AAAL 2023: TBLT at AAAL Colloquium

Convened by Andrea Révész & Youjin Kim

Interdisciplinarity and Collaboration in Task-based Language Teaching Research: Advances and Challenges


Andrea Révész, University College London
Youjin Kim, Georgia State University

Colloquium Abstract

The last three decades have seen a growing interest in the role of tasks in second language (L2) teaching and learning, with pedagogic tasks being increasingly promoted and used as a defining or key organizing unit of syllabi in many instructional contexts in the world. The rationale for task-based language teaching (TBLT) is multifaceted, informed by fields such as education, psycholinguistics, educational technology, and second language acquisition. Inevitably, the research domain of task-based language teaching is interdisciplinary, often involving collaboration between task researchers and experts and stakeholders from other disciplines and domains. This colloquium will showcase a series of interdisciplinary TBLT research projects that were based on collaboration among researchers and/or practitioners from various disciplines. First, Gilabert and Malicka will discuss ways in which joint work between TBLT scholars, domain experts and other stakeholders can contribute to task-based needs analysis. The second presentation by Bryfonski et al. will report on a case study that explored a collaborative project involving researchers, novice teachers and school administrators in the design, implementation, and evaluation of an online task-based teacher training program. Next, Jeong et al. will demonstrate how combining insights from neuroscience and cognitive-behavioural TBLT research can further our understanding of the neuro-cognitive processes that underlie task-based oral production. Finally, Van Gorp et al. will describe an effects of instruction study in which a teacher and TBLT and assessment specialists cooperated to evaluate a TBLT program. Our joint goal is to show how interdisciplinary work can facilitate advances in a wide range of TBLT subdomains. Drawing on our own and others' work, we will also discuss the challenges that such collaborative work may pose and consider ways to overcome these. The colloquium will end with an open discussion and reflection with the audience. 

Researcher-stakeholder collaboration in second language needs analysis and its impact on the transfer from NA to task and syllabus design
Roger Gilabert, University of Barcelona,
Aleksandra Malicka, Universitat Oberta de Catalunya,

Abstract:  In the last three decades, tasks have attracted considerable research attention as vehicles for language production, acquisition, and testing. In parallel to those interests, a systematic enquiry into learner needs via Needs Analysis (NA) has long been proposed as a foundation for the design of language learning curricula (Long, 2005; Munby, 1978; Wilkins, 1976), 

To date there has been surprisingly limited theorizing or empirical work investigating two key aspects of NA: the role of the collaboration between researchers and stakeholders, on the one hand, and the transition from needs analysis to syllabus and task design on the other. While there is a large body of empirical research investigating the language needs of particular learner groups through NA, scarce attention has been paid to how domain experts and other stakeholders may be instrumental in guaranteeing the successful retrieval of information during NA, and how knowledge obtained from NA may inform the challenging process of creating a task-based language learning syllabus (see Malicka, Gilabert & Norris, 2017, for an exception, and also Gilabert & Malicka, 2021; Gilabert & Malicka, forthcoming). The current paper aims to shed light on these two important aspects of NA in relation to task and syllabus design.

Our aim in this theoretical paper is two-fold. First, by means of a research synthesis, we aim to systematically analyze how the collaboration between TBLT researchers and stakeholders has been described in NA literature. To this end, we adopt the synthetic technique of “scoping review”, through which we aim to review a full scope of relevant NA research across a range of studies and provide a systematic overview of how domain expert-stakeholder collaboration informs NA. Second, the paper will focus on the role of such collaboration in the transfer of information from NA to task and syllabus design in the areas of task selection, pedagogical design, task sequencing, task-based methodology, language testing and program evaluation.

Designing an online task-based teacher training course: A collaboration between researchers, teachers, and administrators

Lara Bryfonski, Georgetown University,
Amelie Doyle, Bilingual Education for Central America,
Jaira Figueroa, Bilingual Education for Central America, 
Fran Talavera, The British School Rio de Janeiro/Bilingual Education for Central America, 

Abstract:  This case study examines a collaboration between researchers, novice English teachers and bilingual school administrators to design, implement and evaluate an online task-based teacher training program. Growing attention has been paid recently to examinations of TBLT in computer mediated (CMC) contexts, with many pointing out the advantages of CMC for engaging learners in authentic task performances using innovative digital tools across wider geographic areas (Baralt & Morcillo Gómez, 2017; Bryfonski & Cook, forthcoming; González-Lloret & Ortega, 2014; Lai & Li, 2011; Torres & Yanguas, 2021). However, examinations of teachers’ perceptions about the challenges, affordances and effectiveness of task-based online teaching remain scarce. This is a particular issue of interest for the COVID era, where entire programs have to deliver instruction effectively in online modes, including language teacher education programs.

