COGNITION AND EMOTION, 2016 VOL. 30, NO. 7, 1304–1316 http://dx.doi.org/10.1080/02699931.2015.1061481 Realistic affective forecasting: The role of personality Michael Hoergera,b, Ben Chapmanb and Paul Dubersteinb aDepartments of Psychology, Psychiatry, and Medicine, Tulane University, New Orleans, LA, USA; bDepartment of Psychiatry, University of Rochester Medical Center, Rochester, NY, USA ABSTRACT ARTICLE HISTORY Affective forecasting often drives decision-making. Although affective forecasting Received 3 September 2014 research has often focused on identifying sources of error at the event level, the Revised 7 May 2015 present investigation draws upon the “realistic paradigm” in seeking to identify Accepted 9 June 2015 fl factors that similarly in uence predicted and actual emotions, explaining their KEYWORDS concordance across individuals. We hypothesised that the personality traits Affective forecasting; neuroticism and extraversion would account for variation in both predicted and Personality; Individual actual emotional reactions to a wide array of stimuli and events (football games, an differences; Decision-making; election, Valentine’s Day, birthdays, happy/sad film clips, and an intrusive interview). Judgement As hypothesised, individuals who were more introverted and neurotic anticipated, correctly, that they would experience relatively more unpleasant emotional reactions, and those who were more extraverted and less neurotic anticipated, correctly, that they would experience relatively more pleasant emotional reactions. Personality explained 30% of the concordance between predicted and actual emotional reactions. Findings suggest three purported personality processes implicated in affective forecasting, highlight the importance of individual- differences research in this domain, and call for more research on realistic affective forecasts. When making decisions, people often engage in affec- understanding of strengths and weaknesses in affec- tive forecasting, the process of predicting how future tive forecasting. events will influence their emotional well-being. This investigation provides an illustrative example of People tend to make bad decisions when their affec- the realistic paradigm in affective forecasting research tive forecasts are steeped in error, and good decisions by examining the extent to which personality dually when their forecasts are realistic (Dunn & Laham, 2006; explains predicted and actual reactions to a range of Hsee & Zhang, 2010; Wilson & Gilbert, 2005). The life events and laboratory stimuli, contributing to their Downloaded by [Tulane University] at 16:18 11 November 2017 social-cognitive error paradigm, popular in this concordance. In particular, theoretical evidence (Gray, domain of research, has focused on identifying 1994;alsoCorr,2004, 2008) and empirical findings factors that differentially influence predicted versus (Canli et al., 2001;Costa&McCrae,1980;Gross, actual emotional reactions (Hoerger, Quirk, Lucas, & Sutton, & Ketelaar, 1998; Hoerger & Quirk, 2010; Telle- Carr, 2009, 2010; see also, Mathieu & Gosling, 2012), gen, 1985; Zelenski & Larsen, 2001) on dispositional thus resulting in discrepancies or error. In contrast, emotionality suggest that individuals who are more the “realistic paradigm” (Funder, 1995) seeks to ident- extraverted and less neurotic tend to experience ify factors that comparably explain both predicted and more pleasant emotional reactions. The present investi- actual emotional reactions, accounting for any degree gation builds on this body of literature by examining of congruence across individuals in terms of who pre- whether neuroticism and extraversion are also associ- dicts and experiences more positive or negative reac- ated with predicted emotional reactions. If so, personal- tions. Identifying factors that account for realistic ity could account for some of the relative match across affective forecasting could help to foster a balanced individuals’ predicted and actual reactions. CONTACT Michael Hoerger [email protected] © 2015 Taylor & Francis COGNITION AND EMOTION 1305 A historical perspective: the error paradigm domains of judgement are provided in Table 1.Ineach as the prevailing zeitgeist domain, error research seeks to identify mechanisms that differentially affect predicted versus actual out- Judgement research has traditionally drawn from two comes. In the affective forecasting domain, for complementary perspectives, the realistic paradigm example, emotional regulation strategies can affect and the error paradigm (for reviews, see Funder, actual emotional reactions considerably but bear little 1987, 1995, 2012). In the first half of the twentieth on emotional predictions, resulting in error (Dillard, century, this research was