In The Principles of Instructional Design, Robert Gagné describes learning as a combination internal and external events. In other words, the executive processes of the brain combined with external instructional events. He characterises short-term memory (STM) as being limited in both duration and capacity. He puts forward the idea that this can only be extended through what he calls “internal rehearsal.” In his book, he lists the internal executive processes:
• Reception of stimuli by receptors
• Registration of information by sensory registers
• Selective perception for storage in short-term memory (STM)
• Rehearsal to maintain information in STM
• Semantic encoding for storage in long-term memory (LTM)
• Retrieval from LTM to working memory
• Response generation to effectors
• Performance in the learner’s environment
• Control of processes through executive strategies

He goes on to argue that instruction is best understood as a deliberately organised sequence of external events intended to support these internal learning processes. This idea underpins his Nine Events of Instruction:
• Gain attention – Research shows attention is the gateway to learning. If learners are not engaged in the first 30 seconds, retention drops dramatically. Gagnè argued this must be intentional, not accidental. Open with a provocative question or a striking statistic, or use a short video clip showing a real-world problem. Present a culturally surprising or counter-intuitive scenario. Another strategy could be to include an interactive poll or quick challenge. Begin a communication module by playing a clip of a negotiation going badly wrong, and then ask learners to spot the moment trust broke down.
• Inform learners of objectives – Learners need to know what’s expected of them before instruction begins. Clearly stated objectives create a mental framework (schema) that helps learners organise incoming information and understand what success looks like. In language courses, tie objectives to authentic communicative tasks: ordering in a restaurant, conducting an interview, resolving a complaint call. Learners engage far more when the outcome feels real and personally useful. Tie objectives to real-world tasks learners actually perform.
• Stimulate recall – New learning is built on existing knowledge. Activating prior knowledge creates hooks in long-term memory where new information can attach. Without this bridge, learners must work much harder to make sense of new content. For language training, retrieval practice works best when embedded in context. Ask learners to use previously learned vocabulary or grammar in a new situation rather than isolated recall tests.
• Present the stimulus – This is the core instructional event where new knowledge, concepts, or procedures are introduced in context. Gagnè emphasised that content must be logically sequenced, appropriately chunked to respect working memory limits, and presented through multiple modalities. Break content into lessons of 5–7 key points maximum. Combine narration with supporting visuals (not identical text) in multimedia. Apply the worked example effect for procedural content and sequence from simple to complex, concrete to abstract. Use learner-paced navigation with clear signposting
• Provide guidance – Guidance helps learners process and encode content by providing structure, examples, analogies, mnemonics, and worked demonstrations. Gagnè distinguished this from content presentation, defining guidance as the bridge between receiving information and being able to use it. In language training, ‘noticing’ is a powerful guidance strategy, drawing learners’ attention to specific features of authentic language input (intonation pattern, politeness marker, hedging expression) before asking them to produce similar output.
• Elicit performance – Learning requires doing. This event prompts learners to actually perform the skill or apply the knowledge, not just observe it. Practice encodes new learning into procedural or long-term memory through active retrieval and production. It is a good idea to implement scenario based simulations requiring realistic decision. Make sure that tasks are sequenced: easy → medium → complex application, starting with drag-and -drop, sorting or ordering before role play exercises with branching options. Avoid fill-in-the-blank exercises for higher-order communication skills.
• Provide feedback – Feedback confirms correct performance and corrects errors before they become entrenched habits. Gagnè specified that effective feedback must be informative, going beyond basic correct/incorrect feedback, explaining why a response was right or wrong and how to improve. This needs to be embedded into the learning platform, so that it can deliver feedback for right and wrong answers. Without feedback, learners cannot self-correct, and errors become ingrained.
• Assess performance – Assessment verifies that learners have achieved the stated objectives. Unlike practice (Event 6), assessment is evaluative as it measures competence without feedback or scaffolding. Gagnè argued assessment must match the objective in both content and performance conditions. Assessment that does not match objectives creates misleading learner data. It is important to use multiple assessment points, not just a final exam.
• Enhance retention & transfer – The final event ensures that learning moves from short-term working memory into long-term memory, and that learners can apply it beyond the course context. Gagnè advocated spaced practice, varied application, and real-world transfer activities. Without deliberate retention strategies, 70% of learning is forgotten within 24 hours (Ebbinghaus Forgetting Curve). Transfer is the ultimate goal of all training, and if skills reach the workplace, organisational value is created. The single most powerful retention strategy for language is authentic use outside the course.

