This leaves open the possibility that decision effects previously attributed to acute emotion may be a consequence of acute ERR strategies such as cognitive reappraisal and expressive suppression. In Study 1, we manipulated ERR of laboratory-induced fear and disgust, and found that the cognitive reappraisal of these negative emotions promotes risky decisions (reduces risk aversion) in the Balloon Analogue Risk Task and is associated with increased performance in the presence/hunch period of the Iowa Gambling Task.
In Study 2, we found that naturally occurring negative motions also increase risk aversion in Balloon Analogue Risk Task, but the incidental use of cognitive reappraisal of emotions impedes this effect. We offer evidence that the increased effectiveness of cognitive reappraisal in reducing the experience of emotions underlies its beneficial effects on decision making.
Keywords: emotion regulation, cognitive reappraisal, expressive suppression, decision making, risk aversion we find that the decision effects of emotion vary according to the way in which a person regulates the emotion experience. Various theoretical approaches have indicated that, contrary to traditional hinging in psychology and economics (Enemies, 1 967; Simon, 1956), emotions play an active role in some forms of decision making.
Regardless of whether they have been assimilated to the “goodness” or “badness” of alternatives for action (Slavic et al. , 2007), attributed to activation in basic appetitive or defensive motivational systems (Bradley & Lang, 2007; Leninist & Donahue, 2004), or reduced to somatic markers associated with current or past behavioral outcomes (Beachwear, Damasks, & Damasks, 2000), emotions have been consistently shown to influence decision making.
In the emerging neuroscience literature, brain lesion, functional nonrecurring, and neurophysiology studies in animal models and humans have begun to shed light on the neural foundation of emotion and decision (Coercible, Dolan, & Siring, 2007; Doherty & Boasters, 2008; Ranger, Camera, & Montague, 2008; Seymour & Dolan, 2008). These studies suggest that humans can anticipate the emotional impact of potential future decisions using processes that involve the amazedly as well as the abovementioned frontal cortex (Beachwear, Damasks, Damasks, & Lee, 1 999; De Martina, Kumara, Seymour, &
Dolan, 2006; Weller, Levin, Ship, & Beachwear, 2007). This type of anticipation can be adaptive in that emotions such as anxiety or disgust have been shown to impair decision making (Learner, Small, & Leninist, 2004; Preston, Buchanan, Standstill, & Beachwear, 2007), even when physiological responses properly signal disadvantageous alternatives (MIM, Hellman, & Houses, 2008). The intrinsic role of emotion in decision is all the more important as the value of prospects (i. E. , actions with uncertain rewards) is computed in “emotion- cognition brain hubs” (Passes, 2008) economic decision making (see, e. . Ulster, 1 998; Leninist, 2000; Peters, Va- stoat “II, Ga -riling & Slavic, 2006). People evaluate objective features of alternatives such as expected return in a subjective way (Edwards, 1 962; Keenan & Taverns, 1979), and emotions are understood to influence these subjective evaluations (Leninist & Donahue, 2004; Nazi, Ship, & Beachwear, 2006; Slavic, Finance, Peters, & MacGregor, 2007). The recent literature on emotion regulation (ERR), however, highlights that humans typically make efforts to control emotion experiences (Gross, 2002).
This leaves open the possibility that decision effects attributed to acute emotions ay be mediated by ERR strategies. If so, this raises the additional possibility that different regulation strategies could have different decision implications. Only very recently, however, have scholars begun to investigate these possibilities. We report data from both designed and naturally occurring environments, providing convergent evidence that ERR strategies modulate decision making. In particular, Rennet M.
Hellman, Livid G. Sacristan, Marcia Micelle, and Andrei C. MIM, Emotion and Cognition Neuroscience Laboratory, Program of Cognitive Neuroscience, Department of Psychology, Babes. -Boldly University; and Daniel Houses, Interdisciplinary Center for Economic Science and Department of Economics, George Mason University. We thank Carnelian M. Kuhn (Northwestern University) and Sergei p. Pass (Stanford University) and two anonymous reviewers for useful comments on a version of this article.
