Skip to content

The Vibe Learning framework

Vibe Learning uses generative AI as a tutor, study partner, and project collaborator. It is personalized, interactive, adaptive, conversational, project-based, and structured around human learning.

Core principles

1. Personal Learning Profile (PLP)

Describe your current knowledge, difficulties, available time and explanations you want to try. These preferences can change with the task; they are not fixed learning-style categories.

2. Multi-angle explanation

Ask for stories, metaphors, step-by-step breakdowns, real-world scenarios, and beginner-to-expert versions of the same idea.

3. Productive struggle

Try first. Show your reasoning. Ask AI to critique your thinking before it gives the answer.

4. Project-based learning

Each concept should lead to a small project, artifact, or output. AI helps break work into milestones.

5. Memory and retention loops

Use quizzes, active recall, spaced repetition, summaries, compression, and teach-back methods to convert short-term understanding into long-term memory.

person working with notes, laptop, diagrams, and AI-assisted thought bubbles in a modern bright workspace. This image contains a clean white background, editorial layout, and abstract shapes suggesting active problem solving and learning.

The 4 pillars of a Vibe Learning plan

  • Your Personal Learning Profile
  • A defined skill goal
  • A structured routine with daily micro-rituals
  • A weekly deep-work cycle for reinforcement and review

The knowledge system

  • Compression — distill concepts into clear notes
  • Storage — organize your learning artifacts
  • Retrieval — make your notes easy to review and search
  • Synthesis — connect ideas together
  • Application — use knowledge in real tasks and projects

How does the framework connect to learning research?

Vibe Learning brings familiar learning practices into an AI-assisted workflow. Retrieval practice means bringing an idea to mind before looking at the answer. Feedback then helps identify gaps; a different example checks whether learning transfers.

Try it: Explain a concept from yesterday without your notes, check the explanation, then apply it to a new situation. Record what changed rather than treating fluency as proof of mastery.

For the research background to retrieval practice, see Pooja K. Agarwal’s explanation and research resources. This supports a component of the approach, not a claim that the complete Vibe Learning framework has been independently validated.

Practise retrieval in Recall & Transfer Lab →

YOUR FRAMEWORK LEARNING PATH

Explore an idea. Make it your own.

Follow the eight steps or open the activity you need today. These offline activities use authored guidance, not an AI connection.

  1. 1. Discover the framework
  2. 2. Explore the loop
  3. 3. Compare approaches
  4. 4. Build a profile
  5. 5. Productive struggle
  6. 6. Knowledge system
  7. 7. Apply the framework
  8. 8. Independent challenge

Nothing entered here is sent to AI or analytics. Export a backup before clearing browser data. Preferences describe support for this task, not fixed learning-style categories.

The eight-stage learning loop

Select a stage, try its micro-activity and connect it to your task.

Orient: Choose a useful question and an observable outcome.

A clear destination helps you decide what to practise and what evidence to collect.

Example: Instead of “learn spreadsheets,” aim to total a small expense table and explain how the range works.

Common mistake: Starting a long AI conversation without deciding what you want to be able to do.

Suggested AI coaching prompt

Ask me what I want to do independently. Help me narrow the outcome before suggesting a plan.

Reflect: Which part of your goal is still too vague?

Practise orient with a related resource →

Understand: Build an initial explanation using examples, contrasts and questions.

An explanation gives you a model to test; it is not yet evidence that you can use the idea.

Example: Compare a SUM formula with manually adding each cell, then explain what the range includes.

Common mistake: Treating a clear explanation as proof of your own understanding.

Suggested AI coaching prompt

Explain one example, then ask me to explain why it works. Wait for my response.

Reflect: What term or causal link can you not yet explain?

Practise understand with a related resource →

Predict: State what you expect before seeing the result.

A prediction makes your current model visible and gives feedback something specific to address.

Example: Before running a formula, predict whether adding a row outside the range changes the total.

Common mistake: Looking at the answer first and then claiming it was obvious.

Suggested AI coaching prompt

Give me a small case and ask for a prediction with reasoning. Do not reveal the result yet.

Reflect: What result would challenge your current explanation?

Practise predict with a related resource →

Apply: Make an attempt that requires the idea.

