Handbook of emotion regulation ( PDFDrive )

Emotion Regulation Choice 131 To test our predictions, we manipulated alternative neutral reinterpretation depends emotional intensity with emotional images on the original emotional information. To or unpredictable electric stimulation and examine this disparity we showed that when had participants choose between distraction the generation process was simplified, by and reappraisal (Sheppes et al., 2011). The providing participants concrete regulatory results strongly supported our predictions in options for distraction and reappraisal, both emotional contexts. Specifically, par- reappraisal was more frequently chosen. ticipants preferred to reappraise their emo- tional reactions to low-­negative-­intensity The third determinant of emotion regula- pictures and to a threat of low-i­ntensity tion choice involved investigating the influ- electric shocks, but they preferred to dis- ence of motivational goals (Sheppes et al., tract from their emotional reactions to high-­ 2012). An important motivational goal negative-i­ntensity pictures and to a threat of for choosing a regulatory strategy involves high-­intensity electric shocks. evaluating whether an emotional stimulus will be encountered once or multiple times. In a follow-­up study we wanted to exam- While an emotional stimulus that is encoun- ine the robustness of the effect of emotional tered once can be regulated with strategies intensity on regulation choice (Sheppes such as distraction, which provide short term et al., 2012). To that end, we examined relief, emotional stimuli that are encoun- whether individuals would keep their regu- tered multiple times can be better regulated latory preferences under different emotional with strategies such as reappraisal, which intensities even when offered a potent rein- involves engaging with emotional process- forcement (monetary incentive) to engage ing that results in gradual adaptation (e.g., in a counter preference regulatory option. Wilson & Gilbert, 2008; Blechert et al., Although we found that monetary incentives 2012). As we therefore expected and found, influenced regulatory choices in the expected when participants were told that they would direction, the basic preference to reappraise encounter emotional stimuli more than low-­intensity emotional situations and to once, they preferred to reappraise more than distract high-­intensity emotional situations they did when they expected to encounter an remained evident, even when participants emotional stimulus once. were paid high monetary amounts to choose the contrasting strategy. These results sug- Underlying Mechanisms of Emotion gest that emotional intensity strongly influ- Regulation Choice ences regulatory preferences. The previous section established that cen- tral emotional, cognitive, and motivational The second determinant we examined was factors strongly influence individuals’ pref- the cognitive complexity of generating an erences between two regulation strategies emotion regulation strategy (Sheppes et al., that modulate emotional responding at an 2012). Emotion regulation can be viewed as early attentional stage (distraction) or a late involving several sequential cognitive pro- semantic meaning stage (reappraisal). In this cesses that involve generation, implementa- section I turn to the issue of the mechanisms tion, and maintenance (Kalisch, 2009; Och- at the core of emotion regulation choice. sner & Gross, 2008). Generation involves finding an adequate regulatory option that According to our conceptual framework, can replace emotional information pro- emotion regulation choice should involve a cessing. Implementation involves activat- general ability of deliberate executive con- ing a regulatory strategy, and maintenance trol processes to override competing asso- involves holding it in an active state as long ciative emotional processes. In line with cen- as regulation is required. Because generation tral models of self control (e.g., Muraven & operates early in the sequence of an emo- Baumeister, 2000), an ability to recruit cen- tion regulation episode, it is likely to affect tral control processes that can moderate the emotion regulation choice. In distraction, influence of drives and emotions is crucial the generation process is simple, because the for daily functioning. In addition, the actual neutral thoughts that are produced can be regulatory choice process is heavily influ- of any content as long as they are absorbing. enced by the engagement–d­isengagement However, in reappraisal, the generation pro- cess is complex, because the formation of an

132 COGNITIVE APPROACHES dimension where people are weighting their processes originating from the appetitive preference to employ early attentional disen- system (Sheppes et al., 2012). gagement from emotional processing (dis- traction) versus engagement with emotional While these results support the involve- processing prior to late modulation at the ment of deliberate executive control pro- semantic meaning processing stage (reap- cesses in emotion regulation choice, an praisal). important question remains: What are the dimensions that receive central weight in the Provision of supporting evidence for the choice between distraction and reappraisal? involvement of deliberate executive con- One central dimension in my conceptual trol processes that override associative account is engagement–d­isengagement, in emotional processes in emotion regulation which people who prefer to reappraise may choice is important given a potential alterna- wish to engage in emotional processing, tive, more parsimonious account. According and when they want to distract may want to this more parsimonious account, emotion to disengage from emotional processing. regulation choice can be fully explained by However, in a second potential key dimen- a direct influence from simple associative sion, the amount of cognitive effort people emotional processes (e.g., Bradley, Codis- wish to exert (e.g., Chajut & Algom, 2003; poti, Cuthbet, & Lang, 2001). Specifically, Kool, McGuire, Rosen, & Botvinick, 2010; according to an associative emotional pro- Muraven & Baumeister, 2000), when people cess account, as negative emotional intensity prefer to reappraise they are willing to exert increases, it directly activates a basic defen- considerable effort, and when they want sive system to shift from an engagement (or to distract they wish to preserve cognitive sensory intake) preference, which leads to effort. Accordingly, when people prefer to preferring reappraisal, to a disengagement distract in high-n­ egative emotionally intense (or sensory rejection) preference, which may situations, do they prefer to disengage from result in an increased preference to distract. emotional processing or do they prefer to Alternatively, according to the emotion reg- reserve their cognitive resources and go with ulation choice account, individuals can use the easy regulatory option? deliberate executive processes in preferring to reappraise in low-­negative intensity situa- To determine between these two options, tions and to distract in high-n­ egative inten- we had participants choose between two sity situations. Because both accounts lead types of distractions that involve making to the same regulatory choice prediction in mathematical computations (e.g., see Erber negative emotional situations, in order to & Tesser, 1992; van Dillen & Koole, 2007 determine between them we investigated for similar operationalization of atten- down-­regulation of positive emotional situa- tional distractions): One regulatory option tions—a­ n emotional context where the two was cognitively simple and involved minor accounts diverge. Specifically, the associative disengagement from emotional processing emotional process account would argue that (subtract 2’s) and a second regulatory option as positive emotional intensity increases, it was cognitively effortful yet highly disen- directly activates a basic appetitive system gaged from emotional processing (subtract that would lead to an increased preference 7’s). The logic was that if the cognitive effort to engage. By contrast, the emotion regula- dimension is central in regulatory choice, tion choice account would predict that the then one should expect that the preference to operation of deliberate control processes, use the more simple subtract 2’s distraction whose goal is to provide down-­regulation of would increase as negative emotional inten- positive emotional situations, would involve sity increases. However, if the engagement–­ overriding the associative tendency to disengagement dimension is central in engage, resulting in an increased preference regulatory choice, we should expect that to disengage as positive emotional intensity the preference to use the more disengaging increases. The results, which showed par- (despite it being also more effortful) subtract ticipants’ clear tendency to disengage via 7’s distraction would increase as negative distraction as positive emotional intensity emotional intensity increases. increased, clearly support the involvement of deliberate executive control processes Results strongly supported the centrality that override opposing associative emotional of the engagement–­disengagement dimen- sion, where participants preferred to disen- gage from emotional processing in the high-­

Emotion Regulation Choice 133 intensity condition despite the fact that this emotion regulation have almost exclusively regulatory option was clearly more effort- focused on the consequences of implement- ful (Sheppes et al., 2012). In addition, when ing different emotion regulation strategies participants were performing the subtract (Gross, 2007; see Koole, 2009, for reviews). 7’s option, their actual mathematical perfor- Multiple studies have instructed participants mance was interfered with less by the inten- to engage with different emotion regulation sity of emotional stimuli relative to perform- strategies and have examined the costs and ing subtract 2’s. This result indicates that benefits associated with successful imple- the more effortful subtract 7’s option indeed mentation. Understanding the consequences resulted in a stronger disengagement from of employing different regulation strategies emotional processing relative to subtracting is a crucial and important step toward an 2’s. understanding of the basic elements of emo- tion regulation strategies. Nevertheless, the In a complementary fashion we wished to emotion regulation choice findings extend show that the engagement–Âd

134 COGNITIVE APPROACHES one’s environment. For example, in intertem- Consider first the emotional intensity one poral choice paradigms, such as temporal is facing. As previously described, healthy discounting (Reynolds, 2006), individuals individuals prefer to use reappraisal in low-­ choose between receiving smaller monetary intensity emotional situations and distrac- rewards sooner (e.g., $5 today) and receiv- tion in high-­intensity situations. Regulatory ing bigger monetary rewards later (e.g., preferences that deviate from the flexible $8 tomorrow). Other examples of choice regulation choice observed in healthy indi- behavior involve deciding between differ- viduals might be related to different psy- ent internal processes to deal with external chopathologies. Specifically, deviation demands, such as the mathematical strate- from a preference to choose to disengage gies children choose in order to solve math from very high-­intensity stimuli is expected problems (Siegler, 2005) or the strategies from individuals who are prone to develop adults choose to solve chess problems (de major depression. According to the response Groot, 1978). Emotion regulation choice is style theory, rumination involves engag- a special case of decision making, because it ing with strong emotional experiences, and involves choosing between internal cognitive repeatedly thinking about their causes and processes to control one’s internal emotional consequences in an abstract and repeti- environment. tive way (e.g., Nolen-­Hoeksema, 1991; see Nolen-­Hoeksema et al., 2008, for reviews). Although emotion regulation choice Although rumination has been proven to be appears to be unique in some ways, it shares related to onset maintenance and relapse of with other theories basic assumptions about depression, depressed individuals who pre- strategy choice. Like classic theories in deci- fer to use positive distractions when dealing sion sciences (e.g., Payne, Bettman, & John- with strong emotional experiences show a son, 1988, 1993), the emotion regulation better prognosis (e.g., Nolen-­Hoeksema & choice account argues that individuals are Morrow, 1991, 1993). Emotion regulation sensitive to central factors (emotional, cog- choice is likely to be an important target in nitive, and motivational factors in the case the context of depression, because empirical of emotion regulation choice) when mak- studies have shown that although depressed ing their regulatory selections. In addition, individuals are able to implement distrac- just as other models highlight the role of tion effectively when instructed to do so learning (e.g., Rieskmap, 2006; Rieskmap (e.g., Joormann & Siemer, 2004; Joormann, & Otto, 2006), in the emotion regulation Siemer, & Gotlib, 2007), they hold a favor- choice account regulatory decisions are to able view of rumination, believing that it some extent based on prior knowledge, with helps them understand better the reasons the consequences of implementing different for depressed mood. Nevertheless, an over- strategies in different contexts. generalized preference to engage with emo- tional contents may also depend on impaired Implications for Clinical Psychology ability to shift attention away from negative Emotion regulation choice also has impor- contents (e.g., Joormann & Gotlib, 2008). tant clinical implications. Central concep- Another type of deviation from the healthy tual accounts argue that psychological well- regulatory choice patterns involves diverg- being requires flexibly adapting emotion ing from engagement with low-­intensity regulation strategies to fit with differing situ- emotional stimuli. Common to several anxi- ational demands (Gross, 2007; Kashdan & ety disorders is a tendency to overgeneral- Rottenberg, 2010; Watkins, 2011). The flip ize a disengagement or avoidance regulatory side of flexible regulatory choice is a rigid response (for reviews, see Campbell-­Sills & and maladaptive regulation choice that may Barlow, 2007; Foa & Kozak, 1986). Avoid- be related to various forms of psychopathol- ance usually starts in response to high-­ ogy. intensity emotional stimuli, but over time, it ends up spilling over to seemingly low-­ The recent empirical evidence maps emo- intensity stimuli. As pointed in our concep- tional, cognitive, and motivational determi- tual model, while disengagement strategies nants of regulation choice in healthy adults; are helpful in providing short-term relief, thus, deviations from healthy regulatory they are maladaptive in the long run and can choice can be used to understand different perpetuate anxiety and fears. forms of psychopathology.

Emotion Regulation Choice 135 The cognitive factor highlighted is the individuals’ regulatory choices. Specifically, ease with which a regulation strategy is gen- a temporary state of self-­control resource erated. Specifically, aiding the generation depletion (e.g., Baumeister, Vohs, & Tice, process resulted in increased reappraisal 2007; Muraven & Baumeister, 2000) is choice. These results are relevant for many likely to lead individuals to prefer strategies cognitive-b­ehavioral therapies in which such distraction, which provide short-term patients are encouraged to reappraise strong relief. Therefore, studying how various novel emotional events (e.g., Campbell-­Sills & factors affect regulatory choice is an impor- Barlow, 2007). While the final objective is tant future research direction. to have patients generate their own regula- tion strategies, therapists should be mindful Second, while distraction and reappraisal that their patients might need aid in gener- are considered classic disengagement and ating alternative ways to think about upset- engagement strategies (Parkinson & Tot- ting events, until they can gradually build terdell, 1999), and they are widely used in their skill. Complementarily, one can expect everyday life, studies of choices between that with continued practice in reappraisal, other emotion regulation strategies are patients would improve its generation and urgently needed. In everyday life, individu- choose to use it more frequently when it’s als choose from many more regulatory adaptive. options that should be studied in the future. Several promising avenues involve allowing The motivational factor highlighted indi- people to choose from regulatory options viduals’ goals when choosing to regulate that are considered less adaptive in certain their emotions. It was shown that when contexts (e.g., suppression and rumination). participants expect to encounter emotional Another option involves investigating how events repeatedly, they increase reappraisal individuals switch their regulatory choices choice, which offers long-term benefits. The when dealing with dynamic and prolonged ability to override regulatory preferences emotional events that constitute many of that offer short-term relief (i.e., distraction) our daily experiences (see Aldao & Nolen-­ in favor of regulatory preferences that offer Hoeksema, 2012c, for a related discussion). long-term benefits (i.e., reappraisal) is likely Relatedly, a crucial test of the applicability of to require self-c­ontrol, and impairments the emotion regulation choice paradigm and in self-c­ontrol ability have been linked to its supporting results should involve study- various psychopathologies, including addic- ing regulatory choices in daily emotional tions and eating disorders such as bulimia experiences, such as when patients wait to (for comprehensive reviews, see Heatherton receive potentially stressful news in medi- & Baumeister, 1991; Vohs & Baumeister, cal settings, or when clients face a stimulus 2011). about which they feel anxious. Although distraction and reappraisal are widely used Future Directions in real-life situations, it remains unclear Despite promising preliminary results and whether we would observe somewhat simi- clear, broad implications, research in emo- lar preferences when individuals spontane- tion regulation choice has only begun to ously choose between regulatory options in develop. In closing I wish to point out sev- daily emotional situations. eral potential future research directions. Third, the goals of this study were gener- First, to date, researchers have examined ally to characterize the influence of differ- the influence of only one emotional (emo- ent factors on emotion regulatory choice. tional intensity), one cognitive (generation Nevertheless, it is quite clear that studying of a strategy), and one motivational (short- individual differences in emotion regula- vs. long-term goals) determinant of emotion tion choice is crucial. Recent relevant studies regulation choice. While these factors are have shown that individual differences in the important, future studies should evaluate ability to modify emotions are tightly linked the influence of the many additional factors to long-term adaptation (e.g., Bonanno et that likely influence regulatory preference. al., 2004; Westphal et al., 2010). Therefore, To give just one example, the availability future studies should evaluate how mul- of cognitive resources is likely to influence tiple individual differences can moderate the influence of central factors on emotion regulation choice. One promising venue,

136 COGNITIVE APPROACHES which I mentioned in discussing the clinical References implications, involves studying impairments in regulatory preferences of individuals with Aldao, A., & Nolen-H­ oeksema, S. (2012a). mood, anxiety, and personality disorders, When are adaptive strategies most predictive whose psychopathology revolves around of psychopathology? Journal of Abnormal emotion dysregulation. Psychology, 121, 276–281. Fourth, according to the emotion regula- Aldao, A., & Nolen-H­ oeksema, S. (2012b). The tion choice account, healthy individuals are influence of context on the implementation of sensitive to central costs and benefits of dif- adaptive emotion regulation strategies. Behav- ferent regulation strategies when trying to iour Research and Therapy, 50, 493–501. choose between regulatory options, and in general these individuals showed adaptive Aldao, A., & Nolen-H­ oeksema, S. (2012c). One regulatory choice profiles. Nevertheless, versus many: Capturing the use of multiple multiple demonstrations in general decision-­ emotion regulation strategies in response to making studies have shown important limi- an emotion-­eliciting stimulus. Cognition and tations in human reasoning (e.g., Tversky & Emotion. [E-publication ahead of print] Kahneman, 1974); therefore, future studies should investigate situations in which indi- Aldao, A., Nolen-H­ oeksema, S., & Schweizer, viduals’ regulatory choices would not nec- S. (2010). Emotion regulation strategies essarily lead them to the best outcome. In across psychopathology: A meta-a­nalytic a related vein, while conscious regulation review. Clinical Psychology Review, 30, 217– strategies have been a major focus in the 237. field of emotion regulation, and an integral part of many cognitive-­behavioral thera- Altamirano, L. J., Miyake, A., & Whitmer, A. pies targeting emotion dysregulation (e.g., J. (2010). When mental flexibility facilitates Linehan, 1993), many emotion regulation cognitive control: Beneficial side effects of choices are likely to be determined implicitly ruminative tendencies on goal maintenance. and without deliberate control. Psychological Science, 21, 1377–1382. Finally, until now, emotion regulation Baumeister, R. F., Vohs, K. D., & Tice, D. choice has been investigated separately from M. (2007). The strength model of self-­ studies on the consequences of implement- control.  Current Directions in Psychological ing regulation strategies. Future studies Science, 16, 396–403. should make stronger connections between the developing studies on emotion regula- Blechert, J., Sheppes, G., Di Tella, C., Williams, tion choice and the well-­established studies H., & Gross, J. J. (2012). See what you think: on the consequences of regulation imple- Reappraisal modulates behavioral and neural mentation. For example, studies should responses to social stimuli. Psychological Sci- evaluate whether the effectiveness of imple- ence, 4, 356–363. menting a given strategy is moderated by an ability to override default regulatory choice Bonanno, G. A. (2005). Resilience in the face of preferences. At the same time, future studies potential trauma. Current Directions in Psy- in emotion regulation choice should utilize chological Science, 14, 135–138. multiple levels of analysis that combine the concurrent assessment of the effectiveness of Bonanno, G. A., & Keltner, D. (1997). Facial a chosen regulatory strategy. expressions of emotion and the course of con- jugal bereavement. Journal of Abnormal Psy- Acknowledgments chology, 106, 126–137. This chapter draws on and updates previous Bonanno, G. A., Papa, A., Lalande, K., West- reviews by Sheppes and Gross (2011, 2012) and phal, M., & Coifman, K. (2004). The impor- two recent empirical manuscripts (Sheppes et al., tance of being flexible: The ability to both 2011, 2012). The writing of this chapter was sup- enhance and suppress emotional expression ported by the Israel Science Foundation (Grant predicts long term adjustment. Psychological No. 1393/12). Science, 15, 482–487. Bradley, M. M., Codispoti, M., Cuthbert, B. N., & Lang, P. J. (2001). Emotion and motivation I: Defensive and appetitive reactions in picture processing. Emotion, 1, 276–298. Bradley, M. M., Codispoti, M., Sabatinelli, D., & Lang, P. J. (2001). Emotion and motivation II: Sex differences in picture processing. Emo- tion, 1, 300–319.

