Handbook of emotion regulation ( PDFDrive )

Socialization of Emotion and Emotion Regulation in the Family 181 For children, these parental evaluations also words to say how you feel,” take a deep become internalized guides for how children breath, or think of something nice when judge their own emotional reactions. emotionally aroused. Research in my labo- ratory indicates that in conversation with Developmental studies indicate that their preschoolers about recent emotional children cope more adaptively with their events, mothers directly suggest strategies of emotions in immediate circumstances and emotional self-c­ ontrol and explicitly endorse acquire more constructive emotion regula- certain strategies over others (Thompson, tory capacities when parents respond accept- Waters, Meyer, Raikes, Jochem, & Virmani, ingly and supportively to their negative emo- 2009). They do so in light of their expecta- tional displays. By contrast, outcomes are tions for age-­appropriate emotional behav- more negative when parents are denigrating, ior in social settings (Cassano & Zeman, punitive, or dismissive, or when the child’s 2010). negative emotions elicit parents’ personal distress (for reviews, see Denham, 1998; Parent–child conversations have, however, Denham, Bassett, & Wyatt, 2007; Eisen- broader significance for the growth of emo- berg et al., 1998). In a socioeconomically tion regulation. Consistent with the work disadvantaged sample, for example, mothers of Gottman and his colleagues (1997) on who reported exerting more positive control parental coaching, conversation shapes chil- (using warmth and approval) over their sons dren’s broader representations of emotion. at age 1½ had children who were observed to Understanding the causes and consequences manage their negative emotions more con- of their feelings, comprehending gender structively (e.g., by using self-­distraction) at and cultural expectations for emotional age 3½ (Gilliom, Shaw, Beck, Schonberg, & expression, learning the social functions Lukon, 2002). Eisenberg, Fabes, and Mur- of emotional behavior, and other aspects phy (1996) found that mothers’ self-r­ eported of emotion understanding are advanced problem-s­olving responses to their grade by parent–­child conversation (Thompson, school children’s negative emotions were 2006a, 2006b). These broader aspects of associated with independent reports of their emotion understanding are important to children’s constructive coping with prob- developing an informed comprehension of lems (e.g., seeking support, problem solving, emotion regulation. As their conceptual and positive thinking), while mothers’ puni- skills mature, children begin to appreciate, tive and minimizing reactions to children’s especially through conversations with par- emotions were negatively associated with ents, that they can enlist internal constitu- children’s constructive coping and positively ents of emotion to manage external emotion associated with avoidant coping. Research (e.g., redirecting attention; cognitive reap- in my laboratory indicates that parents who praisal; mental distraction), that one need accepted and validated young children’s feel- not always show what one feels, and that ings during conversations about recent dif- mental processes (e.g., expectations, memo- ficult events had children who were more ries, thoughts) affect how one feels. willing to talk about these experiences and as a result more likely to learn about their Parent–child conversation has further sig- emotions and how to manage them (Waters, nificance for emotion regulation, moreover, Virmani, Thompson, Meyer, Raikes, & as a means of guiding children’s immediate Jochem, 2010). These studies indicate that appraisals of events. By managing infor- how parents respond to children’s emotions, mation the child receives about potentially and the behaviors that result, predict chil- stressful events (e.g., describing an antici- dren’s emotion-­related coping in later assess- pated dental visit as “teeth tickling”) or ments. introducing alternative interpretations of an event (e.g., eliciting sympathy for a physi- Parent–Child Conversation and cally challenged person the child fears or Developing Emotion Representations finds amusing), parents influence emotion-­ From an early age, parents actively coach related appraisals that facilitate emotion children on strategies of emotion regula- regulation. In an experimental probe, Mor- tion when they encourage offspring to “use ris et al. (2011) found that mothers’ use of attention refocusing and cognitive reframing was associated with less sadness and anger

182 DEVELOPMENTAL CONSIDERATIONS in children who received a disappointing the growth of emotion regulation (Cassidy, gift in her presence (see Kliewer, Fearnow, 1994; Thompson, 1994; see also Shaver & & Miller, 1996, for similar results with Mikulincer, this volume). Young children mothers and fathers). Comparable to how in secure relationships have more sensitive parent–­child conversations influence mem- mothers who are accepting of children’s ory through attentional redirection, there- positive and negative feelings, and open to fore, these conversations can influence emo- talking about intense, disturbing, or con- tion through reorienting appraisal processes fusing feelings with them. Consequently, (Thompson, 2006b). like the offspring of emotion-­coaching par- ents, securely attached children are likely It is important to note that conversations to become more emotionally self-aware, with peers and siblings are also important acquire deeper emotion understanding, and catalysts to the growth of emotion regula- develop a more flexible capacity to manage tion in childhood. Young children talk about their emotions appropriate to circumstances. their feelings more frequently with friends Moreover, the security of the parent–c­hild and siblings than they do with their moth- relationship provides a continuing resource ers, and these conversations also contrib- of support on which the child can rely. By ute to developing emotional understanding contrast, young children in insecure rela- (Brown, Donelan-­McCall, & Dunn, 1996). tionships have mothers who are less sensitive Peer conversations are important; there is and more inconsistently responsive to their evidence that adaptive strategies of emotion feelings, and less likely to be comfortable regulation with peers are not the same as talking with them about difficult emotional those with family members and adult inter- experiences. These children are likely to actions (Thompson & Waters, 2010). Vent- have a more limited understanding of emo- ing distress does not typically elicit as much tion and to become more easily emotionally support from peers as from parents, for dysregulated, especially in stressful circum- example. Thus, as children travel between stances, because of the lack of support in the different social settings, they must develop parent–c­ hild relationship. competence in applying different norms for emotional behavior and accepted strategies There is research evidence in support of for emotion regulation. this view. In a longitudinal study of chil- dren’s first 3 years, Kochanska (2001) An Integrative Perspective: reported that over time, insecurely attached Parent–Child Attachment Security children, compared with secure children, The preceding discussion focused on imme- exhibited progressively greater fear and/or diate and longer-­term socialization processes anger, and diminished joy, in standardized by which children become competent at assessments. Even by age 1, the mothers of managing their emotions. But what happens secure infants commented on both positive and who does it are both important in emo- and negative emotions when interacting tion socialization. Most of the socialization with their children, while the mothers of influences on emotion regulation discussed insecurely attached infants rarely discussed in this chapter occur in a relational context, their feelings or commented primarily about and the quality of the parent–c­ hild relation- negative emotions (Goldberg, MacKay-­ ship colors the influence of parents’ direct Soroka, & Rochester, 1994). By early child- interventions, appraisals of children’s emo- hood, securely attached preschoolers talked tions, coaching, and conversations about more about emotions in everyday conversa- emotion and its regulation. In this section, tions with their mothers, and their mothers I discuss the significance of the security of were more richly elaborative than mothers the parent–c­hild relationship, and in doing of insecurely attached preschoolers in their so reintegrate the bottom-u­ p and top-down discussions of emotion with them. This may processes discussed earlier. help to explain why secure children are more advanced in emotion understanding (Laible According to attachment researchers, & Thompson, 1998; Raikes & Thompson, differences in the security of child–p­arent 2006). There is also evidence that children attachment may be especially significant for in secure relationships are better at manag- ing negative emotions. Gilliom and his col-

Socialization of Emotion and Emotion Regulation in the Family 183 leagues (2002) reported that boys who were self-­regulation conform to the functional securely attached at age 1½ were observed requirements of managing difficult every- to use more constructive anger manage- day emotional demands at home. Viewed in ment strategies at age 3½. In a study of the terms of top-down influences, family pro- responses of 18-month-olds to moderate cesses also shape children’s understanding stressors, Nachmias and colleagues (1996) of emotion, and their motivation and com- reported that postsession cortisol elevations, petence to manage their feelings. These pro- indicating stressful responding, were found cesses contribute to significant individual only in temperamentally inhibited toddlers differences in children’s emotional apprais- in insecure relationships with their mothers. als, goals for emotion regulation, and strate- For inhibited toddlers in secure relationships, gies for self-­control. the mother’s presence helped to buffer the physiological effects of challenging events. Although study of the socialization of In another study, Contreras, Kerns, Weimer, emotion regulation has long been a topic of Gentzler, and Tomich (2000) reported that vigorous research interest, a considerable by middle childhood, attachment security research agenda remains. Continuing work was significantly associated with children’s in this field is justified not only for theoreti- constructive coping with stress, and the cal reasons, but also for its importance to extent of coping mediated the association understanding the challenges faced by chil- between attachment and children’s peer dren at risk for problems of emotion-­related competence. Research in our laboratory has psychopathology. also shown that mothers in secure relation- ships with their children are more accurate References in judging what their children are feeling and potentially are thus more capable of Barrett, L. F., & Bar, M. (2009). See it with providing helpful guidance about managing feeling: Affective predictions during object those feelings (Waters et al., 2010). perception. Philosophical Transactions of the Royal Society B: Biological Sciences, 364, These findings indicate that the rela- 1325–1334. tional context in which emotion regulation develops is important not only for the spe- Blair, C. (2010). Stress and the development of cific ways that parents respond to children’s self-r­ egulation in context. Child Development feelings but also for the relational support Perspectives, 4, 181–188. that shapes the growth of emotion self-­ regulation. Brown, J. R., Donelan-­McCall, N., & Dunn, J. (1996). Why talk about mental states?: The Conclusions significance of children’s conversations with In the broadest sense, the research surveyed friends, siblings, and mothers. Child Develop- in this chapter confirms how significantly ment, 67, 836–849. social influences shape the growth of emo- tional experience and emotion regulation. Calkins, S. D., & Johnson, M. C. (1998). Tod- This literature underscores the multifaceted dler regulation of distress to frustrating ways that socialization processes in the fam- events: Temperamental and maternal corre- ily affect children’s emotional reactivity and lates. Infant Behavior and Development, 21, self-­regulation. 379–395. Viewed in terms of bottom-­up influences, Caspi, A., Moffitt, T. E., Morgan, J., Rutter, family processes shape children’s emotional M., Taylor, A., Arseneault, L., et al. (2004). reactivity through the everyday emotional Maternal expressed emotion predicts chil- demands to which children must adapt, dren’s antisocial behavior problems: Using anticipate, and cope. These influences are monozygotic-­twin differences to identify envi- most apparent in studies of at-risk children, ronmental effects on behavioral development. for whom the challenges of family dysfunc- Developmental Psychology, 40, 149–161. tion, marital conflict, and even maltreatment underscore how emotional development and Cassano, M. C., & Zeman, J. L. (2010). Parental socialization of sadness regulation in middle childhood: The role of expectations and gen- der. Developmental Psychology, 46, 1214– 1226. Cassidy, J. (1994). Emotion regulation: Influ-

184 DEVELOPMENTAL CONSIDERATIONS ences of attachment relationships [Special Eisenberg, N., Valiente, C., Morris, A. S., Fabes, issue]. Monographs of the Society for Research R. A., Cumberland, A., Reiser, M., et al. in Child Development, 59(2–3, Serial No. (2003). Longitudinal relations among parental 240), 228–249. emotional expressivity, children’s regulation, Cole, P., Martin, S., & Dennis, T. (2004). Emo- and quality of socioemotional functioning. tion regulation as a scientific construct: Meth- Developmental Psychology, 39, 3–19. odological challenges and directions for child development research. Child Development, El-S­heikh, M., & Erath, S. A. (2011). Family 75, 317–333. conflict, autonomic nervous system function- Contreras, J. M., Kerns, K. A., Weimer, B. L., ing, and child adaptation: State of the science Gentzler, A. L., & Tomich, P. L. (2000). Emo- and future directions. Development and Psy- tion regulation as a mediator of associations chopathology, 23, 703–721. between mother–­child attachment and peer relationships in middle childhood. Journal of Feldman, R., Greenbaum, C. W., & Yirmiya, N. Family Psychology, 14, 111–124. (1999). Mother–­infant affect synchrony as an Cummings, E. M., & Davis, P. T. (2010). Chil- antecedent of the emergence of self-c­ontrol. dren and marital conflict: An emotional secu- Developmental Psychology, 35, 223–231. rity perspective. New York: Guilford Press. Davies, P. T., & Forman, E. M. (2002). Chil- Gekoski, M., Rovee-­Collier, C., & Carulli-­ dren’s patterns of preserving emotional secu- Rabinowitz, V. (1983). A longitudinal analy- rity in the interparental subsystem. Child sis of inhibition of infant distress: The origins Development, 73, 1880–1903. of social expectations? Infant Behavior and Davies, P. T., Harold, G. T., Goeke-Morey, M. Development, 6, 339–351. C., & Cummings, E. M. (2002). Child emo- tional security and interparental conflict. Gianino, A., & Tronick, E. (1988). The mutual Monographs of the Society for Research in regulation model: The infant’s self and inter- Child Development, 67(Serial No. 270), active regulation and coping and defensive 1–127. capacities. In T. Field, P. McCabe, & N. Davies, P. T., & Woitach, M. J. (2008). Chil- Schneiderman (Eds.), Stress and coping (Vol. dren’s emotional security in the interparental 2, pp. 47–68). Hillsdale, NJ: Erlbaum. relationship. Current Directions in Psycho- logical Science, 17, 269–274. Gilliom, M., Shaw, D. S., Beck, J. E., Schonberg, Denham, S. (1998). Emotional development in M. A., & Lukon, J. L. (2002). Anger regula- young children. New York: Guilford Press. tion in disadvantaged preschool boys: Strat- Denham, S., Bassett, H. H., & Wyatt, T. (2007). egies, antecedents, and the development of The socialization of emotional competence. In self-c­ ontrol. Developmental Psychology, 38, J. Grusec & P. Hastings (Eds.), Handbook of 222–235. socialization (pp. 614–637). New York: Guil- ford Press. Goldberg, S., MacKay-S­ oroka, S., & Rochester, Eisenberg, N., Cumberland, A., & Spinrad, T. L. M. (1994). Affect, attachment, and maternal (1998). Parental socialization of emotion. Psy- responsiveness. Infant Behavior and Develop- chological Inquiry, 9, 241–273. ment, 17, 335–339. Eisenberg, N., Fabes, R. A., & Murphy, B. C. (1996). Parents’ reactions to children’s nega- Goldsmith, H. H., Pollak, S. D., & Davidson, tive emotions: Relations to children’s social R. J. (2008). Developmental neuroscience per- competence and comforting behavior. Child spectives on emotion regulation. Child Devel- Development, 67, 2227–2247. opment Perspectives, 2, 132–140. Eisenberg, N., Gershoff, E. T., Fabes, R. A., Shepard, S. A., Cumberland, A. J., Losoya, S. Gottman, J. M., Katz, L. F., & Hooven, C. H., et al. (2001). Mothers’ emotional expres- (1996). Parental meta-e­motion philosophy sivity and children’s behavior problems and and the emotional life of families: Theoretical social competence: Mediation through chil- models and preliminary data. Journal of Fam- dren’s regulation. Developmental Psychology, ily Psychology, 10, 243–268. 37, 475–490. Gottman, J. M., Katz, L. F., & Hooven, C. (1997). Meta-e­ motion: How families commu- nicate emotionally. Mahwah, NJ: Erlbaum. Gross, J. J. (this volume, 2014). Emotion regula- tion: Conceptual and empirical foundations. In J. J. Gross (Ed.), Handbook of emotion regulation (2nd ed., pp. 3–20). New York: Guilford Press. Halberstadt, A. G., Crisp, V. W., & Eaton, K. L.

