  • Godfroid, A., Kim, K., Hui, B., & Isbell, D. (2018). Validation research on implicit and explicit knowledge: A research synthesis. 2018 European Second Language Acquisition. Münster, Germany. Paper presentation.
  • Suzuki, Y., & DeKeyser, R. (2017). The interface of explicit and implicit knowledge in a second language: Insights from individual differences in cognitive aptitudes. Language Learning67(4), 747-790.

13. Eva Koch (Vrije Universiteit Brussel), Aline Godfroid (Michigan State University), and Alex Housen (Vrije Universiteit Brussel): Measuring the usage of the grammatical information encoded through verb inflection during real-time sentence comprehension: A visual-world eye-tracking study

In L1 comprehension, we constantly integrate lexical, morphosyntactic, pragmatic and contextual information to predict upcoming input (Huettig et al., 2011), which facilitates efficient communication. L2 learners, however, manifest difficulties in exploiting grammatical cues to generate such anticipations (Kaan, 2014), possibly relying more on lexical-semantic information (Hopp, 2015). Our study investigates whether advanced L2 German learners can use the grammatical information provided through verb inflection to anticipate upcoming input. We examine both productive (regular verb conjugation, providing morphosyntactic information through affixation) and unproductive morphology (strong verb conjugation, using stem-vowel alternations in addition to affixation). We compare data of 20 learners of German to those of 20 NS in a visual-world eye-tracking experiment. Participants are exposed to two pictures, varying in the number of referents depicted (singular vs. plural), combined with auditory sentences in which the suffix or the stem-vowel of the verb represent the first cue for number of the referent. Successful exploitation of the cue is measured as anticipatory eye-movements towards the correct picture. The findings may shed light on the extent to which L2 speakers are able to exploit (un)productive morphological features in real-time sentence comprehension, and identify processing differences between L2 and L1 speakers.


  • Hopp, H. (2015). Semantics and morphosyntax in predictive L2 sentence processing. International Review of Applied Linguistics, 53, 277-306.
  • Huettig, F., Rommers, R., & Meyer, A. S. (2011). Using the visual world paradigm to study language processing: A review and critical evaluation. Acta Psychologica, 137(2), 151-171.
  • Kaan, E. (2014). Predictive sentence processing in L2 and L1: What is different? Linguistic Approaches to Bilingualism, 4(2), 257-282.

14. Zeyu Li (University of Muenster): How do adult L2 learners acquire English rhoticity? Perception and production of postvocalic-/r/ by Chinese learners of English

This study explores the acquisition of rhoticity by Chinese learners of English, whose L1 Mandarin has various degrees of rhoticity depending on their places of origin in China. 48 first-year undergraduate students majoring in English participated in this project at two data collection times over six months. Based on perception data collected from an oddity task and speech data elicited from a series of production experiments (picture naming, passage reading, and storytelling), the study reveals the interaction between perception and production of rhoticity for Chinese learners and how this changes over time. It further draws on variationist sociolinguistic theory and controls for linguistic and extra-linguistic constraints conditioning rhoticity using mixed-effects logistic regression models. In particular, it aims at contributing to existing theoretical L2 speech models such as Flege’s Speech Learning Model (SLM; 1995) and Eckman’s Markedness Differential Hypothesis (MDH; 1977). Preliminary results suggest that the phonological factor preceding vowel exerts a significant influence on learners’ perception of rhoticity, with the overall degree of accuracy improved over time. Their production is constrained by both internal (preceding vowel, syllable structure) and external (L1 transfer, speech style, target norms, language attitude, etc.) factors.

15. Fernando Martín-Villena and Cristóbal Lozano (Universidad de Granada): Overproduction in topic-continuity: A corpus study of L1 English-L2 Spanish learners

Previous studies have shown deficits with anaphora resolution (AR) at the syntax-discourse interface in the production of L1 English-L2 Spanish learners. In particular, in topic-continuity (TC) learners produce infelicitous overt pronouns and Noun Phrases (NPs) where a null pronoun is expected. This research analyses the instances of overproduction in TC in L2 Spanish written compositions using a developmental corpus of L2 Spanish compared against a native Spanish subcorpus from CEDEL2 ( Each anaphoric form was tagged using UAM Corpus Tool and was assigned a number of different tags regarding 1) their pragmatic (in)felicity; 2) the syntactic patterns in which they occur; and 3) the distance between the anaphoric form and their textual potential antecedent(s). The overall results reveal that learners overuse overt anaphoric forms where null pronouns are expected (Lozano 2009; Lozano 2016). In particular, advanced learners show nonnative-like behaviour, thus supporting the Interface Hypothesis (Sorace 2011). In addition, the results also reveal that the L1 is a modulating factor: learners use felicitous null pronouns mostly with syntactic coordination from early stages simply because this is possible in their L1 English. This indicates that learners gradually acquire the pragmatic constraints of AR at the syntax-discourse interface.