Merrill’s First Principles of Instruction are a research-based framework for designing effective learning around real-world tasks. The model shows that learners learn best when instruction activates prior knowledge, demonstrates new content, supports practice, and helps them transfer learning into their own context. There are very strong similarities between Merrill’s principles and the Present, Practice and Produce (PPP) methodology in English Language Training.

• Problem-centred – instruction organised around meaningful real-world tasks.
• Activation – learners activate prior knowledge and experience.
• Demonstration – examples and models show how to perform the task.
• Application – learners practice the task with guidance.
• Integration – learners apply new skills in their own contexts and reflect.

Constructive alignment rests on two connected ideas: constructivism (learners build meaning through active engagement) and alignment (objectives, teaching, and assessment must fit together). Misalignment, where one thing is taught, another is assessed, and something else is stated as the objective, is one of the most common design flaws. Biggs also developed the SOLO taxonomy to describe the quality and depth of understanding that objectives and assessments should aim for.

• Aligned objectives – learning outcomes should clearly show what learners are expected to understand, perform, or produce.
• Aligned teaching – classroom activities, practice tasks, and materials should directly prepare learners for the intended outcomes.
• Aligned assessment – assessment should measure the same knowledge, skills, or performances described in the objectives.
• Coherent system – objectives, teaching, and assessment should support each other rather than work in separate directions.
• SOLO taxonomy – levels of understanding can help shape objectives and assessments at the right cognitive depth, from simple to complex.

• Pre-structural – the learner misses the point and gives irrelevant or very limited answers.
• Uni-structural – the learner focuses on one relevant idea, but the response is still basic and narrow.
• Multi-structural – the learner includes several relevant ideas, but they are listed separately rather than connected.
• Relational – the learner links the ideas together into a coherent whole and shows deeper understanding.
• Extended abstract – the learner generalises the idea, transfers it to new contexts, and thinks at a higher level of abstraction.

• Misalignment is one of the most frequent and expensive problems in instructional design.
• If a goal cannot be assessed directly, it probably should not be the main focus of the course.
• Alignment checks should be completed before content development starts.
• Strong alignment makes learning more transparent for learners and easier to evaluate for teachers and designers.

ELT methodology often moves from tightly guided input to freer language use. In Present – Practice – Produce (PPP), the teacher presents the target language, guides learners through controlled practice, and then asks them to produce the language more independently. ESA follows a similar logic but is more flexible: learners first engage with the topic, then study the language, and finally activate it in a freer communicative task.

• Notice – learners are exposed to the target language in context and begin to identify meaning, form, and use.
• Practise – learners work through guided, controlled tasks that help them handle the target language accurately.
• Use – learners move into freer production where they apply the language more independently and communicatively.
• Scaffolding fades – support is gradually reduced so learners take more responsibility for language production.
• Independent output – learners eventually use the target language in a more open, realistic task.

• Presentation – the target language is introduced in context through a text, dialogue, video, or example.
• Practice – learners complete controlled activities such as multiple choice questions, matching, gap-fills, categorising, or true/false tasks.
• Production – learners use the language more freely in speaking or writing tasks with less teacher control.

Extended PPP

• Warm-up – set the topic, activate prior knowledge, and introduce the lesson goals using pictures, questions, quizzes, or vocabulary tasks.
• Acquisition – present comprehensible input through texts, audio, video, or example sentences, with attention to the target language.
• Guided Application – provide highly scaffolded tasks with clear answers and strong support, such as matching, ordering, categorising, multiple choice, true/false or gap fills.
• Independent Application – reduce support and let learners make more choices with tasks such as sentence transformation, sentence building, sentence completion, short writing, discussion, or role-play.
• Activation – move into open tasks such as writing, role plays, presentations, simulations, storytelling, reports, or AI-supported conversations.
• Feedback – give delayed or immediate feedback on learner performance and highlight useful language, errors, and improvement points.

Dewey argued that learning should grow out of experience, reflection, and problem solving. Knowledge becomes more meaningful when learners actively connect new ideas to real situations rather than receiving information passively.

