← Think + AI Free tool · Grades 8–12

Thinking Moves

Twelve five-minute thinking tasks, each built on the topic you are teaching tomorrow. AI does the preparation on your side. Your students do the thinking on paper, whiteboards or out loud.

Build your task

Do not include student names or student work. The prompt only needs your subject, topic and grade.

Choose a move and fill in subject and topic to see your prompt.

Copied

The twelve moves

Each move trains one kind of thinking. Open a card to see two worked examples from different subjects and the evidence behind it. Think back closes every move and also works on its own.

Move 1

Many ways

Students produce as many different solutions as they can, then look for the most unusual.

Promising Divergent, associative and lateral thinking

Two worked examples and the evidence

Technology · Grade 8

Task
Find as many ways as you can to keep a packed lunch cold until noon without a fridge or electricity. Aim for ten. Then circle the two most unusual ones.
Look for
Ideas that use different principles, not ten versions of one: insulation (wool, foil, a vacuum flask), cold mass (a frozen bottle, an ice pack), evaporation (a wet cloth), shade and position, and changing the food itself (packing it frozen).
Probe
Which principle does each idea use? Is there a principle nobody in the class has used yet?
Think back
My best idea came after I stopped thinking about …

Music · Grade 10

Task
Take a melody everyone knows, such as "Happy Birthday". List as many ways as you can to make it sound sad without changing a single note of the melody.
Look for
Ideas across several musical elements: tempo, dynamics, instrument and timbre, minor chords in the accompaniment under the same melody, articulation, the rhythm of the accompaniment, and the setting it is played in.
Probe
Which musical element did nobody change? Try one idea from that element.
Think back
I got stuck on … until I switched to thinking about …

Evidence. Creativity training works on average (Haase, Hanel & Gronau 2023: g = 0.53), but that meta-analysis covers adults, and the evidence for school-age students is older and less certain. The label rates research on teaching this kind of thinking regularly, not a single five-minute move. Source: Haase, Hanel & Gronau 2023, Psychology of Aesthetics, Creativity, and the Arts

Use this move
Move 2

Pick and justify

Students agree on criteria first, then choose one option and defend it against each criterion.

Promising Convergent thinking

Two worked examples and the evidence

Geography · Grade 9

Task
A town must choose one of three sites for a wind farm: a hilltop beside a nature reserve, flat farmland 15 km from the nearest road, or a coastal strip next to a holiday village. Agree on three criteria first, then pick a site and justify it against each criterion.
Look for
Criteria set before the choice, such as wind exposure, the cost of roads and grid connection, and the impact on wildlife and people. Strong answers admit what their choice costs and say which criterion they weighted most.
Probe
Which criterion would have to change for you to pick a different site?
Think back
The criterion I almost forgot was …

History · Grade 10

Task
You want to know how ordinary Londoners felt during the Blitz in 1940. You have three sources: a private diary written that week, a newspaper report from the next morning, and a school textbook from 2010. Rank them for this question and justify the order.
Look for
A ranking tied to the question: the diary is closest to private feelings but shows one person; the wartime newspaper was shaped by censorship and the aim of keeping up morale; the textbook summarises with hindsight. Strong answers say what each source can and cannot show.
Probe
Would your ranking change if the question were "How many bombs fell on London?"
Think back
I changed my ranking when I noticed …

Evidence. In the Haase, Hanel & Gronau 2023 meta-analysis, training convergent thinking showed larger effects than training divergent thinking, but the studies are on adults. In school it is part of ordinary problem solving. The label rates research on teaching this kind of thinking regularly, not a single five-minute move. Source: Haase, Hanel & Gronau 2023, Psychology of Aesthetics, Creativity, and the Arts

Use this move
Move 3

Check the claim

Before deciding whether a claim is true, students say what evidence would convince them and what would not.

