Choosing your memes — why role models are the highest-leverage decision
Something clicked today that connects several threads in this project.
The Agency hypothesis says a transmissible mindset — not systems, not institutions — drives social mobility. The Hungarian study and the Kresy case show this mindset persisting across generations regardless of political system. The uprootedness hypothesis explains what happens when families learn to invest in what can't be seized.
But there's a missing piece: how does the mindset actually transmit?
The answer, across multiple independent research traditions, is the same: through the people you observe. Through role models. Not as inspiration — as infection.
The meme is the mechanism
Richard Dawkins coined "meme" in 1976 as a unit of cultural transmission — an idea or behaviour that spreads through imitation. Susan Blackmore went further: humans are "meme machines," brains evolved specifically as selective imitation devices. We don't choose to copy the people around us. We are built to.
The families in the Hungarian study weren't sitting their children down and explaining strategic awareness. They were modelling it — thousands of small decisions per day that their children absorbed into what Daniel Kahneman would call System 1: the fast, automatic, effortless mode of thinking that handles the vast majority of daily life. System 1 runs on heuristics — informal rules of thumb learned from observation. Your role models provide the training data.
This is what makes role model selection so consequential. You don't consciously choose which heuristics to absorb. You choose (or fail to choose) the source, and the heuristics flow in automatically. A role model isn't one idea. It's an entire package: decision patterns, emotional responses, time orientation, risk tolerance, relationship to discomfort. The memetics research calls this a memeplex — a bundle of memes that travel together.
Girard saw it first
René Girard's mimetic desire theory goes deeper than behavioural imitation. His central claim: we don't even know what we want until someone else shows us. "Man is the creature who does not know what to desire, and he turns to others in order to make up his mind."
Desire is triangular — subject, model, object. The model transforms how you see the object by wanting it. A career path seems valuable because your model pursues it. A lifestyle seems desirable because someone you admire lives it. The object hasn't changed. Your perception of it changed because a model mediated it.
This raises a question I keep circling back to: Are the things I'm pursuing genuinely mine, or did I absorb them from someone I happened to be around?
Luke Burgis calls the distinction thick versus thin desires. Thick desires are anchored in deep experience — they survive changes of environment. Thin desires are mimetically absorbed and evaporate when you remove the source. If you stopped following everyone you currently follow, which of your ambitions would remain?
The science is unsettling
The research on how role models affect decisions is more specific than I expected:
- Chartrand and Bargh's chameleon effect (1999): people nonconsciously mimic the postures, mannerisms, and behaviours of those around them. Your brain copies interaction partners automatically.
- Bandura's social learning: children reproduce observed behaviour using identical actions and words, without being rewarded for it. We model people who are high-status, similar to us, and visibly rewarded.
- A 2019 PMC study on role models and self-control: exposure to a role model changed participants' intertemporal choices — how they weighed present versus future rewards — and the effect persisted three months later, even when subjects were only asked to memorise the model's decisions, not adopt them.
- Christakis and Fowler (Framingham Heart Study, 12,067 people): a person's chances of becoming obese rose 57% if a friend became obese. Behaviours spread to three degrees of separation — friends of friends of friends.
The last finding is especially relevant. Role model memes don't just change you. They propagate through you into your network. The families in the Kresy study weren't just shaping their own children — they were creating a memetic environment that radiated outward through the community.
The compounding problem
Here's the part that makes this feel urgent.
James Clear's marginal gains framework: 1% improvement daily = 37x over a year. 1% decline daily = near zero. Dave Brailsford applied this to British Cycling and won 66 Olympic golds.
Now apply it to decision heuristics. Two people wake up every morning with slightly different System 1 defaults — different absorbed patterns about how to spend time, how to react to setbacks, whether to save or spend, whether to persist or quit, whether to learn or scroll. Each individual difference is trivial. But across thousands of decisions per day, compounding over years, the trajectories diverge enormously.
This is the mechanism I've been looking for. The intergenerational persistence that no political system seems able to break — the 0.6–0.8 correlation in social mobility research — might not be mysterious at all. It might simply be the compounding effect of transmitted decision heuristics, absorbed through role-model observation and operating below conscious awareness.
Why do the same families keep ending up on top? Not because of money or connections or genetics. Because the memes running their System 1 — the ones absorbed from watching their parents, who absorbed them from watching their parents — produce slightly better decisions at scale. Every. Single. Day.
Buffett already knew
Warren Buffett: "The best thing I did was choose the right heroes."
And more pointedly: "Tell me your heroes and I'll tell you how your life will end up."
Charlie Munger: "I believe in the discipline of mastering the best that other people have ever figured out. I don't believe in just sitting down and trying to dream it all up yourself. Nobody's that smart."
Seneca, two thousand years earlier: "Associate with those who will make a better man of you. Welcome those whom you yourself can improve."
These aren't motivational quotes. They're statements about the mechanics of memetic transmission. Buffett isn't saying heroes make you feel good. He's saying they install the heuristics that determine your decisions for the rest of your life.
The counterpoint: Emerson and Naval
Emerson warned that "Imitation is Suicide." Naval Ravikant extends this: "When you compete with people, it's because you're copying them."
But this isn't a contradiction. Girard's framework resolves it. The danger isn't learning from models — it's internal mediation, where you imitate someone close enough to become their rival. External mediation — learning from people far enough from your competitive space — lets you absorb their operating system without the rivalry trap.
The goal isn't to become your role model. It's to absorb their decision heuristics while running your own content. Install their System 1 patterns. Keep your own desires.
What this means for the project
I wrote in next steps about "people who are six months ahead of you — close enough in mindset that their experience feels relevant." That's Girard's external mediation. Close enough to be useful. Far enough to be safe.
If the agency mindset transmits through role models, then "choosing your memes" — deliberately curating who you observe, follow, and learn from — is the foundational practice. Not because it's self-help advice, but because it's the mechanism. The Kresy families didn't choose their memes consciously — displacement did it for them, forcing a reorientation toward portable human capital. But if you haven't been uprooted, you need to do the selection deliberately.
This is what memetic hygiene means in practice: auditing your actual meme sources (not the ones you'd name, but the ones that actually occupy your attention), testing whether your desires are thick or thin, and curating exposure toward people whose System 1 defaults you'd want running in your own head.
And the evaluation criterion isn't whether the memes are true — it's whether they're useful. The heuristics the Kresy families transmitted weren't truth claims. They were strategies that compounded. You evaluate a strategy on whether it works, not on whether it corresponds to reality.
The question isn't whether you'll absorb memes from role models. You will. The question is whether you'll choose them, or let them choose you.