Socialization of Emotion and Emotion Regulation in the Family 185 (1999). Family expressiveness: A retrospective Lunkenheimer, E. S., Shields, A. M., & Cortina, and new directions for research. In P. Philip- K. S. (2007). Parental emotion coaching and pot & R. S. Feldman (Eds.), The social context dismissing in family interaction. Social Devel- of nonverbal behavior (pp. 109–155). New opment, 16, 232–248. York: Cambridge University Press. Halberstadt, A. G., & Eaton, K. L. (2003). A Lupien, S. J., McEwen, B. S., Gunnar, M. R., & meta-a­nalysis of family expressiveness and Heim, C. (2009). Effects of stress through- children’s emotion expressiveness and under- out the lifespan on the brain, behaviour and standing. Marriage and Family Review, 34, cognition. Nature Reviews Neuroscience, 10, 35 – 62 . 434 – 445. Hooley, J. M., & Richters, J. E. (1995). Expressed emotion: A developmental perspective. In D. Morris, A. S., Silk, J. S., Morris, M. D. S., Stein- Cicchetti & S. Toth (Eds.), Emotion, cogni- berg, L., Aucoin, K. J., & Keyes, A. W. (2011). tion, and representation (pp. 133–166). Roch- The influence of mother–­child emotion regu- ester, NY: University of Rochester Press. lation strategies on children’s expression of Hooven, C., Gottman, J. M., & Katz, L. F. anger and sadness. Developmental Psychol- (1995). Parental meta-e­motion structure pre- ogy, 47, 213–225. dicts family and child outcomes. Cognition and Emotion, 9, 229–264. Morris, A. S., Silk, J. S., Steinberg, L., Myers, S. Katz, L. F., Maliken, A. C., & Stettler, N. M. S., & Robinson, L. R. (2007). The role of the (2012). Parental meta-­emotion philosophy: A family context in the development of emotion review of research and theoretical framework. regulation. Social Development, 16, 361–388. Child Development Perspectives, 6, 417–422. Kliewer, W., Fearnow, M. D., & Miller, P. A. Nachmias, M., Gunnar, M., Mangelsdorf, S., (1996). Coping socialization in middle child- Parritz, R. H., & Buss, K. (1996). Behavioral hood: Tests of maternal and paternal influ- inhibition and stress reactivity: The moderat- ences. Child Development, 67, 2339–2357. ing role of attachment security. Child Devel- Kober, H., Barrett, L. F., Joseph, J., Bliss-­ opment, 67, 508–522. Moreau, E., Lindquist, K., & Wager, T. D. (2008). Functional grouping and cortical–­ Newton, E. K., & Thompson, R. A. (2010). subcortical interactions in emotion: A meta-­ Parents’ views of early social and emotional analysis of neuroimaging studies. NeuroIm- development: More and less than meets the age, 42, 998–1031. eye. Zero to Three, 31, 10–15. Kochanska, G. (2001). Emotional development in children with different attachment histories: Ochsner, K. N., Ray, R. R., Hughes, B., McRae, The first three years. Child Development, 72, K., Cooper, J. C., Weber, J., et al. (2009). 474 – 490. Bottom-up and top-down processes in emo- Laible, D. J., & Thompson, R. A. (1998). Attach- tion generation: Common and distinct neu- ment and emotional understanding in pre- ral mechanisms. Psychological Science, 20, school children. Developmental Psychology, 1322–1331. 34(5), 1038–1045. Lamb, M. (1981). Developing trust and per- Pollak, S. D. (2008). Mechanisms linking early ceived effectance in infancy. In L. Lipsitt (Ed.), experience and the emergence of emotions: Advances in infancy research (Vol. 1, pp. 101– Illustrations from the study of maltreated chil- 127). New York: Ablex. dren. Current Directions in Psychological Sci- Lamb, M., & Malkin, C. (1986). The develop- ence, 17, 370–375. ment of social expectations in distress–r­elief sequences: A longitudinal study. International Raikes, H. A., & Thompson, R. A. (2006). Fam- Journal of Behavioral Development, 9, 235– ily emotional climate, attachment security, 249. and young children’s emotion understand- Loman, M. M., & Gunnar, M. R. (2010). Early ing in a high-risk sample. British Journal of experience and the development of stress reac- Developmental Psychology, 24, 989–1004. tivity and regulation in children. Neuroscience and Biobehavioral Reviews, 34, 867–876. Ramsden, S. R., & Hubbard, J. A. (2002). Fam- ily expressiveness and parental emotion coach- ing: Their role in children’s emotion regulation and aggression. Journal of Abnormal Child Psychology, 30, 657–667. Repetti, R. L., Robles, T. F., & Reynolds, B. (2011). Allostatic processes in the family. Development and Psychopathology, 23, 921– 938. Rogosch, F., Cicchetti, D., & Toth, S. (2004).

186 DEVELOPMENTAL CONSIDERATIONS Expressed emotion in multiple subsystems of (Eds.), Emotion regulation and psychopathol- the families of toddlers with depressed moth- ogy (pp. 38–58). New York: Guilford Press. ers. Development and Psychopathology, 16, Thompson, R. A., Lewis, M., & Calkins, S. D. 689–709. (2008). Reassessing emotion regulation. Child Saarni, C. (1999). The development of emo- Development Perspectives, 2(3), 124–131. tional competence. New York: Guilford Press. Thompson, R. A., & Meyer, S. (2007). The Shaver, P. R., & Mikulincer, M. (this volume, socialization of emotion regulation in the fam- 2014). Adult attachment and emotion regula- ily. In J. J. Gross (Ed.), Handbook of emotion tion. In J. J. Gross (Ed.), Handbook of emo- regulation (pp. 249–268). New York: Guilford tion regulation (2nd ed., pp. 237–250). New Press. York: Guilford Press. Thompson, R. A., Virmani, E., Waters, S. F., Thompson, R. A. (1990). Emotion and self-­ Meyer, S., & Raikes, A. (2013). The develop- regulation. In R. A. Thompson (Ed.), Socio- ment of emotion self-r­ egulation: The whole and the sum of the parts. In K. Barrett, N. A. Fox, emotional development: Nebraska Sympo- G. A. Morgan, D. J. Fidler, & L. A. Daunhauer sium on Motivation (Vol. 36, pp. 383–483). (Eds.), Handbook of self-r­egulatory processes Lincoln: University of Nebraska Press. in development (pp. 5–26). New York: Taylor Thompson, R. A. (1994). Emotion regulation: & Francis. A theme in search of definition. Monographs Thompson, R. A., & Waters, S. F. (2010). The development of emotion regulation: Parent and of the Society for Research in Child Develop- peer influences. In R. Sanchez-A­ ragon (Ed.), ment, 59(2–3, Serial No. 240), 25–52. Thompson, R. A. (2000). Childhood anxiety dis- Emotion regulation: A crossing from culture orders from the perspective of emotion regula- tion and attachment. In M. W. Vasey & M. to the development of personal relationships R. Dadds (Eds.), The developmental psycho- (pp. 125–157). Mexico City: National Auton- pathology of anxiety (pp. 160–182). Oxford, omous University of Mexico. UK: Oxford University Press. Thompson, R. A., Waters, S. F., Meyer, S. C., Thompson, R. A. (2006a). The development of Raikes, A., Jochem, R. A., & Virmani, E. the person: Social understanding, relation- A. (2009, April). Conversational catalysts to ships, self, conscience. In W. Damon & R. M. developing understanding of emotion regula- Lerner (Eds.), N. Eisenberg (Vol. Ed.), Hand- tion. In L. J. Levine (Chair), How kids keep book of child psychology: Vol. 3. Social, emo- their cool: Pathways to effective emotion reg- tional, and personality development (6th ed., ulation. Symposium presented at the biennial pp. 24–98). New York: Wiley. meeting of the Society for Research in Child Thompson, R. A. (2006b). Conversation and Development, Denver, CO. developing understanding: Introduction to Valiente, C., Fabes, R. A., Eisenberg, N., & Spin- the special issue. Merrill–­Palmer Quarterly, rad, T. L. (2004). The relations of parental 52(1), 1–16. expressivity and support to children’s coping Thompson, R. A. (2011). Emotion and emotion with daily stress. Journal of Family Psychol- regulation: Two sides of the developing coin. ogy, 18, 97–106. Emotion Review, 3(1), 53–61. Waters, S., Virmani, E., Thompson, R. A., Thompson, R. A., Flood, M. F., & Goodvin, Meyer, S., Raikes, A., & Jochem, R. (2010). R. (2006). Social support and developmental Emotion regulation and attachment: Unpack- psychopathology. In D. Cicchetti & D. Cohen ing two constructs and their association. (Eds.), Developmental psychopathology: Vol. III. Risk, disorder, and adaptation (2nd ed., Journal of Psychopathology and Behavioral pp. 1–37). New York: Wiley. Assessment, 32(1), 37–47. Thompson, R. A., & Goodman, M. (2010). Woltering, S., & Lewis, M. D. (2009). Devel- Development of emotion regulation: More opmental pathways of emotion regulation in than meets the eye. In A. Kring & D. Sloan childhood: A neuropsychological perspective. Mind, Brain, and Education, 3, 160–169.

Chapter 12 Emotion Regulation in Adolescence Michaela Riediger Kathrin Klipker Adolescents’ emotional lives are distinct of the current status of research on adoles- from those of children or adults: They react cent emotion regulation in Western societ- more strongly to emotion-e­liciting situa- ies. To set the stage, we briefly characterize tions (e.g., Miller & Shields, 1980; Stroud the developmental period of adolescence and et al., 2009), experience negative and mixed the relevance of emotion regulation for ado- emotions more frequently (e.g., Larson & lescents’ socioemotional adjustment. The Asmussen, 1991; Larson & Lampman-­ main part of the chapter integrates evidence Petraitis, 1989; Riediger, Schmiedek, Wag- on three questions that are important for ner, & Lindenberger, 2009; Riediger, Wrzus, the understanding of emotion regulation in & Wagner, 2012), and fluctuate more rap- adolescence: Which factors contribute to the idly in their emotional states (e.g., Larson, development of emotion regulation skills in Moneta, Richards, & Wilson, 2002). Emo- that life phase? Does adolescents’ motiva- tion regulation is assumed to play a key tions to regulate their feelings differ from role in these characteristics of adolescents’ those of other age groups? Which strategies emotional experiences (e.g., Opitz, Gross, do adolescents use to regulate their emo- & Urry, 2012). It has also been proposed to tions, and how adaptive and effective are be a central component of adolescents’ more these strategies? We conclude with an out- general socioemotional adaptation, that is, look on important future research direc- their ability to adjust to the socioemotional tions. challenges they face (e.g., Silk, Steinberg, & Morris, 2003; Yap, Allen, & Sheeber, Setting the Stage 2007). Studying emotion regulation is there- The life phase “adolescence” refers to the fore essential for understanding the develop- transition from being a child to being an mental phase of adolescence. independent adult. It can roughly be defined as beginning with the physical changes of In line with the widely accepted definition puberty and ending with the assumption of by Gross (e.g., 1999), we conceive of emo- adult social roles. Thus, the particular age tion regulation as comprising those delib- range of adolescence varies widely between erate and automatic processes that allow individuals, depending on the onset of individuals to influence which emotional puberty and life circumstances. Empirical experiences they have, when they have them, and how they experience and express them. This chapter provides an overview 187

188 DEVELOPMENTAL CONSIDERATIONS investigations often approximate adoles- (e.g., Thorne, 2004). These challenges can cence as a period within the time window stipulate the development of skills that are from about 10 to about 25 years of age. necessary for identity formation, autonomy In Western industrialized societies, the life increase, and other developmental tasks of phase of adolescence has undergone a major adolescence (Erikson, 1968). This, however, expansion in recent history. Throughout the is only true as long as the challenges do not past century, the average onset of pubertal overtax the individual (Yap et al., 2007). processes has occurred progressively earlier, Temporary increases in affective reactivity particularly in girls. Nowadays, hormonal (which we address in greater detail when changes of female puberty begin between 9 discussing neurophysiological aspects of and 12 years of age, and most of the physical emotion regulation in adolescence) may changes of puberty are typically completed amplify adolescents’ vulnerability to being by the midteens. The acceleration of puberty overwhelmed by the emotional challenges onset has coincided with a progressive delay they face (e.g., Gunnar, Wewerka, Frenn, of assuming adult roles in Western societ- Long, & Griggs, 2009; Miller & Shields, ies. This is often postponed until around the 1980; Stroud et al., 2009; Sumter, Bokhorst, early 20s (for an overview, see Dahl, 2004). Miers, Van Pelt, & Westenberg, 2010). In fact, increases during adolescence in inter- Adolescence is characterized by vast nalizing (e.g., depressiveness) and external- changes in multiple domains of functioning izing (e.g., aggressiveness) problems, as well (for an overview, see, e.g., Eccles, Temple- as in various forms of psychopathology (e.g., ton, Barber, & Stone, 2003). Puberty-­related Allen & Sheeber, 2009; Shortt, Stoolmiller, hormonal changes, for example, lead to Smith-Shine, Mark Eddy, & Sheeber, 2010; sexual maturation, the development of sec- Silk et al., 2003) demonstrate that develop- ondary gender characteristics, and dramatic mental demands of adolescence may over- changes in body size and composition. Ado- stretch the adaptational capacity of vulner- lescence is also a phase of profound psycho- able individuals. logical change, such as marked growth in cognitive functioning or moral reasoning. Emotion regulation has been proposed Furthermore, many aspects of social experi- to play a core role in adolescents’ ability to ence change as adolescents move away from weather the developmental challenges they their family and toward their peers. Impor- face (e.g., McLaughlin, Hatzenbuehler, tantly, there is large variability in the timing Mennin, & Nolen-­Hoeksema, 2011; Silk et of these various changes, both between ado- al., 2007; Yap et al., 2007). Both over- and lescents (e.g., regarding the onset of puberty) underregulation of emotion are assumed and within adolescents (i.e., development in to imply risks for adolescents’ socioemo- one domain of functioning does not neces- tional adaptation, that is, for their ability to sarily imply parallel development in another; adjust to the socioemotional challenges they e.g., Dahl, 2004). encounter. Available evidence indeed links emotion regulation skills to higher social During adolescence, emotionally chal- competence, more prosocial behavior, bet- lenging situations typically become more ter academic achievements, and fewer inter- frequent and intense. In Western industri- nalizing and externalizing problems (e.g., alized societies, for example, this derives Bell & Calkins, 2000; Buckley & Saarni, from an increased potential for conflict with 2009; Silk et al., 2003; Zeman, Cassano, parents, from adolescents’ greater sensitivity Perry-­Parrish, & Stegall, 2006). Emotion to peer-r­elated social interactions, or from regulation has also been related to more early romantic and sexual experiences. In peer acceptance, higher peer status, and a addition, the physical changes of puberty lower likelihood of experiencing chronic make social demands regarding adolescents’ victimization and bullying by peers (for an behavior more challenging. Social environ- overview, see Buckley & Saarni, 2009). In ments expect adolescents who look like terms of adolescent psychopathology, emo- adults to behave like adults (Dahl, 2004). tion regulation has been implicated in the Furthermore, cognitive growth during development of diverse problems, including adolescence provides the basis for greater anxiety, depressive or conduct disorders, engagement with emotion-r­elevant aspects as well as eating disorders in females (e.g., of one’s own and others’ existence and future

Emotion Regulation in Adolescence 189 McLaughlin et al., 2011; Silk et al., 2007; trol an essential role for effective emotion Sim & Zeman, 2006; Yap et al., 2007). regulation (Gray & Braver, 2007; Ochsner & Gross, 2005, this volume). Cognitive con- To date, the vast majority of studies are trol encompasses the various processes that cross-s­ ectional (but see, e.g., McLaughlin et are necessary for successful goal pursuit, al., 2011). Thus, whether emotion dysregu- which include maintaining a current goal lation is a risk factor for, or a consequence in working memory or shielding it from dis- of adaptational problems in adolescence traction (Best & Miller, 2010). Intentional cannot yet be conclusively anwered. There emotion regulation can be conceived of as an also is substantial variability across studies instance of such goal-d­ irected behavior (e.g., in the particular aspects of emotion regula- to not express one’s anger). Indeed, several tion investigated. Taken together, however, studies with participants from various age ample evidence links emotion regulation groups indicate associations between mea- skills to socioemotional adjustment in ado- sures of cognitive control and the effective- lescence. Insight into the development of ness of emotion regulation (e.g., Compton et emotion regulation skills during adolescence al., 2008; Hoeksma, Oosterlaan, & Schip- can thus have important practical implica- per, 2004; Robinson, 2007). For example, tions. This chapter reviews three fundamen- Schmeichel and Demaree (2010; Schmeichel, tal research topics in this respect that refer to Volokhov, & Demaree, 2008) demonstrated influences on emotion regulation in adoles- that undergraduates with higher working cence, emotion regulation motivations, and memory capacity (a facet of cognitive con- the ways and effectiveness of adolescents’ trol) were better able to suppress negative attempts to regulate emotions. emotional expressions and to reappraise stimuli in an unemotional manner. Influences on Emotion Regulation The development of emotion regulation Cognitive control capacities become skills in adolescence is shaped by the interac- increasingly complex and efficient through- tion of multiple influences. We focus on two out childhood and adolescence and into examples in this chapter, namely, individu- young adulthood (for review, see Yurgelun-­ als’ neurophysiological development and Todd, 2007). Maturation of the prefrontal their familial context. A discussion of other cortex has been associated with the devel- factors—­such as genetic disposition, gen- opment of these skills (e.g., Casey, Getz, der, personality, temperament, attachment & Galvan, 2008; Casey et al., 2010; Dahl, style, or influences from developments dur- 2001, 2004; Steinberg, 2008). Structural ing infancy and childhood—i­s beyond the maturation includes a decrease in gray mat- scope of this chapter (but see, e.g., Shaver & ter. This is reflective of synaptic pruning, Mikulincer, and Mesquita, De Leersnyder, that is, the elimination of unused neural & Albert, this volume; Bariola, Gullone, & connections, which possibly enables more Hughes, 2011). focal activation of these brain regions. These structural developments are also associated Neurophysiology with functional changes. Studies examining and Emotion Regulation performance on tasks requiring cognitive Advances in neuroimaging technologies have control (e.g., Stroop or flanker tasks), for led to a recent upsurge of interest in the role example, suggest an age-­related increase in of neurophysiological development in emo- the activation of the dorsolateral prefrontal tion regulation during adolescence. Research cortex during childhood and adolescence, has shown that similar brain regions are acti- and a further age-r­ elated increase from ado- vated when participants regulate emotions lescence to young adulthood in the focaliza- and when they control cognitive operations tion and efficiency of selective recruitment (Mauss, Bunge, & Gross, 2007; Mohanty et of cognitive control areas (for reviews, see al., 2007; Ochsner & Gross, 2005). Thus, Casey et al., 2008; Steinberg, 2008). Behav- neurophysiologically oriented research typi- ioral studies show that these structural and cally ascribes the exertion of cognitive con- functional changes are associated with per- formance improvements in a variety of cog- nitive tasks from childhood to late adoles- cence (e.g., Best & Miller, 2010; Forman,