Emotion Regulation Choice 137 Campbell-ÂS

138 COGNITIVE APPROACHES the contents of working memory in depres- (2012). Unpacking cognitive reappraisal: sion: Interference from irrelevant negative Goals, tactics and outcomes. Emotion, 12, material. Journal of Abnormal Psychology, 250 –255. 117, 182–192. Muraven, M., & Baumeister, R. F. (2000). Self-­ Joormann, J., & Siemer, M. (2004). Memory regulation and depletion of limited resources: accessibility, mood regulation and dysphoria: Does self-­control resemble a muscle? Psycho- Difficulties in repairing sad mood with happy logical Bulletin, 126, 247–259. memories? Journal of Abnormal Psychology, Nolen-H­oeksema, S. (1991). Responses to 113, 179–188. depression and their effects on the duration of Joormann, J., Siemer, M., & Gotlib, I. H. (2007). depressive episodes. Journal of Abnormal Psy- Mood regulation in depression: Differential chology, 100, 569–582. effects of distraction and recall of happy mem- Nolen-H­ oeksema, S., & Morrow, J. (1991). A ories on sad mood. Journal of Abnormal Psy- prospective study of depression and post-­ chology, 116, 484–490. traumatic stress symptoms following a natural Kalisch, R. (2009).The functional neuroanatomy disaster: The 1989 Loma Prieta Earthquake. of reappraisal: Time matters. Neuroscience Biobehavioral Reviews, 33, 1215–1226. Journal of Personality and Social Psychology, Kashdan, T. B., & Rottenberg, J. (2010). Psy- 61, 115–121. chological flexibility as a fundamental aspect Nolen-H­ oeksema, S., & Morrow, J. (1993). of health. Clinical Psychology Review, 30, Effects of rumination and distraction on 865–878. naturally-o­ccurring depressed mood. Cogni- Kool, W., McGuire, J. T., Rosen, Z., & Botv- tion and Emotion, 7, 561–570. inick, M. M. (2010). Decision making and Nolen-H­ oeksema, S., Wisco, B. E., & Lyubomir- the avoidance of cognitive demand. Journal sky, S. (2008). Rethinking rumination. Per- of Experimental Psychology: General, 139, spectives on Psychological Science, 3, 400– 665 – 682 . 424. Koole, S. L. (2009). The psychology of emotion Ochsner, K. N., & Gross, J. J. (2008). Cogni- regulation: An integrative review. Cognition tive emotion regulation: Insights from social and Emotion, 23, 4–41. cognitive and affective neuroscience. Current Kross, E., & Ayduk, O. (2008). Facilitating Directions in Psychological Science, 17, 153– adaptive emotional analysis: Distinguishing 158. distanced-­analysis of depressive experiences Ochsner, K. N., Ray, R. D., Robertson, E. R., from immersed-­analysis and distraction. Per- Cooper, J. C., Chopra, S., Gabrieli, J. D. E., sonality and Social Psychology Bulletin, 34, et al. (2004). For better or for worse: Neural 924–938. systems supporting the cognitive down- and Lehle, C., & Hübner, R. (2008). On-the-fly up-r­egulation of negative emotion. NeuroIm- adaptation of selectivity in the flanker task. age, 23(2), 483–499. Psychonomic Bulletin and Review, 15, 814– Parkinson, B., & Totterdell, P. (1999). Classify- 818. ing affect regulation strategies. Cognition and Linehan, M. M. (1993). Skills training manual Emotion, 13, 277–303. for treating borderline personality disorder. Pashler, H. (1998). The psychology of attention. New York: Guilford Press. Cambridge, MA: MIT Press. MacNamara, A., Ochsner, K. N., & Hajcak, G. Payne, J. W., Bettman, J. R., & Johnson, E. J. (2011). Previously reappraised: The lasting (1988). Adaptive strategy selection in decision effect of description type on picture-e­licited making. Journal of Experimental Psychol- electrocortical activity. Social, Cognitive, and ogy: Learning, Memory, and Cognition, 14, Affective Neuroscience, 6, 348–358. 534–552. Mancebo, M. C., Eisen, J. L., Grant, J. E., & Ras- Payne, J. W., Bettman, J. R., & Johnson, E. J. mussen, S. A. (2005). Obsessive–c­ompulsive (1993). The adaptive decision maker. New personality disorder and obsessive–­compulsive York: Cambridge University Press. disorder: Clinical characteristics, diagnostic Reynolds, B. (2006). A review of delay-­ difficulties, and treatment. Annals of Clinical discounting research with humans: Relations Psychiatry, 17, 197–204. to drug use and gambling. Behavioural Phar- McRae, K., Ciesielski, B. R. G., & Gross, J. J. macology, 17, 651–667. Richards, J. M., & Gross, J. J. (1999). Compo-

Emotion Regulation Choice 139 sure at any cost?: The cognitive consequences Siegler, R. S. (2005). Children’s learning. Ameri- of emotion suppression. Personality and Social can Psychologist, 60, 769–778. Psychology Bulletin, 25, 1033–1044 Richards, J. M., & Gross, J. J. (2000). Emotion Tamir, M. (2011). The maturing field of emotion regulation and memory: The cognitive costs of regulation. Emotion Review, 3, 3–7. keeping one’s cool. Journal of Personality and Social Psychology, 79, 410–424. Thiruchselvam, R., Blechert, J., Sheppes, G., Rieskamp, J. (2006). Perspectives on probabi- Rydstrom, A., & Gross, J. J. (2011). The tem- listic inferences: Reinforcement learning and poral dynamics of emotion regulation: An an adaptive network compared. Journal of EEG study of distraction and reappraisal. Bio- logical Psychology, 87, 84–92. Experimental Psychology: Learning, Mem- ory, and Cognition, 32, 1355–1370. Troy, A. S., & Mauss, I. B. (2011). Resilience in Rieskamp, J., & Otto, P. E. (2006). SSL: A the- the face of stress: Emotion regulation ability ory of how people learn to select strategies. as a protective factor. In S. Southwick, B. Litz, D. Charney, & M. Friedman (Eds.), Resiliency Journal of Experimental Psychology: Gen- and mental health (pp. 30–44). Cambridge, eral, 135, 207–236. UK: Cambridge University Press. Sheppes, G., Catran, E., & Meiran, N. (2009). Reappraisal (but not distraction) is going to Tversky, A., & Kahneman, D. (1974). Judgment make you seat: Physiological evidence for self under uncertainty: Heuristics and biases. Sci- control effort. International Journal of Psy- ence, 185, 1124–1131. chophysiology, 71, 91–96. Sheppes, G., & Gross, J. J. (2011). Is timing van Dillen, L. F., & Koole, S. L. (2007). Clear- everything?: Temporal considerations in emo- ing the mind: A working memory model of tion regulation. Personality and Social Psy- distraction from negative mood. Emotion, 7, chology Review, 15, 319–331. 715–723. Sheppes, G., & Gross, J. J. (2012). Emotion regu- lation effectiveness: What works when. In H. Vohs, K. D., & Baumeister, R. (Eds.). (2011). A. Tennen & J. M. Suls (Eds.), Handbook of psychology (2nd ed., pp. 391–406). Indianap- Handbook of self-­regulation: Research, the- olis, IN: Wiley-­Blackwell. ory, and applications (2nd ed.). New York: Sheppes, G., & Meiran, N. (2007). Better late Guilford Press. than never?: On the dynamics of on-line regu- Vohs, K. D., Baumeister, R. F., Schmeichel, B. J., lation of sadness using distraction and cogni- Twenge, J. M., Nelson, N. M., & Tice, D. M. tive reappraisal. Personality and Social Psy- (2008). Making choices impairs subsequent chology Bulletin, 33, 1518–1532. self-­control: A limited resource account of Sheppes, G., & Meiran, N. (2008). Divergent decision making, self-­regulation, and active cognitive costs for online forms of reappraisal initiative. Journal of Personality and Social and distraction. Emotion, 8, 870–874. Psychology, 94, 883–898. Sheppes, G., Scheibe, S., Suri, G., & Gross, J. J. Watkins, E. R. (2008). Constructive and uncon- (2011). Emotion-­regulation choice. Psycholog- structive repetitive thought. Psychological ical Science, 22, 1391–1396. Bulletin, 134, 163–206. Sheppes, G., Scheibe, S., Suri, G., Radu, P., Watkins, E. (2011). Dysregulation in level of goal Blechert, J., & Gross, J. J. (2012). Emotion and action identification across psychological regulation choice: A conceptual framework disorders. Clinical Psychology Review, 31, and supporting evidence. Journal of Experi- 260 –278. mental Psychology: General. [E-publication Westphal, M., Seivert, N. H., & Bonanno, G. ahead of print] A. (2010). Expressive flexibility. Emotion, 10, 92–100. Wilson, T. D., & Gilbert, D. T. (2008). Explain- ing away: A model of affective adaptation. Perspectives on Psychological Science, 3, 370 –386.

Chapter 9 Emotion Regulation and Decision Making Alessandro Grecucci Alan G. Sanfey Emotion and Decision Making ematical models of behavior. These models Definition of the Field typically strive to outline what the correct Recent years have seen a notable increase decision might be when selecting from a in interest in examining the processes that specified choice set, with these models gen- underlie human decision making, with one erally assuming that decisions are based on a of the more energetic new directions being “rational,” cognitive evaluation of their con- that of how emotions can impact our deci- sequences (Sanfey, 2007). While the explicit sions and choices (Sanfey, 2007). Though a and analytically tractable solutions have relatively recent development, this research certainly been of use in examining how we area has already uncovered some compelling make decisions, and in particular perhaps findings with regard to the emotional and how we should make decisions, factors such social factors involved in how we make deci- as emotions, mood, and social cues have sions. Indeed, it is now commonly argued typically been excluded from these models. that the principal way in which actual However, in recent decades there has been a decision-­making behavior deviates from the welcome increased focus on the roles these choices prescribed by economic models of factors play in decision making, and experi- decision making may be in large part due mental evidence has begun to clarify greatly to emotional factors that weigh heavily on how emotions impact our decision behavior. our actual decision but are seldom taken into account by the standard models. Hence, Emotions can affect individual’s goals, greater knowledge of the exact role that attitudes, and social decision making (For- emotions play in decision making is invalu- gas, 2003; Zajonc, 2000). For example, able in building complete, accurate, models unpleasant emotions are associated with of choice. lower confidence and a more risk-a­verse and vigilant processing style (Clark & Isen, The Role of Emotion 1982; Isen & Daubman, 1984). in Decision Making Historically, the study of how we make On the other hand, pleasant emotions decisions has been tremendously influenced have been associated with higher confi- by economic theories grounded in math- dence, more optimistic framing, and greater 140 cooperation (Forgas, Bower, & Moylan, 1990). More related to economic decisions is the work done by researchers in the field of neuroeconomics. One of the first well-­

Emotion Regulation and Decision Making 141 characterized demonstrations of the rela- reflect the activation of emotional variables tionship between emotions and economic that can be additionally increased by subtle choices famously showed that damage to incidental mood states (Harlé & Sanfey, particular neural areas can radically bias 2007), and, given that the role of emotions in decision making, with the lesioned areas human social and individual decision mak- of interest thought to contribute important ing has been widely demonstrated (Slovic, emotional information to choice (Damasio, Finucane, Peters, & MacGregor, 2004), 1994; Sanfey, Rilling, Aronson, Nystrom, suggests that the existing models of behav- & Cohen, 2003). Continuing in this vein, ior need to incorporate affective factors to many other basic decision-m­ aking studies arrive at a more comprehensive accounting have now shown that physiological indi- of human decision-­making behavior. ces, such as skin conductance, correlate with other decision-­relevant information, As a result of this understanding, such as imminent losses in a gambling task researchers have begun to incorporate emo- (Bechara, Damasio, Tranel, & Damasio, tional influences into decision process mod- 1997) and anger during negative interper- els by assuming the reciprocal modulation sonal interactions (Van’t Wout, Kahn, San- of affective and more deliberative modes of fey, & Aleman, 2006). thought (Harlé & Sanfey, 2007). One useful framework to interpret the role of emotions Other studies have looked at how emo- in decision making suggests that our choices tional responses to the stimuli themselves may be best understood as the operation of can directly impact decision making. For multiple underlying systems that interact to example, in a functional magnetic resonance form our decisions (Sanfey & Chang, 2008). imaging (fMRI) experiment, Sanfey et al. (2003) showed that activation of the anterior There is a long tradition in psychology for insula, known to be associated with disgust a dual-s­ystems approach in the understand- and anger (Phillips et al., 1997), correlated ing of cognitive processes, and one aspect of with the rejection of inequitable financial dual systems that is relevant for this chap- offers made by an opponent in an economic ter is the distinction between emotion-b­ ased task known as the Ultimatum Game. The and cognitive-b­ased judgments. Emotions interpretation given was that insula activa- are rapid and automatic responses to spe- tion reflected negative emotional reactiv- cific stimuli (Sanfey & Chang, 2008); how- ity in response to the unfair offers (Sanfey ever, in some circumstances we are required et al., 2003). This hypothesis was tested to adopt more flexible and context-­specific in a follow-­up study (Van’t Wout et al., solutions. At a neural level, emotional pro- 2006), which showed that skin conductance cesses activate brainstem reward processing responses, a measure for emotional arousal, regions, such as the basal ganglia; ventral were higher for unfair offers, and this mea- parts of the brain, such as the ventromedial sure discriminated between acceptances and frontal, orbitofrontal, and anterior cingulate rejections of these offers. cortex; and amygdala and insular cortex (Dalgleish, 2004; Sanfey & Chang, 2008). Researchers have also carefully manipu- Deliberative processing engages anterior lated emotions either before or during choices and dorsolateral and posterior regions of to demonstrate the causality of the relation- the brain (Duncan, Emslie, Williams, John- ship (Lerner, Small, & Loewenstein, 2004; son, & Freer, 1996; Smith & Jonides, 1999; Winkielman, Berridge, & Willbarger, 2005; Miller & Cohen, 2001; Stuss & Knight, Harlé & Sanfey, 2007; Andrade & Ariely, 2002), whereas automatic processes acti- 2009; Sokol-H­ essner, Camerer, & Phelps, vate more parietal and subcortical systems. 2013). Recent research in this vein suggests It seems evident that emotions play a role that task-­irrelevant mood states may bias in decision making in general; however, decision making (Harlé, Chang, Van’t Wout, it is still largely unknown how these pro- & Sanfey, 2012). For example, Harlé and cesses interact with deliberative systems and Sanfey (2007) showed that negative emotion whether they can be controlled in a “top- is associated more with lower acceptance down” fashion (Grecucci, Giorgetta, Bonini, rates of unfair monetary. What rational mod- & Sanfey, 2013a, 2013b). One proposal on els consider suboptimal (i.e., turning down how the two systems may interact comes monetary gain) therefore appears likely to from Evans (2009) who conceptualized the

142 COGNITIVE APPROACHES sion making have two notable advantages as conflict between these two styles of thinking compared to basic emotion regulation stud- as a “cognitive control problem,” referring ies: (1) It provides the opportunity to see to the fundamental question of the mecha- how these regulatory strategies can actually nism by which control over the answer is affect behavioral responses themselves and ultimately allocated, and suggested how this not only emotional perception per se, and (2) conflict could be resolved. Evans pointed out it allows for the study of complex emotions that when confronted with a decision, auto- that are often difficult to elicit via standard matic processing produces default intuitive sets of simple perceptual stimuli. responses, unless slower and more reflective deliberative processing intervenes. Impor- Though there has been much important tantly, such an intervention may or may not recent progress on understanding emotion lead to a different decision that in turn may regulation, most studies to date have mainly or may not be the correct one. However, measured the ability to regulate negative the linkage between emotions and decision emotions induced by the viewing of images making may be more complex and multi- of valenced and arousing scenes (see Och- faceted, as suggested by the “fourfold clas- sner & Gross, 2008, for a review). How- sification of emotions and decision-m­ aking” ever, in daily life we are confronted with an model of Pfister and Böhm (2008), accord- often bewildering range of decisions, each ing to which emotions provide information containing an evaluation of gains, losses, about pleasure and pain during preference and associated risks, and usually laden with construction, enable rapid choices under emotional value. Therefore, investigating time pressure, lead attention on relevant whether emotion regulation strategies can aspects of a decision problem, and may gen- have an effect on actual decisions provides erate commitment concerning morally and an exciting opportunity to extend emotion socially significant decisions. regulation research beyond the modulation of emotional responses per se. Emotion Regulation and Decision Making The effectiveness of emotion regulation Overview strategies is dependent on both the specific One useful approach to investigate how type of stimulus and the emotion to be regu- emotion and cognition interact is to exam- lated (Grecucci, 2012; Ochsner & Gross, ine whether regulatory strategies applied to 2008), and the emotions elicited by decision-­ decision-m­ aking tasks have the ability to making situations are substantially differ- alter subjects’ decisions by modulating the ent than those experienced while passively ongoing emotions (Van’t Wout, Chang, & viewing emotional images (Van’t Wout et al., Sanfey, 2010). Emotion regulation refers to a 2010). In addition to the benefits of better set of different strategies by which individu- understanding the processes of emotional als influence which emotions they have, when regulation by examining these mechanisms they have them, and how they experience in concert with decision making, the field of and express them. Emotions can intervene decision making itself also stands to benefit in different ways in our decisions, starting from this research program. Decision mak- from the assessment of potential alternative ing and emotion regulation are linked by the options, the evaluation of a choice, the vari- role that emotion plays in both (Mitchell, ous possible outcomes, and so on. Although 2011), and emotions appear to play a fun- there is surprisingly little empirical work damental role in choices associated with on the link between top-down control of reward or punishment. Moreover, psychiat- emotions and the act of decision making, ric disorders are often associated with emo- the experimental use of emotion regulation tion regulation deficits and abnormal deci- strategies has great potential to elucidate sion making (Mitchell, 2011). For example, this relationship (Van’t Wout et al., 2010), neuropsychological studies have shown that that is, to clarify how emotions and decision damage in brain areas implicated in emo- making interact. Broadly speaking, emo- tional regulation (i.e., ventromedial cortex) tion regulation strategies applied to deci- often result in suboptimal decision making (Koenigs & Tranel, 2007; Moretti, Drag- one, & di Pelegrino, 2009). A recent line of research has studied how emotion regulation

Emotion Regulation and Decision Making 143 strategies can affect decision making directly irrational evaluation of the risk associated and bias our decision-­making behavior as with a given choice (pathological gambling, presented in the following sections. substance abuse, etc.). Individual Decision Making Two recent studies have manipulated One useful organizational distinction when people’s emotion regulation strategies when talking about decision making is to separate making risky decisions. One study from the field into two main branches: individual Heilman, Crişan, Houser, Miclea, and decision making and social decision mak- Miu (2010) used cognitive reappraisal (an ing. This categorization is not trivial, and antecedent-­focused strategy that changes the it has particular implications for emotion meaning of the situation), expressive sup- regulation, because the affective experience pression (a response-­focused strategy that associated with these two kinds of decision involves inhibiting behavioral expressions can be quite different. Individual decision of emotions), and a control baseline condi- making refers to situations in which the tion in a between-­subjects design (Study 1). decision makers assess the available options, Subjects were required to apply these strat- then choose the best course of action for egies while playing the Balloon Analogue themselves based on their own value and Risk Task (BART; Lejuez et al., 2002), beliefs, for example, when choosing to play as well as the Iowa Gambling Task (IGT; a risky bet. Factors typically considered here Bechara, Damasio, Damasio, & Anderson, include risk evaluation, reward, loss aver- 1994). The BART provides a vivid measure sion, regret, and quantity evaluation. Social of risk taking by giving players the opportu- decision scenarios also require that these nity to progressively earn money by pump- factors be taken into account, but what dis- ing up computerized balloons (Lejuez et al., tinguishes these situations is the presence of 2002). While each additional pump provides another (involved) agent. A good example of an extra monetary payoff, the balloons can this is making decisions in a business nego- explode without warning, in which case the tiation context. These contexts additionally participant loses everything. Therefore, the involve evaluations of motivations such as task assesses the degree to which players are fairness, equity, cooperation, and in general prepared to risk an extra pump (for an extra more socially driven emotions. Here, we reward) at the risk of a total loss (Lejuez et first examine the experimental evidence of al., 2002). The well-known IGT is a comput- emotion regulation on individual decision erized card-­searching game that measures making, before turning our attention to the the degree to which individuals shift from social contexts. choices with large immediate gains (which, however, are also associated with even larger The Modulation of Risk Evaluation losses in the long term) toward the more pru- and Risk‑Seeking Behavior dent and eventually more rewarding choices One method to study the interaction between with small, immediate gains (Bechara et al., emotion and decision making is examin- 1994). ing risky choices, for example, when there is uncertainty about the exact outcome of Heilman et al. (2010) found that when the decision and, typically, a safe, relatively compared with the control group, the group low payout or low-cost option is weighed of reappraisers reported fewer negative emo- against an option with both higher costs tions and demonstrated reduced risk aver- and benefits. A canonical example would sion; that is, they made more risky deci- be a choice between a sure gain of $10 and sions in the BART task, operationalized as a risky choice, with a 50% chance of pay- a greater number of pumps on average per ing out $25. The importance of studying unexploded balloon. In the IGT, reapprais- the effect of emotion regulation strategies ers demonstrated improved performance in in modulating risky behaviors becomes par- the prehunch/hunch period of the task, in ticularly salient when considering that many which players are endeavoring to discover pathological behaviors stem from seemingly the exact traits of the various cards; there- fore, they yielded significantly higher earn- ings at the end of the game. Importantly, suppressors and controls showed no sig- nificant difference in both tasks in terms

