As a leader, how can I improve my critical thinking and challenge my own assumptions?

Critical thinking is the disciplined process of examining what you believe, what evidence supports it, what assumptions it depends on, and what might cause you to revise it. That sounds straightforward until the thinking being examined is your own. Most reasonably capable people can find the weakness in a proposal they already disagree with, identify the questionable assumption behind a colleague’s recommendation, or explain why somebody else’s evidence isn’t quite as conclusive as they think it is. The harder test comes when the proposal is yours, the evidence supports what you hoped would be true, and the conclusion already feels right.

At that point, reasoning can quietly change jobs. Instead of helping you discover what is true, it starts helping you explain why what you already believe must be true. The distinction is difficult to notice from inside your own thinking because both processes feel like reasoning. You are still gathering evidence, making connections and constructing arguments. The question is whether those activities are genuinely testing your conclusion or simply making an increasingly sophisticated case for it. Critical thinking isn’t thinking harder. It is making your thinking easier to test.

What this actually buys you as a leader of yourself

A leader who doesn’t challenge their own assumptions can make perfectly rational-looking decisions from a faulty starting point. They interpret a stakeholder’s silence as resistance and start managing opposition that wasn’t there, assume a struggling employee lacks capability when the real problem is unclear priorities, or become so convinced that a customer objects to price that they never discover the real concern is implementation risk. None of those decisions necessarily looks irrational from the outside. Once the original assumption is accepted, everything that follows can be internally coherent.

A leader who can interrogate their own thinking has a different relationship with judgement. They can form a strong view without becoming fused with it, hear disagreement without immediately treating it as something to overcome, and distinguish the confidence required to make a decision from certainty that the decision must be right. An assumption becomes something that can be named and tested rather than quietly embedded inside a plan. An alternative explanation becomes something worth exploring rather than an inconvenience to the preferred one. New evidence can genuinely change a judgement instead of being absorbed into whatever was already believed. The benefit isn’t that you stop making mistakes. It’s that mistakes have fewer places to hide.

Why it matters

Assumptions themselves aren’t the problem. You couldn’t function without them. Every judgement involves filling gaps between what is directly known and what experience, expectation and context suggest is probably true. The difficulty begins when that distinction disappears and an assumption starts receiving the same status as the evidence it was built from. Once an explanation becomes plausible enough to feel like “what is happening”, the way you subsequently gather and interpret information can start reinforcing it.

Confirmation bias is one of the most established findings in the psychology of reasoning, although the term describes several related processes rather than one single error. Nickerson’s extensive review described the tendency to seek or interpret evidence in ways that favour existing beliefs, expectations or a hypothesis already under consideration. Once an explanation has become my explanation, the danger isn’t simply that I will deliberately ignore inconvenient evidence. I may ask questions that are more likely to confirm it, notice supportive information more readily, interpret ambiguous evidence in its favour, and set a higher standard of proof for evidence that challenges it than for evidence that supports it (Nickerson, 1998).

A closely related problem is what Stanovich, West and Toplak call myside bias: evaluating evidence, generating evidence and testing hypotheses in ways that favour our existing attitudes and opinions. The uncomfortable finding is that susceptibility to this kind of thinking shows surprisingly little relationship with intelligence. Being cognitively capable doesn’t automatically protect you from reasoning in favour of your own position, which means the solution cannot simply be to think harder or become cleverer. A highly capable thinker may also be highly capable of constructing an impressive defence of a conclusion they haven’t adequately tested (Stanovich, West and Toplak, 2013).

That distinction between cognitive ability and the willingness to use it differently is central to research on actively open-minded thinking. Work by Baron, Stanovich and Toplak describes thinking standards that include considering alternative opinions, being sensitive to evidence that contradicts your current belief, being willing to postpone closure and engaging in reflective thought. This doesn’t mean refusing to have opinions, assuming your first judgement must be wrong, or treating every possible explanation as equally credible. It means giving competing possibilities enough genuine consideration that your preferred conclusion can lose (Baron, 2019; Stanovich and Toplak, 2019; Stanovich and Toplak, 2023).

Actively open-minded thinking isn’t a guarantee against myside bias, however. Knowing that open-mindedness matters, or even seeing yourself as open-minded, doesn’t ensure you’ll recognise when your preferred conclusion shapes what counts as convincing evidence. The disposition matters, but it still has to translate into what you actually do in a live judgement (Stanovich and Toplak, 2023).

Halpern’s work on critical thinking adds another important part of the picture. Critical thinking involves more than knowing a collection of techniques; it also requires the disposition to use them and the metacognitive ability to recognise when more deliberate scrutiny is needed. A person can understand how assumptions work, know what confirmation bias is and be perfectly capable of constructing an alternative explanation, yet fail to do any of those things at the moment their own judgement most needs testing. The skill has to travel from the workshop, book or classroom into the messy situation in which your interpretation already feels obvious (Halpern, 1998).

