SCHEDULE
Conference Schedule
Day 1: Thursday, May 22
Day 2: Friday, May 23
Day 3: Saturday, May 24
| Time | Event |
|---|---|
| 8:00 – 8:15 | Official Opening |
| COLLOQUIUM 1: L2 WRITING. SLA PERSPECTIVES. EFFECTS OF IDS & TASK-RELATED VARIABLES ON L2 TEXT FEATURES | |
| 8:15 – 9:15 |
Keynote 1: Judit Kormos – Task-Mediated Cognitive Model of L2 Writing and Writing to Learn
As second language (L2) writing is a cognitively demanding task, working memory (WM) resources and aptitude can exert a substantial impact on L2 writing processes and the quality of the written product. L2 writers with different cognitive abilities might also benefit from differing pedagogical approaches and tasks. Therefore, it is important to investigate how L2 writing instruction can support students who might be disadvantaged by their lower WM capacity or language aptitude so that they also achieve the required level of L2 writing expertise.
In my plenary I will discuss the role of cognitive individual differences in L2 writing. I will first review recent advances in research findings on the effect of WM and language learning aptitude on L2 writing processes and performance. Next, I will explore how cognitive factors can mediate learning gains when L2 users engage in writing. Using theoretical accounts of cognitive individual differences, I will also discuss how the characteristics of writing tasks can interact with individual variation in WM and language learning aptitude. Building on these areas, I will describe the Task-Mediated Cognitive Model of L2 Writing and Writing to Learn (Kormos, 2023) that provides a hypothetical account of the role of cognitive factors in L2 writing processes and in learning through writing. The talk will conclude with proposing a research agenda for future studies on the role of cognitive individual differences in L2 writing.
|
| 9:15 – 10:15 |
Keynote 2: Mark Johnson – Formal Genre-Specific Knowledge as a Resource-Dispersing Feature of Cognitive Task Complexity: Implications for Task Complexity Theory and Research
Despite early criticisms that task-based theories of language teaching and learning favor the oral productive mode (Manchon, 2014; Tavakoli, 2014), task complexity frameworks—most notably Robinson’s (2011) triadic componential framework—have been enthusiastically adopted in L2 writing research. Results of research seeking to determine the effect of task complexity features on L2 writing performance—most frequently operationalized as syntactic complexity, accuracy, lexical complexity, and fluency—have demonstrated some trends demonstrating an effect of task complexity features on L2 written production (Johnson, 2017). However, the role of genre—a task feature particularly salient to writing—remains underexamined in the research domain. This presentation attempts to situate formal genre-specific knowledge—a component of genre-specific knowledge (Tardy, 2009, 2012; Tardy et al., 2020)—as a resource-dispersing feature of task complexity in Robinson’s triadic componential framework, arguing that abstract genres make unique demands on the working memory capacity of L1 and L2 writers (Kellogg, 1996), demands which in turn affect the fluent, accurate production of complex forms. Theoretical implications are discussed in terms of their repercussions for theories of task-based language teaching and learning and their application to L2 writing research as well as their repercussions for cognitive models of working memory and its role in L1 and L2 written production. Practical research implications are discussed with regard to research design, research methods, metric selection, and analytic methods.
|
| 10:20 – 11:00 |
Independent and interactive effects of cognitive IDs, task complexity, and proficiency on L2 written performance
M. Dolores Mellado, Lena Vasylets & Rosa M. Manchón
The research reported in this paper intended to extend the growing body of research in 2 domains: the effects of individual differences (IDs) on L2 writing processes and products, on the one hand, and task complexity studies in the domain of writing, on the other. More precisely, the study looked into the independent and interactive effects of cognitive IDs (working memory and aptitude), task complexity, and proficiency on L2 written performance. This ultimate aim was motivated by (i) the conflicting and inconclusive results in previous research on whether or not task complexity effects on text features vary as a function of learner IDs; and (ii) the scant research on language aptitude effects in writing, and especially, on how aptitude effects may be moderated by task complexity (Kormos & Trebits, 2012; Vasylets et al, 2022).
