Regret Without the Replay: What It's For, How It Differs From Guilt, and How to Stop Reliving It

The notification lands at 9:14 on a Tuesday: the colleague who took the transfer you turned down three years ago has been promoted again. Before you finish reading her name your mind has rebuilt the whole decision, the salary call, the lease, the exact sentence you used to decline, and twenty minutes later you have added a chapter where you said yes.

Regret has a second shape. It is 11:40pm and you are replaying something you did say, to your brother, at a dinner in 2019. He laughed it off. You never have.

Both are regret, and both are useful for about a minute. Regret is a counterfactual signal, a flag that a different choice would have turned out better, and it earns its keep only until you have read it; what hurts for years is rarely the regret but the replay.

What is the purpose of regret?

Regret exists to improve your next decision. Zeelenberg and Pieters, in their 2007 theory of regret regulation, define it as a comparison-based emotion of self-blame: you realize or imagine that your situation would be better had you decided differently. It is aversive, and people are motivated to get rid of it in two ways: fix the outcome now, or learn for later.

Roese put it plainly in a 2010 Kellogg Insight article: regret "is like a flag going up," and the signal is supposed to switch off when nothing can be done. The flag itself is not the damage. It marks where the next decision can go differently.

The catch is one line from Epstude and Roese's 2008 functional theory of counterfactual thinking: "perceived opportunity appears to be the master moderator." When you can still act, regret turns into behavior change. When you can't, the signal has nothing to attach to, and it can start to loop.

How is regret different from guilt, shame and rumination?

Guilt is about harm you did to someone. Regret is about a choice that turned out worse than the alternative. Zeelenberg and Breugelmans showed in 2008 that guilt arises mainly from harm to other people, while regret arises from harm to others or to yourself. The transfer hurt nobody but you; the remark at dinner hurt your brother, which is partly why it has outlasted the job.

Regret, guilt, shame: three different targets

Regret judges a decision. Guilt judges a behavior and its effect on someone else. Shame judges you: in Tangney and Dearing's 2002 distinction, guilt says, roughly, that I did a bad thing, while shame says that I am bad. Regret and guilt have exits: a different choice next time, a repaired harm. Shame has none, which is why a regret that has slid into "I'm the kind of person who always chooses wrong" has stopped being a signal and become a sentence.

Anger at someone else's choice rather than yours isn't regret either; it's closer to resentment.

Regret vs rumination

Rumination is not an emotion but a thinking style: repetitive, passive, abstract and self-focused. Nolen-Hoeksema, Wisco and Lyubomirsky's 2008 review found that it makes depression worse, impairs problem-solving, erodes social support, and predicts the onset of depression more than its duration.

The evidence that the two are separable, and that the combination hurts most, is a nationally representative survey Roese and colleagues published in 2009. Regret and repetitive thought were each linked to general distress, but only regret was tied to anhedonic depression (the flat, nothing-feels-good kind) and anxious arousal, and their interaction was "highly predictive of general distress." Epstude and Roese supply the mechanism: self-blaming "if only" thoughts raise negative affect without improving behavior.

Is it better to regret something you did or didn't do?

Neither, and the famous answer has weakened. The pattern most people have heard comes from Gilovich and Medvec's 1995 review: in the short term people regret their actions more, and over the long run their inactions more.

The long-term half rested on a striking number: in the original 1994 study, 84% of the long-term regrets people described were inactions. In 2023, Richardson and Gilovich re-ran the question with 2,600 museum visitors.

Source: Richardson and Gilovich (2023), Royal Society Open Science.

The short-term half held; the 1994 paper reported the direction but, in our sources, not the split. The long-term half did not hold: lifetime regrets split 51 to 49. The authors called it "a partial replication."

Two caveats. First, the pattern reverses after a failure: Zeelenberg and colleagues (2002) found that when a previous decision had already gone badly, people attributed more regret to the person who then did nothing than to the one who acted. Second, which inactions linger. Across six studies, Davidai and Gilovich (2018) found that 72% of listed regrets concerned the ideal self (who you wanted to be) rather than the ought self (who duty said you should be), and 76% named an ideal-self failure as their biggest regret. "The failure to be your ideal self," Gilovich said in Cornell's announcement, "is usually an inaction."

What people actually regret

Opportunity triggers regret. A pooled analysis of 11 data sets by Roese and Summerville (2005) covered 3,041 people and 4,054 regrets: education 32.2%, career 22.3%, romance 14.8%, parenting 10.2%, the self 5.5%, leisure 2.6%. Their explanation: "opportunity breeds regret."

The biggest recent dataset is Daniel Pink's, gathered for his 2022 book The Power of Regret, including a World Regret Survey of more than 23,000 regrets from 109 countries that was not a random sample. More than 8 in 10 of his respondents said they feel regret at least occasionally. In the world survey, inaction regrets were roughly twice as common as action regrets and connection was the largest category, though volunteered regrets are a different measurement from a structured question put to museum visitors.