190 DEVELOPMENTAL CONSIDERATIONS Mäntylä, & Carelli, 2011; Ordaz, Davis, & mate and human postmortem studies but Luna, 2010). needs further empirical confirmation (e.g., Steinberg, 2008). Furthermore, Pfeifer and The pattern of improving cognitive control Allen (2012) recently criticized this assump- associated with maturation of the prefrontal tion as being based on an overly simplifying cortex suggests that self-r­egulation compe- conception of structure–f­unction mappings tence in general, and emotion regulation in the brain. Development of the prefrontal competence in particular, should improve cortex, for example, may also be associated from childhood to adulthood (Casey et al., with more emotional understanding and 2010). However, the well-­documented peak complexity, which in turn could be related in impulsivity and risky/reckless behavior to emotional experiences. In addition, while in adolescents and emerging adults demon- neurophysiological maturation undeniably strates limitations in the ability or willing- plays a central role in emotion regulation in ness to resist temptations and peer influence adolescence, external influences, to which (see also Luerssen & Ayduk, Grecucci & we turn next, are also important for those Salfey, and Cole, this volume) and suggests learning processes. a nonlinear development of self-r­egulation competence (Steinberg, 2008). The tempo- Familial Contexts rary increase in negative emotionality (e.g., and Emotion Regulation Ciarrochi, Heaven, & Supavadeeprasit, Emotion regulation abilities are funda- 2008) and affective instability (e.g., Larson mentally shaped by the continual interac- et al., 2002) during adolescence gives rise tions between individuals and their social to the assumption that a similar nonlinear environments. On the one hand, demands development is characteristic of emotion on emotion regulation skills are often par- regulation skills as well. ticularly high in social situations (Bell & Calkins, 2000). Individuals are required Several contemporary positions on ado- to comply with sociocultural norms and lescent brain development conclude that an expectations regarding appropriate emo- exclusive focus on prefrontal cognitive con- tional experiences and expressions in social trol areas is insufficient for understanding situations. Furthermore, attainment of situ- the neurophysiological correlates of ado- ational goals would often not be possible if lescent emotion regulation. These models individuals were not able to regulate their propose that it is also necessary to take into emotions effectively (e.g., to control their consideration subcortical regions involved in anxiety or anger appropriately; Thompson, the processing of emotional information, as 1994). well as the coordination between these and prefrontal regions (e.g., Casey et al., 2010; On the other hand, social contexts also Dahl, 2001; Steinberg, 2008). Central to present important influences that facilitate these models is the assumption that struc- or hinder emotion regulation skills (Bariola tural and functional maturation of these et al., 2011). Most of the available research various brain systems do not occur in par- focuses on the respective role of familial con- allel. For example, subcortical regions criti- texts. Morris, Silk, Steinberg, Myers, and cal to affective processing are assumed to Robinson (2007), for example, propose that mature earlier in adolescence than cortical family contexts influence the development regions subserving cognitive control. The of emotion regulation during childhood differential timing of maturation in differ- and adolescence in three important ways: ent brain regions is expected to result in a through observation learning (e.g., when temporarily increased disjunction between parents’ own emotion regulation serves younger adolescents’ affective experiences as a social role model for their offspring), and their emotion regulation abilities. This through parenting practices and explicit disjunction is smaller during childhood, instruction (e.g., when parents coach emo- when both systems are still developing, and tion regulation strategies), and through the becomes smaller again during adulthood, emotional climate in the family (e.g., when when both systems have fully matured. parental behaviors and attitudes toward The assumption of differential develop- the offspring’s emotional expressions and mental timing in prefrontal and subcortical regions is consistent with nonhuman pri-

Emotion Regulation in Adolescence 191 experiences reinforce or discourage emotion adolescents, however, indirect influences, regulation). such as talking about possible emotion regulation strategies, fit better with their Empirical research on the role of paren- cognitive and self-­regulation competen- tal influences on the emotion regulation of cies, and their increased need for autonomy offspring has typically focused on families (Bell & Calkins, 2000; Morris et al., 2007). with infants or children (see Thompson, this Thus, parental balancing of the adolescent’s volume). Fewer investigations are available opposing needs for autonomy on the one on the role of familial contexts for emotion hand, and for guidance and structure on the regulation in adolescence. This work sug- other is particularly important for emotion–­ gests that parents’ emotion–­socialization socialization in adolescence. Imbalance in behaviors continue to be important when either direction can have potentially disad- their children become adolescents. This is vantageous effects on adolescents’ socio- the case even though adolescents’ orienta- emotional adaptation. Adolescents who tion toward more autonomy increases the remain overly dependent on their parents relative importance of extrafamilial influ- appear to be at a higher risk of internaliz- ences, such as peers, media, neighborhood, ing problems (e.g., depression; for review, or culture, as socialization agents of emotion see Morris et al., 2007). Lack of emotion regulation (Bariola et al., 2011). Shortt and regulation skills has been proposed to play a colleagues (2010), for example, found that mediating role in this regard (for review, see maternal emotion coaching (e.g., approach- Bell & Calkins, 2000). In contrast, adoles- ing the adolescent when he or she is upset cents who lack or refuse emotional guidance to talk about the situation and experienced from their parents seem to be at a higher risk emotions) was associated with fewer diffi- for externalizing problems (e.g., rule break- culties of their 10- to 13-year-old children ing and delinquency, substance use, aggres- in regulating anger. Better anger control, in sion). These behaviors, again, have been turn, was associated with fewer external- associated with a lack of skills to regulate izing problems of the adolescent. Maternal emotions, particularly anger (for review, see emotion coaching in this study also pre- Morris et al., 2007). dicted fewer externalizing problems of the adolescent 3 years later (i.e., at ages 13–16). Interaction between Internal Sheeber, Allen, Davis, and Sorensen (2000) and External Factors found that maternal approval or affirma- Evidence suggests that a complex set of tion of adolescent depressive behaviors was internal and external factors contributes to predictive of longer episodes of negative the development of emotion regulation skills affect on the part of the adolescent (ages during adolescence. We illustrated this using 12–19 years). The authors assumed that this the examples of individuals’ neurophysi- reflects a poorer ability to regulate negative ological development, which represents an affect. Yap, Allen, and Ladouceur (2008) essential internal factor, and their famil- found a relationship between self-r­eported ial context, which can provide significant maternal tendencies to be restrictive and external influences on emotion regulation. unaccepting of positive affective displays Importantly, the effects of these internal and and adolescents’ use of maladaptive emotion external factors are not unidirectional. They regulation strategies in conflict situations. reciprocally shape each other. Familial influ- More restrictive maternal tendencies were ences on emotion regulation competence, associated with a higher tendency among for example, are modulated by the propen- early adolescents to use maladaptive strate- sity of adolescents to experience and express gies, such as venting or dysregulated expres- intense affect. The latter, as discussed ear- sion of negative affect, which in turn were lier, can in turn be related to a temporary associated with more depressive symptoms. dissociation of neurophysiological changes yielding increased emotional reactivity at Evidence also suggests that emotion–­ a time when control functions have not socialization styles in the family need to be yet fully developed. In a parent–a­dolescent adjusted to the age and developmental status interaction study, Schulz, Waldinger, of the offspring. Direct intervention in the form of soothing or directive instructions is effective in children at younger ages. For

192 DEVELOPMENTAL CONSIDERATIONS Hauser, and Allen (2005) indeed showed (Riediger et al., 2009; Riediger, Wrzus, & that parents were less positively engaged and Wagner, 2013). Participants reported their showed more hostility when their adolescent momentary emotional experiences, on aver- children fully expressed their emotions than age 54 times over 3 weeks, and whether they when they controlled their emotion expres- momentarily wanted to influence their feel- sions. Other studies also indicate that par- ings. Contrahedonic motivation (wanting ents may have difficulties in coping with to maintain or enhance negative affect, or adolescents’ uninhibited displays of intense to dampen positive affect) was considerably emotions, and that these parental difficulties less prevalent than prohedonic motivation can shape emotion coaching and the emo- (wanting to maintain or enhance positive tional climate in the family, and thus feed affect, or to dampen negative affect) in all back to the adolescent (e.g., Dishion, Nel- investigated age groups. There were, how- son, & Bullock, 2004). ever, pronounced age differences. Adoles- cents reported contrahedonic motivation Taken together, the research reviewed so most frequently, namely, in about 25% of far suggests that interactions among a com- the measurement occasions. There were plex set of internal and external factors are steep decreases in the prevalence of contra- associated with the development of emotion hedonic motivation between the adolescent regulation skills during adolescence. Struc- and the young adult subsamples, and a fur- tural and functional changes of the pre- ther decline throughout the adult subsam- frontal cortex subserving cognitive control ples into old age. Prohedonic motivation, abilities might facilitate cognitive emotion in contrast, showed an opposite prevalence regulation skills throughout adolescence. pattern. It was least prevalent among adoles- Simultaneous increases in affective reac- cent and young adult participants, and most tivity associated with the development of prevalent in later adulthood (Riediger et al., subcortical regions, however, might over- 2009, 2013). tax these evolving regulatory abilities. Such neurobiological factors could interact with Furthermore, analyses of associations aspects of the individual’s social context in with within-­person fluctuations in work- promoting or hindering the development of ing memory capacity indicated that contra- emotion regulation skills. hedonic motivation was more cognitively demanding than prohedonic motivation. The Emotion Regulation Motivation more contrahedonic motivation participants To date, only little attention has been paid to reported, the lower their momentary work- the fact that emotion regulatory behaviors ing memory capacity. Prohedonic motiva- are preceded and fundamentally shaped by tion, in contrast, was only weakly associated motivational processes (see John & Eng, this with fluctuations in working memory. This volume). One reason for this is that most was the case irrespective of participants’ investigators seem to assume that emotion ages. Therefore, despite the pronounced regulation is inevitably “prohedonic,” that age-r­elated differences in the prevalence is, always directed at optimizing one’s well- of different affect regulation motivations, being. Only recently has awareness arisen their cognitive requirements appeared to be that occasional exceptions are possible, that independent of the individual’s age (Riedi- is, that emotion regulation can sometimes be ger, Wrzus, Schmiedek, Wagner, & Linden- “contrahedonic.” People may occasionally berger, 2011). be inclined to dwell on or intensify negative emotional experiences, such as anger or sad- Two complementary theoretical perspec- ness, or to lessen positive ones, such as pride tives on possible reasons for contrahedonic or amusement (e.g., Erber & Wang Erber, motivation are currently being discussed in 2000; Riediger et al., 2009; Tamir, 2009). the literature. From an instrumental perspec- tive, it has been argued that there may be sit- Various experience-s­ampling studies uations in which negative affect is useful, or reported by Riediger and colleagues demon- when positive affect is disadvantageous for strated adolescent peaks in the prevalence of the individual (e.g., Tamir, 2009). Anger, for contra-h­ edonic motivation in samples rang- example, can help to assert one’s interests in ing in age from 12 years to late adulthood an argument, whereas expressions of joy can be inappropriate when attending a funeral. Contrahedonic orientation may therefore

Emotion Regulation in Adolescence 193 derive from people being (consciously or ences and contrahedonic motivation were unconsciously) strategic in seeking affec- most prevalent among adolescent partici- tive states that are instrumental in a given pants and least prevalent among older adults context (Ford & Tamir, 2012). Supporting (Riediger et al., 2009, 2013).1 Importantly, this position, Riediger and colleagues (2009) this pattern of age-r­elated differences was found contrahedonic motivation to be less not restricted to participants’ self-r­eport; it strongly related to people’s current emo- was also reflected in implicit representations tional experiences than prohedonic orienta- of affect valence, as assessed with implicit tions. That is, while participants tended to association tests. Compared to adults from report that they wanted to maintain positive various age groups, adolescents associated affect when it was high, to enhance positive positive affect least distinctively with pleas- affect when it was low, and to dampen nega- antness (vs. unpleasantness) and unhappi- tive affect when it was high, contrahedonic ness least distinctively with unpleasantness orientations were less strongly associated (vs. pleasantness). The older the participants, with participants’ momentary affect. This however, the more differentiated their repre- is consistent with the idea that contrahe- sentations of the valence of affective states. donic motivation may serve instrumental Furthermore, the less differentiated people’s functions that are not necessarily related to mental representations of affect valence, the the individual’s current affect. The authors more likely they were to report mixed affect speculated that the relatively higher preva- and contrahedonic motivation in their every- lence of contrahedonic motivation in adoles- day lives (Riediger et al., 2013). Although cence might reflect an instrumental value in causal conclusions are not possible given the tackling the developmental tasks of that life correlational nature of these studies, these phase. Repudiating prevailing hedonic con- findings are in line with the mixed affective ventions, for example, might put adolescents perspective. Thus, the comparatively higher in situations where they have to deal with prevalence of mixed affective experiences in negative emotional experiences. This, in adolescence could be among the factors that turn, might help them to establish emotional contribute to a comparatively higher preva- autonomy from their parents, affirm a sense lence of self-­reported contrahedonic moti- of maturity, develop their sense of identity, vation in that age group. This seems to be or refine their self-­regulatory competencies. associated with relatively more undifferenti- To date, however, these assumptions have ated mental representations of the valence of not yet been empirically verified. affective states in adolescence. A complementary account of possible Emotion Regulation Strategies reasons for contrahedonic motivation is the The research reviewed so far demonstrates mixed-a­ffect perspective. It proposes that that internal and external influences on contrahedonic motivation may also arise emotion regulation abilities undergo pro- when apparently negative emotional states found changes during childhood, adoles- are accompanied or followed by positive cence, and into young adulthood, and that experiences, that is, when the emotional adolescents’ emotion regulation motivation episode is mixed (Andrade & Cohen, 2007; differs from those of adults. But what are Riediger et al., 2009). Such mixed affective the consequences? Are there age-r­elated dif- experiences might motivate individuals, for ferences in the strategies individuals use to example, to seek or maintain a given nega- regulate their emotions, or in their emotion tive affective state because of the positive regulation effectiveness? An integrated and aspects they associate with it. In fact, the interpretable body of research addressing experience sampling studies by Riediger this question has yet to emerge. Only few and colleagues demonstrated that mixed studies have investigated cross-s­ectional age affective experiences in everyday life were differences within or across developmen- related to participants’ being more likely to tal periods, and longitudinal evidence on report contrahedonic orientations. In addi- change within persons over time is scarce. tion, these studies also showed pronounced These studies differ widely in their concep- age-­related differences in the prevalence of tualization of regulatory strategies. This is mixed-­affective experiences that followed the same pattern as those of contrahedonic motivation: Both mixed affective experi-

