An Integrative Model for Affective Forecasts: Understanding Predictions about Future Feelings



We’re presenting a short series of abstracts of the work-in-progress our scholars will present and discuss at their June 2017 Working Group Meeting.  Heather C. Lench is Associate Professor and Associate Head in the Department of Psychology at Texas A&M University and scholar with Virtue, Happiness, & the Meaning of Life.


People try to make decisions that will improve their lives and make them happy, and to do so, they rely on affective forecasts–predictions about how future outcomes will make them feel. The greater the emotional impact people expect a future outcome to have, the more effort and resources they invest in attaining or avoiding it. Understandably then, inaccuracy in affective forecasting has been identified as a major obstacle to making good decisions. Decades of research suggest that people are poor at predicting how they will feel and commonly overestimate the impact that future events will have on their emotions. Although the simplicity of this idea is intuitively attractive, recent studies have revealed that people are actually very good at forecasting some features of their emotional reaction. This investigation tested a new theoretical model that explains past inconsistent results demonstrating that sometimes people overestimate, sometimes underestimate, and are sometimes accurate in their forecasts. The investigation clearly differentiates forecasts of emotional intensity, frequency, and duration for the first time in the real-world setting of a controversial presidential election. Participants accurately forecast the intensity of their reaction, but overestimated how frequently they would feel emotions about the election and how much their mood would be impacted by the election. Consistent with our theoretical model, bias in forecasts of emotion were predicted by cognitive features. Overestimating the importance of the election resulted in overestimating the intensity of responses; overestimating the frequency of thinking about the election resulted in overestimating the frequency of responses; and overestimating the relevance of the election to personal goals predicted overestimating the impact of the election on mood. By allowing researchers to achieve greater precision about the features of emotion being predicted, this study clarifies when and why people overestimate, underestimate, and accurately predict their emotional reactions. Addressing this question is essential, not only for a theoretical understanding of how people think about their futures, but also for understanding how to intervene to improve decisions.


The results inform interventions designed to improve decision-making in applied domains including health, public policy, education, and economics. People making important decisions–such as whether to undergo surgery, listen to public health warnings, or pursue a specific career– will be better informed if they can accurately predict how the outcomes of their decisions will make them feel. Thus, interventions that improve forecasting are critically important for helping people make informed choices with implications for the length and quality of their lives.
*This is a collaborative project with Linda J. Levine, and is funded by the National Science Foundation (#1451297)

*A similar abstract was submitted for the December 2016 meeting; however, discussion of these primary findings was delayed in favor of presenting several serendipitous results given the surprising outcome of the election.