144 COGNITIVE APPROACHES of emotional ratings and decision-­making behavior. Overall, then, this study showed Anticipation and Processing of Losses: that cognitive reappraisal appears to lead to The Regulation of Loss Aversion increased risk taking by reducing the expe- Kahneman and Tversky (1979) famously rience of negative emotions elicited by the codified the notion that losses loom larger decision scenario. than their equivalent gains, terming this phenomenon “loss aversion.” Experimental Another study (Martin & Delgado, 2011) studies over the past 30 years indicate that explored how emotion regulation strategies people demonstrate loss aversion for objects influence participants’ preferences when (Kahneman, Knetsch, & Thaler, 1990) and choosing between two monetary options: money (Tversky & Kahneman, 1992). Two a gamble (risky option) and a guaranteed studies from Sokol-­Hessner and his col- amount (safe option). Here, participants leagues (2009, 2013) showed that applying engaged in imagery-­focused regulation strat- emotion regulation strategies to decision egies during the presentation of a cue that making can affect the experience of loss preceded an economic decision. The emotion aversion. In the first study (Sokol-H­ essner et regulation strategies placed subjects in the al., 2009), participants were asked to reap- following experimental conditions: “Look,” praise, by taking a different perspective, “Relax,” or “Excite.” When instructed to about a choice between a binary gamble “Relax,” participants were asked to imagine and a guaranteed amount. Outcomes were a calming scene; when in the “Excite” condi- displayed following the decision phase (e.g., tion, participants were to imagine an excit- “You won” or “You lost”). For half of the ing scene, in order to increase their reactiv- trials, participants applied an “Attend” ity. After the imagery phase, participants strategy, which considered each choice in made a choice between a risky and a safe isolation, and for the other half, a “Regu- monetary lottery while undergoing fMRI. late” strategy, which emphasized choices in Participants who successfully used the emo- their broader context, such as “thinking like tion regulation strategy “Relax” made fewer a trader.” In other words, they were asked to risky choices in comparison to the “Look” imagine themselves to be experienced pro- condition. The “Excite” condition was not fessional stock traders. Risk aversion and effective in producing any change in deci- loss aversion indices were then computed, sion behavior. To better understand the based on participants’ actual choice behav- effect of the “Relax” strategy on decision ior. The authors were able to show that the making, the authors divided participants “Regulate” strategy had a strong effect in into two groups on the basis of their success decreasing individuals’ initial levels of loss at implementing the strategy. Regulators aversion but not risk aversion. Importantly, were then compared with nonregulators, to identify the degree of change in loss aver- and the former showed a significant pro- sion according to differences in regulation portion of fewer risky choices as compared success, subjects were divided into “regu- with “Look” trials, suggesting that emotion lators” and “nonregulators.” Reduced loss regulation strategies are able to modulate aversion was also associated with reduced risky decisions (Martin & Delgado, 2011). physiological arousal responses to loss out- Moreover, at the neural level, responses in comes, further showing that the reappraisal the striatum, a key region associated with strategy was also associated with decreased risky choice, were attenuated during deci- emotional reaction to bad outcomes. sion making when participants adopted the “Relax” emotion regulation strategy (Mar- In the second study, using a similar para- tin & Delgado, 2011). In contrast, partici- digm, Sokol-­Hessner et al. (2013) found pants who did not effectively use emotion that success in regulating loss aversion by regulation strategies (i.e., the nonregulator using emotion regulation strategies (reap- group) failed to show neural differences praisal), was associated with a reduction during decision making. In conclusion, the in amygdala responses to losses but not to authors argued that imagery-f­ocused strate- gains. Moreover, the application of the reap- gies “can foster more goal-d­ irected behavior praisal strategy increased the baseline activ- and promote safer, compared with riskier, ity in prefrontal cortices and the striatum, in decision-­making.” line with previous studies (Ochsner, Bunge,

Emotion Regulation and Decision Making 145 Gross, & Gabrieli, 2002; Banks, Eddy, Ang- to be won (€0.50, €0.10), followed by a wait- stadt, Nathan, & Phan, 2007; Eippert et al., ing period. A target (square or triangle) was 2007; Delgado, Gillis, & Phelps, 2008; Del- then presented, after which subjects became gado, Nearing, LeDoux, & Phelps, 2008). eligible to win the announced amount of These results support the idea that different money. Immediately afterwards, a feed- modes of thinking can alter both choices and back display informed the subjects whether arousal responses related to loss aversion. they had actually won (€0.50, €0.10) or not (€0.00). Subjects in one condition were Regulation of Rewards, Expected Values, asked to rethink an emotion-e­liciting situ- and Prediction Error ation in nonemotional terms by distancing Another line of research has involved the themselves from feelings and behaving as study of rewards in an economic decision-­ neutral observers (distancing; Gross, 2002). making context. A reward can be defined as In another condition (“Permit”) they were “an object or event that generates approach allowed to experience their feelings. Dis- and consummatory behavior, represents tancing significantly lowered subjects’ per- positive outcomes of decisions and may ceived feelings. More importantly, a modu- engage positive emotions” (cf. Schultz, 2009, lation of expected value and its associated p. 323). In other words, rewards are positive prediction error was observed in the right outcomes that occur after specific events. ventral striatum of subjects while applying the distancing strategy. The comparison of The importance of studying emotion reg- the “Distance” and “Permit” conditions ulation strategies of reward stems from the showed a significant activation in the right-­ fact that dysregulation of reward seems to hemispheric frontoparietal network (includ- be associated with pathological addiction ing the rostral cingulate zone, lateral orbi- syndromes (Heinz, 2002; Robinson & Ber- tofrontal cortex [OFC] and inferior parietal ridge, 2003; Volkow & Wise, 2005; Wrase lobe/temporal parietal junction). Interest- et al., 2007). At the neural level, when the ingly, reappraisal success was positively cor- reward is greater than expected, there is a related with activation in the rostral cingu- positive prediction error (greater ventral late zone and the lateral OFC (Staudinger et striatal activity), and when the reward is al., 2009). The authors were therefore able less than expected (or missing altogether), to show that emotion regulation strategies there is a negative prediction error (less ven- generally impact reward coding, and also tral striatal activity) (Hollerman & Schultz, modulate expected value and prediction 1998; McClure, Berns, & Montague, 2003; error encoding in the ventral striatum. Staudinger, Erk, Abler, & Walter, 2009). Prediction errors have been interpreted as In the second study, Staudinger and col- “teaching” signals, helping participants leagues (2011) used a similar task but to adapt their behavior so as to maximize explored whether emotion regulation would future rewards acquisition (Sutton, 1988; also be effective in decreasing the moti- Montague, Hyman, & Cohen, 2004; vation to obtain a reward. The authors Staudinger et al., 2009). found reduced positive evaluative feelings for larger gains (€1) during the “Distance” Two studies from Staudinger and col- instruction (from 4.08 to 2.50 points), as leagues (2009; Staudinger, Erk, & Walter, well as a slowdown in reaction times, which 2011) addressed the question of whether they interpreted as a lowered motivation to emotion regulation might also impact reward obtain the reward. At a neural level, using processing. In the first study, Staudinger psychophysiological interaction analyses, et al. (2009), the authors used a monetary they found that the dorsolateral prefrontal incentive delay (MID) task (Knutson, Tay- cortex played a fundamental role in exerting lor, Kaufman, Peterson, & Glover, 2005) direct modulatory control over the striatum to investigate whether regulation strategies by promoting low-r­eward responses. In the affect the neural processing of reward and same fashion, another experiment showed its associated motivated behavior. Subjects that imagery-­based strategies successfully were scanned with fMRI while playing the modulate reward expectation in a monetary MID task, in which subjects observed an reward–c­onditioning procedure (Delgado, abstract cue indicating the amount of money Nearing, et al., 2008).

146 COGNITIVE APPROACHES the more well-known tasks in this field is Overall, these experiments showed that the Ultimatum Game (UG; Guth, Schmitt- emotion regulation strategies applied to indi- berger, & Schwarz, 1982), in which two vidual decisions have the power to alter our players must agree on the division of a sum of choices. Emotions such as risk perception, money provided by the experimenter (Van’t reward, and loss evaluation are modulated Wout et al., 2010). One player, the proposer, by the usage of different kind of strategies makes an offer as to how this money should (cognitive or relaxation-b­ ased imagery), and be split. The responder must then make a this alters our decision behavior toward safer decision to either accept or reject this offer. behavior (Martin & Delgado, 2011), toward If the offer is accepted, then the money is less sensitivity to outcomes (Sokol-H­ essner split as proposed, but if the responder rejects et al., 2013), and toward lower motivation the offer, then neither player receives any- to obtain an economic reward (Staudinger et thing. A central focus of the literature has al., 2011). been the well-­replicated result that low offers (i.e., less than 30% of the total amount) are Social Decision Making rejected approximately 50% of the time Despite the extensive literature on emotional (Camerer, 2003). That is, UG choices are “self-r­egulation” (see Ochsner & Gross, not purely driven by financial self-i­nterest, 2008), evidence of emotion regulation in but also incorporate affective information social interactive situations is still relatively about the social interaction, with evidence scant. Indeed, interest in such “social regu- that negative emotions play an important lation” has been explored only recently in role in punishment behavior (Pillutla & a study examining the emotion regulation Murnighan, 1996; Xiao & Houser, 2005). of subjects looking at pictures depicting Put simply, the UG gives us the opportunity social versus nonsocial scenes. Studying the to elicit anger reactions in subjects via real mechanisms involved in the regulation of social interactions, then to measure their real social situations is particularly impor- willingness to punish the perpetrator (and, tant when considering the failures to regu- of course, themselves) by examining rejec- late interpersonal interactions exhibited in tion rates of these unequal offers. psychiatric disorders such as borderline per- sonality disorder or social anxiety disorder One study from Van’t Wout and col- (Grecucci, 2012, p. 135); Ochsner, 2008; leagues (2010) showed that participants Ochsner & Gross, 2008). Recent experi- who were asked to reappraise their emotions ments on emotion regulation and social deci- accepted a greater number of unfair offers sion making hold the promise to elucidate (i.e., $1 or $2 offered from a pot of $10) the mechanisms subserving interpersonal than participants who did not use any regu- emotion regulation (Grecucci, Giorgetta, lation strategy (Experiment 1). Importantly, Bonini, & Sanfey, 2013). Moreover, emo- in another condition, subjects were asked tions elicited in social interactions may well to use a response-f­ocused strategy, namely, be of a qualitatively different nature than suppression of any negative emotions toward those experienced while making individual the proposer; however, this strategy was not decisions about monetary gains, rewards, or effective in altering the decision-­making of risky choices. subjects. Moreover, when acting later in the experimental session as proposers them- Anger selves (Experiment 2), participants who had A potentially useful approach to measuring used reappraisal strategies made more gen- socially driven emotions is to adopt tasks erous offers to their previously unfair part- derived from “game theory.” Game theory is ners than those who had use the strategy of the study of mathematical models of conflict suppression, demonstrating an interesting and cooperation between decision makers “follow-o­ n” effect of reappraisal. (Myerson, 1991), and offers useful models for the psychological investigation of social Following this line of research, Grecucci, exchange (Sanfey & Dorris, 2009). One of Giorgetta, Van’t Wout, Bonini, and Sanfey (2013) applied a similar paradigm to that of Van’t Wout et al. (2010) but also employed fMRI to explore the neural bases of regulated

Emotion Regulation and Decision Making 147 interpersonal decisions. In addition they similar to the UG, with the important excep- studied the effects of both up- and down-­ tion that responders cannot reject the divi- regulation in participants. This allowed sion of money made by the proposer; that them to explore whether the negative emo- is, they must accept whatever the proposer tions induced by an inequitable proposal can decides to give them from the experimenter-­ be both reduced and increased depending provided pot. Here, subjects were trained to on the strategy, and, importantly, whether reappraise the other player (a form of “men- this regulation in turn affects the decision to talizing”) in one condition, and to distance punish the other player for his or her unfair from the situation (“distancing”) in another. treatment. When using a down-r­egulation This kind of reappraisal comprised refram- strategy (rethink the situation as less nega- ing the intention and behavior of the other tive), a greater number of unfair offers were player in either a more or less negative way. accepted (i.e., fewer punishment behaviors). Following both altruistic and selfish divi- Up-­regulation, on the other hand, yielded a sions of money, emotional ratings were col- greater number of rejections of unfair offers lected separately for arousal and valence. (i.e., more punishment behaviors), with both Reappraisal strategies increased the valence strategy conditions compared with a base- ratings (less negative perception) for self- line “Look” condition. Subjective ratings ish offers ($1, $2, $3, from a total of $10), analyses clarified that subjects’ anger about and subjective ratings also indicated that the the unfair proposed division was the main main emotion involved when receiving such emotion regulated by the reappraisal strat- offers was “disappointment” (but also to egies. At the neural level, significant acti- some extent “anger” and “disgust”) in rela- vations during reappraisal were observed tion to the selfish proposers, while distanc- in the dorsolateral prefrontal cortex and ing did not produce any change compared cingulate gyrus, but only for unfair offers. with the “Look” condition. This study “Regulated” decisions activated the left infe- provisionally demonstrates that disappoint- rior frontal gyrus for up-­regulation and the ment regarding selfish behavior can be suc- middle frontal gyrus for down-­regulation cessfully modulated by mentalizing strate- strategies, respectively. The insula, one key gies (reappraisal of the intention of others), region that was active when participants whereas distancing from others (an interper- were treated unfairly (Sanfey et al., 2003), sonal strategy that is typical of social and showed less activation for down-r­egulation generalized anxiety disorders) is not effec- and more activation for up-r­egulation. In tive in modulating these reactions. conclusion, these studies provide strong evidence that interpersonal emotions (e.g., Overall, these experiments have begun anger when treated unfairly) can be success- to show that emotions in reaction to com- fully modulated by emotion regulation strat- plex social motivations such as inequity and egies. And, as an important consequence, selfishness can be successfully modulated by these studies also clearly show that emotion emotion regulation strategies, and that these regulation can alter our social behavior (in strategies can lead to changes in observed this case, punishment of unfair behavior), decision behaviors. with these effects observed in within-s­ ubject interactions. This has important implica- Conclusion tions for the regulation of social behavior in Our aim in this chapter has been to review interactive contexts. the extant studies that have used regula- tory strategies to moderate affective inputs Selfishness to the decision-m­ aking process. Though Another recent study by the same group relatively few in number to date, there has (Grecucci, Giorgetta, Bonini, et al., 2013) been encouraging success with these type of used a similar socioeconomic game, namely, manipulations, and overall these studies pro- the Dictator Game, to measure emotional vide compelling evidence that emotion regu- responses to both the altruistic and selfish lation strategies can have a strong impact behavior of others. The Dictator Game is on not only perceptions of stimuli, but also

148 COGNITIVE APPROACHES real, consequential, decisions in both social getta, Bonini, et al., 2013a, 2013b). Interest- and nonsocial contexts. This turn can lead ingly, dlPFC and vlPFC have been previously to interesting hypotheses about how our reported as active when reappraising emo- daily decisions can be modulated by emo- tional pictures (Ochsner et al., 2004). One tion regulation strategies. Though it seems interpretation might therefore be that dlPFC evident that emotional factors play a role in is involved in modulating both behavioral decision-m­ aking situations, the mechanisms and emotional outputs in order to satisfy integrating affective and more deliberative contextual demands (Mitchell, 2011). information in decision making are still relatively understudied. Recent experiments Interestingly, there is an overlap between have demonstrated changes in decision “decision-­making” and “regulatory” brain making following both incidental emotion areas. Indeed, the role of the lateral prefron- induction (Harlé & Sanfey 2007; Bonini et tal cortex in decision making has been con- al., 2011) and neurophysiological disruption ceptualized along four main lines: (1) nega- via transcranial magnetic stimulation (Van’t tive feedback encoding; (2) motor response Wout, Kahn, Sanfey, & Aleman, 2005; inhibition; (3) stimulus–­response reconfigu- Knoch, Pascual-­Leone, Meyer, Treyer, & ration; and (4) attention to behaviorally sig- Fehr, 2006), but emotion regulation studies nificant stimuli (Mitchell, 2011). The same of decision making have already extended considerations apply to the role of the lat- these findings in important ways by dem- eral PFC in emotion regulation. For exam- onstrating that emotions directly related to ple, Mitchell hypothesized that the lateral the choice situation itself can be controlled prefrontal cortex is involved in modulating through top-down effort, leading to changes behavioral and emotional output given par- in how the choice sets are represented and ticular contextual demands. Another key the associated decisions themselves. More- region connected to both emotion regulation over, these studies are important because and decision making is the cingulate cortex, they extend the range of emotions that can which may best be interpreted in terms of be affected by emotion regulation strategies. performance monitoring. One of these mon- itoring functions is the detection of response Neuroanatomical Bases of Emotion conflict between top-down “deliberative” Regulation of Decision Making signals and bottom-u­ p “emotional” signals. Overall, the brain regions implicated in the Other relevant regions for these processes aforementioned studies can be grouped into are the temporoparietal junction (TPJ) and two main functional clusters: the regions the inferior parietal lobe (IPL). Activation that implement the regulation strategies, of the right TPJ/IPL is in line with previous and the regions regulated by the strategies. reappraisal studies (Ochsner et al., 2002; Delgado, Nearing, et al., 2008; Staudinger For the former, regions acting as “regula- et al., 2009) and might indicate allocation tors” are primarily the dorsolateral (dlPFC) and dislocation of attention (Corbetta & and ventrolateral (vlPFC) portions of the Shulman, 2002). prefrontal cortex, and in some cases also the ventromedial regions. The dlPFC and The regions modulated by the strategies are vlPFC have been implicated in active cogni- generally connected with the emotions elic- tive control and inhibition (Miller & Cohen, ited in the tasks. For example, tasks involving 2001; Smith & Jonides, 1999), and may risk assessment found reduced activity in the underlie the generation and maintenance ventral striatum (Martin & Delgado, 2011); of reappraisal strategies (Ochsner et al., tasks involving reward found alterations in 2002, 2004). Importantly, the ventral part the activity of OFC, cingulate cortex, and of the prefrontal cortex is well-­connected the ventral striatum (Staudinger et al., 2009, with brain structures such as the insula, the 2011); experiments involving the experience amygdala, and the striatum. One hypothesis of monetary loss found reduced amygdala derived from these activations is that the pre- activity (Sokol-H­ essner et al., 2013); and frontal cortex actively modulates the activity social decision-m­ aking tasks such as the UG in more emotional regions under conditions yielded reduced activity in the insula, known of emotional regulation (Grecucci, Gior- to be associated with anger and feelings of disgust (Grecucci, Giorgetta, Bonini, et al., 2013a). See Table 9.1 for a summary.

