Anticipated emotion
Also known as: Expected emotion
A cold-headed prediction of how a future outcome will make you feel.
What it means
Anticipated emotions are cognitive predictions about the emotional consequences of future outcomes, generated deliberatively and used as inputs to present decisions. Unlike anticipatory emotions, they are not felt in the moment but inferred, which makes them vulnerable to the well-documented errors of affective forecasting such as the impact bias and immune neglect. People routinely choose options to secure anticipated pleasure or avoid anticipated regret and disappointment, the core mechanism in regret and disappointment theories of choice. Because the forecasts are often wrong, decisions built on them can leave people worse off than expected. The distinction between anticipated and anticipatory emotion is a cornerstone of the affective science of decision making.
How the forecast is built
To predict a feeling, you build a mental simulation of the outcome and read your reaction off the picture. The picture is the problem. It contains the focal event and almost nothing else, so it omits the ordinary business of life that will actually occupy most of your attention when the day arrives. It also omits your own resilience: the rationalizing, reframing and adapting that quietly blunt bad outcomes. The forecast is not a faulty memory of feeling but an accurate reading of an impoverished image, which is why, on the standard account, the errors run in a predictable direction, toward emotions more intense and longer-lasting than the ones that actually arrive.
What the evidence shows
Anticipated regret is the best-measured case. Brewer, DeFrank and Gilkey (2016) pooled 81 studies and roughly 45,000 participants, finding anticipated regret correlated .50 with intentions and .29 with behavior, and predicting better than other anticipated negative emotions or people's own risk estimates. Sandberg and Conner (2008) found a similar .47 with intentions across 25 effect sizes, where anticipated regret added about 7% to the variance explained in intentions beyond the theory of planned behavior's existing variables, but only about 1% for behavior, and nothing once past behavior was controlled. So the forecast tracks behavior reliably enough to be worth measuring, though whether it drives behavior is a further question the correlational designs cannot settle. Whether it is accurate is a separate and unsettled question. Levine and colleagues (2012) argued the overestimation is partly a procedural artifact, because people predict their feelings about an event and are later asked how they feel in general. Wilson and Gilbert (2013) replied that the bias survives the correction, and Levine's group answered in the same issue that the impact bias is both dead and alive, so the exchange closed without settling it.
Where it shows up
Whole categories of product are sold on anticipated emotion rather than expected value. Extended warranties and some insurance trade on the imagined regret of the uncovered breakdown, not on the arithmetic. Free returns and trial periods work the other way, stripping out the anticipated regret of a bad purchase before it can block the sale. Vaccination and screening campaigns lean on the same lever. Inside organizations it shapes hiring and procurement: a committee picks the unremarkable candidate or the incumbent vendor because the regret of a visible failure is easy to picture, while the cost of the good option never chosen leaves no trace to imagine.
Limits and caveats
Most of the evidence is self-report gathered in one sitting, where the regret rating and the stated intention come from the same person minutes apart, which inflates the correlation and leaves the causal direction open. Asking the question is also not neutral: prompting people to rate how much they would regret inaction can itself shift the behavior, so the measure doubles as an intervention. The anticipation is asymmetric, too, but not in the direction the classic action effect suggests: people attribute more regret to acting than to failing to act when they judge a described protagonist's immediate reaction, whereas Brewer and colleagues (2016) found anticipated inaction regret rated substantially higher than action regret (d = -1.11), with inaction regret associated with higher intentions and action regret with lower ones. And the forecast keeps its grip on the choice even when the feeling that eventually arrives is milder than promised.
Using it in practice
The useful move is to stop asking people to imagine. Gilbert and colleagues (2009) found, across two experiments (n = 25 and n = 88), that people predicted their reaction to an event more accurately from a single report by someone who had just been through it than from a description of the event itself, and that neither the forecasters nor the observers believed this would work. The samples were modest, but the direction is instructive: prefer a report from someone already in the outcome to your customer's forecast of it. Where you must use forecasts, widen the frame by asking what else will be happening that week, and read stated anticipated regret as a signal of what people will do rather than of how they will feel.
Examples
Choosing the safer job because you predict you would feel crushing regret if a risky venture failed is acting on an anticipated emotion.
You buy the bigger car picturing how pleased the extra space will make you: a prediction, not a feeling, and one that routinely overshoots how long the pleasure lasts.
A student pictures the relief of walking into the exam already prepared, and starts revising in October; the imagined relief, not any present feeling, gets her to the desk.
A surgeon orders one more scan she believes is unnecessary, picturing how she would feel defending the omission if the rare tumor were missed. The imagined regret, not the base rate, drives the order.
An investor holds a falling stock because she pictures the specific sting of watching it rebound the week after she sold. The forecast of that one imagined moment outweighs the fundamentals in front of her.
First described in Distinction drawn by Loewenstein et al. (2001); regret theory, Loomes & Sugden (1982).
Key references
- 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
- Wilson, T. D., & Gilbert, D. T. (2013). The impact bias is alive and well. Journal of Personality and Social Psychology, 105(5), 740-748. doi.org/10.1037/a0032662
- Levine, L. J., Lench, H. C., Kaplan, R. L., & Safer, M. A. (2012). Accuracy and artifact: Reexamining the intensity bias in affective forecasting. Journal of Personality and Social Psychology, 103(4), 584-605. doi.org/10.1037/a0029544
- Gilbert, D. T., Killingsworth, M. A., Eyre, R. N., & Wilson, T. D. (2009). The surprising power of neighborly advice. Science, 323, 1617-1619. doi.org/10.1126/science.1166632
- Sandberg, T., & Conner, M. (2008). Anticipated regret as an additional predictor in the theory of planned behaviour: A meta-analysis. British Journal of Social Psychology, 47(4), 589-606. doi.org/10.1348/014466607X258704
- Loewenstein, G. F., Weber, E. U., Hsee, C. K., & Welch, N. (2001). Risk as feelings. Psychological Bulletin, 127(2), 267-286. doi.org/10.1037/0033-2909.127.2.267