An attempt exposes decisions and gaps that passive reading may hide.

Example: Create your own expense table and write the formula without copying the worked example.

Common mistake: Asking AI to produce the entire artifact before trying a manageable part.

Suggested AI coaching prompt

Let me attempt this task first. If I get stuck, ask one guiding question rather than completing it.

Reflect: Which step was yours, and where did you need support?

Practise apply with a related resource →

Feedback: Compare your attempt with evidence or criteria and decide what to revise.

Feedback becomes useful when it informs an action you can take.

Example: A manual check reveals that the final expense row was omitted. Revise the range and test again.

Common mistake: Collecting comments without making or checking a revision.

Suggested AI coaching prompt

Compare my attempt with these criteria. Identify one important gap and ask me to repair it.

Reflect: Did the revision address the reason for the error?

Practise feedback with a related resource →

Recall: Reconstruct an idea without the answer visible.

Recall reveals what is available without immediate cues and helps target later review.

Example: Close the example and explain how to choose the start and end of a formula range.

Common mistake: Recognizing an answer on a card and counting that as retrieval.

Suggested AI coaching prompt

Ask a short-answer question. Wait for my attempt before showing a comparison answer.

Reflect: Which cue did you rely on before closing the source?

Practise recall with a related resource →

Connect: Relate the idea to other concepts, tasks or prior knowledge.

Connections can help you choose when an idea is relevant, provided you explain the relationship.

Example: Connect formula ranges to the idea of boundaries in a loop: both can miss the final item.

Common mistake: Drawing a dense map without being able to explain any link.

Suggested AI coaching prompt

Ask me to connect this idea to one thing I already know, and test whether the analogy has limits.

Reflect: Does the connection clarify the idea or merely sound clever?

Practise connect with a related resource →

Transfer: Use the idea in a new situation and inspect what needs adaptation.

A different context checks whether you can recognize and use the relevant structure.

Example: Build a summary for a differently arranged table and explain how you chose the range.

Common mistake: Repeating the identical worked example and calling it independent mastery.

Suggested AI coaching prompt

Give me a different case using the same principle. Let me choose the method and explain my reasoning.

Reflect: What transferred, what changed and what remains untested?

Practise transfer with a related resource →

Passive or active? Compare the decisions.

Explore six authored situations. Compare an immediate answer with an attempt, hint, revision and independent check.

Build a practical learning profile

Choose the right amount of help

Support should fit the task and the observed gap. A full explanation is sometimes appropriate; productive struggle does not mean withholding needed help.

Rewrite: “Give me the complete answer so I can finish.”

Build a usable knowledge system

Follow a percentages example through all five stages, then build a workflow for your own topic.

Compression

Distill meaning without removing necessary conditions.

Worked example: From a source on percentages: “A percent compares an amount with a named base.”

Storage

Keep the explanation with its source and a useful retrieval cue.

Worked example: Store the note under “percentages,” with the worked example and a link to its source.

Retrieval

Try bringing the idea to mind before reopening the note.

Worked example: Ask: why is a change from 20% to 30% not simply a 10% relative increase?

Synthesis

Connect ideas while preserving distinctions.

Worked example: Connect percentages to ratios, and distinguish percentage points from relative change.

Application

Use the knowledge in a useful task and check the result.

Worked example: Compare two fictional price changes, name the base and explain the difference to a colleague.

A scenario-based self-check

Eight questions sample useful decisions. They do not establish mastery.

You can follow an AI explanation but have not tried a task. What next?
Your project is polished, but AI made every important decision. What evidence is missing?
A learner is stuck on one identifiable step. What support is proportionate?
Your notes are neatly organized but hard to recall. What change helps inspect recall?
A confident AI answer cites a source. What should you check?
Your learning goal is “understand AI.” What makes it more usable?
A familiar method fails in a new situation. What should you inspect?
A short quiz went well. What can you conclude?

Independent challenge

Choose a real task. Define a criterion, make a first attempt, use one targeted piece of support, then complete a different example unaided. Explain what the evidence supports and what remains uncertain.

Your framework workbook

Export includes your notes, stage activities, profile and generated workflows.