Listening and Speaking:
The Science Behind Confident Communication

Abstract

A learner can know the grammar of English yet struggle to follow an unscripted conversation or respond quickly to a question. This gap between linguistic knowledge and communicative performance is central to second-language education. This evidence-informed narrative review examines listening comprehension, speech perception, pronunciation, spoken output, interaction, feedback, fluency and online teaching. It draws on selected peer-reviewed studies and research syntheses, distinguishing evidence for particular instructional interventions from wider pedagogical interpretation. The argument is not that listening and speaking universally outrank reading and writing, nor that progress in one domain automatically transfers to another. Rather, spoken communication imposes real-time perceptual, linguistic and social demands that deserve explicit attention. The review closes by considering how these principles may inform personalised online tuition, coordinated teaching teams and independent practice at C2 English School. It does not claim that C2's own programme has been independently validated.

1. From knowing English to using it

1.1 The gap between declarative knowledge and performance

A student may explain a conditional construction accurately during a grammar lesson but struggle to use the same construction while negotiating a deadline. Another may read a business email without difficulty yet miss a key qualification in a fast meeting. These apparent inconsistencies are not surprising. The tasks differ in the time available, the degree of predictability and the number of processes that must operate simultaneously. Written work permits inspection, correction and sometimes consultation. Conversation often demands immediate interpretation and a response while the speaker is still processing the previous turn.

The distinction should not become a simplistic attack on explicit grammar teaching. Grammatical and lexical knowledge are indispensable resources. The problem arises when performance on exercises is taken as sufficient evidence that a learner can use those resources spontaneously. Spoken interaction requires listening, interpreting intention, retrieving vocabulary, choosing structures, planning articulation, monitoring the listener and repairing misunderstandings. These operations compete for attention. A learner who pauses frequently may lack vocabulary, but may instead be spending too much processing capacity on retrieving familiar forms.

A credible assessment must therefore examine what the learner can do in relevant situations. An accurate rehearsed presentation does not establish that the learner can answer an unexpected question. Equally, hesitation in an unfamiliar topic does not prove a general lack of fluency. The teacher needs evidence from different contexts and should avoid treating one task as a complete representation of ability. The goal is to identify which processes most limit participation and design opportunities to improve them.

1.2 Why focus on oral communication?

Listening and speaking are especially visible in workplaces, travel and social life because misunderstandings happen immediately. A learner may possess the language needed to express an idea but fail to recognise an interlocutor's rapid question. Conversely, they may understand the question but lack the fluency to formulate a timely response. These difficulties can be emotionally costly: repeated hesitation may reduce willingness to participate, even when the learner has substantial knowledge.

It would be inaccurate to claim that listening and speaking are the hardest skills for everyone. Academic writing, extended reading and literacy present substantial challenges in their own right. The importance of each skill depends on the learner's purposes, prior experience and educational setting. The more defensible argument is that oral comprehension and production involve distinctive, time-sensitive demands and cannot be assumed to develop sufficiently through grammar exercises and written practice alone.

Interaction creates a particularly interesting environment for development. A learner's question elicits new input; a misunderstanding prompts clarification; an attempted explanation may reveal a gap in vocabulary or structure. Mackey and Goo's (2007) research synthesis provides a foundation for examining interaction and second-language acquisition. It does not justify the stronger claim that every conversation produces learning. Outcomes depend on tasks, attention, feedback, learner readiness and the opportunity to use newly noticed forms.

1.3 What counts as evidence?

This publication is a narrative research review, not a systematic review, meta-analysis or original experiment. It integrates influential theory with selected empirical studies and syntheses, but it does not claim to exhaust the literature. Where a cited study concerns learners of French, children or a particular pronunciation task, we identify that context rather than assuming the result transfers unchanged to all adult English learners. Where research supports an instructional principle but not C2's particular implementation, we say so.

The difference between research-informed design and research-validated outcomes is important. A school may use activities compatible with published findings without having demonstrated that its programme produces superior results. A designated lead teacher, small classes and supplementary practice can be coherent design choices; their effectiveness must ultimately be evaluated through learner outcomes, not inferred from their existence. The review therefore treats research as a way to formulate good questions and improve teaching decisions, not as a decorative endorsement.

2. The cognitive challenge of listening

2.1 Continuous speech and word recognition

Writing provides visible spaces between words. Speech does not. A listener must infer word boundaries from a continuous acoustic signal while pronunciation changes according to stress, speed, context and neighbouring sounds. Learners who recognise every word in a transcript may still fail to identify those words when spoken naturally. This is one reason listening can feel unexpectedly difficult even for students with a broad written vocabulary.

Consider a familiar expression such as “What are you going to do?” Its careful pronunciation differs from many everyday realisations, in which unstressed syllables become shorter and less distinct. A learner expecting a sequence of fully articulated dictionary forms may mis-segment the phrase. The problem is not necessarily lack of vocabulary: the learner may know all the words but not recognise their acoustic shapes in context. The teacher must distinguish unknown words from known words that were not heard.

Field (2008) argues for greater attention to the processes underlying listening rather than relying on comprehension questions alone. If a learner answers a question incorrectly, replaying the entire recording may help, but it may not explain the error. More diagnostic work can isolate the relevant phrase, compare the learner's initial interpretation with the transcript, identify reduced sounds and then test recognition in a different example. The objective is to improve a reusable listening process, not merely to recover one answer.

2.2 Bottom-up and top-down processing

Listening combines evidence from the sound signal with expectations derived from context. Bottom-up processing includes recognising sounds, word forms and grammatical relationships. Top-down processing includes topic knowledge, situational expectations and likely speaker intentions. Neither is sufficient alone. Someone hearing a station announcement can predict that a destination or platform may be mentioned, but that expectation must be checked against what was actually said.

Context helps listeners resolve ambiguity; it can also encourage mistaken certainty. If a listener expects a familiar number or destination, they may interpret a weak acoustic signal accordingly. Instruction should therefore encourage predictions as hypotheses rather than substitutes for hearing. A useful activity asks students to identify which words they recognised directly and which conclusions they inferred from the situation. The distinction is educationally valuable because two learners may arrive at the same correct answer through different routes.

Graham (2017) examines listening strategies in instructed settings and the difficulty of translating research into classroom practice. Strategy instruction should not be reduced to slogans such as “listen for keywords”. A learner cannot implement that advice reliably if rapid word recognition is itself the problem. Strategy awareness needs to work alongside perceptual practice, vocabulary development and repeated opportunities to verify interpretations.