Strong Critical thinking

Two worked examples and the evidence

Biology · Grade 10

Task
Claim: "Cracking your knuckles gives you arthritis." Before deciding whether it is true, write down what kind of evidence would convince you, and what kind would not.
Look for
The difference between an anecdote ("my grandad did it and has arthritis") and a comparison of people who crack their knuckles with people who don't. Strong answers ask for a comparison group, a large sample and years of follow-up. Studies comparing the two groups have not found more arthritis among knuckle-crackers.
Probe
One doctor cracked the knuckles of one hand for more than 50 years and never the other. What does that show, and what can it not show?
Think back
The evidence I would have accepted too quickly was …

English · Grade 11

Task
An advert says: "80% of dentists recommend our toothpaste." List every question you would need answered before believing what the advert wants you to think.
Look for
Questions about who was asked, how many, the exact wording of the question, whether dentists could name more than one brand, and who paid for the survey. In 2007 the UK advertising regulator ruled a claim like this misleading because the dentists surveyed could recommend several brands.
Probe
Rewrite the advert so it says only what such a survey could prove.
Think back
The question I would not have thought of on my own was …

Evidence. Explicit teaching of critical thinking improves it across subjects and ages (Abrami et al. 2015: g = 0.30 over 341 effect sizes), most when combined with discussion and authentic problems. The label rates research on teaching this kind of thinking regularly, not a single five-minute move. Source: Abrami et al. 2015, Review of Educational Research

Use this move
Move 4

Argue it

Students build an argument with a claim, evidence and the reasoning that links them, and answer the other side.

Promising Argumentation from evidence

Two worked examples and the evidence

Civics · Grade 11

Task
Claim: "The voting age should be lowered to 16." Write one argument for and one against. Each needs a claim, a piece of evidence or an example, and the reasoning that links them.
Look for
Evidence, not just opinion. For example, Austria lowered the voting age to 16 for national elections in 2007, so there is real experience to point to. Strong answers explain why the evidence supports the claim and answer the strongest argument on the other side.
Probe
What evidence would make you change sides?
Think back
My argument got stronger when …

Chemistry · Grade 10

Task
Claim: "Recycling an aluminium can saves more than recycling a glass bottle." Decide whether you agree and argue it with a claim, evidence and reasoning. Say what "saves" means in your argument.
Look for
A definition of "saves" (energy, raw material or emissions). Key evidence: making aluminium from recycled metal uses about 95% less energy than making it from ore, because the electrolysis step is skipped; recycled glass saves far less, since it has to be melted either way. Strong answers say what the claim leaves out, such as weight and transport.
Probe
Which part of your reasoning depends on a number you would want to check?
Think back
Defining … first changed my answer.

Evidence. School interventions that teach scientific reasoning and argumentation show a medium effect on average across 30 interventions (Engelmann, Neuhaus & Fischer 2016). The label rates research on teaching this kind of thinking regularly, not a single five-minute move. Source: Engelmann, Neuhaus & Fischer 2016, Educational Research and Evaluation

Use this move
Move 5

Like what?

Students set a new idea beside a familiar one, map what matches, and find where the comparison breaks.

Strong Analogical thinking

Two worked examples and the evidence

Physics · Grade 9

Task
An electric circuit is often compared to water flowing in pipes. Fill in: battery = ?, wire = ?, current = ?, resistance = ?. Then find at least two places where the comparison breaks.
Look for
Battery as a pump, current as the flow, resistance as a narrow section of pipe. Where it breaks: cut a pipe and water pours out, but cut a wire and the current simply stops; electrons in a wire drift slowly even though a lamp lights almost instantly. Strong answers say what the comparison is good for and when to stop using it.
Probe
Can you find a better comparison for the part where water pipes fail?
Think back
The comparison helped me until …

Religion and ethics · Grade 10

Task
Is a promise like a contract? List what they have in common, then where the comparison breaks down. Give one example of each.
Look for
Both create obligations and expectations. Where it breaks: a contract can be enforced by a court and a promise usually cannot; a contract needs both sides to agree, while a promise can be one-sided; breaking a promise is judged morally rather than legally. Strong answers ask what the comparison hides.
Probe
Which kind of promise comes closest to being a contract? Why?
Think back
The difference I did not notice at first was …

Evidence. Learning by comparing cases improves understanding across 57 experiments (Alfieri, Nokes-Malach & Schunn 2013), and comparing solution methods built flexibility in Grade 7–8 algebra (Rittle-Johnson & Star 2007). The label rates research on teaching this kind of thinking regularly, not a single five-minute move. Source: Alfieri, Nokes-Malach & Schunn 2013, Educational Psychologist, Rittle-Johnson & Star 2007, Journal of Educational Psychology

Use this move
Move 6

Best explanation

Students meet a puzzling observation, propose several explanations, and say what evidence would tell them apart.

