Five checks for the moments when your first answer is most likely to be wrong, with myths, common questions and a skill check to rate yourself.
Doctor's Note: These checks are for everyday choices. If worst-case thinking, constant worry or harsh self-judgment fills your days or stops you from deciding at all, talk to a doctor or therapist. If you ever feel unsafe or think about harming yourself, find a free crisis line in your country at findahelpline.com (988 in Canada and the US), or call your local emergency number now.
Here's What's Actually Going On in Your Head
Try this before you read on. A bat and a ball cost 1.10 in total. The bat costs 1.00 more than the ball. How much is the ball? Most people answer "10 cents", because 1.10 splits so neatly into 1 and 10 cents. The right answer is 5 cents, since 0.05 plus 1.05 makes 1.10 [4].
Your mind has a quick, automatic way of thinking and a slow, deliberate one. The quick one hands you a plausible answer before the slow one has time to check it [4]. Most of the time that works well. The trouble starts with a big price, a heated argument or a plan you are attached to.
Being clever does not protect you. World Bank staff read the same data less accurately when it was about minimum wage laws than when it was about a skin cream, and ability and seniority did not explain the gap [4].
So the aim is not to memorise a list of biases. It is to use five checks at the right moments.
Step 1: Catch the Moments That Need Slow Thinking and Take 10 Seconds
WHY THIS MATTERS
You do not need to slow down for every choice. Time pressure and money stress make it harder for the slow mind to step in [4], so the skill is to spot the few moments that need a check and give them 10 seconds.
- Learn the three signals by heart: high stakes (money, health, a relationship, a job, anything hard to undo), strong feelings (angry, scared, excited, desperate to be right) and an answer that came too easily (it "just feels obvious").
- When one signal fires, stop for 10 seconds before you say yes, buy, send or commit.
- In those 10 seconds, ask one question: "Why might my first answer be wrong?" Name one reason, even a weak one.
- If someone pushes you to decide now, say: "I don't make decisions this size on the spot. I'll get back to you." A fair offer will still be there.
Within 2 weeks you catch at least one signal before you act, not after, and you notice which of the three signals you miss most.
Step 2: Test a Belief: Search the Best Evidence Against It and Ask the Mirror Question
WHY THIS MATTERS
Once you hold a view, you read new evidence in its favour. In a classic set of experiments, people judged new evidence on social issues in a biased way, and the fix that worked best was getting them to consider the opposite: to bring to mind the possibilities at odds with what they already believed [2]. Simply trying to be fair worked less well [2].
- Pick the belief you are about to act on: a diet, an investment, a claim about a person or a group, a news story.
- Search for the strongest case against it, in neutral words. Type "evidence against [topic]" or "[topic] pros and cons", never "why [topic] works".
- For each piece of evidence, on either side, ask the mirror question: "Would I accept this evidence if it pointed the other way?" If the answer is no, give it less weight, whichever side it supports.
- Write one sentence: "The best argument against my view is ___." If you cannot write it, you have not finished the check.
Within 3 weeks you can write down the best argument against at least one belief you hold, and you have softened or changed at least one view.
Step 3: Resist an Anchor: Write Your Own Number Before You See Theirs
WHY THIS MATTERS
An anchor is a number that should not matter but still pulls your judgment towards it [4]. When people had 5 seconds to estimate 1 x 2 x 3 x 4 x 5 x 6 x 7 x 8, the median guess was 512; when the same numbers came in reverse order, starting with 8, it was 2,250. The real answer is 40,320 [4]. Have your own number ready, and when theirs comes first, list reasons it is wrong: this step reduced the anchor's effect, including for experts in a real-world setting [5].
- Before you see a price, offer, quote, salary or estimate, write your own number on paper or in your phone.
- Under it, write one line on how you got it: two or three real prices you checked, your budget, or your walk-away point.
- When their number comes first anyway (a list price, a "was 199" sticker, a first offer), write 2 reasons it could be too high or too low before you reply [5].
- Answer from your own number, not from theirs. If theirs is far from yours, say: "That's not what I found. I was looking at ___."
In your next 5 purchases, offers or estimates, at least 3 start from a number you wrote first.
Step 4: Run a Pre-Mortem: Imagine It Failed a Year From Now, Then Fix the Top Risks
WHY THIS MATTERS
Once you commit to a plan, it is hard to see its weak spots, and people in a group often keep their doubts to themselves. A pre-mortem flips the question. You imagine the plan has already failed and work out why, then change the plan to deal with those causes [3]. In a study comparing it with ordinary brainstorming, teams using the pre-mortem found better quality risks, made more useful changes to the plan and spotted more rare, high-impact risks [3].
- Before a big plan (a move, a launch, a wedding, a new business, a training programme), say: "It is one year from now. This plan failed. What went wrong?"
- Everyone writes reasons alone first, for 5 minutes, before anyone speaks. On your own, write for 5 minutes without stopping.
- Share them, then pick the top 2 by how likely and how costly they are.
- Change one thing in the plan for each of the top 2: a backup supplier, a savings buffer, a second person who can step in, a test run first.
- Write one early warning sign for each top risk. Then go to Step 5 and set a stopping rule.
Your next big plan starts with 2 fixes already in place and one warning sign for each.