The answer, then, isn’t to distrust your thinking or try to eliminate assumptions. It is to create conditions in which assumptions remain visible, evidence can still surprise you, and your preferred explanation has to compete.

How an assumption quietly becomes a conclusion

The problem isn’t that you made an assumption. You couldn’t function without assumptions. The problem is losing sight of where the evidence ended and the assumption began. None of the stages above requires irrationality either. Your interpretation may be entirely reasonable, your assumptions may come from years of relevant experience, and the evidence you collect may genuinely support your conclusion. Critical thinking isn’t the insistence that your first explanation must therefore be wrong; it is the discipline of keeping the distinction between plausible and established visible for long enough that another explanation has a genuine opportunity to compete.

This is also where critical thinking connects to cognitive distortions without becoming the same thing. In my article on catching cognitive distortions, the focus is on noticing when an interpretation such as catastrophising, mind reading or all-or-nothing thinking has quietly become confused with observation. Here the question comes one step later. You have an interpretation that may be perfectly reasonable. How do you test it well enough to know how much confidence it deserves?

Five practices for improving your critical thinking

Exercise 1: The fact-assumption split

Most conclusions contain more assumption than they initially appear to because the mind is very good at filling gaps without labelling the material it has added. “The project is slipping because the sponsor has lost interest” may feel like one observation, when the project slipping is observable and almost everything after “because” is an interpretation built from incomplete information. The purpose of this exercise isn’t to eliminate assumptions, since decisions would be impossible without them. It is to stop assumptions receiving the same status as facts simply because they sit inside the same sentence.

Steps:

1. Choose one judgement you’re currently reasonably confident about and write it down in a single sentence.
2. List the pieces of information you can genuinely observe or verify without needing to interpret what they mean.
3. Separately list everything your conclusion requires you to assume, including assumptions that feel obvious because experience tells you they’re probably true.
4. Ask which assumption, if it turned out to be wrong, would most change your conclusion.
5. Decide whether that assumption can be checked directly, tested indirectly, or simply needs to remain visible as an uncertainty in the judgement.

Variations: Do the split with somebody who isn’t involved in the situation. Ask them to stop you every time something in your “facts” column actually contains an interpretation.

Why it matters: Critical thinking becomes much easier once the invisible parts of an argument are visible. You don’t need to abandon an assumption simply because it is an assumption; you need to know how much of your conclusion is resting on it.

Exercise 2: The rival explanation test

Generating alternatives sounds straightforward until you notice how easily the preferred explanation gets a strong case and the alternatives get weak ones. If you believe a colleague is resisting your proposal because they’re protecting their territory, “perhaps they just don’t understand it” isn’t much of a rival if you don’t genuinely believe that explanation fits the evidence. A useful alternative has to be strong enough that, if it were true, you would substantially revise what you currently think is happening. If your rival explanation couldn’t change your decision even if it were true, it isn’t much of a rival.

Steps:

1. Write your preferred explanation for the situation as clearly and specifically as possible.
2. Imagine an intelligent person with access to the same information disagrees with you. Write the explanation they could reasonably offer instead.
3. Strengthen that explanation until it genuinely fits the available evidence rather than leaving it deliberately easy to dismiss.
4. For both explanations, write down what you would expect to observe next if that explanation were true.
5. Look for the next piece of information that would best discriminate between them, rather than simply asking which explanation currently feels more convincing.

Variations: Where the stakes justify it, generate three explanations rather than two. Stop once the alternatives become merely possible rather than genuinely plausible; the purpose isn’t to create endless uncertainty.

Why it matters: A conclusion that has never competed with a credible alternative hasn’t been tested very hard. The rival explanation gives your preferred story something capable of beating it.

Exercise 3: The search-against-yourself test

Confirmation bias doesn’t require deliberately hiding evidence from yourself. It can start much earlier, in the questions you choose to ask. If you believe a new approach is working, you naturally look for signs of progress; if you believe someone isn’t performing, you notice examples of poor performance. The search itself starts producing an increasingly one-sided evidence base, which then appears to justify the conclusion that shaped the search in the first place. This exercise changes the direction of search before adding more evidence to the pile.

Steps:

1. Write down the conclusion you currently favour.
2. Before collecting any more supporting evidence, ask: “If this conclusion were wrong, what might I expect to see?”
3. Identify three pieces of evidence, observations or perspectives that could genuinely weaken your current position.
4. Look for those first, particularly in places or people you’re least likely to consult when you’re trying to prove the conclusion.
5. Only after that search, review the evidence for and against together and decide whether your confidence should rise, fall or stay where it was.