149 TABLE 9.1.  Emotion Regulation and Decision Making Studies and Their Main Findings Emotion or emotion- Decision-making Emotion Main regulated regions Regulating Reference related behavior task regulation strategy Behavioral effect regions Heilman et al. — (2010) Risky behavior Balloon Analogue Suppression, Reduced risk aversion; — Risk Task; Iowa reappraisal increased monetary Gambling Task performance Risk perception Gambling Imagery-focused Reduced risky choices Ventral striatum Martin and Delgado regulation (2011) Loss aversion Gambling Reappraisal Reduced loss aversion Amygdala dlPFC, vmPFC Sokol-Hessner et al. (2009, 2013) Reward Monetary Incentive Distancing Attenuation of expected value; Ventral striatum, rostral dlPFC, CC, Staudinger et al. Delay Task modulation of prediction cingulate, orbitofrontal IPL/TPJ (2009, 2011) error; attenuation of pleasant cortex, putamen anticipation of gains Anger Ultimatum Game Reapprasial, Reduced rejection rates Insula dlPFC, CC Van’t Wout et al. suppression (2010); Grecucci, Giorgetta, Van’t Wout, et al. (2013) Responses to selfish Dictator Game Reapprasial, Less negative emotions Insula, Striatum, dIPFC, TPJ Grecucci, Giorgetta, behavior of others distancing posterior cingulate Bonini, and Sanfey (2013a, 2013b) Note. dlPFC, dorsolateral prefrontal cortex; vmPFC, ventromedial PFC; CC, cingulate cortex; IPL, inferior parietal lobe; TPJ, temporoparietal junction.

150 COGNITIVE APPROACHES Concluding Remarks positive emotions may also have a large and Future Directions impact on decision making, and future stud- In this chapter, we have reviewed the cur- ies could productively examine these affec- rent experimental evidence concerning emo- tive influences. Another intriguing finding tion regulation as applied to decision mak- is that, in some studies, augmented activity ing. While relatively few experiments to in the more “decision-­related” brain areas date have explicitly used regulation strate- has been reported. At least two studies have gies in the context of decision and choice, reported a baseline shift in dlPFC (Grecucci the results thus far are encouraging and et al., 2013; Sokol-­Hessner et al., 2013). The point to the usefulness of regulation in exact interpretation of this shift is a matter potentially modifying decision making. The for future research studies, but one specula- experimental findings are consistent with a tive account is that this may be involved in key role of emotional regulatory process in increased intentional control (Sokol-H­ essner decision making. For example, expressing et al., 2013) or more flexibility in satisfying emotions when bargaining in the UG has contextual demands (Grecucci, 2012; Gre- effects on the subsequent punishment behav- cucci, Giorgetta, Van’t Wont, et al., 2013). ior (Xiao & Houser, 2005). Individual and One important point that remains unsolved social decision-m­aking experiments have is how emotion regulation strategies affect shown that emotions connected to relevant decisions. Do regulatory strategies alter deci- decision factors such as loss, risk, reward, sions themselves, without impacting emo- interpersonal anger, and moral disgust can tions? Or does emotion regulation influence be successfully regulated by various strate- decision-­making processes by altering the gies. Notably, not all the strategies result emotion, which in turn alters decision mak- in equally successful changes in subjects’ ing? Even if the latter seems to be the more decision-­making behavior. While rein- plausible explanation, the two hypotheses terpretative strategies such as reappraisal have rarely been tested in the same experi- appear effective at altering decision mak- ment. Finally, recent studies of experimental ing and associated emotions, the same does psychopathology are providing important not necessarily apply to expressive strate- insights into how emotional dysregulation gies such as suppression (Heilman et al., (excessive anxiety or depression, etc.) affects 2010; Van’t Wout et al., 2010) or distancing our decisions. These studies may provide (Grecucci, Giorgetta, Bonini, et al., 2013; further proof of the importance of regula- but see Staudinger et al., 2009, for a posi- tory mechanisms (and their failure) in deci- tive result). Although researchers are now sion making. beginning to investigate the role of regula- tion in decision making, several factors still Overall, studying the processes and mech- need to be understood and explored. For anisms involved in how emotions are regu- example, there are many different types of lated in concert with actual decision-m­ aking decision making that have been hitherto behavior offers an appealing template for unexamined in terms of how regulation may future research. Demonstrating how regula- impact choice, such as impulsivity, temporal tion can actually impact both behavior and discounting, charitable donations, assess- subsequent choice outcomes extends the cur- ments of uncertainty and so on. Moreover, rent research topic in interesting directions some (e.g., Sokol-H­essner et al., 2009) and also has the potential to outline practi- have argued that ability to regulate may be cal strategies that can be usefully employed related to more efficient decision-m­ aking in real-life situations. behavior. One interesting question is there- fore whether individual differences in the References way we treat and regulate our emotions are predictive of more optimal choices and the Andrade, E. B., & Ariely, D. (2009). The endur- better avoidance of detrimental behaviors. ing impact of transient emotions on decision-­ making. Organizational Behavior and Human To date, researchers have focused their Decision Processes, 109(1), 1–8. attention primarily on the effect of nega- tive emotions on decision making. However, Banks, S. J., Eddy, K. T., Angstadt, M., Nathan, P. J., & Phan, K. L. (2007). Amygdala–­frontal

Emotion Regulation and Decision Making 151 connectivity during emotion regulation. Social affect infusion model (AIM). Psychological Cognitive and Affective Neuroscience, 2(4), Bulletin, 73(117), 39–66. 303–312. Forgas, J. P. (2003). Affective influences on atti- Bechara, A., Damasio, A. R., Damasio, H., & tudes and judgments. In R. J. Davidson, K. J. Anderson, S. W. (1994). Insensitivity to future Scherer, & H. H. Goldsmith (Eds.), Hand- consequences following damage to human book of affective sciences (pp.€596–618). New prefrontal cortex. Cognition, 50, 7–15. York: Oxford University Press. Bechara, A., Damasio, H., Tranel, D., & Dama- Forgas, J. P., Bower, G. H., & Moylan, S. J. sio, A. R. (1997). Deciding advantageously (1990). Praise or blame?: Affective influences before knowing the advantageous strategy. on attributions for achievement. Journal of Science, 275(5304), 1293–1295. Personality and Social Psychology, 59, 809– Bonini, N., Hadjichristidis, C., Mazzocco, K., 818. Demattè, M. L., Zampini, M., Sbarbati, A., Grecucci, A. (2012). Il Conflitto Epistemologico. et al. (2011). Pecunia olet: The role of inciden- tal disgust in the Ultimatum Game. Emotion, Psicoanalisi e Neuroscienze dei Processi anti- 11(4), 965–969. conoscitivi. Francavilla, Italy: Edizioni Psi- Camerer, C. F. (2003). Behavioral game theory. conline. New York: Russell Sage Foundation. Grecucci, A., Giorgetta, C., Bonini, N., & San- Clark, M. S., & Isen, A. M. (1982). Towards fey, A. (2013a). Living emotions, avoiding understanding the relationship between feel- emotions: behavioral investigation of the regu- ing states and social behavior. In A. H. Has- lation of socially driven emotions. Frontiers in torf & A. M. Isen (Eds.), Cognitive social psy- Psychology, 3(616). chology (pp.€73–108). New York: Elsevier. Grecucci, A., Giorgetta, C., Bonini, N., & Corbetta, M., & Shulman, G. L. (2002). Control SÂ

152 COGNITIVE APPROACHES making: A mediating role of the anterior Miller, E. K., & Cohen, J. D. (2001). An inte- insula. NeuroImage, 61(1), 32–40. grative theory of prefrontal cortex function. Heilman, R. M., Crişan, L. G., Houser, D., Annual Review of Neuroscience, 24, 167– Miclea, M., & Miu, A. C. (2010). Emotion 202. regulation and decision-m­ aking under risk and uncertainty. Emotion, 10(2), 257–265. Mitchell, D. G. V. (2011). The nexus between Heinz, A. (2002). Dopaminergic dysfunc- decision-­making and emotion regulation: tion in alcoholism and schizophrenia—­ A review of convergent neurocognitive sub- Psychopathological and behavioral correlates. strates. Behavioral Brain Research, 217, 215– European Psychiatry, 17, 9–16. 231. Hollerman, J. R., & Schultz, W. (1998). Dopa- mine neurons report an error in the temporal Montague, P. R., Hyman, S. E., & Cohen, J. D. prediction of reward during learning. Nature (2004). Computational roles for dopamine in Neuroscience, 1, 304–309. behavioural control. Nature, 431, 760–767. Isen, A. M., & Daubman, K. A. (1984). The influence of affect on categorization. Jour- Moretti, L., Dragone, D., & di Pellegrino, G. nal of Personality and Social Psychology, 47, (2009). Reward and social valuation deficits 1206–1217. following ventromedial prefrontal damage. Kahneman, D., Knetsch, J. L., & Thaler, R. H. Journal of Cognitive Neuroscience, 21(1), (1990). Experimental tests of the endowment 128–140. effect and the Coase theorem. Journal of Polit- ical Economy, 98, 1325–1348. Myerson, R. B. (1991). Game theory: Analysis of Kahneman, D., & Tversky, A. (1979). Prospect conflict. Cambridge, MA: Harvard University theory: An analysis of decision under risk. Press. Econometrica, 47, 263–291. Knoch, D., Pascual-L­ eone, A., Meyer, K., Treyer, Ochsner, K. (2008). The social-e­motional pro- V., & Fehr, E. (2006). Diminishing reciprocal cessing stream: Five core constructs and their fairness by disrupting the right prefrontal cor- translational potential for schizophrenia and tex. Science, 314, 829–832. beyond. Biological Psychiatry, 64(1), 48–61. Knutson, B., Taylor, J., Kaufman, M., Peterson, R., & Glover, G. (2005). Distributed neural Ochsner, K. N., Bunge, S. A., Gross, J. J., & representation of expected value. Journal of Gabrieli, J. D. (2002). Rethinking feelings: An Neuroscience, 25, 4806–4812. fMRI study of the cognitive regulation of emo- Koenigs, M., & Tranel, D. (2007). Irrational tion. Journal of Cognitive Neuroscience, 14, economic decision-m­ aking after ventromedial 1215–1229. prefrontal damage: Evidence from the Ulti- matum Game. Journal of Neuroscience, 27, Ochsner, K. N., & Gross, J. J. (2008). Cogni- 951–956. tive emotion regulation: Insights from social Lejuez, C. W., Read, J. P., Kahler, C. W., Rich- cognitive and affective neuroscience. Current ards, J. B., Ramsey, S. E., Stuart, G. L., et al. Directions in Psychological Science, 17, 153– (2002). Evaluation of a behavioral measure of 158. risk taking: The Balloon Analogue Risk Task (BART). Journal of Experimental Psychology: Ochsner, K. N., Ray, R. D., Cooper, J. C., Rob- Applied, 8, 75–84. ertson, E. R., Chopra, S., Gabrieli, J. D., et al. Lerner, J. S., Small, D. A., & Loewenstein, G. F. (2004). For better or for worse: Neural sys- (2004). Heart strings and purse strings: Car- tems supporting the cognitive down- and up-­ ryover effects of emotions on economic deci- regulation of negative emotion. NeuroImage, sions. Psychological Science, 15(5), 337–341. 23(2), 483–499. Martin, L. N., & Delgado, M. R. (2011). The influence of emotion regulation on decision-­ Pfister, H.-R., & Böhm, G. (2008). The multi- making under risk. Journal of Cognitive Neu- plicity of emotions: A framework of emotional roscience, 23(9), 2569–2581. functions in decision making. Judgment and McClure, S. M., Berns, G. S., & Montague, P. R. Decision Making, 3(1), 5–17. (2003). Temporal prediction errors in a passive learning task activate human striatum. Neu- Phillips, M. L., Drevets, W. C., Rauch, S. L., & ron, 38, 339–346. Lane, R. (2003). Neurobiology of emotion perception: II. Implications for major psychi- atric disorders. Biological Psychiatry, 54(5), 515–528. Phillips, M. L., Young, A. W., Senior, C., Bram- mer, M., Andrew, C., Calder, A. J., et al. (1997). A specific neural substrate for perceiv- ing facial expressions of disgust. Nature, 389, 495 – 498. Pillutla, M. M., & Murnighan, J. K. (1996). Unfairness, anger, and spite: Emotional rejec-

Emotion Regulation and Decision Making 153 tions of ultimatum offers. Organization Dorsolateral prefrontal cortex modulates Behavior and Human Decision Process, 68, striatal reward encoding during reappraisal of 208–224. reward anticipation. Cerebral Cortex, 21(11), Robinson, T. E., & Berridge, K. C. (2003). 2578–2588. Addiction. Annual Review of Psychology, 54, Stuss, D. T., & Knight, R. T. (2002). Principles 25–53. of frontal lobe function. New York: Oxford Sanfey, A. G. (2007). Social decision-­making: University Press. Insights from game theory and neuroscience. Sutton, R. (1988). Learning to predict by the Science, 318(5850), 598–602. methods of temporal differences. Machine Sanfey, A. G., & Chang, L. J. (2008). Multiple Learning, 3, 9–44. systems in decision making. Annals of the Tversky, A., & Kahneman, D. (1992). Advances New York Academy of Sciences, 1128, 53–62. in prospect theory: Cumulative representation Sanfey, A. G., & Dorris, M. (2009). Games in of uncertainty. Journal of Risk and Uncer- humans and non-human primates: Scanners tainty, 5, 297–323. to single units. In P. W. Glimcher, C. F. Cam- Van’t Wout, M., Chang, L. J., & Sanfey, A. G. erer, E. Fehr, & R. A. Poldrack (Eds.), Neu- (2010). The influence of emotion regulation on roeconomics (pp. 63–80). London: Academic social interactive decision-­making. Emotion, Press. 10(6), 815–821. Sanfey, A. G., Rilling, J. K., Aronson, J. A., Nys- Van’t Wout, M., Kahn, R. S., Sanfey, A. G., & trom, L. E., & Cohen, J. D. (2003). The neural Aleman, A. (2005). Repetitive transcranial basis of economic decision-m­ aking in the ulti- magnetic stimulation over the right dorsolat- matum game. Science, 300, 1755–1758. eral prefrontal cortex affects strategic decision-­ Schultz, W. (2009). Midbrain dopamine neurons: making. NeuroReport, 16, 1849–1852. A retina of the reward system? In P. W. Glim- Van’t Wout, M., Kahn, R. S., Sanfey, A. G., cher, C. F. Camerer, E. Fehr, & R. A. Poldrack & Aleman, A. (2006). Affective state and (Eds.), Neuroeconomics (p. 323). London: decision-m­aking in the ultimatum game. Academic Press. Experimental Brain Research, 169(4), 564– Slovic, P., Finucane, M. L., Peters, E., & Mac- 568. Gregor, D. G. (2004). Risk as analysis and risk Volkow, N. D., & Wise, R. A. (2005). How can as feelings: Some thoughts about affect, rea- drug addiction help us understand obesity? son, risk, and rationality. Risk Analysis, 24(2), Nature Neuroscience, 8, 555–560. 311–322. Winkielman, P., Berridge, K. C., & Wilbarger, Smith, E. E., & Jonides, J. (1999). Storage and J. L. (2005). Unconscious affective reactions executive processes in the frontal lobes. Sci- to masked happy versus angry faces influ- ence, 283(5408), 1657–1661. ence consumption behavior and judgments of Sokol-H­ essner, P., Camerer, C. F., & Phelps, E. value. Personality and Social Psychology Bul- A. (2013). Emotion regulation reduces loss letin, 31(1), 121–135. aversion and decreases amygdala responses to Wrase, J., Schlagenhauf, F., Kienast, T., Wusten- losses. Social Cognitive and Affective Neuro- berg, T., Bermpohl, F., Kahnt, T., et al. (2007). science, 8(3), 341–350. Dysfunction of reward processing correlates Sokol-­Hessner, P., Hsu, M., Curley, N. G., Del- with alcohol craving in detoxified alcoholics. gado, M. R., Camerer, C. F., & Phelps, E. NeuroImage, 35, 787–794. A. (2009). Thinking like a trader selectively Xiao, E., & Houser, D. (2005). Emotion expres- reduces individuals’ loss aversion. Proceeding sion in human punishment behavior. Proceed- of the National Academy of Sciences, 106(13), 5035–5040. ings of the National Academy of Sciences, Staudinger, M. R., Erk, S., Abler, B., & Walter, 102, 7398–7401. H. (2009). Cognitive reappraisal modulates Zajonc, R. B. (2000). Feelings and thinking: expected value and prediction error encod- Closing the debate over the independence of ing in the ventral striatum. NeuroImage, 47, affect. In J. P. Forgas (Ed.), Feeling and think- 713–721. Staudinger, M. R., Erk, S., & Walter, H. (2011). ing: The role of affect in social cognition (pp. 31–58). New York: Cambridge University Press.



