L2 task engagement: A research agenda

Abstract

In this research agenda, we first review the thematic landscape of task engagement research, providing definitions and elaborating on the core theoretical infrastructure for task engagement. We then summarize consensus perspectives from this body of work and identify important contributions that task engagement research stands to make to second language (L2) learning and teaching research. Following this, we outline five key research tasks that we believe will broaden the field’s understanding of task engagement, sharpen insights from empirical work, and accelerate the contribution of this research. Our goals are, first, to highlight for readers the shared understandings that exist in this important area of language learning research and, second, to draw attention to specific areas where additional L2 task engagement research is needed to push the field forward productively.

1. Introduction

1.1. The landscape of L2 task engagement: Definitions, theoretical infrastructure, and research contributions

Task engagement is a construct that relates to how individuals focus on, interact within, and learn from a specific task or series of tasks both in and outside of the language classroom. It provides researchers interested in instructional design, learner activity, and L2 task performance and production with insights into the amount, intensity, and quality of learners’ integrated cognitive, social, emotional, and agentic effort and activity in service of completing a task (Hiver & Wu, 2023). It will come as no surprise that a grounding theoretical framework for task engagement research is the interactionist approach and its accompanying insights from task-based language teaching (TBLT) research (Philp & Duchesne, 2016). The point of departure in this framework is that active learner involvement, whether through L2 interaction or other socially embedded and cognitively oriented means of participation, is a driver of L2 learning (Mackey, 2020).

Multiple theoretical perspectives are useful for looking at task engagement, and though it is the most prominent, task engagement is not limited to a cognitive-interactionist perspective. Related theoretical frameworks for language development, such as sociocultural theory and complex dynamic systems perspectives, also see the learner and their engagement with various affordances for learning (e.g. context, input, formal linguistic information, feedback, interlocutors, and technology) through improvisation, appropriation, and co-regulation, as central to the learners’ growth (e.g. Behney & Gass, 2021). By framing the learner as the locus of learning, and tasks as affordances for learning that are responsive to learners’ needs and purposes for using the language, task-based pedagogy retains the core idea that learning is an active and participatory process where learners must engage directly in and with the subject (Long, 2015; Robinson, 2011). Central to the thematic area of task engagement are the notions of task and engagement, each of which we unpack in turn.

Tasks are the object of task engagement and the granular unit in which that engagement unfolds. Accordingly, the way in which tasks are conceptualized matters (Erlam & Tolosa, 2022; Teng, 2025). There is, of course, considerable variation around definitions, shared features, and conceptualizations of tasks in the field (e.g. Ellis, 2018; González-Lloret & Ortega, 2014; Long, 2016; Samuda & Bygate, 2008; Skehan, 2016; Willis & Willis, 2013). But, despite concerns about the ambiguous and, at times, uncritical use of the term task (Ellis, 2017), there is also some consensus regarding what this term refers to in the context of L2 learning and teaching and what key characteristics must be satisfied for a learning activity to be labeled atask. One useful distinction is between the workplan itself and the process that ensues. Ellis et al. (2020, p. 10) propose, following Breen (1989), that tasks are workplans specified in terms of their design features intended to create a context and provide opportunities for meaningful language use regardless of the modality. Various processes resulting from a task such as learners’ language use, their interaction, task performance, and learning are meaningfully distinct from definitions of the task itself. Following Ellis et al. (2020), the features that are most often part of the theoretical rationale for tasks as a unit of learning and analysis include the following (see also Long, 2015; Willis & Willis, 2013):

• • tasks have a primary focus on meaning such that learners are concerned mainly with encoding (through output and interaction) and decoding (through input) those meanings (e.g. making plans using given information);

• • tasks involve a non-linguistic goal or outcome that learners set out to achieve through task participation (e.g. making a decision, solving a problem);

• • tasks feature a gap that creates a need or authentic purpose (e.g. to convey information, to reason, or to express an opinion) for learners to communicate in the target language;

• • tasks provide the opportunity for learners to rely on their own (para)linguistic resources in attempting to meet that outcome and bridge that gap (i.e. using multimodal communication and comprehension strategies).

Central to this thematic area is the notion of a task as both the object of engagement and a provider of affordances for task engagement at different scales of activity (Jackson, 2025b; Pusey, 2025). That is, task engagement does not emerge ex nihilo; rather it is ‘the availability in the context of “things to talk about” and the availability of resources to engage with them [that] stimulate[s] further action’ (van Lier, 2004, p. 81).

Engagement is often thought of as energy in action (Lawson & Lawson, 2013; Skinner & Raine, 2022). Accordingly, task engagement is learners’ participatory action or involvement in service of completing a task or series of tasks (Philp & Duchesne, 2016). The engagement part of the equation refers, equally, to the integrated activity and mental effort that learners deliberately expend through their participation in a task. More specifically, tasks provide learners with the opportunity to engage behaviourally, by adopting behaviours that drive a task forward to successful completion; cognitively, by attending to the quality of interaction, the task demands, and their on-task performance (e.g. in negotiating meaning or attending to the formal features of language); affectively, by monitoring and regulating emotions aroused during task involvement; and socially, through collaborative, relational moves, interacting with and supporting others (e.g. negotiating with and scaffolding a peer). Tasks may also offer opportunities for learners to contribute to and shape the conditions of their personal task experience in ways that better support their motivation and learning. This proactive involvement with tasks, or agentic engagement, has received relatively little empirical attention but is increasingly recognized as an important dimension of learner engagement (Reeve & Jang, 2022).

Although an obvious dimension of task engagement relates to learner action, it is also clear that at least some degree of cognitive attention and information processing takes place during L2 task performance (e.g. N. C. Ellis, 2019; Leow, 2015, 2019). Task engagement should, therefore, be understood as the degree to which individuals invest the integrated physical, cognitive, and emotional energies at their disposal in a specific L2 task (Hiver & Wu, 2023). Clearly, task engagement involves more than just ‘doing’ or ‘completing’ a task, so this added dimension of proactive intention and focused participation distinguishes genuine engagement from mere task compliance (Gijsen, 2021; Newton et al., 2020). It is, therefore, possible to think of degrees of engagement and qualitatively different profiles of task engagement which, as we outline below, merits specific research focus. As with tasks, the way we define and look at the engagement dimension affects how we measure it, and this has implications for how we research it and ultimately help teachers and learners work on it (Jackson, 2025b).