  • Lozano, C. (2009). Selective deficits at the syntax-discourse interface: Evidence from the CEDEL2 corpus. In: N. Snape, Y-k. I. Leung & M. Sharwood-Smith (eds). Representational Deficits in Second Language Acquisition (pp. 127–166). Amsterdam: John Benjamins.
  • Lozano, C. (2016). Pragmatic principles in anaphora resolution at the syntax-discourse interface: advanced English learners of Spanish in the CEDEL2 corpus. In: M. Alonso Ramos (ed.), Spanish Learner Corpus Research: Current Trends and Future Perspectives (pp. 236–265). Amsterdam: John Benjamins.
  • Sorace, A. (2011). Pinning down the concept of “interface” in bilingualism. Linguistic Approaches to Bilingualism, 1(1), 1–33.

16. Joana Moscoso (Native Scientist), Patrick Rebuschat (Lancaster University & University of Tübingen), and Tatiana Correia (Native Scientist): Strengthening heritage languages in migrants pupils through science outreach

Education is critical for the integration of migrants (to avoid poverty and marginalisation), yet migrant pupils suffer from educational disadvantage. In science and maths, for example, 40% of migrant pupils in EU schools are underachievers, compared to 16% of non-migrant pupils (EU Monitor 2017). In relation to languages, most schools do not provide heritage language teaching, which means that pupils waste the potential of becoming multilingual and put at risk the communication with family members. Through the organisation of ‘Science+Language’ workshops that bring together pupils and scientists with the same heritage language, Native Scientist drives a societal change where i) pupils learn new vocabulary and are prouder of speaking more than one language; ii) girls and boys both think that becoming a scientist is a real viable option; and iii) multiculturalism is accepted and celebrated by all, not marginalised or repressed. Since 2013, work from 150 workshops has reached 3000 pupils. Impact surveys showed that 2 in 3 pupils feel prouder of speaking more than one language while almost 50% of pupils are inspired to become a scientist. Overall, this work helps teachers, pupils and parents to celebrate and value heritage languages, promoting a positive attitude towards diversity.


  • European Commission (2017). Education and Training Monitor 2017. Luxembourg: Publications Office of the European Union.

17. Athina Ntalli, Theodora Alexopoulou, Henriette Hendriks, and Ianthi Tsimpli (University of Cambridge): Age in child L2: the case of finiteness

This study investigates the acquisition of verbal tense and agreement morphology in L2 English by Chinese child learners. Specifically, we aim to identify the potential impact of the age of onset for child learners in an EFL environment. To shed light on the age issue, we looked at two groups of learners with an average of 5 years of instruction but differing in AoO. Specifically, in our Group A we tested 39 9-year-olds with AoO=4 while in our Group B, we tested 34 12-year-olds with AoO=7, all recruited in private language schools in Shanghai. To investigate the acquisition of tense and agreement morphology we used the Test of Early Grammatical Impairment (TEGI, Rice & Wexler, 2001). Results showed that older children are more accurate with both 3rd person agreement and past tense than younger ones, but both groups showed a low performance especially in the agreement condition. We also found different strategies between the two groups which are mainly attested through the use of be+ verb strategy, which seems to signal early sensitivity to inflection. We interpret these results in light of current theories and debates (e.g. Meisel 2009 vs Schwartz’s DAM model, Interpretability Hypothesis vs Hawkins & Liszka, 2003).


  • Rice, L. M., & Wexler, K. (2001). Test of Early Grammatical Impairment. New York: The Psychological Corporation.