Part IV Developmental Considerations



Chapter 10 Self‑Regulation, Effortful Control, and Their Socioemotional Correlates Nancy Eisenberg Claire Hofer Michael J. Sulik Tracy L. Spinrad In this chapter, we focus on the nature and many of the elements of others’ definitions development of temperamentally based but is broader in some respects and narrower effortful control (EC) processes and the in others. For us, it refers to processes used role of effortful self-r­egulation in children’s to manage and change whether, when, and socioemotional development. First we con- how (e.g., how intensely) one experiences sider definitional and conceptual issues. emotions and emotion-­related motivational Then we briefly summarize the normative and physiological states, as well as how development of self-r­egulation (including emotions are expressed behaviorally. Thus, EC), followed by review of research illustrat- it includes processes used to change one’s ing the importance of individual differences own emotional state, to prevent or initiate in children’s self-­regulatory skills (including emotional responding (e.g., by selecting or effortful control) for their socioemotional changing situations), to modify the signifi- development. We conclude by discussing a cance of an event for the self, and to modu- few current conceptual and methodological late the behavioral expression of emotion issues and gaps in the research. (e.g., through verbal or nonverbal cues). Our definition is consistent with Gross’s (this Conceptual Issues volume) model, in which emotion regulation Our discussion of EC must be considered includes situation selection, situation modi- in the broader context of conceptions of fication, attentional deployment, cognitive children’s self-­regulation. In the work with change, and response modulation. children, there has been a strong focus on self-­regulatory processes involved in We have used the term emotion-­related the modulation of emotion—­an aspect of self-­regulation because many of the pro- regulation labeled as emotion regulation. cesses frequently involved in emotional self-­ Definitions of this construct vary consider- regulation (see below) are also used for the ably (Gross, this volume). Our definition self-­regulation in the absence of emotion. of emotion-­related self-­regulation includes Although we fully acknowledge that emo- tional control/regulation often is externally imposed, especially early in life, we believe it is useful to differentiate between inter- 157

158 DEVELOPMENTAL CONSIDERATIONS nally generated self-­regulatory processes to which one attends, and planning—­are and those processes that the child does not involved in EC; thus, EF and EC are par- execute (e.g., parental soothing of an infant; tially overlapping constructs. In our view, Eisenberg et al., 2004; Eisenberg, Spinrad, EC and, as a result, aspects of EF, include & Eggum, 2010). Thus, we tend to use the an array of processes or capabilities that can term self-­regulation only for internally gen- be used to manage emotion and behavior; erated regulatory processes (although in the however, EC is not completely analogous to past we used the term regulation), similar emotional self-r­egulation because EC can to Gross’s (this volume) distinction between be used for other purposes (e.g., to focus intrinsic and extrinsic regulatory processes. on schoolwork). Moreover, self-r­egulation capacities likely include more than just EC; There are other variations in how the term for example, self-­regulation may include the regulation or even self-­regulation is used. motivation to utilize EC/EF in the service of Some people use it to refer to a host of pro- goals (Eisenberg & Zhou, in press). We view cesses that have a controlling effect on cog- the skills inherent in EC as building blocks nition, attention, or behavior. As we discuss for the development of self-­regulation, emo- later, we believe that it is useful to make a tional and otherwise, across the lifespan, distinction between more and less volitional and thus, as the core of internally based, processes involved in regulation/control. volitional self-r­egulation. Therefore, despite Although it is clear that many nonvolitional some distinctions between EC and self-­ processes have important modulating (even, regulation, the constructs are overlapping in a sense, regulating) effects on attention, and we tend to use the terms interchange- cognition, behavior, and physiological ably in this chapter. responding, it would be useful to differenti- ate the two types of processes semantically. The fact that EC is effortful or willful A term such as potentially volitional self-­ does not mean that the individual is always regulatory processes would be more precise aware that he or she is modulating emotion than self-­regulation in referring to volitional or behavior. Some aspects of EC, especially processes; however, such terminology is after rehearsal and practice, may often be quite awkward. We ask readers to focus less automatic and executed without much con- on the use of the word self-­regulation than scious awareness in contexts with relevant on the distinctions we address. trigger cues (Mischel & Ayduk, 2011); however, if needed, the individual is able to Because it is extremely difficult to differen- shift into a more volitional mode of func- tiate emotion from its regulation, we believe tioning. Effortful regulatory processes are it is useful to focus on the processes used distinguished from less voluntary reactive to manage attention, emotion, and associ- processes (see below) by the ease with which ated behavior rather than on the amount of they can be volitionally controlled. emotion experienced or expressed. In this respect, we find EC to be a useful construct. Any particular form of self-­regulation (or EC) is not necessarily good or bad in terms Effortful Control of its consequences. Nonetheless, we believe EC, the regulatory component of tempera- that effortful self-­regulatory processes are ment, is defined as “the efficiency of execu- more likely than some less volitional aspects tive attention, including the ability to inhibit of control (see below) to result in adaptive a dominant response and/or to activate a outcomes in normative contexts, or at least subdominant response, to plan, and to detect in desired goals, be they actually adaptive or errors” (Rothbart & Bates, 2006, p. 129). It not. This is because effortful capacities can involves the ability to deploy attention will- be applied at will and adapted flexibly to fully (often called attention focusing and the demands of specific contexts. Of course, shifting, and including cognitive distraction) individuals may also use EC in a manner and to inhibit or activate behavior willfully that is not adaptive. For example, EC can be (inhibitory control and activational control, used to achieve socially inappropriate goals, respectively). Some executive functioning such as a youth planning and carrying out a (EF) skills—e­specially effortful deployment series of well-r­egulated actions to humiliate of attention, integration of information a peer or to steal items to impress friends. Moreover, an effortful mode of regulation

Self-Regulation, Effortful Control, and Their Socioemotional Correlates 159 could be adaptive in the short run but not undercontrol and overcontrol are somewhat in the long run, or vice versa. In part, the similar to what the Block and Block (1980) adaptiveness of EC depends on the goals labeled as the extremes of ego control (i.e., that an individual is striving to achieve and ego undercontrol and overcontrol), which is the social context. defined as the “threshold or operating char- acteristic of an individual with regard to Control and Regulation: Is Control the expression or containment of impulses, Always Regulation? feelings, and desires” (p. 43). In the Blocks’ Like a number of other investigators (e.g., model, ego resiliency provides control and Block & Block, 1980), we have argued that modulation of impulses (i.e., ego control). well-­regulated people are neither overly con- trolled nor undercontrolled; rather, they can We consider EC, but not reactive control, respond flexibly to varying demands with to be part of volitional self-r­egulation. As a range of responses, depending on the cir- noted previously, we certainly acknowledge cumstances. Optimally self-r­egulated chil- that reactive control processes influence and dren can effortfully initiate or inhibit behav- even have controlling or “regulatory” effects iors when appropriate or when required to on individuals’ functioning, but we prefer achieve goals, but they can also be sponta- not to equate these effects with volitional neous (uncontrolled) when control is not self-r­egulation (see Compas, Connor-Smith, needed. Thus, their regulation is flexible and Saltzman, Thomsen, & Wadsworth, 2001, generally (although not always) adaptive. for a similar distinction when defining cop- ing; also see Carver, 2005, for a review Because of the importance of both will- of similar perspectives, including dual-­ fulness and flexibility in EC, it is useful to processing models). try to differentiate EC from less voluntary over- or undercontrolled processes. Some Rothbart and colleagues (e.g., Derryberry control is nearly always automatic and & Rothbart, 1997) linked what we have nonvolitional, and often it is emotionally labeled reactive overcontrol and undercon- based; such control is likely less flexible and trol processes to reactive emotional pro- often is less adaptive than volitional self-­ cesses—fear versus desire, hope, and relief, regulation. An example is children who have respectively—and their associated moti- been labeled “behaviorally inhibited.” They vational systems, defensive and appetitive, tend to be wary and overly constrained in respectively. Although we agree that these novel or stressful situations and have dif- associations between behavioral inhibi- ficulty modulating (e.g., relaxing) their tion or impulsivity and emotion/motivation inhibition (Kagan & Fox, 2006). Although exist, we wish to differentiate between emo- they may appear to be self-­regulated, their tional reactivity and the aspects of behavior inhibition or constraint is relatively involun- relevant to control that typically are associ- tary or so automatic that it is difficult for ated with emotion. It is quite possible that them to modulate effortfully (Eisenberg & children who display behavioral inhibition Morris, 2002). In addition, the impulse to (reactive overcontrol) do not experience fear approach people or inanimate objects in the or anxiety every time they display overly environment (sometimes called surgency; inhibited behavior. Such behavior, although Rothbart & Bates, 2006) often may be rela- probably originally based on fear or a reac- tively involuntary. For example, individuals tion to novelty (Kagan & Fox, 2006), seems may be “pulled” toward rewarding or posi- to become a habitual style of response to tive situations or stimuli, and this pull may novel or potentially stressful contexts. More- be nonvolitional; such people generally are over, highly inhibited children often look viewed as “impulsive.” We have labeled such controlled (restrained) in their behavior but overly inhibited and impulsive behavior as tend to be prone to fear and anxiety; thus, two aspects of reactive control—reactive control of overt behavior is not the same overcontrol and undercontrol—­based on the as control in regard to emotional reactivity distinction by Rothbart and Bates (2006) (although both are elements of temperamen- between effortful and reactive tempera- tal reactivity). Similarly, impulsive behavior mental processes. The constructs of reactive that may be linked with both desire/posi- tive affect (e.g., Rothbart, Ahadi, Hershey, & Fisher, 2001) and anger in some contexts

160 DEVELOPMENTAL CONSIDERATIONS might, in other contexts, not be linked to any We would argue that some responses that clear emotion. Thus, emotional reactivity of are relatively automatic usually can be voli- a particular sort is not strictly paired with tional if necessary, so the critical distinction impulsivity, and it is worthwhile to differ- is whether a capacity can easily become voli- entiate between impulsivity and emotional tional if needed for adaptation, not whether reactivity, and among associated motiva- it can be automatic. In addition, Ursache et tions in specific interactions. In support of al. seem to include what we have labeled as this argument, in factor analyses, disposi- reactive control in the construct of regulation tional negative emotionality tends to load (“the bottom-u­p, less volitional” aspects). on one factor, whereas reactive processes Although we prefer to confine the term self-­ that involve avoidance or approach behavior regulation to the aspects that can be voli- (e.g., shyness, impulsivity) tend to load on tional when needed, as a semantic way to a separate surgency factor (Rothbart et al., differentiate between volitional and reactive 2001). control processes, we agree that both pro- cesses occur. We also agree that the relation A disadvantage to invoking the distinction between more effortful and less volitional between effortful self-­regulation and reac- processes is bidirectional; moreover, reactive tive control is that it is difficult to catego- over- and undercontrol systems likely affect rize some aspects of regulation/control. For one another and are affected by emotions example, it is difficult to know whether the (Derryberry & Rothbart, 1997). Not com- child who is constrained in a new context is pletely in accord with Ursache et al. (2012), self-r­egulated (i.e., voluntarily constrained) we (Eisenberg & Zhou, in press) suggested or overly controlled (behaviorally inhibited). that aspects of EF related to inhibition (e.g., In our view, a major challenge for the field is in regard to thoughts or behavior and the differentiating between these two contribu- ability to shift attention) overlap with the tors to constrained behavior. construct of self-r­egulation (as well as the construct of EC), whereas other parts of Although impulsivity and EC are fairly EF, such as working memory, facilitate self-­ consistently negatively related (Aksan & regulation but are separate capacities from Kochanska, 2004; Valiente et al., 2003), the both self-­regulation and EC. two constructs have been found to load on different latent constructs (Eisenberg et al., In summary, numerous conceptual issues 2004, 2013; Valiente et al., 2003). The rela- are unresolved regarding the definition of tion between EC and reactive overcontrol self-r­ egulation and the relations among vari- is less consistent (e.g., Aksan & Kochan- ous processes that have controlling/regulat- ska, 2004; Spinrad et al., 2007), and there ing effects. We believe that more rather than is a need for research testing the relation less conceptual differentiation is useful, even between reactive over- and undercontrol. if in reality the various processes are inter- connected in complex and intricate ways. A comparison with Ursache, Blair, and Raver’s (2012) definition helps to highlight The Normative Development some of the subtle differences in definitions of Self‑Regulation/EC (some other definitions differ from ours Because EC is viewed as providing the build- in numerous, more obvious ways, includ- ing blocks for self-r­egulation, its develop- ing some of those mentioned earlier). They ment is of importance to many aspects of defined self-r­ egulation as children’s functioning. Children begin to regulate their own emotions and emotion-­ the primarily volitional management of related behavior in the first few years of arousal or activity in attention, emotion, or life, and there are improvements in regula- stress response systems in ways that facilitate tion skills across childhood and adolescence the use of EF abilities in the service of goal-­ (see Eisenberg, Spinrad, et al., 2010). Even directed actions. . . . EF is the top-down or voli- young children have been found to modulate tional component of self-r­egulation. . . .This their distress using a variety of methods. For volitional component, however, is reciprocally example, by 6 months of age, infants some- related to and dependent on bottom-­up, less volitional, and more automatic regulation of responses to the environment through atten- tion, emotion, and stress response systems. (p. 123)

Self-Regulation, Effortful Control, and Their Socioemotional Correlates 161 times reduce their own distress in response responses. For example, Diamond (1991) to novelty by looking away from the novel found that infants were able to inhibit pre- object and by using self-­soothing strate- dominant response tendencies when reach- gies (Crockenberg & Leerkes, 2004). Such ing for objects (e.g., move around an object), behaviors appear to be effective in reduc- an ability believed to involve the execution ing arousal in the first year of life because of intentional behavior, planning, and the they are associated with lower negativity in resistance of more automatic or reactive frustrating situations (Stifter & Braungart, action tendencies. 1995). Stifter and Braungart found that self-s­oothing was the most preferred regu- According to Posner and Rothbart (2000), latory strategy at both 5 and 20 months of a transition in the development of execu- age. In contrast, Mangelsdorf, Shapiro, and tive attention and the effortful inhibition of Marzolf (1995) found that (1) 6-month-old behavior can be seen around 30 months of infants tended to use gaze aversion as their age. Much of the relevant work involved the primary regulatory strategy, (2) 12-month- use of a Stroop-like task that required tod- olds engaged in more self-­soothing (e.g., dlers to switch attention and inhibit behav- thumb sucking and hair twirling) than did ior. Children show significant improvement 18-month-olds, and (3) 12- and 18-month- in performance on such a task between 24 old toddlers used more behavioral avoid- and 30 months of age and often perform ance and self-­distracting strategies than did with high accuracy by 36 to 38 months 6-month-olds. of age, and this ability has been positively related to adults’ ratings of EC (Rothbart There appears to be a decline in the use & Bates, 2006). In addition, with the matu- of self-s­ oothing between 24 and 48 months, ration of attentional mechanisms, the abil- coupled with the emergence of new and more ity to inhibit motor behavior effortfully complex use of objects and interactions to greatly improves between 22 and 44 months regulate emotional state (see Diener & Man- (Kochanska, Murray, & Harlan, 2000; gelsdorf, 1999, for a review). By 24 months Rothbart & Bates, 2006) and is fairly good of age, self-d­istraction may be the most by age 4 (Posner & Rothbart, 2000). Wil- common and successful regulatory strategy loughby, Wirth, Blair, and Family Life Proj- in fearful and frustrative situations (Grol- ect Investigators (2012) found substantial nick, Bridges, & Connell, 1996). Advances improvements in a latent factor of EF abili- in cognitive, sociocognitive, motor, and ties in children between ages 3 and 5 years, language development that occur between with the average child improving more than 2 and 5 years of age contribute to the emer- one standard deviation in EF skills over any gence of more sophisticated, diverse, and 12-month period. Moreover, there appears successful models of self-r­egulation and EF to be a further increase in the use of inter- skills (Garon, Bryson, & Smith, 2008), such nal mental or cognitive regulatory strategies that many children are relatively skilled at in the school years. Nonetheless, the capac- managing impulses by age 4 or 5 (Mischel ity for EC continues to improve in adoles- & Ayduk, 2011; Posner & Rothbart, 2000; cence and may even continue to develop at a Rothbart, 2011). slower pace into adulthood (Murphy, Eisen- berg, Fabes, Shepard, & Guthrie, 1999; Wil- Researchers who have focused on the liams, Ponesse, Schachar, Logan, & Tan- development of EC processes such as atten- nock, 1999; also see Eisenberg, Spinrad, et tion shifting and attention focusing or the al., 2010). ability to inhibit behavior voluntarily (rather than as specific reactions to distress or The Relation of EC frustration) have also noted developmental to Socioemotional Development changes in their use. Eight- to 10-month- Children who differ in their self-r­egulatory olds demonstrate some capacity to focus capacities are expected to differ in aspects their attention (Kochanska, Coy, Tjebkes, of socioemotional functioning involving & Husarek, 1998), and voluntary control self-r­egulation. Based on the distinction of attention increases somewhat between between reactive control and EC, Eisen- 9 and 18 months of age (Ruff & Rothbart, 1996). Around 12 months of age, infants develop the ability to inhibit predominant

162 DEVELOPMENTAL CONSIDERATIONS berg and Morris (2002) constructed a heu- all modes of EC and, in regard to reactive ristic model including descriptions of three control, are neither overcontrolled (inhib- styles of control—­overcontrolled, undercon- ited) nor undercontrolled (impulsive). These trolled, and optimally controlled—t­o gen- individuals are expected to be well adjusted, erate predictions regarding the relations of socially competent (including moral and various aspects of EC and reactive control prosocial), and resilient when faced with to children’s socioemotional adjustment and stress because they typically regulate their development. In this heuristic model, the behavior in a goal-d­ irected manner but can authors described different constellations of also be spontaneous and unconstrained. self-­regulation/reactive control expected to cluster together for some children and the Consistent with the heuristic model, EC expected socioemotional correlates of each and self-r­ egulation more generally have been constellation. related in predictable ways to a variety of children’s developmental outcomes. Next In the model, overcontrolled individuals we provide a selective summary of studies are those high in involuntary reactive con- on the relation of EC and/or other measures trol (e.g., high in behavioral inhibition) and of self-r­ egulation to some positive and nega- low in reactive undercontrol (impulsivity), tive outcomes expected to be affected by average in the ability to inhibit behavior EC and/or reactive control, including social effortfully as needed (inhibitory control), competence, social cognition, (mal)adjust- low in effortful attentional regulation (e.g., ment, school-r­elated outcomes, internaliza- the abilities to shift and focus attention tion of parental rules/demands and guilt, effortfully), and low in the ability to activate empathy/sympathy, and prosocial behavior. behavior effortfully as needed (i.e., effort- ful activational control). Volitional shifting Social Competence or focusing attention is expected to be use- In a number of studies with preschoolers, ful in reducing the negative emotions, such children, and adolescents, EC has been asso- as fear, associated with highly inhibited ciated with relatively high levels of social behavior. Individuals with this constellation competence (see Eisenberg, Smith, & Spin- are expected to be prone to internalizing rad, 2011, for a review). Children’s abilities problems (e.g., depression, anxiety, social to modulate their attentional resources likely withdrawal) and relatively low in social lead to optimal levels of emotional arousal in competence, especially if they are also pre- stressful contexts and, thus, accurate infor- disposed to experience negative emotions. mation processing linked to understanding These children might also lack the ability the causes of emotion, effective planning, to be relaxed, socially interactive, and spon- and action suited to the situation. Moreover, taneous in all but very familiar settings, the abilities to inhibit and activate behavior which also is likely to undermine their social effortfully allow children to implement or attractiveness and peer status. inhibit behaviors as is adaptive in a given context, hence facilitating socially appropri- In contrast, undercontrolled individuals ate responses. Consistent with these argu- are those high in reactive approach tenden- ments, EC (assessed with EF and/or delay cies (i.e., impulsive); low in reactive overcon- of gratification tasks) in the preschool years trol; and low in all types of EC, including has been found to predict parent-r­eported attentional, inhibitory, and activation con- social competence 6 months later (Lengua, trol. People with this style of control are Honorado, & Bush, 2007), as well as socio- predicted to be relatively low in social com- emotional competencies in adolescence and petence and prone to externalizing behav- adulthood (Mischel & Ayduk, 2011). Eisen- ior problems such as aggression, defiance, berg and colleagues (2011) have found sup- and antisocial behaviors (e.g., delinquency). port for the positive relation between EC These children would also be expected to and social competence in multiple longi- be low in empathy, prosocial behavior, and tudinal samples, and this relation has also moral development, because of deficits in been observed in other cultures (e.g., Hofer, social cognition and regulated behavior. Eisenberg, & Reiser, 2010; Zhou, Eisenberg, Wang, & Reiser, 2004). Finally, optimally regulated individuals—­ those who are most flexible and appropriate in their regulation—­are those fairly high in