The research followed a within-between-participant factorial design, with two levels of task complexity as the within-participant variable, and L2 proficiency, language aptitude (LA), and working memory (WM) as between-participants variables. The outcome measure was L2 writing performance as measured by CAF indices. Two groups of undergraduate students from a degree in English Studies were invited to complete the simple and complex version of the “Fire-Chief” task (Gilabert, 2005). Task complexity was operationalized in terms of reasoning demands and tasks were counterbalanced to avoid unwanted order effects. Participants also completed the Oxford Placement Test, working memory test (n-back test; Kane et al., 2007) and aptitude tests (LLAMA tests). Results indicate (i) lack of WM effects on L2 writing performance but presence of aptitude effects -which were consistent across proficiency levels; (ii) no significant interaction between WM, proficiency, or task complexity; and (iii) variation of aptitude effects as a function of task complexity.
|
| 11:00 – 11:30 | COFFEE BREAK |
| 11:30 – 12:10 |
Testing Kormos’s Task-Mediated Cognitive Model of L2 Writing and Writing to Learn. Empirical evidence. Aitor Garcés, Judit Kormos, Bryan Koronkiewicz, Rosa M. Manchón & Lena Vasylets.
This study is dually motivated by the inconsistent and at times limited findings in previous empirical research on the interaction between individual differences (IDs) and task complexity, as well as by the intention to widen the scope of research in the domain by testing tenets on such interaction of variables as formulated in a recent theoretical model (Kormos, 2023).
From the first perspective, the empirical evidence assembled to date shows that more complex writing tasks induce higher levels of processing and enhance the complexity, accuracy, and fluency of written performance (Cho, 2018; Michel et al., 2019; Révesz et al., 2017; Tabari & Hui, 2024; Vasylets et al., 2017, 2020; Xu et al, 2023; Zalbidea, 2017, 2020, 2021). Yet, contradictory findings exist on whether or not task complexity effects on text features vary as a function of leaner-related variables, including L2 proficiency and cognitive IDs (working memory -WM- and language aptitude -LA). Thus, regarding WM, limited (and at times contradictory) findings exist on the interaction between WM effects and task complexity (e.g., Kormos & Trebits, 2011; Manchón et al, 2023; Zalbidea, 2017), and on the L2 proficiency-dependency of WM effects (Kormos & Sáfár, 2008; Lu, 2015; Vasylets & Marín, 2021). Similarly, regarding aptitude effects, only 4 studies to date (Kormos & Trebits, 2012; Mujtaba et al., 2021; Vasylets et al. 2022; Yang et al.,2019) have inspected LA effects on text features, of which only 2 (Kormos & Trebits, 2012; Vasylets et al, 2022) looked into LA effects as moderated by cognitive task complexity. Further research is hence warranted.
The study additionally attended Manchón and Sanz’s (2023) call to test Kormos’s (2023) Task-mediated cognitive model of L2 writing and writing to learn. The model theorizes the role of WM and LA in language learning through writing, and puts forward novel predictions on potential interactions of WM and LA and task-related and learner-related variables. Task-related variables encompass dimensions of writing tasks subsumed under the umbrella concept of “task environment”, which includes task demands (cognitive, linguistic, and genre-related), time-on-task factors, and the transcribing technology (i.e. paper-based or screen-based writing). Writer-related variables include L2 proficiency and relevant L1 literacy skills. Our study was designed to test the predictions in the model regarding the impact of explicit language aptitude components and working memory on L2 text features, as well as potential moderating effects of task complexity and language proficiency. A group of 47 university-level L1-Spanish L2-English writers participated in the study, where they completed two argumentative writing tasks manipulated for complexity through varying content support (simple vs. complex). For internal validity consideration and generalizability of findings, the design included two different prompts that were counterbalanced across the conditions (+/- complex), mitigating unwanted task repetition and task sequencing (simple-to-complex/complex-to-simple) effects. Tasks prompts were given in the participants’ L1 to avoid the potential influence of language support as a confound. Participants were given 30 minutes to complete their tasks on the computer using Inputlog.
