6 Ways Your Brain Tricks You Into Making Bad Decisions
In 1979, two psychologists at Stanford and the University of British Columbia published a paper called “Prospect Theory: An Analysis of Decision Under Risk” in the journal Econometrica. The authors were Daniel Kahneman and Amos Tversky. The paper was dense, technical, and unusual for the journal — which mostly published mathematical economics. It became one of the most-cited papers in the social sciences. Twenty-three years later, Kahneman won the Nobel Prize in Economics for the ideas inside it. Tversky had died of cancer six years earlier and could not share the prize.
What Kahneman and Tversky showed, across that paper and decades of work surrounding it, was that the standard economic model of the human being — a rational actor who weighs costs and benefits and chooses the optimal outcome — was incorrect at a fundamental level. People do not work that way. Their decisions are systematically distorted by predictable cognitive patterns, and those patterns are so consistent that the deviations can be measured to a decimal place. Below are six of the most thoroughly documented examples. None of them are signs that you are stupid. They are how the human brain is built. Knowing they exist does not eliminate them. It only gives you a fighting chance.
Confirmation Bias: You Test Your Beliefs by Trying to Confirm Them
In 1960, a British psychologist named Peter Wason set up an experiment that would become one of the most famous in cognitive psychology. He told participants he had a rule in mind that applied to sequences of three numbers, and gave them an example that fit the rule: 2, 4, 6. He asked the participants to figure out the rule by proposing their own three-number sequences. Each time they proposed one, he would tell them whether it fit or not. They could propose as many as they wanted.
The actual rule was simple: any three numbers in ascending order. But almost no one figured it out. The participants would form a hypothesis — “it’s even numbers going up by twos” — and then propose sequences designed to confirm it. 8, 10, 12. Yes, that fits. 20, 22, 24. Yes, that fits. They would then announce their rule, confidently, and be told they were wrong. Of Wason’s 29 subjects, only 6 reached the correct conclusion without first reaching at least one incorrect one.
The participants weren’t dumb. They were doing what humans almost always do when testing a belief: looking for evidence that supports it. The faster route to the right answer was to propose sequences that should fail — 1, 2, 3, or even 6, 4, 2 — to see whether the rule could be broken. Almost nobody did this. The brain’s default mode is to search for confirmation, not refutation. This pattern, which Wason named confirmation bias, has been replicated thousands of times across every domain in which humans hold beliefs. Politics, medicine, science, relationships, investing. The information that contradicts your view is harder to see, harder to remember, and easier to dismiss — and you are barely aware that this filter is operating.
The Sunk Cost Fallacy: You Keep Paying for Decisions You’ve Already Lost
In 1985, the psychologists Hal Arkes and Catherine Blumer ran a now-classic study at Ohio University. They sold season tickets to the campus theater under three randomly assigned conditions: full price ($15), a $2 discount, and a $7 discount. The shows were the same. The seats were the same. The only variable was how much each ticket holder had paid. Then Arkes and Blumer tracked how often each group actually attended the performances.
The full-price ticket holders attended more performances. The deeper the discount, the lower the attendance. From a rational standpoint, this makes no sense — the cost of attending a show you’ve already paid for is the same regardless of what you paid. The money is gone either way. But the money wasn’t actually gone in the minds of the ticket holders. The more they had paid, the more strongly they felt the pull to “get their money’s worth,” even though their money was equally unrecoverable in all three conditions.
This is the sunk cost fallacy. The phenomenon is sometimes called the Concorde fallacy, after the supersonic airliner that the British and French governments continued to fund in the 1960s and 1970s long after it had become clear the project would never recover its development costs. The plane flew until 2003. It never broke even. The phenomenon was named in 1976 by the biologist Richard Dawkins and his student Tamsin Carlisle in a paper for Nature, observing the same logic in animals: a parent bird that has invested heavily in feeding a chick will keep feeding it even when the energy could be better spent on a new clutch.
Loss Aversion: Losing a Hundred Dollars Hurts Twice as Much as Finding One Feels Good
The central finding of Kahneman and Tversky’s prospect theory was that the brain processes gains and losses through different machinery, and that the machinery for losses is much louder. In their original 1979 paper, and confirmed in their 1992 follow-up, they measured the asymmetry at a coefficient of roughly 2.25 — meaning losses are weighted about 2.25 times more heavily than equivalent gains. Lose $100 and the pain is roughly equivalent to finding $225 worth of pleasure.
This is loss aversion, and its consequences are everywhere. Investors hold losing stocks far too long because selling them would convert a paper loss into a realized one — even though the paper loss is just as real. People stay in unhappy jobs because the certain loss of leaving outweighs the uncertain gain of something better. Negotiations stall because both sides weight their concessions (losses) more heavily than the other side’s concessions (gains), so neither feels the deal is fair even when it is mathematically balanced. A 2020 multi-country replication study confirmed Kahneman and Tversky’s original findings across cultures with a 90% replication rate on the key tests.