2.3 Working memory and the disappearing signal

Listening places demands on attention and working memory because incoming information must be processed while earlier material is retained. If word recognition is slow, the learner may spend so much effort decoding the current phrase that the previous one is forgotten. A short utterance may be manageable while an extended explanation becomes overwhelming. This is not necessarily poor motivation or concentration; it may reflect processing demands that exceed the learner's current efficiency.

Instruction can help by controlling task length and complexity without permanently simplifying authentic speech. Early practice might involve short segments, a clear purpose and optional replay. Later tasks can introduce longer turns, unfamiliar speakers, less predictable topics and interactional pressure. The objective is graduated independence. The teacher should be able to explain what the additional difficulty is intended to develop rather than simply making material harder.

Different genres create different demands. A customer-service call, academic lecture, casual conversation and engineering meeting vary in vocabulary, organisation and turn-taking. Success in one genre does not guarantee success in another. Assessment should therefore sample contexts that matter to the learner, while teaching should gradually expand the range of situations in which comprehension remains reliable.

2.4 Metacognitive listening instruction

Listening also involves monitoring one's own understanding. Skilled listeners notice uncertainty, revise an interpretation and decide whether to request clarification. Metacognitive instruction makes these decisions more visible. Vandergrift and Tafaghodtari (2010) compared a process-based approach with a control condition using the same French listening texts and the same number of listenings. In their study of 106 learners, the experimental group performed better on the final comprehension measure after initial differences were controlled, with particular benefits for less skilled listeners.

This finding supports the potential value of guided planning, monitoring and evaluation, but it does not establish a universal treatment effect for every language or age group. Goh and Taib (2006) also explored metacognitive listening instruction with a small group of primary pupils; their findings are suggestive, while the sample size limits generalisation. Both studies encourage a more thoughtful question than “How many recordings did students complete?”: did the learner become better able to diagnose and manage comprehension difficulty?

A practical sequence might begin with an uninterrupted listening attempt, followed by a record of what was understood and what remained uncertain. The teacher then selects one difficulty for analysis, perhaps a reduced phrase, an unfamiliar discourse marker or a missed contrast. A final listening attempt tests whether the learner can now recognise the relevant information. Such an approach preserves meaningful comprehension while making the underlying learning process explicit.

3. Pronunciation, perception and being understood

3.1 Pronunciation is a communicative skill

Pronunciation is often reduced to the question of whether someone sounds like a first-language speaker. This is an unhelpfully narrow objective. The educational goal is usually to help a learner communicate intelligibly and understand others. Pronunciation includes individual consonants and vowels, but also word stress, rhythm, prominence, intonation and timing. These features affect both the listener's interpretation and the speaker's ability to make intended meanings clear.

Speech perception and production are related but not identical. A learner may recognise a contrast without producing it consistently, or reproduce a phrase carefully without recognising the same feature in fast conversation. Teachers should therefore assess both domains when the learner's needs require it. A useful lesson might begin with distinguishing two similar words, continue with controlled production and finish with a communicative task in which the contrast carries meaningful information.

English stress and reduction illustrate the connection. Learners accustomed to equally strong syllables may have difficulty recognising words whose unstressed vowels are shortened or centralised. Instruction that links written forms, careful speech and natural connected speech can make these patterns easier to notice. The point is not to teach one artificial “correct” accent, but to develop flexible perception and sufficiently clear production across relevant varieties.

3.2 Accent, comprehensibility and intelligibility

Derwing and Munro (2005) argue for pronunciation teaching guided by intelligibility rather than accent elimination. A noticeable accent is not equivalent to being difficult to understand. Conversely, a small difference in a crucial word can have a disproportionate effect on communication. The distinction matters because learners may invest effort trying to erase harmless features of their accent while overlooking patterns that repeatedly cause misunderstanding.

Teaching priorities should depend on communicative consequences. For an engineer, clear numbers, quantities, technical terms and stress patterns may be especially important. For a healthcare worker, understanding questions and communicating instructions precisely may be central. For a teenager preparing for international study, turn-taking, classroom discussion and recognising different accents may matter more immediately. The same universal pronunciation syllabus cannot perfectly serve all three without adaptation.

Intelligibility is also relational. Listeners vary in their familiarity with accents and their willingness to collaborate. Misunderstanding should not automatically be attributed to the speaker. Teaching can develop both sides of communication: clearer production and respectful listening, including requests for repetition, confirmation and reformulation. This is especially important in international English, where successful communication frequently occurs between speakers with different linguistic backgrounds.

3.3 What pronunciation research supports

Saito and Plonsky (2019) reviewed pronunciation-teaching studies using a framework that distinguishes specific from global measures, controlled from spontaneous performance and different scoring methods. Their synthesis of 77 studies found clearer effects when instruction targeted monitored production of particular segmental or suprasegmental features. Evidence was less clear when outcomes were measured globally through subjective judgments of spontaneous speech.

This result should shape how schools interpret improvement. A student may pronounce a target sound accurately during a focused drill but still fail to use it consistently during an unpredictable discussion. Both achievements matter, but they are not interchangeable. Progression from controlled recognition to controlled production, guided conversation and spontaneous use should be monitored rather than assumed. If gains disappear when attention shifts towards meaning, further practice is required.

The evidence also argues against extravagant promises. Pronunciation development varies with prior experience, language background, motivation, opportunity for practice and the target feature. Teachers can offer systematic feedback, carefully chosen tasks and ongoing assessment; they should not promise universally rapid or accent-free outcomes. A credible approach makes success criteria explicit and selects measures aligned with the learner's real needs.

3.4 Prosody and interaction

Meaning in speech is carried by more than words. Prominence can indicate contrast; intonation can signal uncertainty, enthusiasm or an invitation to continue; pauses can support or disrupt turn-taking. A learner who focuses exclusively on individual sounds may remain difficult to follow in conversation despite careful articulation. Instruction should therefore connect pronunciation to the communicative work it performs.

A simple activity can use the sentence “I thought you sent it” with prominence on different words. The learner listens to contrasting versions, explains what each implies and then produces a version that communicates a chosen meaning. Another activity asks the learner to give a short explanation and observe whether the listener understood the intended contrast. This links sound, meaning and feedback rather than treating pronunciation as mimicry.

Repair strategies deserve equal attention. When a learner cannot find a word, paraphrase may maintain the conversation; when a listener misses a detail, a precise clarification request can prevent misunderstanding. Such behaviours are not evidence of deficient English. They are normal parts of successful communication and can be taught explicitly alongside sound recognition and production.

4. Meaningful input and purposeful attention

4.1 Exposure is necessary but not sufficient

Learners need substantial contact with spoken language to build familiarity with vocabulary, pronunciation patterns and discourse. Yet minutes of audio played are not the same as learning achieved. Material that is almost entirely incomprehensible may offer few usable cues, while endlessly repeating trivial material may fail to extend ability. The instructional task is to select input that is meaningful, manageable and sufficiently challenging to support growth.

Recorded interviews, podcasts, stories and live conversations offer different benefits. Recordings permit controlled replay and close analysis; live conversation allows clarification and adaptation. A balanced course can use both. A transcript can help connect acoustic forms with known words, but providing it before the first attempt may turn a listening task into reading. A sequence of listening, focused transcript analysis and renewed listening can preserve the auditory goal while adding support.

A useful distinction is between listening for a communicative purpose and listening merely to identify isolated forms. Students might first listen to decide which solution a colleague recommends, then revisit a difficult phrase that changed the meaning. This retains a real reason to understand while allowing targeted attention to form. The two goals should reinforce rather than displace each other.

4.2 Noticing and instructional judgment

Exposure does not guarantee that every relevant feature is noticed. A learner may follow the overall message while missing a grammatical ending, stress contrast or reduced pronunciation. Equally, a task focused exclusively on form may become detached from the communication it is intended to support. Teachers must direct attention without destroying meaningful engagement.

A brief intervention after a task can address a recurring difficulty. If a learner repeatedly confuses quantities, the teacher can isolate number contrasts and test them in new contexts. If the learner misunderstands disagreement because of intonation, comparing spoken alternatives may help. The intervention should be tied to an observed need rather than inserted merely because a grammar point appears in the syllabus.

This requires restraint. Constant interruption may reduce willingness to speak and overload attention. Never correcting anything may leave persistent problems unaddressed. The teacher must decide when to protect the flow of communication and when to pause for a teachable moment. That decision should be revisited as the learner becomes more independent.

4.3 The importance of curricular balance

Nation's (2007) four-strands framework divides course activity into meaning-focused input, meaning-focused output, language-focused learning and fluency development. It is an important counterweight to any slogan suggesting that listening and speaking alone are sufficient. Learners also require vocabulary growth, form-focused attention and opportunities to use familiar language with increasing efficiency.

A listening-and-speaking priority need not exclude reading and writing. Students can read a briefing before a discussion, write a reflection after a meeting simulation or use written feedback to notice recurring problems. These activities support oral communication while developing literacy in its own right. The aim is a coherent learning ecology, not a rigid hierarchy in which two skills receive all the attention.

For an exam candidate, the balance may differ from that of a professional preparing for meetings. A learner targeting academic writing needs substantial writing practice even if speaking remains a personal concern. Individualised teaching should reflect such differences transparently. Research-informed instruction aligns methods with goals and evidence instead of applying the same preferred technique to everyone.

5. Speaking, retrieval and fluency

5.1 The demands of production

Speaking requires a learner to decide what to communicate, choose words, organise them grammatically and articulate the message. In spontaneous conversation these processes overlap with listening to another person and planning the next turn. Even when vocabulary and grammar are known, retrieving them quickly enough can be difficult. Hesitation may reflect processing demands rather than lack of understanding.

The educational implication is that knowledge must be practised under conditions resembling intended use. Controlled exercises can establish a form, but learners also need opportunities to select that form while attending to meaning. A progression can move from preparation to guided production and then to tasks with less predictable language and genuine information exchange. Teachers should observe whether clarity survives as planning time decreases.

Skehan (2009) discusses complexity, accuracy and fluency as related but distinct dimensions of second-language performance. A learner who speaks faster is not necessarily more accurate; a learner attempting complex structures may hesitate more. Assessment should consider these trade-offs rather than relying on one global impression of improvement.

5.2 Fluency is not simply speed

Fluency is often confused with talking quickly. Effective speech involves appropriate pacing, manageable pauses, connected ideas and responsiveness to a listener. A rapid stream of poorly organised language may be less useful than a slightly slower explanation that listeners can follow. The goal is purposeful communication, not maximum words per minute.

Teachers can distinguish hesitation caused by searching for ideas from hesitation caused by linguistic retrieval. A learner discussing an unfamiliar technical issue may need planning time rather than a fluency drill. Another who repeatedly pauses before common phrases may benefit from meaningful repetition of those phrases in varied contexts. These difficulties call for different responses.

Progress measures can include the length of meaningful turns, responses to follow-up questions, successful paraphrase when a word is missing and clarity of explanations. These observations may be more useful than a vague judgment that the learner “sounds more fluent”. They also make feedback actionable and allow teachers to choose appropriate next tasks.

5.3 Repetition with variation

Repetition can help learners access familiar language more efficiently, but repetition without a purpose can become mechanical. A learner might explain a process to a supportive teacher, explain it again to a different listener with fewer prompts, and finally answer unexpected questions about it. The core topic remains familiar while the communicative demands change. This can consolidate language and gradually increase flexibility.

Variation matters because success in rehearsed performance does not establish transfer. A learner who memorises an interview answer may perform smoothly until the interviewer asks something unexpected. A course should include supported preparation and opportunities to handle novelty. The contrast helps identify which language is accessible and which remains dependent on rehearsal.

Reflection after a task can also support development. The learner identifies a moment of hesitation, reformulates the intended message and practises a more efficient expression. The teacher selects a limited number of high-value targets rather than supplying an overwhelming list of corrections. Performance becomes information for the next teaching decision.

6. Interaction, feedback and participation

6.1 Conversation as a learning environment

Conversation is not simply a test of speaking. It creates opportunities to negotiate meaning, ask questions, receive clarification and notice differences between intended and understood messages. These opportunities are valuable when the interlocutor adjusts language in response to difficulty. The learner may hear a reformulation, compare it with their own attempt and try again where the meaning matters.

Mackey and Goo (2007) synthesise research on interaction and acquisition, supporting serious attention to interaction as a potential learning mechanism. Yet the evidence does not guarantee that any amount of conversation produces the same result. A friendly discussion can be enjoyable without addressing recurring difficulties. A highly corrective exchange can be informative but discourage participation. Task design and teacher judgment influence whether interaction becomes productive learning.

A practical distinction is between conversation that continues and conversation that occasionally reveals a useful gap. If a student says “He explained me the problem”, the listener may understand. The teacher can decide whether the form merits attention immediately or after the task. The goal is not error hunting but connecting feedback with meaningful use.

6.2 Feedback that leads somewhere

Oral corrective feedback can take different forms. A teacher may reformulate an utterance, request clarification, explicitly identify a problem or prompt self-correction. Lyster, Saito and Sato (2013) review these practices and emphasise the importance of instructional context and learner perceptions. No single feedback technique should be assumed optimal for every target or learner.

Feedback can concern meaning, organisation, pronunciation, pragmatics or task strategy. A learner preparing for an interview may benefit more from comments on clarity and relevance than correction of every article error. Another whose pronunciation causes repeated confusion about numbers may need targeted sound practice. Teachers should explain why a particular target has been chosen.

A useful feedback cycle includes evidence, explanation, practice and reassessment. The teacher identifies a difficulty, provides a manageable intervention and creates a later opportunity to use the improved form. Without that opportunity, feedback may remain a comment rather than a change in performance. Students should know what to improve, how to improve it and when they will try again.

6.3 Confidence and participation

Willingness to communicate depends on more than linguistic competence. Learners may avoid speaking because they fear embarrassment, believe their accent is unacceptable or expect every mistake to be judged. A classroom can be demanding while remaining respectful. Teachers should make clear that repair, clarification and hesitation are normal parts of communication.

Small-group formats may provide more opportunities to participate, but are not automatically superior in every respect. Group dynamics, teacher skill, task design and the number of active participants matter. A pair activity dominated by one learner may offer less useful practice than a well-managed larger class. The relevant quality indicator is how participation and feedback are organised, not just advertised group size.

In online tuition, responsiveness is especially important because some social cues are reduced or delayed. Clear turn-taking, visible listening, appropriate wait time and consistent feedback routines make interaction easier to manage. Continuity may help learners attempt unfamiliar language, but its benefits should be evaluated rather than assumed.

7. Listening and speaking as a coordinated system

7.1 Why the two skills belong together

Listening and speaking are often separated in textbooks, but real conversation alternates between them continuously. A learner listens to interpret another person's intention, responds, monitors the response and adjusts. Speech production can prompt new input, while listening supplies models and information that influence later production. Coordinated instruction reflects a basic feature of communicative use.

Coordination does not mean every lesson must contain equal minutes of each skill. A learner with acute listening difficulties may need intensive perceptual work; another may understand well but struggle to organise extended speech. The point is to consider progress in one domain alongside demands in the other. Focused listening analysis can lead into a discussion requiring newly noticed phrases, while a speaking task can lead into listening to contrasting models.

The “point of convergence” is a useful metaphor for the experience of understanding and expression reinforcing one another. It is a pedagogical image, not a neurological event or scientifically established threshold. Development is gradual and uneven. The metaphor is credible when it describes an experience without pretending to identify a single measurable instant at which English becomes effortless.

7.2 Transfer must be tested

An improvement in recognising a phrase during a controlled listening exercise may not transfer immediately to a noisy meeting. Accurate pronunciation in a prepared sentence may not appear in spontaneous conversation. A learner who performs confidently with a familiar teacher may hesitate with a new speaker. These differences reveal that performance conditions matter.

Teachers can design transfer checks. After practising reduced forms with one voice, learners can listen to a different speaker. After explaining a familiar topic, they can answer unexpected questions. After pronunciation practice, the teacher can evaluate intelligibility during an information-gap task. Such checks prevent premature claims of mastery and help select the next intervention.

Listening and speaking are not the sole foundations of proficiency. Reading can enrich vocabulary and discourse knowledge; writing can support reflection and precise formulation. For some learners these activities may strengthen oral performance, while transfer may be limited for others. Cross-skill benefits should be investigated rather than guaranteed.

7.3 An integrated teaching cycle

An integrated cycle can begin with meaningful listening, continue with analysis of a difficult phrase and move into a speaking task where that phrase serves a communicative purpose. Feedback identifies a few issues, and later practice checks whether the learner can use the language with less support. The sequence combines exposure, noticing, production and assessment without treating any single stage as sufficient.

For example, a professional might listen to colleagues disagree politely about a deadline. The teacher checks factual comprehension and the cues signalling disagreement. The learner then practises expressing a reservation in a simulated meeting, receives feedback on clarity and intonation, and tries again in a different scenario. The connection between listening and speaking is purposeful rather than merely decorative.

The same architecture can be adapted for younger learners, exam candidates and advanced professionals. Contexts, vocabulary and support differ, but the principle remains: listen for meaning, notice what matters, use language purposefully and test improvement under new conditions.

8. Individual differences, accessibility and challenge

8.1 Why one method cannot fit every learner

Learners differ in language background, prior schooling, motivation, goals and opportunities to use English. They also differ in how much support they need when listening, reading or speaking under pressure. A course suitable for a confident intermediate professional may not suit a beginner or a learner with specific processing difficulties. Personalisation should influence tasks, pace, feedback and assessment rather than merely the topics selected.

Initial assessment should explore real needs. Someone who reports “I cannot understand native speakers” may struggle with reduced forms, unfamiliar accents, vocabulary, attention or several factors together. A teacher can sample short tasks, compare performance with and without transcripts and ask where comprehension failed. The objective is an educational profile, not a medical diagnosis.

Speaking difficulties require equally careful interpretation. Hesitation may reflect anxiety, limited vocabulary, pronunciation concerns or difficulty organising ideas. Treating every pause as a fluency problem risks selecting the wrong intervention. Teachers need evidence from different tasks and should revise the plan as new information emerges.

8.2 Dyslexia and accessible design

Some learners benefit from adjustable text presentation, reduced visual distraction and explicit support connecting written and spoken language. However, dyslexia is not a single learning profile, and no font, colour or digital tool should be presented as a universal solution. Accessibility options increase learner control and reduce avoidable barriers; their usefulness must be judged individually.

In combined listening-and-reading tasks, teachers can separate the demands of decoding print from interpreting speech. A student may understand an oral explanation but struggle with dense written instructions, or the reverse. Alternative ways to access instructions can make assessment fairer without changing the intended learning objective. This is an instructional design principle, not a claim that any interface has been clinically validated.

C2's developing Focus tools illustrate how reading assistance and evidence-based comprehension activities might be incorporated into a learning environment. Sentence focus, adjustable spacing and evidence selection are design choices that warrant usability testing and outcome evaluation. They should not be described as treatments for dyslexia or as proven to improve examination scores without appropriate evidence.

8.3 Challenge without overload

Instruction must balance productive difficulty with the learner's ability to engage. If a task is so demanding that the message cannot be identified, additional pressure may offer little benefit. If it is always effortless, it may fail to reveal gaps or develop flexibility. Appropriate challenge depends on the learner, skill and stage of instruction.