Early scholarship positioned task engagement as related to the quantity of L2 learners’ on-task behaviour and their actual language output (Dörnyei & Kormos, 2000), as well as the ways in which L2 learners marshal ‘appropriate [language] resources in specific situations’ (Platt & Brooks, 2002, p. 369) and construct meaning by strategically adopting ‘tools and practices’ (p. 372) for problem solving (see also Bygate & Samuda, 2009). Building on this seminal work and drawing on educational psychology, Philp and Duchesne (2016) published an influential paper on L2 task engagement in which they emphasized that TBLT research had focused a great deal of attention on how tasks were conceptualized, designed, sequenced, implemented, and assessed, but that ultimately it is what L2 learners do within the task (i.e. their task engagement) that matters for their learning (see also Ellis, 2024; Ellis et al., 2020; Lambert et al., 2023a for similar arguments). Research on task engagement has since increased considerably, focusing on the optimal conditions for learners’ engagement and how these differ across individuals and tasks (Lambert, 2017; Lambert et al., 2023a). A shared characteristic of many of these studies is that they focus heavily on the relationship between task design, task implementation, and learners’ task engagement. Recent synthetic work has now begun to take stock of cumulative evidence on this topic (Namkung & Kim, 2024), and higher-level publications have also distilled strategies for enhancing learners’ engagement in tasks across different learning modalities (Dao, 2024; Egbert & Panday-Shukla, 2024). From this perspective, the consensus is that language learners’ deeper involvement directed to a task, without discounting the value of incidental learning conditions and implicit processing mechanisms (Loewen, 2025; Long, 2015), can lead to the creation of more learning opportunities. That is, L2 learners’ use of the target language in interactive settings and the learning opportunities that arise during interaction (e.g. receiving modified input, hypothesis testing, negotiation of meaning, attending to features of the task and the language) are key to successful language learning (Swain, 1998; Swain & Lapkin, 1998). In other words, learners’ task engagement is a prerequisite for most successful learning through tasks.

Related frameworks for examining language learner involvement with formal features of the target language are engagement with language (EWL; Svalberg, 2009) and engagement with corrective feedback (Ellis, 2010). Both of these can occur within the context of task-based learning. Engagement with language places focus on explicit knowledge about language both in the process of learning and as the outcome of learning (Svalberg, 2012, 2018). This has been linked to task-based learning, for instance, when learners complete tasks requiring them to think and talk about language, its forms, functions, and how it works (Baralt et al., 2016; Carver et al., 2021). Engagement with corrective feedback, for its part, extends evidence that corrective feedback (CF) can effectively draw learners’ attention to mismatches in their own language use while performing a task. Research shows that the most influential aspects of CF relate to what learners actually do with the feedback they receive rather than exclusively to the type, timing, and provider of that CF (Han & Gao, 2021; Shen & Chong, 2023). Research in TBLT also shows that learners differ in how they seek feedback, select what feedback to attend to, how they perceive, process, and respond to it, and the extent to which they use it to restructure their L2 knowledge and increase their control and accurate use of the target forms (Ellis et al., 2020).

Having laid out some of the underlying theoretical foundations, we now turn to examining the contribution of this thematic area for L2 learning and teaching research. In all areas of language education, the premise is application by design: researchers and practitioners alike are interested in what works best and what matters most (Hiver & Nagle, 2024; Sato et al., 2025). Instructed language learning research, for example, explores how deliberately and selectively manipulating instructional conditions, including tasks, can influence learning (Loewen, 2025). Decades of research show that one of the greatest sources of variance in learning relates to learners themselves and what they bring to the learning process as individuals (Gurzynski-Weiss, 2017; Li et al., 2022). There are, thus, large and well-developed bodies of research on L2 learning processes and outcomes, on language instruction, and on language learners (see Fig. 1 for examples) (see also Gurzynski-Weiss, 2024). Although there is occasionally some relevance between them and links in their research designs, for the most part these important thematic domains of research tend to exist and develop separately and in parallel to each other (Dörnyei, 2009). Scholars problematizing persistent thematic and theoretical silos in the field have observed that research to date in language instruction examines questions about the nature and process of language learning, and independently from that focuses on language learners to establish the reasons for differential levels of success (Larsen-Freeman, 2015; Larsen‐Freeman, 2018; Sato et al., 2025; Zheng et al., 2025). Theory and research in the field tends to treat the learner as an afterthought that can provide qualifications and caveats to the applicability of a theory or the utility of pedagogy.

Figure 1. Siloed lines of L2 research.

We should also acknowledge that there have been attempts in TBLT research to examine language learners by linking, for example, their cognitive individual differences (IDs) to various instructional techniques and L2 learning processes, as well as their roles in relation to instructional interventions (see e.g. Révész, 2011, 2012, 2014). However, this line of cognitive IDs research does not appear to explicitly address the integration of these three key domains (i.e. language learners, L2 instruction, and learning processes/outcomes). Instead, it largely focuses on exploring learners’ IDs as fixed, trait-like features that serve as optional add-ons or as mediators of instructional effects, rather than as central elements of instructional design (see Hiver & Nagle, 2024, for a detailed discussion of the limitations of this approach and the remaining challenges in advancing the field). More importantly, there remains a notable gap in research that seeks to connect these domains specifically in the context of task engagement, where the goal is to place the language learner at the center of task design and implementation within specific spatiotemporal contexts (see Fig. 2). In looking for a way forward, a central question is how to address this fragmentation and achieve synergies with meaningful coordination and overlap between all three areas (Michel et al., 2025; Papi & Hiver, 2025). Notably, attempts have been made to do just that, especially in TBLT research, by using tasks as a focal point to integrate various thematic areas of research (e.g. language instruction, learners, and L2 learning processes). In this approach, tasks serve as a point of departure for linking and understanding (a) learner factors, (b) syllabus content, (c) classroom second language acquisition, and (d) the measurement of outcomes. While this task-centered approach offers valuable insights into the relationships among these aspects (e.g. Li, 2024), as we argue above, it is the learners, rather than the tasks, who are a source that introduces great variation in task performance, L2 learning and teaching processes, and language instruction outcomes. Thus, recentering on the learner (i.e. learners’ task engagement) as a starting point offers a promising alternative avenue for addressing these siloed lines of inquiry, as depicted in Fig. 1.

Figure 2. Task engagement renews focus on tasks, learners, and context.

The premise of task engagement research is that it centers the individual learner and their learning activity in the context and parameters of a task, and, further, that a task entails certain learning activity and language use within a particular classroom context with some specific instructional features (see Fig. 2). Consequently, task engagement is a natural way of integrating or embedding the individual (who) in research on processes of learning (what, how, when, where) to inform pedagogy (how, why). In their conceptual overview, Philp and Duchesne (2016) cautioned that task engagement varies depending on the task, the participants, and the setting. Recent empirical work based on their framing provides clues for how task engagement research can integrate learners, learning processes, and instruction with a renewed focus on three dimensions: learner, task, and time (Aubrey, 2022; Dao & Sato, 2021; Henry & Thorsen, 2020). First, task engagement manifests differently for different learners, and learners determine their own engagement (Berry, 2022). Task engagement is an embodied construct ‘done’ by a learner. This means that differences in the amount, type, and intensity of task engagement and the resulting learning are inherently tied to the agent/learner doing the task. Second, task engagement is task-specific and context-dependent which means that tasks themselves impact engagement. Characteristics of task design and task implementation have an important influence on generating and sustaining task engagement and the resulting learning. Finally, task engagement unfolds as a process and emerges in the context of the L2 interaction or task performance. Task engagement is a momentary and granular phenomenon (Symonds et al., 2024) that requires a more targeted, micro-level time frame to tap into the momentary lived experience of learning activity. Task engagement also allows researchers to investigate all the language skills and modalities. But, unless the individual is a central part of the unit of study in task engagement, what results is a somewhat artificial separation of issues with engagement and learning behaviour clearly in focus, and the learner as an afterthought to those abstract processes (Hiver, 2022). Task engagement is grounded in learners’ present learning behaviour and activity, and it is oriented to war future growth and development. It is, therefore, one way of ending the previously mentioned disjointed research agenda by examining questions about the nature and process of language learning with the learner and their learning behaviour as the point of departure (see Fig. 3) (see also Gurzynski-Weiss, 2024).