We are grateful to the International Foundation for Research in Experimental Economics CONCISE for supporting this research. Correspondence concerning this article should be addressed to Andrei C. MIM, 37 Republic, Club-Anaphora, CO 40001 5, Room -Nina. E-mail: [email protected] Org 257 258 HELLMAN, CRISIS. AN, HOUSES, MICELLE, AND MIM such as mandarin deprogramming regions and their targets (i. E. , ventral and dorsal stratum, abovementioned and ventilator frontal cortex, anterior Cingular cortex).
Therefore, impressionists have emphasized that the interaction of emotion and decision making is profitably studied in environments that include risk (where the decision maker has perfect information regarding the stochastic relationship be;en actions and outcomes) and uncertainty (where the decision maker does not have full information about the stochastic environment; see, e. G. , Ranger et al. , 2008). When a human anticipates or experiences an emotion, s/he will often use strategies to control that experience.
It follows that ERR, a concept subsuming the processes controlling which emotions we have, when we have them, and how we experience and express them (Gross, 2002), could be crucial to decision making as well as other cognitive processes (e. G. , memory; Richards & Gross, 1 999, 2000). Although many ERR strategies may be in use (e. G. , Garnering, Legers, Kraal, Van Den Comer, & Teeter, 2002), most recent research has focused on more commonly used ERR strategies that act either before emotions become activated (interconnectedness ERR) or after emotions arise (response-focused ERR; Gross & Thompson, 2007).
The process model that distinguishes ERR strategies according to when they act in the emotion generative process (Gross, 1 Bibb, 2002) has been exceptionally useful to cognitive scientists and generated pioneering discoveries on the effects of ERR on cognition, physiology, and health (for recent reviews, see Canal, Ferrier, & Adman, 2009; cross, 2008; Phillips, L adductor, & Drivers, 2008). Two specific ERR strategies have been extensively investigated over the past decade (Schooners & Gross, 2005, 2008).
One of these, cognitive reappraisal, is an antecedent-focused ERR strategy that alters the trajectory of emotional responses by reformulating the meaning of the situation. The other strategy, expressive suppression, is a response-focused strategy that involves inhibiting behaviors (e. G. , facial expressions, verbal utterances, gestures) associated with emotional responding (Gross, 2002; Gross & Thompson, 2007). Whereas both reappraisal and suppression decrease the expression of emotions, their effectiveness in decreasing the experience of emotion differs due to their timing with respect to the emotion generative process.
In particular, reappraisal and suppression effectively decrease the experience of positive emotions, but only the former is as effective in reducing the experience of negative emotions (Gross, 1 AAA; Gross & Eleven’s, 1997). In addition, reappraisal diminishes emotion at an early stage and without the need for sustained effort over time, whereas suppression involves active efforts to inhibit proponent emotional responses (Gross, 2002; Maureen, Twice, & Bandmaster, 1998).
By way of increased effort or “ego depletion” (Bandmaster, 2003), suppression impairs explicit memory (Richards & Gross, 1999, 2000). Similarly, suppression of negative emotions (e. G. , anger, embarrassment, but not sadness) accompanied by high arousal is associated with impulsive decision making (Elite & Bandmaster, 1996). Recent research suggests that reappraisal and suppression might affect decision making differently. For instance, habitual use of cognitive reappraisal accounts for over half of the variance in the profit of simulated economic negotiations (Yourselves, 2004, 2008).
Similarly, individual differences in cognitive reappraisal and expressive suppression predict health risk behaviors (Magma, Phillips, & Hosier, 2008). The effects of reappraisal and suppression on decision making, however, have not previously been directly compared. We suggest that there may be TV key routes by which reappraisal and suppression might generate different consequences for decision making. One, an “emotional” route, stems from differences between reappraisal and suppression in their effectiveness in mitigating the experience of negative and positive emotions (Gross, 2002).