Teachers can vary one demand at a time: speaking rate, listening length, number of speakers, topic familiarity or unpredictability of follow-up questions. This helps identify what caused difficulty. When several demands increase simultaneously, it becomes harder to know whether a poor result reflects language knowledge, task complexity or momentary overload.

Learners should also have a voice in selecting support. Some prefer transcripts after an initial listening; others find text distracting. Some welcome immediate correction; others benefit from delayed feedback. Preferences do not prove effectiveness, but they matter for engagement and should be considered alongside observed outcomes.

9. Live online language teaching

9.1 Affordances and limitations

Online tuition can provide access to teachers and peers who would otherwise be geographically unavailable. It can support screen sharing, controlled playback, collaborative documents and flexible scheduling. These affordances are useful only when they serve a learning purpose. A lesson crowded with digital features is not necessarily more effective than a well-managed conversation with clear feedback.

For listening, digital tools make short segments easy to replay and compare. For speaking, teachers can structure turn-taking, role-play and pronunciation feedback, and where appropriate review recordings with consent and privacy safeguards. The instructional value lies in attention, practice and feedback rather than the novelty of the technology.

Online communication also introduces constraints. Audio compression, microphone quality, delay and background noise affect comprehension. Teachers must distinguish technical problems from language ability. A learner who cannot hear a consonant contrast because of poor audio should not be judged as having a perceptual deficit. Technical reliability is part of fair assessment.

9.2 Continuity and teaching teams

A designated lead teacher can provide continuity by tracking goals, recurring difficulties and progress. Additional teachers may expose learners to different voices, interaction styles and pronunciation patterns. These are plausible advantages of a coordinated team, but the model works only when communication between teachers is systematic and guidance remains coherent.

A useful teaching record identifies the current objective, evidence from recent tasks, feedback targets and what the next teacher should reinforce. Without such coordination, supplementary lessons may become disconnected experiences. A learner should not have to explain the same needs from the beginning every time they meet a new teacher.

Research does not establish that a particular staffing arrangement is inherently superior in all circumstances. Team teaching is an organisational choice whose effectiveness depends on planning, teacher expertise, communication and learner experience. Schools should evaluate whether the additional perspectives produce measurable benefit rather than assuming that more teachers automatically means better results.

9.3 Independent practice between lessons

Live lessons provide interaction and expert feedback, but many learners need more opportunities to consolidate language between meetings. Independent practice can involve focused listening, vocabulary review, pronunciation discrimination, short speaking recordings or reflection on feedback. The task should be connected to current learning goals and manageable within the learner's available time.

Digital applications can support repeated practice, immediate responses and varied task formats. Yet an app is not a substitute for human judgment in every domain. Automated scoring may identify certain features while missing meaning, appropriateness or the effect on a real listener. Practice tools should therefore be presented as supplements to instruction, with limitations explained clearly.

The quality of independent practice depends on what the learner actually does. Completing a large number of exercises does not prove improvement if tasks are too easy, feedback is ignored or skills fail to transfer. Teachers can assign a small number of purposeful activities and revisit the result in the next lesson. This closes the loop between independent work and live teaching.

10. From research to the C2 classroom

10.1 Thoughtful assessment

The research reviewed here suggests that successful instruction begins by identifying what the learner can and cannot yet do under relevant conditions. At C2, an initial conversation and appropriate assessment can inform the starting plan. A learner who wants to participate in meetings needs a different performance sample from a teenager preparing for a speaking examination. The teacher should consider listening, production, interaction, accuracy and the learner's own account of difficulty.

Assessment should not be presented as an infallible measurement of an underlying trait. Performance varies with topic, fatigue, task familiarity and interlocutor. A single score is useful only as one part of a broader picture. Teachers should use early lessons to test assumptions and revise the plan when evidence suggests a different need.

Where an assessment identifies listening difficulty, the next question is what kind of difficulty. Is the learner missing familiar words in connected speech? Misinterpreting a speaker's intention? Losing track of extended explanations? Different answers lead to different interventions. Personalisation becomes meaningful when it changes the teaching decision.

10.2 A lead teacher and supporting specialists

C2's model assigns a lead teacher to follow the learner's progress and coordinate the course. Support teachers can provide additional opportunities for listening and pronunciation work and expose students to different voices. The educational rationale is continuity combined with varied interaction, not a claim that this exact staffing arrangement has been scientifically proven superior.

For the model to work, teachers need shared objectives and a practical record of recent evidence. A support teacher should know whether the current target is number recognition, connected speech, intonation or spontaneous response. The lead teacher should be able to see whether the additional practice changed performance. Otherwise the learner may experience variety without coherent progression.

Students should also understand the purpose of each activity. If a listening session focuses on reduced forms, the teacher can explain why this matters for real conversation. If a speaking task revisits a familiar topic, the reason may be to reduce retrieval demands and develop flexibility. Transparent objectives help learners see the course as a connected journey rather than isolated appointments.

10.3 Individual, paired and mini-group tuition

C2 offers individual, semi-individual and mini-group formats rather than crowded classes. These formats can make it easier to observe performance and allocate speaking time, but their value depends on task design and teacher skill. An individual lesson can be highly personalised; a paired task can create meaningful information exchange; a mini-group can introduce multiple perspectives and varied interaction.

Each format has trade-offs. Individual tuition may offer more immediate feedback but fewer peer interaction patterns. A pair can generate authentic negotiation but may be affected by mismatched proficiency or participation. A mini-group can offer richer conversation but requires careful facilitation. Choosing a format should involve learner goals and practical constraints, not a universal claim that one arrangement is always best.

The promise of small teaching formats is an opportunity for responsiveness. Teachers can adapt the task when a learner is overwhelmed, increase challenge when performance becomes secure and notice patterns that might be missed in a crowded setting. Those opportunities must be converted into actual practice to become educational advantages.

10.4 Feedback that informs the next lesson

C2's emphasis on feedback is strongest when comments lead to a new task and an observable change. A teacher might identify a recurring difficulty in recognising reduced forms, provide short focused practice and revisit the same contrast in a new recording. A speaking task might reveal unclear word stress; the teacher can model alternatives and then evaluate whether the learner communicates more clearly in a subsequent conversation.

Feedback should be selective and intelligible. A long list of corrections can overwhelm learners, especially when they are trying to build confidence. A few priorities connected to communicative consequences are often more actionable. Teachers should distinguish errors that obstruct meaning from those that are noticeable but currently less important to the learner's goals.

The student also has a role in evaluating progress. A short reflection can identify which expressions felt easier, which listening segments remained difficult and what practice was useful. Such reflection should complement rather than replace performance evidence. Learners may feel more confident before objective measures change, or improve in a narrow task without yet noticing broader benefits.