18. Doğuş Can Öksüz, Vaclav Brezina, and Patrick Rebuschat (Lancaster University): Exploring L1 and L2 collocational processing in typologically different languages: Evidence from corpus-based and experimental data

We explored the processing of adjective-noun collocations in Turkish and English. Turkish is an agglutinating language with rich morphology, it is therefore valid to ask if agglutinating structure of Turkish affects collocational processing in L1 Turkish, and whether there is influence of agglutinating structure of Turkish on Turkish L1 users’ processing of collocations in L2 English. A contrastive corpus-based analysis using a general corpus of Turkish (TNC; 47 million words) and English (BNC; 112 million words) was conducted prior to the psycholinguistic experiment. The corpus data suggested that the agglutinative structure of Turkish affects both the frequency of occurrence of the collocations, and transitional probabilities between the component words of the collocations. 78 L1 Turkish, advanced level L2 English participants completed an acceptability judgment task, in which adjective-noun collocations were presented as two-word units. 32 of the participants completed the AJT in English, and 46 of them in Turkish. 30 English native speakers also completed the AJT in English. The preliminary analyses of RTs showed evidence that L1 Turkish-L2 English participants processed strong and weak collocations in both languages in a more serial and disjointed way, that is word-by-word. Meanwhile English native speakers processed the collocations more holistically.


  • McCauley, S. M., & Christiansen, M. H. (2017). Computational investigations of multiword chunks in language learning. Topics in Cognitive Science. 9(3), 637-652.

19. Diana Pili-Moss (Lancaster University): The earliest stages of second language learning: A behavioral investigation of long-term memory and age

Two studies, one with 40 L1 Italian 8-9 year-old children and its replication with 36 L1 Italian adults, investigated the role of declarative and procedural learning ability (Decl/Proc) in the early stages of L2 learning. The studies investigated the extent to which memory-related abilities predicted L2 learning of form-meaning mapping between syntax and thematic interpretation and L2 learning of word order and case marking. In the context of a computer board game in incidental instruction conditions, the participants were aurally trained in the artificial language BrocantoJ over three sessions. Standardized memory tasks, vocabulary learning ability (VocLearn), and an alternating serial reaction time task provided measures of visual/verbal declarative and procedural learning ability. Language learning was assessed via a measure of comprehension during practice and a grammaticality judgment test (GJT). Generalized mixed-effects models fitted to both experimental datasets revealed that, during practice, Decl predicted accuracy in both groups, but Proc significantly increased only in children. Moreover, the Proc effect emerged again significantly only in the child GJT dataset. Overall, the findings support age-related differences in the way abilities related to long-term memory predict L2 development.

20. Teresa Quesada and Cristóbal Lozano (University of Granada): Using corpus method to test the position of antecedent strategy in L1 English – L2 learners

The Position of Antecedent Strategy (Carminati, 2002) is a structural parsing strategy proposed for Italian (a null subject language) whereby a null form biases towards an antecedent in a structurally higher (subject) position, whereas an overt pronoun biases towards a non-subject antecedent. The PAS has been extensively studied in Spanish, both native and L2ers (Alonso-Ovalle et al. 2002, Bel et al. 2016, Filiaci et al. 2014). Advanced and near-native learners of Spanish typically show deficits when processing PAS, arguably as a result of their limitations at the syntax-discourse interface, as predicted by the Interface Hypothesis (Sorace 2011). However, most of these studies are experimental and have investigated PAS in decontextualized scenarios. We present evidence from naturalistic data in the CEDEL2 corpus (L1 English-L2 Spanish), as production data offers contextually richer scenarios. A sample of late bilinguals and a control group of natives were annotated following a fine-grained tagset. Results from natives and late bilinguals (at advanced levels) confirm the PAS, revealing that null pronouns show a bias towards subject antecedents. However, overt pronouns show a rather optional bias (subject/non-subject antecedent) and, importantly, NPs show a stronger bias towards antecedents in non-subject position, a finding which has been overlooked in previous experimental research.