The analysis examined the effects of task complexity and IDs, including L2 proficiency (via the Oxford Test of English), WM (via the Backward Digit Span Test), and LA (via the LLAMA B, D, E, and F tests), on various aspects of written products, such as accuracy (errors per 100 words), lexical diversity (D-value), and syntactic complexity (subordination, complex nominals). The overall MANOVA revealed statistically significant effects for the following predictors: L2 proficiency, V = .27, F(4, 73) = 6.90, p < .001; WM, V = .18, F(4, 73) = 3.93, p = .006; and LLAMA B, V = .17, F(4, 73) = 3.73, p = .008. However, task complexity did not show a significant effect, V = .02, F(4, 73) = 0.39, p = .816. Univariate ANOVA results indicated that task complexity did not significantly impact any of the individual writing measures, while L2 proficiency, WM, and LLAMA B emerged as significant predictors for accuracy and subordination.
|
| 12:10 – 12:50 |
Task complexity effects in L2 writing: How anxiety modulates perceived task difficulty Abbie Finnegan & Cristina Sanz
While research has often compared task complexity (TC) effects in speaking vs. writing (Zalbidea, 2017; Kuiken & Vedder, 2011; Tavakoli, 2014), less attention has been given to L2 writing alone (Manchón, 2014; Manchón & Sanz, 2023; Tavakoli, 2014). Prominent complexity frameworks like the Cognition Hypothesis (Robinson, 2001, 2011) and the Limited Capacity model/Tradeoff Hypothesis (Skehan, 1998, 2009) focus on speaking, while writing models primarily address L1 writing (Hayes, 2012; Kellogg, 1996), leaving gaps in the understanding of how TC affects L2 writing. Importantly for this study, TC’s influence is intertwined with genre (Yoon, 2017) as well as with learner-specific factors, including perceived task difficulty (PTD), motivation, L2 writing anxiety (L2WA), and working memory (Ishikawa, 2011; Gilabert, 2007).
The research reported in this presentation is part of an ongoing study on how the complexity of writing tasks impacts the quality of L2 writing across different genres among adolescent learners, an under-researched demographic in SLA research (Andringa & Godfroid, 2020). The study included 160 participants who completed four writing tasks in two distinct genres across two levels of task complexity in a within-group design. L2 writing was evaluated using the CALF framework, and individual differences evaluated by questionnaires taken from the literature. The report focuses on the relationship between PTD and L2WA in adolescent L2 writers, a population particularly sensitive to cognitive and affective task demands (Duchesne & McMaugh, 2016). Adolescents, undergoing significant physical, emotional, and social changes, may be especially vulnerable to anxiety-driven distortions in task perception, influencing engagement and persistence in L2 writing. While increased TC can enhance engagement (Révész et al., 2016), it may also lead to distorted PTD and heightened L2WA (Rahimi & Zhang, 2019). A small number of studies has provided insights into PTD’s impact on writing performance (Cho, 2018; Robinson, 2001; Tavakoli & Skehan, 2005). And while L2WA, a well-documented individual difference in SLA (Horwitz et al., 1986), influences linguistic performance and cognitive resource allocation, its role in shaping PTD in L2 writing remains underexplored. Some studies suggest that high-anxiety learners perceive tasks as more difficult, even when objective demands remain constant (Eysenck, 1979; MacIntyre & Gardner, 1994). However, little research has examined how anxiety affects PTD in L2 writing, particularly among adolescents. By exploring whether L2WA amplifies PTD in adolescent L2 writers, this study provides insights into the interaction among task design and learner affect, offering pedagogical implications for reducing task-related anxiety and supporting learners through task design.