Promising Abductive thinking

Two worked examples and the evidence

Biology · Grade 9

Task
The bean plants on the windowsill all lean toward the glass. The ones on the table in the middle of the room grow straight up. Write three possible explanations. For each, name one thing you could observe that would support it.
Look for
Light from one side (phototropism), plus at least one real alternative such as warmth or a draught, each with a test that separates it from the others.
Probe
Which explanation would still stand if we turned the pots every day?
Think back
The explanation I gave up first was … because …

History · Grade 11

Task
Archaeologists find an English village that was lived in for 300 years and empty by 1500. Give three explanations, and the evidence that would make each one more likely.
Look for
Plague (the population fell sharply after 1349, though villages rarely emptied from plague alone; look for burials and tax records), land turned into sheep pasture (gradual, mostly in the 1400s; enclosure records and new field boundaries), and failing soil or climate. Strong answers match each cause with a kind of evidence.
Probe
Which single piece of evidence could rule out two explanations at once?
Think back
I changed my mind when …

Evidence. Generating and testing hypotheses is part of scientific reasoning, which school interventions improve on average (Engelmann, Neuhaus & Fischer 2016). Studies of abductive reasoning on its own are few and small. The label rates research on teaching this kind of thinking regularly, not a single five-minute move. Source: Engelmann, Neuhaus & Fischer 2016, Educational Research and Evaluation

Use this move
Move 7

Strip it down

Students set aside the rule they were given and rebuild it from what they know for certain.

Framework First-principles thinking

Two worked examples and the evidence

Maths · Grade 10

Task
Without using the formula, show why the area of any triangle is half of base times height. Start only from what you know for certain: the area of a rectangle.
Look for
A triangle drawn inside a rectangle with the same base and height, split along its height into two right-angled triangles, each half of a smaller rectangle. Strong answers check the case where the height falls outside the triangle.
Probe
Does your argument still work if the triangle has an angle greater than 90°? What would you have to change?
Think back
The step I could not justify at first was …

Art · Grade 8

Task
Forget what you have been taught about drawing in 3D. On a flat page, what are the basic things that make one object look closer than another? List them, then test each one with a quick sketch.
Look for
Overlap, size, height on the page, detail and sharpness, colour and contrast fading with distance, lines that converge, shading and cast shadows. Strong answers test each cue on its own.
Probe
Which cue is strong enough to work on its own, without any of the others?
Think back
I found the basics by asking …

Evidence. There are no intervention studies of first-principles thinking itself. The closest research is productive failure, where students try to build a method before being taught it. Across 53 studies it helped conceptual understanding more than teaching first, but not procedural skill (Sinha & Kapur 2021). The label rates research on teaching this kind of thinking regularly, not a single five-minute move. Source: Sinha & Kapur 2021, Review of Educational Research

Use this move
Move 8

Estimate it

Students reach a rough answer to a big question by breaking it into steps they can guess, and saying every assumption out loud.

Framework Estimation (Fermi problems)

Two worked examples and the evidence

Maths · Grade 9

Task
Roughly how many times will your heart beat in a lifetime? Work in pairs, write down every assumption and round freely. Aim for the right power of ten, not an exact answer.
Look for
A chain such as 70 beats a minute × 60 × 24 × 365 × 80 years, which gives about 3 billion. Strong answers state their assumptions, round sensibly and check whether the result is plausible.
Probe
Which assumption changes the answer most if you get it wrong?
Think back
Writing my assumptions down helped me notice …

PE and health · Grade 10

Task
Estimate how many kilometres a professional football midfielder runs in a whole league season. Show each step and the assumption behind it.
Look for
A distance per match (top-level outfield players often cover around 10–12 km), the number of league matches (38 in a 20-team league), and a decision about cup games and training. Strong answers say which number they are least sure of.
Probe
Would training add more distance than matches? How could you find out?
Think back
My estimate changed most when I …

Evidence. Estimation carries real weight in maths and science curricula, but research on Fermi problems mostly describes how students solve them (Albarracín & Ärlebäck 2019) rather than testing whether they improve students' estimation. The label rates research on teaching this kind of thinking regularly, not a single five-minute move. Source: Albarracín & Ärlebäck 2019, For the Learning of Mathematics

Use this move
Move 9

Model it

Students turn a real situation into a simple mathematical model, use it, and say where it stops matching reality.