• Experience – learning begins with meaningful activity and interaction with the world.
• Reflection – learners think about what they did, what happened, and why it mattered.
• Problem solving – tasks should involve authentic problems rather than isolated facts.
• Continuity – each learning experience should build on previous experience.
• Social learning – learning is strengthened through communication and collaboration with others.

Maslow proposed that human motivation is shaped by a hierarchy of needs, ranging from basic survival needs to higher-level goals such as self-esteem and self-actualisation. In learning, this means motivation and engagement are influenced by whether learners feel safe, supported, and valued.

• Physiological needs – basic comfort and well-being must be in place before deeper learning can happen.
• Safety needs – learners need a secure environment where they are not afraid of embarrassment or failure.
• Belonging – people learn better when they feel included and connected to others.
• Esteem – confidence, recognition, and a sense of competence support motivation.
• Self-actualisation – learners are more engaged when learning supports personal growth and meaningful goals.

Dörnyei’s model explains language learning motivation through the learner’s future self-image, current learning experience, and sense of desire or obligation to learn. It is especially useful in language education because it connects motivation to identity, goals, and imagined use of the target language.

• Ideal L2 self – the learner’s vision of the person they want to become in the target language.
• Ought-to L2 self – the learner’s sense of obligation, expectations, or external pressure to learn.
• L2 learning experience – motivation is shaped by the quality of the learning process itself.
• Future orientation – learners stay more motivated when they can picture a meaningful future use of the language.
• Identity and emotion – language learning is tied to how learners see themselves and how they feel while learning.

Gardner proposed that intelligence is not a single ability but a set of different capacities that people may develop in different ways. In education, this encourages teachers to vary how learners access material, practise skills, and show understanding.

• Linguistic intelligence – sensitivity to words, meaning, and language use.
• Logical-mathematical intelligence – reasoning, pattern recognition, and problem solving.
• Spatial intelligence – thinking in images, layout, and visual relationships.
• Bodily-kinaesthetic intelligence – learning through movement and physical activity.
• Interpersonal and intrapersonal intelligence – understanding other people and understanding oneself.

Behaviourism explains learning as a change in observable behaviour shaped by stimulus, response, repetition, and reinforcement. In language learning, it treats language as a set of habits that can be built through imitation, drilling, correction, and practice.

• Stimulus–response – learners respond to prompts, models, or cues in a predictable way.
• Repetition – repeated practice helps form automatic language habits.
• Reinforcement – correct answers and successful responses are strengthened through praise, feedback, or reward.
• Imitation – learners copy accurate language models before producing language independently.
• Error correction – mistakes are treated as habits that should be corrected quickly.

Andragogy explains how adults learn differently from children. It emphasises self-direction, experience, relevance, problem solving, and immediate application, because adult learners usually want to understand why something matters and how they can use it right away.

• Self-direction – adults prefer to have some control over how they learn.
• Experience – adult learners bring rich prior knowledge that should be used in learning.
• Readiness – adults learn best when content fits current roles, goals, or problems.
• Relevance – learning should feel practical and immediately useful.
• Problem-centred learning – adults are more motivated by real tasks than by abstract information alone.

Generative grammar explains language as a rule-based system that allows speakers to produce and understand an unlimited number of new sentences. Rather than memorising fixed phrases, learners internalise patterns and structures that let them create language creatively and accurately.

• Rule system – language is organised by underlying grammatical rules.
• Creativity – speakers can generate sentences they have never heard before.
• Universal Grammar – humans are thought to share an innate capacity for language structure.
• Deep and surface structure – meaning and sentence form are connected but not identical.
• Competence – the theory focuses on the internal knowledge of language, not just observable performance.