Figure 3. Task engagement integrates learners, learning processes, and instruction.

In addition to providing a natural way to unify existing research on language learners with their L2 learning processes and language instruction, task engagement research helps to inform teachers in several important ways (Dao et al., 2021; Nguyen et al., 2023). Task engagement research expands teachers’ understanding of individuals’ learning processes and how learners interact with real time learning activities (see e.g. Jackson, 2021). It can also reveal potential barriers to learner participation, enabling more tailored instructional approaches that help teachers maintain learners’ focus in diverse classroom settings (Berry, 2022; Erlam & Tolosa, 2022). Such insights are useful both for initial teacher preparation, where participants’ learning is embedded in authentic teaching contexts and knowledge is constructed through practical experience, and for continuing teacher education, where critical assessment, reflection on classroom interactions, and continuous improvement are the goals (Dao, 2024; Dao et al., 2021).

1.2. Overview of research tasks

In this research agenda piece, we highlight five core research tasks, each of which presents various objectives and avenues with potential to clarify and consolidate the field’s collective understanding in this area. We introduce these briefly here before discussing each in detail below.

Research task 1 relates to understanding what the various levels, dimensions, and scales of task engagement are and how they are linked and interact. A small body of work unpacks definitions of task engagement, and the dominant models propose that task engagement is multidimensional in nature. Yet, beyond these multicomponential descriptions, there is little work that explores the substantive links or the implications of such relationships. Ultimately, by seeing task engagement as more of a system than an inert construct, there is also considerable work to be done on the context–motivation–engagement–learning cycle, which will help to make better sense of task engagement and its proximal and distal antecedents and outcomes for L2 learning.

Research task 2 relates to measurement. Once explicit definitions and operationalizations exist, measurements are needed that match these. A cumulative body of evidence is built on a foundation of coherent and precise concepts and measurement. Still, many initial questions remain concerning the conceptualization of task engagement and the accompanying operational issues. Greater clarity is needed about the object of task engagement as it manifests through communication and in collaboration across varied modalities, through various tools and affordances, with peers, with language features in multiple domains, and with the content and instructional material.

Research task 3 relates to individuals, their role in learning groups, and how task engagement unfolds over time and in different task contexts. Arguably, the learner is the key to understanding task engagement. Yet the role of the learner and how they influence task engagement at the individual and group level, at various momentary and continuous timescales, within various tasks, have almost been entirely neglected in task engagement research.

Research task 4 relates to task design and implementation. Task design and implementation features have been the focus of task engagement research. Task engagement research emphasizes the importance of task design and task implementation that meets learners’ needs and generates personal investment. However, continued work is needed to focus further on the necessary and high-leverage task conditions for engagement and the interaction between learners themselves and tasks. There are also, undoubtedly, other important features of tasks that have not yet been explored that deserve attention.

Research task 5 relates task engagement to instruction and to teacher education. Current task engagement research directs significant attention to task designs and, to a lesser extent, the resulting influences on learner involvement. While the central focus in task engagement research has been learners and their learning, given the renewed urgency of bridging the research-practice interface (Hwang et al., 2024; Sato & Loewen, 2022), there is an important role for teachers in this domain that can inform how they pursue their professional and instructional goals.

On balance, there is a good deal that is known about task engagement, and yet it is also clear that some of the most intriguing questions in this domain remain unexamined or underexplored. We turn now to an in-depth description of these five major research tasks.

2. Research tasks

2.1. Task 1 – Conceptualization, links, and interactions

Investigate the context–motivation–engagement–learning cycle using L2 tasks.

Since Philp and Duchesne’s (2016) landmark paper, many studies in the field have adopted a multidimensional perspective of L2 task engagement comprizing three or more dimensions (e.g. behavioural, cognitive, affective, etc.). Conceptually, this viewpoint has a solid rationale and is based on established work in educational psychology and the learning sciences (Reschly & Christenson, 2022). However, the preliminary psychometric and construct validation work that exists presents a mixed picture: task engagement is multidimensional in theory but not always in practice (Dao et al., 2025; Hiver et al., 2024). Learners and teachers also tend to see the proposed dimensions more as a unified phenomenon (e.g. Aubrey, 2022), and this lends support to calls to treat task engagement as a meta-construct that emerges only when these different categories combine (e.g. Mercer, 2019; Oga-Baldwin, 2019; Zhou et al., 2021). This research task is, therefore, intended to better understand the different dimensions of task engagement and to explore whether and how these are linked and interact. Before going further into the objectives of this research task, it is important to note here that when examining how the dimensions of task engagement interact and connect, it is possible to task engagement both as having trait-like characteristics and as a state-like interactionist construct. Engagement can show some stability across time and contexts, and there may be some individuals who consistently demonstrate higher levels of engagement in learning activities regardless of the specific learning targets or environment. This trait-like aspect is often captured through self-reports, while the state-like interactionist dimension is more accurately measured using direct observational measures such as discourse analytic methods (Hiver & Dao, 2025a) (see Research task 2 for further discussion on the measurement of task engagement). This dual perspective offers a more comprehensive and inclusive understanding of task engagement and can address the limitations of research that considers it solely as either trait-like (e.g. Vo, 2024) or as a state-like interactionist construct. Such one-sided approaches often dictate the choice of measurement, that is, favoring either subjective self-reports or direct observational tools exclusively. By adopting both perspectives, our approach provides a more holistic and nuanced account of task engagement.

A primary objective of this first research task will be to clarify for our field what the various dimensions of task engagement are and what they entail in concrete terms. For instance, what is behavioural engagement in the context of a range of language learning tasks, and how is that substantively different from cognitive or social engagement? Pursuing this objective will provide insight into whether any of the core domains have primary importance, and which domains are necessary and/or sufficient for explaining learners’ deliberate involvement in language learning tasks. An example project would be to analyze a collection of L2 tasks being implemented in a course, with clear task characteristics (i.e. some output-based, some input-based; some focused tasks, some purely interactive tasks), and devise a detailed coding scheme to identify specific indicators for cognitive, affective, social engagement, and so forth, that might be expected to arise as learners undertake those tasks. This could also be achieved by piloting tasks and collecting think-aloud protocols to capture such conceptual categories in real-task terms.