194 DEVELOPMENTAL CONSIDERATIONS partly due to the fact that these studies were of antecedent-­focused emotion regulation. conducted either in the research tradition Central to these strategies is that they are of emotion regulation or (more frequently) directed at factors influencing the process in that of coping, with the latter including of emotion generation. Emotion-­focused emotion regulation as one of many ways of coping, in contrast, overlaps with response-­ dealing with stress (Compas, 2009; Gross, focused emotion regulation in Gross’s model, 1999). Despite considerable overlap in as both conceptualizations refer to strategies research interests, there has been little cross- directed at modifying an emotional response talk between these two traditions. after it has been activated. Research within the emotion regula- Use of Emotion Regulation Strategies tion tradition has most typically proceeded Zimmer-G­ ebeck and Skinner (2011) inte- from the process model of emotion regula- grated findings from 58 studies compar- tion (e.g., Gross, 1999), which distinguishes ing coping strategies within the age range antecedent-f­ocused strategies (directed at from childhood to adolescence. Overall, the changing the emotional input prior to the authors concluded that age differences in actual emotional experience) from response-­ regulatory behaviors are characterized by focused strategies (directed at modifying two trends. emotional responses after they have been elicited). Most research has centered around First, there appears to be an age-g­raded two prototypical examples, namely, “cog- increase in regulatory capacities. This is nitive reappraisal” (i.e., changing one’s reflected in not only an increased under- thinking about an emotion-e­liciting situa- standing of emotional situations (Labouvie-­ tion or about the capacity to manage it) as Vief, DeVoe, & Bulka, 1989) but also a an instance of antecedent-f­ocused emotion broader and more sophisticated repertoire regulation, and “expressive suppression” of regulatory strategies. For example, from (i.e., inhibiting the outward expression of childhood to adolescence, instrumental an emotional experience) as an instance of action in response to stressors appears to response-­focused emotion regulation. become gradually supplemented by planful problem solving. Furthermore, distraction More research on regulatory strategies tactics seem to become more diverse and in adolescence has been conducted within increasingly include cognitive distraction, the coping research tradition, which views in addition to behavioral distraction. Also, emotion regulation as one of various ways compared to children, adolescents appear of dealing with stress. These studies, how- to be better able to attend to and reflect on ever, often referred to different classifica- their own emotional states. They also seem tion systems of regulatory strategies, mak- to use progressively more sophisticated cog- ing an integration of findings difficult. Two nitive strategies to deal with emotions, such widely used classification systems are the as positive self-talk and reappraisal. primary–s­econdary control model (e.g., Rothbaum, Weisz, & Snyder, 1982) and The second developmental trend from the ways-of-­coping model (e.g., Lazarus & childhood to adolescence, according to the Folkman, 1984). The primary–s­econdary analysis by Zimmer-­Gebeck and Skinner control model distinguishes regulatory strat- (2011), reflects age-­related improvement in egies according to their underlying goals. tailoring regulatory attempts to the situa- The aim of primary coping is to influence tion. That is, adolescents seem to become characteristics of the stress-e­liciting condi- increasingly able to engage in those regu- tions, whereas the aim in secondary coping latory strategies that are most effective in is to adapt oneself to the conditions as they dealing with particular kinds of stressors. are. The ways-of-c­ oping model, in contrast, For example, evidence suggests increasing distinguishes regulatory strategies according use of both problem solving to deal with to their respective target. Problem-­focused modifiable difficulties (e.g., in school or coping refers to modifying the source of sports) and distraction to deal with uncon- stress, whereas emotion-­focused coping trollable stressors (e.g., parental illness) refers to modifying the elicited emotional from childhood to adolescence. Another response. Primary and secondary coping, as example derives from a longitudinal study well as problem-f­ocused coping, partly over- on anger regulation among friends. Salisch lap with Gross’s (1999) conceptualization

Emotion Regulation in Adolescence 195 and Vogelgesang (2005) observed a decline regulatory styles, as characterized by their from older childhood to adolescence in indi- association with self- or other-r­eports of viduals’ self-­reported use of confrontation various indicators of psychological adjust- and harming, redirection of attention, and ment, such as emotional or behavioral prob- ignoring and self-blame when dealing with lems, psychosomatic health, subjective well- situations involving anger toward a friend. being, and social or academic competence. The authors also found an increase in par- ticipants’ use of explanation and reconcilia- Within the research tradition on emotion tion in such situations. regulation, investigations of the adaptive- ness of specific regulatory strategies have Zimmer-­Gembeck and Skinner (2011) most typically focused on reappraisal and also emphasized that the apparent improve- expressive suppression. Investigations with ment in emotion regulatory competence young adults suggest that habitual use of does not necessarily need to be linear. Some reappraisal is associated with a healthier of the reviewed evidence points to tempo- profile of socioemotional adjustment than is rarily increased struggles with regulation in habitual use of expressive suppression, and response to stressors during the transition that the latter is also associated with greater from late childhood to early adolescence. cognitive and physiological costs (for an Compared to older adolescents, for exam- overview, see Gross, 2002). Investigations in ple, younger adolescents have occasionally adolescent samples are rare, but they have shown lower levels of help-s­ eeking behaviors yielded similar patterns of findings (Betts, and effort expenditure in domains in which Gullone, & Allen, 2009; Hughes, Gullone, this would be helpful (e.g., regarding school-­ Dudley, & Tonge, 2010). related stressors). The authors also reviewed evidence of a temporary rise in the use of More research on the adaptiveness of potentially more maladaptive regulation regulatory styles in adolescence has been strategies during early adolescence, such as conducted in the research tradition on cop- cognitive escape, rumination, verbal aggres- ing. Compas and colleagues (2001) reviewed sion, or venting. 63 of these studies and concluded that the majority of the studies demonstrate that Most of the extant research investigated problem-­focused and engagement coping the age range from childhood to adolescence. (which overlap with antecedent-­focused Only a few available studies compare the use emotion regulation) were associated with of regulatory strategies in adolescence and better psychological adjustment during adulthood. Overall, this research suggests childhood and adolescence. The specific continued change into young adulthood subtypes most consistently associated with (e.g., Blanchard-­Fields & Coats, 2008; Gar- better adjustment included the generation nefski, Legerstee, Kraaij, van den Kommer, of positive and hopeful thoughts (i.e., cog- & Teerds, 2002). Garnefski and colleagues, nitive reappraisal), careful analysis of the for example, found that adolescents around stressful situation, and selective attention 13 years of age reported using various cog- to positive aspects of the situation. In con- nitive change strategies (e.g., positive reap- trast, disengagement and emotion-f­ocused praisal or refocus on planning) significantly coping (which overlap largely with response-­ less frequently than did adults of various age focused emotion regulation) were associated groups. with poorer psychological adjustment in most studies. However, it was not focusing Effects of Emotion Regulation on one’s emotions in general that was related Strategies: Adaptiveness to lower psychological adjustment. Rather, and Effectiveness these effects were due to disengagement Most studies on the effects of emotion regu- from the stressor or from one’s emotions lation strategies proceed from the idea that it (e.g., expressive suppression, avoidance, is meaningful to conceive of an adolescent’s or withdrawal), negative cognitions about regulatory style as a trait. These studies the self and the situation, and unregulated characterize adolescents according to their release or ventilation of emotions. overarching tendency to use particular types of regulatory strategies more than others. Compas and colleagues (2001) also The focus is on the adaptiveness of different emphasized that notable differences in results across studies may be associated with the nature of the investigated stressor, par-

196 DEVELOPMENTAL CONSIDERATIONS ticularly its controllability. This is consistent engagement in this study was unrelated to with the view that an adaptive regulatory mood improvement after social rejection. style may not be characterized primarily by In contrast, better mood improvement was more frequent use of particular regulation associated with more behavioral distrac- strategies relative to others, but by a high tion and less passive behaviors. Silk and col- level of flexibility (e.g., Gross, 1999). This leagues (2003) used experience sampling in flexibility should be evident in positive asso- 12- to 17-year-old participants to investigate ciations between socioemotional adaptation the role of regulatory strategies for mood and adolescents’ ability to tailor specific dynamics after participants had encoun- regulation attempts to the nature (e.g., the tered negative events in their daily lives. controllability) of a given situation. While When adolescents used disengagement-­ this claim seems intuitively appealing, future related strategies in response to subjectively research needs to strengthen the empirical severe negative events, or when they reacted foundation supporting it. involuntarily (e.g., by ruminating or act- ing impulsively), they maintained higher The research on the adaptiveness of affect levels of negative affect over time, suggest- regulation styles takes a person-­centered ing limited effectiveness of these strategies. (i.e., trait-o­riented) approach. Another Engagement in primary and secondary con- approach is process-­centered (i.e., state-­ trol attempts, however, was unrelated to oriented) and focuses on how effective indi- the dynamics of negative affect. The latter, viduals actually are in influencing their emo- again, is not consistent with what one would tional experiences in intended ways. Hardly expect based on the majority of studies on any evidence is available on the effectiveness associations between regulatory styles and of emotion regulation in adolescence. This more general outcomes of socioemotional is probably due to the methodological chal- adaptation. Differences in findings across lenges involved in operationalizing emo- studies on the effectiveness versus the adap- tion regulation effectiveness. In fact, there tiveness of regulatory strategies might be is an ongoing debate whether and to what attributable to differences in assessment degree it is possible to disentangle emotion approaches (process- vs. person-c­entered) regulation from the respective emotional and/or in the investigated emotion-e­liciting experience. While some researchers regard situations. Future research is necessary to emotion regulation as being experientially better integrate the findings from these two and structurally indistinguishable from the lines of research. emotional experience, others claim that it is possible and meaningful to differenti- A second perspective on emotion regula- ate between emotion and emotion regula- tion effectiveness in adolescence focuses on tion (for an overview of positions, see, e.g., respective age-­related differences. In two Matarazzo, 2008). recent investigations (McRae et al., 2012; Silvers et al., 2012), participants between 10 In line with the latter position, emotion and 23 years of age were presented neutral regulation effectiveness in adolescence has and aversive picture stimuli from the Inter- recently been addressed from two different national Affective Picture System (IAPS). perspectives. One focuses on the effective- Viewing instructions varied across studies ness of particular emotion regulation strat- but generally included a reappraisal condi- egies in adolescence. Here, adolescents’ use tion (e.g., thinking about the picture in a of regulatory strategies is not conceived of way that makes one feel less negative) and as a trait but as a process that can fluctuate a comparison condition (e.g., reacting natu- within persons over time. The covariation rally to the picture, but not reappraising). of this process with fluctuations in affec- After viewing each picture, participants tive experiences is of particular interest. rated their current negative affect. Regula- Reijntjes, Stegge, Terwogt, Kamphuis, and tion effectiveness was determined as the rel- Telch (2006), for example, experimentally ative decrease in negative affect after view- manipulated experiences of social rejec- ing aversive stimuli in the reappraisal versus tion during an ostensible computer game in the comparison condition. Silvers and col- 10- to 13-year-olds. Contrary to what one leagues observed an age-r­elated increase in would expect from research on the adaptive- regulation effectiveness from late childhood ness of regulatory styles, problem-­focused

Emotion Regulation in Adolescence 197 to late adolescence that tapered off in young varying affective states. Empirical evidence adulthood. McRae and colleagues (2012), that directly links emotion regulation to the in contrast, found relative age invariance in characteristic emotional experiences of ado- regulation effectiveness from ages 10 to 17 lescents is rare. Providing such evidence rep- years, but an increase in effectiveness from resents an important undertaking for future late adolescence to young adulthood. These studies. Knowledge about emotion regula- inconsistencies in findings might be due to tion in adolescence can thus have important methodological differences between the practical implications. We believe that an studies. Procedures in both studies varied, integration of findings from three research for example, in the selection of stimuli for areas—i­nfluences on the development of the analyses, and in whether or not partici- emotion regulation, emotion regulation pants were trained in the reappraisal task. motivation, and emotion regulation strate- More empirical investigations are necessary gies—i­s crucial for a better understanding of to clarify and expand the empirical picture adolescents’ emotion regulation. This chap- on emotion regulation effectiveness through- ter has reviewed contemporary positions out adolescence. within each of these areas of inquiry. Wrapping Up and Looking Forward A complex set of internal and external Adolescence is characterized by an upsurge factors contribute to the development of in the frequency and intensity of emotional emotion regulation skills during adoles- challenges, in a time of life when individuals cence. A temporal dissociation in the matu- have to manage such challenging situations ration of prefrontal and subcortical brain more and more independently. This provides regions, for example, is assumed to result in developmental stimulation, but it can also a temporary decrease in emotion regulation increase vulnerability to adaptational dif- capacity when cognitive control capacities ficulties. Emotion regulation has been pro- have not yet sufficiently improved to provide posed to play an important role in how ado- adequate control of the progressively more lescents adapt to these challenges. Empirical active emotion system. It seems plausible investigations of this claim vary considerably that this may be one of the factors contribut- regarding their operationalizations of what ing to the unique quality of emotional expe- “better” emotion regulation entails. System- riences observed during adolescence. How- atic comparisons of implications of different ever, predictions about relations between facets of emotion regulation for adolescents’ neurophysiological changes during adoles- socioemotional adaptation are still lacking. cence and adolescents’ development of emo- In addition, causal relations between emo- tion regulation abilities are still in need of tion regulation and developmental adapta- systematic empirical verification. So far, lit- tion need more systematic investigation in tle is known about the possibility that differ- the future. Another open task for the future ent emotion regulation strategies might vary is a more rigorous investigation of possible in terms of how much cognitive control they interindividual differences in adaptational require, and that people can recruit external outcomes of emotion regulation. There are, resources when internal resources are insuf- for example, indications that outcomes may ficient for effective emotion regulation (e.g., differ between male and female adolescents by asking somebody else to distract one- (Bowie, 2010; Perry-P­arrish & Zeman, self from emotion-­eliciting thoughts when 2011). attempts of self-i­nitiated distraction are not successful, Opitz et al., 2012). Furthermore, In addition to assuming implications of emotion regulation skills in adolescence are emotion regulation for adolescents’ socio- also influenced (i.e., facilitated or impaired) emotional adjustment in general, it has also by many other factors, such as varying been proposed that emotion regulation is degrees of parental structure and supervi- involved in the unique emotional lives of sion. Future research needs to provide more adolescents in particular. Compared to chil- empirical evidence on the interplay of these dren and adults, adolescents report more various internal and external factors to negative, more mixed, and more rapidly arrive at a more integrated understanding of influences on emotion regulation in adoles- cence.

198 DEVELOPMENTAL CONSIDERATIONS In comparison to adults from various chapter currently represent relatively inde- age groups, adolescents have been found to pendent bodies of research. A more explicit report more contrahedonic and less prohe- integration of these various research per- donic motivation. This suggests that part spectives will help us to arrive at a more of the negative emotionality that is charac- integrated and interpretable picture of the teristic of adolescence may be intentionally development of emotion regulation during sought and maintained by the individual. adolescence. In addition, future research is Many questions, however, still remain open. necessary to address various limitations of For example, future research will need to the currently available research. For exam- employ well-c­ontrolled experiments to dis- ple, more longitudinal evidence is necessary entangle the causal mechanisms involved in to portray developmental changes within these associations. This will also contribute persons as they move from childhood to to a better understanding of the reasons for adolescence and young adulthood. Such por- adolescent patterns of pro- and contrahe- trayals should also take into account more donic orientations. Another open question explicitly the possibility of nonlinear trends pertains to potential functions of contrahe- in development, such as temporary pertur- donic motivation, for example, related to the bations or deviations from growth trajecto- development of autonomy, self-r­egulatory ries. Furthermore, in addition to adolescents’ skills, or other aspects of socioemotional chronological ages, the role of their pubertal development of adolescents. Finally, rela- status (which can vary substantially by age) tions between adolescents’ emotion regula- should be taken into consideration more tion motivation and their actual engagement comprehensively, because affective pro- in emotion regulatory behaviors, as well as cesses during adolescence are strongly asso- the effectiveness of these emotion regula- ciated with puberty-­related developments tion attempts, need to be explored in future (Steiner, Dunn, & Born, 2003). Finally, investigations. multimethodological enrichment of research approaches would be desirable to overcome Available research on age-­related differ- the methodological limitations associated ences in the use, adaptiveness, and effective- with the currently prevailing dominance of ness of adolescents’ strategies for regulating self-­report approaches in most research on their emotions differs widely in the con- emotion regulation in adolescence. ceptual frameworks employed. This leads to difficulty in deriving a cohesive picture Note about possible developmental changes dur- ing adolescence and adjacent developmental 1. Note that research on age differences in the periods. Tentatively, however, this research frequency of mixed affect during adulthood suggests an age-r­ elated increase in regulation has yielded inconclusive findings. Results competence from childhood to adolescence, from these studies range from age-r­elated as reflected in an increasingly sophisticated decrease, to age invariance, to age-r­elated understanding of emotional situations and a increase in the propensity to experience mixed broadening repertoire of regulatory strate- affect. Heterogeneity between studies regard- gies. This appears to coincide with an age-­ ing the operationalization of mixed affect may related increase in the ability to tailor regu- be among the reasons for the inconclusiveness latory strategies to the specific requirements of findings with this measure (for a review, see of the emotion-e­liciting situation. Future Riediger & Rauers, in press). research should further clarify the complex empirical picture on adolescent development References in the use and adaptiveness of regulatory strategies. Meta-a­nalyses of the available Allen, N., & Sheeber, L. (Eds.). (2009). Adoles- empirical evidence would be helpful in this cent emotional development and the emer- respect. Furthermore, more empirical evi- gence of depressive disorders. New York: dence is needed on differences in the actual Cambridge University Press. effectiveness of regulatory attempts between children, adolescents, and adults. Andrade, E. B., & Cohen, J. B. (2007). On the The three themes on emotion regulation development in adolescence reviewed in this