  • Alonso-Ovalle, L., Fernández-Solera, S., Frazier, L., & Clifton, C. (2002). Null vs. overt pronouns and the topic-focus articulation in Spanish. Journal of Italian Linguistics, 14(2), 151–169.
  • Bel, A., García-Alcaraz, E., & Rosado, E. (2016a). Reference comprehension and production in bilingual Spanish: The view from null subject languages. In A. A. de la Fuente, E. Valenzuela, & C. Martínez Sanz (Eds.), Language Acquisition Beyond Parameters (pp. 37–70). Amsterdam: John Benjamins.
  • Carminati, M. N. (2002). The processing of Italian subject pronouns. PhD thesis: University of Massachusetts at Amherst.
  • Filiaci, F., Sorace, A., & Carreiras, M. (2014). Anaphoric biases of null and overt subjects in Italian and Spanish: A cross-linguistic comparison. Language, Cognition and Neuroscience, 29(7), 825–843.

21. Hannes Schröter, Theresa Geppert, Josef Schrader (German Institute for Adult Education – Leibniz Centre for Lifelong Learning); Zarah Weiss, Sabrina Galasso, Detmar Meurers (University of Tübingen); Simone Jambor-Fahlen, Waltraud Steinborn, Marco Triulzi, Michael Becker-Mrotzek (University of Cologne): Competence-adaptive, user-oriented search engine for authentic language learning texts (KANSAS)

Adult educators teaching literacy or German as a second language (GSL) typically deal with linguistically and culturally diverse classrooms and/or learners with heterogeneous biographical and educational backgrounds. Furthermore, there is a lack of appropriate textbooks, especially so for adult literacy training courses. Hence, teachers often create their own teaching materials and search the World Wide Web in order to find suitable learning texts. Standard search engines, however, do only consider the content but not the linguistic complexity of texts. The search engine KANSAS is designed to support teachers in adult literacy and GSL in searching suitable German learning texts in the web or electronic corpora. KANSAS will identify specific linguistic constructions and classify the overall linguistic complexity of search results based on the relevant reference frames (Alpha Levels, CEFR). Search results can then be reranked according to the selected complexity level and/or the (de)prioritization of grammatical constructions (see Chinkina, Kannan, & Meurers, 2016, for the search engine FLAIR for English texts). The interdisciplinary project includes the development and test of the identification and classification algorithms. Furthermore, several usability and quasi-experimental studies with literacy and GSL teachers will evaluate the efficacy of the search engine under everyday conditions.


  • Chinkina, M., Kannan, M. & Meurers, D. (2016). Online information retrieval for language learning. In ACL – The Association for Computational Linguistics (Eds.), Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics – System Demonstrations (pp. 7-12). Berlin, Germany. Retrieved from:

22. Laura Stiefenhöfer (Lancaster University): Investigating the relationship between peer interaction and writing processes and outcomes in computer supported collaborative L2 writing

In recent years, an increasing amount of research has been conducted on second language (L2) peer interaction during computer-mediated collaborative writing (CW) tasks (e.g. Elola & Oskoz, 2010; Li & Zhu, 2013). Most of these studies, however, have looked at CW in asynchronous modes of writing, such as Wikis (Li, 2018). This poster will report on ongoing research aiming to investigate I) the dynamics of collaboration in a synchronous Google Docs writing environment and II) the potential relationship between the various dimensions of the CW process (Storch, 2002, Meier et al. 2007) and the quality of the writing product. Employing a two-phase mixed methods research design, data will be collected from approximately 100 advanced EFL learners, who complete a collaborative writing task in Google Docs and interact via the embedded chat function. The resulting texts will be analysed regarding complexity and accuracy, using both quantitative and qualitative measures. In the follow-up phase, a subset of participants will perform an additional writing task and data will be collected by the means of eye-tracking and stimulated recall interviews. Ultimately, the poster will address methodological affordances and challenges of using eye-tracking methodology and triangulating different data types in conducting mixed-methods CW research.

23. Giulia Sulis (Lancaster University): Directed motivational currents in the second language (L2) classroom

Directed Motivational Currents (DMCs) are defined as intense motivational drives which absorb learners and transport them towards the achievement of a personally meaningful goal (Dörnyei, Muir & Ibrahim, 2014). As such, they are seen as a potential device to sustain learners’ motivation and as an ‘optimal form of engagement’ (Dörnyei, Henry & Muir, 2016, p.33). The present study aims at investigating how the motivation and engagement of L2 learners of French and Spanish at a British university fluctuate during four naturally occurring lessons spread across two academic terms, and how motivation and engagement can be sustained over time adopting a DMC perspective. First, each lesson was video-recorded and classroom engagement of students observed. Second, while watching the replay of the video-recording, students were asked to rate their level of motivation at 2.5-minutes intervals. Finally, during individual interviews students explained their rationale for changes in motivation and engagement cued by their charts. Data were analysed qualitatively through a combined inductive and deductive coding approach. Findings suggest that intensely motivated and engaged students displayed features similar to those of DMCs. Examining these characteristics can help determine how to create the optimal conditions for sustained engagement and motivation in the L2 classroom.

