Anticipated regret
We choose to avoid the regret we expect to feel.
What it means
Anticipated regret is the influence that the expectation of future regret exerts on present choices, leading people to favor options less likely to produce painful 'what if' comparisons after the fact. The mechanism is that decision-makers mentally simulate how they will feel looking back, and because regret is an aversive, counterfactual emotion, they weight the avoidance of it alongside the objective outcomes — the basis of regret theory, which augments expected-value reasoning with anticipated emotional consequences. This can be adaptive, prompting prudent caution and due diligence, but it can also tip into excessive conservatism, inaction, or omission bias, since errors of action are often anticipated to sting more than equivalent errors of inaction. A nuance is that what people anticipate regretting can diverge from what they actually regret, and the salience and reversibility of an outcome shape how heavily regret is weighted. It matters in consumer choice, medical and financial decisions, and risk-taking, where designing options and information to manage anticipated regret — or simply recognizing its pull — can improve both decisions and satisfaction with them.
Why it happens
Regret needs something to compare an outcome against, so anticipated regret tracks the feedback a decision-maker expects to get. Choosing between two funds whose prices you will both watch invites far more of it than a choice where the road not taken stays dark. Closing off that feedback genuinely lowers the emotion rather than hiding it. Agency is the second ingredient. Regret is the emotion of self-blame, which separates it from disappointment: disappointment measures an outcome against what chance might have delivered, regret measures it against what you should have done. Decisions people feel authorship over therefore generate more of it, which is why defaults, expert advice and delegating the call all defuse anticipated regret: they move the authorship elsewhere, and with it the blame.
What the evidence shows
The largest synthesis, a meta-analysis of 81 studies covering 45,618 people, found anticipated regret correlated .50 with health intentions and .29 with health behaviour, out-predicting perceived severity and worry — and, for intentions, perceived likelihood too. Direction depends on which regret is cued: anticipated regret about failing to act pushes people toward protective behaviour, while anticipated regret about acting pushes them away from it. Most of that evidence (58 percent) was cross-sectional, though the longitudinal subset produced similar associations. The harder problem is what happens when you try to manipulate the emotion at scale. ARTICS randomised 60,000 invitees to Scotland's bowel-screening programme, posting an anticipated-regret questionnaire to one arm; the intention-to-treat analysis found no overall effect on uptake — 57.4 percent against 57.3 percent in the control arm — with a gain only in the subgroup that started with low intentions. The famous temporal pattern — actions regretted in the short run, inactions in the long run — has held up only in part. A 2023 replication with 2,600 museum visitors, against 32 in the original, reproduced short-term action regret but not the long-run reversal, splitting 51 to 49 where the original found 84 percent inaction; a 2022 MTurk replication recovered the pattern in the hypothetical-scenario studies but not the real-life-experience one, which is the same seam.
Limits and caveats
People are worse at forecasting regret than at feeling it. Across near-miss studies, participants overestimated how much more regret they would feel from nearly winning than from clearly losing, and subway riders overpredicted the sting of just missing a train. The reason is that we rationalise faster than we expect to, reattributing the miss to circumstance rather than to ourselves, so the regret never fully arrives. Caution spent to avoid it insures against a feeling that would not have happened. The economic case has its own crack: much of the cyclical choice behaviour that motivated regret theory turned out to be explained by event-splitting, where breaking one outcome into sub-events inflates its weight. Some of the early evidence was a framing artifact, not regret.
Using it in practice
Three moves follow. Discount your own forecasts, especially for near-misses, and ask what you will still care about in a year rather than what stings to imagine now. Sell reversibility rather than reassurance, but be honest about what it does: a generous cancellation or return window moves the regret from before the purchase to after it, keeping the counterfactual live and observable rather than closing it off. That is good for the seller's conversion and not necessarily for the buyer's satisfaction — people are less satisfied with outcomes they can still revise, and fail to anticipate that they will be. And name the regret you are cueing, because the default framing quietly favours inaction: a bad decision you made is vivid, while a good one you declined is invisible and produces no feedback. Where doing nothing is itself a choice, write down the counterfactual you will never observe. That is what stops anticipated regret from silently becoming a preference for standing still.
Examples
Sticking with your original exam answer to avoid the sharper regret of 'changing it to a wrong one.'
People keep playing the same lottery numbers every week, not because the odds improve but because they can vividly picture watching those numbers come up on the week they skipped.
Parents delay a vaccine because harm they caused by acting feels more regrettable than identical harm from doing nothing, anticipated regret feeding straight into omission bias.
A retailer's free 90-day return window sells the shopper an exit from the regret they are simulating at checkout, not a better product.
A hiring panel picks the solid, unremarkable candidate over the unusual one: a visible bad hire they chose is easy to picture regretting, while the brilliant applicant they passed over never reports back.
First described in Regret theory: Loomes & Sugden (1982); Zeelenberg.
Key references
- Richardson, J., & Gilovich, T. (2023). A very public replication of the temporal pattern to people's regrets. Royal Society Open Science, 10(6), 221574. doi.org/10.1098/rsos.221574
- Yeung, S. K., & Feldman, G. (2022). Revisiting the temporal pattern of regret in action versus inaction: Replication of Gilovich and Medvec (1994) with extensions examining responsibility. Collabra: Psychology, 8(1), 37122. doi.org/10.1525/collabra.37122
- Brewer, N. T., DeFrank, J. T., & Gilkey, M. B. (2016). Anticipated regret and health behavior: A meta-analysis. Health Psychology, 35(11), 1264-1275. doi.org/10.1037/hea0000294
- O'Carroll, R. E., Chambers, J. A., Brownlee, L., Libby, G., & Steele, R. J. C. (2015). Anticipated regret to increase uptake of colorectal cancer screening (ARTICS): A randomised controlled trial. Social Science & Medicine, 142, 118-127. doi.org/10.1016/j.socscimed.2015.07.026
- Gilbert, D. T., & Ebert, J. E. J. (2002). Decisions and revisions: The affective forecasting of changeable outcomes. Journal of Personality and Social Psychology, 82(4), 503-514. doi.org/10.1037/0022-3514.82.4.503
- Bleichrodt, H., & Wakker, P. P. (2015). Regret theory: A bold alternative to the alternatives. The Economic Journal, 125(583), 493-532. doi.org/10.1111/ecoj.12200
- Gilbert, D. T., Morewedge, C. K., Risen, J. L., & Wilson, T. D. (2004). Looking forward to looking backward: The misprediction of regret. Psychological Science, 15(5), 346-350. doi.org/10.1111/j.0956-7976.2004.00681.x
- Loomes, G., & Sugden, R. (1982). Regret theory: An alternative theory of rational choice under uncertainty. The Economic Journal, 92(368), 805-824. doi.org/10.2307/2232669