Emotion Regulation in Adolescence 199 consumption of negative feelings. Journal of Compas, B. E., Connor-Smith, J. K., Saltzman, Consumer Research, 34, 283–300. H., Thomsen, A. H., & Wadsworth, M. E. Bariola, E., Gullone, E., & Hughes, E. K. (2011). (2001). Coping with stress during childhood Child and adolescent emotion regulation: and adolescence: Problems, progress, and The role of parental emotion regulation and potential in theory and research. Psychologi- expression. Clinical Child and Family Psy- cal Bulletin, 127, 87–127. chology Review, 14, 198–212. Bell, K. L., & Calkins, S. D. (2000). Relation- Compton, R. J., Robinson, M. D., Ode, S., ships as inputs and outputs of emotion regula- Quandt, L. C., Fineman, S. L., & Carp, J. tion. Psychological Inquiry, 11, 160–163. (2008). Error-m­ onitoring ability predicts daily Best, J. R., & Miller, P. H. (2010). A develop- stress regulation: Research article. Psychologi- mental perspective on executive function. cal Science, 19, 702–708. Child Development, 81, 1641–1660. Betts, J., Gullone, E., & Allen, J. S. (2009). An Dahl, R. E. (2001). Affect regulation, brain examination of emotion regulation, tempera- development, and behavioral/emotional health ment, and parenting style as potential predic- in adolescence. CNS Spectrums, 6, 60–72. tors of adolescent depression risk status: A correlational study. British Journal of Devel- Dahl, R. E. (2004). Adolescent brain develop- opmental Psychology, 27, 473–485. ment: A period of vulnerabilities and oppor- Blanchard-F­ ields, F., & Coats, A. H. (2008). The tunities. Annals of the New York Academy of experience of anger and sadness in everyday Sciences, 1021, 1–22. problems impacts age differences in emotion regulation. Developmental Psychology, 44, Dishion, T. J., Nelson, S. N., & Bullock, B. M. 1547–1556. (2004). Premature adolescent autonomy: Par- Bowie, B. H. (2010). Understanding the gender ent disengagement and deviant peer process in differences in pathways to social deviancy: the amplification of problem behavior. Jour- Relational aggression and emotion regulation. nal of Adolescence, 27, 515–530. Archives of Psychiatric Nursing, 24, 27–37. Buckley, M., & Saarni, C. (2009). Emotion regu- Eccles, J. S., Templeton, J., Barber, B., & Stone, lation: Implications for positive youth devel- M. (2003). Adolescence and emerging adult- opment. In R. Gilman, E. S. Huebner, & M. hood: The critical passage ways to adulthood. J. Furlong (Eds.), Handbook of positive psy- In M. H. Bornstein, L. Davidson, C. L. M. chology in schools (pp. 107–118). New York: Keyess, & K. A. Moore (Eds.), Well-being: Routledge. Casey, B., Getz, S., & Galvan, A. (2008). The Positive development across the life course adolescent brain. Developmental Review, 28, (pp. 383–406). Mahwah, NJ: Erlbaum. 62–77. Erber, R., & Wang Erber, M. (2000). The self-­ Casey, B., Jones, R. M., Levita, L., Libby, V., regulation of moods: Second thoughts on the Pattwell, S. S., Ruberry, E. J., et al. (2010). importance of happiness in everyday life. Psy- The storm and stress of adolescence: Insights chological Inquiry, 11, 142–148. from human imaging and mouse genetics. Erikson, E. E. (1968). Identity: Youth and crisis. Developmental Psychobiology, 52, 225–235. New York: Norton. Ciarrochi, J., Heaven, P. C., & Supavadeeprasit, Ford, B. Q., & Tamir, M. (2012). When getting S. (2008). The link between emotion identifi- angry is smart: Emotional preferences and cation skills and socio-e­ motional functioning emotional intelligence. Emotion, 12, 685– in early adolescence: A 1-year longitudinal 689. study. Journal of Adolescence, 31, 565–582. Forman, H., Mäntylä, T., & Carelli, M. G. Cole, S. W. (this volume, 2014). Emotion regula- (2011). Time keeping and working memory tion and gene expression. In J. J. Gross (Ed.), development in early adolescence: A 4-year Handbook of emotion regulation (2nd ed., follow-­up. Journal of Experimental Child pp. 571–585). New York: Guilford Press. Psychology, 108, 170–179. Compas, B. E. (2009). Coping, regulation, and Garnefski, N., Legerstee, J., Kraaij, V., van den development during childhood and adoles- Kommer, T., & Teerds, J. (2002). Cognitive cence. New Directions for Child and Adoles- coping strategies and symptoms of depression cent Development, 2009, 87–99. and anxiety: A comparison between adoles- cents and adults. Journal of Adolescence, 25, 603 – 611. Gray, J. R., & Braver, T. S. (2007). Integration of emotion and cognitive control: A neurocom- putational hypothesis of dynamic goal regula- tion. In S. C. Moore & M. Oaksford (Eds.),

200 DEVELOPMENTAL CONSIDERATIONS Emotional cognition: From brain to behavior The role of emotion and emotion regulation in (pp. 289–316). Amsterdam: Benjamins. the ability to delay gratification. In J. J. Gross Grecucci, A., & Sanfey, A. G. (this volume, (Ed.), Handbook of emotion regulation (2nd 2014). Emotion regulation and decision mak- ed., pp. 111–125). New York: Guilford Press. ing. In J. J. Gross (Ed.), Handbook of emo- Matarazzo, O. (2008). Use and effectiveness of tion regulation (2nd ed.). New York: Guilford three modalities of emotion regulation after Press. negative life events: Rumination, distraction Gross, J. J. (1999). Emotion regulation: Past, and social sharing. In A. B. Turley & G. C. present, and future. Cognition and Emotion, Hofmann (Eds.), Life style and health research 13, 551–573. progress (pp. 87–138). Hauppauge, New York: Gross, J. J. (2002). Emotion regulation: Affec- Nova Biomedical Books. tive, cognitive, and social consequences. Psy- Mauss, I. B., Bunge, S. A., & Gross, J. J. (2007). chophysiology, 39, 281–291. Automatic emotion regulation. Social and Per- Gunnar, M. R., Wewerka, S., Frenn, K., Long, sonality Psychology Compass, 1, 146–167. J. D., & Griggs, C. (2009). Developmental McLaughlin, K. A., Hatzenbuehler, M. L., Men- changes in hypothalamus–­pituitary–a­ drenal nin, D. S., & Nolen-H­ oeksema, S. (2011). activity over the transition to adolescence: Emotion dysregulation and adolescent psy- Normative changes and associations with chopathology: A prospective study. Behaviour puberty. Development and Psychopathology, Research and Therapy, 49, 544–554. 21, 69–85. McRae, K., Gross, J. J., Weber, J., Robertson, Hoeksma, J. B., Oosterlaan, J., & Schipper, E. R., Sokol-H­ essner, P., Ray, R. D., et al. E. M. (2004). Emotion regulation and the (2012). The development of emotion regula- dynamics of feelings: A conceptual and meth- tion: An fMRI study of cognitive reappraisal odological framework. Child Development, in children, adolescents and young adults. 75, 354–360. Social Cognitive and Affective Neuroscience, Hughes, E. K., Gullone, E., Dudley, A., & Tonge, 7, 11–22. B. (2010). A case control study of emotion Mesquita, B., De Leersnyder, J., & Albert, D. regulation and school refusal in children and (this volume, 2014). The cultural regulation adolescents. Journal of Early Adolescence, 30, of emotions. In J. J. Gross (Ed.), Handbook 691–706. of emotion regulation (2nd ed., pp. 284–302). Labouvie-­Vief, G., DeVoe, M., & Bulka, D. New York: Guilford Press. (1989). Speaking about feelings: Conceptions Miller, E. M., & Shields, S. A. (1980). Skin con- of emotion across the life span. Psychology ductance response as a measure of adolescents’ and Aging, 4, 425–437. emotional reactivity. Psychological Reports, Larson, R., & Asmussen, L. (1991). Anger, 46, 587–590. worry, and hurt in early adolescence: An Mohanty, A., Engels, A. S., Herrington, J. D., enlarging world of negative emotions. In M. Heller, W., Ho, M. H., Banich, M. T., et al. E. Colton & S. Gore (Eds.), Adolescent stress: (2007). Differential engagement of anterior Causes and consequences (pp. 21–41). New cingulate cortex subdivisions for cognitive and York: de Gruyter. emotional function. Psychophysiology, 44, Larson, R., & Lampman-­Petraitis, C. (1989). 343–351. Daily emotional states as reported by children Morris, A. S., Silk, J. S., Steinberg, L., Myers, S. and adolescents. Child Development and Psy- S., & Robinson, L. R. (2007). The role of the chopathology, 60, 1250–1260. family context in the development of emotion Larson, R., Moneta, G., Richards, M. H., & regulation. Social Development, 16, 361–388. Wilson, S. (2002). Continuity, stability, and Ochsner, K. N., & Gross, J. J. (2005). The cogni- change in daily emotional experience across tive control of emotion. Trends in Cognitive adolescence. Child Development, 73, 1151– Sciences, 9, 242–249. 1165. Ochsner, K. N., & Gross, J. J. (this volume, Lazarus, R. S., & Folkman, S. (1984). Coping 2014). The neural bases of emotion and emo- and adaption. In W. D. Gentry (Ed.), Hand- tion regulation: A valuation perspective. In J. book of behavioral medicine (pp. 282–325). J. Gross (Ed.), Handbook of emotion regula- New York: Guilford Press. tion (2nd ed., pp. 23–42). New York: Guilford Luerssen, A., & Ayduk, O. (this volume, 2014). Press.

Emotion Regulation in Adolescence 201 Opitz, P. C., Gross, J. J., & Urry, H. L. (2012). Rothbaum, F., Weisz, J. R., & Snyder, S. S. Selection, optimization, and compensation (1982). Changing the world and changing the in the domain of emotion regulation: Appli- self: A two-pÂ

202 DEVELOPMENTAL CONSIDERATIONS Silvers, J. A., McRae, K., Gabrieli, J. D., Gross, A theme in search of definition. Monographs J. J., Remy, K. A., & Ochsner, K. N. (2012). of the Society for Research in Child Develop- Age-r­elated differences in emotional reactiv- ment, 59, 25–52. ity, regulation, and rejection sensitivity in ado- Thompson, R. A. (this volume, 2014). Socializa- lescence. Emotion, 12(6), 1235–1247. tion of emotion and emotion regulation in the family. In J. J. Gross (Ed.), Handbook of emo- Sim, L., & Zeman, J. (2006). The contribution tion regulation (2nd ed., pp. 173–186). New of emotion regulation to body dissatisfaction York: Guilford Press. and disordered eating in early adolescent girls. Thorne, A. (2004). Putting the person into social Journal of Youth and Adolescence, 35, 219– identity. Human Development, 47, 361–365. 228. Yap, M. B., Allen, N. B., & Ladouceur, C. D. (2008). Maternal socialization of positive Steinberg, L. (2008). A social neuroscience per- affect: The impact of invalidation on adolescent spective on adolescent risk-­taking. Develop- emotion regulation and depressive symptom- mental Review, 28, 78–106. atology. Child Development, 79, 1415–1431. Yap, M. B., Allen, N. B., & Sheeber, L. (2007). Steiner, M., Dunn, E., & Born, L. (2003). Hor- Using an emotion regulation framework to mones and mood: From menarche to meno- understand the role of temperament and fam- pause and beyond. Journal of Affective Disor- ily processes in risk for adolescent depressive ders, 74, 67–83. disorders. Clinical Child and Family Psychol- ogy Review, 10, 180–196. Stroud, L. R., Foster, E., Papandonatos, G. D., Yurgelun-T­ odd, D. (2007). Emotional and cog- Handwerger, K., Granger, D. A., Kivlighan, nitive changes during adolescence. Current K. T., et al. (2009). Stress response and the Opinion in Neurobiology, 17, 251–257. adolescent transition: Performance versus peer Zeman, J., Cassano, M., Perry-P­arrish, C., & rejection stressors. Development and Psycho- Stegall, S. (2006). Emotion regulation in chil- pathology, 21, 47–68. dren and adolescents. Journal of Developmen- tal and Behavioral Pediatrics, 27, 155–168. Sumter, S., Bokhorst, C., Miers, A., Van Pelt, Zimmer-­Gembeck, M. J., & Skinner, E. A. J., & Westenberg, P. (2010). Age and puberty (2011). Review: The development of coping differences in stress responses during a public across childhood and adolescence: An integra- speaking task: Do adolescents grow more sen- tive review and critique of research. Interna- sitive to social evaluation? Psychoneuroendo- tional Journal of Behavioral Development, crinology, 35, 1510–1516. 35, 1–17. Tamir, M. (2009). What do people want to feel and why? Pleasure and utility in emotion regu- lation. Current Directions in Psychological Science, 18, 101–105. Thompson, R. A. (1994). Emotion regulation:

Chapter 13 Emotion Regulation and Aging Susan Turk Charles Laura L. Carstensen Developmental psychologists have docu- clear that circumstances that restrict strat- mented age-­related changes in the regulation egy usage may have especially dire outcomes of emotions across childhood that paral- for older adults. lel biological maturation. As children and adolescents acquire and develop abilities For our purposes herein, we define emo- to control impulses and gain awareness of tion regulation as efforts to experience rela- themselves and others, they begin to achieve tively low levels of negative affect and elicit mastery over their environments and become relatively high levels of positive affect (see increasingly effective at describing and regu- Gross, this volume, for a more complete lating their emotions. Brain maturation and description of emotion regulation). The patterns of neurological functioning related study of age differences in emotion regula- to emotional processes continue to develop tion has grown, although direct comparisons throughout childhood and adolescence. across the lifespan remain limited. Similarly, Greater inhibitory control, for example, is studies that focus on thoughts and behav- reflected in increased activity in the prefron- iors that enable people to avoid or mitigate tal cortex from childhood into the early 20s exposure to distressing events rarely link (e.g., McRae et al., 2012). In the latter half such strategies to affective outcomes. Only of the lifespan, however, the role of biologi- a few studies, which we describe below, cal development in emotional experience and examine age differences in regulation when regulation is relatively understudied. Physi- people are intentionally increasing (i.e., up-­ ological functioning is gradually degraded regulating or amplifying) or decreasing (i.e., with age, and it has long been assumed that down-r­ egulating) their positive and negative such changes affect emotional functioning. emotional states. A growing literature, however, suggests that aging is associated with relatively positive We focus our review on normal aging as emotional experience in everyday life. As the opposed to disease-­related aging, and on the understanding of emotional aging improves, general experience of positive and negative it is becoming increasingly clear that older affect as opposed to discrete emotions. We people typically engage in strategies that begin with a brief overview of age differ- maintain positive emotional experiences ences in physiological processes involved in and limit negative ones. It is also becoming emotional experience. We then describe the scope of the existing literature examining age and affective well-being. Next, we pres- 203

204 DEVELOPMENTAL CONSIDERATIONS ent socioemotional selectivity theory (SST) Age Changes in Brain Structure and strength and vulnerability integration and Functioning (SAVI), both of which guide our understand- Researchers have documented normative ing of emotional development in adulthood. age-­related changes in both brain structure SST has received considerable empirical and functioning. Brain volume decreases support since it was proposed in the early with age, with reductions in the synaptic 1990s (see Charles & Carstensen, 2010). Its density of white matter and a consequent core tenets concern the ways that time hori- slowing of neural processing that accom- zons influence goals. SST initially addressed panies demyelination (e.g., Westlye et al., emotion somewhat indirectly, based on the 2010). These age-r­elated decreases are premise that as people become increasingly observed after the fourth decade of life, and focused (selectivity) on emotionally mean- exhibit an accelerating decline beginning in ingful goals, well-being likely benefits. As the early to mid-60s (Westyle et al., 2010). evidence for social selection accrued along with evidence that emotional well-being is Brain activation in response to emotional relatively high in later life, we began to con- stimuli shifts throughout childhood and into sider selection as a highly effective form of early adulthood toward less activation in the antecedent emotion regulation (Carstensen, amygdala and greater activation of the pre- Fung, & Charles, 2003). SAVI, in contrast, frontal cortex, findings that researchers use was proposed relatively recently. It aims to to explain developmental increases in emo- integrate the relatively positive profile of tion regulation abilities (McRae et al., 2012). emotional well-being with concurrent evi- The same difference in relative activity of dence of repercussions for older adults who these brain regions is found in comparisons experience sustained and inescapable expo- of younger and older adults, a phenomenon sure to toxic situations. We review empirical referred to as the posterior–a­ nterior shift in findings that are consistent with the premise aging (PASA; Davis, Dennis, Dasellar, Fleck, that motivational shifts lead to differences & Cabeza, 2008). In memory and percep- in attention, memory, social partner prefer- tual studies, with and without emotional ences, and problem-s­olving strategies that stimuli, older adults show greater activity in serve to maintain or optimize emotional the prefrontal regions and reduced activity experience. Next, we discuss situations pos- in medial and temporal regions compared to ited by SAVI, in which benefits of age may younger adults (Davis et al., 2008). be less pronounced and may even reverse in direction. We then review areas that we pre- Age Changes in Cardiovascular viously highlighted as especially important and Neuroendocrine Functioning and how they have been addressed since the Cardiovascular and neuroendocrine previous edition of this volume. We conclude responses also change with age. Although that the emerging literature suggests that age the overall pattern of cardiovascular reactiv- differences in emotional experience, cogni- ity to stressors is the same, aging is related tive processes, and physiological functioning to relative differences in the degree of this operate in concert to yield relatively well-­ reactivity. Specifically, older age is related to preserved if not enhanced emotional well- a smaller increase in heart rate in response being in later life. to stressors but greater increases in blood pressure reactivity (see review by Uchino, Emotion and the Aging Birmingham, & Berg, 2010). The reduced Physiological System heart rate reactivity most likely represents Emotional experience, like all psychological age-r­elated reductions in heart rate variabil- phenomena, is inextricably linked to physi- ity (HRV) that are normative and begin in ological functioning. Aging in three bio- adolescence, and that are a risk factor for logical systems—­the central nervous system mortality (Zulfigar, Jurivich, Gao, & Singer, (specifically the brain), the cardiovascular 2010). system, and the neuroendocrine system—i­s relevant for emotional aging. The neuroendocrine response also varies with age (Björntorp, 2002). The perception of a real or imagined threat mobilizes the