• Twelve evidence-based principles for designing effective multimedia instruction that respect cognitive architecture.
• 01: Coherence – People learn more deeply when extraneous material is excluded rather than included. This means removing interesting but irrelevant stories, background music decorative graphics, irrelevant details, and unnecessary text, and what Mayer called seductive details, even when they seem engaging. If they do not support the learning goals or use the target language, they have to go. Every element in e-learning competes for a limited working memory capacity. Even entertaining or interesting content that is irrelevant to the learning goals consumes cognitive resources. Ruthlessly audit every asset in your e-leaning module: images, audio, video, or text. Does it directly help learners understand the target concept. If not, delete it! This seems counter intuitive for designers who equate richness of content with quality.
  A pronunciation module on English stress patterns does not need a picture of a London city street. The cultural backdrop is a seductive detail which activates knowledge about London rather than stress patterns competing with the target learning.
• 02: Signalling – People learn more deeply when cues are added that highlight the organisation and key ideas of the material. Signalling includes structural signals (outlines, headings, lists, bullet points, transitions), emphasis cues (bold text, arrows, colour highlighting), and pointer words in narration (The key point here is…).
  In language e‑learning, use headings and numbered lists to make structure explicit. Use colours consistently to signal the same type of material throughout. Use bold fonts to highlight a key term when it is first defined. Use arrows or callouts to direct attention in complex diagrams. Use discourse markers to structure information. In a narrated video with slides combine verbal and visual signalling. Display a simple numbered list or bullet points, but use discourse markers in the audio channel to add verbal clues. The main point is to be totally consistent throughout the learning product.
• 03: Redundancy – People learn better from graphics and narration than from graphics, narration, and identical on‑screen text. When narrated text is simultaneously displayed word-for-word, learners must process the same content twice in the visual channel (reading text AND view the graphic), creating redundant processing that overwhelms the working memory. In language e‑learning, pair visuals with spoken language rather than duplicating every sentence as on‑screen text; use short captions only where necessary. Avoid narrating bullets word-for-word while they’re visible. The most common redundancy is when trainers/presenters read bullet points aloud. Replace bullet text with keywords or headings, let the narration carry the majority of meaning. Subtitles should not be displayed automatically, and should only be supplied for accessibility purposes. Transcripts should only be used for post-listening reviews.
• 04: Spatial contiguity – People learn more deeply when corresponding words and pictures are near each other rather than far from apart on the page or screen. When explanatory text and a related graphic are spatially separated, learners must use working memory to hold one in mind while scanning to find the other. The spacial search consumes resources needed for learning. This is the split attention effect that Sweller also wrote about. Place cations directly below or beside the image they describe. Label diagram components with callouts, not a separate legend at the bottom. Position the explanatory text immediately adjacent to the graph or figure. Use leader lines (arrows) to connect text labels directly to the diagram elements. In language training, the most common violation of this principle is when there is a text on one slide, and then questions on the slide after. It is important to place the text and the relevant questions on the same slide.
• 05: Temporal contiguity – People learn better when corresponding narration and visuals are simultaneous rather than successive. This is similar to spatial contiguity but relates to only to time. It is also an important aspect of presenting slides.
  In language e‑learning, it is better to time audio explanations so they play while learners can see the relevant text or image, instead of explaining long after the visual is gone. The synchrony of action and explanation is a critical design decision.
• 06: Segmenting – People learn more deeply from a multimedia lesson when it is presented in learner-paced segments rather than as a continuous unit. Divide long listening or video tasks into short scenes, and let learners start, pause, and replay each segment. Add learner controlled “continue” or “next” buttons between segments. Allow learners to restart or replay individual segments without restarting the whole lesson. Use progress bars to indicate to learners where they are in a sequence of activities. A common issue in language training is where a listening exercise uses a 20 minute authentic podcast. Break it down into 5 minute listening segments each with a focus question, and a “listen again” / “continue” choice.
• 07: Pre‑training – People can learn better if they already know names, Key vocabulary and concepts. Pre-teaching reduces the cognitive demand during the main part of the lesson by ensuring that learners are not simultaneously learning vocabulary AND concepts. For example, pre‑teach critical vocabulary (key negotiation phrases) before doing a complex business meeting role play. Before a listing activity for medical professionals, pre-teach 8 key clinical terms with images. When the learners then hear the actual consultation they recognise the vocabulary immediately and can devote full attention to other aspects like discourse stucture, information flow and pragmatic meaning.
• 08: Modality – People learn better from graphics with spoken words than from graphics with on‑screen text. When a visual is accompanied by narration, the two processing channels (visual for the graphic, auditory for the narration) are used complementary. When text replaces narration, the visual channel must handle both the graphic AND the text, creating visual channel overload. In language e‑learning, use visuals plus audio narration or dialogues, keeping written text minimal so you don’t overload the visual channel. The bottom line is that narration should carry meaning that is not duplicated by the visuals. Together they should compliment each other and be greater than the sum of their parts.
• 09: Multimedia – People learn better from words and pictures than from words alone. This is Mayer’s foundational principle, and is the basis of multimedia e-learning. Visual are not a decoration, as they activate the pictorial processing channel and enable dual-channel encoding which produces richer and more transferable mental models. In language e‑learning: illustrate dialogues or texts with contextual images or short video clips that show the communicative situation.
• 10: Personalization – People learn more deeply when words are presented in conversation style rather than formal style. This includes using you, I or we in narration. This also mans using contractions and creating the social presence of a human conversation, even in self-paced e-learning without a live trainer. Write prompts and feedback in a friendly, learner‑centred tone, and use “you” to address the learner directly. Avoid referring to learners in the third person and using the passive voice or substituting “one” for “you.”
• 11: Voice – People learn well from a clear, friendly human voice rather than an AI generated voice. The quality, warmth, and perceived authenticity of the narrator activates social agency. Learners engage more when they perceive a real person is communicating with them. This is a real problem with text-to-speech voice generation, which often alienates learners. Use a narrator who sounds genuinely interested in the content, and avoid monotone text-to-speech generation. As AI voice technology improves, the principle is becoming more nuanced. There are AI voices around that now pass the “human threshold” for social agency activation. Many of these are real voices that have been cloned. This is further enhanced by the use of genuine looking video avatars.
• 12: Image Mayer found out that people do not necessarily learn more deeply from a multimedia lesson when the instructor’s image is added to the screen. Despite the many online courses that only provide narrator video, it more satisfying when the screen used for meaningful visuals (context, diagrams, etc.) over “talking head” video, unless the speaker’s image adds specific communicative and social value. In language training, it is better to use screen space primarily for instructional visuals, diagrams, and content. Consider picture-in-picture instructor video only for highly social or motivational content or trust building at the beginning of a course/module. If showing instructor, keep the image small (corner thumbnail) so content dominates.