10.5 The private learning environment

C2's developing student area and Learning Lab can provide structured practice between lessons. Exercises involving vocabulary, collocations, sentence construction, listening or pronunciation can reinforce specific targets. The educational advantage comes from alignment: the student knows why an activity was assigned and the teacher knows what to do with the result.

Not every digital activity measures communicative competence. A word-level checker can inform vocabulary awareness but does not establish that a learner can use the word spontaneously. A sentence-building exercise can strengthen controlled production but may not transfer immediately to conversation. A reading-support interface can improve access to text without necessarily improving listening. The school should be explicit about what each tool does and what remains to be assessed through live interaction.

This distinction is also important for responsible claims about technology. An application may be useful, enjoyable and carefully designed without having undergone controlled validation. C2 can explain its design rationale and gather learner feedback while avoiding statements that imply scientific endorsement of the software itself.

11. Limitations and competing perspectives

11.1 What the evidence does not prove

The research discussed here supports attention to listening processes, interaction, pronunciation and feedback, but it does not prove that listening and speaking must always be prioritised above reading and writing. The appropriate balance depends on the learner's objectives. An academic researcher may require advanced reading and writing more urgently than conversational fluency; a professional preparing for client calls may need the opposite emphasis.

Nor does research on particular tasks automatically validate an entire course model. A meta-analysis of pronunciation teaching does not demonstrate that every pronunciation app improves spontaneous speech. A study of metacognitive listening instruction in French does not establish identical outcomes for all English learners. A synthesis of interaction research does not prove that small classes always outperform larger ones.

Measurement choices matter. Improvement on a controlled pronunciation item, a listening comprehension test and a spontaneous conversation represent different outcomes. Teachers and schools should avoid combining them into a vague promise of universal progress. Better claims specify what changed, under which conditions and whether the result transferred to a new task.

11.2 Alternative explanations

Learners improve for many reasons: increased exposure, independent study, changing motivation, growing familiarity with task formats and experiences outside class. When a student progresses during a course, it is difficult to attribute the change to one component without an appropriate comparison. A lead teacher, extra listening practice and learner motivation may all contribute, but the relative contribution cannot be inferred from an individual success story.

Research also contains disagreements about the role of explicit instruction, the conditions under which feedback is effective and the degree to which controlled practice transfers to spontaneous performance. A credible pedagogy should remain open to revision. Teachers should use research to sharpen questions and select plausible interventions, not to freeze a method into dogma.

Finally, language learning is social and personal. Confidence, identity, time, fatigue and access to meaningful interaction affect participation. A technically precise exercise may fail if it is irrelevant or discouraging; an enjoyable conversation may support engagement without producing immediate measurable gains. Good teaching must balance these considerations while retaining clarity about intended outcomes.

12. Conclusion: from knowledge to confident communication

Listening and speaking deserve systematic attention because real-time communication places distinctive demands on perception, retrieval, articulation and interaction. Research on listening processes, metacognitive instruction, pronunciation and feedback provides useful principles for designing learning activities. It also warns against simplistic promises: no single exercise, teacher arrangement or technology guarantees spontaneous fluency.

A defensible teaching approach identifies the learner's real needs, provides meaningful exposure, directs attention to consequential difficulties, creates opportunities for purposeful production and checks whether improvements transfer to new contexts. Reading and writing remain important partners in this process. The strongest claim is not that two skills magically unlock all others, but that understanding and expressing spoken English can reinforce one another when taught with care.

At C2, these principles inform a commitment to personalised assessment, a coordinating lead teacher, supporting teaching expertise, small teaching formats, feedback and independent practice. These are research-informed design choices, not independently validated outcomes of the C2 model. The ultimate test is whether learners become more capable of understanding others and making themselves understood in the situations that matter to them.

13. Worked examples: what research-informed teaching looks like

13.1 The learner who knows the words but cannot hear them

Imagine an adult learner who reads business English confidently but repeatedly misses short phrases in meetings. Their teacher notices that the learner can explain the meaning of a sentence when shown a transcript, yet cannot recognise it at normal speed. The initial hypothesis is not that the student lacks vocabulary, but that the acoustic form of familiar words is insufficiently accessible. This distinction changes the intervention.

The teacher begins with a short recording containing a relevant phrase. The learner listens without a transcript and writes what they believe they heard. The teacher asks which words were clear and which were inferred. Only then is the transcript introduced. Together they compare the expected pronunciation with the actual connected speech, paying attention to stress, reductions and boundaries. The learner listens again and then encounters the same pattern in a different voice.

This sequence is compatible with Field's (2008) process-oriented account of listening. It does not establish that the student has mastered the feature. A follow-up task is needed: can the learner recognise the pattern in a new recording without textual support? A later conversation can test whether recognition survives competing demands. Progress is demonstrated through performance across contexts, not simply through a feeling that the explanation made sense.

The teacher might assign a brief independent activity using carefully selected clips. The assignment should have a specific purpose: identify three instances of a target reduction and record which one remained difficult. At the next lesson, the teacher checks whether the learner can hear the feature in unfamiliar speech. This closes the loop between classroom diagnosis and home practice.

13.2 The learner who understands but hesitates to speak

A second learner follows conversation comfortably but produces very short responses. They know vocabulary and grammar, yet spend considerable time searching for expressions. It would be easy to prescribe more grammar exercises, but the evidence suggests a different priority: accessible retrieval and opportunities for extended meaningful turns.

The teacher selects a familiar topic and asks the learner to explain a decision. After the first attempt, the teacher identifies two expressions that would make the explanation more efficient. The learner practises them briefly and repeats the explanation to a different listener or with a different constraint. Later, the teacher asks an unexpected follow-up question. The progression introduces novelty while retaining enough familiarity for the learner to build fluency.

Skehan's (2009) discussion of complexity, accuracy and fluency helps frame the evaluation. The learner may initially speak more smoothly but use simpler language. That is not necessarily a regression; it may reflect a different allocation of attention. The teacher should assess whether the learner can gradually increase both flexibility and clarity, rather than expecting every dimension to improve simultaneously.

Feedback should also address the learner's communicative goals. If the explanation is understandable but too indirect for a workplace setting, the teacher can model a concise alternative. If hesitation arises from uncertainty about how to disagree politely, the next lesson might focus on interactional language rather than speed. Personalisation is a process of revising hypotheses about what the learner needs.

