Why Do I Act Different Around My Family?

Why do I act different around my family? It is a context switch, not a personality switch. What the research shows, and what to do at the table.

Gizella Nagyne Palinkas

11/2/20268 min read

Why do I act different around my family? Because the self you bring to a room is partly built by the room — and your family is the oldest, most specific room you have. It is a context switch, not a personality switch: the person who runs meetings is still in there, but the cues at that table are from 1998, and they call up the version of you who lived through 1998.

The reframe that helps, before you have worked anything else out: this is who I am with them, not who I am.

What is happening

Most of us assume personality is a fixed object you carry everywhere, like a passport. If that were true, being thirty-eight and sulking in your parents' kitchen would be a failure of character.

The more useful picture is that the self is conditional. It has if-then shapes: if I am here, with these people, then this is the version that comes forward. You are not inconsistent for having several. You are responsive, which is what a social animal is supposed to be.

Families are unusually powerful at this for three reasons that have nothing to do with how healthy the family is.

The cues are dense and old. The smell of the house, who sits where, the pitch of your mother's "mmm". These are not neutral details; they are the exact context in which an earlier version of you was learned.

The roles were assigned before you could negotiate them. The responsible one, the funny one, the difficult one, the baby. Those labels were handed out when you were six and nobody has called a meeting to review them since.

Everyone is doing it at once. Your sibling slips into their old shape, which calls up yours, which confirms theirs. Four people can re-enact a dynamic nobody wants, and all four can leave convinced it was the others.

What the research says

I want to be careful here, because this is a subject where plausible-sounding studies get invented.

The strongest thing I can point you to is a line of social-cognitive psychology on what its authors call the relational self. In a 2016 review in Frontiers in Psychology, Andersen and colleagues argue that "the self is intrinsically embedded in an interpersonal context such that it varies in IF–THEN terms, as the relational self." They describe how "representations of the significant other and the relationship with that other are automatically activated by situational cues and that this activation affects both experienced and expressed aspects of the self and personality."

Two honest caveats. First, that paper is a review of the authors' own body of work, not a single experiment with a headline number, so I cannot give you a percentage or an effect size from it. Second, its examples are mostly about new people who happen to resemble someone significant — a coworker with your brother's laugh — rather than sitting down with your actual brother. It supports the mechanism. It does not, on its own, prove the family-dinner version of it.

What I can give you with numbers is how common the experience is.

Ahead of the 2021 holidays, the American Psychiatric Association polled 2,119 US adults and asked what they felt anxious about. 26% said "feeling like you can't be yourself with your family." And it fell steadily with age: 35% of adults aged 18–34, 33% of those 35–44, 24% of those 45–64, and 14% of those 65 and over.

Nearby in the same poll: 32% were anxious about feeling socially uncomfortable, 28% about seeing family members they do not like, and 27% about fighting with family. That poll is five years old and was taken in a pandemic year, so read it as evidence that this is widespread, not as a current measurement.

An everyday example

You are home for two days. Within an hour your father asks whether you have thought about your tires.

At work, a question about your competence gets a calm answer. Here you hear the 1998 version — you are not quite capable — and something answers it before you decide to. You are curt. He becomes more insistent, because insistence used to work. Now you are both in it.

Nobody in that kitchen chose any of that. The cue arrived, the old shape loaded, and the conversation ran on rails. The only striking thing is how fast it happened — and it is fast because it is practiced.

What this does not necessarily mean

It does not mean your family is toxic. This happens in warm families. Density of history is enough.
It does not mean you have not grown. A version of you that activates is not a version of you still in charge of your life. You can revert at the table on Thursday and run your department on Monday.
It does not mean they are doing it on purpose. They are almost certainly doing their own version of this, with as little choice in it as you.
It does not mean anyone here has a condition. This is a pattern, not a verdict on anybody — not on you, not on your mother, not on your sibling. Nothing here can diagnose a person, and it is not trying to.
It does not mean you have to cut anyone off. That is a serious decision about a real relationship, and regressing at Thanksgiving is not evidence for it.

Practical things to try

Notice the switch instead of fighting it. The sentence is "I am in the old shape", said silently. Naming it does not stop it, but it puts a sliver of distance between cue and reply.

Pick one small thing to do differently — one. Not "be my adult self". Something with edges: answer the tire question with one friendly sentence and no defense. Offer to drive. Ask your sibling a real question. One is enough, because you are proving the dynamic has some give, not rebuilding it.

Change the physical setup. Sit somewhere else. Arrive separately. Walk with one person instead of talking to everyone. Half of what loads the old role is the staging.

Give yourself an exit with a reason attached. "I'm going to get some air for ten minutes" is a complete sentence, and leaving before you are annoyed is easier than leaving after.

Write down what happened afterwards. Two lines, not a diary: what triggered the switch, and what you did. Three visits in you will know your own cue, which is more useful than any general theory, including this one.

When professional help may be appropriate

Reverting to an old role for a weekend is ordinary. It is worth talking to a qualified professional if being around your family reliably leaves you in a state that takes weeks rather than days to settle; if visits bring up feelings you cannot put down afterwards; if you dread contact for months in advance; or if anything in that history was frightening or unsafe. Those are not harder versions of this subject — they are a different subject, and the right person is a therapist or your doctor, not a blog.

Take the quiz: which old role loads first?

