Aneta Naumoska is an Associate Professor of Linguistics at the English Department, Faculty of Philology (Ss. Cyril and Methodius University in Skopje, N. Macedonia). She is a CELTA-holder and NILE-certified Teacher Trainer. She has been in EFL teaching for 20 years, presented at many Conferences both home and abroad, and (co-)authored University coursebooks. Email: aneta.naum@flf.ukim.edu.mk  

Biljana Naumoska-Sarakinska is a Full Professor of Linguistics at the Department of English Language and Literature within the "Blaže Koneski" Faculty of Philology, Ss. Cyril and Methodius University in Skopje, N. Macedonia. She has more than 25 years of experience teaching EFL and ESP, has presented at many interational conferences, and (co-)authored a number of scholarly articles and university coursebooks.

Email: biljananaumoska@flf.ukim.edu.mk 

 

Introduction

In an era of abundant English - streaming media, social platforms, academic publishing, and AI-mediated communication - it is easy to assume that learners are saturated with input and that classroom time should prioritize language use. Yet EFL teachers often observe the opposite: plentiful exposure can co-exist with fossilized patterns, limited lexical range, and persistent form-meaning mismatches. The problem is not simply a lack of input, but it is the gap between input (what is available) and intake (what is actually processed in ways that change the learner’s interlanguage) (Corder 1967; Gass 2018).

Task-based language teaching (TBLT) in the SLA literature offers a principled response because it treats meaning-oriented action as the organizing unit of instruction, while also making room for form-focused work when it is psycholinguistically timely. Long (2014) argues that a task-based approach is not simply a communicative ‘coating’ but an attempt to align pedagogy with what we know about acquisition mechanisms and constraints. Moreover, the success of TBLT depends on a central condition that is sometimes underplayed in classroom discussion: tasks only drive acquisition if they organize input in ways that learners can process, and if they create interactional pressures that promote noticing and deeper form-meaning mapping (Gass 2018; VanPatten 2020).

The aim of this article is to connect input research to TBLT in a way that is usable for EFL teachers, with a practical application in higher education (for English majors). Throughout, the guiding question is: How can tasks be designed so that the input they mobilize is not merely encountered but converted into intake?

 

From input to intake: literature

Corder’s (1967) classic contribution was not merely to legitimize learner error as evidence of development, but to insist that the environmental language stream is filtered by learner processing. In his terms, input is what is available, amd intake is what is taken in. Pedagogically, this means that more English does not automatically equate to more learning. It also means that teaching can intervene not only by supplying input but by shaping the conditions under which learners notice, interpret, and store form-meaning connections.

Krashen’s (1985) input hypothesis is often invoked as a call for comprehensible exposure. His central claim is that acquisition is driven by understanding messages in the target language that are slightly beyond the learner’s current level (i+1). In classroom terms, this focuses attention on two design issues. First, comprehensibility is not a property of texts alone, but it is a property of texts in contexts with support and scaffolding. Teachers can make demanding authentic input comprehensible through pre-task priming, multimodal scaffolding, and purposeful comprehension goals. Second, the +1 is pedagogically meaningful only if tasks are calibrated so that learners can comprehend while still being ‘stretched’. Too little challenge results in comfortable comprehension with limited development, while too much results in breakdown and avoidance strategies.

Long’s (2014) interaction hypothesis adds a crucial layer: input becomes more learnable when it is interactionally modified through negotiation of meaning. In interaction, learners signal non-understanding, request clarification, confirm interpretations, and reformulate, among other things. Such moves increase comprehensibility and create moments where learners are pushed to attend to form because form becomes relevant to meaning. When a learner asks “Do you mean…?” or negotiates the interpretation of a modal verb, form is no longer ornamental, but it is rather part of solving the communicative problem. This is why interaction matters even in advanced university classes: not because students need more speaking, but because interactive work can reorganize input to make it processable and salient.

