How do we learn?

Ask students to think of a lesson that an instructor explained very well. Then ask what the student could do independently afterwards. Turn their responses into observable verbs such as explain, compare and justify. Student-centred learning does not mean the instructor withdraws; appropriate guidance is needed. Build an example together: a student can compare two transport mechanisms in terms of energy use and direction. Later, you will ask how this activity could be assessed.

Use this as an ungraded activity and do not publicly expose individual answers. Everyone should think independently first, then compare their explanation with a neighbour. Treat identifying what is not yet known as a positive learning behaviour. When students reopen the source, ask them to note how their explanation differs rather than simply copying the correct answer. This brief experience is not, by itself, a research finding; it is a classroom exercise demonstrating the retrieval practice method we will use later.

Do not limit the example to exam scores or competitions for rewards. Choose an observable behaviour, such as attaching a genuine source to every claim. Discuss how the consequence following a behaviour might affect how often it occurs in the future. Explain that not all feedback technically counts as reinforcement; its effect must be evident in the behaviour. Emphasise that students should be asked not only for the correct answer but also why it is correct, and that this approach alone does not explain all learning.

Ask students to imagine an overloaded slide: unfamiliar terms, moving images and a long spoken explanation all at once. Ask which information should come first. Invite suggestions for explaining a mechanism by presenting its basic structure before adding detail. Show the transition from a worked example to a partially completed example. Do not confuse simplification with giving inaccurate information. Emphasise that the support needed may change as students' prior knowledge grows, and that a single arrangement will not suit everyone.

Choose a familiar concept such as osmosis. Ask students to write their own explanation, then compare it with a source or model. Put the first and revised explanations side by side. Constructing meaning does not make every view scientifically correct; sources and feedback matter. Ask students to explain which connection they changed and why.

Invite a student to talk through their source-searching process: I am checking the title, opening the publication and looking for the section relevant to the claim. Make their decisions visible, not just the result. Ask about the difference between copying a peer and learning by understanding the process. Give quieter students time to write first. A group's success should not depend on one person doing everything; rotate roles and ensure each student can explain the reasoning behind a decision.

Do not present the four approaches as competing boxes. One lesson may include a brief drill, a structured explanation, a discussion of prior knowledge and peer modelling. To prevent groups from choosing only tools they enjoy, ask what learning task each suggestion enables. Consider how much support beginners need. A method does not necessarily require an app or video; a reasoned comparison on paper can also be a powerful task.

Examine the evidence for the claim without criticising the person who made it. A student may like visuals, but preference and improved achievement are different outcomes. Ask groups to consider comparing different presentation formats for the same learning goal using a common assessment. Keep accessibility requirements distinct from learning-style labels. Support for vision or hearing must not be reduced to a preference test. Invite students to conclude by recommending multiple representations appropriate to the topic alongside active learning tasks.

The answer should be more than a single word; ask students to explain a mechanism or relationship. Provide a brief rationale after the response. Ask learners to identify why they made an error: confusion between concepts, missing knowledge or misreading the question. Explain that an error is not a sign of personal inadequacy; it is information that guides the next study decision. Effects depend on context, so avoid suggesting that one exercise guarantees lasting learning. Plan another attempt later.

You can suggest an example schedule of tomorrow, three days later and a week later, but explicitly state that these are not universally optimal intervals. Groups should try to create different questions rather than reread the same text. Ask first for a definition, then a comparison, then application to a new example. An incorrect answer may call for an earlier revisit. Keep the total time realistic; even a one-minute explanation can help students see their progress.

Ask students to consider the difference between tracking time and monitoring learning. Sitting at a desk for two hours is not, by itself, evidence of understanding. Elicit signs such as explaining without the source, solving a new example or giving a reasoned answer to a friend's question. If a plan is not completed, examine the scope of the task, available support and circumstances rather than blaming the person. Explain that the cycle is individual but can be strengthened by peer and instructor support. Ask everyone to write one monitoring question.

Revise an example together on screen: replace ‘I will study the cell membrane’ with ‘I will compare two types of transport using examples’. Add a checkpoint in which the student closes the source and produces a brief explanation. Let the student choose the date of the next attempt. A peer should check whether the plan is observable. Scale down overambitious goals. Comparing estimated confidence with an actual answer can be a useful opportunity for reflection; do not present this activity as a psychological test or a precise prediction of success.