24. Anaïs Tack (Université catholique de Louvain & KU Leuven), Thomas François (Université catholique de Louvain), Piet Desmet (KU Leuven), and Cédrick Fairon (Université catholique de Louvain): CEFR-based complex word identification for French and Dutch L2

Vocabulary competence is an important determinant in reading in a foreign language since we know extensive lexical coverage is needed for adequate comprehension. When assessing a text’s readability, it is therefore crucial to deliver correct estimates of vocabulary knowledge. In this view, current research has resorted to the use of annotations of word difficulty by learners to predict which words in a text will be unknown, also known as complex word identification (Yimam et al., 2018). The aim of our research is to further current advances in complex word identification, focusing on French and Dutch as a foreign language (L2). In particular, we examine the use of a CEFR-grounded lexical knowledge base to automatically identify complex words in a text. We employ a method which uses CEFR-graded lexicons for French L2 (François et al., 2014) and for Dutch L2 (Tack et al., 2018). These lexicons describe the frequency distributions of lexemes in a corpus of reading materials graded along the CEFR scale, informing us about what words should be understood a priori at a given level. Our pilot user studies for French indicate that the method achieves moderate effectiveness on word difficulty compared to a more comprehensive statistical model.


  • François, T., Gala, N., Watrin, P., & Fairon, C. (2014). FLELex: a graded lexical resource for French foreign learners. In Proceedings of the 9th International Conference on Language Resources and Evaluation (LREC 2014) (pp. 3766–3773). Reykjavik, Iceland: European Language Resources Association (ELRA).
  • Tack, A., François, T., Desmet, P., & Fairon, C. (2018). NT2Lex: A CEFR-Graded Lexical Resource for Dutch as a Foreign Language Linked to Open Dutch WordNet. In Proceedings of the 13th Workshop on Innovative Use of NLP for Building Educational Applications. New Orleans, Louisiana: Association for Computational Linguistics.
  • Yimam, S. M., Biemann, C., Malmasi, S., Paetzold, G., Specia, L., Štajner, S., Tack, A. & Zampieri, M. (2018). A Report on the Complex Word Identification Shared Task 2018. In Proceedings of the 13th Workshop on Innovative Use of NLP for Building Educational Applications. New Orleans, Louisiana: Association for Computational Linguistics.

25. Nour Elhouda Toumi (Lancaster University): Towards an efficient meta-level processing: The effects of concept mapping and summarization on L2 readers’ comprehension monitoring and metacognitive accuracy

Active reading comprehension depends not only on readers’ decoding and comprehension (cognitive) skills, but also on their ability to evaluate and regulate their comprehension (comprehension monitoring) during reading. An extensive body of research literature exists about the nature, development and contribution of these cognitive processes to first (L1) and second language (L2) reading. To date, however, little is known about comprehension monitoring in the L2 context. Many L2 reading researchers have adopted an exploratory approach in the investigation of this metacognitive ability and there is a lack of converging empirical evidence on the most effective ways to improve readers’ comprehension monitoring when they read in another language. Using a mixed-method intervention research design, this study examines the effects of two types of instructional approaches, concept mapping and summarization, on L2 readers’ self-evaluation and self-regulation processes during reading as well as their accuracy in judging their comprehension level (metacognitive accuracy). This research also investigates how working memory mediates the effects of the two instructional approaches on L2 readers’ comprehension monitoring and metacognitive accuracy. Besides to the use of offline measures, this research uses a technology-supported online instrument (eye tracking technology) to detect the effects of the intervention on readers’ metacognitive abilities.