Six questions on what happens in the first hour, and which cue does it.

Frequently Asked Questions

Q: Why do I feel like a teenager again around my parents?

A: Because the context that is doing the work is the same context in which being a teenager was your job. Your internal sense of yourself is partly cued by the situation you are in, and nowhere matches the original situation as closely as the house, the people and the running jokes you grew up with. The regression is a measure of how precise the match is, not of how far you have failed to come. It tends to be sharpest in the first hour and after alcohol, and it loosens as the visit stops being an event.

Q: Is this the same thing as code-switching?

A: They overlap but they are not identical. Code-switching usually describes a deliberate or semi-deliberate adjustment of how you speak and present yourself for a particular audience, often with real social stakes attached. What this article describes is less chosen than that: an older configuration of the self being cued into place before you have decided anything. You can code-switch on purpose at a family dinner and still be ambushed by the old role underneath it.

Q: How do I stop acting different around my family?

A: You probably cannot stop it, and aiming to is what makes the weekend exhausting. What you can change is what you do in the gap after it happens — one friendly sentence instead of a defense, one walk instead of three hours at the table, one early exit with a reason attached. Our Mind Decoded's general position on patterns like this is that interrupting them once in a specific moment is worth more than deciding to be a different person at the door.

Q: My sibling brings out the worst in me. Are they the problem?

A: Probably neither of you is "the problem", which is the frustrating answer. Two people who learned their roles opposite each other will cue each other automatically, and each will experience the other as starting it, because from inside, your own response always feels like a reaction. The only lever you actually hold is your own half. Changing one move in it often changes more of the dynamic than seems reasonable, and it does not require them to agree to anything.

Q: Does this get better with age?

A: The one measurement I have points that way, though it is indirect. In the American Psychiatric Association's 2021 poll, the share of adults anxious about not being able to be themselves with their family fell from 35% among 18- to 34-year-olds to 14% among those 65 and over. That is a snapshot across different age groups rather than the same people followed over time, so it cannot tell you whether individuals ease off or whether older cohorts simply answer differently. Treat it as encouraging, not as a promise.

Q: Why does it hit hardest at holidays specifically?

A: Holidays stack every amplifier at once: the original setting, the full cast, a longer stretch than usual, disrupted sleep, alcohol, and an explicit expectation that everyone will enjoy it. The expectation is its own pressure, because now a flat moment feels like a failed occasion. In the same 2021 poll, 41% of US adults said their stress level increases during the holidays, against 7% who said it decreases.

Free: the Our Mind Decoded newsletter

One piece a week on why people do what they do, including you. No cost and no product attached to it.

Read Our Mind Decoded →

Mood & Gratitude Journal (printable)

Seven printable pages: a daily mood and three-good-things page, a weekly reflection, a 31-day color grid and a "what helps me" page. A4 and US Letter, instant PDF, for noticing your own pattern across several visits rather than arguing with it from memory. €5.00.

See the Mood & Gratitude Journal →

Sources

Andersen SM, Tuskeviciute R, Przybylinski E, Ahn JN, Xu JH. "Contextual variability in personality from significant-other knowledge and relational selves." Frontiers in Psychology, published 7 January 2016 (Volume 6 – 2015). DOI: 10.3389/fpsyg.2015.01882. https://www.frontiersin.org/journals/psychology/articles/10.3389/fpsyg.2015.01882/full — the two quoted sentences, and the if-then description of the relational self. This is a review article, not a single experiment; it reports no effect size or sample size for the relational-self findings, and its own examples concern new acquaintances who resemble a significant other rather than family gatherings. Both limits are stated in the body above.
American Psychiatric Association, "Holiday Stress" polling presentation, November 2021. https://www.psychiatry.org/File%20Library/Unassigned/APA_Holiday-Stress_PPT-REPORT_November-2021_update.pdf — the 26% figure and its age breakdown (35% / 33% / 24% / 14%), the neighboring anxiety items (32%, 28%, 27%) and the 41% versus 7% stress figure. Sample: 2,119 US adults, interviewed online 17–21 November 2021, weighted to population targets, margin of error ±2 percentage points. Note that this is the American Psychiatric Association, not the American Psychological Association, and that the poll is from 2021 and was taken during the pandemic.
American Psychiatric Association, "As Holiday Season Begins, America's Stress Rises, But Less About COVID-19," 1 December 2022. https://www.psychiatry.org/news-room/news-releases/as-holiday-season-begins-americas-stress-rises — read for context (December 2022 Healthy Minds Monthly Poll, 2,209 US adults, fieldwork 9–14 November 2022, Morning Consult, ±2 points). No figure from it is quoted in the article, because its family items are COVID-specific and would not transfer to 2026.
The date of US Thanksgiving 2026 (Thursday 26 November) was confirmed by search in the Trend Radar of 2 November 2026 and is arithmetic: November 2026 begins on a Sunday, so its Thursdays fall on 5, 12, 19 and 26, and the fourth is the 26th.

Not used, and why: Andersen & Chen (2002), the primary theory paper, could not be retrieved in this run (the PubMed page returned a bot challenge), so nothing is attributed to it. A specific experiment on context-dependent memory with exact recall figures was sought and not obtained, so no such figures appear. No number is given for how many people revert to childhood roles, how long it lasts, or how often, because no source fetched for this article states one.