13.3 The learner whose pronunciation affects understanding

A third learner speaks willingly but is sometimes misunderstood when giving numbers or technical information. The teacher begins by collecting examples rather than assuming that the learner's accent is the problem. Which words caused confusion? Did listeners mishear a vowel contrast, a stress pattern or a sequence of digits? Was the audio connection clear? Did the listener have enough context?

The teaching goal is intelligibility, consistent with Derwing and Munro's (2005) emphasis. The learner practises a small set of consequential contrasts and then uses them in a task where the listener must act on the information. The listener's response provides evidence of communicative success. This is more informative than asking the learner to repeat a word until the teacher approves of its sound.

The teacher can measure controlled improvement and spontaneous use separately. Saito and Plonsky's (2019) findings caution against assuming that gains on monitored pronunciation tasks will automatically appear in spontaneous speech. A student may perform accurately while focusing on the target but revert to an earlier pattern during a complex explanation. The next teaching decision should be based on that observation, not on an exaggerated claim of mastery.

An effective lesson can also teach repair. If the listener misunderstands a quantity, the speaker can repeat it more clearly, rephrase it or confirm it in another way. Repair strategies protect communication while pronunciation continues to develop. They are a mark of competence rather than an admission of failure.

13.4 The learner preparing for an examination

Examination candidates need a different balance of priorities. A learner preparing for a speaking component must understand the task format, organise responses, interact appropriately and demonstrate language ability under time constraints. Practice should resemble examination demands without reducing the course to rehearsed scripts. A memorised answer can create false confidence if the examiner asks a slightly different question.

Listening preparation also needs process work. Repeatedly completing past-paper questions may reveal a score trend but not the source of persistent mistakes. A teacher can classify errors: missed sound contrasts, unknown vocabulary, incorrect inference, failure to track reference or loss of attention during a long passage. The resulting intervention should target the difficulty while retaining sufficient exam-format practice.

Reading and writing remain essential in examination preparation. An IELTS candidate cannot compensate for weak writing simply by becoming a confident speaker. A balanced course must honour the assessment criteria and the learner's target scores. The argument for integrated oral development is therefore a complement to, not a substitute for, preparation across the required components.

13.5 The younger learner developing confidence

Children and teenagers bring different developmental, motivational and educational needs. A young learner may participate readily in games but struggle to monitor a listening strategy explicitly; another may benefit from a simple reflection routine. Goh and Taib's (2006) small-scale study suggests that process-oriented listening instruction can be meaningful for younger learners, but its limited sample does not justify broad claims about all ages.

For younger learners, the teacher can use concrete questions: What did you hear? Which word helped you? What would you listen for next time? Short tasks with clear outcomes may be more appropriate than lengthy theoretical explanations. Spoken interaction can involve storytelling, information gaps, collaborative problem-solving or explaining a choice to a partner.

Confidence should not be confused with the absence of challenge. A child who enjoys every activity may still need explicit support with vocabulary, pronunciation or listening. Equally, an activity that produces frequent errors may be too demanding to support productive engagement. Teachers should observe both participation and performance and adapt the balance as the learner develops.

14. A framework for evaluating progress responsibly

14.1 Define outcomes before choosing activities

A course that promises “better English” without defining the target makes evaluation difficult. Better listening might mean recognising familiar words in connected speech, following longer explanations, understanding a wider range of accents or responding accurately to questions. Better speaking might mean clearer pronunciation, longer meaningful turns, more flexible vocabulary or more effective interaction. These outcomes overlap but are not interchangeable.

Teachers should begin with a small number of observable goals. For example, a professional learner might aim to summarise the main decisions in a five-minute meeting extract and ask appropriate clarification questions. Another might aim to explain a familiar technical process without a prepared script. Such goals can be assessed with repeated tasks that are comparable but not identical.

A single score should not be treated as a complete portrait. Task familiarity, topic knowledge and confidence can affect performance. A useful evaluation combines multiple samples and notes the conditions under which success occurred. It should distinguish supported performance from independent performance and controlled tasks from spontaneous communication.

14.2 Monitor process and product

Product measures include comprehension accuracy, successful completion of a speaking task and whether a listener understood the intended message. Process evidence includes where comprehension broke down, which strategies the learner used, whether feedback was incorporated and how much support was required. Both matter. A learner can answer correctly through a plausible guess, or answer incorrectly while demonstrating improvement in a previously weak perceptual skill.

Process data should not become intrusive surveillance. Teachers need only information that helps them make better decisions. A brief note on recurring listening errors or useful pronunciation targets may be more valuable than a large dashboard of loosely interpreted metrics. Where digital systems record attempts, schools should explain what is collected, why it is needed and how it is protected.

Learner reflection can complement teacher observation. Students can report which tasks felt easier and where they remain uncertain. These reports provide information about confidence and perceived difficulty, but they should not replace evidence of performance. Someone may feel more fluent without becoming more intelligible, or improve objectively before noticing the change.

14.3 Test generalisation

A central risk in language teaching is mistaking practice effects for broader learning. Students often improve on familiar materials, especially when they know the format and likely answers. The more important question is whether the skill transfers to a new voice, topic, interlocutor or communicative situation. Teachers should therefore include periodic unfamiliar tasks that test generalisation.

Transfer can be examined progressively. After targeted listening practice, introduce a different speaker or recording. After pronunciation drills, assess the target in a spontaneous explanation. After repeated speaking practice, change the listener or task constraints. A learner who succeeds only in the original exercise has made a narrower gain than one who can use the skill flexibly.

Delayed checks are also useful. A performance improvement observed immediately after feedback may not persist. Returning to a target in a later lesson can reveal whether it remains accessible. The appropriate interval and task depend on the learner and objective; there is no universal schedule established by the studies reviewed here.

14.4 Avoid false precision

Educational measurement is vulnerable to numbers that appear more precise than they are. A pronunciation score generated by software may look authoritative even when its relationship to listener comprehension is unclear. A speaking-speed metric may increase while accuracy or organisation deteriorates. A listening percentage may conceal whether errors were caused by vocabulary, perception or inference.

The measurement framework discussed by Saito and Plonsky (2019) is a reminder that the construct being measured matters. A controlled production score and a spontaneous intelligibility judgment answer different questions. Teachers should choose measures aligned with the learner's aims and explain their limitations rather than combining everything into a single unexplained index.

For a small school, evaluation need not be complicated to be responsible. A clear initial sample, a documented learning objective, periodic comparable tasks and a short account of observed change can provide a useful basis for discussion. Such records also make it easier for a lead teacher and supporting teachers to coordinate instruction.