The moment regret turns into replay

Regret has done its job when you can say what you would do differently; replay is running the tape after that point, or without ever reaching it. Four signs:

1. The "if only" has stopped producing anything new. The fortieth run of the transfer decision contains no information the third one didn't.
2. It has moved from the decision to you. "I should have asked about the relocation package" is a regret; "why do I always sabotage myself" is a verdict, the self-blaming counterfactual Epstude and Roese found raises distress without changing behavior.
3. Nothing can be done, and the flag is still up. The signal has outlived its purpose.
4. It happens at bedtime. In 101 people, Schmidt and Van der Linden (2009) found that bedtime counterfactual thinking explained the link between a tendency to act rashly and insomnia symptoms.

If most of these are familiar, the problem is no longer the regret but the rumination that has adopted it, and that needs addressing more than the decision does. If the loop runs through most of your day, the overthinking page covers the wider pattern.

How do I stop living with regret?

You stop by doing two things in order: read the regret for its lesson, then put the rest down. Epstude and Roese describe two routes from "if only" to behavior: content-specific, where the causal lesson becomes an intention, and content-neutral, through mood, motivation and a sense of control. When action is still possible, people shift toward regulating behavior; when it isn't, toward regulating the feeling. Doing the second while the door is open is avoidance; doing the first after it has shut is replay.

Read the regret: turn the lesson into an if-then plan

1. Write the counterfactual as one sentence. "If I had X, then Y." Not "I should have been braver." A regret you cannot write this way is probably shame wearing regret's clothes.
2. Check that the lesson is a behavior, not a trait. "Ask for the full package before declining" is usable; "be less of a coward" is not. Mr. Relentless would put it bluntly: a hundred replays and no behavior you would change is not learning, it is sentencing. If the regret is about something you tried that failed, our post on getting the lesson out of a failure walks through a short review built for this step.
3. Write the if-then plan. "If I'm offered a role and my first instinct is to decline, then I ask for the full package in writing before I answer." That is an implementation intention. Gollwitzer and Sheeran's 2006 meta-analysis of 94 tests found it improved goal attainment by d = .65, a medium-to-large standardized difference between groups, not "65% more likely." Our guide to following through covers how to build one.
4. Tell someone who will ask. A lesson in your head is still a counterfactual; one you have told a person who will ask "did you do it?" is a commitment. If the real reason you keep declining is one you haven't admitted yet, start with how to stop making excuses.

Put it down: self-distancing, self-compassion and writing (not at bedtime)

The lesson does not switch off the feeling, especially where the door is shut. Three approaches have evidence, each with fine print.

Self-distancing. Kross and Ayduk distinguish self-immersed reflection (reliving it through your own eyes) from self-distanced reflection (watching from across the room). In their 2011 review, distanced reflection moved people from recounting what happened to reconstruing what it meant and lowered distress both immediately and a week later, with less rumination. The cheapest version is a pronoun. In seven studies with 585 participants, Kross and colleagues (2014) found that people who talked themselves through a stressful event by name or as "you" rather than "I" showed more distance, less distress and less rumination afterwards. Nearly all of this comes from one lab, so treat it as promising rather than settled.

Self-compassion. One series of three studies by Zhang and Chen (2016) found that people who met their regrets with self-compassion reported more acceptance, forgiveness and personal improvement, with acceptance as the route to improvement. Compassion did not let them off the hook; it made change more likely.

Writing, with two warnings. Expressive writing's average effect is small: Frattaroli's 2006 meta-analysis of 146 randomized studies found an average r of .075, with better results from three or more sessions of at least 15 minutes and for events not yet fully processed; among healthy college students, six studies found a benefit and 32 did not. In a randomized study of 176 people, Schmidt and Van der Linden (2013) found that writing about one's most burdensome regret before sleep delayed falling asleep in habitually regretful people. Write in daylight, with a stop time, about regrets you have not processed, not nightly about one you have.

When the door is closed

Some regrets have no next time: the parent is gone, the window has passed, the business failed and you are 58. The behavior route is closed; the task is regulating the feeling.

Age changes what works. In 122 adults aged 20 to 87, Wrosch and Heckhausen (2002) found that believing a regret had been within your control went with lower regret in the young and higher regret in the old: "that was up to me" helps when you can use the lesson and corrodes when you cannot. Wrosch, Bauer and Scheier (2005) found regret intensity predicted lower quality of life only in older adults, and that disengaging from the lost goal and having other goals available were both protective.

So the work is to name what the regret says you valued, find where that value can still go, and stop re-litigating a decision whose evidence is all in. Processing a difficult emotion rather than arguing with it is the skill underneath.