Emotion Regulation and Aging 205 body for action through a series of reac- depressive symptoms in advanced old age tions along the hypothalamic–Âp

206 DEVELOPMENTAL CONSIDERATIONS people do experience emotions, the inten- their goals as they age, so that experiences sity is similar to that of younger people. are more emotionally meaningful and satis- Older and middle-aÂ

Emotion Regulation and Aging 207 situations, presumably because there is preference for negative material in the young informational value in negative experiences as a preference for positive material in the that prepares people for the future. When old. The positivity effect refers to the age-­ confronted with minor events, older adults related pattern of change. appraise them as less stressful (e.g., Charles & Carstensen, 2008). When older adults Brain Activity and the Positivity Effect cannot employ strengths of aging to avoid or Studies of brain activity during the process- mitigate their exposure to a negative event, ing of emotional information support the however, they experience distress. During positivity effect. Mather et al. (2004) specu- these demanding and highly arousing situ- late that subconscious changes in motivation ations, SAVI posits that the age-r­elated ben- influence activation of the amygdala, con- efits disappear or even reverse in direction. sistent with their finding that older adults Because age is related to biological changes show greater amygdalar activity when view- that present greater threats to a less robust ing positive rather than negative images. physiological system, SAVI posits that high Other brain regions also show age differ- levels of physiological reactivity and conse- ences in activation depending on the emo- quent recovery in response to sustained neg- tional valence of the information (Samanez-­ ative arousal are more costly to older adults. Larkin et al., 2007). In a study examining brain activity in response to decision mak- Age, Cognitive Functioning, ing, older adults exhibited less activation in and Antecedent‑Focused insula and caudate regions than did younger Emotion Regulation adults when they anticipated loss, but Emotional Salience similar activation when anticipating gains and Emotion Regulation (Samanez-L­ arkin et al., 2007). Brains do not operate like computers, treat- ing all information equally. Rather, cogni- The general PASA in processing infor- tive resources are directed to goal-­relevant mation discussed previously (St. Jacques, information. SST posits that older adults Bessette-­Symons, & Cabeza, 2009) may place more importance and prominence on serve emotion regulation goals. For exam- emotional than on nonemotional aspects ple, one study revealed that older adults of stimuli they encounter in the labora- display less amygdalar activity and greater tory or in their daily lives, and they do so activation of areas associated with emotion without explicit awareness of their actions. regulation (prefrontal cortex) when viewing Laboratory studies using incidental memory negative images compared to positive images paradigms reveal that the proportion of relative to younger adults (Leclerc & Kens- emotional versus nonemotional informa- inger, 2011). Another study revealed age tion increases linearly as a function of age differences in connectivity of the amygdala (Carstensen & Turk-C­harles, 1994), and to other brain regions, consistent with age-­ emotional details are recalled better than related increases in emotion regulation pro- perceptual details of stories (e.g., Mather, cesses (St. Jacques, Dolcos, & Cabeza, 2009, Johnson, & De Leonardis, 1999). In addi- 2010). Specifically, older adults showed more tion, the salience of emotion is greater with activity in a brain region associated with age when people recall autobiographical emotion regulation (ventral anterior cingu- information (Alea, Bluck, & Semegon, late cortex), and less activity in areas asso- 2004). Overall, this pattern of findings sug- ciated with perceptual processing (visual gests that older adults pay attention to emo- cortex) than did the younger adults (St. tionally relevant information that in and of Jacques et al., 2010). Age-r­elated reductions itself may aid emotion regulation. in activities related to perceptual processing are consistent with the well-e­ stablished age-­ Rather than “positivity bias” we coined related reductions in facial emotion recogni- the term positivity effect to describe the tion in the literature (e.g., Lambrecht, Krei- age-­related pattern of change (Carstensen & felts, & Wildgruber, 2012). Similar patterns Mikels, 2005; Mather & Carstensen, 2005). emerge with brain activation and consequent Age differences are as frequently driven by a memory for these pictures (St. Jacques et al., 2009). Older adults exhibit greater activity

208 DEVELOPMENTAL CONSIDERATIONS in frontal regions associated with emotional adults (e.g., Scheibe & Blanchard-­Fields, control (bilateral dorsolateral prefrontal cor- 2009). In one study, Scheibe and Blanchard-­ tex) and less activity in regions associated Fields asked younger and older adults to with later memory encoding (hippocampus perform a working memory task after hav- and bilateral ventrolateral prefrontal corti- ing been induced into a negative mood and ces) for negative stimuli relative to neutral asked to down-r­egulate or maintain current stimuli than do younger adults. Behavioral mood, or were provided no specific instruc- data again reflected these findings: Older tions. Although the performance of younger adults were less likely than younger adults adults deteriorated when they were asked to to remember the negative stimuli relative to down-­regulate their emotions, the perfor- the neutral stimuli (St. Jacques et al., 2009). mance of older adults did not vary from that in the other conditions. Similarly, another Age‑Related Changes in Attention study found that emotional suppression was and Working Memory related to worse memory for younger adults The positivity effect explains how chronic but not for older adults (Emery & Hess, activation of emotion-­related goals directs 2011). cognitive resources systematically toward positive and away from negative stimuli—­ Evidence That Memory Supports often without conscious awareness. Older Emotion Regulation adults focus their attention on more posi- Antecedent-­focused emotion regulation tive and less negative stimuli than do strategies prevent or reduce exposure to younger adults in studies of simple atten- events and thoughts that threaten well- tion, as assessed by reaction times (Mather being. To the extent that memories influence & Carstensen, 2003) and eye gaze (e.g., current mood states, thoughts, and behav- Isaacowitz, Toner, Goren, & Wilson, 2008; ior (see review by Levine & Pizarro, 2004), Nikitin & Freund, 2011). memory can be a powerful regulation strat- egy. This shift in valence toward positive SST predicts that top-down motivational information with age has been documented changes with age lead to changes in atten- in studies examining memory for both life tion and memory that serve emotion regu- events and laboratory stimuli. Memory lation goals, such that performance may be about past events is more positive for older better for emotional information than for compared to younger adults (Kennedy, nonemotional information. One such exam- Mather, & Carstensen, 2004). Even when ple lies in working memory, which is vital asked about prior negative events, older for emotional functioning and planning, adults report them as having been more because working memory allows an individ- positive than do younger adults (Schryer & ual to focus on the potential for rewards and Ross, 2012). enhances goal-­seeking behavior (Davidson, Jackson, & Kalin, 2000). In a test of SST, Laboratory studies underscore the shift the working memory performance of older toward older people remembering events and younger adults was examined on tasks more positively. In a study in which younger, in which participants compared images middle-a­ ged, and older adults viewed posi- based on either their brightness intensity tive, negative, and neutral images, and were or emotional intensity (Mikels, Larkin, later asked to recall and to recognize these Reuter-L­ orenz, & Carstensen, 2005). Find- previously viewed images from a larger set ings reveal the usual age-­related decrement of images, younger adults recalled similar in working memory for visual brightness but levels of positive and negative stimuli, and no age difference in working memory for they recognized a greater proportion of neg- affective arousal. Moreover, the lack of age ative images than positive images (Charles, differences was driven by older adults’ bet- Mather, & Carstensen, 2003). Older adults ter performance for positive stimuli than for displayed no such negative bias in their recog- negative stimuli compared to younger adults nition, and they recalled significantly fewer (Mikels et al., 2005). Other studies suggest negative than positive images (Charles et that emotion regulation is less cognitively al., 2003). This age-a­ ssociated shift toward taxing for older adults than for younger positive emotion and away from negative

Emotion Regulation and Aging 209 information has been replicated in a num- the plans; that is, they gave the participants ber of studies from many laboratories (e.g., a specific goal, namely, to pursue informa- Petrican, Moscovitch, & Schimmack, 2008; tion. And in this condition, age differences Piguet, Connally, Krendl, Huot, & Corkin, were eliminated. Reed and Carstensen 2008; see reviews by Reed & Carstensen, (2013) recently reviewed the positivity effect 2012; Scheibe & Carstensen, 2010). literature and concluded that positivity pref- erences are strongest during controlled pro- Cognitive Appraisal of Negative Events cessing when motivations are hypothesized Older adults not only focus their attention to play a strong role (for discussion, see on less negative aspects of the environment Kensinger, 2004). but they also appraise ambiguous or unpleas- ant events more benignly than do younger Whether or not positivity in cognitive pro- adults. Older adults perceive daily stressors cessing is adaptive or maladaptive depends as less severe and less threatening than do on the task. Certainly there are situations younger adults in a sample ranging from where selective attention to positivity may 25 to 74 years (Charles & Almeida, 2007). disadvantage performance. Decision mak- They appraise aversive stimuli less nega- ing likely benefits from evenhanded process- tively than do younger adults, and are more ing of advantages and disadvantages associ- likely to focus their thoughts away from the ated with options. Focusing on the positive negative content (Charles & Carstensen, interpersonal qualities of a charming con- 2008). Even when asked about their diag- man could have disastrous financial conse- nosis and treatment of cancer, older adults quences. Ignoring negative information or report lower levels of threat appraisals (e.g., appraising a situation as less negative than the amount of tension people experience in it is may lead to short-term benefits to well- response to their illness) and are more likely being but longer term problems that lead than younger adults to frame the event as to distress (Scheibe & Carstensen, 2010). a challenge rather than a threat (Hart & Focusing on positive aspects and avoiding Charles, 2013). negative realities of fiscal concerns may lead to “optimistic, but not necessarily realistic, The Positivity Effect Is Malleable financial planning” (Weierich et al., 2011, SST maintains that—all things being equal—­ p. 197). Ignoring negative information may chronically activated goals differ for older contribute to misremembering negative and younger people. However, because posi- information as more positive than it actually tivity reflects top-down motivational shifts was (Shamaskin, Mikels, & Reed, 2010). as opposed to bottom-­up brain-d­ riven shifts, the effect is theoretically malleable. Younger Finally, there is some evidence that not people sometimes pursue emotional goals, only emotional valence but also arousal con- and older adults sometimes pursue informa- tribute to age differences in positivity. Grühn tional goals. In circumstances where stakes and Scheibe (2008) found that older people are high, younger and older people may perceived positive pictures in the Interna- pursue goals that direct attention toward or tional Affective Picture System (IAPS) as away from emotional material accordingly. more positive and less arousing than did Indeed, Löckenhoff and Carstensen (2007) younger adults. Keil and Freund (2009) later examined whether experimental instruc- observed that older adults prefer less arous- tions provided to participants would elimi- ing positive and negative images and words nate age differences. In an initial phase of than do younger adults, and Streubel and the study, participants were simply asked Kunzmann (2011) reported that the subjec- to review information about a variety of tive experience of positivity in older adults alternative health care plans. In this condi- is reduced under high arousal. Using experi- tion, older people focused more on positive ence sampling data, Scheibe, English, Tsai, than on negative characteristics. In a second and Carstensen (2013) recently examined condition, researchers explicitly asked older the effect of age differences in the value adults to focus on informational aspects of placed on high- versus low-­arousal states and concluded that older people placed more value on low-­arousal positive states and were more successful in achieving them than their younger counterparts. Taken together, these

210 DEVELOPMENTAL CONSIDERATIONS findings suggest that older adults may orient 2001; but see Villamil et al., 2006). On the to positive material that is more calming and other hand, older people face many chal- peaceful than to positive material associated lenges that younger people are less likely to with excitement and surgency. face, such as chronic health problems, loss of loved ones, and age discrimination. Social Partners and Emotion Regulation Although questions about environmental This review has focused largely on age-­ demands are important and there is some related changes in cognitive processes that evidence that changes in life circumstances serve emotion regulation goals. SST also may contribute to the decreased number of makes specific predictions about how age-­ stressors older adults report in their daily related behaviors, and explicitly changes in lives (e.g., Charles & Almeida, 2007), there social partner preferences and interaction is evidence that older adults are also active patterns, serve emotion regulatory goals. agents in influencing their well-being. Com- Older adults have smaller and more care- pared to younger people, older adults report fully pruned social networks that contain better control over their emotions (Gross larger percentages of emotionally close et al., 1997), enhanced abilities to regulate social partners than do those of younger emotions in general (Kessler & Staudinger, adults (Lang & Carstensen, 1994). Younger 2009), as well as more success at calming peoples’ networks, by way of contrast, often strategies and other thoughts to control their include many social partners who were not feelings of anger (Phillips, Henry, Hosie, & chosen freely but have been incorporated Milne, 2008). Although the veracity of sub- because of relationships to work or offspring jective reports alone may be questioned, they (Lang & Carstensen, 1994; Lansford, Sher- are consistent with low levels of psychiatric man, & Antonucci, 1998). Thus, the social disease and reported experience, and behav- networks of older adults may be easier to ioral evidence in the literature. For exam- navigate emotionally. Because most strong ple, when people ranging in age from 18 to emotions occur in social contexts, careful 94 years reported the emotions they were selection of social partners is arguably a experiencing at five random times each day highly effective strategy. When faced with throughout the course of a week, the proba- interpersonal conflict in their networks, bility of continuing to feel a negative emotion older adults engage in more conflict avoid- from one reporting interval to the next was ance, with the goal of preserving harmony in lower with advancing age (Carstensen et al., their relationships, than do middle-­aged and 2000). Researchers have discussed how older younger adults (Blanchard-F­ields, Mienal- adults select and are adept at using emotion towski, & Seay, 2007). regulation strategies given their decrements in others area of their lives, such as cognitive Responding to Negative Events functioning (Urry & Gross, 2010). The previously discussed literature points to older adults using—e­ither knowingly or Questions remain, however, as to how not—t­houghts and behaviors that fall under well older adults regulate their emotions the category of antecedent-­focused emo- when they cannot avoid or mitigate expo- tion regulation strategies. Some researchers sure to a negative event. SAVI posits that argue, however, that older adults’ greater when older adults cannot avoid experienc- well-being reflects not actions by the older ing high and sustained levels of emotional adults but instead age-­related differences distress, this arousal reduces any age-­related in environmental demands (see review by benefits previously observed for regulation Lawton, 2001). According to this view, the abilities. One study explicitly compared age end of child-­rearing responsibilities and the differences in affective outcomes in two situ- onset of retirement decreases the frequency ations: one in which people disengaged from of negative events more in the lives of older a negative event and another in which they than younger adults (see review by Lawton, did not (Charles, Piazza, Luong, & Almeida, 2009). When older adults encountered an unpleasant situation in which they chose to avoid further conflict, they reported lower levels of affective reactivity than did younger adults. When they engaged in a conflict, no

Emotion Regulation and Aging 211 age differences in reactivity were apparent tive affect in older (60 to 75 years old) (Charles et al., 2009). and younger (20 to 30 years old) adults in response to a film clip that elicited disgust Other studies also suggest that high lev- (Scheibe & Blachard-­Fields, 2009). After els of arousal reduce age-r­elated benefits having viewed the film clips, participants in emotional outcomes and may even show were given instructions to attempt to change age-r­elated decrements. When older and the negative feelings into positive ones or to younger adults report stressors of equally continue feeling their negative emotions and high severity, older adults display greater not try to change them, or they were given no affective reactivity than do younger adults instructions. Results indicated no significant (Mroczek & Almeida, 2004). Longitudinal age differences in the reported affect during data confirm this age-r­ elated pattern, in that the recovery period. However, older adults when people report experiencing a stressor, showed less deterioration in performance on affective reactivity increases with age (Sli- one of the three trials of a n-back task fol- winski, Almeida, Smyth, & Stawski, 2009). lowing the induction, suggesting that fewer resources were required for regulation. The studies described earlier use informa- tion gathered from the daily lives of adults. A third study examined three types of Researchers attempting to capture emo- regulation strategies in older (60–69 years), tion regulation in the laboratory face the middle-­aged (40–49 years) and younger challenge of devising paradigms that elicit adults (20–29 years) in response to emo- responses sufficiently strong to demand tional film clips (Shiota & Levenson, 2009). active and sustained regulatory efforts. The three strategies were detached appraisal, Although use of weak stimuli, such as pic- in which participants were told to watch tures of facial expressions, has become objectively in order to feel less negative emo- commonly accepted, we strongly question tion; positive reappraisal, in which they the validity of this approach when research- were told to think about positive aspects of ers are interested in participants’ efforts to what they were viewing so that they would regulate emotional responses actively. Stron- feel less negative emotion; or emotional sup- ger inductions are possible with film clips, pression, in which they were asked to hide personal memories, music, and interactions negative facial expressions. When compar- with confederates or spouses. Of course, a ing subjective reports across different con- tradeoff exists between carefully controlled ditions within each age group, researchers elicitors and responsivity. Only a handful of found that detached appraisal resulted in such studies that focus on age differences in significantly lower levels of distress among regulation effectiveness have been published the younger adults, marginally signifi- to date. cant lower levels of distress in middle-a­ ged adults, but the same level of distress in older In one study, subjective reports of distress adults when compared to the “just watch” were compared across groups of younger (18 condition. Only older adults showed sig- to 40 years old) and older adults (60 to 88 nificantly lower levels of subjective distress years old) who watched emotional film clips in the positive reappraisal condition com- with instructions to react naturally, suppress pared to the “just watch” condition. When all facial expressions, or positively refocus examining mean levels of the target emo- on a pleasant memory if they felt any nega- tions (sadness or disgust) across age groups, tive emotions (Phillips et al., 2008). Results subgroups reporting the lowest mean levels indicate that older adults reported less were the youngest and middle-a­ged adults intense negative affect in the positive reap- in the detached appraisal group (3.25 and praisal condition compared to the other con- 4.08, respectively), and the two groups with ditions. When comparing across age groups, the highest reported target emotions were older and younger adults had similar ratings the middle-­aged and oldest adults in the in the positive reappraisal condition, but positive reappraisal groups (5.42 and 5.20, younger adults reported less negative affect respectively). than older adults in the conditions where they were instructed to react naturally or to One recent study compared younger (18- suppress their facial expressions. to 30-year-olds) and older (60- to 80-year- olds) participants’ responses as they viewed Another study examined the baseline, reactivity, and recovery ratings of nega-