Promising Mathematical modelling

Two worked examples and the evidence

Maths · Grade 11

Task
A rumour starts with one student. Every hour, everyone who knows it tells one new person. How many know after 1, 2, 3 … hours? Then think about a school of 600 students and change your model to fit.
Look for
Doubling at first (1, 2, 4, 8 … passing 600 after about 10 hours), then the realisation that the spread must slow down as fewer people are left to tell, giving an S-shaped curve. Strong answers state the model's assumptions and where it stops being realistic.
Probe
Which assumption in your model is least like real life?
Think back
My first model broke when …

Economics · Grade 12

Task
The school concert has 300 seats. At €5 a ticket, 300 people come; for every €1 more, 30 fewer people come. Build a model of the ticket income and find the best price. Then name one thing the model ignores.
Look for
Income = price × audience. The best whole-euro price is €7 or €8 (both give €1,680); €7.50 gives €1,687.50. Strong answers question the assumption that the audience falls by the same amount for every euro, and ask what the concert is for: money or a full hall.
Probe
If the goal were a full hall instead of the most money, what would your model recommend?
Think back
The assumption I accepted without checking was …

Evidence. Teaching mathematical modelling improved problem solving and achievement in a meta-analysis of 20 studies (Hidayat et al. 2025). The field is small. The label rates research on teaching this kind of thinking regularly, not a single five-minute move. Source: Hidayat et al. 2025, STEM Education

Use this move
Move 10

Then what?

Students follow a decision through its first consequence, the consequence of that, and one more.

Framework Second-order and causal thinking

Two worked examples and the evidence

History · Grade 10

Task
Hanoi, 1902. To fight rats that spread plague, the authorities pay a reward for every rat tail handed in. Write the first consequence, then the consequence of that, then one more.
Look for
More than "people kill more rats": people cut off tails and let the rats go free to breed, and some started breeding rats for the reward, as the historian Michael Vann describes. Strong answers trace how the reward changed what people wanted.
Probe
How would you design the reward so the chain does not go wrong?
Think back
The consequence I missed at first was …

Geography · Grade 12

Task
In 1989 Mexico City banned each car from driving on one weekday, based on its number plate, to cut air pollution. Write the chain of consequences: first, second, third.
Look for
First, fewer cars on the road. Then some households bought a second, often older car to drive on their banned day. The economist Lucas Davis found no evidence that the ban improved air quality. Strong answers ask how people adapt to a rule.
Probe
What would you have to measure to know whether the ban worked?
Think back
I started thinking further ahead when …

Evidence. Teaching students to see patterns of cause and effect has been developed through design research (Grotzer, Harvard Project Zero), but there are no effect sizes for it yet. The label rates research on teaching this kind of thinking regularly, not a single five-minute move. Source: Grotzer, Understandings of Consequence, Project Zero

Use this move
Move 11

Map the loop

Students draw how the parts of a system affect each other, and find the loops that make it grow or settle.

Promising Systems thinking

Two worked examples and the evidence

Biology · Grade 10

Task
Draw the connections between sea otters, sea urchins and kelp forests on the Pacific coast. Then show what happens to all three if the otters are hunted out.
Look for
Otters eat urchins and urchins eat kelp: fewer otters means more urchins and less kelp. Estes and Palmisano showed this in 1974 by comparing islands with and without otters. Strong answers add a link back, such as the kelp sheltering fish and other prey.
Probe
If the otters returned, what would happen first, and what would take years?
Think back
Drawing it showed me …

Social studies · Grade 9

Task
A new café opens near the school and becomes popular. Draw a loop with three parts: how popular it is, how long the queue gets, and how many customers give up and go elsewhere. Does the loop keep growing or balance itself?
Look for
More popular, longer queue, more people give up, fewer customers, shorter queue: a loop that balances itself. Strong answers add a loop that grows (crowds attract more customers) and say which one wins.
Probe
What could the café owner change to break the balancing loop?
Think back
The arrow I nearly drew the wrong way was …

Evidence. About 50 Grade 8 students taught with an earth-systems approach progressed through levels of systems thinking, and a third reached the highest level (Ben-Zvi Assaraf & Orion 2005). There is no meta-analysis yet, and few secondary-school trials. The label rates research on teaching this kind of thinking regularly, not a single five-minute move. Source: Ben-Zvi Assaraf & Orion 2005, Journal of Research in Science Teaching

Use this move
Move 12

Think back

Students describe how they thought, not what they found, and decide what to do differently. It closes every move and works on its own.