Apply usability heuristics and user-centred design principles so that the interface supports learning instead of getting in the way. In language learning, every extra click, confusing label, or hidden control takes attention away from comprehension and practice.

• Visibility of system status – the system should always tell learners what is happening, such as progress, loading, feedback, or completion.
• Match between system and the real world – use language, icons, and layouts that feel familiar and natural to learners.
• User control and freedom – let learners pause, go back, replay, and recover from mistakes without feeling trapped.
• Consistency and standards – keep buttons, navigation, labels, and interactions the same across the course.
• Error prevention – design tasks and assessments so avoidable mistakes are less likely to happen in the first place.
• Recognition rather than recall – make options, instructions, and key controls visible so learners do not have to remember them.
• Flexibility and efficiency of use – support both beginners and faster learners with shortcuts, speed controls, and optional support.
• Aesthetic and minimalist design – remove clutter so only the information needed for learning stays on screen.
• Help users recognize, diagnose, and recover from errors – use clear feedback and plain-language error messages.
• Help and documentation – make help easy to find and focused on the learner’s current task.

• Usability testing reveals interface barriers that designers often miss.
• Navigation effort competes with the mental effort needed for learning.
• Audio controls such as replay, pause, and speed adjustment should always be easy to find.
• Consistency reduces cognitive load and helps learners focus on language, not interface rules.
• Assessments should include error prevention, such as review screens and confirmation before submission.

Universal Design for Learning applies the idea of universal design to education: instead of designing for a single “average” learner and adding accommodations later, you build flexibility into the course from the start. UDL is based on the idea that learners vary in how they perceive information, show what they know, and stay engaged, so learning should offer choices from the outset.

• Multiple Means of Representation – provide information in more than one way so learners can access content through audio, text, visuals, captions, diagrams, or interactive support.
• Multiple Means of Action and Expression – allow learners different ways to demonstrate understanding, such as speaking, writing, selecting, recording, drawing, or completing interactive tasks.
• Multiple Means of Engagement – give learners choice, vary challenge, and support motivation, persistence, and self-regulation so they can stay involved in the learning process.