In the case study reported in this presentation, a team of researchers, teachers and administrators developed and evaluated a 4-week, online training program designed for pre-service teachers preparing to teach grades K-9 in English in a network of Honduran bilingual schools. Using the platform Google Classroom, the team designed a series of modules that enabled teachers to navigate course content and collaborate with one another in real time. The modules covered principles of SLA and TBLT, curricular mapping and task-based lesson planning, as well as using tasks to teach content such as math, science, and literacy. The program evaluation included an analysis of teachers’ reactions to the training gathered from pre- and post-training surveys, as well as an analysis of teacher-created task-based lesson plans and their reflections on those lesson plans.

This presentation will provide an overview of how the task-based online training was collaboratively designed and implemented and will conclude with a discussion of the affordances and challenges of multinational, researcher-practitioner partnerships for developing effective task-based teacher education programs.  

Investigating the neural correlates of task complexity during L1 and L2 speech production

Hyeonjeong Jeong, Tohoku University, Japan,
Andrea Révész, University College London,
Shungo Suzuki, Waseda University, Japan,
Haining Cui, Tohoku University, Japan,
Shunsui Mastuura, Tohoku University, Japan,
Kazuya Saito, University College London,
Motoaki Sugiura, Tohoku University,

Abstract:  The role of task complexity has received much attention in the past three decades. Previous studies, however, have solely used cognitive-behavioural and linguistics tools to investigate the effects of task complexity on task-based production. No direct evidence exists about the neural processes associated with task complexity manipulations. Therefore, this study aimed to examine the neural correlates of task complexity, focusing on silent pausing during L1 and L2 speech. Specifically, we examined the extent to which neural processes during pauses may differ depending on whether learners perform less or more conceptually demanding tasks. We hypothesised that greater task complexity would relate to increased activation in conceptualisation-related brain areas. 

The participants were 26 intermediate Japanese students of L2 English, who performed monologic decision-making tasks in English and Japanese (e.g., selecting individuals for vaccination when vaccines are limited). Each participant completed two simple and two complex task versions in each language, which posed lower/higher reasoning demands depending on the difficulty of decision-making. While the participants were speaking, their brain activity was recorded by fMRI scanner. Transcripts of their speech were annotated for silent pauses. Brain imaging data, for the duration of pauses, were analysed by using SPM12 in the conventional two-level analysis. First, we modelled two regressors of simple and complex tasks in each participant as an independent variable. Second, for the group analysis, we tested the effect of task complexity and L1/L2 with one-sample t-tests.

Our preliminary analyses indicate that, as predicted, the left precuneus and left inferior parietal lobule (conceptualisation-related areas) were more active when participants engaged in more cognitively demanding tasks. We will discuss the implications of the study for models of task-based learning and speech production. We will also consider the benefits and challenges of combining linguistics, behavioural, and neuroimaging data. 

Measuring the Effectiveness of TBLT: Longitudinal Evidence from Multiple Assessment Measures

Koen Van Gorp, Michigan State University,
Xinyi Guan, Guangdong Shunde Desheng Primary School,
Matt Coss, Michigan State University, 
Aysen Tuzcu, Michigan State University,

Abstract:  TBLT’s effectiveness has been debated, especially for Asian educational contexts with high-stakes standardized testing and teacher-centered pedagogies. Whereas Bryfonski and McKay’s (2019) meta-analysis provided a strong case for TBLT even in Asian contexts; Boers et al.’s (2021) critical reflection on that meta-analysis concluded that “much more (and more solid and replicable) empirical work on the comparative effectiveness of TBLT needs to be done” (p. 15). 

Responding to Boers et al. ’s (2021) concern, this project collected and compared longitudinal data of student performances in two intact classes in Guangzhou, China: a task-based group (n = 42) with instructor-developed materials and a control group (n = 42), which followed the mandated curriculum. Both groups were given Oxford Placement Tests (OPT) and task-based performance assessments at three time points during a two-semester course. Task-based performance assessments allow for direct assessment of second language (L2) learners’ ability to use their L2 in authentic tasks (Van Gorp & Deygers, 2014), whereas the OPT is a test of general language proficiency. The task-based performances were scored by three raters using an analytic rubric focusing on task completion, comprehensibility, message delivery, and language use, as well as a holistic, comparative judgment of students’ speaking performances. 

This study sits at the intersection of task-based syllabus design and task-based language assessment (TBLA), and was realized as a collaborative effort between a novice English teacher in China, an experienced TBLT and assessment scholar in the U.S., and two Ph.D. students with an interest in and different levels of TBLT and assessment expertise. This presentation will focus on how our diverse team addressed some of the major challenges in measuring the effectiveness of TBLT programs, for example, the relationship between general language ability and task-based language performance tests, and the analytic versus holistic (cf., communicative effectiveness) rating of task performances.

>>>Back to AAAL 2023 Invited Colloquia