dominated by the realistic Fagerlin, Dal Cin, Zikmund-Fisher, & Ubel, 2010;Gilbert, paradigm, which emphasises the ways in which judge- Pinel, Wilson, Blumberg, & Wheatley, 1998;Hoerger, ments are “good”,defined as concordant across raters, 2012). Relatedly, an attentional bias called focalism, stable, beneficial, or predictive of later behaviour which leads people to focus on the most salient (Dymond, 1949; Taft, 1955; Vernon, 1933). With the feature of an event, can affect predicted emotional reac- rise of social and cognitive psychology in the 1980s, tions considerably but in some studies has been found the error paradigm gained prominence (Funder, to play less of a role in actual reactions, also resulting 1995, 2012; Swann & Seyle, 2005), emphasising the inerror(Hoergeretal.,2009, 2010;Lench,Safer,& ways in which judgements are “bad”, or discordant Levine, 2011; Wilson, Wheatley, Meyers, Gilbert, & from objective measurements, informant reports, Axsom, 2000). These examples, focused on explaining actuarial data, or optimal reasoning. The prevailing differential variance in predicted versus actual reactions, focus on error is readily apparent in the emerging contrast with research from the realistic paradigm field of research on affective forecasting. For seeking to identify mechanisms that explain substantive example, consistent with the error paradigm, 66% of variance in both predicted and actual ratings simul- articles in two recent meta-analyses (Levine, Lench, taneously. For example, in the context of weather fore- Kaplan, & Safer, 2012; Mathieu & Gosling, 2012) had casting (see Table 1), geographic elevation affects titles that described affective forecasting using expli- regional variation in temperature, and because this infor- citly negative terms (e.g., error, bias, failure, ignorance, mation is widely documented, geographic elevation and emotional innumeracy1); no title described affec- similarly affects weather forecasts, contributing towards tive forecasting positively. them being realistic. Analogously, in the affective fore- Affective forecasting studies have proceeded at casting domain, some factors such as personality that two different levels of abstraction, increasingly shifting shape actual emotional reactions might also underlie from descriptive to mechanistic research. Namely, emotional predictions, contributing to realistic affective descriptive studies have focused on the general level forecasting in terms of the relative positivity or negativity of the emotional event, aiming to identify whether of reactions across individuals. Any particular forecast is forecasts are generally realistic versus error-prone. likely multidetermined by a combination of concor- The central conclusions have been that affective fore- dance-enabling mechanisms that foster realistic fore- casts can be prone to biases towards overestimating casts and discordance-enabling mechanisms that foster or underestimating the emotional intensity of future Downloaded by [Tulane University] at 16:18 11 November 2017 erroneous forecasts. Although recent trends in psychol- events (Wilson & Gilbert, 2013), and simultaneously, ogy have favoured error research, understanding both rank-order concordance is often moderate to high realistic and error mechanisms is needed as these two (rs from .30 to .50; Mathieu & Gosling, 2012), complementary perspectives seek to answer different meaning that people can to some extent realistically questions about the nature of judgement, such as under- gauge whether their emotional reactions will be standing strengths versus weaknesses. more or less intense than the reactions of other indi- viduals. Studies have moved towards the more specific question of what mechanisms—situational Personality and affective forecasting or individual-difference factors—influence predicted accuracy and actual emotional reactions. Examples of mechanistic research from both the The realistic paradigm suggests new avenues for realistic and error paradigms across three common incorporating personality research into studies on 1The exhaustive list of other error-focused terms includes the following: cannot predict, contrary to affective forecasts, do not anticipate, do not learn, focalism, mispredict, misunderstand, more regret than imagined, neglect, unaware, and underestimating. 1306 HOERGER, CHAPMAN AND DUBERSTEIN Table 1. Hypothetical examples of the error paradigm and realistic paradigm in three forecasting contexts. Paradigm Weather forecasting Political forecasting Affective forecasting Error Goal: Identify sources of bias in Goal: Identify sources of bias in political polls Goal: Identify sources of bias in emotional paradigm weather forecasts (forecasts) predictions Examples: Wet bias (overprediction Examples: