When she moved apartments last spring, she hired movers rather than ask a single person for help — not because nobody would have come. Plenty would have. That was never the question.
She’s the one who mentioned her surgery three weeks after it happened, casually, as logistics. Who’s warm with everyone and known by no one. Who ends every story about handling things alone with the same five words: I didn’t want to bother anyone.
Unlikable? The opposite — people like her enormously, which makes the moat around her all the more confusing. But psychology has a precise account of what’s happening here, and it starts with a piece of mental machinery every human being runs, whether they know it or not.
Hers is just calibrated differently. And it was calibrated a long time ago.
Everyone runs the same ledger
Relationship scientists Sandra Murray, John Holmes and Nancy Collins built their careers on what they framed as the basic dilemma of interpersonal life: we’re wired to need closeness, but closeness requires making yourself rejectable — and the psychological costs of rejection only grow as closeness grows. Their famous opening question: how do people find the courage to love when getting closer just raises the stakes of getting hurt?
Their answer — the risk regulation model — is that everyone carries an internal system running this exact ledger, constantly and automatically: how safe is it to move toward this person? When the system reads acceptance as likely, it green-lights connection — the disclosure, the ask, the reaching out. When it reads rejection as possible, it flips priorities to self-protection: pull back, reveal less, need nothing.
Everyone has the system. What differs between people isn’t the machinery — it’s the setting. And research is clear about who runs theirs pinned to self-protection: people whose histories have left them chronically unsure of acceptance have regulation systems calibrated to prioritize self-protection over connection, essentially as a standing policy.
Which raises the real question about the woman with the movers: who set her dial?
The eight-year-old actuary
Think of the childhood version of this system as a small insurance actuary, pricing a brand-new market — closeness — with no data except what the household provides.
In some households, the early claims paid out: tears got comfort, needs got met, reaching out worked. That actuary prices closeness as affordable and writes generous policies for life. But in a household where vulnerability led to pain — where the reached-out hand met anger, or mockery, or air; where needing things cost love instead of earning it — the actuary does exactly what any honest analyst would do with that dataset: prices closeness as catastrophically risky, and sets the premium accordingly. Distance is safer isn’t a mood. It’s an underwriting decision, made young, from real losses.
The trouble is what happens next: the policy never gets re-underwritten. Childhood conclusions don’t stay filed under “that house, those people” — they generalize into how the world works. So the adult walks through decades of genuinely safe rooms, full of people who would gladly have helped her move, still paying a premium priced from one disaster year in 1993. The market changed completely. The actuary never went back to check — because the whole point of the policy is that you never run the experiment that would update it.
The cruel twist: the policy causes the crash
Here’s where the research delivers its most uncomfortable finding, and it deserves to be read slowly.
You’d think the self-protectors, whatever they lose in closeness, at least gain what they’re paying for: safety from rejection. The data says otherwise. As researchers testing the model put it, ironically, people who prioritize self-protection goals end up more vulnerable to eventual rejection than people who prioritize connection.
The mechanism is visible the moment you look for it. Distance doesn’t read as self-protection from the outside — it reads as disinterest. The declined help, the unmentioned surgery, the friendship held permanently at arm’s length: each one tells the people who like her that the liking isn’t wanted. So, reasonably, they stop offering it. Invitations thin. The warm acquaintances stay acquaintances, then drift. And the system, watching people pull away, logs it all as confirmation: see — they leave. Good thing we never got attached. The insurance policy doesn’t just fail to prevent the crash. Over enough years, it manufactures it — the one form of rejection the premium was supposed to make impossible, delivered slowly, in installments, with no one to blame.
That’s the real cost of “distance is safer.” Not loneliness as bad luck. Loneliness as the policy working exactly as written.
Re-underwriting is allowed
Now the part the eight-year-old actuary could never have known: the system updates. Not through insight — you can’t talk a risk model out of its priors — but through evidence, which is precisely what the policy has spent decades preventing anyone from collecting.
So the repair is claims-testing, in the smallest denominations that count. One real ask, aimed at the safest person on the roster — the ride, the favor, the actual answer to “how are you.” Small enough to survive if it goes badly; genuine enough to register when it doesn’t. Each safe landing is a data point the old model can’t ignore forever, and attachment researchers have a name for where this road leads: earned security — adults who started with every reason for the high-premium life and recalibrated anyway, ending up, by the measures that matter, indistinguishable from people who were secure all along. The setting is a history, not a sentence.
And if you love someone who lives behind the premium: your job isn’t to storm the distance or take the “I’m fine” personally. It’s to be boringly, repeatedly consistent — and when the first small bid for help finally comes, to treat it like the underwriting review it secretly is.
Because the woman with the movers was never avoiding people.
She was honoring a policy written by a child who was doing her best with terrible data — and the kindest thing anyone can do, including her, is help the actuary see the current market.
It’s been safe out here for years.