Step 5: Get Out of Sunk Costs: "Would I Choose It Today?" and a Stopping Rule Set in Advance
WHY THIS MATTERS
Sunk cost bias is the pull to keep going with something because you have already put money, time or effort into it [4]. When World Bank staff were asked about a failing project, the more of its money had already been spent, the more likely they were to keep funding it [4]. The fix is to make stopping easy to justify, with clear and public criteria for ending [4].
- For anything you keep doing mainly because you started (a course, a renovation, a car, a business idea, a subscription), ask: "If I didn't have this, would I choose it today, at today's price?"
- Before you start any new big commitment, write a stopping rule with a date and a number: "If by [date] I have not reached [result], or I have spent more than [amount], I stop or rethink."
- Tell one person your rule, or write it where others will see it. Public criteria make stopping easier [4].
- On the date, judge by the rule, not by what you have already spent.
Every new big commitment this month has a written stopping rule with a date, and you have dropped or changed at least one thing you kept only because you had started.
Step 6: Track Your Five Checks for 60 Days
WHY THIS MATTERS
Bias training that lasted was practice-based, and it beat just watching a video [6]. Your own real choices are your practice ground.
- Keep a decision-check tally for 60 days in a notebook or phone note. For each big choice or plan, write one line: the date, the choice, which of the 5 checks you used (10-second pause, mirror question, your number first, pre-mortem, stopping rule) and whether it changed anything.
- On day 7, count the choices you checked and how many changed. Do it again on day 30 and day 60.
- Aim for 20 or more big choices checked over 60 days, with at least 5 where a check changed your decision or plan.
- Every 3 months, rate yourself on the Skill Check table below.
By day 60 you use at least one check on most big choices without needing to remind yourself, and you can point to 5 choices that came out better.
Conclusion
You will never switch biases off, but you can choose when the slow mind gets its turn. Five checks cover the costly moments: pause on the three signals, test your belief with the mirror question, write your number first, run a pre-mortem, and set a stopping rule. Start with Step 1 this week. It needs no extra time.
Myths
Myth 1: "If I know the names of the biases, I won't fall for them."
Knowing helps only a little. Teaching students about biases gave a small effect, and it is unclear whether it reaches real-life decisions [1]. Practising checks on real problems works better [7].
Myth 2: "Smart, experienced people are less biased."
Professional World Bank staff read the same data less accurately when the topic touched their views, and ability and seniority did not explain it [4]. Experienced judges were pulled by numbers from rigged dice [4].
Myth 3: "Just trying to be fair is enough."
People told to be fair and unbiased corrected less than people told to consider the opposite [2]. A specific question beats good intentions.
Myth 4: "Quitting means everything I put in was wasted."
What is spent is spent either way. Carrying on with a failing plan can waste more [4]. Write down what you learned, and stopping becomes something you can justify [4].
FAQ
Do I need to run all five checks on every choice?
No. Run the 10-second check in Step 1 when a signal fires. Use the others when they fit: the mirror question for beliefs, your number first for prices, a pre-mortem and a stopping rule for big plans.
How long before this works?
A single training session reduced confirmation bias on a later surprise case with no reminder [7]. Lasting change needs repeated practice [6], so give the tally 60 days.
What if the other side's number really is fair?
Then your own research will show it, and you lose nothing. The aim is that your answer comes from your reasons, not from whichever number you heard first [5].
Is a pre-mortem just being negative?
No. You look for failures in order to fix them before you start, and teams that did this found better risks and made more useful changes than brainstorming teams [3].
Do sunk costs affect everyone the same way?
No. Younger adults fell for them more often than older adults, and the pull to use something you paid for faded with time [8].
Skill Check
| Skill | Starting | Solid | Strong |
|---|---|---|---|
| Slow down in time: of your last 5 high-stakes or emotional choices, how many had the 10-second pause before you decided? | 0 or 1 of 5 | 3 of 5 | 5 of 5 |
| Consider the opposite: in the last month, how many strong beliefs did you test against the best evidence against them? | None | 2 | 4 or more, and you changed or softened at least one |
| Your own number first: in your last 5 purchases, offers or estimates, how many started from a number you wrote first? | 0 or 1 of 5 | 3 of 5 | 5 of 5 |
| Pre-mortem: of your last 3 big plans, how many had a pre-mortem with 2 fixes? | 0 | 2 of 3 | 3 of 3 |
| Stopping rules: of your last 3 big commitments, how many had a written stopping rule with a date? | 0 | 2 of 3 | 3 of 3, and you acted on at least one rule |
What Next
- Start your 60-day decision-check tally today and run the 10-second check on the next choice where a signal fires.
- At day 30, pick the check you skip most and practise it on purpose.
- When you want more tools, use the Go further list below.
Go further
- Fast Mind, Slow Mind and Your Bias Blind Spot: Why You See Others' Biases First: why you spot others' biases first.
- Catch Biases Before They Cost You: A Decision Checklist, Real Dissent, a Bias Log and an 8-Week Plan: a fuller checklist and bias log.
- Why You Self-sabotage, and How to Catch It Before It Costs You: the pre-mortem in more depth.
- When to Trust Your Gut: Test Your Intuition Against Outcomes and Train It: when the quick mind is worth trusting.
- Stop Waiting for Certainty: Decide Without All the Answers and Count the Cost of Not Deciding: for when checks turn into stalling.
- Trust Your Own Judgement Again: End Self-doubt and Self-gaslighting: if doubting yourself is the bigger problem.