Gass’s (2018) work systematizes the relationship among input, interaction, and learner internal processing by emphasizing limited attentional resources and the role of cognitive operations such as attention and memory. If attention is limited, then task design becomes a form of attentional engineering: teachers choose what learners are likely to attend to at a given moment and how those attentional choices are distributed over time. This aligns with a simple observation: when tasks are cognitively heavy, students prioritize meaning and may sacrifice accuracy, and when tasks are linguistically dense, students may sacrifice fluency.

VanPatten’s (2020) input processing perspective makes this mechanism even more explicit by arguing that acquisition is, to a significant degree, a by-product of comprehension, but that comprehension is guided by processing strategies and constraints. Learners typically prioritize meaning over form, therefore, some grammatical cues are routinely ignored unless instruction or task conditions make them necessary or salient. In practical terms: if learners can complete a task by relying on lexical content and world knowledge, they may not attend to discourse markers, tense-aspect morphology, or stance devices, even if these are critical to advanced proficiency.

Hence, input is the language available in and around tasks, while intake is the part of that language that learners process with sufficient depth to modify their interlanguage. TBLT is one attempt to build that depth of processing into the architecture of instruction (Ellis 2003; Ellis et al. 2019). Understanding TBLT requires more than a list of classroom activities, rather it requires a clear view of what tasks are designed to do to input: make it comprehensible (Krashen 1985), interactionally modified (Long 2014), attentionally manageable (Gass 2018), and processing-relevant for form-meaning mapping (VanPatten 2020).

 

What TBLT is and what it is not

TBLT is frequently reduced to ‘group work’ or “’communicative activities’, but in the research tradition it refers to an approach in which tasks are the central unit of syllabus design and classroom organization (Ellis 2003; Long 2014). A task, in this tradition, is a meaning-oriented workplan that requires learners to use the target language to achieve a non-linguistic outcome, e.g. solving a problem, making a decision, or synthesizing information, under conditions that resemble real communication in one key way: language choices are constrained by meaning and purpose, not merely by a grammar point chosen in advance.

In this context, three characteristics should be noted: (1) Tasks are not the same as exercises. An exercise typically targets a predetermined linguistic form and measures success by correct production, while a task measures success by achievement of an outcome and allows multiple linguistic routes (Ellis 2003; Ellis et al. 2019). Exercises can still have a place in task-supported teaching, but in a task-based syllabus the task is primary, and form-focused work is subsidiary or strategically embedded. This matters because if tasks are simply disguised grammar drills, they no longer mobilize the processing mechanisms that give TBLT its theoretical rationale (Long 2014; VanPatten 2020). (2) TBLT is not equivalent to ‘no teaching.’ Teachers do not simply assign tasks and step back, but actually they curate input, design constraints, provide scaffolding, and orchestrate interaction so that tasks become learning opportunities (Ellis et al. 2019). This includes selecting texts and audio as input, designing pre-task planning that supports comprehension, setting interaction norms that promote negotiation of meaning, and using post-task reflection to consolidate learning. Teacher agency is central in TBLT, but it is exercised as design and orchestration, not as constant direct explanation. (3) TBLT is not automatically real life. A task can be pedagogically task-like without being identical to a real-world activity as long as it creates a communicative need and involves authentic processing demands (Robinson 2011; Skehan 2018). In higher education, where English majors must engage with contemporary discourse, tasks can be designed around academic and public genres, e.g. a conference poster, a literature review, a policy brief, or a podcast episode. The authenticity lies in discourse demands, such as stance-taking, audience awareness, source integration, and coherence under communicative constraints - the kinds of demands that move learners towards advanced competence.