• UDL is proactive: it reduces the need for individual accommodations by anticipating learner variability from the beginning.
• It makes learning environments more flexible, accessible, and inclusive for a wider range of learners.
• Captions, transcripts, adjustable pace, and multiple response options can help all learners, not only those with formal access needs.

Open Educational Resources are freely accessible teaching and learning materials released under open licences, usually Creative Commons. They can be retained, reused, revised, remixed, and redistributed, which makes them highly adaptable for different teaching contexts and learner groups.

• Retain – keep a copy of the resource for your own use.
• Reuse – use the resource in its original form in teaching or learning.
• Revise – adapt, edit, translate, or improve the material.
• Remix – combine the resource with other open materials to create something new.
• Redistribute – share the original or adapted resource with others.

• OER reduce duplication of effort and lower the cost of content creation.
• They support localisation, adaptation, and collaborative authoring.
• Quality can vary, so checking accuracy, currency, and pedagogical fit is always necessary.
• Discoverability is also important, because useful resources can be hard to find without good repositories and search tools.

• Learners who monitor, evaluate, and regulate their own learning processes achieve significantly better outcomes.
• Metacognition: (thinking about thinking) involves awareness of one’s own cognitive processes and the ability to regulate them. Flavell identified metacognitive knowledge (knowing about tasks, strategies, and oneself as a learner) and metacognitive regulation (planning, monitoring, and evaluating). Zimmerman’s self-regulated learning model adds motivational and behavioural dimensions: goal-setting, strategy selection, self-monitoring, and self-reflection in iterative cycles.

• Dual Coding Theory proposes that humans have two distinct cognitive systems: a verbal system that processes language and a non-verbal system that processes images, sounds, and sensory experiences. When both systems encode the same information simultaneously, recall and comprehension improve because there are two retrieval routes. The theory supports using visual aids, diagrams, and imagery alongside spoken or written language.

• Spaced practice (distributing study sessions over time rather than massing them) and retrieval practice (testing yourself rather than re-studying) are the two most empirically robust strategies for improving long-term retention. Roediger and Karpicke demonstrated the ‘testing effect’: taking a test on material improves retention far more than spending equivalent time re-studying. Bjork coined ‘desirable difficulties’—challenges that slow immediate performance but strengthen long-term retention.

• Jerome Bruner viewed learning as an active process in which learners build new knowledge from previous experience. He stressed discovery, scaffolding, and revisiting content in increasingly complex ways so learners can gradually develop deeper understanding. He introduced the idea of a spiral curriculum, where topics are revisited at increasing levels of complexity as the learner’s understanding deepens. Bruner’s believed that people learn best when they build new understanding from what they already know, rather than just memorising facts.

• Lev Vygotsky emphasised the Zone of Proximal Development (ZPD), which is the gap between what a learner can do independently and what they can achieve with guidance. This is closely related to Bruner’s concept of scaffolding, where a more knowledgeable other (teacher or peer) provides support to help the learner move through the ZPD. As the learner becomes more capable, this support is gradually withdrawn, allowing them to become independent.
• This is achieved with Scaffolding, which temporarily supports enabling performance within the ZPD.
• The focus is on Mediation & social interaction since learning is mediated by language and collaboration.

• Bandura argued that human learning occurs though observation and modelling rather than direct reinforcement. His four conditions for effective modelling are: attention, retention, reproduction and motivation.
His central ideas are:

• Observational learning – learning by watching models.
• Modelling – demonstrations provide behaviours to imitate.
• Self‑efficacy – beliefs about capability influence effort and persistence.

• Lave and Wenger challenged classroom-based learning models by showing that real learning happens through legitimate peripheral participation. Newcomers join communities of practitioners, observe and assist experts, and gradually take on more complex tasks until they are full participants. Learning is inseparable from practice; knowledge is not stored in heads but distributed across people, tools, and routines. Wenger later elaborated on Communities of Practice (CoPs) as groups united by a shared domain, community, and practice. CoP is the basis of most modern tailor made business language training. The trainers coordinate the language courses with a Subject Matter Expert (SME), and immerse themselves in the client’s CoP. This involves a longer and more in depth assignment.

Three Dimensions:
• Domain: Shared interest that defines the community.
• Community: Relationships and interactions among members.
• Practice: Shared repertoire of resources, stories, and tools.

