Your Brain Would Rather Blame You Than Accept a Meaningless Coincidence

A colleague walks past with a stony face and doesn't say hello.

Within half a second you have a theory: I did something wrong. Then comes the search for what exactly: that email, that comment on the call, refusing to help yesterday.

The version where something happened in her life and you have nothing to do with it also exists. But it doesn't get serious consideration, because there's nothing in it you can fix. The version with you at fault at least provides work.

Helplessness is harder to bear than guilt

In 1975 Ellen Langer ran an experiment that's easiest to summarise like this.

Office workers were sold lottery tickets for a dollar. Some got to pick their own ticket, others were handed one chosen by the experimenter. The draw was random, the odds identical, the tickets interchangeable; the only difference was whose finger pointed.

Before the draw, everyone was offered the chance to sell the ticket back and name their price.

The people who'd been handed a ticket asked for about two dollars on average. The people who'd chosen their own asked for nearly nine.

What's the four-and-a-half-fold premium for? For the sense that the choice decides something. Langer called it the illusion of control: an expectation of personal success pitched higher than the objective probability warrants, with no grounds at all.

An honest caveat: later work suggested part of the effect is explained less by belief in your own odds than by reluctance to part with a ticket that's already yours. The argument about the mechanism is still running; the phenomenon itself replicates reliably.

And one more detail, useful for this cluster. Langer worked out what strengthens the illusion: choice, competition, familiar surroundings, personal involvement. The more a situation resembles one where skill decides, the more the brain behaves as if skill decides here too.

The pigeons who invented a ritual

The second experiment, 1948, Burrhus Skinner.

A pigeon's cage was fitted with a device that delivered food at regular intervals. Not for behaviour, just on a timer. There was nothing the pigeon could do to speed it up or slow it down.

After a while the pigeons started behaving oddly. One turned counter-clockwise, another bobbed its head, a third pushed its head into a corner of the cage. Each was repeating whatever it happened to be doing when the food first arrived, and repeating it more insistently, because the food did indeed keep coming.

Skinner called it superstitious behaviour. The mechanism is simple: a system that looks for cause and effect will find it where there isn't any, if the material allows.

You know this in another phrasing: if I worry hard enough, nothing bad will happen. The pigeon turned in circles, you worry. In both cases the food arrives on its own schedule.

Why this is cheaper than admitting randomness

The article on the spec describes the reason as the priority given to the worst-case scenario. What works here is its continuation.

A predator in the bushes is manageable: run, climb, shout to your people. A world where misfortune happens for no reason is unmanageable in principle, and for a system whose only job is choosing an action, that's a dead end. So the brain prefers any cause to no cause, and checks first whichever cause is easiest to reach. You.

Which gives an unobvious conclusion. Self-blame isn't always about low self-worth. Often it's the reverse side of something else: "I'm the cause" means "I can influence this", and that's easier to carry than "something happened to me that I don't affect".

The formula for this cluster: where there's no manageable cause, the brain appoints one, and the first candidate for the role is you. The four traps below are versions of a single trade: control in exchange for guilt.

Four traps in this cluster

1. Personalisation. Everything that happens is somehow about you

What it feels like. Someone else's behaviour, the weather, a failed project, the mood in the room, all of it gets explained through you. Not as pride, but as responsibility: so I missed something, said it wrong, did the wrong thing.

Example. A colleague looks grim. You run back through your actions for the week. Her child is ill and she isn't mentioning it, because you aren't that close.

When it isn't a trap. When you really are the cause and it can be verified. The project fell over because you didn't send the data in time: that isn't personalisation, that's a fact, and it needs working on. The trap starts where the link isn't checked but assigned.

What helps. Ask, if you can. If you can't ask, invent three versions with you absent from them entirely and rate their likelihood honestly. It usually turns out that people have their own lives, in which you occupy considerably less space than you do in your own head.

2. Magical thinking. Anxiety disguised as action

What it feels like. A sense that worrying prevents something. Relaxing is frightening: what if the bad thing happens precisely then, because I stopped holding it up.

Example. Someone close is flying, and you don't sleep all night. The plane lands normally. The connection between your insomnia and the smooth landing registers as confirmed, although it's the same pigeon turning counter-clockwise.