A productive starting point for this research task is to take a more expansive stance and look at task engagement as more of a system with multiple, interacting parts than as an inert construct or ID variable (see also Research task 1 for additional discussion). For many, engagement exists on a comparable level to other meta-constructs such as emotions, motivation, or beliefs, rather than at a level of discrete emotions, such as enjoyment or discrete beliefs, such as attributions. At the task level, engagement can be treated as equivalent in specificity to these meta-constructs (e.g. task motivation). Accordingly, it can be understood not only as an outcome variable, as commonly conceptualized in existing L2 research, but also as an independent variable that influences or leads to differential effects on various dependent variables such as interactional behaviours, satisfaction, and learning outcomes (see Hiver & Dao, 2025a; 2025b; Ji et al., 2023; Wang et al., 2020). But while various models and theories exist to account for many other meta-constructs, including well-known frameworks such as goal theories and identity theories of motivation, and dimensional or developmental perspectives of emotion, these do not exist for engagement. Instead, the complexity involved in deciding on the appropriate conceptual level of engagement relates to the diverse components it has. One template for treating task engagement as an integral part of a larger network is a relation-intensive research design (Hilpert & Marchand, 2018) that investigates the links between various dimensions and interrogates what the necessary and sufficient dimensions, or combinations of dimensions, of task engagement are. For example, research using network analysis or directed acyclic graphs could map the dynamic relationships between different engagement dimensions during collaborative tasks and examine how changes in one dimension influence others in real time.

Beyond multicomponential descriptions of task engagement, work is also needed to explore ‘engagement paradoxes’ (Pusey, 2025) where there are trade-offs between different dimensions (e.g. +affective -cognitive/social engagement; -affective +cognitive/social) and ‘engagement gaps’ where the presence of one dimension (e.g. positive affect in the form of enjoyment and relaxation) does not materialize into genuine task engagement (i.e. it is accompanied by a lack of attention and focus or effort). For example, case studies could investigate engagement paradoxes by following individual learners who show high affective engagement with low cognitive engagement (or vice versa) with specific tasks to understand how these trade-offs develop and what triggers shifts between different engagement profiles across different tasks. These dilemmas require careful theorizing and coherent conceptual thinking but can also be seen as important empirical questions. There are, therefore, several ways to accomplish this research task that involve focusing specific attention on elements of the context–motivation–engagement–learning cycle within and beyond classroom settings.

Learning activity takes place in particular contexts, and studies are needed to link these instructional contexts (e.g. online L2 course delivery) to task engagement. Tasks themselves also provide affordances for using, interacting with, and attending to language (van Lier, 2004). The tangible and situated learning activity that most scholars are interested in emerges in this context. Although a task is a more bounded ‘context’ than something like a language classroom or a virtual learning environment, additional work is needed to understand how these spatiotemporal parameters provide the necessary social and psychological conditions for task engagement. Learners also bring with them various goals, affective dispositions, and cognitive and non-cognitive assets that influence their involvement in that task. For instance, task motivation and task emotions can both be thought of as precursors to learners’ subsequent behaviour on tasks, as well as the by-products that arise out of engaging in a task. Thoughtful study design must include such precursors and explicitly investigate how the initial energy and intention (task motivation) that catalyzes and directs task engagement impacts learners’ performance in L2 tasks and, by extension, how task engagement feeds back into those. For example, comparing task engagement in similar tasks across different delivery modalities (face-to-face vs. online synchronous vs. asynchronous) can help illuminate how specific task affordances create opportunities for language use and how that context shapes the motivation-engagement relationship.

Another objective, considering the underlying interactionist and sociocultural perspectives that inform L2 tasks as a unit of learning, is to address what learning is and what it looks like in the context of L2 tasks, independent from task engagement and/or performance. Engagement, by definition, creates opportunities for successful learning and is the medium through which learning occurs. ‘Learning’ itself can be defined and operationalized in many task-specific ways (Ellis et al., 2020; Larsen-Freeman, 2017; Leow, 2015). However, the notion of learning as a by-product of task engagement has still not garnered much research. Studies could use (quasi-)experimental or case-study designs to investigate whether and how incidental learning occurs in unfocused tasks as well as the patterns of learning particular focal items, structures, and information (e.g. discourse, grammatical, lexical, phonological) that result from engaging in form-focused tasks (see e.g. Lambert et al., 2023b; Qahl & Lambert, 2025). Studying relationships and interactions between all these facets will help the field make better sense of relevant proximal and distal antecedents and outcomes of task engagement.

A final objective for this research task is to feature prominent links between task engagement and L2 learning activity more broadly. There are, of course, many other complementary levels of engagement that deserve additional research. Engagement, at these different timescales and levels of specificity, is directed toward different targets and purposes than task engagement, and they may be hierarchical in nature, with the more micro forms and levels of engagement nested within more macro areas (Pusey, 2025). Incorporating task engagement into study designs that examine engagement with the object of language itself (Han & Gao, 2021; Svalberg, 2018), engagement in projects and in online learning spaces (Dao, 2024), extramural engagement in naturalistic settings (Papi & Hiver, 2025), engagement with language institutions and with communities of language users (Sundqvist, 2024; De Wilde, 2024) will contribute greatly to the field. For example, we might imagine a community-based research project that follows learners from classroom task engagement to their engagement in local heritage language communities, using methods to analyze and capture how task-based engagement transfers to authentic community participation and vice versa.

2.2. Task 2 – Measurement

Develop rigorous, subjective, indirect measures and valid direct, observational measures of task engagement.

As outlined above, significant challenges in conceptualizing and measuring task engagement still confront most scholars who seek to research this domain. Once clear and explicit definitions and operationalizations of task engagement are available, measurements that match these are needed. The first objective for this research task would be to develop and validate skill- and domain-specific measures of task engagement that incorporate both verbal and non-verbal measures. Systematic reviews in this area (e.g. Dao et al., 2025; Hiver et al., 2024) show that most research has examined task engagement in L2 interaction and L2 production (e.g. oral L2 tasks, L2 writing tasks), but it is still unclear how task engagement should be measured in receptive language domains (i.e. input-based tasks in the reading and listening modalities) or for specific language systems (e.g. discourse, vocabulary, speech perception and production) in focused tasks. Interested researchers could pursue this task by designing and validating multi-modal measurement protocols with fine-grained coding systems to capture on-task and off-task indicators for input-based and language-focused task engagement. Measures in all areas of task engagement would ideally include standardized observational frameworks to capture verbal (e.g. interaction moves and patterns, task-related comments, language use and reflection) and non-verbal indicators (e.g. physiological indicators, body language and positioning, micro-expressions and dispositions, attentional activation patterns) and undergo validity testing in multiple sites and contexts of language learning.