Self-Regulation, Effortful Control, and Their Socioemotional Correlates 163 EC (or aspects thereof) frequently has awareness allows children to immediately been examined as a mediator of the rela- attach feelings to events, which can then tion of parenting to social competence. facilitate successful and appropriate regula- For example, Mintz, Hamre, and Hat- tion. Researchers have found that children field (2011) found that mothers’ reports of who are able to understand and communi- 54-month-olds’ inhibitory control, but not cate about emotions, and who know how to attentional focusing, partially mediated the manage emotions, are better able to regulate relation between early maternal sensitiv- themselves (Denham & Burton, 2003). ity (prompt and appropriate responses to the child, nonintrusive, positive regard for With the development of EF, children also child) and observed and teacher-­rated social improve on theory of mind (ToM) tasks, and relational competence, as well as social which involve an understanding of others’ and relational competence in first grade. internal mental representations. Researchers Similarly, EC has been found to mediate have identified inhibitory control as espe- the relation between maternal support and cially relevant to the development of ToM children’s social competence concurrently because it enables children to suppress irrele- (at 18 and 30 months), albeit not over time vant perspectives (Carlson, Moses, & Clax- (Spinrad et al., 2007). In addition, other ton, 2004). In a number of studies using a competencies appear to mediate the rela- variety of observed and reported measures, tion of self-r­egulation to social competence. inhibitory control has been positively related For example, the positive relation between to ToM skills in preschool and kindergar- EC and peer popularity appears to be medi- ten children (Blair & Razza, 2007; Carlson ated by ego-­resiliency (personality resil- et al., 2004). Moreover, inhibitory control iency; the ability to cope with and rebound appears to precede the development of ToM from stress), both in longitudinal work with in the preschool years (e.g., Flynn, 2007). In American 6- to 8-year-olds (Spinrad et al., brief, self-r­egulated children appear to have 2006) and concurrently with French ado- an advantage in regard to emotional under- lescents (Hofer et al., 2010). Children with standing and ToM skills. high EC are expected to be able to bounce back easily in new or stressful situations, Maladjustment and this resiliency is expected to improve Children with adjustment problems tend to their social competence. have difficulties managing and controlling their attention, behavior, and emotions in a Social Cognition willful and flexible manner. These difficul- Children who can avoid emotional over- ties can result in processing information and arousal and focus their attention are more modulating emotions and behaviors in ways likely than less regulated children to focus that hinder adaptive functioning (Eisenberg, on relevant information about emotions Spinrad, et al., 2010). In addition, because and the environment in social interactions children with high EC are probably better and, hence, develop a better understand- able to avoid negative interactions and mod- ing of emotions. In support of these ideas, ulate negative emotions such as sadness and preschoolers’ self-r­egulation has predicted anger, it is not surprising that low EC has their understanding of emotion 2 years later been related to maladjustment (for reviews, (Schultz, Izard, Ackerman, & Youngstrom, see Eisenberg et al., 2011; Eisenberg, Spin- 2001), and emotion understanding mediates rad, et al., 2010). the relation of self-r­egulation to adaptive social behavior (Izard, Schultz, Fine, Young- Consistent with our heuristic model, strom, & Ackerman, 1999). It is also likely many investigators have found negative rela- that emotion understanding fosters, as well tions between EC and externalizing prob- as stems from, emotion-­related regulation. lems in preschoolers, elementary schoolchil- Denham and Burton (2003) suggested that dren, and adolescents (Eisenberg, Spinrad, emotion understanding gives children labels et al., 2010). For example, Kochanska and for their internal feelings, which can then be Knaack (2003) reported negative associa- made conscious. Such conscious emotional tions between EC (assessed with a battery of behavioral measures) and externalizing problems, concurrently and across time,

164 DEVELOPMENTAL CONSIDERATIONS in young children (Kochanska & Knaack, were removed from the scales (Lemery, 2003). In multiple longitudinal samples, Essex, & Smider, 2002). Similarly, negative Eisenberg and colleagues have found nega- relations between EC and depressive symp- tive relations of teacher-r­eported, parent-­ toms have been reported for 11- to 18-year- reported, and observed measures of EC olds (Verstraeten, Vasey, Raes, & Bijttebier, with children’s externalizing behaviors 2008). Moreover, EC seems to moderate the (Eisenberg et al., 2011; Eisenberg, Spinrad, relation between affectivity and internaliz- et al., 2010). For example, Eisenberg et al. ing problems. In one study, adolescents’ low (2009) found that change in at-risk elemen- positive and high negative affectivity pre- tary schoolchildren’s externalizing over the dicted more depressive symptoms for youth 4 years—­whether pure externalizing prob- with low EC (Verstraeten et al., 2008); in lems or co-o­ccurring with internalizing another study, EC buffered the relation of problems—w­ as related to change in EC over fearfulness to adolescents’ internalizing the same period. Such associations appear to problems (Oldehinkel et al., 2007). continue throughout adolescence, including in European samples (Hofer et al., 2010). For School‑Related Outcomes example, Bakker, Ormel, Verhulst, and Old- The role of EC in school-r­elated outcomes ehinkel (2011) found that parent-­reported has aroused increasing interest (Eisenberg, EC at age 13½ appeared to buffer Dutch Valiente, & Eggum, 2010; Ursache et al., adolescents against externalizing problems 2012), and both direct and indirect rela- at age 16 in families with high adversity. tions between EC and school-r­elated out- comes have been predicted. Consistent with EC also appears to act as a mediator or our heuristic model, because children high moderator when predicting maladjustment in EC are expected to manage their atten- (e.g., Hofer et al., 2010; Spinrad et al., 2007). tion, behavior, and emotion flexibly, they For example, Hardaway, Wilson, Shaw, and are predicted to have an advantage when Dishion (2012) found that inhibitory control learning. Moreover, the effects of EC on at age 4 mediated the positive and negative academic outcomes are hypothesized to be relations, respectively, between chaos in mediated by children’s adjustment, quality the home and positive parental behavior at relationships with peers and teachers, and age 3 to externalizing behaviors at age 5½. school engagement (Eisenberg, Valiente, et Moreover, EC appears to reduce the positive al., 2010). relation of inconsistent discipline and physi- cal punishment to 8- to 12-year-olds’ exter- In fact, EC predicts a variety of school-­ nalizing problems (Lengua, 2008) and the related outcomes, both academic and social, relation of adolescents’ proneness to frustra- concurrently and longitudinally (Allan & tion to externalizing problems (Oldehinkel, Lonigan, 2011; Eisenberg, Valiente, et al., Hartman, Ferdinand, Verhulst, & Ormel, 2010; Ursache et al., 2012). In recent work, 2007). mediators of these relations have been exam- ined. For example, in a 1-year longitudinal Theoretically, as delineated in the afore- study with 3- to five-year-olds, Silva et al. mentioned model, reactive control and low (2011) found that a high-q­uality teacher–­ EC are related to internalizing problems: child relationship midyear (i.e., low conflict Children with internalizing problems are and high closeness) mediated the positive often overcontrolled and struggle with the relation between high EC measured early in flexible control of their emotions, cognition, the year and positive school attitudes mea- attention, and behaviors. Although the lit- sured at the end of the year with reports erature is not highly consistent (Eisenberg, of school liking and avoidance. In another Spinrad, et al., 2010), a number of studies study with multiple reporters, high social support a negative relation between EC and competence and low externalizing prob- internalizing problems. For example, moth- lems at ages 6–10 fully mediated the positive ers’ reports of preschoolers’ attention focus- relation between high EC at ages 4–8 and ing and inhibitory control have been mod- high academic achievement at ages 10–14 estly, negatively related with internalizing (Valiente et al., 2011). In brief, optimally problems at 5½ years old, even when items regulated children tend to have better school that overlapped between the constructs of self-r­egulation and internalizing problems

Self-Regulation, Effortful Control, and Their Socioemotional Correlates 165 outcomes, and these relations appear to be related self-r­egulation tend to experience both direct and indirect. sympathy (an other-o­riented response to another’s emotion or condition) rather than Internalization of Parental Rules/ personal distress (i.e., a self-­focused, aver- Demands and Guilt sive response to another’s emotional state or EC and related measures of emotion-­related condition). Personal distress is believed to be regulation or control have also been linked associated with empathic overarousal (pos- to a variety of measures of moral develop- sibly due to underregulation), which in turn ment, including compliance, internalization can lead to a self-focus and self-c­oncerned of rules, and conscience. Committed com- behavioral response (rather than concern for pliance—when children appear to comply others; see Eisenberg, Wentzel, & Harris, willingly and even enthusiastically with 1998). Consistent with this premise, Guthrie adults’ expectations/requests—is thought to et al. (1997) found that children rated high reflect children’s rudimentary internaliza- on EC exhibited greater facial sadness (but tion of their mothers’ commands. In addi- not distress)—presumed to reflect sympa- tion, when toddlers comply with their moth- thy/empathy—­during an empathy-­inducing ers’ rules when left alone, they are believed film than did less regulated peers. Children’s to have internalized the rules (Kochanska, postfilm reports of sadness and sympathy Coy, & Murray, 2001). during the film were also positively related to parents’ ratings of EC. Conversely, chil- Toddlers who exhibit more committed dren low in parent-r­ated EC were prone to compliance tend to have higher EC than experience personal distress (e.g., anxiety their less compliant peers (e.g., Kochanska and tension) during the film. et al., 2001). In addition, EC at 30 and 42 months predicts higher committed compli- Children’s EC (or aspects thereof) also has ance 1 year later, even after researchers con- been positively related to self- or other-r­ eport trol for earlier levels of committed compli- measures of dispositional empathy/sympathy ance (although these findings did not hold in children (Eisenberg et al., 1998; Panfile in models accounting for time-i­nvariant & Laible, 2012; Rothbart et al., 1994; also covariates; Spinrad et al., 2012). Infants’ see Eisenberg, Fabes, & Spinrad, 2006). In attention regulation (components of EC) an 8-year longitudinal study, relatively high predicts later committed compliance (Hill & levels of EC and growth in EC were related Braungart-R­ ieker, 2002; Kochanska et al., to high sympathy, although this pattern was 1998). In addition, Kochanksa and Knaack somewhat stronger for boys than for girls (2003) have found a positive relation, often (Eisenberg et al., 2007). Similar concurrent across time, between EC and the develop- relations have been found in studies in which ment of conscience (i.e., moral self, refrains adults reported on their own sympathy and from cheating on a task, and moral reason- regulation, although sometimes the associa- ing) during the toddler, preschool, and early tion was not significant until the effects of school years (Kochanska & Knaack, 2003). individual differences in negative emotion- ality were controlled (see Eisenberg et al., There are few data on the relation between 1998). In addition, EC has been negatively conscience and EC in schoolchildren. How- related to children’s and adults’ reports of ever, parent reports of children’s EC have personal distress (Bandstra, Chambers, been related to reports of their 7-year-olds’ McGrath, & Moore, 2011; see Eisenberg et guilt (Rothbart, Ahadi, & Hershey, 1994). al., 1998, 2006). More research is needed to clarify the rela- tion of EC to conscience and other aspects Consistent with the empirical relation of moral development in older elementary between EC and sympathy, adults’ ratings of school-a­ ge children and adolescents. elementary school children’s effortful atten- tional control and/or a behavioral measure Empathy/Sympathy of persistence have been correlated with and Prosocial Behavior peers’ ratings (e.g., Eisenberg et al., 1997) Eisenberg and Fabes (1992) hypothesized or teachers’ ratings (Diener & Kim, 2004) that individuals high in effortful emotion-­ of prosocial behavior. Thus, people who are skilled at regulating their emotion and behavior are not only more likely to feel con-

166 DEVELOPMENTAL CONSIDERATIONS cern for others but also they are relatively tor to performance on EF tasks. In contrast, likely to help them. working memory has generally not been considered a contributor to performance on Conclusions and Future Directions measures of EC, even though it likely plays an The study of emotion regulation and related important role for very young children and processes has progressed tremendously in in tasks with more complex working mem- the last decade. Yet definitional and meth- ory demands. Finally, researchers studying odological issues need to be addressed, and EF have made the distinction between hot there are gaps in the literature, as well as (i.e., emotional, for example, in delay tasks) new directions to exploit. and cold (i.e., cognitive, for example, in typi- cal EF measures such as flanker or Stroop Terminology tasks) aspects of EF, which may have impor- Currently, there is considerable variation in tant implications for emotion-r­elated self-­ what behavioral scientists mean when they regulation (e.g., Willoughby, Kupersmidt, use words such as regulation, self-­regulation, Voegler-­Lee, & Bryant, 2011). However, we and emotion regulation. The term regulation would argue that it is not EF that is hot or is used so broadly by different investigators cold, but that EF may function somewhat that it is often hard to understand what is differently in hot and cold contexts (Eisen- meant. Even though there may be somewhat berg & Zhou, in press; see also Kim, Nor- of a continuum, we suggest that researchers dling, Yoon, Boldt, & Kochanska, 2013). clearly distinguish volitional self-­regulatory processes (e.g., EC) from other, less voli- Researchers studying EF can also benefit tional processes that also “regulate” behav- from theoretical distinctions made in the ior (e.g., reactive control). It is also impor- EC literature. For example, temperament tant for researchers to distinguish between researchers have begun to study the devel- self-r­ egulation and external regulation. opment and correlates of activational con- trol, which is the ability to activate behavior Distinguishing between EC and EF willfully. This is an aspect of EC that has Despite the fact that EF and EC typically no corresponding construct in the field of EF have been examined in two separate litera- but merits further study due to the impor- tures, researchers have begun to recognize tance of up-­regulating emotional reactions that there is considerable overlap between in certain situations—­for example, show- these constructs (Eisenberg & Zhou, in ing gratitude when receiving a disappointing press). The integration of these two bodies gift—as well as its likely role in persistence of knowledge has great potential to be mutu- on unpleasant tasks and the achievement of ally informative. For example, considerable long-term goals. attention has been given to the factor struc- ture of EF (Garon et al., 2008), whereas sim- Design of Research Studies ilar work on EC is only now beginning (e.g., There is growing recognition of the bidi- Allan & Lonigan, 2011; Sulik et al., 2010). rectional nature of the relations between The conceptualization of EF as compris- temperament and environmental influences. ing interrelated but independent abilities—­ Researchers have come to understand that working memory, response inhibition, and although temperament has a biological set shifting—­has led to the understanding basis, it is also susceptible to environmental that task performance may be differentially influences. This may be especially true for related to these abilities and consequently to EC, which develops more slowly than more a more fine-g­rained analysis of what indi- emotionally reactive aspects of tempera- vidual tasks are measuring (Best & Miller, ment. Conversely, there is also evidence that 2010). For example, working memory, which children’s characteristics may affect parent- is considered an important aspect of EF but ing behavior (e.g., Eisenberg, Vidmar, et al., not EC, has been examined as a contribu- 2010). Longitudinal designs in which tem- poral hypotheses can be tested are becoming more common and can help to clarify the degree to which the early environment influ- ences children’s temperament and vice versa.

Self-Regulation, Effortful Control, and Their Socioemotional Correlates 167 Although such research designs have clear the relations among tasks and to measure strengths, even well-­designed correlational more accurately the underlying construct. studies cannot definitively test causal rela- However, a limitation of the majority of tions. There is still a need for experimental these measures is that performance quickly studies, such as intervention and prevention reaches a ceiling in early school-­age chil- trials that promote self-r­egulatory skills in dren. One challenge for EC researchers is to children, to determine how these skills are establish measures that can be used across causally related to self-­regulation processes a relatively wide age range to establish con- and other child outcomes. School-based tinuity in measurement. In particular, mea- prevention trials have demonstrated effects sures appropriate for adolescents and adults on self-r­egulation in low-i­ncome preschool- are needed, and it would be advantageous if ers (Diamond, Barnett, Thomas, & Munro, these measures also could be used with chil- 2007), but more research is needed to deter- dren. There is now evidence that it is pos- mine whether gains are maintained over long sible to develop tasks appropriate for a wide periods of time, the extent to which increases age range, including adolescents and adults, in EC are causally related to a number of dif- making it possible to measure more accu- ferent child outcomes, whether such effects rately the development of EC beyond early can be achieved in other age groups, and childhood (Lagattuta, Sayfan, & Monsour, whether the intervention effects generalize 2011; Tottenham, Hare, & Casey, 2011). to other populations (e.g., middle-i­ncome Although there has been recent progress children or children in other cultures). Simi- in this area, further work is still needed larly, experimental studies that modify par- because a single laboratory measure of EC enting behavior are needed to understand is relatively unreliable, and a battery of the causal effects of parenting on the devel- tasks is recommended in order to assess the opment of EC. construct more accurately (Kochanska & Knaack, 2003). Genetically informed research designs could also contribute to an understanding Psychophysiological methods have been of the influences on, and effects of, chil- underutilized in the existing research, even dren’s self-r­egulation. Not only can twin, though it is possible to assess many psycho- adoption, or sibling studies be used to assess physiological variables inexpensively and the role of genetics and environment in self-­ noninvasively. Of particular interest for regulation (see Rothbart & Bates, 2006), but the assessment of EC is respiratory sinus they also can be used to assess environmen- arrhythmia (RSA), a cardiac measure of tal influences, especially when genetic influ- parasympathetic nervous system function. ences are held constant. For example, one There is now a consistent body of evidence could examine factors related to differences linking resting RSA to EC (e.g., Marcovitch between identical twins (who are genetically et al., 2010). RSA reactivity—a­ change in the same) in EC. A relatively new and prom- RSA in response to task demands—h­ as also ising area of research is the use of measured been found in some studies to correlate with genes in studies of EC (Kochanska, Kim, EC and EF, but the exact nature of this rela- Barry, & Philibert, 2011; Voelker, Sheese, tion is still not clear. Although RSA reactiv- Rothbart, & Posner, 2009). More work is ity may also be a promising as a means of needed in this area to determine whether assessing emotion-­related regulation, fur- such effects are replicable, but such work ther research is needed to determine how may help investigators target interventions task demands relate to the interpretation of and understand the biological mechanisms this measure (Obradović, Bush, & Boyce, underlying the development of emotion and 2011). self-­regulation. Process Models of Emotion‑Related The Measurement of Emotion‑Related Self‑Regulation Self‑Regulation Finally, there is a need to understand how EC It is now common for studies of self-­ supports emotion-­related self-r­ egulation. EC regulation to include a number of behavioral likely plays a supporting role in a wide vari- measures, allowing researchers to study ety of emotion regulation strategies, includ-

168 DEVELOPMENTAL CONSIDERATIONS ing reappraisal, attentional strategies such as P. J., & Moore, C. (2011). The behavioural distraction or avoidance, choosing or modi- expression of empathy to others’ pain versus fying situations, and inhibiting emotionally others’ sadness in young children. PAIN, 152, motivated behavior. Although there is ample 1074–1082. evidence that EC is related to successful Best, J. R., & Miller, P. H. (2010). A develop- emotion regulation, we know relatively little mental perspective on executive function. about the extent to which the effectiveness Child Development, 81, 1641–1660. of specific emotion regulation strategies is Blair, C., & Razza, R. P. (2007). Relating effort- dependent on individual differences in EC, ful control, executive function, and false belief or whether specific aspects of EC such as understanding to emerging math and literacy activational control, inhibitory control, and ability in kindergarten. Child Development, attention focusing are differentially related 78, 647–663. to emotion-­related self-r­ egulation. Block, J. H., & Block, J. (1980). The role of ego-­ control and ego-r­esiliency in the organization Conclusion of behavior. In W. A. Collins (Ed.), Minne- sota Symposia on Child Psychology (Vol. 13, In summary, although the study of children’s pp. 39–101). Hillsdale, NJ: Erlbaum. self-­regulation is flourishing, there is much Carlson, S. M., Moses, L. J., & Claxton, L. J. yet to be explored and understood. A multi- (2004). Individual differences in executive method approach, influenced by conceptual functioning and theory of mind: An investiga- work in developmental, clinical, personality, tion of inhibitory control and planning ability. social, and neurophysiological psychology, Journal of Experimental Child Psychology, including greater integration of research and 87, 299–319. theory in the adult and child literatures, is Carver, C. S. (2005). Impulse and constraint: likely to be productive in future attempts to Perspectives from personality psychology, expand our knowledge. convergence with theory in other areas, and potential for integration. Personality and Acknowledgments Social Psychology Review, 9, 312–333. Compas, B. E., Connor-Smith, J. K., Saltzman, Work on this chapter was supported by grants H., Thomsen, A. H., & Wadsworth, M. E. from the National Institute of Child Health and (2001). Coping with stress during childhood Human Development to Nancy Eisenberg, Tracy and adolescence: Problems, progress, and L. Spinrad, and Carlos Valiente. potential in theory and research. Psychologi- cal Bulletin, 127, 87–127. References Crockenberg, S. C., & Leerkes, E. M. (2004). Infant and maternal behaviors regulate infant Aksan, N., & Kochanska, G. (2004). Links reactivity to novelty at 6 months. Develop- between systems of inhibition from infancy mental Psychology, 40, 1123–1132. to preschool years. Child Development, 75, Denham, S. A., & Burton, R. (2003). Social and 1477–1490. emotional prevention and intervention pro- gramming for preschoolers. New York: Klu- Allan, N. P., & Lonigan, C. J. (2011). Examin- wer Academic/Plenum Press. ing the dimensionality of effortful control in Derryberry, D., & Rothbart, M. K. (1997). Reac- preschool children and its relation to academic tive and effortful processes in the organization and socioemotional indicators. Developmen- of temperament. Development and Psychopa- tal Psychology, 47, 905–915. thology, 9, 633–652. Diamond, A. (1991). Neuropsychological insights Bakker, M. P., Ormel, J., Verhulst, F. C., & into the meaning of object concept develop- Oldehinkel, A. J. (2011). Adolescent family ment. In S. Carey & R. Gelman (Eds.), The adversity and mental health problems: The epigenesis of mind: Essays on biology and cog- role of adaptive self-r­egulation capacities: The nition (pp. 67–110). Hillsdale, NJ: Erlbaum. TRAILS study. Journal of Abnormal Child Diamond, A., Barnett, W. S., Thomas, J., & Psychology, 39, 341–350. Munro, S. (2007). The early years: Preschool program improves cognitive control. Science, Bandstra, N. F., Chambers, C. T., McGrath, 318, 1387–1388.

