Critical Thinking Skills: 9 Skills With Exercises

What are critical thinking skills?

Critical thinking skills are the habits that let you decide what to believe and what to do based on reasons rather than impulse, authority or repetition. They boil down to a few questions asked well: What exactly is being claimed? How do we know? What else could explain this? What would change my mind?

The skills below are grouped roughly in the order you’d use them on a claim: first understand it, then test it, then test yourself. Each comes with a short exercise. Do one a day for nine days, or pick the one that matches your weakest spot.

1. Clarifying the claim

Clarifying means pinning down exactly what is being said before you agree or disagree. Most arguments that go nowhere are two people arguing about different claims.

Vague words hide disagreements: “better”, “safe”, “most”, “significant”, “natural”. Ask what they mean in this sentence. Ask what would count as the claim being false. A claim that nothing could contradict isn’t saying much.

Exercise: Take three headlines from today’s news. Rewrite each as a precise, testable sentence. “Coffee is good for you” becomes something like “People who drink X cups a day have a lower rate of Y than people who drink none.” Notice how many details the headline left out.

2. Analyzing arguments

Analyzing an argument means separating its conclusion from its premises (the reasons given) and finding the assumptions that connect them.

Most real arguments are compressed. “We should hire her; she worked at a top company” has a hidden premise: people from top companies perform well here. Once it’s written out, you can judge it. Unstated premises are where weak arguments hide.

Exercise: Find an opinion piece or a persuasive email. Write the conclusion in one sentence. List the premises as bullet points. Then add the unstated assumption that would be needed for the premises to lead to the conclusion. Ask whether you believe that assumption.

3. Evaluating evidence

Evaluating evidence means asking how strong the support for a claim actually is: where it came from, how it was gathered and whether it could have come out differently.

Useful questions:

• Source: Who says so, and how would they know?
• Method: An experiment, a survey, a single anecdote, an expert’s opinion?
• Size: One case or many?
• Comparison: Compared with what? A number with no baseline means little.
• Incentive: Does the source gain if you believe it?

Exercise: Pick a statistic you’ve seen quoted this week. Trace it back to its original source. Note how many steps it took, and whether the claim changed along the way. It often does.

4. Inference

Inference is drawing conclusions that actually follow from the evidence, no more and no less.

The common errors are going too far (“sales rose after the ad campaign, so the campaign worked”) and not going far enough (refusing to conclude anything when the evidence is clear). Distinguish what is shown, what is suggested and what is merely possible.

Exercise: Take a short news report about a study. Write three sentences: what the study showed, what it suggests, and what it does not show. If you can’t write the third sentence, read the report again; the limits are usually there.

5. Spotting logical fallacies

A fallacy is a flaw in reasoning that makes an argument seem stronger than it is. Knowing the common ones lets you name the problem instead of just feeling that something is off.

The most frequent in everyday life are the straw man, ad hominem, false dilemma, appeal to popularity, slippery slope and post hoc reasoning. Our complete list of logical fallacies covers them with examples.

Exercise: Watch ten minutes of a debate, a talk show or a comment thread. Each time you spot a fallacy, write down its name and the sentence. Then do the harder part: find one in an argument from the side you agree with.

6. Spotting cognitive biases

Cognitive biases are systematic patterns in how human judgment departs from careful reasoning. Unlike fallacies, which are errors in arguments, biases are errors in the thinker, and they work quietly.

The most important one for critical thinking is confirmation bias: we search for, notice and remember evidence that supports what we already believe. Others to watch include the availability heuristic (judging how common something is by how easily examples come to mind) and the overconfidence bias. See more cognitive biases with examples.

Exercise: Pick a belief you hold strongly. Spend 15 minutes looking only for the best evidence against it. Write down the strongest point you found. You don’t have to change your mind; you have to be able to state it.

7. Considering alternatives

Considering alternatives means generating other explanations, options or interpretations before you settle on one.

The first explanation that comes to mind is usually the most available, not the most likely. Steelmanning is the discipline of building the strongest version of the opposing view, not the weakest. Steelmanning protects you from winning arguments against positions nobody holds.

Exercise: Next time something surprises you (a colleague’s odd decision, a drop in sales, a friend’s silence), list three different explanations before you act. Rank them by likelihood. Note which one you would have assumed without the exercise.

8. Probabilistic reasoning

Probabilistic reasoning means thinking in degrees of confidence rather than yes or no, and anchoring those degrees in how often things usually happen.

Two habits do most of the work. First, start from base rates: before judging a specific case, ask how often this kind of thing happens in general. Second, put a number on your confidence. “Probably” can mean 55% or 90%, and the difference matters. More on this in probabilistic thinking.

Exercise: Write down five predictions about the coming month (a project deadline, a team result, the weather on a given day) with a confidence percentage for each. At the end of the month, check them. If your 90% predictions come true only half the time, you’ve learned something about your calibration.

9. Updating your beliefs

Updating means changing your mind in proportion to new evidence: a little for weak evidence, a lot for strong evidence, and not at all for none.

This is the skill that makes the other eight matter. Many people evaluate evidence well when it threatens other people’s views and poorly when it threatens their own. Cognitive dissonance, the discomfort of holding conflicting beliefs, pushes us to explain away the evidence instead.

Exercise: Before your next important decision or debate, write one line: “I would change my mind if I learned ___.” If you can’t fill in the blank, you are not holding a belief; you’re holding a position.

Critical thinking skills at a glance

How to practice critical thinking every day

Critical thinking improves with regular, small practice on real material. Three habits make it automatic:

1. The pause. Before sharing, agreeing or deciding, ask “how do we know?” once.
2. The written prediction. For decisions that matter, write what you expect and how confident you are. A decision journal makes this a routine.
3. The opposite search. Once a week, look for the best case against something you believe.

Critical thinking also connects directly to decisions. The decision-making hub shows how to apply it to choices, and the mental models library gives you tools to reason with.

The Mentarium take

Most critical thinking advice treats it as a weapon for evaluating other people’s claims. That’s the easy half. The hard half, and the half that pays, is skills 6 and 9: spotting your own biases and actually updating.

Here is the test we use: can you name one belief you changed in the last year because of evidence? If not, your critical thinking is probably running in one direction only.

That’s why the Stress-test step of the Mentarium Decision Loop (Frame → Map → Stress-test → Decide → Review) runs the Bias Radar against your favorite option, not the ones you already dislike. Point the skills inward first. The arguments you’re most likely to accept without checking are the ones you wanted to hear.