The rationale for TBLT is rooted in the idea that language develops through use under conditions that engage comprehension, interaction, and attention. Krashen’s (1985) emphasis on comprehensible input suggests that tasks should be built around texts and interactions that learners can understand with support, while still pushing them beyond current competence. Long’s (2014) interaction hypothesis suggests that tasks should trigger negotiation of meaning and interactionally modified input. Gass (2018) highlights attention and memory: learning is not simply exposure, but exposure plus processing under attentional constraints. VanPatten (2020) emphasizes that learners’ default processing strategies prioritize meaning, so tasks must sometimes make form relevant to meaning to promote intake. Skehan (2018) adds a cognitive lens by emphasizing that tasks impose attentional and memory demands and that performance reflects trade-offs among complexity, accuracy, and fluency. When students are pushed to speak fluently, accuracy may drop; when they are pushed to be accurate, fluency may be inhibited; when the content is complex, lexical variety may shrink. Tasks must be sequenced and supported so that development is more likely than overload. Robinson’s (2011) work on task complexity provides complementary guidance by arguing that task demands can be manipulated to direct attention and stimulate development, but that complexity must be handled carefully to avoid overload. In practice, this means teachers can adjust features such as reasoning demands, number of elements, time pressure, and information organization. A task sequence can, therefore, be planned as a developmental pathway: early tasks offer stronger support and fewer competing demands, later tasks increase discourse complexity once learners have resources to handle it.

 

Core components of TBLT

The pre-task phase prepares learners by activating topic knowledge, clarifying goals, and providing key input. Importantly, pre-task work should not frontload the entire language needed, but it should make the main task doable while leaving room for learners to stretch and discover. Pre-task work can include: purpose setting (“What decision must you reach?”), guided comprehension of a model text, strategic planning time, and noticing prompts that draw attention to useful language in input (Krashen 1985; VanPatten 2020).

The during-task phase is where interaction and outcome pressure shape processing. Teachers can decide whether tasks are performed under time pressure, whether learners can consult resources, and whether the output is spoken, written, or multimodal. These decisions matter because they redistribute attentional resources (Skehan 2018) and influence whether learners engage in negotiation of meaning (Long 2014). During-task monitoring also offers teachers data for later consolidation: the language that learners struggled with, the breakdowns that required negotiation, and the discourse moves that learners attempted but did not control.

The post-task phase is where TBLT’s reputation for being only communicative can be corrected. Post-task work can include repetition of the task with improved performance, reflection on strategies, analysis of language from the task, and focused practice of emergent language as needed (Ellis 2003; Ellis et al. 2019). From an input perspective, post-task work also provides new input: models, reformulations, and peer output become language data that learners can process with greater awareness because they now have task experience and a clearer sense of what language was needed (Gass 2018; VanPatten 2020).

 

Input in TBLT: Where it comes from and how tasks make it learnable

In task-based classrooms, input comes from multiple channels: teacher talk, peer talk, texts, audio/video, and the learners’ own emerging output that becomes input for others. The pedagogical issue is not simply ‘more input’, but the quality of input and the processing opportunities attached to it (Corder 1967; Krashen 1985). Since tasks are outcome-driven, they can provide a natural reason to engage with authentic input: learners read, watch, or listen because they need information to complete the task. This purposefulness matters because it invites deeper engagement than read-and-answer-question routines that often encourage surface scanning rather than interpretive processing (Gass 2018).

Teacher language and classroom routines are an often underestimated input source. For university EFL students, teacher talk is most helpful when it is not simply slow and clear but strategically purposeful: teachers can model academic discourse moves (summarizing, evaluating, qualifying), rephrase student contributions in more target-like ways, and recycle key lexical bundles across weeks so that students meet them repeatedly in meaningful contexts. During the pre-task phase, brief teacher narratives or mini-lectures can anchor the topic and provide an interpretable model of register. During the task, teachers can offer reformulations and confirmation checks that keep interaction moving while also making form-meaning mappings clearer (Gass 2018; Long 2014). After tasks, teacher summary talk can consolidate groups’ output into a coherent, more target-like version, adding one or two upgraded forms for students to notice and reuse. However, outcome focus can also allow students to bypass target forms. VanPatten’s (2020) processing perspective predicts this: when learners can rely on content words, world knowledge, or context, they may not allocate attention to grammatical morphology, discourse markers, or less salient lexico-grammatical patterns. The challenge is to design tasks where those patterns become functionally necessary. When forms are tied to outcome success, learners are more likely to process them deeply enough for intake (Robinson 2011).