Variations: Ask another person to nominate the evidence you should look for without first telling them what evidence you’ve already collected. This reduces the temptation to steer them towards your existing frame.

Why it matters: Looking for disconfirming evidence doesn’t assume your conclusion is wrong. It gives being wrong a fair chance to reveal itself before the cost of discovering it increases.

Exercise 4: The mind-change test

One of the clearest tests of critical thinking is whether you can answer a deceptively simple question: what would change your mind? If the honest answer is “nothing”, you no longer have a conclusion you’re testing. You have a position you’re defending. A subtler version happens when contradictory evidence is technically allowed to count but the standard keeps moving as it arrives: one piece is an exception, another needs more context, a third isn’t quite comparable. Each explanation may be individually reasonable, but collectively they can make the conclusion almost impossible to challenge. The mind-change test sets the conditions for updating before you know whether the next evidence will support you or not.

Steps:

1. Choose a judgement that matters enough that being wrong would change what you do.
2. Complete the sentence: “I currently believe…” and state the conclusion without qualification.
3. Then complete: “I would become less confident in this if…” and identify something you could realistically observe.
4. Where possible, decide in advance roughly how much that evidence should change your confidence, rather than deciding after it arrives.
5. Return to the statement when the evidence becomes available and update against the criterion you set beforehand, without quietly rewriting the criterion.

Variations: For a team decision, have each person privately write their mind-change condition before discussing the issue. Compare the conditions afterwards; differences often reveal assumptions that the discussion itself hadn’t surfaced.

Why it matters: Deciding what would change your mind before the evidence arrives protects against moving the goalposts afterwards. A belief that no imaginable evidence could weaken isn’t functioning as a testable judgement anymore.

Exercise 5: The confidence calibration

Leadership decisions often require commitment before certainty is available, which creates an understandable temptation to turn “this is currently the best explanation” into “this is what is happening”. Confidence calibration separates the need to act from the need to pretend the evidence is stronger than it is. Actively open-minded thinking doesn’t require low confidence for its own sake; Baron argues for appropriate confidence, including recognising when low confidence is warranted. More thinking isn’t automatically better either, which means the goal is not permanent hesitation but a better match between what the evidence supports and how strongly you hold the conclusion (Baron, 2019).

Steps:

1. Take a judgement you’re about to act on and give your confidence in it a rough percentage, for example 60%, 75% or 90%.
2. Ask what evidence is doing most of the work in producing that number, rather than simply listing everything you know.
3. Identify what would move your confidence down by ten percentage points and what would move it up by ten.
4. Ask whether the decision itself requires greater confidence than you currently have, or whether it is sufficiently reversible that you can act and learn.
5. When new evidence arrives, update the number before defending the original decision.

Variations: Keep a small record of repeated judgements and their confidence levels. The purpose isn’t mathematical precision; it is noticing whether “I’m 70% confident” repeatedly turns into “I knew it” once an outcome becomes known.

Why it matters: A confidence estimate creates room between you and the conclusion. You can act decisively on a 70% judgement while leaving 30% of the door open for the world to tell you something different.

Exercise wrap up

These five exercises work at different points in the same reasoning process: exercise 1 exposes the assumptions that have been travelling disguised as facts, exercise 2 makes your preferred explanation compete with something strong enough to challenge it, exercise 3 changes the direction of evidence search, exercise 4 decides in advance what would genuinely warrant updating, and exercise 5 keeps the strength of your confidence connected to the strength of the evidence. None requires you to become permanently doubtful or endlessly analytical. They are ways of making your thinking easier to test before commitment makes it harder to change.

When critical thinking becomes overthinking

There is an obvious trap in all of this. If every assumption needs examining, every explanation needs three rivals and every decision requires a search for contradictory evidence, critical thinking quickly becomes a sophisticated form of paralysis. Critical thinking isn’t measured by how long you think; it is measured by whether the thinking you do improves the quality of the judgement. Actively open-minded thinking doesn’t mean maximising the amount of thought devoted to every decision either. Baron explicitly argues that more thinking isn’t always better. The amount of scrutiny should be proportionate to what is at stake, how uncertain the situation is, and how difficult the decision will be to reverse (Baron, 2019).

A low-consequence, reversible decision often deserves a reasonable judgement and quick action. Choose, see what happens and adjust. A consequential decision built on several uncertain assumptions, with substantial cost attached to reversing it, deserves more deliberate challenge before commitment. In situations where uncertainty can’t be reduced through further analysis, the better test may be action itself: a small probe, experiment or reversible step that generates information you couldn’t obtain by thinking for another hour. Critical thinking should improve your ability to act intelligently, not become a reason to avoid acting until the world has removed all uncertainty for you.

