Can You Learn to Think Critically? | Lesson 2.1, Free Course

What it means, and how much teaching helps.

The idea

If a friend says a new study app helps you learn, the careful question is: 'What is that based on?' That question sits at the heart of this lesson. Critical thinking is careful, steady thought about what a belief rests on. It also asks where the belief would lead if it were true. One short definition calls it reasonable, reflective thinking about what to believe or do. People who study it don't fully agree where it stops. For some it ends once you've reached a judgement; others include what you then do about it.
Can it be taught? Yes, a little. Studies of classes that set out to teach it find a real but modest gain, on average, not a transformation. Results vary a lot from one study to the next. Classes that taught it both on its own and inside other subjects did better than classes doing only one. But that difference may have been down to chance.
What helps? Teaching it goes better when the goal is spelled out than when it's left to happen by itself. Three things stood out: talking it through, real problems and examples, and a mentor. Picture a class arguing over a real news story, with a teacher who keeps asking 'how do you know?' Methods that worked turned up at every level of schooling and in every subject.
Here's the catch: the skill may not move easily from one subject to another. Thinking well about something seems to depend heavily on knowing it. A keen gardener can judge a claim about soil, and still be fooled by one about cars. It can carry over, but when and why it does is complicated. One popular idea is to teach it across many subjects, pointing out the skills they share. One philosophy encyclopedia calls the evidence for that 'scanty': there isn't much of it.

Want the research, names and dates? They are in Go deeper, at the end.

A sharp mechanic, and a sailing boat.

Picture it

Think of a good mechanic. Told 'the car makes a noise', they don't just swap a part. They ask what could cause it, and what each cause would mean. Put the same mechanic in front of a sailing boat, though, and they're much closer to a beginner. Their skill is real. It just runs on what they know. Their habit of asking 'what could cause this?' comes along. The answers don't.

For a child: A girl hears there's a ghost in the old shed. Before running, she asks, 'Who saw it?' Nobody did: her cousin heard it from a friend, who heard it from someone else. The story was big, but what it rested on was tiny.

Four moves between hearing a claim and believing it.

How it applies

Careful thinking comes down to four moves. Two are about the belief: what a belief rests on, and where it leads. One is about you: how well you know the subject. And one is a decision about what to believe or do, which can include waiting. How well you know a subject often shapes how well you can judge it. Say a headline claims crime in a city has doubled. What it rests on is who counted, and how. Where it leads is what else you'd expect to see if it were true.

Four moves · Drawing: Hacks Vitae

At home, at work and online.

Where you’ll see it

At home: a relative forwards a "life-changing" cleaning trick. It has been shared again and again, and nobody has asked where it came from.
At work: a sales figure in a meeting looks great, until someone asks what it was compared with. Last year? A rival? A target set in a hurry?
Online: you spot the holes in a post about your own job at once. A confident post about medicine slips straight past, because you have nothing to check it against.

Signs it’s happening

You can say what you believe, but not what it rests on.
A claim feels obviously true, and you can't quite say why.
You might feel just as sure in a field you barely know as in one you know well.

Same news, two different first questions.

Before and after

A friend says a new app doubles your phone's battery life.

The quick answer

"Brilliant, I'll share it." The claim gets passed on just as it arrived, with nothing added but speed.

The second look

"Where does that come from, and what would follow if it were true?" That question doesn't call the friend wrong. It just checks the ground first. Asking what a claim rests on comes before passing it on.

He printed the story, then asked if it was true.

A story from history: The professor who doubted his own story

The iron maiden is said to be an iron figure that killed the condemned. That story begins with a note printed in 1793, not with any medieval record.
A law professor, Johann Christian Siebenkees, printed it, headed only "from a chronicle". He did not name the chronicle.
The note said that in 1533 an iron figure was set up in Nuremberg. When a step was touched, it hacked the condemned to pieces.
In the same breath, he doubted it. He had never read of it being used, he wrote, and asked: "Might the whole thing perhaps be a legend?"
In the 19th century, spikes were added to old cloaks of shame. One museum, in Rothenburg, dates the spikes on its iron maiden to 1813–1815. It says the object served as a cloak of shame for women, who were released physically unharmed. With spikes on, the legend now had an object to point to.