The second, a “emotional” route, stems from differences in the level of effort (cognitive load) required to implement reappraisal or oppression, which could perhaps be related to differences between their respective contributions to ego depletion (Bandmaster, 2003; Richards & Gross, 1999). Regarding the emotional route, recent studies offer evidence that the acute use Of reappraisal effectively decreases physiological arousal related to the anticipation of reward (Delegated, Gillis, & Phelps, 2008) and loss aversion (Solo;Hester et al. , 2009).
The emotional route is emphasized in studies of self-regulation, including evidence that the expressive suppression of emotion modulates risk taking through ego depletion (Elite & Bandmaster, 996; Maureen et al. , 1998). In this research, we manipulated ERR strategies under controlled conditions to provide rigorous evidence on their decision-making effects. We focused on instructed and incidental reappraisal and suppression used to regulate negative emotions induced by movies (Study 1) and naturally occurring negative and positive emotions (Study 2).
We evaluated the effects of these ERR strategies on decision making under both risk and uncertainty. We studied decisions in the Balloon Analogue Risk Task (BART; Lezzy et 2002) as well as the Iowa Gambling Task (CIT; Beachwear, Darwinian, Darwinian, & Anderson, 1994). BART is a computer-based measure of risk taking in which participants can earn financial rewards by pumping balloons presented on a screen; different balloons have variable explosion points, and once a balloon explodes, the money deposited for pumping that balloon is lost (Lezzy et al. 2002). The other economic game, CIT, simulates real-life decision making in the Way it factors uncertainty of premises and outcomes as well as reward and punishment (Beachwear et al. , 1994). It measures the degree to which individuals come to choose small immediate gains (associated in the long ERM with smaller losses) over large immediate gains (associated in the long term with larger losses; Beachwear et al. , 1994).
We next report (a) whether manipulations of ERR of two negative emotions induced by films influence decision-making performance, and (b) whether the incidental use of ERR strategies on naturally occurring positive and negative emotions affects decisions. We found statistically identical decision patterns between those who use suppression and a control group using no ERR strategy. In contrast, in relation to both the control and suppression groups, participants using reappraisal displayed systematically and statistically efferent decision patterns.
In particular, we found that reappraisal promotes increased risk taking (or equivalently, reduces risk aversion), and we traced the source of this effect to the emotional route. Study 1 This study was designed to investigate the effects of ERR on decision making in laboratory conditions that involved the controlled induction of two negative emotions. Fear and disgust were 259 EMOTION REGULATION AND DECISION MAKING induced by movies, and the effects of suppressing or reappraising these emotions were tested in CIT and BART.
We chose to study fear and disgust for three reasons. First, they are both negative emotions that, according to the previous literature on ERR (e. G. , Gross & Eleven’s, 1 997), should be effectively downhearted by reappraisal, but not suppression. Therefore, this study allowed us to test the emotional route by which ERR may influence decision making. Second, little is known about the way different emotions of the same valence differentially influence judgment and choices (Learner & Kilter, 2000).
This is relevant because although both fear and disgust involve unpleasant states, they differ in important dimensions including certainty and attention activity (Smith & Ellsworth, 1985). In contrast to disgust, which involves high levels of certainty and is associated with a strong unwillingness to attend to the situation, fear involves maximal uncertainty reflected in the apparent indecisiveness regarding whether to attend to the situation. In addition, disgust may have evolved to ward off contamination, and this might increase its effects on risk attitudes (Fester, Pillowslip, & Flagman, 2004).