Key Points:
• Legitimate peripheral participation: real but low-stakes early participation
• Learning is embedded in authentic practice, not decontextualised exercises
• Identity and belonging in the community are as important as knowledge
• CoPs require shared domain (what), community (who), and practice (how)
• Digital CoPs (Discord servers, LinkedIn groups) can support language learners

• Brown, Collins, and Duguid argued that schooled knowledge becomes ‘inert’ because it is learned in artificial contexts stripped of authentic use. Real expertise develops through activity in the context where knowledge will be applied, using the tools and social structures of the target community. Cognitive apprenticeship—their instructional model—extends traditional apprenticeship into cognitive domains through modelling, coaching, scaffolding, articulation, reflection, and exploration.

Key Points:
• Schooled knowledge often fails to transfer because it lacks authentic context
• Authentic activities are those practitioners actually perform, not school-like tasks
• Cognitive apprenticeship: model, coach, scaffold, articulate, reflect, explore
• Tools and social structures of a community are part of learning, not accessories
• TBLT (Task-Based Language Teaching) is situated learning applied to language

• Self-Determination Theory (SDT) distinguishes intrinsic motivation (driven by interest and enjoyment) from extrinsic motivation (driven by rewards or pressures). SDT holds that three basic psychological needs must be satisfied for intrinsic motivation to develop:
• Autonomy – feeling of volition and choice.
• Competence – feeling effective and capable.
• Relatedness – feeling connected to others.
• Intrinsic motivation is strongest when these three needs are supported.

• Expectancy-Value Theory holds that a learner’s motivation to engage in a task is jointly determined by their expectancy of success (Do I believe I can do this?) and their perception of task value. Task value has four components: Intrinsic value (Is it interesting?), Attainment value (Is it important to my identity?), Utility value (Will it be useful to me?), and Cost (What do I have to give up or risk?). Both low expectancy and low value independently reduce engagemen

• Flow is the state of complete immersion in an activity, which is characterised by loss of self-consciousness, distortion of time, and intrinsic enjoyment. Flow is produced when challenge level matches skill level exactly: too easy produces boredom; too hard produces anxiety. Csikszentmihalyi’s flow channel is a narrow band between these states. For flow to occur, goals must be clear, feedback must be immediate, and the person must have perceived control.

• Goal Orientation Theory distinguishes mastery goals (learning for growth, improvement, and intrinsic interest) from performance goals (looking competent compared to others or avoiding looking incompetent). Mastery-oriented learners seek challenges, persist through difficulty, and attribute failure to effort (fixable). Performance-oriented learners avoid challenge if failure risks their image and attribute failure to fixed ability. Dweck’s growth vs. fixed mindset theory is a popular application of this framework.

• Authentic assessment requires learners to demonstrate knowledge and skill in tasks that reflect real-world application, not school-like exercises.
  Characteristics of authenticity:
  • real or realistic audiences
  • genuine purposes
  • complex tasks requiring integration of multiple skills,
  • learner choice in approach
  • performance criteria that match professional standards
• Performance-based assessment is a related term emphasising observable demonstration rather than written tests.

• Level 1: Reaction – measures how learners feel about the training, including satisfaction, engagement, relevance, and perceived usefulness. This level looks at whether participants found the course clear, interesting, and worth their time.
• Level 2: Learning – measures changes in knowledge, skills, confidence, and attitudes. This level asks what learners actually gained from the course and whether they can demonstrate the intended learning outcomes.
• Level 3: Behaviour – measures whether learners apply what they learned back on the job. It focuses on transfer, looking for real changes in workplace performance after the training has ended.
• Level 4: Results – measures the broader organisational impact of the training. This may include improved communication, better customer service, fewer errors, stronger collaboration, or other outcomes that matter to the organisation.

• Jack Phillips extended Kirkpatrick’s model with Level 5: ROI. The ROI methodology isolates the impact of training from other business variables (control groups, trend analysis, expert estimation), converts impact to monetary value, tallies programme costs (direct, indirect, opportunity), and calculates ROI as a percentage. The approach requires agreement with business sponsors on measurement methodology before the programme begins. Intangible benefits (morale, engagement, culture) are reported separately.