Self-Regulation, Effortful Control, and Their Socioemotional Correlates 169 Diener, M. L., & Kim, D. (2004). Maternal and (2004). The relations of effortful control and child predictors of preschool children’s social impulsivity to children’s resiliency and adjust- competence. Journal of Applied Developmen- ment. Child Development, 75, 25–46. tal Psychology, 25, 3–24. Eisenberg, N., Valiente, C., & Eggum, N. D. (2010). Self-rÂ

170 DEVELOPMENTAL CONSIDERATIONS Dishion, T. J. (2012). Family functioning and Lagattuta, K. H., Sayfan, L., & Monsour, M. externalizing behaviour among low-­income (2011). A new measure for assessing execu- children: Self-­regulation as a mediator. Infant tive function across a wide age range: Chil- and Child Development, 21, 67–84. dren and adults find happy–sad more difficult Hill, A. L., & Braungart-­Rieker, J. M. (2002). than day–night. Developmental Science, 14, Four-month attentional regulation and its pre- 481– 489. diction of three-year compliance. Infancy, 3, 261–273. Lemery, K. S., Essex, M. J., & Smider, N. A. Hofer, C., Eisenberg, N., & Reiser, M. (2010). (2002). Revealing the relation between tem- The role of socialization, effortful control, perament and behavior problem symptoms and ego resiliency in French adolescents’ social by eliminating measurement confounding: functioning. Journal of Research on Adoles- Expert ratings and factor analyses. Child cence, 20, 555–582. Development, 73, 867–882. Izard, C. E., Schultz, D., Fine, S. E., Youngstrom, E., & Ackerman, B. P. (1999). Temperament, Lengua, L. J. (2008). Anxiousness, frustration, cognitive ability, emotion knowledge, and and effortful control as moderators of the adaptive social behavior. Imagination, Cogni- relation between parenting and adjustment in tion and Personality, 19(4), 305–330. middle-­childhood. Social Development, 17, Kagan, J., & Fox, N. A. (2006). Biology, culture, 554–577. and temperamental biases. In W. Damon & R. M. Lerner (Series Eds.) & N. Eisenberg (Vol. Lengua, L. J., Honorado, E., & Bush, N. R. Ed.), Handbook of child psychology: Vol. 3. (2007). Contextual risk and parenting as pre- Social, emotional, and personality develop- dictors of effortful control and social compe- ment (6th ed., pp. 167–225). Hoboken, NJ: tence in preschool children. Journal of Applied Wiley. Developmental Psychology, 28, 40–55. Kim, S., Nordling, J. K., Yoon, J. E., Boldt, L. J., & Kochanska, G. (2013). Effortful control in Liew, J., Eisenberg, N., Spinrad, T. L., Eggum, “hot” and “cool” tasks differentially predicts N. D., Haugen, R. G., Kupfer, A., et al. (2011). children’s behavior problems and academic Physiological regulation and fearfulness as performance. Journal of Abnormal Child Psy- predictors of young children’s empathy-­related chology, 41(1), 43–56. reactions. Social Development, 20, 111–134. Kochanska, G., Coy, K. C., & Murray, K. T. (2001). The development of self-r­egulation in Mangelsdorf, S. C., Shapiro, J. R., & Marzolf, the first four years of life. Child Development, D. (1995). Developmental and temperamental 72, 1091–1111. differences in emotion regulation in infancy. Kochanska, G., Coy, K. C., Tjebkes, T. L., & Child Development, 66, 1817–1828. Husarek, S. J. (1998). Individual differences in emotionality in infancy. Child Development, Marcovitch, S., Leigh, J., Calkins, S. D., Leerks, 69, 375–390. E. M., O’Brien, M., & Blankson, A. N. (2010). Kochanska, G., Kim, S., Barry, R. A., & Philib- Moderate vagal withdrawal in 3.5-year-old ert, R. A. (2011). Children’s genotypes inter- children is associated with optimal perfor- act with maternal responsive care in predict- mance on executive function tasks. Develop- ing children’s competence: Diathesis–s­tress or mental Psychobiology, 52, 603–608. differential susceptibility? Development and Psychopathology, 23, 605–616. Mintz, T. M., Hamre, B. K., & Hatfield, B. E. Kochanska, G., & Knaack, A. (2003). Effort- (2011). The role of effortful control in medi- ful control as a personality characteristic of ating the association between maternal sen- young children: Antecedents, correlates, and sitivity and children’s social and relational consequences. Journal of Personality, 71, competence and problems in first grade. Early 1087–1112. Education and Development, 22, 360–387. Kochanska, G., Murray, K. T., & Harlan, E. T. (2000). Effortful control in early childhood: Mischel, W., & Ayduk, O. (2011). Willpower in Continuity and change, antecedents, and a cognitive-a­ffective processing system: The implications for social development. Develop- dynamics of delay of gratification. In K. D. mental Psychology, 36, 220–232. Vohs & R. F. Baumeister (Eds.), Handbook of self-­regulation: Research, theory, and applica- tions (2nd ed., pp. 83–105). New York: Guil- ford Press. Murphy, B. C., Eisenberg, N., Fabes, R. A., Shepard, S., & Guthrie, I. K. (1999). Consis- tency and change in children’s emotionality and regulation: A longitudinal study. Merill– Palmer Quarterly, 46, 413–444.

Self-Regulation, Effortful Control, and Their Socioemotional Correlates 171 Obradović, J., Bush, N. R., & Boyce, W. T. to children’s social competence: A longitudinal (2011). The interactive effect of marital con- study. Emotion, 6, 498–510. flict and stress reactivity on externalizing and Spinrad, T. L., Eisenberg, N., Gaertner, B., Popp, internalizing symptoms: The role of labora- T., Smith, C. L., Kupfer, A., et al. (2007). Rela- tory stressors. Development and Psychopa- tions of maternal socialization and toddlers’ thology, 23, 101–114. effortful control to children’s adjustment and social competence. Developmental Psychol- Oldehinkel, A. J., Hartman, C. A., Ferdinand, ogy, 43, 1170–1186. R. F., Verhulst, F. C., & Ormel, J. (2007). Spinrad, T. L., Eisenberg, N., Silva, K. M., Effortful control as modifier of the association Eggum, N. D., Reiser, M., Edwards, A., et al. between negative emotionality and adoles- (2012). Longitudinal relations among mater- cents’ mental health problems. Development nal behaviors, effortful control and young and Psychopathology, 19, 523–539. children’s committed compliance. Develop- mental Psychology, 48, 552–566. Panfile, T. M., & Laible, D. J. (2012). Attach- Stifter, C. A., & Braungart, J. M. (1995). The ment security and child’s empathy: The medi- regulation of negative reactivity in infancy: ating role of emotion regulation. Merrill– Function and development. Developmental Palmer Quarterly, 58, 1–21. Psychology, 31, 448–455. Sulik, M. J., Huerta, S., Zerr, A. A., Eisenberg, Posner, M. I., & Rothbart, M. K. (2000). Devel- N., Spinrad, T. L., Valiente, C., et al. (2010). oping mechanisms of self-­regulation. Devel- The factor structure of effortful control and opment and Psychopathology, 12, 427–441. measurement invariance across ethnicity and sex in a high-risk sample. Journal of Psychopa- Rothbart, M. K. (2011). Becoming who we are: thology and Behavioral Assessment, 32, 8–22. Tottenham, N., Hare, T. A., & Casey, B. J. Temperament and personality in develop- (2011). Behavioral assessment of emotion ment. New York: Guilford Press. discrimination, emotion regulation, and cog- Rothbart, M. K., Ahadi, S. A., & Hershey, K. nitive control in childhood, adolescence, and L. (1994). Temperament and social behavior adulthood. Frontiers in Developmental Psy- in childhood. Merill–Palmer Quarterly, 40, chology, 2, 1–9. 21–39. Ursache, A., Blair, C., & Raver, C. C. (2012). Rothbart, M. K., Ahadi, S. A., Hershey, K., & The promotion of self-­regulation as a means of Fisher, P. (2001). Investigations of tempera- enhancing school readiness and early achieve- ment at three to seven years: The children’s ment in children at risk for school failure. behavior questionnaire. Child Development, Child Development Perspectives, 6, 122–128. 72, 1394–1408. Valiente, C., Eisenberg, N., Haugen, R., Spinrad, Rothbart, M. K., & Bates, J. E. (2006). Tempera- T. L., Hofer, C., Liew, J., et al. (2011). Chil- ment. In W. Damon & R. M. Lerner (Series dren’s effortful control and academic achieve- Eds.) & N. Eisenberg (Vol. Ed.), Handbook ment: Mediation through social functioning. Early Education and Development, 22, 411– of child psychology: Vol. 3. Social, emo- 433. tional, and personality development (6th ed., Valiente, C., Eisenberg, N., Smith, C. L., Reiser, pp. 99–166). Hoboken, NJ: Wiley. M., Fabes, R. A., Losoya, S., et al. (2003). Ruff, H. A., & Rothbart, M. K. (1996). Atten- The relations of effortful control and reactive control to children’s externalizing problems: A tion in early development: Themes and varia- longitudinal assessment. Journal of Personal- tions. London: Oxford University Press. ity, 71, 1172–1196. Schultz, D., Izard, C. E., Ackerman, B. P., & Verstraeten, K., Vasey, M. W., Raes, F., & Bijt- Youngstrom, E. A. (2001). Emotion knowl- tebier, P. (2008). Temperament and risk for edge in economically disadvantaged children: depressive symptoms in adolescence: Media- Self-r­egulatory antecedents and relations to tion by rumination and moderation by effort- social difficulties and withdrawal. Develop- ful control. Journal of Abnormal Child Psy- ment and Psychopathology, 13, 53–67. chology, 37, 349–361. Silva, K. M., Spinrad, T. L., Eisenberg, N., Sulik, Voelker, P., Sheese, B. E., Rothbart, M. K., & M. J., Valiente, C., Huerta, S., et al. (2011). Posner, M. I. (2009). Variations in catechol- Relations of children’s effortful control and teacher–c­hild relationship quality to school attitudes in a low-­income sample. Early Edu- cation and Development, 22, 434–460. Spinrad, T. L., Eisenberg, N., Cumberland, A., Fabes, R. A., Valiente, C., Shepard, S., et al. (2006). Relation of emotion-­related regulation

172 DEVELOPMENTAL CONSIDERATIONS O-methyltransferase gene interact with par- Willoughby, M. T., Wirth, R. J., Blair, C. B., enting to influence attention in early develop- & Family Life Project Investigators. (2012). ment. Neuroscience, 164, 121–130. Executive function in early childhood: Longi- Williams, B. R., Ponesse, J. S., Schachar, R. J., tudinal measurement invariance and develop- Logan, G. D., & Tannock, R. (1999). Develop- mental change. Psychological Assessment, 24, ment of inhibitory control across the life span. 418 – 431. Developmental Psychology, 35, 205–213. Willoughby, M. T., Kupersmidt, J., Voegler-­Lee, Zhou, Q., Eisenberg, N., Wang, Y., & Reiser, M., & Bryant, D. (2011). Contributions of hot M. (2004). Chinese children’s effortful con- and cool self-­regulation to preschool disruptive trol and dispositional anger/frustration: Rela- behavior and academic achievement. Develop- tions to parenting styles and children’s social mental Neuropsychology, 36, 162–180. functioning. Developmental Psychology, 40, 352–366.

Chapter 11 Socialization of Emotion and Emotion Regulation in the Family Ross A. Thompson The development of emotion regulation is cess of considerable importance to parents one of the central goals of early socialization but also as one in which parents’ efforts are because of its importance to social compe- leading, sometimes significantly, children’s tence, academic achievement, and psycho- developing behavioral and neurobiological logical well-being (Saarni, 1999). It is also capacities. important to parent–­child relationships (Cas- sidy, 1994). Parents typically overestimate As the study of the development of emo- how much young children can manage their tion regulation has expanded, research on feelings, especially when their judgments are the socialization of emotion regulation has compared with what research has shown changed. Emotion regulation was conven- about the development of self-­regulatory tionally viewed in terms of the top-down ability (Newton & Thompson, 2010). In a imposition of cognitive or neurobiological recent national survey, for example, many control on basic emotional processes, and parents reported that children under the age socialization was perceived primarily as of 3 could control their emotions when frus- helping children acquire the understand- trated, and most thought that children of this ing, self-­control strategies, motivation, and age could competently share and take turns competence needed to enact these control with other children (Newton & Thompson, processes. In recent years, however, devel- 2010). This may help to explain why paren- opmental researchers have recognized that tal efforts to foster emotional self-­control in emotion regulation also involves bottom- offspring are so concerted and multifaceted. u­p processes, such as how basic emotion Their efforts include proactive strategies to appraisals originating in limbic structures manage the everyday emotional demands influence higher control processes (Thomp- that children face, direct interventions to son, 2011). From this view, therefore, the regulate children’s emotions, parents’ evalu- socialization of emotion regulation involves ations of children’s emotional displays, sanc- family processes that influence the growth tions on unregulated emotional displays, of emotional reactivity and the demands and conversation in which emotional control this imposes on emotional self-­control. This strategies are discussed and the benefits of chapter addresses both top-down and bot- self-m­ anagement are described. The social- tom-­up facets of the socialization of emo- ization of emotion regulation in the family is tion regulation, because each is important particularly interesting, therefore, as a pro- to how family experiences affect children’s emotional self-­control. 173

174 DEVELOPMENTAL CONSIDERATIONS The chapter begins with a definition of tion regulation usually alters the dynamics emotion regulation that highlights its rel- of emotion rather than changing its quality. evance to understanding these socialization In other words, individuals alter the inten- processes. Next, the discussion turns to sity, escalation (i.e., latency to onset and rise family processes that influence bottom-u­p time), or duration of an emotional response, aspects of emotion regulation, with partic- speed its recovery, or reduce or enhance the ular attention to experiences of stress that lability or range of emotional responding in can shape the neurobiology of emotion. In particular situations depending on the indi- the third section, the socialization of emo- vidual’s goals for that situation (Thompson, tion regulation as a top-down process is con- 1990). We usually think of emotionally well-­ sidered with respect to the diverse parental regulated people as those who are capable influences that affect how children inter- of altering how long, how intensely, or how pret and appraise their feelings, learn about quickly they feel as they do, rather than strategies of emotion management, achieve transforming the valence of emotion (e.g., competence in controlling their feelings, and changing anger into happiness). acquire cultural and gender expectations for emotion regulation (for a more extended Second, consistent with broader under- discussion of these influences, see Morris, standing of emotional development, emotion Silk, Steinberg, Myers, & Robinson, 2007; regulation can occur through the influence Thompson & Meyer, 2007). In the fourth of other people or through the individual’s section I integrate these perspectives with own efforts. Both extrinsic and intrinsic respect to the broader security of parent–­ processes of emotion regulation are impor- child attachment, then conclude the chapter tant. From birth, children’s emotional lives with final reflections. are managed by others, and this remains so throughout life. Moreover, extrinsic influ- Defining Emotion Regulation ences on emotion regulation can be either Definitions of emotion regulation are built facilitative (i.e., those that assist the child on broader conceptualizations of emotion in emotional self-c­ ontrol) or hindering (i.e., (see Gross, this volume). A developmental those that impair the child’s emotional self-­ approach to emotion regulation highlights control), with the latter occurring often in the changes that occur over time in multiple contexts of family disruption. In addition, components of emotion and their interac- extrinsic influences on emotion regulation tion, and how these emotion components can be either deliberate (i.e., intentionally unfold through biological maturation, social managing the child’s emotions) or unin- influences, the growth of self-r­eferential tended (i.e., not meant to have the effects on processes, and many other developmental children’s emotions that they do). influences (Thompson, 1990, 1994, 2011). The following definition of emotion regula- Third, consistent with a functionalist tion reflects this developmental approach: approach to emotion, strategies of emotion Emotion regulation consists of the extrinsic regulation are rarely inherently optimal or and intrinsic processes responsible for moni- maladaptive. Rather, emotion regulation toring, evaluating, and modifying emotional strategies must be evaluated in relation to reactions, especially their intensive and tem- the individual’s goals for the situation. This poral features, to accomplish one’s goals functionalist orientation is especially impor- (Thompson, 1994, pp. 27–28). Incorporated tant for developmental analysis. A toddler’s within this definition are several assump- petulant crying or an adolescent’s sullenness tions about emotion and emotion regula- may be intuitively interpreted by others as tion. revealing deficient skills in emotion regu- lation until one realizes that the toddler’s First, emotion regulation processes target crying causes parents to accede and the ado- positive as well as negative emotions and can lescent’s sullenness causes peers to provide entail diminishing, heightening, or simply support, each of which may be the child’s maintaining one’s current level of emotional goal (even if these emotional reactions have arousal in particular circumstances. Emo- other, negative consequences). It is impor- tant, however, to recognize that a person can have multiple goals in a situation requir- ing emotion regulation (e.g., managing fear

Socialization of Emotion and Emotion Regulation in the Family 175 and defending oneself when in the presence negative arousal that impairs the child’s of a bully). These alternative goals may con- emotion management. flict, which further complicates evaluating whether specific emotion regulation strate- Emotion regulation can be studied at gies are adaptive. multiple levels of analysis, from genetic and neurobiological processes to developing cog- This functionalist orientation is also nition and personality, to family processes important for understanding emotion regu- and overarching cultural values. A devel- latory processes relevant to developmental opmental analysis highlights that although psychopathology. Children with anxiety psychologists tend to regard “emotion regu- disorders are typically viewed as deficient lation” as if it were a single, coherent per- in emotion regulation, for example, but sonality construct or developmental process, their hypervigilance to threatening events, the growth of emotion regulation is actually fear-­oriented cognitions, and sensitivity to based on a multidimensional network of internal visceral cues of anxiety are part of loosely allied processes arising from within a constellation of self-r­egulatory strategies and outside the child. Developmental study for anticipating and avoiding encounters of emotion regulation requires that theo- with fear-­provoking events. In light of their rists keep the broader view in mind even as genetic vulnerability and family processes they are studying specific components of the that heighten risk for anxious pathology, growth of emotion regulation, such as its these emotion regulatory strategies may be socialization in the family. the most adaptive options available to the child (Thompson, 2000). Characterizing The developmental study of emotion such children simply as emotionally dysreg- regulation requires distinct methodologi- ulated seems oversimplified. To be sure, the cal strategies. Studies of emotion regulation same emotion regulatory strategies that pro- in adults or adolescents typically rely on vide immediate relief can exact long-term respondent self-r­eport, commonly through costs that make anxious children vulnerable questionnaires, to index individual differ- to continued pathology. This double-­edged ences in emotion self-r­egulation. Infants sword is typical of emotional regulatory and young children are not very informative processes for many forms of developmental reporters. Thus, developmental researchers psychopathology (Thompson & Goodman, must use other procedures. These include 2010). Understanding emotion regulation detailed observations of emotional reac- for children at risk requires comprehend- tions in carefully structured experimental ing the emotional goals the child is seeking situations, often with convergent behavioral to achieve and interpreting his or her self-­ and psychophysiological measures, some- regulatory efforts in light of these goals. times accompanied by the reports of moth- ers and other secondary sources concern- Fourth, consistent with a developmental ing the child’s emotional qualities. These analysis, emotion regulation includes under- methods are informative, but behavioral standing how people monitor and evaluate, measures (whether of infants or adults) are as well as modify, their emotions. Children’s also complicated by the following interpre- developing capacities for emotional self-­ tive difficulties: (1) behavior that reflects the awareness and for appraising their feelings influence of emotion regulatory processes in light of personal and cultural expectations is multidetermined; (2) emotional reactions are core features of developing emotion reg- and emotional regulatory influences are not ulation (Saarni, 1999). Emotion monitoring easily distinguished behaviorally; and (3) and evaluation function nonconsciously as the situational context can have a profound well as consciously to affect self-r­egulation. effect on children’s emotional reactions A preschooler arriving at child care after (Cole, Martin, & Dennis, 2004). Studying witnessing her parents’ morning argument emotion regulation in vivo in this manner may not be consciously monitoring how is thus conceptually and methodologically upset she is, for example, but the argument more difficult than enlisting self-­report. might affect her emotional self-­control when she is later provoked by a peer. In this situa- Adding further complexity to the study tion, antecedent emotional arousal creates a of emotion regulation is the functional- lowered emotional threshold for subsequent ist requirement of understanding the goals motivating self-r­egulatory efforts. Although