|
| 12:50 – 13:15 | Discussants: Marije Michel & Andrea Révesz |
| 13:15 – 13:45 | Q&A Session |
| 13:45 – 15:00 | LUNCH BREAK |
| COLLOQUIUM 2: FEEDBACK IN L2 WRITING. SLA PERSPECTIVES | |
| 15:00 – 16:00 |
Keynote 3: Ron Leow – Leow's (2020) Feedback Processing Framework. Implications for ISLA-oriented L2 Writing Theory and Research
The role of feedback (oral or written), and more specifically, corrective feedback (CF), in the L2 learning process has permeated the (I)SLA literature for decades (e.g., Lalande, 1982; Long, 1996). Corrective feedback is a response (whether oral, written, computerized or digital, in the L1 or L2) that is provided by a teacher, a researcher, or a peer in reaction to an error committed by the L2 learner (Leow, 2020). The provision of CF, whether oral and written, has been a pedagogical staple for teachers in the classroom across educational contexts, and has been extended empirically to what Loewen (2015) calls a "systematic manipulation" to L2 learning conditions. Meta-analyses conducted to investigate the pedagogical effectiveness of both oral (e.g., Mackey & Goo, 2007; Li, 2010; Lyster & Saito, 2010; Russell & Spada, 2006) and written feedback (e.g., Kang & Han, 2015; Russell & Spada, 2006) have generally reported an overall medium effect size for its beneficial role in L2 learning. However, most studies in these meta-analyses have relied on L2 learners’ performances on post-exposure writing assessment tasks to infer how some type of CF was processed. Not surprisingly, there is not only theoretical debate as to the usefulness of CF (e.g., Krashen, 1985 and Truscott, 1996 vs. Leow, 2015 and Long, 1996) but also numerous cognitive-based theoretical underpinnings cited for its role (e.g., the Schmidt’s (1990 and elsewhere) Noticing Hypothesis, Swain’s (2005) Output Hypothesis, DeKeyser’s (2015) Skill Acquisition Theory, Bitchener’s (2019, 2021) Model of the Cognitive Processing Stages of a Single Written CF Episode, and Leow’s (202) Feedback Processing Framework. To establish the rationale for Leow’s (2020) Feedback Processing Framework, I will first provide 1) a broad yet succinct overview of the WCF strand of research and 2) a critical review of theoretical underpinnings cited in the WCF strand of research to account for the role of WCF in L2 learning. I will then report 3) Leow’s (2020) Feedback Processing Framework, and premised on the postulations of the Framework, provide implications for 4) ISLA-oriented L2 writing theory, and 5) future ISLA-oriented WCF research.
|
| 16:00 – 16:40 |
Manipulating feedback modality and processing conditions. Effects on revisions. Sophie McBride & Rosa M. Manchón
The study reported in this paper investigated the independent and interactive effects of two types of feedback (electronic and paper-based) and three feedback processing conditions (think aloud, written languaging, and a combination of both) on the texts written and revised by a group of 36 college English learners. The study employed a pre-test/treatment/post-test design over four 50-minute sessions. Participants were randomly assigned totwo feedback treatment groups (18 received electronic feedback and 18 paper-based feedback) and three feedback processing conditions (with 6 participants for each feedback processing condition within each feedback group). All participants completed a picture-based, problem-solving writing task before and after engaging with the feedback. Potential impact of the predictor variables were gauged by changes in complexity, accuracy, and fluency between the original and rewritten texts. Results show (i) differential effects for electronic feedback and paper-based feedback (with the former being more propitious for fluency and the latter leading to increased accuracy); and (ii) an effect of feedback processing conditions (a combination of think-aloud protocols and written languaging emerging as the optimum processing condition for increased accuracy in rewritten texts). These results will be interpreted in the light of previous feedback theory and research.
|
| 16:40 – 17:10 |
Writing and written corrective feedback processing: Does type of writing condition matter? María Laura Zalazar and Ronald P. Leow
Written corrective feedback (WCF) remains a popular strand of research in ISLA. However, studies have mainly addressed the type and scope of WCF and there is a paucity investigating the role of type of writing condition, including collaborative writing (CW) that may have potential to aid L2 learning. Most studies have also employed a product-oriented and non-curricular approach to investigate the learning potential of WCF and CW. Indeed, the question remains whether it is the writing condition or how L2 writers process during such conditions (Leow et al., 2022). To this end, the study is situated within Leow’s (2020) Feedback processing framework and seeks to 1) investigate how deeply Beginner learners of Spanish process, in different writing conditions, unfocused indirect + metalinguistic WCF, 2) examine the relationship between depth of processing and subsequent accuracy of two Spanish linguistic items (gustar and gender agreement), and 3) examine the impact of type of writing condition on the accurate production of the targeted items. Importantly, the study is also situated within the L2 classroom syllabus and curriculum (ISLA applied, see Leow, 2020). Participants will be Beginning writers of Spanish randomly assigned to one of three writing conditions (individual, pair, group) and requested to write a composition with well-designed prompts to elicit the targeted items under the normal curricular writing component. They will receive the usual unfocused WCF + metalinguistic feedback to guide their rewrites. One week later, participants will write a second composition individually. Think aloud protocols will be gathered from the individual L2 writers while the pair and group writing conditions will be visually- and audio-recorded. Depth of processing will be coded based on Leow’s (2015) coding scheme while accuracy will be addressed on the targeted linguistic items produced on the second composition. Pedagogical and curricular ramifications will be discussed.