26. Neil Walker (University of Central Lancashire and University of Lancaster), Padraic Monaghan, and Patrick Rebuschat (University of Lancaster): The distributed practice effect and the explicit and implicit learning of simple and complex grammar and vocabulary

The distributed practice effect is a phenomenon whereby the spacing out of the presentation and/ or practice of to-be-learned items confers greater long-term learning effects than massing them (see Küpper-Tetzal, 2014, for an overview).  Recent studies into the distributed practice effect and second language grammar learning have produced mixed results (Bird, 2010; Suzuki and DeKeyser, 2017).  This poster outlines a proposed study that investigates possible causes of this discrepancy, namely the complexity of the to-be-learned items, whether rules are presented explicitly or incidentally and how the distributed practice effect interacts with declarative and procedural memory systems. Four experimental groups will differ in the length of gap between exposure blocks (either 1 day or 7 days) and the length of time between the final exposure block and the delayed post-test (either 7 days or 28 days). Participants will be exposed to an artificial language with a task to focus on meaning.  The artificial language will contain simple and complex vocabulary and grammar, with some items presented explicitly and others incidentally.


  • Bird, S. (2010). Effects of distributed practice on the acquisition of second language English syntax. Applied Psycholinguistics, 31, 635–650.
  • Küpper-Tetzel, C. E. (2014). Strong effects on weak theoretical grounds: Understanding the distributed practice effect. Zeitschrift für Psychologie/Journal of Psychology, 222, 71-81.
  • Suzuki, Y., & DeKeyser, R. M. (2017). Effects of distributed practice on the proceduralization of morphology. Language Teaching Research, 21, 166–188.

27. Lihua Xia (University of Edinburgh): Task order effects on attention tasks in English and Chinese speakers

 In this study, the order in which different experimental tasks are administered is hypothesized to create a temporary training for individuals in terms of selective and shifting attention. Previously, very few studies explored task order effects on different language speakers. This study aims at examining these effects on visual and auditory attentional conflict and response competing in native English speakers and native Chinese speakers. To explore this issue, we compared the performance of speakers in two widely used tasks: the attentional network task (ANT) and the test of everyday attention (TEA). Two groups of Chinese speakers and two groups of English speakers were recruited who received the two tasks in the ANT-TEA and TEA-ANT orders, respectively. The results suggest that there is an order effect on visual and auditory attention. Specifically, in ANT, the group who received TEA before ANT performed better on alerting; for the TEsubtasks, ANT administered before TEA poses a negative influence on English speakers while the influence is positive on Chinese speakers. Language script could be responsible for the different influece on Chines and English speakers.

28. Deborah Yapp (Hogeschool Leiden): Improving L2 reading skills, an L2 reading strategy programme at Hogeschool Leiden

 For a student today to succeed academically in higher education the ability to read effectively is essential. There is some concern whether students in Dutch higher education are reading less effectively (Chall 2009) and whether Dutch students lack essential second language (L2) skills. Furthermore, some students find L2 texts challenging having had little experience with reading in English (Beeker, 2012). A programme of L2 reading strategies combined with training in metacognitive awareness can help improve L2 reading for struggling students (Macaro & Erler, 2008). This one year reading strategy study at a Dutch polytechnic of 631 first year students from five different departments used a regression discontinuity design (Cook et. al., 2002) with participants functioning as own control. Students received two hours weekly reading strategy instruction for 7 weeks. Results were analyzed using multilevel and contrast analysis and point to the benefits of an L2 reading strategy programme with a mean improvement in reading scores of 4.99 between pretest and posttest (p= .002: ES= 1.25).  Teachers attended weekly reading strategy training sessions for one year and contributing actively to the programme. The level of improvement varied between instructors, calling for more research into fidelity implementation issues of reading interventions.

29. Jorik Fidler (Free University of Brussels): Cognitive multilingual advantage and the role of cognates using the example of German as a foreign language

 The multilingual advantage debate has become a popular research topic. However, less research addresses the question whether or not similarity between two known languages affects the multilingual advantage. This study investigates the influence of Dutch-German cognates on controlled language processing. Monolingual and bilingual Stroop tasks were performed by Dutch participants who can and cannot speak German, French participants who can speak German and German participants. In the German Stroop task, we found additional advantages in congruent as well as in incongruent trials for the two Dutch-speaking groups, which verifies the existence of a cognate facilitation effect, even when participants only know one of the two cognate languages. The results suggest the existence of a cognate warning mechanism in the brain. This mechanism can temporarily inactivate the L2 in an L1-context when the multilingual brain is confronted with cognates.