212 DEVELOPMENTAL CONSIDERATIONS negative and neutral images in three condi- fewer studies using human samples (Björn- tions: a “just watch” condition, in which torp, 2002). Still, findings suggest that nega- they were told simply to look at the image; a tive affect is related more strongly to cortisol condition in which each image was preceded output for older than for younger adults. by a string of digits they need to remember For example, one study found that high lev- immediately after their 10-second exposure els of negative affect were related to higher to the image; and a condition in which they level of daily cortisol output for relatively were instructed to reinterpret the picture in older participants (54 to 74 years old), but such a way that they would feel less negative unrelated to this biomarker for people 53 affect (Tucker, Feuerstein, Mende-­Siedlecki, and younger (Piazza, Charles, Stawski, & Ochsner, & Stern, 2012). After each image, Almeida, 2013). participants rated how much negative emo- tion they felt. Findings indicate that younger When Avoiding Negative adults reported less negative reactivity in Emotions Does Not Occur: the reappraisal condition than did the older Aging with Negative Affect adults, but age groups were similar in levels SAVI posits that affective well-being remains of reactivity in the distraction condition. relatively stable and may increase with age, with one important caveat. When older In summary, few studies have examined people find themselves in unavoidable cir- emotion regulation strategies that target cumstances that elicit sustained physiologi- decreasing subjective distress when people cal arousal, the consequences for well-being are exposed to unpleasant stimuli under con- will be exacerbated. Researchers have long trolled laboratory conditions. Findings from discussed the powerful role of the environ- the handful of existing studies do not point to ment in shaping emotion-r­elated outcomes consistent evidence for age-­related improve- and how age is related to different life cir- ments or impairment in emotion regulation. cumstances. These circumstances are pos- As we note below, however, we believe that ited to explain why not all groups of older more of this type of research is needed. adults show continuing, positive trajecto- ries of emotional experience. Loss of social Physical Costs of Emotion Regulation belonging, prolonged caregiving, and neuro- SAVI further posits that circumstances logical dysregulation may result in an upturn marked by higher levels of sustained arousal in depressive symptoms and negative affect will have greater physiological consequences in the last decades of life (Charles, 2010). for older adults. Support for this tenet comes For example, in the last years of life—often from literature examining both cardiovas- characterized by neurological and physi- cular and HPA reactivity responses to emo- ological dysregulation—­there are reliable tional events in younger and older adults. declines in emotional well-being. Indeed, For example, blood pressure reactivity is Gerstorf and colleagues (2010) found that more pronounced in older adults relative terminal decline is better predicted by close- to younger adults in response to stressors ness to death than by chronological age. On of daily life and laboratory-b­ ased stressors the one hand, age and closeness to death such as social evaluative threat (Ong, Roth- are strongly coupled in the population, but stein, & Uchino, 2012; Uchino et al., 2010) Gerstorf and his colleagues found that years and when dealing with cognitively com- from birth may be less important than the plex stressors of daily life (Wrusz, Müller, circumstances that define the years to death Wagner, Lindenberger, & Riediger, 2013). in predicting emotional well-being. In addition, the effect of a more prolonged negative experience—l­ oneliness—i­ s more In addition, experiencing chronically high strongly related to blood pressure recovery levels of trait-n­ egative affect also makes peo- in older than in younger adults (Ong et al., ple more vulnerable to emotional distress 2012). with age. For example, two studies have found that declines in subjective reports of Age-­related changes in HPA reactivity negative affect do not continue into old age have been studied less frequently. Although for those high in neuroticism (Charles et al., researchers document prolonged HPA reac- 2001; Ready, Akertedt, & Mroczek, 2012). tivity in animal studies, they have conducted

Emotion Regulation and Aging 213 Future Directions and Conclusions 2012). It will be important to consider both Empirical findings amassed over the past 30 valence and arousal. We mentioned the need years have dramatically changed the land- to examine discrete emotions in the previ- scape of research on emotion and aging. ous edition of this volume, and provided Among healthy adults, emotional experi- the example that no large study to date has ences and emotion regulation do not decline reported increases in levels of anger with age, across adulthood but are instead well-­ but one study did observe age differences in maintained and, in some ways, improve. We reports of depressive symptoms (e.g., Teach- review below the aims for future studies we man, 2006) although others have not. A recommended 5 years ago. growing number of studies has examined discrete emotions (Grühn et al., 2010) and We stated that future studies will need to especially emotion recognition (e.g., Slessor, link hypothesized emotion regulation strat- Miles, Bull, & Phillips, 2010), yet theories egies (e.g., attention allocation) directly to still focus mostly on positive and negative emotion regulation. This area has begun to valence and arousal when examining emo- receive some attention (e.g., Isaacowitz & tion regulation processes (see Charles & Blanchard-­Fields, 2012), but more work and Carstensen, 2010). conceptual clarity are needed. SST, for exam- ple, focuses on selection as the key mecha- One methodological issue that frequently nism involved in greater well-being with age, arises in lifespan developmental research is and views selective attention and memory the issue of cohort effects. Cross-s­ectional as consequences of changes in goals. In the research points to potentially interesting theoretical context of SST, the positivity developmental phenomena, but only longi- effect, rather than being a coping strategy, tudinal research can confirm developmen- reflects goal-­directed attention. Whether or tal processes postulated by socioemotional not older people also deplore positive atten- selectivity theory. Researchers have made tion in response to distress remains an open great strides in longitudinal investigation of question. SAVI posits that when selection more complex emotional stimuli in the past is not possible, age advantages disappear 5 years (e.g., Carstensen et al., 2011; Ger- or even reverse. Further study is needed to storf et al., 2010; Sliwinski et al., 2009) and examine how emotion regulation strategies the maintenance of large longitudinal data- and effectiveness change with age, relative sets promises to illuminate the field in this to the way age reflects both birth cohort important area. and age-a­ssociated experience. In addi- tion, we mentioned the need to study mul- Methodological innovations in the past tiple response-­focused regulation strategies 5 years have advanced the field in several and include conditions specifying main- ways. For example, functional imaging of tenance, exacerbation, and attenuation of brain activity has illuminated associations emotional experience in the investigations. between behavior and brain activity, and As we reviewed earlier, several studies have promises to provide insights into the neuro- begun to address different types of strategies logical correlates of age differences. As this among people of different ages (e.g., Shiota area grows, the study of emotion and aging & Levenson, 2009). We hope that research- will continue to benefit. A second innova- ers will continue to investigate age differ- tion that will shape the future of psycho- ences in online emotion regulation tasks. logical science is the growing area of health psychology and new techniques allowing We mentioned the need to examine the for affordable assessments of cardiovascu- importance of the arousal of emotional stim- lar, neuroendocrine, and immunological uli when studying age differences (Wurm, responses in the laboratory and in daily life. Labouvie-V­ ief, Aycock, Rebucal, & Koch, With these studies of physiological stabil- 2004), as well as the complexity of emo- ity and reactivity, researchers will be able tional stimuli (Charles, 2005). Researchers to examine how thoughts and behaviors are who take into account the role of arousal related to physiological arousal and reactiv- in processing and regulating emotions find ity accompanying emotional experience. In that more arousing conditions may be more addition, smart phones provide the ability to challenging for older adults (e.g., Ong et al., assess cognitive functioning rapidly, along with momentary sampling of emotion across

214 DEVELOPMENTAL CONSIDERATIONS the course of the days, which provides more References online collection of cognitive and emotional experiences (see Riediger, Schmiedek, Wag- Alea, N., Bluck, S., & Semegon, A. B. (2004). ner, & Lindenberger, 2009). Young and older adults’ expression of emo- tional experience: Do autobiographical nar- Finally, psychologists have long recog- ratives tell a different story? Journal of Adult nized the importance of the environment Development, 11, 235–250. in shaping behavior (see review by Hess, 2005), but we need to continue to recognize Björntorp, P. (2002). Alterations in the ageing the importance of context when interpret- corticotropic stress-r­ esponse axis. In Novartis ing age differences. Although notable excep- Foundation Symposium (Ed.), Novartis Foun- tions exist, such as studies examining age dation Symposium 242—Endocrine facets of differences in social cognition (e.g., Hess, ageing (pp. 46–58). London: Wiley. 2005) and in health-­related contexts (e.g., Consedine & Magai, 2006), further stud- Blanchard-­Fields, F., Mienaltowski, A., & Seay, ies are needed to understand how similar R. B. (2007). Age differences in everyday behaviors or cognitions may have different problem-s­olving effectiveness: Older adults outcomes according to age-­specific contexts. select more effective strategies for interper- Perceptions are guided by experiences, and sonal problems. Journals of Gerontology: Psy- these experiences may alter the meaning and chological Science, 62, P61–P64. interpretation of emotional stimuli. A pic- ture of a surgical procedure, for example, Carstensen, L. L. (1992). Social and emotional may elicit more sadness in older adults but patterns in adulthood: Support for socio- more disgust in younger adults (e.g., Kun- emotional selectivity theory. Psychology and zmann & Grühn, 2005). The death of a par- Aging, 7, 331–338. ent is often an imagined event for a college student and a long past reality for an older Carstensen, L. L. (2006). The influence of a adult. To this end, the study of emotions and sense of time on human development. Science, aging is an examination of how cognition 312, 1913–1915. and behavior shape, and are shaped by, the interplay between physiological processes Carstensen, L. L., Isaacowitz, D. M., & Charles, and the environment across the lifespan. S. T. (1999). Taking time seriously: A theory of socioemotional selectivity. American Psy- In conclusion, emotions arouse physi- chologist, 54, 165–181. ological systems and direct attention, and they motivate people to action. Emotions Carstensen, L. L., Fung, H., & Charles, S. are critical for successful adaptation across (2003). Socioemotional selectivity theory and the lifespan, but developmental processes the regulation of emotion in the second half alter multiple facets of emotional experi- of life. Motivation and Emotion, 27, 103–123. ence, including cognitive appraisals, behav- ioral responses, physiological reactivity, and Carstensen, L. L., & Mikels, J. A. (2005). At the environmental context. Research on emo- intersection of emotion and cognition: Aging tion and aging has revealed that shifts in and the positivity effect. Current Directions in cognitive processing can offset physiologi- Psychological Science, 14, 117–121. cal decline, such that even though biologi- cal functioning peaks relatively early in life, Carstensen, L. L., Pasupathi, M., Mayr, U., & emotional experience remains vital and well Nesselroade, J. R. (2000). Emotional experi- maintained into late adulthood. ence in everyday life across the adult lifespan. Journal of Personality and Social Psychology, Acknowledgments 79, 644–655. This work was supported by National Institute Carstensen, L. L., Turan, B., Scheibe, S., Ram, on Aging Grant No. R37-AGO8816 to Laura L. N., Ersner-­Hershfield, H., Samanez-L­arkin, Carstensen. G. R., et al. (2011). Emotional experience improves with age: Evidence based on over 10 years of experience sampling. Psychology and Aging, 26, 21–33. Carstensen, L. L., & Turk-­Charles, S. (1994). The salience of emotion across the adult lifes- pan. Psychology and Aging, 9, 259–264. Charles, S. T. (2005). Viewing injustice: Age dif- ferences in emotional experience. Psychology and Aging, 20, 159–164.

Emotion Regulation and Aging 215 Charles, S. T. (2010). Strength and Vulnerabil- sequences of expressive regulation in older ity Integration (SAVI): A model of emotional adults. Psychology and Aging, 26, 388–396. well-being in later adulthood. Psychological Gerstorf, D., Ram, N., Mayraz, G., Hidajat, Bulletin, 136, 1068–1091. M., Lindenberger, U., Wagner, G. G., et al. (2010). Late-life decline in well-being across Charles, S. T., & Almeida, D. M. (2007). adulthood in Germany, the United Kingdom, Genetic and environmental effects on daily and the United States: Something is seriously life stressors: More evidence for greater varia- wrong at the end of life. Psychology and tion in later life. Psychology and Aging, 22, Aging, 25, 577–485. 231–240. Gross, J. J. (this volume, 2014). Emotion regula- tion: Conceptual and empirical foundations. Charles, S. T., & Carstensen, L. L. (2008). In J. J. Gross (Ed.), Handbook of emotion Unpleasant situations elicit different emo- regulation (2nd ed., pp. 3–20). New York: tional responses in younger and older adults. Guilford Press. Psychology and Aging, 23, 495–504. Gross, J. J., Carstensen, L. L., Pasupathi, M., Tsai, J., Götestam Skorpen, C., & Hsu, A. Charles, S. T., & Carstensen, L. L. (2010). Y. C. (1997). Emotion and aging: Experi- Social and emotional aspects of aging. Annual ence, expression, and control. Psychology and Review of Psychology, 61, 383–409. Aging, 12, 590–599. Grühn, D., Kotter-Grühn, D., & Röcke, C. Charles, S. T., Mather, M., & Carstensen, L. L. (2010). Discrete affects across the lifespan: (2003). Aging and emotional memory: The Evidence for multidimensionality and multidi- forgettable nature of negative images for older rectionality of affective experiences in young, adults. Journal of Experimental Psychology: middle-­aged and older adults. Journal of General, 132, 310–324. Research in Personality, 44, 492–500. Grühn, D., & Scheibe, S. (2008). Age-­related Charles, S. T., & Piazza J. R. (2009). Age differ- differences in valence and arousal ratings of ences in affective well-being: Context matters. pictures from the International Affective Pic- ture System (IAPS): Do ratings become more Social and Personality Psychology Compass, extreme with age? Behavior Research Meth- 3, 1–14. ods, 40, 512–521. Charles, S. T., Piazza, J., Luong, G., & Almeida, Hart, S. L., & Charles, S. T. (2013). Age-r­elated D. M. (2009). Now you see it, now you don’t: patterns in negative affect and appraisals Age differences in affective reactivity to social about colorectal cancer over time. Health Psy- tensions. Psychology and Aging, 24, 645–653. chology, 32, 302–310. Charles, S. T., Reynolds, C. A., & Gatz, M. Hess, T. M. (2005). Memory and aging in con- (2001). Age-­related differences and change text. Psychological Bulletin, 131, 383–406. in positive and negative affect over 23 years. Isaacowitz, D. M., & Blanchard-F­ields, F. (2012). Linking process and outcome in the Journal of Personality and Social Psychology, study of emotion and aging. Perspectives on 80, 136–151. Psychological Science, 7, 3–17. Consedine, N. S., & Magai, C. (2006). Emotion Isaacowitz, D. M., Toner, K., Goren, D., & Wil- development in adulthood: A developmental son, H. R. (2008). Looking while unhappy: functionalist review and critique. In C. Hoare Mood congruent gaze in young adults, posi- (Ed.), The Oxford handbook of adult develop- tive gaze in older adults. Psychological Sci- ment and learning (pp. 209–244). New York: ence, 22, 147–159. Oxford University Press. Keil, A., & Freund, A. M. (2009). Changes in the Davidson, R. J., Jackson, D. C., & Kalin, N. sensitivity to appetitive and aversive arousal H. (2000). Emotion, plasticity, context, and across adulthood. Psychology and Aging, 24, regulation: Perspectives from affective neuro- 668 – 680. science [Special issue]. Psychological Bulletin, Kensinger, E. A. (2004). Remembering emotional 126, 890–909. experiences: The contribution of valence and Davis, S. W., Dennis, N. A., Daselaar, S. M., arousal. Reviews in the Neurosciences, 15(4), Fleck, M. S., & Cabeza, R. (2008). Que 231–307. PASA?: The posterior–a­ nterior shift in aging. Cerebral Cortex, 18, 1201–1209. Diener, E., & Suh, E. M. (1998). Subjective well-being and age: An international analysis. Annual Review of Gerontology and Geriat- rics, 17, 304–324. Emery, L., & Hess, T. M. (2011). Cognitive con-

