Why We Overestimate How Much Others Notice Us

Sometime in the late 1990s, Thomas Gilovich at Cornell had an idea for an experiment that, on paper, sounds like a college prank. He handed his participants T-shirts featuring a photograph of Barry Manilow — chosen specifically because the participants found him deeply uncool — and asked them to walk into a room full of other students. After a few minutes, he pulled them out and asked them to estimate how many people in the room had noticed and could identify the shirt.

The participants, on average, guessed around 50 percent. The actual number was about 23 percent.

Then Gilovich ran the same study with shirts featuring people the participants liked (Bob Marley, Martin Luther King Jr.). Same overestimation. The participants' expectation that they would be noticed and judged was roughly double what actually happened. He published the results in 2000 in the Journal of Personality and Social Psychology and named the phenomenon the spotlight effect.

It is one of the more reliably replicated findings in social psychology. And it has more practical use in ordinary life than most people realize.

What the spotlight effect actually says

The technical version: people consistently overestimate how much others notice, remember, and judge their appearance and behavior. The everyday version: nobody is paying as much attention to you as you think they are.

This bias does not require any particular pathology. It happens in confident people and shy people, in extroverts and introverts — though it interacts with the way groups reward confident speakers regardless of accuracy. It is a default feature of having a first-person point of view. From inside your own head, you are the most salient object in every room. So when you try to imagine how the room sees you, you start from your own vivid self-perception and adjust downward — but not nearly enough.

Gilovich's lab spent the next several years stress-testing this. They ran similar experiments with:

• People who had a bad hair day, before going to class
• Students who arrived late to a discussion group
• People who had to wear obviously different clothes than the rest of the room
• Speakers who stumbled over their words in front of an audience

In every variation, participants believed they had been noticed and remembered far more than they actually had. The Barry Manilow study just happened to be the catchiest framing.

A related bias: the illusion of transparency

In a companion paper, Gilovich and his collaborators identified another version of the same egocentric pattern, this one about internal states. The illusion of transparency is the tendency to assume that your hidden thoughts and emotions are more visible to others than they actually are.

In one study, participants tried to hide their disgust while tasting an unpleasant drink. They believed observers could tell. Observers mostly could not. In another study, public speakers who felt visibly nervous were rated by audiences as much calmer than they thought they appeared.

If you have ever walked away from a presentation convinced everyone saw how anxious you were — and then watched a recording and noticed you looked fine — that is the illusion of transparency. The internal weather is much louder than the external signal.

Why it happens

The mechanism is probably very simple: when you try to estimate what someone else perceives, you start from your own perception and try to adjust. But adjusting away from a vivid starting point is hard. You cannot fully discount how obvious your shirt feels to you when guessing how obvious it feels to a stranger.

This is the same kind of egocentric anchoring that shows up in many other places — like the curse of knowledge (experts forget what novices don't yet know) or the planning fallacy (you estimate task duration from your own optimistic perspective). Once you see the pattern, you start spotting it everywhere.

The other factor is that you remember your own stumbles. Other people don't. To you, the moment you forgot someone's name at the party is a vivid, replayable scene. To everyone else at that party, it was one of fifty things that happened that night and is now gone.

What this is useful for

Knowing about the spotlight effect does not magically remove it — you cannot reason your way out of feeling watched. But it changes what you do with the feeling.

Before a presentation or speech, the spotlight effect says you will feel more visibly nervous than you look. The audience will not see your trembling hands the way you feel them. Knowing this won't stop the trembling, but it will stop you from compounding the anxiety with "and everyone can see how anxious I am."

After a small embarrassing moment — tripping in public, saying the wrong word, sending an email with a typo — the spotlight effect says your future memory will badly overweight this event. Almost no one else will think about it after today. The way to test this is to ask yourself: how many embarrassing moments do you remember from strangers you observed last week? Probably none. The same is true for you in their memory.

In day-to-day social anxiety, the spotlight effect is the engine behind a lot of avoidant behavior — not going out because you don't have the right outfit, not asking the question because everyone will notice you don't know. The actual data is that almost nobody notices, and the ones who do mostly don't care.

On social media, the spotlight effect mutates. Likes, follower counts, and view metrics make attention feel measurable, which can both intensify the feeling that you are being watched and distort how you interpret silence. A post with no engagement is not evidence that everyone saw and ignored it — it is mostly evidence that the algorithm did not surface it to many people in the first place.

How to recalibrate

The cleanest way to actually reduce the spotlight effect, as opposed to just intellectually understanding it, is to collect direct evidence against it. Two ways that work:

• Deliberately do something mildly embarrassing in low-stakes situations — wear something a little weird, ask an obvious question in a meeting, walk into a coffee shop and order something unusual. Notice how almost nothing happens. Each data point adjusts your calibration slightly.
• Audit your memory of other people's mistakes. Try to list five embarrassing things strangers did in front of you in the past month. You will almost certainly fail. Then ask: why would your own mistakes be different?

The point is not to stop caring how you come across. Caring is normal and useful. The point is to stop assuming the audience is paying the level of attention you fear, because they almost never are. They are busy being self-conscious about their own shirt.