14.5 Evidence-informed does not mean evidence-complete

No school can derive every classroom decision directly from a definitive experiment. Teachers work with individual learners whose goals and circumstances rarely match research samples exactly. Evidence-informed practice therefore combines published findings, professional judgment, learner preferences and observation of local outcomes. The aim is to make decisions transparent and revisable, not to pretend that uncertainty has disappeared.

When evidence is strong for a narrow outcome, teaching claims should remain narrow. When evidence is mixed or indirect, the school can describe the rationale without claiming proof. When a digital tool is still in development, it can be presented as an emerging support resource rather than a validated intervention. These distinctions strengthen credibility and protect learners from unrealistic expectations.

The long-term ambition should be a cycle of improvement: identify a need, select a plausible intervention, collect meaningful evidence, review the result and adjust. This is compatible with an individual, personal teaching philosophy and does not require the school to imitate a large automated platform. What matters is that care and expertise become visible in the quality of instructional decisions.

15. Questions for further research and classroom inquiry

15.1 Which listening difficulties respond best to targeted instruction?

A major question for practitioners is how to distinguish difficulties caused by speech perception from those caused by vocabulary, inference or attention. Field's (2008) analysis provides a rationale for process-oriented diagnosis, but teachers still need feasible ways to identify the source of a particular failure. Future classroom inquiry could compare a small number of targeted perceptual exercises with ordinary replay and comprehension checking, using new recordings to assess whether any improvement transfers.

Such inquiry should be modest in its claims. A learner who recognises one reduced phrase after practice has demonstrated improvement on that item, not a general transformation in listening proficiency. Repeated samples across speakers, topics and time would be needed before stronger conclusions were justified. Teachers can nevertheless use carefully recorded observations to improve the fit between diagnosis and instruction.

15.2 When does pronunciation practice become spontaneous ability?

Pronunciation instruction often produces visible progress in monitored tasks. The harder question is when, and under what conditions, that progress becomes available during spontaneous communication. Saito and Plonsky's (2019) synthesis makes this distinction central. Future work should attend to how controlled practice, meaningful interaction, varied interlocutors and delayed assessment contribute to generalisation.

For classroom practice, this suggests separating three milestones: the learner can perceive the target feature, produce it with attention and use it while focusing mainly on meaning. Each milestone is worthwhile, but the third is closest to the learner's real communicative goal. A teacher who documents all three can offer more precise feedback than one who reports a single pronunciation score.

15.3 How should different teachers coordinate?

Coordinated teaching teams are appealing because continuity and varied interaction may complement each other. Yet more teachers do not automatically mean better outcomes. Research is needed on how shared goals, teacher notes, feedback consistency and learner relationships affect progress. A school could begin by examining whether supplementary sessions reinforce the targets identified by the lead teacher and whether learners perceive the course as coherent.

The most useful measures would be connected to learner outcomes rather than staff activity. The number of teachers involved is not itself an educational achievement. Evidence might include improved recognition of varied voices, more successful interaction with unfamiliar speakers or better retention of feedback targets. Such inquiry would help distinguish the genuine benefits of a team from the appearance of additional support.

15.4 What do independent digital tools add?

Practice applications can offer repetition, immediate responses and flexible access. Their educational value, however, depends on the construct they address and whether improvement extends beyond the app. A vocabulary exercise may strengthen recognition while leaving spontaneous retrieval unchanged. A pronunciation activity may improve controlled production without changing intelligibility in conversation. Evaluation should compare the specific skill practised with the learner's broader goal.

A practical question is whether independent tasks are more useful when teachers select them in response to observed needs and review them afterwards. This is plausible, but the advantage should not be assumed. Schools can compare completion, learner engagement and subsequent performance when tasks are linked to teaching goals versus assigned generically. Privacy, accessibility and the clarity of automated feedback should form part of that evaluation.

15.5 What would count as success for a learner?

The answer depends on the learner. A professional may value being able to challenge a proposal politely in a meeting. A parent may value understanding conversations at school. A teenager may value making friends and participating in class. An exam candidate may need performance that satisfies specific assessment criteria. Research-informed teaching should respect these different outcomes instead of imposing one idealised image of fluency.

Success also includes knowing how to respond when communication is imperfect. Asking for clarification, checking a number, paraphrasing an unfamiliar term and adapting an explanation to a listener are valuable abilities. A learner who can manage misunderstanding may participate more effectively than one who speaks quickly but cannot repair a breakdown. These interactional outcomes deserve attention alongside accuracy and fluency.

The most credible conclusion is therefore both ambitious and restrained. Listening and speaking can be taught deliberately, supported by research and evaluated with care. They are central to many learners' goals, but they are not magic keys that make every other skill effortless. The purpose of teaching is to make real communication more accessible through thoughtful, responsive and evidence-informed practice.

References

The bibliography below contains ten identifiable publications checked against publisher, DOI or university records. This is a selected narrative review, not a systematic review of all available evidence. References are linked to original records where possible.

1. Derwing, T. M., & Munro, M. J. (2005). Second language accent and pronunciation teaching: A research-based approach. TESOL Quarterly, 39(3), 379–397. Source ↗
2. Field, J. (2008). Listening in the Language Classroom. Cambridge University Press. Source ↗
3. Goh, C., & Taib, Y. (2006). Metacognitive instruction in listening for young learners. ELT Journal, 60(3), 222–232. Source ↗
4. Graham, S. (2017). Research into practice: Listening strategies in an instructed classroom setting. Language Teaching, 50(1), 107–119. Source ↗
5. Lyster, R., Saito, K., & Sato, M. (2013). Oral corrective feedback in second language classrooms. Language Teaching, 46(1), 1–40. Source ↗
6. Mackey, A., & Goo, J. (2007). Interaction research in SLA: A meta-analysis and research synthesis. In A. Mackey (Ed.), Conversational Interaction in Second Language Acquisition (pp. 407–453). Oxford University Press. Source ↗
7. Nation, P. (2007). The four strands. Innovation in Language Learning and Teaching, 1(1), 2–13. Source ↗
8. Saito, K., & Plonsky, L. (2019). Effects of second language pronunciation teaching revisited: A proposed measurement framework and meta-analysis. Language Learning, 69(3), 652–708. Source ↗
9. Skehan, P. (2009). Modelling second language performance: Integrating complexity, accuracy, fluency, and lexis. Applied Linguistics, 30(4), 510–532. Source ↗
10. Vandergrift, L., & Tafaghodtari, M. H. (2010). Teaching L2 learners how to listen does make a difference: An empirical study. Language Learning, 60(2), 470–497. Source ↗

The C2 perspective