Additionally, synthetic work shows that the field needs to develop and validate task-specific measures that provide clarity about the object of engagement across various task designs and implementation options (Dao et al., 2025). Take, for instance, an individual task performed in a face-to-face (F2F) setting versus a collaborative task in a synchronous computer-mediated communication (SCMC) setting. Technology-supported measures of task engagement across different task formats are likely to be meaningfully different given that such measures capture communication and collaboration across these distinct modalities, through various tools and affordances, mediated by peers, with language features in multiple domains, focused on the content and instructional material. This part of the research task could be pursued by creating detailed taxonomies of engagement components for specific task types and by developing standardized protocols that systematically capture engagement markers for specific modes of task delivery and that show variant and invariant engagement characteristics across task delivery modes.

Missing from the field, too, are person-specific measures that reflect more idiographic perspectives (Hiver, 2022) and can account for learners’ and teachers’ own perceptions of task engagement in particular contexts (Philp & Duchesne, 2016). We leave distinctions between individual versus collective group-levels of task engagement to the next research task, but one intriguing method of developing person-specific measures we highlight here is to use participatory research designs that capture participant-generated descriptions of engagement. For example, studies can enlist participants in crafting the focal measures of task engagement through a structured narrative elicitation method, by asking participants about their phenomenological experience (i.e. what they do, think, and feel when they deeply engage) with tasks, and by integrating personal reflection and self-assessment tools into task engagement measurements (Yamazaki, 2025).

A related objective is to integrate both direct and indirect measures due to the common mismatch between these measures in empirical work. As in research using such measures in the learning sciences and educational psychology (Symonds et al., 2024), there is little reason to expect perfect symmetry and consistency across direct, observational measures and indirect, self-report measures of task engagement. Subjective self-ratings and direct objective measures (e.g. discourse analytic measures) may in fact be getting at different things, each with a different time referent (retrospective vs. in-the-moment). Equally important, an overreliance on any single measure of task engagement poses a threat to the validity of results that arise from that data (Albert, 2025; Jackson, 2025b). There are dimensions of task engagement that lend themselves more to certain measures, for instance, making indirect measures more appropriate for capturing affective engagement than other dimensions. However, limitations and potential biases inherent in self-report engagement data are well documented (e.g. Dao et al., 2025; Hiver et al., 2024; Zhou et al., 2021) and relying exclusively on observational measures also raises the possibility of misleading conclusions. In the case of observational measures, particularly for the behavioural dimension of task engagement, learners may sometimes engage in pseudo-engagement behaviours in which they appear to be attentive and on-task but in fact are merely complying procedurally and are not engaged substantively or authentically with the task (Mercer et al., 2021). Consequently, triangulating across direct, observational and subjective, self-report measures is a useful countermeasure. Balancing direct measures with indirect ones is relatively straightforward because both overt and surreptitious aspects of task engagement (e.g. cognitive engagement) manifest through concrete, measurable learner activity even if such manifestations are used as a proxy for specific dimensions of task engagement (see Lambert & Aubrey, 2023 for a review; see also Hiver & Dao, 2025a; 2025b for empirical examples). Unlike some latent constructs that are not observable and have to be measured indirectly (e.g. psychological states such as motivation), by its very definition, task engagement is tangible and observable.

Yet another objective that lends itself to this research task is to develop novel ways of tapping into various timescales on which task engagement emerges and accounts for the temporal processes that are part of all L2 learning activity (Larsen-Freeman & Hiver, 2025). One way to go about this is to develop concurrent measures of task engagement in addition to retrospective, non-concurrent measures. This is crucial for certain dimensions of task engagement, especially the cognitive aspects, because the online (real time) elements of participation (e.g. learners’ processing) are an indispensable part of task engagement. Another way to advance this task is to adopt psychophysiological measures of task engagement (Lambert, 2023, 2025). Although many scholars appeal to the core of engagement as ‘action’, behavioural markers of engagement are often surface level manifestations (Dao, 2017). With learning tasks, one of the important foundations of deep participation and involvement is mental activity, and measures are needed to help uncover this hidden aspect (Gijsen, 2021).

Through this task, we envision the field developing ‘gold-standard’ or benchmark measures of task engagement, as other domains of L2 research (e.g. testing types of L2 knowledge with elicited imitation tasks, timed/untimed grammaticality judgments, oral narrative tasks, or metalinguistic knowledge tests) have done over recent decades. A cautionary note: in pursuing this research task, scholars should beware of using different labels and measures for very similar operationalizations of task engagement (aka the jangle fallacy) as this may create greater confusion about concepts. One remedy for this is to avoid co-opting measures for task engagement that are already conventionally used for other purposes. In current task engagement research, there seems to be a possibility for research to uncritically adopt measures of task engagement that are originally devised for measuring other constructs. These measures, for instance, include CALF/FA, which are used to gauge L2 production or task performance, or serial reaction time measures and other psychophysiological proxies that are employed to tap into cognitive processes (see also Albert, 2025; Jackson, 2025b). Arguably, if task engagement is perceived as a distinct construct, its measures must also be distinct and need not be borrowed from those of other constructs. Adopting the measures of other prominent constructs under the rubric of task engagement is only likely to introduce ambiguity into the empirical record. Addressing this critical gap in measuring task engagement with carefully operationalized and validated measurement tools will ultimately help enhance clarity, build empirical consensus, and allow for cross-study insights.

2.3. Task 3 – Individuals, groups, time, and contexts

Investigate individual and group engagement profiles and their variations across learners, groups, time, and contexts.

Given that this research task is concerned with individual and group dynamics in task engagement (rather than interaction) across time, space, and context, it is important to distinguish task engagement from interaction. Learners’ task engagement is neither considered synonymous with nor an alternative term for interaction. Task engagement is unique because it is viewed as a learner-focused construct that reflects cognitive, social, emotional, and agentic dimensions, whereas interaction is seen as the context or level within which task engagement occurs and unfolds. Thus, the first objective for this research task is to uncover commonalities in how individual learners engage (not interact) cognitively, emotionally, agentically, and/or behaviourally with peers and with/without the teacher’s participation in diverse contexts and across tasks. Task engagement has been conceptualized, operationalized, and measured in diverse ways, as highlighted in various synthetic reviews (e.g. Dao, 2024; Dao et al., 2025; Hiver et al., 2024; Namkung & Kim, 2024). While examining task engagement incrementally from multiple theoretical and analytical perspectives is valuable, following the principles of cumulative science, we will eventually need consensus evidence for synthetic work and meta-analyses (e.g. Reeve et al., 2025). While research has generated initial insights into some aspects of task engagement, findings reveal a complex picture of learner engagement profiles. Therefore, it is essential to continue to establish a foundational understanding of the common cross-study patterns in how learners engage across different contexts and tasks (Oga-Baldwin, 2019). This will enable subsequent research syntheses to provide a more comprehensive picture of what individual task engagement is, how it can be effectively promoted, and how it relates to its precursors (i.e. facilitators) and to consequential learning outcomes. One way this can be accomplished is to systematically review existing task engagement research to benchmark these commonalities, despite variations in theoretical frameworks, conceptualizations, and measurement approaches (see Dao et al., 2025; Namkung & Kim, 2024). These commonalities can then be re-examined in future studies involving different learners, tasks, lessons, schools, curricula, and socio-geographic contexts. Advancing task engagement research in these meaningful and reliable ways can minimize the risks of non-reproducible results, hype-driven research, and knowledge fragmentation.

