Learning from Experience: Do People Adjust Affective Forecasts Following Feedback?
ABSTRACT:
Affective forecasting is the ability to predict how one will feel in response to future events. Prior research has shown that individuals are poor at predicting their emotional reactions to a premeditated situation. Most of these studies focus on why these errors happen; however, far less attention has been paid to whether people correct their emotion predictions after living through the experience. This study examined whether individuals update their emotional predictions after receiving performance feedback, if that process looks different depending on whether the feedback was positive or negative, and whether anxiety or depression symptoms are associated with less accurate emotion updating.
A total of 208 undergraduates (89% White, 80.8% female; mean age 18.94) from the University of New Hampshire completed a timed anagram task (rearranging scrambled letters to create words), purposefully framed as measures of cognitive intelligence, with performance scores supposedly compared to their peers. Before each anagram task, participants predicted how they would feel if they performed better or worse than the majority of their peers on this task across thirteen emotion terms (e.g., sad, disappointed, depressed, upset, anxious, angry, ashamed, frustrated, excited, proud, relieved, happy and neutral) each rated on a 7-point scale. Following the first anagram task, participants received false feedback concerning their performance in the task (e.g., outperformed 97% of others), which were fabricated scores assigned regardless of how participants actually did on the anagram task, intended to elicit emotional reactions. Participants were randomly assigned to receive either positive feedback (outperformed 97% of their peers; n = 102) or negative feedback (outperformed only 26% of their peers; n = 106). They rated their current feelings immediately after receiving this feedback using the same thirteen emotion terms listed above. Participants then made a second set of predictions, about how they would feel if they performed better or worse than their peers on a second anagram task. Participants then completed a second timed anagram task.
Anxiety and depression symptoms were measured using the Generalized Anxiety Disorder Scale-7 (GAD-7; Spitzer et al., 2006) and the Patient Health Questionnaire-8 (PHQ8; Kroenke et al., 2009). The GAD-7 measures anxiety symptom severity across seven items (e.g., "feeling nervous, anxious, or on edge," "being so restless it's hard to sit still"). They are rated by frequency over the prior two weeks by participants on a 0 ("not at all") to 3 ("nearly every day") scale. Meanwhile, the PHQ8 is a questionnaire that measures depression symptom severity. The scale presents eight items (e.g., “Little pleasure or interest in doing things,” “Feeling bad about yourself – or that you are a failure or have let yourself or your family down”). They are rated by frequency over the prior two weeks by participants on a 0 ("not at all") to 3 ("a lot") scale.
Results showed that participants overestimated how strongly they would feel in response to feedback, with this overestimation being larger following positive outcomes than negative ones (t(206) = 4.14, p < .001, d = −.57). For example, before receiving false positive feedback, a participant predicted they would feel very “happy” (e.g., a 6 on a 7-point scale) if they learned they outperformed most other students on the task. However, after receiving that feedback, their actual happiness rating was often lower (e.g., a 3), indicating they had overestimated how good the outcome would make them feel. After experiencing the feedback about how they scored on the task, participants did adjust their second predictions downward; that is, they lowered their predicted emotion ratings for future feedback to more closely match what they had actually felt (e.g., predicting a 5 instead of a 6 for “happy”). This adjustment was particularly pronounced in the positive condition (t(206) = 2.35, p = .020, d = −.33). However, even after adjusting, their revised predictions, the positive feedback condition remained anchored closer to their original expectations than to what they actually felt (t(103) = 4.09, p < .001, d = .14), a pattern that failed to hold in the negative feedback condition (t(105) = 1.42, p = .158, d = .40). Finally, participants reporting higher levels of anxiety or depression showed less accurate prediction updating, with their revised predictions remaining further from their lived emotional experience than those lower in anxiety and depression symptom severity (anxiety: r = .21, p = .003; depression: r = .16, p = .020). These findings suggest people struggle to learn from emotional experience, especially after positive outcomes, and that psychological distress may be associated with lessened tendency to update emotional predictions following lived experience.
I'd like to sincerely thank Dr. Jolie Wormwood and Kristen Petagna for their generous guidance and support throughout this project, and for the invaluable opportunity to grow as a researcher under their mentorship.
Author and Mentor Bios
Melinda Ricardo came to the United States from France as a teenager and has called New Hampshire home for the past eleven years. Her passion for psychology led her to work full-time as a mental health worker at New Hampshire Hospital while completing her bachelor’s degree full-time at the University of New Hampshire, an experience that only deepened her commitment to this field. In fall 2026, she is excited to begin the Counseling Psychology (PsyD) program at Springfield College in Springfield, Massachusetts.
Jolie Wormwood is an associate professor in the Department of Psychology at the University of New Hampshire. She is a social psychologist and peripheral psychophysiologist, and her research broadly concerns affect and emotion.
Copyright 2026 © Melinda Ricardo