|
| 17:10 – 17:40 | COFFEE BREAK |
| 17:45 – 18:15 | Discussant: Melissa Bowles |
| 18:15 – 18:45 | Q&A Session |
| 19:00 | Welcome Reception |
| 20:30 | Festival Murcia Tres Culturas |
| Time | Event |
|---|---|
| COLLOQUIUM 3: L2 WRITING. MONOMODAL AND MULTIMODAL PERSPECTIVES. TASK VARIABLES AND EFFECTS ON L2 PROCESSES AND PRODUCTS | |
| 8:30 – 9:30 |
Keynote 4: Matt Kessler – Digital Multimodal Composing: Reflecting on Research Trends and Charting Future Directions
Due to recent advancements in digital technologies, what it means “to write” – whether in a first or a second language (L1/L2) – has fundamentally changed. Specifically, during the past two decades, writing has become increasingly digital and multimodal in nature, often requiring people to communicate by leveraging digital tools to manipulate various modal resources (e.g., text, speech, images, colors, music) (Lim & Kessler, 2023; Tardy, 2005). The shift from monomodal to digital multimodal writing, which is referred to as digital multimodal composing (DMC), has also become commonplace in both academic and professional settings. For instance, in academic contexts, L1 and L2 students are now frequently asked to compose multimodal genres such as digital posters, e-portfolios, lab reports, slideshow presentations, and websites, among others (Lim & Polio, 2020). In turn, these developments have led to an explosion of scholarly activity, in which researchers have investigated a variety of topics pertaining to DMC and L2 teaching and learning (e.g., Hafner & Ho, 2020; Jiang, 2017; Li, 2022; Zhang et al., 2023). In this presentation, I open with a brief discussion of DMC and its relevance to the L2 writing curriculum. Then, I provide an overview of prior empirical research that has been published during the past 20 years. This review is organized thematically, with foci involving explorations of DMC and learners’ writing processes, evidence of L2 learning, teachers’ and students’ perceptions, and DMC’s impact on individual differences. Finally, the presentation closes with a discussion of future research directions. Multiple research tasks are proposed, which are aimed at fostering connections between theory, research, and practice.
|
| 9:30 – 10:15 |
Multimodal viewing-to-write test tasks: English L2 learners’ processing and performance. Tineke Brunfaut and Judit Kormos
Language use is becoming exponentially multimodal, not the least due to technological advancements. Hence, theories of language and meaning-making are expanding beyond linguistic code and language purely as speech/text, towards multichannel events including written, oral, and visual semiotic systems. Language teaching and assessment tasks have similarly been broadened from involving individual to integrated language-skills, but few teaching and test tasks currently reflect the complex multimodality present in real-world domains.
In this presentation, we report on a study that explored ways of representing multimodal communication in second language assessments. Specifically, we designed two types of viewing-to-write test tasks: 1) viewing-to-describe tasks, in which learners watch a recording that orally explains and visually and verbally depicts how something is made, and then they write a text describing the production process; 2) viewing-to-compare-and-contrast tasks, in which learners watch a videocast with two experts discussing a topic, accompanied by visuals, and then the learners write a report comparing and contrasting the experts’ opinions. We administered the test tasks to 134 English-L2 learners (CEFR B1-C1 proficiency). Their performances were marked using a purpose-developed rating scale. Twenty learners additionally participated in post-task interviews, using the viewing inputs as stimuli for recalling their task-completion processes.
To gain insights into the inter-relationships of the constructs of L2 writing and listening operationalised and assessed by the tasks, we ran statistical analyses on the test-task scores and we qualitatively analysed the post-task recall data. This showed that our viewing-to-write test tasks and rating scales elicited, reflected, and evaluated multimodal integrated language use and are practical for testing and developing such abilities of intermediate and advanced level learners. In our talk, we will illustrate these findings with concrete examples from our dataset, with a particular focus on L2 writing processes and implications for multimodal integrated skills development.
|
| 10:20 – 11:00 |
Exploring the effects of multimodal and monomodal pre-task planning on writing behaviours and text quality. Raquel Criado, Aitor Garcés, & Alberto Sánchez
Building on the social semiotic theory of multimodality (Kress, 2010) and the rise of digital communication, research has increasingly examined the affordances of multimodal composing for writing pedagogy and language learning (Kim et al., 2023). However, concerns persist about reduced attention to linguistic aspects when integrating multiple semiotic modes during composition (Manchón, 2017). A possible compromise, framed within a weak version of multimodality (Grapin, 2018), is the use of multimodal pre-task planning before monomodal writing. While research on multimodal pre-task planning is warranted (Lim & Kessler, 2022), studies on monomodal pre-task planning are extensive and have yielded mixed results on CALF measures (Johnson & Abdi Tabari, 2023). Additionally, scholars have emphasized the need to examine the cognitive processes underlying text production following pre-task planning (Wu & Ellis, 2023).