A second objective for this research task is to explore the variation in how individual learners within the same group of peers engage with and benefit differently from identical tasks. Pursuing this will shed light on between-learner differences in task engagement and learning outcomes. The options to be explored through the first and second objectives here are how different learners engage with the same or different tasks in either the same or different context(s), and these suggest research designs that capture inter-individual variability. Studies are needed to examine how cognitive, affective, social, and agentic factors interact to shape these differences between learners, considering variables such as learners’ prior knowledge and experience (including the potential impact of previous or imagined interlocutors in shaping their internal cognitive processes) (Back, 2020; Lantolf, 2006), task motivation, language proficiency, and interactional dynamics (Yamazaki & Hiver, 2024). Innovative mixed-methods approaches, incorporating real time engagement measures (e.g. eye-tracking, psychophysiological responses) alongside qualitative insights from learner reflections and stimulated recalls, can help provide a more comprehensive understanding of how different tasks elicit varied levels of engagement and success across learners (Gurzynski-Weiss, 2024). While we discuss task design and implementation at length in the next research task, here we call attention to the need for research that investigates how instructional design, task complexity, and peer collaboration jointly provide the necessary conditions for task engagement (e.g. Albert, 2025; Hiver & Dao, 2025a, 2025b), thereby offering pedagogical implications for optimizing task effectiveness for diverse learners.

A third objective in pursuing this research task is to examine within-learner (aka intra-individual) differences in task engagement and learning outcomes over time, given that the same language learning task can yield different levels of engagement and success for the same learners at different times (Larsen-Freeman & Hiver, 2025). Specifically, studies can focus on how learners engage with and benefit from repeated encounters with the same task over time (e.g. Hiver et al., 2025). This line of inquiry can also explore the dynamic interaction between person-specific factors (e.g. individuals’ cognitive resources, motivation, emotional states) and salient external influences (e.g. task conditions, social interactions) that contribute to fluctuations in engagement and success across different task iterations. Longitudinal and microgenetic studies, incorporating real time measures such as eye-tracking, keystroke logging, and psychophysiological responses, alongside retrospective accounts like stimulated recall, could provide insights into how engagement evolves over time. Additionally, research is needed to identify patterns of stability and variability in individuals’ engagement, uncovering factors that promote sustained or declining engagement across repeated task exposures (e.g. Zhou et al., 2023). For example, empirical evidence suggests that the amount, intensity, and quality of learner engagement within a single task could fluctuate on a moment-to-moment timescale (e.g. Dao & Sato, 2021; Lambert & Aubrey, 2025). Findings from this work would inform pedagogical approaches to task design and task sequencing, ensuring tasks remain engaging and effective for learners at different points in their learning trajectory. Exploring these within-task variations can provide a more nuanced understanding of the momentary nature of task engagement (Symonds et al., 2024). One way to approach this is by investigating learner engagement across different phases of a single, preferably structured, task (e.g. information-sharing phase, decision-making phase). Time-sensitive data collection methods, such as the experience sampling method (Csikszentmihalyi, 2014) and the idiodynamic method (MacIntyre, 2023), could be of particular use for this purpose.

A final objective for this research task is to explore group-level or collective task engagement (i.e. what it is, what it looks like, what it does, how it can be measured, etc.). Because much of task-based language learning and use occurs relationally in dyads or groups with and/or without teachers’ involvement (Philp et al., 2013), collective forms of task engagement are also worth investigating. Consequently, future task engagement research should systematically explore group-level or collective task engagement, moving beyond individual engagement to examine how groups as a whole interact with and sustain engagement in tasks (see e.g. Jackson, 2025b on the topic of joint attention). This includes defining the construct of group engagement, identifying its various dimensions (see also Research task 2), and investigating its impact on learning processes and outcomes. Current research on collaborative group tasks in which dyads, triads, or larger groups complete tasks still relies heavily on measurement and analysis templates for individual task engagement. However, task engagement at a collective level may be qualitatively different when thinking about those same categories at the individual level. Empirical studies should explore how group engagement emerges through task performance and interaction drawing on insights from existing research on group work dynamics (e.g. Hiromori, 2024; Leeming, 2024). Methodologically, new tools and frameworks are needed to measure group engagement holistically, potentially incorporating multimodal data such as interactional patterns, gaze alignment, physiological synchronization, and discourse analysis. By distinguishing collective engagement from individual engagement, research can provide theoretical clarity and pedagogical guidance for designing tasks that foster meaningful group engagement in task performance.

2.4. Task 4 – Task design and implementation for task engagement

Investigate the impact of key task design features and task implementation options on task engagement.

The major research tasks outlined thus far have focused on two key dimensions of task engagement mentioned previously: learners (i.e. who is doing the engaging) and time (i.e. when, where, and on what timescale task engagement occurs). In this research task, we shift our attention to the third critical dimension: the task itself. Tasks have received substantial attention from scholars in L2 learning and teaching research, and continued growth is anticipated in this area (Namkung & Kim, 2024). Tasks take various forms according to their design and implementation features, and learners’ engagement with tasks is influenced by these task features (Ellis, 2024; Lambert, 2017). Scholars must remain mindful of the somewhat varying definitions of tasks that can create confusion in synthetic work on the topic (Ellis, 2017; Long, 2016). As a first step, it is thus imperative for task engagement studies to clarify what constitutes a task by offering concrete definitions and elucidating the theoretical and pedagogical underpinnings of the study (Teng, 2025). Thereafter, integrating meaningful adjacent constructs and models can expand and accelerate applications in practice. Pursuing this research task will offer more coherent insights by focusing on the relationships between task design features, task implementation choices, and learners’ task engagement.

A preliminary objective of this research task is to leverage instructional design to reach consensus on the key conditions for task engagement. Although the current lack of consensus partly stems from the complex interactions among the learner, task, and spatiotemporal context in which task engagement unfolds, another contributing factor is the sheer volume of task design and implementation features requiring investigation (Ellis et al., 2020). Given the almost limitless ways tasks can be designed (e.g. monologic or dialogic, open or closed, convergent or divergent) and implemented (e.g. F2F or SCMC, individual or dyad or group, with or without technology-based tools or generative AI, the absence or presence of planning time, etc.), an incremental approach to examine each feature’s unique and combined impact on task engagement is impractical and not an economical use of limited researcher resources. Instead, a more selective and targeted approach may be warranted, one which prioritizes meaningful combinations of task design and implementation features that are likely to create high-leverage conditions and have significant theoretical and pedagogical implications. In Research task 2, we outlined some examples of such features, including modality (i.e. input-based tasks) and linguistic orientation (i.e. focused tasks), that merit closer attention. In addition to these, we advocate for further work on learner-generated content (LGC). A growing body of work in this area has yielded promising insights for igniting learners’ personal investment by allowing them to generate task content (Lambert & Aubrey, 2025). Unlike many topics in task engagement research, which are often investigated sporadically, LGC has received consistent attention and has a solid evidentiary basis which future research can capitalize on and extend. Also, recent advances in technology have highlighted the potential of generative AI (GenAI) to create facilitative and high-leverage task conditions that enhance learner engagement. However, to date, research in TBLT and task engagement has yet to explore the integration of GenAI. Given its capabilities, GenAI holds promise for supporting engagement at various stages of task implementation, such as during planning (e.g. idea generation), reflection (e.g. providing feedback), and performance (e.g. acting as an interlocutor).