Interaction is the second major input channel. Long (2014) emphasizes that interaction does not merely provide more talk, but it modifies input in ways that support comprehension and noticing. Clarification requests, confirmation checks, and reformulations can make forms more salient and can direct learners’ attention to precise mappings between form and meaning. This matters particularly in higher education, where students may already comprehend a lot but still need to refine accuracy, pragmatic appropriateness, and genre control. For English majors at university, subtle choices, like evaluative vs. descriptive stance, are not optional, but they are central to contemporary discourse competence. Interactional tasks can make these choices visible and consequential (Ellis et al 2019; Skehan 2018).

A third dimension is input structuring and enhancement. While early discussions of TBLT sometimes framed task work as naturalistic, contemporary accounts accept that teachers can make input more learnable without abandoning task authenticity (Ellis et al. 2019). Teachers can select texts with a high density of target patterns, visually enhance forms in written input (bolding, underlining), or create noticing prompts that ask learners to attend to how a speaker hedges or how a writer signals limitation. These supports align with Krashen’s insistence on comprehensible input but also address VanPatten’s concern that comprehension alone does not guarantee attention to form (Krashen 1985; VanPatten 2020).

 

Practical design principles for deliberate input-to-intake conversion

Comprehensible input remains a precondition for deeper processing (Krashen 1985). In university contexts, ‘comprehensible’ does not mean simplified, but rather supported. Pre-task work can supply background knowledge, preview key lexis, and set a clear purpose for reading/listening so that students’ resources are available for meaning-making (Ellis et al. 2019; Krashen 1985). Once comprehension is secured, teachers can add a precision pressure: tasks can require fine distinctions that cannot be solved through gist alone, e.g. differentiating certainty from probability, attributing claims to sources, or distinguishing evaluation from description. This aligns with VanPatten’s (2020) view that tasks must sometimes force attention to form by making it relevant to meaning.

Negotiation of meaning is not an automatic by-product of pair work, rather it is an outcome of task constraints (Long 2014). Information distribution is a robust constraint: give each student different pieces of evidence, conflicting data, or complementary texts. Require a jointly authored product. Assign roles that create productive asymmetries (e.g. one student is editor, another fact-checker). These structures increase the likelihood of clarification requests, confirmation checks, and reformulations, which are precisely the interactional moves that Long (2014) argues make input more learnable. They also help advanced students move beyond polite agreement toward precision-oriented academic talk.

VanPatten (2020) warns that learners’ default processing priorities can cause them to miss grammatical cues unless tasks make those cues meaningful and salient. Noticing can be built into tasks in two complementary ways. First, embed ‘input-to-output bridges’: after reading or listening to a model text, learners must produce a short segment that uses specific discourse resources they noticed (e.g. three hedges, two contrastive connectors). Second, embed guided comparison: learners compare two versions of a text (e.g. an assertive vs. hedged claim) and decide which is appropriate for a given audience and why. Such noticing routines are compatible with TBLT because they remain outcome-oriented: noticing is framed as a tool for better task performance, not as isolated grammar study (Ellis 2003; Ellis et al. 2019).

Skehan’s trade-off view (2018) and Robinson’s complexity framework (2011) together suggest that harder tasks are not automatically better, but what matters is how demands are sequenced. Teachers can increase demands on one dimension while reducing another. Early tasks can allow planning time and provide strong input support while learners learn a new genre; later tasks can reduce planning time, increase reasoning demands, or introduce additional sources once language resources are established. This sequencing principle is especially relevant in university courses where task outcomes are complex (synthesis, evaluation, argumentation), so without careful sequencing, cognitive load can crowd out attention to language, producing fluent but underdeveloped output.

Input becomes intake partly through repeated, varied encounters. A task sequence can recycle the same language feature across modalities: read an op-ed (written input), listen to a panel discussion (audio input), debate in groups (interactive input), and then produce a podcast or policy brief (output that becomes input for peers) (Krashen 1985; Long 2014). From a TBLT perspective, recycling should be planned as genre movement: learners encounter language in one genre and then redeploy it in another, strengthening form-function mapping and discourse flexibility (Ellis 2003; Ellis et al. 2019).