The endowment effect is loss aversion’s most amusing manifestation. In a series of experiments by Daniel Kahneman, Jack Knetsch, and Richard Thaler in the early 1990s, undergraduates were randomly given coffee mugs and then offered the chance to sell them. The selling price they demanded was reliably about twice the buying price participants without mugs were willing to pay. The mug had value because they owned it. Twenty seconds earlier, they hadn’t owned it, and it would have been worth half as much. Ownership itself had created the inflation.
Anchoring: The First Number You Hear Sticks, Even When It’s Random
In one of the strangest experiments in the Kahneman-Tversky body of work, the two psychologists spun a wheel of fortune rigged to land on either 10 or 65 in front of their participants. After the wheel stopped, they asked the participants two questions: was the percentage of African nations in the United Nations higher or lower than the number on the wheel, and what did the participants think the actual percentage was?
The number on the wheel was obviously random. The participants had watched it spin. They knew it had nothing to do with Africa or the United Nations. And yet: the participants who saw the wheel land on 10 estimated, on average, that 25% of African nations were in the UN. The participants who saw the wheel land on 65 estimated 45%. A meaningless number, generated by a rigged wheel of fortune in front of their faces, had moved their answer by 20 percentage points.
This is the anchoring effect, and once you know it exists, you start to see it everywhere. The “manufacturer’s suggested retail price” on a product tag is an anchor designed to make the actual price feel like a deal. The first number named in a salary negotiation defines the range the rest of the conversation will inhabit. Real estate agents use it on buyers and sellers simultaneously, by presenting comparable properties at chosen price points before discussing the property in question. The anchor doesn’t have to be accurate. It doesn’t even have to be plausible. It just has to be the first number your brain encounters, after which all subsequent judgments will adjust insufficiently away from it.
The Availability Heuristic: You Estimate Risk by What You Can Easily Picture
In 1973, Tversky and Kahneman published a paper introducing what they called the availability heuristic — the brain’s habit of estimating the frequency or probability of an event based on how easily examples come to mind. The shortcut works most of the time. Things that happen often are easier to remember. But the shortcut also fails predictably whenever something is unusually memorable for reasons unrelated to frequency.
Plane crashes are the canonical example. They are vanishingly rare on a per-mile basis. They are also televised, narrated, and replayed for days. Every plane crash you can recall is occupying a disproportionate footprint in your memory relative to its actual statistical weight. Your brain reads the footprint and concludes that flying is dangerous. The math says otherwise. The math is not what your brain consults when it makes the snap judgment in the moment.
This is why news coverage of any rare event — terrorism, shark attacks, child abductions by strangers — creates a wave of public concern that vastly outstrips the actual change in risk. The event itself didn’t get more common. The availability of examples got more saturated. The same heuristic explains why doctors who’ve recently had a patient die of a rare disease are more likely to order tests for that disease in the next few weeks, why people fear flying after a crash but not driving after the much larger weekly toll of road deaths, and why the things you worry about often correlate with whatever you happened to watch on the news yesterday rather than with what actually threatens you.
The Planning Fallacy: You Believe Your Next Project Will Go Differently This Time
In 1979, Kahneman and Tversky introduced a phenomenon they called the planning fallacy: the human tendency to underestimate how long a task will take, how much it will cost, and how much can go wrong, even when previous similar tasks have always overrun on every dimension. The forecast is consistently optimistic. The execution is consistently not.
The Sydney Opera House is the textbook case. Original budget: $7 million. Original timeline: 1963. Actual cost: $102 million. Actual completion: 1973. A factor of fourteen over budget, ten years late. This is not exceptional. Studies of public infrastructure projects routinely find cost overruns averaging 50% to 100% and time overruns of similar magnitude. The forecasters are not lying — at least not always. They are subject to the same cognitive trap as everyone else: imagining the project succeeding, plotting the steps required, and failing to budget for the fact that on every comparable project in history, something unexpected has eaten months and millions.
The fix Kahneman proposed is called “reference class forecasting”: instead of asking how long your project will take, ask how long projects like yours have historically taken, and trust that number more than your own estimate. Almost nobody does this. The brain’s intuition that this time will be different is too powerful. The renovation will be different. The dissertation will be different. The startup will be different. They almost never are. And when they aren’t, you don’t learn the right lesson; you blame the specific circumstances, file the experience under “unusual,” and start the next project with the same optimism.
What these six patterns have in common is that they all feel like thinking. The judgment arrives smoothly, with confidence, accompanied by a story explaining itself. The story is plausible. The story is also generated after the fact, by a brain that already made the call. Kahneman spent a career trying to teach himself out of these patterns. By his own admission, in his late seventies, he was still falling into them. The most you can hope for is to recognize the moment of the trap and slow down. The trap is not a malfunction. It is what your brain does when it’s working normally.