Another related objective, in line with what we argued at the outset of this piece, is to examine task design and task implementation features that have clear associations with research on L2 learning processes and language instruction. For example, task complexity (Révész, 2014; Robinson, 2001, 2011), task repetition (Bygate, 2018; Rogers, 2023), and task sequencing (Baralt et al., 2014; Robinson, 2005, 2021) are well-established topics in TBLT research, but these topics have garnered little attention in task engagement (Albert, 2025; Jackson, 2025b). Investigating how these features might enhance learner engagement has multiplicative benefits for both task engagement research and TBLT research (e.g. Jackson, 2025a; Qiu, 2024). Key insights will come not from asking how engaged learners were in a task (i.e. the quantity and amount) but investigating how learners engaged in a task (i.e. the qualitative differences in task engagement) (see also, Gijsen, 2021; Pusey, 2025).

For instance, task engagement research is needed to integrate and test the Cognition Hypothesis and the Limited Attentional Capacity Model (i.e. task complexity), insights from Cognitive Load Theory and Load-Reduction Instruction (i.e. task repetition), and the SSARC model (i.e. task sequencing). Because task features have received substantial attention from scholars in L2 learning and teaching research, our position is that the field needs more research that focuses on ‘how?’ and ‘why?’ questions with fewer ‘yes/no’ and ‘whether?’ questions. Attempts to answer such questions can be accomplished through mixed- and multi-method research, such as explanatory sequential designs and exploratory sequential designs, and are invaluable to unifying the aforementioned separate research strands.

A final objective for this research task is to link theoretical constructs that underlie how learners orient to and participate in tasks with tangible task design and implementation features. Even a well-designed and meticulously implemented task may not generate or sustain engagement for all learners (Lambert et al., 2023a; Philp & Duchesne, 2016). Consequently, a comprehensive understanding of optimal conditions for task engagement cannot be reached unless the task and the learner constructs are considered in tandem. In this regard, we previously outlined task motivation and task emotions as proximal antecedents and outcomes (see Research task 2). Examples of theoretical constructs that need work to articulate how they translate to tangible task design include learners’ situational interest, subjective task value, perceived task costs, and task competence beliefs (e.g. Yamazaki & Hiver, 2024). Concepts from task emotions that need explicit linking to task design and implementation elements also include enjoyment, anxiety, boredom, and other affective responses. Work that translates how tangible elements of task structure, task familiarity, appraisals of task conditions and difficulty, and choice of interlocutor implicate theoretical constructs related to the situation, the self, the goal, the action, the outcome, and the consequences of task performance will provide important insights for classroom practice (Dao et al., 2021).

2.5. Task 5 – Task engagement, L2 instruction, and teacher education

Explore teachers’ understanding, application, and promotion of task engagement to bridge research and classroom practice.

Teachers play a crucial role in translating task engagement research insights into effective instructional practices. And, while much is now understood about task engagement, particularly related to task design factors that influence it, there is a growing need to bridge the gap between research and practice (Hwang et al., 2024; Sato, 2024; Sato & Loewen, 2022). This raises key questions about how task engagement research can inform teachers’ professional development and instructional practices, as well as its implications for teacher preparation. To advance task engagement at both empirical and practical levels, this final research task aims to address teachers’ understanding, application, and promotion of task engagement in ways that will bridge research and classroom practice (see also Jackson, 2021).

The very first objective here is to explore how teachers of different learner groups in diverse teaching contexts understand the concept of engagement, how they assess it (i.e. which aspects they focus on) during task performance, and what strategies they use to promote engagement in various teaching settings (e.g. in-person, online synchronous and asynchronous delivery, and out-of-class settings). Too few studies have investigated how teachers understand and perceive the concept of learner engagement and the strategies they use to promote it (e.g. Acosta-Manzano & Mercer, 2024; Dao et al., 2021; Mystkowska-Wiertelak, 2022; Yuan & Dao, 2025). While providing some initial insights, this body of research has focused on small groups of teachers in specific contexts, mostly at the tertiary level, and largely relies on qualitative and other self-report measures (e.g. interviews and surveys). As a result, the generalizability of these findings is limited, as is our understanding of teachers working with different groups of learners (e.g. high school and primary school students, or adults vs. young learners) across various geographical contexts (e.g. where L2 learning is a compulsory foreign language vs. a second language) and instructional settings (e.g. F2F vs. online synchronous/asynchronous teaching). Pursuing this research task will advance our understanding of teachers’ perceptions of task engagement, the factors affecting it, and strategies for promoting it by investigating a wider range of teacher groups in diverse contexts and incorporating larger-scale data to enhance the generalizability of the findings.

The second objective of this research task is to examine how and to what extent teachers learn about task engagement research findings, how they apply these findings in their teaching practices, and the challenges they face in designing and implementing engaging tasks (Jackson, 2021). Surprisingly, L2 task engagement research has rarely explored how teachers and practitioners in the field use and apply research findings. Task engagement research is often positioned as directly relevant to teaching practice (Dao, 2024), particularly since teachers commonly use engagement as a general measure of instructional effectiveness and student learning. However, little is known about how task engagement research findings reach teachers, how they are translated into practice, and the challenges teachers face in applying these findings in their specific contexts with different learner groups. Work taking stock of the research-practice interface has pointed out that many pedagogical implications in L2 research are written primarily to fulfil journal requirements (Coss & Hwang, 2024) rather than to provide actionable recommendations for teachers. A key step in this research task is to explore how effectively teachers understand and apply task engagement research findings in their practice and how this application varies across different teachers and teaching contexts.