 

Two applications: university EFL context

The following examples take a 12–14-week semester course in Contemporary English Language for English majors (C1). Both aim to develop students’ ability to analyze and produce contemporary academic and public discourse, while using tasks to convert input into intake.

Example 1: “Stance in Contemporary Public Discourse” Podcast Project

Task outcome: a 6-8-minute group podcast episode that evaluates a contemporary issue (e.g. AI in education, climate communication, digital privacy) for a defined audience, using appropriate stance and hedging.

Input package (Weeks 1-2): Students receive two short opinion articles with contrasting stances, a transcript excerpt from a panel discussion, and a model podcast segment. The input is not simplified, but authentic and supported through a pre-task glossary of stance markers (e.g. it seems, arguably, there is evidence that), guided questions, and a clear comprehension purpose (Ellis et al. 2019; Krashen 1985). Students complete an input-based task: they highlight stance expressions and classify them by function (hedging, strengthening, attributing, evaluating). A brief noticing prompt requires them to identify where the speaker/writer shifts stance and why (VanPatten 2020).

During-task interaction (Weeks 2-3): Each student brings a different source to the group (information distribution) and must persuade the group which evidence should appear in the episode. Because sources conflict, negotiation is unavoidable: students must clarify interpretations, challenge claims, and reformulate. The teacher monitors for interactional moves and provides occasional prompts that connect form to meaning (“Is that definitely true or is it more cautious?”; “Can you attribute that claim?”), supporting negotiation and precision (Gass 2018; Long 2014).

Post-task focus on form (Week 3): Using excerpts from student outlines, the teacher runs a short workshop on hedging patterns, reporting verbs, and cohesion devices common in evaluative discourse. The language focus is reactive: it targets what students needed but did not yet control (Ellis 2003; Ellis et al. 2019). Students revise their outlines and mark where stance choices are intentional.

Recycling and performance (Weeks 4-5): Students script and record the episode. A second task requires peer groups to listen and provide feedback using a rubric with both outcome criteria (coherence, evidence integration, audience alignment) and language criteria (range and appropriateness of stance devices, pragmatic politeness, clarity of attribution). This dual focus reflects the idea that task success is meaning-based but development can be tracked through performance features (Robinson 2011; Skehan 2018). Students submit a brief reflection identifying three stance devices they intentionally adopted from the input, and one stance device they overused or misused, referencing examples from their script.

How this turns input into intake: The sequence forces repeated encounters with stance markers across reading, listening, discussion, and production; it also makes those markers functionally necessary for success; negotiation of meaning and feedback create salience, while post-task analysis supports deeper processing and retention (Krashen 1985; Long 2014; VanPatten 2020). For English majors, the outcome is not only fluent speech but rhetorically controlled speech.

Example 2: Corpus-Informed “Contemporary Academic English” Style Portfolio

Task outcome: an individual 900-1,100-word literature review (or research commentary) that uses contemporary academic conventions, such as cautious claims, synthesis, and appropriate collocations.

Input package (Weeks 6-7): Students explore a small, curated corpus (10-12 short research article introductions in linguistics) and a set of annotated excerpts showing common rhetorical moves. The pre-task goal is comprehension of genre expectations: students identify how writers position claims, attribute sources, and signal limitation (Ellis et al. 2019). An input-based task guides processing: students search the corpus for reporting verbs (e.g. argue, suggest, claim, demonstrate), stance bundles (e.g. it is widely accepted that), and limitation markers (e.g. however, nevertheless, despite). They annotate examples with ‘function notes’ (what the form is doing in the argument), supporting form-meaning mapping (Corder 1967; VanPatten 2020).

During-task interaction (Weeks 7-8): In groups, students build a shared style bank, i.e. a collaboratively written resource with three ways to synthesize studies, three ways to express limitation, and three ways to signal cautious interpretation. Information distribution is built in: each student is responsible for a different rhetorical move and must teach it to the group using evidence from the corpus (Long 2014). Since students must explain patterns, not just find them, processing depth increases. Group discussion also creates opportunities for negotiation, linking interaction to input processing (Gass 2018; Long 2014).

