What the data actually shows
A central reason is the norm of reciprocity — the deeply held social expectation that a favour should be returned. Accepting help can create a felt sense of indebtedness, and research on help-seeking has long noted that this anticipated obligation is one of the main psychological costs of receiving support. When you cannot easily repay, accepting can feel worse, not better.
There is also a self-image dimension. Asking for or accepting help can feel like an admission that you cannot manage on your own, and for people who tie their worth to self-reliance, that can register as a threat to competence or autonomy. The discomfort, in other words, is often less about the help itself than about what accepting it seems to say about you.
At the same time, a well-replicated line of research suggests we misread the other side of the exchange. Across studies, people asked to make a request systematically underestimate the likelihood that others will agree to help (the work of Flynn and Bohns is frequently cited here). Separately, research on prosocial behaviour finds that helping reliably tends to make the helper feel good — so the imposition we fear is, for many helpers, closer to a small gift they are glad to give.
Why this feels different from how it actually is
It feels harder than it is partly because we judge ourselves from the inside while imagining how others see us — and we tend to assume they are watching more critically than they are. Research on the so-called beautiful-mess effect suggests that we view our own moments of vulnerability, including asking for help, as weaker and more exposing than observers do, who are more likely to see them as courageous or relatable.
The cost of accepting is also vivid and immediate — the awkwardness, the sense of owing something — while the benefit is spread out and easy to discount. So in the moment, declining feels like the safe, self-protective choice, even when it leaves a real need unmet and quietly distances you from the people who would have stepped in.
And because so much cultural messaging frames independence as a virtue and needing help as a deficit, the instinct to say "I'm fine" can feel less like a personal quirk and more like the correct thing to do. That framing tends to make refusing help feel principled rather than costly.
The imposition we fear is, for many helpers, closer to a small gift they are glad to give.
What the research says to do about it
The most consistent corrective is to recalibrate your prediction of how others will respond. Because people reliably underestimate both the willingness of others to help and how good helping makes them feel, making the request smaller and more concrete in your mind — and reminding yourself that a "yes" is more likely than it feels — tends to lower the barrier to asking.
Reframing the exchange from one-directional debt to mutual connection also helps. Research on relationships suggests that allowing others to help you is part of how closeness is built, not a withdrawal from it — being needed and being trusted with someone's vulnerability is something many people value. Asking can be a gift to the relationship rather than a drain on it.
Where indebtedness is the sticking point, small, ongoing reciprocity over time tends to feel more comfortable than trying to repay every favour immediately. The evidence here is more about general relationship dynamics than a single tidy study, but the pattern is that relationships with a loose, two-way flow of help feel more secure than strictly balanced ledgers.
What the research says does not help
Waiting until you are certain you won't be a burden before asking rarely helps, because the burden is usually overestimated. The research suggests the fear of imposing is largely a misprediction, so deferring the request mostly just leaves the need unmet for longer.
Insisting on full self-reliance as a matter of pride tends to backfire on the relationship side. Refusing all help can read to others as keeping them at arm's length, and it removes the moments of mutual reliance that research links to closeness. Protecting your independence this way can quietly cost connection.
Repaying help instantly and in full, out of discomfort with owing anyone anything, can also undercut the point. It can signal that you experienced the help as a transaction to be cleared rather than a kindness to be received, which is not how warm relationships tend to operate.
Relationships with a loose, two-way flow of help feel more secure than strictly balanced ledgers.
What this looks like in real life
The 'yes' you don't expect
The cost of accepting is vivid and immediate — the awkwardness, the sense of owing something — while the benefit is spread out and easy to discount. So declining feels like the safe, self-protective choice in the moment. But across studies, people asked to make a direct request expected far fewer yeses than they actually received: the willingness of others is usually larger, and the imposition smaller, than it feels.
'I'm fine' as a reflex
Because so much cultural messaging frames independence as a virtue and needing help as a deficit, saying 'I'm fine' can feel principled rather than costly. Yet refusing all help can read to others as keeping them at arm's length, quietly removing the moments of mutual reliance that research links to closeness. Asking can be a gift to the relationship rather than a drain on it.
Real numbers in context
This is an area where the honest answer is qualitative more than numerical: the core findings are about direction and pattern, not precise percentages. The most robust quantitative thread is that people consistently underestimate how likely others are to agree to a direct request for help — across several experiments, those asking expected far fewer yeses than they actually received.
The reciprocity and self-image costs of accepting help are well documented in the help-seeking literature, but they are described as tendencies rather than fixed figures, and they vary a lot by culture, relationship, and the size of the favour. Treat the takeaway as a reliable general pattern — the cost of asking is usually smaller, and the willingness of others usually larger, than it feels — rather than a single statistic.