A logical extension is this next objective which focuses on the role of continuous professional development and teacher education in preparing teachers to design engaging tasks and promote learner engagement effectively. Recent research has provided initial evidence regarding the impact of specific task design and implementation features on learner engagement (see Research task 4). However, training teachers to design empirically validated tasks that effectively promote engagement remains a challenge (Erlam & Tolosa, 2022). Another limitation in this area is that existing studies are mostly exploratory (i.e. teachers’ perceptions) rather than experimental and classroom-based (i.e. teachers’ instructional actions), making it unclear whether teacher training in designing, manipulating, and implementing tasks can effectively equip them with sufficient skills to design tasks that promote learner engagement. Previous research has shown that even after receiving systematic training on designing effective classroom tasks, teachers still struggle to understand and create engaging tasks (Erlam, 2016). A key part of this research task is to investigate how research findings can inform teacher training programs to enhance teachers’ ability to design and implement tasks that have been empirically shown to promote engagement. In practice, Dao (2024) suggests that engagement-based pedagogy (EBP) can serve as a tool for teachers to assess engagement-related aspects of their tasks during both the design and implementation stages. For instance, during the design phase, teachers might reflect on questions such as:

• • Affective engagement: ‘Will my students enjoy the tasks I am designing?’

• • Cognitive engagement: ‘Will the tasks I am designing activate my students’ deep thinking, encourage the use of their prior knowledge, create opportunities for them to attend to and discuss language issues, and help them invest focused effort in completing the tasks?’

• • Social engagement: ‘Will the tasks I am designing encourage collaboration among my students and enhance their positive relationships throughout the task?’

• • Agentic engagement: ‘Will the tasks I am designing enable students to shape the task according to their preferences while still staying on track and focusing on what is expected?’

To date, no research has exclusively focused on working with teachers to design engagement-enhancing tasks. Therefore, it is crucial to explore how task engagement research insights can be embedded into initial teacher education across different teaching contexts and integrated into ongoing professional development to ensure that research findings translate into real life classroom practices.

The final objective for this research task is to explore how researchers can effectively engage with teachers to help them apply task engagement research frameworks in practice. Investing in researcher-practitioner dialogue can be mutually beneficial: practitioners stand to benefit researchers as much as they stand to gain from research. While this issue extends beyond task engagement to other areas of L2 research (see also Gurzynski-Weiss, 2024), it is particularly relevant here. Recently, there has been a growing call for increased dialogue between researchers and teachers to translate research findings into practice and foster researcher-practitioner collaboration (Sato & Cárcamo, 2024; see also Hwang et al., 2024; Rose & McKinley, 2017). Since task engagement research directly relates to teaching practices by offering pedagogical implications, it is essential that these insights reach teachers in accessible ways. If research findings remain inaccessible to practitioners, task engagement research will fail to fulfil its practical purpose. To bridge this gap, this research task can help move toward collaborative research-practice partnerships, in which teachers actively participate in research processes, practitioners generate research questions based on problems and possibilities of practice, and pedagogical implications are co-constructed to be immediately applicable in real world classrooms (see also Michel et al., 2025; Sato et al., 2025; Zheng et al., 2025).

3. Conclusion

This research agenda has outlined five essential tasks to advance our understanding of L2 task engagement. We began by addressing conceptualization needs that will further understanding of the context-motivation-engagement-learning cycle (Research task 1). We then emphasized the importance of developing rigorous measurement protocols that capture the multidimensional nature of task engagement across different contexts (Research task 2). Our agenda further highlighted the need to investigate individual and collective engagement profiles and their variations across diverse learners, groups, time periods, and contexts (Research task 3). Additionally, we explored how task design features and implementation options impact engagement, and then stressed the importance of integrating established L2 research frameworks (Research task 4). Finally, we emphasized the critical role of teachers as well as research-practice partnerships in bridging the research-practice gap through teachers’ enhanced understanding, application, and promotion of task engagement in language classrooms (Research task 5).

While many studies have contributed significantly to growing the field’s understanding of task engagement, numerous avenues for productive research remain ahead. The five research tasks outlined in this agenda, each with a set of objectives, provide a roadmap for addressing current limitations and expanding our knowledge of how task engagement functions within L2 learning and teaching. Pushing task engagement forward in these ways will help build a more cohesive, empirically grounded foundation for task engagement research that benefits both theoretical understanding and classroom practice. As the field continues to evolve, integrating these research priorities will help establish task engagement as a central concept in understanding effective L2 learning processes and outcomes.

Looking forward, if the research agenda outlined in this manuscript is systematically pursued over the next several years to a decade, L2 task engagement research would likely experience considerable development and maturation. We can reasonably expect a research landscape where task engagement evolves toward a coherent, empirically informed domain that better serves both theoretical understanding and practical application.

First, the field will likely achieve greater theoretical clarity through sustained attention to conceptual debates about the nature, make-up, and diverse manifestations of task engagement. Research will also make substantial progress toward establishing clearer taxonomies that specify when and under what conditions different task engagement dimensions emerge and influence learning outcomes. This theoretical development would be supported by more refined measurement protocols that capture both individual and collective engagement across various task types and contexts. The development of more standardized measures for task engagement would enable systematic meta-analyses, providing the field with more reliable estimates of engagement-learning relationships and their boundary conditions. The integration of advanced methodological approaches, including real time psychophysiological measures, multimodal interaction analysis, and longitudinal designs and analyses, will also contribute to a more nuanced understanding of the relationships between contextual factors, motivational precursors, dimensions of engagement, and learning outcomes.

Second, in a future likely characterized by increased technological integration, GenAI-mediated instruction, and ongoing challenges in teacher professional development, the insights generated by this research agenda could provide valuable guidance for maintaining effective language learning environments. For example, with GenAI as a tutor and a tool (Levy, 2024), and virtual reality applications and adaptive learning platforms becoming more prevalent, understanding how different engagement dimensions manifest in technology-mediated contexts would be increasingly important for TBLT. This research agenda’s emphasis on developing task-specific measures across different delivery modalities would help ensure that engagement-based principles can be applied effectively as the L2 educational landscape continues to evolve. The development of engagement-based pedagogy tools would support the development of more individualized and adaptive approaches to learner assessment and instruction. Ongoing assessment of learners’ task engagement patterns will enable teachers to more systematically enhance their instructional practices, even within existing constraints. And as educational systems seek meaningful measures of program effectiveness, validated task engagement indicators could provide useful alternatives or complements to traditional achievement metrics. This would allow stakeholders to evaluate L2 learning programs based on their capacity to promote sustained task engagement and create responsive learning environments for diverse student populations.

Additionally, in an era where teachers face increasing demands and precarity accompanied by diminished levels of professional support, this research agenda’s focus on bridging research and practice would be particularly relevant. Teachers will have access to more evidence-based engagement frameworks that provide clearer guidance for task design and implementation. Perhaps most significantly, this research agenda will contribute to how educators understand their role in task-based language learning. Rather than viewing task engagement as a characteristic inherent to learners, teachers will have a more sophisticated understanding of task engagement as an emergent phenomenon that can be influenced through thoughtful task design and implementation. The realization of this research agenda will, we hope, result in a field where theoretical development is closely connected to practical application (Michel et al., 2025), where research findings are increasingly accessible to educators, and where both practitioners and researchers work more collaboratively to improve L2 learning experiences for diverse student populations in evolving educational contexts (Sato et al., 2025; Zheng et al., 2025).