Post-task focus on form (Week 8): The teacher provides targeted feedback on recurrent issues (e.g. lack of attribution, missing hedges). Rather than presenting a full lecture, the teacher uses brief micro-lessons triggered by student drafts and illustrated with corpus lines. This is task-based focus on form as it is responsive to performance and tied to the communicative goal of writing credibly in the discipline (Ellis 2003; Ellis et al. 2019).

Recycling and assessment (Weeks 9-10): Students submit two portfolio drafts with structured peer review. Peer review is designed as an information-gap task: reviewers must locate and annotate three instances of hedging and two synthesis moves, then meet the writer to discuss whether the choices match the intended argument and audience. Because reviewers need to interpret the writer’s meaning, negotiation occurs naturally (“Do you mean this is certain or tentative?”; “Is this your claim or the source’s claim?”), connecting interaction to intake (Gass 2018; Long 2014). Final assessment considers both the communicative outcome (clarity, synthesis, audience alignment) and language development (range and appropriateness of bundles, stance consistency, cohesion), aligning assessment with task performance and developmental indicators (Robinson 2011; Skehan 2018).

How this turns input into intake: Corpus exploration makes patterns visible; collaborative teaching of patterns forces deep processing; drafting and feedback recycle language across time; and the genre outcome makes form-function choices consequential for credibility (Corder 1967; Gass 2018; VanPatten 2020).

 

Conclusion

If TBLT is understood as an attempt to align classroom practice with acquisition mechanisms, then providing input is not enough. Teachers must design input pathways, i.e. select and structure texts, engineer interaction, and build noticing and recycling into a task sequence (Ellis et al. 2019; Long 2014). Krashen (1985) reminds us that comprehension is the gateway, Long (2014) reminds us that interaction can modify input and create moments of form-relevant attention, Gass (2018) reminds us that attention is limited and must be managed through task design, and VanPatten (2020) reminds us that learners’ processing priorities can cause them to miss critical cues unless tasks make those cues meaningful and salient.

The practical takeaway is to treat tasks as the vehicle and input as the engine. When tasks are designed so that learners must engage deeply with authentic input, negotiate meaning, and recycle language across genres, intake becomes more likely. In higher education contexts such as Contemporary English Language courses for English majors, this design logic supports not only general proficiency but also the discourse competence students need to participate in contemporary academic and public communication. Finally, TBLT provides teachers with classroom-researchable indicators of input-to-intake conversion: whether learners’ interaction shows negotiation moves, whether repeated tasks show greater accuracy and complexity without loss of meaning, whether recycled language appears in new genres, and whether post-task reflection shows finer form-meaning awareness. In this sense, TBLT is not only a methodology but also a framework for making input visible, learnable, and accountable.

 

Bibliography

Corder, S. P. (1967). “The Significance of Learners’ Errors.” International Review of Applied Linguistics in Language Teaching, vol. 5. pp. 161-170.

Ellis, R. (2003). Task-Based Language Learning and Teaching. OUP.

Ellis, R., Skehan, P., Li, S., Shintani, N., & C. Lambert. (2019). Task-Based Language Teaching: Theory and Practice. CUP.

Gass, S. M. (2018). Input, Interaction, and the Second Language Learner. Routledge, 2 ed.

Krashen, S. D. (1985). The Input Hypothesis: Issues and Implications. Longman.

Long, M. (2014). Second Language Acquisition and Task-Based Language Teaching. Wiley Blackwell.

Robinson, P. (Ed). (2011). Second Language Task Complexity: Researching the Cognition Hypothesis of Language Learning and Performance. John Benjamins.

Skehan, P. (2018). Second Language Task-Based Performance: Theory, Research, Assessment. Routledge.

VanPatten, B. (2020). “Input Processing in Adult L2 Acquisition.” In VanPatten, B., Keating, G. D., & S. Wulff (Eds). Theories in Second Language Acquisition: An Introduction. Routledge. pp. 105-127.

 

Coming soon! Please check the Pilgrims in Teacher Training courses in 2027 at Pilgrims website.