This study investigated the effects of multimodal and monomodal pre-task planning on writing behaviours and linguistic performance in L2 writing. Forty-one undergraduate students, comparable in their typing speed, general L2 English proficiency, and L2 writing competence, completed an expository writing task after engaging in multimodal (n = 14), monomodal (n = 14), or no pre-task planning (n = 13). Writing behaviours were recorded using Inputlog 8.0 (Leijten & Van Waes, 2013). Lexical and syntactic complexity, as well as fluency, were assessed using automated natural language processing tools. Accuracy was manually coded, ensuring high intra- and inter-rater reliability. Writing behaviours were analysed using several indices of fluency and pausing. Results showed that the lexical diversity of texts in both pre-task planning conditions was significantly higher than that of the control group. No other significant effects were observed in the remaining CALF measures or writing behaviours. These findings provide limited support for Kellogg’s Overload Hypothesis (1990) and Skehan’s Trade-off Hypothesis (2009). The study potentially holds theoretical and pedagogical implications for researchers and practitioners.
|
| 11:00 – 11:30 | COFFEE BREAK |
| 11:30 – 12:10 |
Exploring the relationship between lexical complexity and propositional complexity as mediated by pre-task planning. Olena Vasylets & Raquel Criado
Pre-task planning is crucial in second language (L2) writing, influencing the quality and complexity of output (Ellis, 2021). A key yet underexplored factor in writing assessment is the type of measures used. Most studies have focused on holistic assessments or discrete measures of complexity, accuracy, and fluency, with little attention to propositional complexity—the amount of information conveyed, measured by the number of idea units (Ellis & Barkhuizen, 2005). Propositional complexity is vital to discourse complexity, reflecting cognitive effort and planning (Bulté & Housen, 2012). Chafe (1985, 1994) described idea units as meaningful segments writers plan and encode. Examining idea units as a measure of propositional complexity bridges a methodological gap and aligns with the view that planning involves structuring meaningful information.
This study explores how individual and collaborative pre-task planning shape L2 writing performance, focusing on propositional complexity (number and length of idea units). It also examines the relationship between propositional and lexical complexity across planning conditions. An experiment with 66 Spanish secondary students (age = 14.9, SD = 0.28) compared three conditions: individual planning (n = 15), collaborative planning (n = 16), and no planning (n = 15). Essays were analyzed for idea unit number and length, lexical diversity (D-value), accuracy, writing speed, and syntactic complexity (subordination index, nominal complexity). Results showed no significant differences in idea unit number or length across conditions. However, in the individual planning condition, idea unit number correlated positively with lexical diversity, suggesting that individual planning fosters independent and varied idea generation. These findings indicate that while pre-task planning affects idea organization, it does not necessarily increase the amount of information encoded.
|
| 12:10 – 12:50 |
Analyzing young CLIL learners’ collaborative digital multimodal composing processes Florentina Nicolás & Yvette Coyle
Research on digital multimodal composing (DMC) is a fast-growing area of inquiry in second language (L2) (Jiang & Hafner, 2024). Scholars working within a social semiotics framework (Halliday, 1978) have explored the ways in which adult and adolescent learners engage in meaning-making by orchestrating diverse semiotic modes into multimodal “designs” (Cimasko & Shin, 2017; Kim et al., 2021; Shin et al., 2020; Smith et al., 2017). These studies show how DMC can raise learners’ awareness of the affordances of modal resources for different purposes and audiences. Yet, despite the increasingly common presence of DMC practices in primary schools, very few studies have examined how younger learners handle such tasks. Addressing this research gap is crucial since we cannot assume that findings obtained in university writing classes or in multilingual immersion contexts can be routinely extended to child EFL learners whose linguistic and digital competence is still underdeveloped. One recent study with Korean EFL high school students examined the relationship between their collaborative discussions during multimodal composing and the quality of their final products. Students were found to focus on language issues, particularly grammar, while composing, suggesting that they prioritized the linguistic mode over other modal resources (Kim & Kang, 2020). However, it remains to be seen whether these results might be extended to other instructional contexts in which disciplinary content and language learning are integrated in the curriculum (CLIL), and with younger primary school learners. The present study aims to expand available research by analyzing the composing processes young CLIL learners engage in during a collaborative DMC task. In doing so, we intend to shed light on how teachers might support these processes effectively.
