Why Habits Are Contagious: The Science of Peer Effects

You probably think of your habits as personal. Your decision to skip the gym, your third coffee, the late-night scroll. Yours, chosen freely.
The research says something less flattering and more useful: a meaningful share of what you do every day is borrowed. Habits move through social networks the way accents do. Slowly, invisibly, and mostly without permission.
This is not a metaphor. It is one of the better-documented findings in behavioral science, and it has practical consequences for anyone trying to change how they sleep, move, eat, or work.
The running study that made it measurable
For years, researchers suspected that behaviors spread between people, but it was hard to separate contagion from the obvious alternative: similar people simply cluster together. Runners befriend runners. That is selection, not influence.
Then two MIT researchers, Sinan Aral and Christos Nicolaides, got access to five years of data from a global fitness network: 1.1 million runners, roughly 359 million kilometers of logged runs, plus their social ties. Published in Nature Communications, the study used weather as a natural experiment. If your friend in another city gets a sunny day and runs farther, and you then run farther in your rainy city, that is not clustering. That is influence.
The effect was real and quantifiable. On average, an extra kilometer run by a friend on the same day nudged people to run about 0.3 kilometers more themselves. Two details stand out. First, the influence ran mostly in one direction: less active runners influenced more active ones, not the reverse. Comparison seems to work harder downhill. We appear more motivated to protect our position ahead of someone than to chase someone far in front. Second, the effect was strongest between people who saw each other's activity regularly.
In other words: your behavior is data in someone else's environment, and theirs is data in yours. Every visible run, every skipped workout, every "let's just order in" transmits.
Why healthy habits spread differently than gossip
Here is the nuance most summaries skip. Not everything spreads the same way.
Network researchers distinguish simple contagion from complex contagion. A virus, a rumor, a meme: one contact can be enough. But effortful behavior change, the kind that costs something, tends to spread as a complex contagion. It usually needs reinforcement from multiple people before it takes hold.
Seeing one acquaintance take up morning walks rarely moves you. Seeing three people you actually interact with do it starts to shift what feels normal. Your brain is constantly calibrating "what people like me do," and it weights repeated, close-range signals far more than distant ones.
This explains a common frustration: you follow fifty fitness accounts and nothing changes. Distant, one-way signals are weak. One colleague who walks at lunch every day and occasionally says "coming?" is a stronger intervention than an entire feed.
The uncomfortable half of the finding
Peer effects are symmetric. The same channel that can carry a running habit toward you also carries late nights, desk lunches, sarcasm about rest, and the quiet norm that answering messages at 23:40 is what committed people do.
Teams are especially efficient at this. A founder who visibly never stops teaches everyone the real policy, regardless of what the wellness deck says. Research on peer effects among adolescents and students finds the same pattern in both directions: friends' exercise predicts your exercise, and friends' fast-food frequency predicts yours.
None of this means your environment controls you. It means your environment votes, constantly, and willpower is what you spend when all the votes go against you. That is a losing budget over time.
Using contagion on purpose
You do not need to audit your friendships or engineer your social circle. That advice is both unpleasant and impractical. Smaller moves can shift the signal balance.
Make one healthy behavior visible. Contagion needs a signal. If you walk, run, or lift, let it exist in shared space in a low-key way: a walking meeting, a mention, a calendar block others can see. You are not performing; you are transmitting. Someone in your network is one reinforcing signal away from starting.
Borrow a synchronous partner. The strongest effects in the running data came from same-day, visible activity. A standing appointment with one person, even remotely, in different cities, may do more for consistency than any app. You are lending each other your peer effect.
Count the signals you receive. For one week, notice what the five people you interact with most are transmitting about sleep, movement, food, and rest. Not to judge them, just to see your inputs clearly. Behavior you thought was a discipline problem often turns out to be a signal-environment problem, and those are easier to fix.
Add one countervailing voice. If your daily environment is all urgency and no recovery, one regular contact who treats rest as normal can function as a corrective input. Complex contagion cuts both ways; sometimes a single reinforcing source is what tips a behavior from "thing I keep failing at" into "thing people around me do."
The system view
In the Sapiens OS framework, Social Connection is not the soft pillar. It is infrastructure. Connection regulates stress physiology, shapes mood, and, as peer-effect research shows, quietly sets the default settings for most of the other pillars: how you move, when you sleep, what you eat, whether rest is allowed.
This is why trying to overhaul habits alone, through pure self-discipline, is fighting uphill. The system matters more than the insight. Adjust the signals around you, even slightly, and the same amount of effort starts producing more change.
Start with a clear picture
Before adjusting the social inputs, it helps to know which pillars are actually running low. Energy, sleep, stress load, connection, sense of agency: they interact, and the loudest symptom is not always the real bottleneck.
That is what our free self-assessment is for. It takes a few minutes and gives you a snapshot of where your inner operating system is supported and where it is strained. Start with the free self-assessment at https://osapiens.expert/how-i-feel, then pick one signal worth changing.
O!Sapiens is an educational project. This article is for informational purposes and is not medical or psychological advice. If you are struggling with your mental health, please consider speaking with a qualified professional.
Sources: Aral & Nicolaides, Exercise contagion in a global social network, Nature Communications (2017); MIT Sloan, Turns out exercise is contagious; Frontiers in Psychology (2025), peer effects and adolescent physical exercise meta-synthesis; PLOS ONE, Weight-related behavior among adolescents: the role of peer effects.