Strong Metacognitive thinking

Two worked examples and the evidence

Modern languages · Grade 8

Task
Right after the vocabulary quiz, before you see your score, write: How did you study? Which words do you think you got wrong, and why those? Then compare with your result.
Look for
Specific descriptions of how they studied (reading the list or testing themselves), predictions about which words they missed, and an honest comparison afterwards. Strong answers name one change for next time.
Probe
Was your prediction better or worse than you expected? What does that tell you?
Think back
Next time I will study by … instead of …

Maths · Grade 12

Task
After solving a hard problem, write three lines: where I got stuck, what I tried, and what finally moved me forward. Then swap with a partner and compare strategies.
Look for
Named strategies, such as drawing a diagram, trying a simpler case or working backwards, rather than "I just thought harder". Strong answers notice a strategy they could use on the next problem.
Probe
Which of your partner's strategies would have saved you time?
Think back
When I get stuck next time, I will first try …

Evidence. Teaching students to plan, monitor and evaluate their own learning is one of the strongest strands in the EEF Teaching and Learning Toolkit: +8 months of progress on average, with 4 of 5 padlocks for the strength of the evidence. The label rates research on teaching this kind of thinking regularly, not a single five-minute move. Source: EEF Teaching and Learning Toolkit: Metacognition and self-regulation

Use this move

Five minutes in the lesson

  1. Task (2 min). Students work alone, on paper or a whiteboard.
  2. Pair talk (1 min). They compare with a partner and improve one answer.
  3. Share and probe (1 min). Take two or three answers and ask the Probe question.
  4. Think back (1 min). Each student finishes the sentence starter about how they thought.

Check the AI's answer before the lesson, especially anything listed under CHECK. The AI prepares; you decide.

Six kinds of thinking

Popular lists name thirteen or more modes of thinking. Several are the same thing in research terms, so the moves group them into six clusters.

1 Generating ideas

Producing many and unusual options.

  • Divergent
  • Associative
  • Lateral
  • Janusian No move of its own: the evidence is thin and comes almost entirely from one researcher.

Moves: Many ways

2 Evaluating and choosing

Weighing evidence, testing claims and deciding.

  • Critical
  • Convergent
  • Argumentation from evidence added

Moves: Pick and justify, Check the claim, Argue it

3 Comparing and transferring

Seeing the structure one case shares with another.

  • Analogical

Moves: Like what?

4 Explaining and reasoning from basics

Finding the best explanation and rebuilding from what is certainly true.

  • Abductive
  • First principles
  • Mathematical modelling added
  • Estimation added

Moves: Best explanation, Strip it down, Estimate it, Model it

5 Causes and systems

Chains, loops and side effects.

  • Systems
  • Second-order
  • Inversion No move of its own: the evidence is thin and comes from adult lab studies of the pre-mortem.

Moves: Then what?, Map the loop

6 Regulating your own thinking

Watching and adjusting how you think. It makes every other cluster sharper.

  • Metacognitive

Moves: Think back

Thinking taught inside subject content does better than stand-alone thinking programmes. In large UK trials, Let's Think Secondary Science and Philosophy for Children showed no effect on attainment. That is why every move is built on your lesson topic.

How strong is the evidence?

  • Strong Meta-analyses of classroom interventions show that teaching this kind of thinking improves it.
  • Promising Controlled studies show gains, but results are mixed, small in number or dependent on context.
  • Framework A useful structure for thinking, with no intervention studies behind it yet.

Labels rate the research on teaching this kind of thinking regularly, not a single five-minute move. One move is one small piece of that practice.

Sources

Each source is also in its move's card. For research on many other teaching strategies, see the Evidence Toolkit on this site.

AI tools and age rules

You use AI to prepare the task. Your students do not need to. If older students use AI themselves, these rules apply (October 2026):

Rules differ between countries and change often. Check your own national rules and your school policy before students use any AI tool.

The card deck

All twelve moves with their worked examples, one per A4 page, plus a fill-in prompt for when you are offline. Free to print and share.

Download the card deck (PDF)