Some regret needs more than a technique. If it comes with weeks of flat mood, loss of interest in what you used to enjoy (the anhedonic pattern Roese's 2009 survey tied to regret), trouble functioning, or a sense that nothing you do matters, see a doctor or therapist; hopelessness attached to regret is a clinical sign, not a character assessment. If the replay ever turns into thoughts of not wanting to be here, call or text 988 in the US, 9-8-8 in Canada, or find your country's line at findahelpline.com.

Frequently asked questions

Is it better to regret something you did or didn't do? Neither. In Gilovich and Medvec's work, actions were regretted more in the short term and inactions more over a lifetime, but a 2023 replication with 2,600 people found lifetime regrets split 51% to 49%. After a prior failure, people regret doing nothing more than acting.

What is the difference between regret and guilt? Guilt is mainly about harm you caused another person; regret is about a decision that turned out worse than the alternative, whether it hurt someone else or only you. Zeelenberg and Breugelmans (2008) found this "who was harmed" line separates the two.

Can regret cause depression? The evidence is correlational, so "cause" is too strong. Roese and colleagues (2009) found regret linked to anhedonic depression, with regret plus repetitive thinking the strongest predictor of general distress. Weeks of low mood alongside a regret are a reason to see a clinician, not to think harder.

What are the 4 types of regret? They are Daniel Pink's, from the surveys behind his 2022 book The Power of Regret: foundation, boldness, moral and connection. Connection was the largest group and moral the smallest, at about 10%. It sorts volunteered regrets; it is not a scientific classification, and his larger survey was not a random sample.

Sources and further reading

• Zeelenberg and Pieters (2007). "A theory of regret regulation 1.0." Journal of Consumer Psychology
• Kellogg Insight (2010). "Learning to use regret." Kellogg School of Management, Northwestern University
• Epstude and Roese (2008). "The functional theory of counterfactual thinking." Personality and Social Psychology Review
• Zeelenberg and Breugelmans (2008). "The role of interpersonal harm in distinguishing regret from guilt." Emotion
• Tangney and Dearing (2002). Shame and Guilt. Guilford Press (book).
• Nolen-Hoeksema, Wisco and Lyubomirsky (2008). "Rethinking rumination." Perspectives on Psychological Science
• Roese et al. (2009). "Repetitive regret, depression, and anxiety: findings from a nationally representative survey." Journal of Social and Clinical Psychology
• Gilovich and Medvec (1995). "The experience of regret: what, when, and why." Psychological Review
• Richardson and Gilovich (2023). "A very public replication of the temporal pattern to people's regrets." Royal Society Open Science
• Zeelenberg, van den Bos, van Dijk and Pieters (2002). "The inaction effect in the psychology of regret." Journal of Personality and Social Psychology
• Davidai and Gilovich (2018). "The ideal road not taken." Emotion
• Cornell Chronicle (2018). "Woulda, coulda, shoulda: the haunting regret of failing our ideal selves." Cornell University
• Roese and Summerville (2005). "What we regret most... and why." Personality and Social Psychology Bulletin
• Behavioral Scientist (2022). "What is the power of regret? A conversation with Daniel Pink."
• Psychology Today (2022). "The 4 major kinds of regret." (reporting Pink's survey)
• Schmidt and Van der Linden (2009). "The aftermath of rash action: sleep-interfering counterfactual thoughts and emotions." Emotion
• Schmidt and Van der Linden (2013). "Feeling too regretful to fall asleep: experimental activation of regret delays sleep onset." Cognitive Therapy and Research
• Gollwitzer and Sheeran (2006). "Implementation intentions and goal achievement: a meta-analysis of effects and processes." Advances in Experimental Social Psychology
• Kross and Ayduk (2011). "Making meaning out of negative experiences by self-distancing." Current Directions in Psychological Science
• Kross et al. (2014). "Self-talk as a regulatory mechanism: how you do it matters." Journal of Personality and Social Psychology
• Zhang and Chen (2016). "Self-compassion promotes personal improvement from regret experiences via acceptance." Personality and Social Psychology Bulletin
• Frattaroli (2006). "Experimental disclosure and its moderators: a meta-analysis." Psychological Bulletin
• Wrosch and Heckhausen (2002). "Perceived control of life regrets: good for young and bad for old." Psychology and Aging
• Wrosch, Bauer and Scheier (2005). "Regret and quality of life across the adult life span." Psychology and Aging

The step most people skip is the one in the middle: turning "if I had asked for the package" into a plan for the next offer. That is what an ILTY conversation is built around. You talk the regret through with a companion that pushes back when the lesson is really a verdict on your character, you leave with two to four concrete action items, and the next time you open the app the companion asks whether you did them. ILTY is not therapy, and a regret that arrives with weeks of low mood belongs with a clinician. Start the 7-day trial.