Third, fear and disgust are among the emotions that can be reliably induced using movies in the laboratory (Kiering, Wilhelm, Roth, & Gross, 2007; Rottener, Ray, & Gross, 2007). In summary, laboratory-induced fear and disgust can plausibly inform the effects of ERR on decision making in controlled conditions. In light of the known differences in their ability to downhearted negative emotions, we hypothesized that participants using reappraisal would make riskier decisions than those using suppression. Method Sixty participants (56 women; mean age D 21. 5 years) from the Babies-Boldly University campus volunteered for this study. They were randomly distributed in six groups defined by the induced emotion (i. E. , fear or disgust) and ERR strategy (i. E. , cognitive appraisal, expressive suppression, or control/no ERR instructions). Immediately prior to the experiment, participants completed the Positive and Negative Affect scales of the Positive and Negative Affect Schedule-?Expanded Form (PANS-X; Watson & Clark, 1999) to control for their dispositional mood.
They also completed the Specific Affect scales of the same questionnaire as a pretest measure of emotion. Then, according to the experimental condition, the participants viewed one of two short (i. E. , approximately 2 min) movie excerpts (either Gothic, 2003, or Pink Flamingos, 1972), both of which reliably elicit fear or august (Rottener et al. , 2007). Participants viewed the movies with standard instructions to reappraise, suppress their emotions, or in the absence Of instructions related to ERR (see also Richards & Gross, 2000).
Immediately after the movie, they completed the Specific Affects scales of PANS-X (posters), enabling us to measure the induced emotion. They also completed the Emotion Regulation Questionnaire (ARQ; Gross & John, 2003) to confirm that participants employed the instructed ERR strategy. Next, participants played computer versions of BART and CIT. The average number of pumps per unexploded balloon in BART ND the difference between advantageous (i. E. , C and D) and disadvantageous (i. E. , A and B) selections in CIT were the dependent variables, as described in the standard protocols (Beachwear et al. 1994; Lezzy et al. , 2002). In addition, CD – ABA scores were calculated for each 20-selection block of CIT, in light of the observation that these blocks reflect stages of decision-making optimization, which can be differentially influenced by emotions (see Beachwear, Damasks, Travel, & Damasks, 1997). The order of BART and CIT was counterbalanced between the two groups. Rest Its Manipulation checks. The Positive and Negative Affect scores indicated that immediately before the movies, there were no significant differences in dispositional mood be;en the conditions.
The movies succeeded in specifically inducing fear and disgust, as indicated by the significant increases in the corresponding specific affect scores in the posters over all the groups (see Table 1). ARQ scores confirmed that the participants successfully used the strategy that they were instructed to use during the movies (see Table 1). Emotion experience. The type of ERR that participants used during the movie affected the emotion they experienced. In comparison to the control group, reappraisals, but not suppressors, displayed significantly decreased fear and disgust (see Table 1 Decision-making performance.
A 3 (ERR: reappraisal vs.. Suppression vs.. Control) C 2 (emotion: fear vs.. Disgust) analysis of variance (NOVA) indicated a significant main effect of ERR on BART performance, F(2, 57) 22. 69, p . 01, 02 CLC . 8 The effects of Emotion or ERR O Emotion on BART were not statistically significant. Post hoc analyses showed that reappraisals had significantly higher mean pumps per unexploded balloons in BART than suppressors and controls (see Figure IA). Similar statistical analyses on the total CD – ABA scores indicated no significant effects of ERR or emotion on CIT performance.
However, by including the block of selections as an additional factor in the NOVA we found a significant effect of ERR, 82, 57) 19. 8, p . 05, 02 -4, and block, F(4, 55) n 9. 5, p O . 01, 0 . 3, on CIT performance. Post hoc analyses indicated that reappraisals had significantly higher scores in the third block Of trials compared with suppressors and controls (see Figure AAA). Study 2 This experiment was designed to replicate the effects of ERR on decision aging in conditions that involved naturally occurring emotions and extend these effects to positive emotions.
It is widely accepted that naturally occurring emotions are more salient and valence than those induced in the laboratory, and their influences on cognition may be more conspicuous (for various perspectives, see Bradley & Lang, 2007; Damasks, 2005; Afraid, 1988). Fortunately, intense emotions such as those triggered by having just learned the result of a final exam can be measured and studied (see also Grammar, Willard, & Mended, 2008). We investigated the effects of incidental ERR on decision making in this natural context.