Keep Miller's framework separate from the list of learning theories; here the focus is on assessment and levels of performance. Give first-year examples involving definitions, reasoning and demonstration in a safe educational setting. Explain that performance in real patient care requires different observation and responsibility. Give students four example cards and ask them to justify the appropriate level for each. Focus on which evidence supports which claim rather than memorising level names. Do not infer comprehensive competence from a single test.

Ask groups to write a learning outcome, then identify the student work they would need to see to determine whether it has been achieved. Producing an attractive presentation alone may not demonstrate understanding of the topic. Add a ‘why’ question for an explanation goal and observation criteria for a demonstration goal. You do not have to award a formal grade at this stage; the aim may be feedback during learning. Each group should have a one-sentence success criterion. Make it clear and visible before students begin.

Do not reduce evidence-based medicine to finding papers. Ask who took part in the study, what was measured and under what conditions the findings might apply. Use an example about a learning method rather than making a real treatment decision. A study conducted with physicians does not guarantee the same effect in first-year students. Explain how research, expertise and the person's circumstances are considered together. Stating uncertainty explicitly is part of scientific thinking.

Use this as an educational research question, not a clinical treatment recommendation. Ask students to mark the population, intervention, comparison and outcome in different colours. Discuss the difference between choosing satisfaction as the outcome and choosing learning performance. Make the timing explicit too: an immediate outcome differs from one measured a week later. PICO is not a compulsory template for every research question; here it is a tool for making thinking concrete.

Show why a title or abstract alone may be insufficient. Ask students to record who was studied, which method was used and which outcome was measured. Briefly explain why the sample, comparison, participant attrition and timing of measurement matter. You do not need to teach detailed statistics in this session. The aim is to avoid making a claim stronger than its source supports. Each group should add at least one limitation or uncertainty beside a claim it has checked. In the three-minute lecture slot, use only item B from Source checking: 30 seconds to decide, 60 seconds to justify, 90 seconds to compare and discuss the limit. Complete all four items in the workshop.

Have groups write a brief draft first, then ask AI for an alternative explanation or question. This prevents the tool from taking over the starting point of their thinking entirely. The prompt should include the learner level, learning outcome, time available and expected product. Asking it not to invent sources is not a guarantee. Remind students that the output must be checked against real sources. Groups without access can do the same task using a peer's draft. Record the tool's name, the date and its contribution.

Label this as an example claim written for teaching; do not present it as a live response from a real model. One group can check whether the source exists, another whether it supports the claim and another the limits of the generalisation. Then ask them to write one revised explanation. To discourage trusting the first link they encounter, ask students to visit the publisher or original record. Removing an unverifiable statement or marking it as uncertain is an acceptable outcome.

Explain that removing names is not always enough; a combination of details may identify someone. For this workshop, use examples invented from scratch rather than real records. Note that tools may have different data retention practices. Discuss how AI can produce fluent but incorrect answers and why responsibility for errors cannot be transferred to the tool. Provide an equivalent route for students without access. Present disclosure of the assistance used in a product as normal academic practice.

Create three columns, but do not force every task onto just one person. For source verification, the student does the work, the instructor provides methods and feedback, and the institution supports access and policies. To prevent shared responsibility from becoming unclear responsibility, ask groups to identify who takes the first step for each task. Ask how students can challenge a decision, request help or report a problem. This matrix is a local workshop design, not an official allocation of responsibilities developed by WHO.

Groups can be formed according to the number of devices available. Roles may include facilitator, recorder, source checker, timekeeper and presenter. In smaller groups, roles can be combined and can rotate during the activity. Ask students to write the topic and learning outcome first, then complete the activity, checking question, sources and responsibilities. Introducing the tool alone is not enough in the presentation: groups must explain their decisions, their reasons and what they changed after feedback. Keep the time visible.

Ask everyone to complete the exit ticket individually; do not finish with only the group presenter's perspective. Students should record a useful method, an idea they revised and the date of their next attempt. Return to the initial goal and ask what evidence has been produced. Keep a separate list of unresolved questions. Remind students that what transfers when groups change is not just a file, but the habit of checking sources and explaining decisions. State that the plan will be reviewed in the next activity.