Siebenkees looked at what his note rested on, and openly doubted it. An object made later can seem to prove a story that was doubted from the start.

Quick check3 questions · nothing is saved

1. Which short description fits critical thinking?

2. When classes set out to teach it, what happened on average?

3. In the iron maiden story, what did Siebenkees do?

One question worth keeping.

In one line

Critical thinking asks what a belief rests on, and it seems to work best on things you know well.

Which of your firm beliefs have you never asked 'what does this rest on?'

Go deeperThe research, names and dates

Where this comes from

The idea as a goal for schools goes back to John Dewey (1910), an American philosopher. He more often called it reflective thinking. For him it meant careful, persistent thought about a belief, weighing its grounds and what it leads to. Robert Ennis, also a philosopher, has a shorter version: "reasonable and reflective thinking focused on deciding what to believe or do". Between 1988 and 1989 a panel of 46 scholars, educators and others agreed a longer definition, published in 1990. It describes purposeful, self-regulating judgment that interprets, analyses, evaluates and infers, and explains its reasons. Not every definition agrees. Some end at a judgement; others count actions as well as beliefs.
Philip Abrami and colleagues pooled classroom studies twice. Their 2008 review covered 117 studies and 20,698 participants. It found an average effect of 0.341 standard deviations, and the results varied widely. Their 2015 review, built on 341 effect sizes, came out much the same, at 0.30. On average that is a real but modest gain, not a transformation.
The first review's authors said improvement can't simply be expected. Teachers have to make the goals explicit. The second review found methods that worked at every level of schooling and in every subject. Dialogue, real-world problems and examples, and mentoring stood out.
Daniel Willingham is a scientist who studies the mind. In 2007 he argued that critical thinking isn't a kit to carry anywhere. It leans heavily on knowing the subject. He still wrote that "such transfer does occur", though when and why is complex. One popular answer is teaching it across many subjects, with attention to the skills they share. The Stanford Encyclopedia of Philosophy calls the evidence for it "scanty".

The skill you trained vs skills you didn't

Compared	The trained skill	Skills not trained
What it means	Getting better at the skill you trained	Gains spreading to skills you didn't train
What it's called	Near transfer	Far transfer
Children trained in one thinking skill	That skill improved	No convincing evidence the gains spread

A review of chess, music and working-memory training found that the better the study design, the smaller the effect. Its authors "conclude that far transfer of learning rarely occurs". Action games and music training are the main exceptions claimed.

More examples

If you hear 'This is the top school in town', the question underneath is 'Top by what measure?'
If you hear 'Everyone is switching to this app', the question underneath is 'Who is everyone, and how do we know?'
If you hear 'This old saying has always held true', the question underneath is 'How would we know if it hadn't?'

What if…?

What if AI does the thinking for me?: One field experiment involved nearly 1,000 high-school students in Turkey. With GPT-4, a generative AI tool, helping on maths exercises, their grades went up. Some had a plain version of the tool. Others had a tutor version, set up to guide them step by step rather than hand over answers. Once the tool was removed, students who'd had the plain version scored 17% lower than students who never had it. The tutor version largely avoided that drop. The authors' explanation is that the plain tool was used "as a 'crutch'". The study measured maths learning across a few sessions, not critical thinking as a whole.
Doesn't university teach it anyway?: Huber and Kuncel's 2016 review found that thinking skills and habits grow substantially over an ordinary college experience. Whole-curriculum programmes aimed at boosting them didn't necessarily add long-term gains on top.
Is it the skill employers want most?: The World Economic Forum's October 2020 report put "critical thinking and analysis" first among the skill groups employers saw rising in importance. Among that report's top 15 individual skills for 2025, it came fourth. Its 2023 and 2025 reports put analytical thinking at the top.

Words you may meet

far transfer: When practising one skill also improves other, more distant skills or tasks that were not practised. Glossary →

generative AI: AI that learns the patterns in examples it was given and uses them to make new content, such as pictures, video, sound or text. Glossary →