176 DEVELOPMENTAL CONSIDERATIONS these goals are usually assumed in studies influence higher-o­rder emotion processing of adults (e.g., diminishing negative affect (Barrett & Bar, 2009; Ochsner et al., 2009; and enhancing positive emotion), behavioral Woltering & Lewis, 2009). studies of emotion regulation with chil- dren require carefully designed assessments Consistent with the view that early expe- in which the goals for managing emotion riences shape brain architecture, research are either explicit or incorporated into the has shown that experience is important to design (e.g., observing children as they cope the functioning of higher and lower brain with a disappointing gift). In short, because regions relevant to emotion. The best evi- of the special methodological challenges it dence concerns the development of stress presents, developmental research into emo- responding, which involves neural systems tion regulation is not for the fainthearted. shared with emotion. Developmental stud- ies with humans and animals indicate that Family Influences on Emotional children growing up in adverse conditions Reactivity and Self‑Regulation develop altered autonomic nervous system Two conclusions from recent research have functioning that causes them to become changed how developmental researchers more sensitive to environmental demands, understand the growth of emotion regula- more likely to become biologically and tion (Thompson, 2011; Thompson, Lewis, emotionally reactive (or underresponsive) & Calkins, 2008). First, as noted earlier, to challenge, and less capable of adaptive neurobiological systems that are higher and self-­regulation (Loman & Gunnar, 2010; lower on the neuroaxis exert mutual regu- Lupien, McEwen, Gunnar, & Heim, 2009). latory influences related to emotion. “Top- These consequences have been documented down” regulatory control (e.g., projections in the limbic–h­ ypothalamic–p­ ituitary–­ from the prefrontal cortex to the amygdala) adrenocortical (LHPA) axis (which is rel- is important, but so is the “bottom-­up” evant to affect and stress responding), regulatory influence (e.g., from the limbic the parasympathetic system, and multiple system to higher cortical regions). Second, areas of the prefrontal cortex, indicating these higher and lower neurobiological sys- that early stress can have diverse and per- tems are shaped by the quality of early expe- vasive effects on emotion and emotion reg- rience, particularly family interactions. ulation (Blair, 2010; El-S­heikh & Erath, 2011). Importantly, early family experiences The view that emotion regulation involves have been the model for early adversity in bidirectional influences between higher and this research literature, related both to the lower systems is consistent with current work stresses directly imposed by parental behav- in the neuroscience of emotion. Neuroimag- ior (e.g., experiences of abuse or chronic ing studies show that responses to emotion neglect) and the failure of caregivers to buf- tasks are widely distributed throughout the fer the effects of other stresses on children. brain, including areas commonly regarded as Children with a history of abuse, for exam- relevant to affective activation (including the ple, exhibit multiple indications of difficulty amygdala, hypothalamus, brainstem, and in emotion processing relevant to their mal- striatum) and areas often viewed as impor- treatment: They are hypersensitive to adult tant to emotion regulation (including medial expressions of anger, have a lower threshold and lateral prefrontal cortex and anterior for detecting anger in maternal vocal expres- cingulate). Neuroimaging studies show that sions, and have more difficulty attentionally cortical and limbic areas are coactive in disengaging from perceived angry cues when responses to emotion probes (Kober et al., these responses are measured both behavior- 2008; Ochsner et al., 2009), and that lim- ally and neurobiologically (see Pollak, 2008, bic systems exert influence over cortical for a review). Hypersensitivity to anger and systems in affective arousal, as well as the threat is probably an adaptive emotional reverse. Amygdala activation is associated, regulatory strategy in contexts in which chil- for example, with enhanced perceptual sen- dren cannot avert an adult attack but may be sitivity to cues of danger, consistent with its capable of anticipating it and flee, avoid, or role in affective learning and appraisals that otherwise prepare for it. But it comes with a cost: Children who are hyperreactive to cues of threat are less likely to be capable

Socialization of Emotion and Emotion Regulation in the Family 177 of managing their arousal, as this research neurobiological growth in two areas: par- indicates. Moreover, outside the home, ents’ direct interventions to manage chil- hyperreactivity to potential threat is likely dren’s emotions, and effects of the broader to be a liability in dealing with other adults emotional climate of the family. and peers, whose social cues are more likely to be misinterpreted and imbued with hos- Parents’ Interventions tile intent (Thompson & Goodman, 2010). to Manage Emotion It is important to note that these deficien- From birth, parents exert considerable effort cies in emotion regulation are not pervasive, to manage the emotions of their infants and and abused children are not hyperreactive young children by soothing distress, provok- to cues of adult sadness or happiness (Pol- ing positive emotion, and allaying fear. They lak, 2008). Rather, children’s self-r­ egulatory proactively seek to make the everyday emo- problems are related to the circumstances tional demands on children manageable and of their family adversity, consistent with a predictable by organizing the child’s daily functionalist view of emotion regulation. routine, scheduling meals and naps, choos- ing appropriate child care arrangements, The research does not permit an evalu- and in other ways. Parents also seek to alter ation of the nature, severity, or chronicity children’s direct appraisals of emotionally of early stresses in the family that prompt evocative situations, such as distracting chil- these neurobiological and behavioral adap- dren away from potentially frightening or tations. There is some evidence that the distressing events, assisting in solving prob- association between family stress and its lems that children find frustrating, striving effects on behavioral functioning increment to alter the children’s interpretations of neg- in a dose–­response fashion (Repetti, Robles, atively arousing experiences (e.g., “It’s just a & Reynolds, 2011). Nor is there clear evi- game”), and through social referencing. By dence concerning the extent to which early directly intervening in these ways, parents experiences have privileged influence on the act as external regulators of young children’s growth of neurobiological and behavioral emotions. systems related to emotion and its regula- tion, or whether later experiences of adver- Beyond their immediate consequences, sity have comparable effects. There is reason these parental interventions have broader to believe, however, that early experiences consequences for early emotion and emo- may be especially important because of the tion regulation. With respect to parental plasticity of immature brain systems related soothing of infant distress, for example, to emotion and stress (Goldsmith, Pollak, & Lamb (1981) has argued that such distress Davidson, 2008). Taken together, however, relief sequences are easily learned by infants this research indicates that one way emo- and contribute to the emergence of rudi- tion regulation is socialized in the family mentary expectations that the parent’s is through children’s exposure to the emo- arrival will relieve distress. By 6 months of tional demands and stresses of family life, age, for example, distressed infants begin especially in the context of the (lack of) quieting in anticipation of the arrival of support that exists for children’s competent their mother when they hear her approach- management of the emotional demands on ing footsteps; infants also protest loudly if them. These experiences shape the func- mother approaches but does not pick them tioning of lower and higher neurobiological up to soothe them (Gekoski, Rovee-­Collier, systems associated with emotional arousal & Carulli-­Rabinowitz, 1983; Lamb & Mal- and its regulation, and equip children with kin, 1986). The parent’s sensitive respond- response propensities that are likely to influ- ing thus contributes to anticipatory soothing ence their emotional responding outside of (an early self-­regulatory response) before the the home. parent’s arrival. Similarly, the mother’s sen- sitivity during episodes of animated, face- Such a conclusion is consistent with con- to-face play with the baby contributes to the siderable behavioral research conducted in growth of early capacities for self-­regulation the context of less extreme family contexts. as the infant learns how to maintain man- In the following sections, I discuss develop- ageable arousal in the context of social play mental studies of emotion regulation that are consistent with the foregoing account of

178 DEVELOPMENTAL CONSIDERATIONS with the caregiver (Feldman, Greenbaum, & emotion, therefore, parental responses that Yirmiya, 1999; Gianino & Tronick, 1988). are critical of a child’s emotional behavior can contribute to the child’s risk of devel- Parents who provide support to very oping psychopathology involving emotion young children facing emotional challenges dysregulation. Risk is enhanced because have children who are more emotionally the parent’s critical demeanor is stressful, competent. Calkins and Johnson (1998) undermines opportunities for the child to found that 18-month-olds who became learn more adaptive forms of emotional cop- more distressed during frustration tasks had ing, and creates a more difficult family emo- mothers who were independently observed tional climate for troubled children. to be more interfering when interacting with their offspring, while children who could Direct parental interventions to manage use problem solving and distraction during children’s emotions decline in frequency in frustration had mothers who had earlier early childhood as young children acquire offered greater support, suggestions, and their own self-­regulatory strategies. How- encouragement. In another study, mothers ever, direct interventions remain an impor- who insisted that their toddlers approach tant source of extrinsic influence on emo- and confront potentially fearful objects in tion regulation throughout life as they are the laboratory had children who exhibited supplemented by other socialization influ- greater stress, as indexed by postsession cor- ences. Adults commonly rely on the support tisol levels (Nachmias, Gunnar, Mangels- of family and friends when facing emotional dorf, Parritz, & Buss, 1996). These findings stresses. Social support is thus a significant suggest that parental assistance is important buffer of psychosocial stress, and social iso- not only for managing immediate emotional lation is a risk factor for stress-­related prob- behavior but also for the development of lems (Thompson, Flood, & Goodvin, 2006). children’s emotional self-c­ontrol through the child’s expectation that distress is man- Emotional Climate of Family Life ageable and that adults can assist in manag- The emotional climate of family life can ing emotionally challenging situations. make emotion management easier or more difficult because of the emotional demands These studies focus primarily on paren- that children encounter in the home. As sug- tal interventions that buffer children’s emo- gested by the research on expressed emotion, tional arousal and support self-­regulatory when children must cope with frequent, capacity. As the research on stress indicates, intensive negative emotion from other fam- however, parental behavior may not always ily members, particularly when it is directed be so supportive, and when this occurs there at them, it can contribute to their hypersen- is greater risk for problems with emotional sitivity to threat and undermine emotion reactivity and self-r­egulation. Another regulation capacities. The family emotional research literature supporting this view con- climate is also relevant to emotion regula- cerns parental expressed emotion, which is tion because of the models of emotion reg- an index of parental attitudes of criticism ulation to which children are exposed and or emotional overinvolvement in the child’s how the family environment shapes chil- problems that can undermine competent dren’s developing schemas for emotion and emotional functioning (e.g., Hooley & Rich- its management (e.g., Are emotions threat- ters, 1995). Although expressed emotion has ening? Empowering? Uncontrollable?). been examined most extensively in clinical studies of schizophrenia, depression, and Another illustration of the importance bipolar disorder, because of its relevance to of the family emotional climate for devel- the maintenance or relapse of clinical symp- oping child emotional reactivity and self-­ tomatology, developmental studies have regulation is research on marital conflict found that expressed emotion is associated (Davies & Woitach, 2008). According to with the onset of psychological problems the Cummings and Davies (2010) emo- in children (Caspi et al., 2004). For exam- tional security hypothesis, children seek to ple, one study found expressed emotion to reestablish the emotional security they have be particularly prevalent in homes with a lost in families with marital conflict, even depressed parent (Rogosch, Cicchetti, & when parental conflict is not directed at Toth, 2004). In the context of expressed

Socialization of Emotion and Emotion Regulation in the Family 179 them. They do so by intervening in parental child or not, the frequency and intensity of arguments to end them, monitoring parental these emotions, and whether negative emo- moods to anticipate the outbreak of argu- tions are negative dominant (e.g., anger and ments, and otherwise striving to manage hostility) or negative submissive (e.g., sad- their emotions in a conflicted home environ- ness and distress), with the former more ment. As a consequence, they show height- likely to elicit the child’s fear or defensive- ened sensitivity to parental distress and ness. One can see how emotion regulatory anger, tend to become overinvolved in their skills might be enhanced (rather than under- parents’ emotional conflicts, have difficulty mined) by a child’s exposure to nonhostile managing the strong emotions that conflict negative emotions of moderate intensity in arouses in them (in a manner resembling the contexts in which negative feelings can be “emotional flooding” described by emotion safely expressed and managed, in contrast theorists), and exhibit signs of early devel- with exposure to a more hostile, threatening opment of internalizing problems. Research family emotional environment. derived from this view has found that grade school children who witness the most An important influence on the emotional intense marital conflict not only exhibit the climate of the family—w­ hich also affects greatest enmeshment in family conflict but how parents evaluate and respond to the also go to greater efforts to avoid conflict, emotions of offspring—­is parental beliefs while also showing more signs of internal- about emotion and its expression. These izing symptomatology (Davies & Forman, beliefs can be considered part of a parent’s 2002; see also Davies, Harold, Goeke- “meta-e­ motion philosophy,” which includes Morey, & Cummings, 2002). An impor- the adult’s awareness of her or his own emo- tant facet of the family emotional life is, of tions, an understanding and acceptance of course, parents’ emotional expressiveness the child’s emotions, and the parent’s own (Halberstadt, Crisp, & Eaton, 1999; Hal- approach to managing the child’s feelings berstadt & Eaton, 2003). A series of stud- (Gottman, Katz, & Hooven, 1997; Hooven, ies by Eisenberg and her colleagues showed Gottman, & Katz, 1995). Based on paren- that children’s social competence is affected tal interviews, Gottman and his colleagues by how mothers convey positive or nega- distinguish between “emotion coaching” tive feelings in the home—and this associa- and “emotion dismissing” parenting styles. tion is mediated by differences in children’s Emotion-­coaching parents are attentive self-­regulatory behavior (Eisenberg et al., to their own emotions and the child’s feel- 2001, 2003; Valiente, Fabes, Eisenberg, & ings, and do not believe that feelings should Spinrad, 2004). These findings suggest that be stifled. They consider the child’s emo- a family climate characterized by moderate tional expressions an occasion to validate to high amounts of positive emotion among the child’s feelings, and an opportunity to family members contributes to the growth of teach the child about emotions, expression, emotion regulation, perhaps because emo- and coping. Thus, emotion-c­oaching par- tional demands are manageable within the ents foster the growth of emotion regula- home and children are exposed to models of tion in offspring by offering warm support skillful emotion self-­regulation. and specific guidance for managing feel- ings. Dismissive parents tend to ignore their With respect to the influence of negative own emotions or belittle the importance of emotional expressiveness in the family, sev- emotions, and they may not constructively eral studies report that negative maternal attend to their children’s feelings. They view expressions of emotion are associated with emotions (especially negative ones) as poten- children’s lack of self-r­egulation and cop- tially harmful and believe that parents are ing, but others have found a positive asso- responsible for promptly subduing negative ciation (Eisenberg, Cumberland, & Spinrad, outbursts in offspring and teaching their 1998; Eisenberg et al., 2001, 2003; Valiente children that negative emotions are fleeting et al., 2004). Viewed in the context of the and unimportant. research on expressed emotion and mari- tal conflict, the effects of negative parental Several studies testing this provocative emotional expressiveness may depend on formulation have yielded supportive but whether emotions are directed toward the mixed results (Katz, Maliken, & Stettler, 2012). One study indicated that emotion

180 DEVELOPMENTAL CONSIDERATIONS coaching and emotion dismissing interact tive and attentional strategies to manage complexly as influences on children’s emo- emotional reactions accompanies the devel- tion regulation (Lunkenheimer, Shields, opment of these neurobiological control & Cortina, 2007; see also Gottman, Katz, processes. These top-down influences on & Hooven, 1996; Ramsden & Hubbard, emotion regulation highlight another avenue 2002). Research in my laboratory confirmed for socialization influences in the family. In that the association between maternal rep- direct and indirect ways, parents guide how resentations of emotion and children’s emo- children learn to appraise their feelings, tion regulation was mediated by mothers’ confront the demands of emotion regulation constructive emotion-r­elated socialization at home or in other social settings, acquire practices (Thompson, Virmani, Waters, specific skills for managing their feelings, Meyer, & Raikes, 2013). Mothers’ emotion- understand cultural expectations for emo- ally supportive representations (i.e., atten- tional self-c­ontrol, and represent emotion tion to her feelings; her focus on construc- processes relevant to self-­regulation. In con- tive emotion self-r­ egulation) were associated trast to the family influences (e.g., marital with children’s constructive emotion regu- conflict) that shape developing emotional lation (e.g., problem-f­ocused and emotion-­ reactivity and usually have unintended focused strategies) primarily through their effects on children’s emotion regulation, association with positive, supportive emo- these socialization influences are usually tion regulation socialization behaviors (i.e., deliberate. In coaching, conversation, and maternal problem-­solving encouragement, other modes of family interaction, parents emotion-f­ocused comfort, validation of chil- encourage the development of well-r­ egulated dren’s feelings, and positive family expres- emotion and children’s understanding, moti- sivity). vation, and competence to accomplish this. Taken together, the emotional climate of Because these socialization processes family life influences the development of extend throughout life and mediate cultural emotion regulation through both the emo- and gender differences in emotion manage- tional demands and supports it provides ment, individuals reach adulthood with that affect children’s developing emotional emotion self-r­ egulation skills that have been, reactivity. In studies of the development of to a large extent, socially constructed. In the stress reactivity, the effects of marital con- following sections, two general socialization flict, and the consequences of parental posi- influences on developing emotion regulation tive and negative emotional expressivity, it are considered: parental evaluations of chil- is clear that skills in emotion management dren’s emotions, and parent–­child conversa- develop, in part, in reaction to the emo- tion about emotion and its regulation. tional demands with which children must daily cope. The research on parental meta-­ Parental Evaluations emotion philosophy highlights one source of of Children’s Emotions variability in the family emotional climate, Emotion regulation can be facilitated or and it also identifies the basis for parental impaired by how others evaluate one’s feel- coaching of emotion regulatory strategies ings. Sympathetic, constructive responses in offspring. In the next section, I consider affirm that one’s feelings are justified and these and other kinds of top-down regula- provide social support in understanding and tory influences. advice that help people cope with difficult situations. However, denigrating, critical, or Family Influences on Emotion dismissive responses contain implicit mes- Self‑Regulatory Strategies sages that demean the appropriateness of the As prefrontal regions of the brain progres- feelings or their expression, or the compe- sively mature, these higher brain systems tence of the person feeling this way. Indeed, enable children to become more capable others’ dismissiveness can exacerbate the of deliberately managing their feelings, negative emotions that one is trying to man- impulses, and behavior. Growth in children’s age (in part by arousing further emotion), as capacities to understand and enact cogni- well as diminish opportunities for acquiring more adaptive modes of emotion regulation.