When it isn't a trap. When the worry turns into action: you don't sleep, and over that night you assemble documents, find a doctor, write a letter. Anxiety that produces action works. Anxiety that only produces more anxiety runs on empty.

What helps. Split it: what can I do about this right now, and what can't I. Do the first. With the second, all that's left is to live alongside it, which is more honest than paying with insomnia for the illusion of influence. And keep the price in mind: anxiety isn't free, it spends a resource you'll need tomorrow for something real.

3. Just world. Good behaviour as prepaid insurance

What it feels like. A persistent sense that effort and decency ought to convert into how you're treated. I tried hard, so it should be recognised. I've done nobody harm, so nothing bad should happen to me.

Example. You carried a project, worked weekends, and someone else got the promotion. The reaction isn't "the odds didn't fall my way" but "this is unfair", and in that word you can hear a contract being broken.

When it isn't a trap. When fairness genuinely applies and there's someone to invoke it with: a contract, a law, company policy. Demanding that agreements be honoured is normal. The trap starts where the agreement existed only in your head.

What helps. Accept that the world isn't built as a reward system for behaviour. It sounds cold and it frees you up: if good treatment can't be bought in advance, it doesn't have to be paid for. What's left is the practical question of where and with whom you spend your time, and that genuinely is your choice.

4. Shoulds. Rules nobody signed

What it feels like. Speech made of "should", "must", "have to". The rules register as absolutes, with no source and no expiry date. Breaking one produces guilt before you've had time to ask where the rule came from.

Example. "I should have known." Should have, to whom, when did I take that on, how was I supposed to know? None of these questions get asked, because "should" has already been said and it works as a verdict.

When it isn't a trap. When the obligation is real: a signed contract, a promise given, a duty under law. There aren't many of those, and as a rule they're written down. Everything you actually owe is usually listed in the tax code or on a bank statement.

What helps. Replace "should" with "want to" or "it would be good if", and see what survives. "I should have replied immediately" turns into "I'd like to reply faster", and that's a conversation rather than a sentence. And ask about the source: who set this rule, and when did I agree to it?

What they have in common

Four traps, one trade.

Personalisation appoints you the cause of other people's behaviour. Magical thinking appoints anxiety as an instrument of influence. Just world appoints behaviour as the currency that buys a good outcome. Shoulds appoint the rules whose observance is meant to guarantee it.

In every case one thing is exchanged for another: you take on guilt or obligation, and in return you get the sense that the situation can be influenced. The trade only looks bad from outside. From inside it looks like the only way to avoid living in a world where anything can happen to you for no reason at all.

Which is why reassurance like "it isn't your fault" doesn't work. It removes the guilt and gives nothing back, and what's needed back is exactly the control. Something else does work: separate where influence genuinely exists from where it doesn't, and reclaim the first instead of clinging to the second.

What to do: separate influence from control

The difference between those words is practical, not philosophical. Control is when the outcome depends on you. Influence is when the probability depends on you.

Split the list in two. Literally, on paper, at the moment it's grinding. On the left, where action is possible: write, ask, prepare, decline. On the right, where it isn't: someone else's decision, someone else's mood, the weather, the outcome of a competition. The right-hand column doesn't require your participation, and that isn't a defeat, it's a resource freed up.

Verify causation instead of assigning it. The question "how do I know this is because of me" sounds boring and works better than any reassurance.

Watch what the anxiety is doing. If it turned into action, it's working. If it's been producing itself for three hours, you're the pigeon, and the food is on a timer regardless.

Look for where the rule is written down. A real obligation has a source, a date and a second party. A should has nothing but a tone of voice.

And here I'll say the professional thing, because this is what I've done for twenty years. Engineering has the notion of a zone of responsibility: what the unit answers for, what the operator answers for, and what belongs to conditions neither of them affects. A system where the operator considers themselves responsible for everything, including voltage fluctuations in the grid, works worse rather than better: attention is smeared thin, and the real points of control get lost in a general sense of guilt.

A head is built the same way. Taking on everything doesn't mean controlling more. It means losing sight of the little that's genuinely in your hands.

All cognitive traps >

Next: Your Reaction Is Not Evidence - the fifth cluster, on why the strength of a feeling says nothing about the truth of a conclusion.