The study was conducted in three intact Year 6 classes (n= 65) in an urban bilingual school in Spain. Following their performance on a standardized English language test, nine focal pairs of low, average and high proficiency levels were selected for in-depth analysis of their multimodal composing processes. The children created a presentation using Canva to explain the effects of climate change on the planet. The data collection took place over two days during which time the pairs worked autonomously outside the main classroom under the supervision of the researchers. Data were collected using laptops installed with Open Broadcaster Software to record all screen activity including mouse movements, websites visited, and content created. Audio recorders to capture students’ conversations were also used. The screen capture data were coded by both researchers using an adaptation of Smith et al.’s (2017) multimodal code-meshing timescapes and Kim and Kang’s (2020) interaction episodes. Frequency counts were computed for each proficiency group. The results indicate that the pairs, regardless of proficiency, prioritized the visual over the linguistic mode during DMC, investing more effort in searching for and designing images and customizing the text on the screen than attending to and discussing content and language-related issues. Pedagogical implications for digital multimodal composing in CLIL classrooms with young learners will be discussed.
|
| 12:50 – 13:15 | Discussant: YouJin Kim |
| 13:15 – 13:45 | Q&A Session |
| 13:45 – 15:00 | LUNCH BREAK |
| COLLOQUIUM 4: L2 WRITING. FEEDBACK IN L2 WRITING: MULTIMODALITY PERSPECTIVES | |
| 15:00 – 16:00 |
Keynote 5: Idoia Elola & Ana Oskoz – Multimodal Feedback in L2 Writing. Issues in Research & Practice
Multimodal feedback, the use of different modes such as written, oral and visual semiotic resources for feedback provision, has been increasingly examined in the heritage (HL) and second language (L2) classroom. In recent years, there has also been a significant effort to investigate whether different stages of the composing process and/or final product can be better addressed by multimodal feedback (Bakla, 2020; Ducate & Arnold, 2012; Elola & Oskoz, 2016; Harper et al., 2018), especially in a time when composing is also multimodal as well. Up to now, studies on multimodal feedback have examined, among others, the extent to which instructors have combined oral (screencast-o-matic) and written comments (track changes) for local and global issues (e.g., Harper et al., 2018; Elola & Oskoz, 2016), or the use of visual and oral comments (recordings) for indirect and direct feedback (e.g., Cunningham, 2018; Román Zuñiga, 2024; Valentín-Rivera & Yang, 2021).
To move beyond previous research and practices, in this presentation we will discuss the integration of multimodal feedback in three important areas of investigation: (1) how to provide multimodal feedback to monomodal and/or multimodal texts either during the writing process or to the final products, (2) how to use different tools and their affordances (e.g., screencast software, ChatGPT) to provide feedback, and (3) whether multimodal feedback influences the nature of learners’ revisions. We will also reconsider approaches to multimodal feedback, paying greater attention to the linguistic and nonlinguistic elements of monomodal and multimodal composing as well as its impact on linguistic and non-linguistic revision. We will reflect on research agendas that explore the use of multimodal feedback and provide guidelines that address multimodal ways of providing feedback that are specific to the process and products of our learners’ work and needs.
|
| 16:00 – 16:40 |
Traditional error correction versus screencast video feedback: effects on L2 written accuracy Florentina Nicolás-Conesa, Alba Cánovas, & Lourdes Cerezo
The introduction of real-world technologies in the classroom has provided students with newer and realistic ways to practice and learn the L2 using the language in combination with different semiotic modes such as images, sound, or music in multimodal tasks. Likewise, technology has allowed teachers to use more innovative methods to provide feedback, such as multimodal feedback, which involves a combination of communication modes (e.g., written and oral language). A type of multimodal feedback is screencast feedback (SCF): “digital video-recordings of a [teacher’s] computer’s on-screen activities […] accompanied by a narration recorded while the video is created […]” (Séror, 2012, p. 106). SCF is supported by the dual channel assumption (part of Mayer’s (e.g., 2012) theory of multimedia learning), according to which learning is promoted through two channels and communication modes: the eyes/written language and the ears/oral language, respectively, as opposed to WCF, based exclusively on written language and the visual channel. Empirical evidence on the learning potential of SCF is still limited, with studies focusing on how teachers provide SCF (e.g., Elola & Oskoz, 2016) or on students’ perceptions of SCF (e.g., Cunningham, 2019). There is a need to investigate whether multimodal feedback (i.e., oral and written) provided through screencast technology is more or less effective than traditional WCF to enhance L2 accuracy.
