Thank You — or: I Owe YouWhy the same gift produces two entirely different feelings, and what decides between them

Someone brought you something. Someone stepped in when things got tight. Someone paid before you could reach for your wallet.

There are two ways this can feel. In one, you are moved and you think of the person. In the other, something starts calculating in the background — when can you return it, and with what.

In everyday speech we call both gratitude. In research they are two different feelings, and the difference between them is one of the most stable findings in the whole field.

An experiment that changes exactly one thing

The setup is simple enough to remember. People are told a short story: a friend offers to pay for your textbooks. You badly need them, so you say yes.

Exactly one sentence is varied — what you know about his motives:

• Benevolent: “You can tell that your friend is really concerned about you and wants to help you out.”
• Ulterior: “You know that your friend is doing you this favour in order to borrow your car next weekend.”
• Ambiguous: You say yes. The next weekend that same friend asks if he can borrow your car.

In 2025 this study was repeated with 759 people — preregistered, with openly available data, that is, to the strictest standards this field knows. The result is unusually clear-cut.

Gratitude shifts enormously. On a seven-point scale: 6.54 for benevolent intent, 4.05 for ulterior. In the three-level version: 6.59 — 6.14 — 5.48.

Indebtedness shifts by nothing. 3.91 versus 3.49 — statistically indistinguishable. And in the three-level version all three groups sat at practically the same value, with a test statistic of exactly zero.

So whether your friend means well or wants your car decides how grateful you are — and not at all how much you feel in his debt.

The finding I find most interesting

Across all conditions the two feelings are related — people who feel more grateful also tend to feel more indebted. The association is moderate.

But break it down and something remarkable happens. Where the intention is clearly benevolent, the association vanishes entirely — it sits at practically zero. Where the intention is ulterior, it is strong.

Put differently: when it is clear that someone means well, the two feelings decouple. You can then be deeply grateful without feeling indebted. Only when an ulterior motive is in the room do they fuse.

Indebtedness, then, is not the shadow of gratitude. It is what remains when goodwill is not legible.

What that feeling is actually made of

You might assume it is about an account — a social norm saying that whoever receives must give back. A 2024 study looked more closely, using three methods: a survey of 1,619 people about real experiences, a laboratory game with real money, and a brain-imaging study.

The result: the feeling has two components, and the larger one is not the norm. 91.9 per cent of respondents attributed their sense of indebtedness to having been a burden to the other person. Only 39.2 per cent named the suspicion that the other might want something.

So it is mostly not an IOU. It is a bad conscience about having caused someone trouble.

And one detail from the lab component that runs against intuition: when the helper knows you can reciprocate, indebtedness in the sense of the norm rises — and gratitude falls. Being able to repay does not make things easier. It turns them into a transaction.

What that costs in everyday life

Here it becomes practically relevant. In that same lab study participants could also decline the help offered. And they did so more often when repayment seemed possible: the refusal rate rose from 0.30 to 0.37.

People forgo support they could use in order to avoid the feeling. Interestingly this did not hold when the help cost the other person a great deal — the more effortful it was, the less often it was declined. What deters is not the size of the gesture. It is the foreseeable bill.

A finding from another line of research fits with this, and it gave me the most pause. In one experiment the motivation of the helpers was varied: some helped because they wanted to, others because they felt they had to. Measurements were taken from the recipients.

Those who received help from someone who felt pressured into it were afterwards worse off than people who had received no help at all — in mood, in energy, in self-esteem, and in their sense of connection. Not “less well”. Worse than nothing.

People evidently notice very precisely whether someone helps gladly. And reluctantly given help is not a small kindness — it is a burden.

What gratitude actually responds to

If not the value of the gift — then what? A field study followed real gifts between people and then measured what best predicted gratitude.

1. 01Whether it fits meThe strongest factor was whether the recipient liked the gift — and close behind, whether they felt the other person had thought of them. Not effort as such, but attention in the literal sense: that someone knows who I am.
2. 02The price is almost irrelevantThe actual monetary value was the weakest of all the factors examined — and its influence vanished entirely once perceived thoughtfulness was included in the model. An expensive gift works not because it is expensive, but only insofar as its price is read as care. Where it isn't, nothing of the price remains.
3. 03Voluntariness counts — but only for gratitudeIn a further experiment a colleague helped either of his own accord or on his manager's instruction. The voluntary version produced more gratitude and markedly more desire for closeness. On indebtedness it had no effect.
4. 04Size counts — but mainly for indebtednessConversely: the bigger the favour, the stronger both — but in one study the cost to the giver raised indebtedness only, and gratitude not at all. A large gesture whose intention remains unclear mainly produces pressure.

What the research does not say

Without this section the article would be dishonest — and it is particularly needed here, because almost all the obvious conclusions go beyond the data.

• It does not say that indebtedness damages relationships. That would be the most attractive message and it is the least supported. In the study that tested the two feelings directly against each other, indebtedness was not associated with less desire for closeness — if anything the opposite. What is established is only that people decline help in advance to avoid the feeling.
• It does not say that indebtedness is always unpleasant. The researcher whose theory underpins all of this writes in a footnote that it can occasionally be pleasant and even sought after — namely when it establishes a relationship in the first place. That sentence is almost never quoted alongside.
• It does not say that gratitude makes people more helpful. The best-known behavioural experiment on this was preregistered and repeated in 2020 with twice the sample — it could not be confirmed. Neither gratitude nor indebtedness predicted actual behaviour there.
• It does not say the two feelings exclude each other. They arise from the same event and usually occur together. The difference lies in what triggers them — not in only one of them being possible.
• And it says nothing about us. In this entire field of research there is not a single German-speaking sample. The data come from the US, the Netherlands, Japan, Thailand, China and Spain — and on the topic of obligation in particular, cultural differences are wide.

What this means for saying thank you

The last article was about gratitude working mainly when you express it — and that sentences about the other person (“you really go out of your way”) land, while sentences about your own benefit (“that helped me so much”) do not.

Against this background the difference now makes sense. A sentence about your own benefit describes a service received — it has the form of a receipt. A sentence about the other person describes who someone is. Only the second cannot be settled up.

There is an uncomfortable postscript, though. In the worldwide study of gratitude exercises, indebtedness rose measurably alongside gratitude — and in one study of thank-you letters that was even the strongest single effect of all. Whoever practises gratitude practises both.

That is not an argument against it. It is an argument for expecting the aftertaste rather than treating it as a personal failing.

And it may explain a constellation that often comes up in conversations. In a survey of 2,450 adult children, the support received from parents predicted not gratitude but guilt. About a fifth reported such feelings towards their ageing mother.

You can read that as an accusation. You can also read it as what, according to the data, it probably is: not a lack of gratitude, but the worry of having been a burden — in a relationship whose helping intention was never spoken aloud across decades, because it went without saying.

And that is exactly where something can be done. Not about the volume of what passes between people. But about whether the reason gets said.

This article reflects the state of psychological research and is intended as everyday orientation. The coaching methods described support general wellbeing and holistic balance. They do not constitute a medical diagnosis, promise no cure, and do not replace medical or therapeutic treatment.