Your Brain Keeps Two Sets of Books

Take an ordinary day apart. You woke up, nobody died, the coffee came out right, the commute took a normal amount of time, you closed a task, a colleague said thanks, dinner was good. And one sentence from your manager, that the report needs redoing.

By evening, what's in your head is the report.

Everything else didn't exactly get forgotten. It just never made it into the accounts.

Different currencies for income and expenditure

In 1979 Daniel Kahneman and Amos Tversky described the thing Kahneman later won the Nobel Prize in economics for. The idea is almost indecently simple.

Picture a bet: we flip a coin. Heads, you win 150 euros. Tails, you lose 100. Are you in?

Most people decline. Even though the arithmetic is unambiguous: on average that bet returns 25 euros a flip. Turning it down means turning down money.

The researchers then worked out what the win would have to be before people said yes. It came out at roughly twice the loss. Losing a hundred registers about as strongly as gaining two hundred, only with a minus sign.

This is called loss aversion, and in prospect theory it shows up as a curve that's noticeably steeper on the loss side than on the gain side. The same size step down is felt more than the same size step up.

An honest caveat is needed here. "Roughly double" became the textbook number, but it isn't a law. Later reassessments produced values nearer 1.3, often statistically non-significant, and showed that the size depends on the stakes, the domain, and even on how the question is worded. In 2018 Gal and Rucker proposed reconsidering the concept altogether. Almost everyone accepts the effect exists; the argument is about its size and its universality.

Why the books are skewed in the first place is covered in the article on the spec: a loss could cost you your life, a gain could improve your dinner.

The reference point decides what counts as a loss

The second half of the same theory, and for this cluster it matters more than the first.

The brain doesn't evaluate an outcome in itself. It evaluates a change relative to a reference point. The same salary is a win if you compare it with your last job and a defeat if you compare it with a colleague. The same flat is an achievement or a cramped box, depending on what's standing next to it on an imaginary scale.

Which leads somewhere uncomfortable: the reference point isn't chosen by you deliberately, it's whatever turned up first. Usually that's someone else's result, displayed from its best side.

The formula for this cluster: income and expenditure are counted in different currencies, and the exchange rate is set by a reference point you didn't choose. The four traps below are different ways of arriving at a minus where the facts say plus.

Four traps in this cluster

1. Discounting the positive. Income written off as luck

What it feels like. Successes aren't exactly denied. They're moved to a different category: got lucky, good timing, the task was easy, people helped. Failures meanwhile stay personally yours and go into the accounts in full.

Example. The promotion was a favourable set of circumstances and a good economy. The task you botched was the real you. Both things happened to the same person, but they were posted to different accounts.

When it isn't a trap. When you really did get lucky, and admitting it is clear sight rather than self-erasure. The difference is whether you apply the same rule to failures. If "got lucky" explains the success, then "got unlucky" should explain the failure. If it only ever explains the success, that isn't modesty, that's bookkeeping.

What helps. A wins file, literally a file. Record each result at the moment it happens, because in hindsight memory will mark it down. Read it once a month. This isn't affirmations, it's restoring the second column of the ledger.

2. Mental filter. One item out of a hundred makes it into the report

What it feels like. Attention is a limited resource and it isn't handed out evenly. Out of a hundred events, one stays in view: the one that hooked anxiety. The other ninety-nine physically go past.

Example. Ninety-nine kind comments and one criticism. You remember the criticism. And you don't choose to remember it, it's simply the only one left.

When it isn't a trap. When one event genuinely outweighs the rest in consequences. One missed defect on a production line matters more than ninety-nine good parts. The filter becomes a trap when importance is assigned by the strength of the anxiety rather than by the consequences.

What helps. Count. Not recall, but literally recount: open all hundred comments and state the ratio. Memory is no help here, memory is the source of the distortion. A number you see with your eyes works, because it arrives from outside.

3. Minimising. Yours is cheap, theirs is expensive

What it feels like. One yardstick for yourself, another for everyone else. She has talent, I got lucky. He managed because he's good at it, I managed because I nearly killed myself. The double standard runs automatically and usually goes unnoticed.

Example. You ran a difficult project for six months, it ended well, and the explanation is ready: the team was decent, the deadline moved, the client was reasonable. If a colleague had done exactly the same thing, you'd have said she's a strong manager.

When it isn't a trap. When you honestly rate someone else's work higher because it's objectively harder. Simple check: do you extend the same discounts to their result that you extend to your own?

What helps. The friend trick. Imagine a friend described her result in exactly the words you use about yours. What would you say to her? The answer usually arrives instantly, and it's nothing like what you say to yourself.

4. Comparison. A reference point chosen to produce a minus

What it feels like. Comparing yourself with others is a basic mechanism; it's how the brain works out your position in a group. The problem is the material. You compare your inside experience with someone else's outside result. Your chaos, doubts and three rewrites against their finished shop window.

Example. She organises her life effortlessly and I can't find my keys. You saw her outcome. You didn't see how many times she's lost them, what she forgot last week, or what her mornings look like.

When it isn't a trap. When the comparison yields data rather than a verdict: someone does something learnable, and you go and ask how. Comparison works for you right up to the moment it turns into a judgement about yourself.

What helps. Switch to an honest reference point: yourself a year ago. That's the only comparison where you can see both sides in full, the inside experience and the result. And keep the asymmetry of the material in mind: of other people's lives you only ever see the cover.

What they have in common

Four traps, one accounting principle.

Discounting the positive writes off the income. The mental filter stops it reaching the ledger. Minimising undervalues your own contribution. Comparison picks a reference point that makes the total come out negative anyway.

The same thing happens every time: events are counted, but by different rules depending on their sign. And because the rules run automatically, the report looks objective. You don't think "I've decided not to count that promotion". You simply remember the report and don't remember the rest of the day.

Which is why arguing with the conclusion is useless. Convincing yourself the day was fine while one sentence from your manager sits in memory means arguing with the bottom line without touching the entries.

What to do: fix the accounting, not the mood

Record in the moment instead of recalling later. Memory keeps one-sided books; that's a property of memory, not laziness on your part. Anything not written down on the day will be marked down. A wins file, a note on your phone, a line in the calendar, the format doesn't matter. What matters is that the record is made outside your head.

Count rather than assess. A ratio instead of an impression. How many normal days this month against bad ones? How many tasks closed against failed? The impression says "everything's terrible", the count says "eighteen against three".

Choose the reference point deliberately. Since there's going to be one anyway, let it be yours. Yourself a year ago, not somebody's shop window.

Apply the same rule to yourself that you apply to others. If someone else's success is explained by ability, so is yours. If your failure is explained by circumstances, so is theirs.

And the engineering part, which applies here literally. In manufacturing you measure what you decided in advance to measure, by a method described in advance. Otherwise you end up with reporting that counts only the defects, while good output disappears because nobody set up a line for it. A head behaves exactly the same way: what gets counted is whatever has a row in the ledger. Add the second row.

All cognitive traps >

Next: Your Brain Would Rather Blame You Than Accept a Meaningless Coincidence - the fourth cluster, on why guilt is easier to carry than helplessness.