5.8 — How to Stop Taking Things Personally

Students were asked to put on an embarrassing T-shirt and walk into a room of other students. Afterwards they estimated what proportion of people in the room had noticed the shirt.

They guessed roughly double the real figure.

This is the spotlight effect, and it has been shown for embarrassing clothing, for good clothing, for mistakes during a talk, for a bad haircut, for contributions to a group discussion. You consistently overestimate how much other people are attending to you, for a simple structural reason: you are the centre of your own experience and cannot easily model a world in which you are peripheral.

There is a matching finding for the inside. The illusion of transparency is the tendency to believe your internal state is visible — that everyone can see you are nervous, or angry, or lying. They mostly cannot. People are much worse at reading you than you assume, which is inconvenient when you want to be understood and extremely useful when you are anxious.

Both point at the same conclusion, and it is the foundation of this page. Most of what feels aimed at you was not aimed at all.

The attribution error

When someone does something, there are two kinds of explanation: something about them, and something about their situation.

For other people's behaviour, you systematically overweight the person. He is rude. She is careless. They do not respect me. This is the fundamental attribution error and it is one of the more robust findings in social psychology, though its size varies across cultures — it is somewhat weaker in societies that attend more to context, including much of South and East Asia.

For your own behaviour, you correctly weight the situation. You were short with someone because you were late and had not eaten and the day had already gone wrong. You do not conclude that you are a short-tempered person.

The asymmetry is not hypocrisy. It is information. You can see your own situation in full and you can see almost none of theirs. From outside, all that is visible is the behaviour, so the behaviour becomes the person.

The correction is a habit, not an insight: whenever you find yourself explaining someone's behaviour by their character, spend ten seconds on what their situation might be. Not to excuse it. To get the estimate right.

What is actually going on with most people, most of the time

The base rates matter here (1.4), and they are not flattering to the assumption that things are about you.

People are absorbed in their own concerns. Their own money, their own health, their own argument from this morning, their own child, their own bad night. The short reply is nearly always about the eleven other things in their head.

People are often physically off. Tired, hungry, in pain, hungover, sick, too hot. All of these produce curtness, and none of them are about you (2.8).

People are frequently bad at communication. Blunt because they are efficient, cold in writing because writing strips tone, brief because they type slowly. Written messages in particular lose nearly all the signal that would have told you it was fine — the same words that would be neutral spoken read as sharp on a screen, and studies of email find that senders reliably overestimate how clearly their tone comes through.

And people are inconsistent. The same person is warm on Tuesday and distant on Thursday for reasons that have nothing to do with anyone.

Put together: if you assume "this is not about me" as your default and are wrong ten per cent of the time, you will be more accurate than someone who assumes the opposite — and vastly less miserable.

The three questions

When something lands as personal, run these in order. Thirty seconds.

One: what actually happened, in camera language?

Strip every interpretation. "He replied with two words." Not "he brushed me off." "She didn't say hello." Not "she ignored me."

This step alone resolves a surprising share of it, because the interpretation was doing all the damage and it was not in the evidence.

Two: what are three other explanations that fit these same facts?

Force three. The first one is easy, the second takes effort, and the third is where it breaks — because producing three makes it obvious that your original reading was one of several, chosen not because it fit best but because it arrived first (2.6).

He replied with two words. He is in meetings all day. He types badly on his phone. He answers everyone in two words. He is annoyed about something unrelated. He is annoyed with me.

Five options. One of them is the one you assumed.

Three: is there evidence for the personal reading, from more than this one event?

One data point is noise. A pattern across weeks, from the same person, in the same direction, is signal. This is the question that keeps you from being naive — because sometimes the answer is yes, there is a pattern, and then you have a real situation to address rather than a reaction to manage.

The three cases, and what each needs

Case one: it genuinely was not about you. Most of the time. Let it go, and let it go properly rather than nursing it while telling yourself you have let it go.

Case two: it was about you and it was fair. The criticism was accurate, the annoyance was earned. Then the correct response is the one in 4.8: take the act, not the self. Fix the thing. Notice how much of the sting in this case is shame rather than injustice, and treat it accordingly.

Case three: it was about you and it was not fair. This happens. Someone is genuinely dismissive, genuinely undermining, genuinely unkind. Then not taking it personally is exactly the wrong advice — it is about you, and the response is to name the behaviour and state what you require (5.7, 13.4).

The mistake people make is applying one case's response to all three. The person who takes everything personally is running case three's response on case one's events. The person who never stands up for themselves is running case one's response on case three's events. The three questions exist to tell you which case you are in.

When it is about you but not aimed at you

A category worth separating, because it causes a specific and avoidable kind of hurt.

Sometimes you are genuinely the target of an emotion that has almost nothing to do with what you did. Displacement (5.1) — the difficult client's anger arriving at your family. Someone's trigger firing (5.2) on a phrase you used innocently. Someone's shame turning into attack (4.8).

In these cases the anger is real, it is pointed at you, and it is not about you. Recognising this is the difference between "why does he think I am incompetent" and "he is furious about something and I said the wrong word".

The response is 4.9, not defence. Acknowledge the state, do not argue the content, and address the substance later. And afterwards, if it is a pattern, that is 13.5.

The special difficulty: silence and absence

The hardest things to not take personally are the ones with no content at all. The message that goes unanswered. The invitation that does not come. The person who is warm to everyone and neutral to you.

Absence is where your mind fills in most freely, because there is no evidence to constrain the story (2.2's blind spot, socially).

Three things help.

Notice that the interpretation is unconstrained. With nothing to go on, whatever you conclude was generated entirely internally.

Consider the arithmetic of other people's inboxes. Most unanswered messages were seen at a bad moment, mentally deferred, and buried. This is the overwhelmingly common explanation and it feels insufficient because it is so mundane.

And where it matters, ask. One plain sentence, without accusation: "Haven't heard from you — everything okay?" Asking is nearly always better than a fortnight of hypothesis, and the reluctance to ask is usually the fear that the bad interpretation will be confirmed, which is precisely the fear that keeps it alive.

What to do with this

Adopt the default: it is probably not about me, and I will find out if it repeats. This is not naivety, it is a base rate (1.4), and it is more accurate than the alternative.

Force three alternative explanations, out loud, every time something stings. Three, not one. The third is the one that does the work.

Wait for the pattern before concluding. One event is noise. Three events in one direction from the same person is data, and then you act on it deliberately rather than reacting to each instance.

Say the two sentences. "That may have nothing to do with me." And, if it might: "I'm going to ask rather than guess."

And reduce the surface area (5.6). Someone whose sense of being all right does not depend on this person's approval finds a short reply mildly puzzling rather than wounding. That is the structural fix, and everything else on this page is technique for while you build it.

Next: 5.9 turns from a specific reaction to a general capacity — patience and calm treated as trainable skills rather than as a temperament you either have or do not.