This study compares the effectiveness of multimodal feedback (SCF) versus monomodal feedback (WCF) to improve advanced L2 learners’ accuracy following a pre-test/post-test design. Our participants (N=35) were university EFL students. As part of the study, they were required to write a personal narrative essay. One week later, they were sorted in two groups to receive unfocused indirect teacher feedback using either SCF or traditional WCF. Six days after processing their corresponding types of feedback, the participants rewrote their essays. Two researchers coded errors in the pre-test essays as grammatical, lexical, or mechanical errors as well the number of (un)successful incorporation of learners’ corrections, no incorporation, or deletion of erroneous forms. The number and percentage of errors and corrections in the original texts and rewritten texts were computed for both feedback groups. Non-parametric statistics were conducted. Significant differences were found between the two types of feedback. SCF was superior to WCF in terms of successful correction of global errors. It was also found that the participants who received WCF left significantly more errors uncorrected in their final versions. In terms of specific types of errors, however, WCF was found superior to SCF for the successful correction of grammatical errors. Pedagogical implications about the effectiveness of SCF compared to WCF will be discussed.
|
| 16:40 – 17:10 |
Analyzing learner engagement with monomodal and multimodal corrective feedback by means of screen capture technology Sophie McBride, Florentina Nicolas-Conesa & Lourdes Cerezo.
Corrective feedback in L2 writing instruction has traditionally been provided in the form of written corrections/comments on students’ errors, hence the term written corrective feedback (WCF), using a single mode of communication —namely, written language. Recent technological advances are allowing teachers more innovative approaches to corrective feedback, using multiple modes of communication (typically, visual, auditory, and/or textual elements). This type of feedback is known as multimodal feedback. One type of multimodal feedback is screencast feedback (SCF), which consists of a recording of the instructor’s screen activities and voice while revising students’ texts. Screencast technology can also be used with a more methodological aim in empirical research, e.g., to analyze learner engagement with feedback (e.g., authors, 2024).
This paper thus has a two-fold aim. One aim is to examine the methodological affordances of using screencast technology to analyze L2 writers’ engagement with monomodal and multimodal feedback, as screencasts, often referred to as “digital windows” into learners’ writing processes (Séror, 2013), are said to be able to provide insights into how students interact with feedback. The other aim is to explore how the two feedback modes (monomodal and multimodal) affect learner engagement. We will therefore first describe the types of data obtained through screencast technology and discuss how analyzing these data can enhance our understanding of the linguistic, cognitive, and affective processes L2 learners engage in while responding to different feedback modalities (monomodal and multimodal) and we will then discuss how the tripartite engagement of our participants differed when processing monomodal or multimodal feedback. The paper begins with a brief overview of the aims guiding the empirical study within which our analysis of screencast technology is situated. We then provide a concise rationale behind the study's aims and methodology, including data collection decisions and instruments. The core of the paper focuses on (i) the types of data obtained via screencast technology concerning learner engagement with both multimodal and monomodal feedback, and (ii) the observed differences in student processing between the two feedback modalities (monomodal/multimodal). We conclude with methodological implications for future research aiming to use screencast technology to analyze learner engagement with different feedback modalities and pedagogical implications of using SCF in L2 writing instruction.
|
| 17:10 – 17:30 | COFFEE BREAK |
| 17:30 – 18:10 | Discussant: Melissa Bowles |
| 18:10 – 19:00 | Q&A Session |
| 20:30 | Conference Dinner |
| Time | Event |
|---|---|
| 9:00 – 11:00 | Parallel Sessions |
| 11:00 – 11:30 | COFFEE BREAK |
| 11:30 – 13:00 | Parallel Sessions |
| 13:00 – 13:15 | CONFERENCE CLOSING |