You think of an old colleague who helped you once. You open a message, type “thank you,” and close it again. It feels too late, too small, a little strange. Five studies from the last few years tested that kind of hesitation. Four point the same way: the person on the other end feels better about it than the sender expects. The fifth shows the forecast can change with practice.
This post walks through the five studies, one by one. Then it turns them into one thing to do tonight, with or without a deck of cards.
The short answer
Researchers call this a miscalibration. Before a kind act, people predict how the other person will feel. Afterward, the other person reports how they felt. In the first four studies below, the prediction came out lower than the report. Senders expected less warmth, less surprise, less interest, and more awkwardness than there was.
The gap matters because the prediction decides what happens next. In several of these studies, a low prediction went with less willingness to act. So the thank-you stays in drafts, and the other person never learns it existed.
1. Gratitude letters: more welcome than the writer expects
Amit Kumar and Nicholas Epley asked people in three experiments to write gratitude letters. Writers then predicted how surprised, happy, and awkward the recipient would feel. Recipients reported how they felt. Writers underestimated how surprised recipients would be about why they were grateful. They overestimated how awkward recipients would feel, and they underestimated how positive recipients would feel (Kumar and Epley 2018).
Expected awkwardness and expected mood were both linked to how willing people were to express gratitude. A familiar worry is that thank-yous lose meaning when you say them often. The surprise in this study came from the reason for the thanks. That may point to the value of a specific reason, such as “the way you explained the budget to me in my first week.”
2. Small kindness: a cup of hot chocolate goes further
In a later paper, Kumar and Epley looked at random acts of kindness. In a park, people gave away a cup of hot chocolate. In the lab, they gave away a gift. Givers predicted how positive the recipient would feel, and recipients said how they felt. Givers consistently underestimated it (Kumar and Epley 2023).
The authors trace part of the gap to the warmth of the act. Recipients responded to the warmth the gesture conveyed. Givers focused on the act itself, a cup or a small gift, and their expectations barely moved with its warmth. Givers also underestimated how much recipients would pass the kindness on. If the worry is that you can’t repay someone, this finding helps: small and warm counts for more on the receiving end.
3. Reaching out: a short message lands as a surprise
Peggy Liu and colleagues ran a series of preregistered experiments on reaching out: a brief check-in message or a small gift to someone in your social circle. They found a robust underestimation of how much people appreciate being contacted (Liu et al. 2023).
Their explanation centers on surprise. The person who gets the message focuses on the surprise of hearing from you, and that surprise predicts appreciation. The gap grew when the contact was more surprising and when the two people were more socially distant. The old colleague, the cousin you lost touch with, the friend from a former city: those are the cases where the gap is widest.
4. Deep talk: strangers care more than we predict
Michael Kardas, Amit Kumar, and Nicholas Epley paired strangers for conversations. The talks felt less awkward and created more connection and happiness than participants expected. People were especially likely to overestimate how awkward deep conversations would be compared with shallow ones. After having both kinds, they felt more connected to their deep conversation partners (Kardas et al. 2021).
The gap came from expecting others to care less about what they shared than the others did. With close friends, family, or partners, predictions were more accurate, because their interest is clearer. The authors also found that more accurate expectations led people to choose deeper topics. For more on the urge to ask and listen, see how to be more curious about people.
5. Strangers: practice corrects the prediction
The fifth study asks whether the gap can shrink. Gillian Sandstrom, Erica Boothby, and Gus Cooney built a week-long scavenger hunt game in which participants repeatedly found, approached, and talked to strangers. Compared with a control group, players became less pessimistic about rejection and more optimistic about their conversational ability. The effect lasted at least a week after the study ended (Sandstrom et al. 2022).
Daily reports showed expectations growing more positive and more accurate day by day. The fix in this study was repetition. The authors note that earlier attempts to correct these fears had failed. Doing the thing again and again moved the forecast.
Why the gap keeps people quiet
The studies point to a few reasons the message stays unsent. The sender looks at the act itself: a short text, a plain card, a small gift. The receiver feels the warmth and the surprise. The sender pictures an awkward moment that recipients report less of. And the further apart two people have drifted, the more the sender expects a shrug.
These are findings from experiments with specific tasks and samples. They don’t promise that every message lands well. They do show that the usual forecast leans low, and it leans low in the same direction across thank-yous, gifts, check-ins, and conversations.
Three words for this conversation
A card game gives the same kind of practice the scavenger hunt did: repeated small turns of saying something personal and seeing how it lands. In Thesis, each card has an idea on one side and an everyday thing on the other. Three cards fit this topic.
Gratitude has a Fence on its other side. The idea side invites the specific thank-you the 2018 study tested: who, when, and what exactly they did. The Fence side suits people who find thanks hard to say head-on. A fence you built with someone, or one a neighbor fixed, leads to the same story by a side path.
Kindness has a Book on its other side. With the 2023 study in mind, the idea side brings out small acts someone did for you, and ones you did that felt too small to count. The Book side opens a quieter story: a book someone lent, gave, or read aloud.
Friendship has a Ball on its other side. This is the card for the friend you haven’t written to. The idea side invites a story about a friendship as it is now. The Ball side goes back to the playground or the park, and to who was there.
Try it at the table
Playing works with two people or with six. The rules start with “Before you start, agree together how passing works.” Then draw a card, choose a side, and roll the die. Its faces are MEMORIES · BODY · PEOPLE · VALUES · CHILDHOOD · FUTURE. “Look at the word from that side.” Say as much or as little as feels right.
The others ask what they’re curious about, gently, without judging. The rule card offers three follow-ups: What did that feel like? Who else was there? Has it always meant that to you? “Who else was there?” can bring up the person you meant to thank.
The last round fits this topic well. Each player names one thing they discovered about someone at the table today. It’s a small, specific appreciation said out loud, the kind the studies say others value more than we expect.
What to do tonight
Pick one person you’ve thought about this week and haven’t contacted. Write two sentences: one specific thing they did, and what it changed for you. Send it before you rehearse the reply. The research suggests the reply will be warmer than your forecast, and if it isn’t, you’ve still said it.
If you’d rather start at a table, write Gratitude, Kindness, or Friendship on a scrap of paper and take turns. See the rules and the box, or open the free online table. The words have their own pages: Gratitude, Kindness, and Friendship. For a small group of close people, see Thesis for friends. The research behind the deck explains how the words were chosen.