Why other people can improve your critical thinking

Sometimes the most efficient way to challenge your own assumptions is to borrow somebody else’s. Other people don’t begin inside your chain of reasoning. They haven’t necessarily made the same assumptions, noticed the same evidence or become invested in the same explanation, which makes them capable of seeing weaknesses that are almost invisible from inside your own frame. Mercier and Sperber’s argumentative account of reasoning is particularly interesting here. They argue that human reasoning has an important social and argumentative function, with people not only producing reasons but evaluating reasons offered by others. Their account is a theoretical position rather than something that should be treated as settled fact, but it helps explain why reasoning can sometimes improve when genuinely different perspectives are allowed to challenge one another rather than each person trying privately to reason their way out of their own assumptions (Mercier and Sperber, 2011).

Putting more people in a room doesn’t automatically create better reasoning, though. If everyone shares the same assumptions, if disagreement carries a social cost, or if the most senior person’s interpretation becomes the starting point everyone else is expected to refine, the group may simply become better at collectively defending one explanation. Asking “Does anyone disagree?” after you’ve spent ten minutes explaining your preferred answer isn’t much of a test either. Ask people what they see before you tell them what you see, invite competing explanations before evaluating them, and make it legitimate for someone to challenge the frame rather than merely improve your solution within it. Diversity of people only improves thinking when it produces diversity of interpretation that can actually be voiced.

Five micro-practices to start today

Five things you can start doing today, no dedicated time required.

• When you hear yourself say “obviously”, pause and ask what assumption has just become too obvious to mention.
• Before accepting a piece of evidence that supports you, ask whether you’d find the same evidence convincing if it supported the opposite conclusion.
• When one explanation immediately feels right, name one other thing that could plausibly produce the same observation.
• Before saying “I know”, ask yourself whether you actually mean “this is my best current explanation”.
• When someone disagrees with you, ask “what are you seeing that I’m not?” before explaining why you think they’re wrong.

Conclusion

Improving your critical thinking isn’t primarily about acquiring more sophisticated ways to analyse other people’s arguments. It starts with becoming more disciplined about your own. Separate evidence from assumption. Make your preferred explanation compete. Search for what could weaken it. Decide what would change your mind. Match your confidence to what you actually know. Critical thinking isn’t thinking harder; it is making your thinking easier to test.

None of that requires permanent doubt. You still need to form views, trust experience, make decisions and act without complete information. The difference is that your conclusions remain revisable after you’ve made them. Good judgement isn’t demonstrated by never changing your mind. It is demonstrated by knowing what should cause you to change it, noticing when that evidence arrives, and being willing to let it count.

Reflective questions

• Think of a judgement you’re currently confident about. What part of it is directly supported by evidence, and what part only works if one of your assumptions is true?
• What would you genuinely need to see to become less confident in that judgement, and have you actually looked for it?
• Who around you is most likely to see the situation differently from you, and what makes it easy or difficult for them to tell you?

Explore further

• How do I catch the thinking patterns that are quietly skewing my decisions, for recognising when an interpretation has started to feel more like an observation than a judgement.
• Decision-making under uncertainty, for what to do when better thinking still can’t give you enough information to predict what will happen.
• How do I stop imposter syndrome from discounting my track record?, for an applied example of how existing beliefs can determine which evidence gets allowed to count.
• Edward DeBono’s Direct Attention Tools are a nice collection of ways to think about a topic.

Sources

Baron, J., 2019. Actively open-minded thinking in politics. Cognition, 188, pp.8–18.

Halpern, D.F., 1998. Teaching critical thinking for transfer across domains: Dispositions, skills, structure training, and metacognitive monitoring. American Psychologist, 53(4), pp.449–455.

Mercier, H. and Sperber, D., 2011. Why do humans reason? Arguments for an argumentative theory. Behavioral and Brain Sciences, 34(2), pp.57–74.

Nickerson, R.S., 1998. Confirmation bias: A ubiquitous phenomenon in many guises. Review of General Psychology, 2(2), pp.175–220.

Stanovich, K.E. and Toplak, M.E., 2019. The need for intellectual diversity in psychological science: Our own studies of actively open-minded thinking as a case study. Cognition, 187, pp.156–166.

Stanovich, K.E. and Toplak, M.E., 2023. Actively open-minded thinking and its measurement. Journal of Intelligence, 11(2), p.27.

Stanovich, K.E., West, R.F. and Toplak, M.E., 2013. Myside bias, rational thinking, and intelligence. Current Directions in Psychological Science, 22(4), pp.259–264.