216 DEVELOPMENTAL CONSIDERATIONS Kennedy, Q., Mather, M., & Carstensen, L. L. J. M. (1994). The influence of age and gender (2004). The role of motivation in the age-­ on affect, physiology, and their interrelations: related positivity effect in autobiographical A study of long-term marriages. Journal of memory. Psychological Science, 15, 208–214. Personality and Social Psychology, 67, 56–68. Levine, L. J., & Pizarro, D. A. (2004). Emotion Kessler, E., & Staudinger, U. M. (2009). Affec- and memory research: A grumpy overview. tive experience in adulthood and old age: The Social Cognition, 22, 530–554. role of affective arousal and perceived affect Löckenhoff, C. E., & Carstensen, L. L. (2007). regulation. Psychology and Aging, 24, 349– Aging, emotion, and health-­related decision 362. strategies: Motivational manipulations can reduce age differences. Psychology and Aging, Kliegel, M., Jäger, T., & Phillips, L. H. (2007). 22, 134–146. Emotional development across adulthood: Magai, C., Consedine, N. S., Krivoshekova, Y. Differential age-­related emotional reactivity S., Kudadjie-­Gyamfi, E., & McPherson, R. and emotion regulation in a negative mood (2006). Emotion experience and expression induction procedure. International Journal across the adult life span: Insights from a mul- of Aging and Human Development, 64, 217– timodal assessment study. Psychology and 244. Aging, 21, 303–317. Mather, M., Canli, T., English, T., Whitfield, S., Kobau, R., Safran, M. A., Zack, M. M., Mori- Wais, P., Ochsner, K., et al. (2004). Amygdala arty, D. G., & Chapman, D. (2004). Sad, responses to emotionally valenced stimuli in blue, or depressed days, health behaviors older and younger adults. Psychological Sci- and health-­related quality of life, Behavioral ence, 15, 259–263. Risk Factor Surveillance System, 1995–2000. Mather, M., & Carstensen, L. L. (2003). Aging Health and Quality of Life Outcomes, 2, 40. and attentional biases for emotional faces. Psychological Science, 14, 409–415. Kunzmann, U., & Grühn, D. (2005). Age differ- Mather, M., & Carstensen, L. L. (2005). Aging ences in emotional reactivity: The sample case and motivated cognition: The positivity effect of sadness. Psychology and Aging, 20, 47–59. in attention and memory. Trends in Cognitive Sciences, 9, 496–502. Kunzmann, U., Little, T. D., & Smith, J. (2000). Mather, M., Johnson, M. K., & De Leonardis, Is age-r­ elated stability of subjective well-being D. M. (1999). Stereotype reliance in source a paradox?: Cross-­sectional and longitudinal monitoring: Age differences and neuropsycho- evidence from the Berlin aging study. Psychol- logical test correlates. Cognitive Neuropsy- ogy and Aging, 15, 511–526. chology, 16, 437–458. McRae, K., Gross, J. J., Weber, J., Robertson, Kunzmann, U., & Richter, D. (2009). Emotional E. R., Solkol-H­ essner, P., Ray, R. D., et al. reactivity across the adult life span: The cogni- (2012). The development of emotion regula- tive pragmatics make a difference. Psychology tion: An fMRI study of cognitive reappraisal and Aging, 24, 879–889. in children, adolescents and young adults. Scan, 7, 11–22. Lambrecht, L., Kreifelts, B., & Wildgruber, D. Mikels, J. A., Larkin, G. R., Reuter-­Lorenz, P., (2012). Age-­related decrease in recognition & Carstensen, L. L. (2005). Divergent trajec- of emotional facial and prosodic expressions. tories in the aging mind: Changes in working Emotion, 12, 529–539. memory for affective versus visual information with age. Psychology and Aging, 20, 542–553. Lang, F. R., & Carstensen, L. L. (1994). Close Mroczek, D. K., & Almeida, D. M. (2004). The emotional relationships in late life: Further effect of daily stress, personality, and age on support for proactive aging in the social daily negative affect. Journal of Personality, domain. Psychology and Aging, 9, 315–324. 72, 355–378. Mroczek, D. K., & Kolarz, C. M. (1998). The Lansford, J. E., Sherman, A. M., & Antonucci, T. effect of age on positive and negative affect: A C. (1998). Satisfaction with social networks: developmental perspective on happiness. Jour- An examination of socioemotional selectivity nal of Personality and Social Psychology, 75, theory across cohorts. Psychology and Aging, 1333–1349. 13, 544–552. Lawton, M. P. (2001). Emotion in later life. Cur- rent Directions in Psychological Science, 10, 120 –123. Leclerc, C. M., & Kensinger, E. A. (2011). Neural processing of emotional pictures and words: A comparison of young and older adults. Devel- opmental Neuropsychology, 26, 519–538. Levenson, R. W., Carstensen, L. L., & Gottman,

Emotion Regulation and Aging 217 Nikitin, J., & Freund, A. M. (2011). Age and is not so costly for older adults. Psychology motivation predict gaze behavior for facial and Aging, 24, 217–223. expressions. Psychology and Aging, 26, 695– Scheibe, S., & Carstensen, L. L. (2010). Emo- 700. tional aging: Recent findings and future trends. Journals of Gerontology B: Psycho- Ong, A. D., Rothstein, J. D., & Uchino, B. N. logical and Social Sciences, 65, 135–144. (2012). Loneliness accentuates age differences Scheibe, S., English, T., Tsai, J. L., & Carstensen, in cardiovascular responses to social evalua- L. L. (2013). Striving to feel good: Ideal affect, tive threat. Psychology and Aging, 27, 190– actual affect, and their correspondence across 198. adulthood. Psychology and Aging, 28, 160– 171. Petrican, R., Moscovitch, M., & Schimmack, Schryer, E., & Ross, M. (2012). Evaluating the U. (2008). Cognitive resources, valence, and valence of remembered events: The impor- memory retrieval of emotional events in older tance of age and self-­relevance. Psychology adults. Psychology and Aging, 23, 585–594. and Aging, 27, 237–242. Seider, B. H., Shiota, M. N., Whalen, P., & Lev- Phillips, L. H., Henry, J. D., Hosie, J. A., & enson, R. W. (2011). Greater sadness reactiv- Milne, A. B. (2008). Effective regulation of ity in late life. Social Cognitive and Affective the experience and expression of negative Neuroscience, 6, 186–194. affect in old age. Journals of Gerontology B: Shamaskin, A. M., Mikels, J. A., & Reed, A. E. (2010). Getting the message across: Age differ- Psychological Sciences and Social Sciences, ences in the positive and negative framing of 63, P138–P145. health care messages. Psychology and Aging, Piazza, J., & Charles, S. T. (2006). Mental health 25, 746–751. and the baby boomers. In S. K. Whitbourne Shiota, M. N., & Levenson, R. W. (2009). & S. L. Willis (Eds.), The baby boomers at Effects of aging on experimentally instructed detached reappraisal, positive reappraisal, and midlife: Contemporary perspectives on mid- emotional behavioral suppression. Psychology dle age (pp. 111–148). Mahwah, NJ: Erlbaum. and Aging, 24, 890–900. Piazza, J. R., Charles, S. T., Stawski, R. S., & Slessor, G., Miles, L. K., Bull, R., & Phillips, L. Almeida, D. M. (2013). Age and the asso- H. (2010). Age-­related changes in detecting ciation between negative affective states and happiness: Discriminating between enjoyment diurnal cortisol. Psychology and Aging, 28, and nonenjoyment smiles. Psychology and 47–56. Aging, 25, 246–250. Piguet, O., Connally, E., Krendl, A. C., Huot, Sliwinski, M. J., Almeida, D. M., Smyth, J., & J. R., & Corkin, S. (2008). False memory in Stawski, R. S. (2009). Intraindividual change aging: Effects of emotional valence on word and variability in daily stress processes: Find- recognition accuracy. Psychology and Aging, ings from two measurement-­burst diary stud- 23, 85–92. ies. Psychology and Aging, 24, 828–840. Ready, R. E., Akerstedt, A. M., & Mroczek, D. St. Jacques, P., Dolcos, F., & Cabeza, R. (2009). K. (2012). Emotional complexity and emo- Effects of aging on functional connectivity of tional well-being in older adults: Risks of neu- the amygdala during subsequent memory for roticism. Aging and Mental Health, 16, 17–26. negative pictures: A network analysis of fMRI Reed, A., & Carstensen, L. L. (2012) The theory data. Psychological Science, 20, 74–84. behind the age-r­elated positivity effect. Fron- St. Jacques, P., Dolcos, F., & Cabeza, R. (2010). tiers in Cognitive Science, 3, 1–9. Effects of aging on functional connectivity of Riediger, M., Schmiedek, F., Wagner, G., & Lin- the amygdala during negative evaluation: A denberger, U. (2009). Seeking pleasure and network analysis of fMRI data. Neurobiology seeking pain: Differences in prohedonic and of Aging, 31, 315–331. contra-­hedonic motivation from adolescence St. Jacques, P. L., Bessette-S­ ymons, B., & Cabeza, to old age. Psychological Science, 20, 1529– R. (2009). Functional neuroimaging studies of 1535. aging and emotion: Fronto-­amygdalar differ- Samanez-­Larkin, G. R., Gibbs, S. E. B., Khanna, ences during emotional perception and epi- K., Nielsen, L., Carstensen, L. L., & Knutson, sodic memory. Journal of the International B. (2007). Anticipation of monetary gain but Neuropsychological Society, 15, 819–825. not loss in healthy older adults. Nature Neuro- science, 10, 787–791. Scheibe, S., & Blanchard-F­ields, F. (2009). Effects of regulating emotions on cognitive performance: What is costly for young adults

218 DEVELOPMENTAL CONSIDERATIONS Stone, A. A., Schwartz, J. E., Broderick, J. E., B: Psychological Sciences and Social Sciences, & Deaton, A. (2010). A snapshot of the age 65, P154–P162. distribution of psychological well-being in the Urry, H. L., & Gross, J. J. (2010). Emotion regu- United States. Proceedings of the National lation in older age. Current Directions in Psy- Academy of Sciences USA, 107(22), 9985– chological Science, 19, 352–357. 9990. Villamil, E., Huppert, F. A., & Melzer, D. (2006). Low prevalence of depression and anxiety is Story, T. N., Berg, C. A., Smith, T. W., Beveridge, linked to statutory retirement ages rather than R., Henry, N. J. M., & Pearce, G. (2007). Age, personal work exit: A national survey. Psycho- marital satisfaction, and optimism as predic- logical Medicine, 36, 999–1009. tors of positive sentiment override in middle-­ Weierich, M. R., Kensinger, E. A., Munnell, A. aged and older married couples. Psychology H., Sass, S. T., Dickerson, B. D., Wright, C. and Aging, 22, 719–727. I., et al. (2011). Older and wiser?: An affec- tive science perspective on age-r­elated chal- Streubel, B., & Kunzmann, U. (2011). Age dif- lenges in financial decision making. Scan, 6, ferences in emotional reactions: Arousal and 195–206. age-­relevance count. Psychology and Aging, Westlye, L. T., Walhovd, K., B., Dale, A. M., 26, 966–978. Bjørnerud, A., Due-Tønnessen, P., Engvig, A., et al. (2010). Life-span changes of the human Teachman, B. A. (2006). Aging and negative brain white matter: Diffusion tensor imaging affect: The rise and fall and rise of anxiety and (DTI) and volumetry. Cerebral Cortex, 20, depressive symptoms. Psychology and Aging, 2055–2068. 21, 201–207. Wrzus, C., Müller, V., Wagner, G. G., Linden- berger, U., & Riediger, M. (2013). Affective Tsai, J. L., Levenson, R. W., & Carstensen, L. and cardiovascular responding to unpleasant L. (2000). Autonomic, expressive and subjec- events from adolescence to old age: Complex- tive responses to emotional films in younger ity of events matters. Developmental Psychol- and older adults of European American and ogy, 49, 384–397. Chinese descent. Psychology and Aging, 15, Wurm, L. H., Labouvie-­Vief, G., Aycock, J., 684 – 693. Rebucal, K. A., & Koch, H. E. (2004). Perfor- mance in auditory and visual emotional Stroop Tucker, A. M., Feuerstein, R., Mende-­Siedlecki, tasks: A comparison of older and younger P., Ochsner, K. N., & Stern, Y. (2012). Dou- adults. Psychology and Aging, 19, 523–535. ble dissociation: Circadian off-peak times Zulfiqar, U., Jurivich, D. A., Gao, W., & Singer, increase emotional reactivity; Aging impairs D. H. (2010). Relation of high heart rate vari- emotion regulation via reappraisal. Emotion, ability to healthy longevity. American Journal 12, 869–874. of Cardiology, 105, 1181–1185. Uchino, B. N., Birmingham, W., & Berg, C. A. (2010). Are older adults less or more physi- ologically reactive?: A meta-a­nalysis of age-­ related differences in cardiovascular reactivity to laboratory tasks. Journals of Gerontology

Part V Social Aspects



Chapter 14 Social Baseline Theory and the Social Regulation of Emotion James A. Coan Erin L. Maresh To survive and reproduce, organisms must own resources in light of expected returns. take in more energy than they expend, This perspective (1) highlights the impor- a principle of behavioral ecology called tance of emotion as a source of information economy of action (Krebs & Davies, 1993). about current and predicted resources, and Social baseline theory (SBT), a framework (2) entails a view of psychological measure- based on this principle, organizes decades ment that extends beyond the individual to of observed links between social relation- systemic processes within dyads and groups ships, health, and well-being, in order to (Schilbach et al., in press). understand how humans utilize each other as resources to optimize individual energy Social Proximity and the Economy expenditures. This general strategy helps us of Action manage the costs of our very long period of The economy of action suggests that adap- ontogenetic development and, we argue, the tations to prevailing environmental con- many behavioral and psychological capa- ditions must, at the very least, ensure that bilities of our uniquely powerful and costly more energy is acquired than lost (Krebs brain (Smith, 2003). Below, we review evi- & Davies, 1993). The alternative—­more dence that social relationships serve the energy out than in—leads ultimately to energy-­saving functions that we claim and death. But because environments are rife describe the reasons SBT refers to social with danger and competition, the economy proximity as a “baseline” condition. For of action tends toward optimization, where this chapter we place special emphasis on the resources are both acquired and conserved role of social proximity in the regulation of whenever possible. For example, foraging emotion. In our view, the social regulation animals optimize energy intake per unit of emotion serves, in the aggregate and on of time by abandoning even plentiful food average, to decrease the cost of coping with sources if the amount of energy (e.g., calo- many of life’s difficulties (Cohen & Hober- ries) acquired is not optimal (MacArthur man, 1983; Cohen & McKay, 1984)—a & Pianka, 1966; Schmidhempel, Kacelnik, function that involves the brain’s ability & Houston, 1985). Optimization strategies to use both internal and external informa- tion to make “bets” about how to deploy its 221

222 SOCIAL ASPECTS are likely mediated through changes in sen- his or her behavior (which might encompass sory perception. So simply wearing a heavy social history with the other person, an offer backpack can make distances appear far- of aid, or simply close physical proximity) ther away and hills seem steeper (Stefanucci, influences another’s emotional response to Proffitt, Banton, & Epstein, 2005). Indeed, some additional current or potential stimu- it is increasingly apparent that perception is lus or situation. Second, we acknowledge influenced by not only the properties of sen- that people vary in the extent to which they sory stimuli but also emotions—­emotions both seek out and derive benefit from social that are themselves embodied instantia- contact (leading to individual differences in tions of prevailing circumstances, goals, the broad processes described here). But on and physiological states (Stefanucci, Proffitt, average and in the aggregate, it is probably Clore, & Parekh, 2008). the case that for humans and other social mammals, proximity to social resources SBT is rooted in the proposition that, for reduces the net cost of survival. Third, humans, social proximity and interaction although most of the empirical evidence we powerfully optimize energy expenditures review entails the regulatory impact of social (e.g., actual and perceived effort) devoted to proximity and interaction on the regulation navigating potentially dangerous environ- of negative affect, we do not intend to argue ments. Moreover, the best evidence suggests that negative affect is the only broad form of these effects are mediated through rela- emotional responding that can be regulated. tively automatic, bottom-u­ p, unconditioned Indeed, below we address some instances of perceptual mechanisms that modulate—­ the social regulation of positive emotion. and regulate—e­motional responding, as opposed to mechanisms often associated Our Social Baseline with the self-regulation of emotion (e.g., the There is little doubt that supportive social dorsolateral [dlPFC] and ventromedial pre- behaviors can modify or quell our subjec- frontal cortex [vmPFC]). SBT invokes the tive, behavioral, and physiological responses economy of action here, too, by suggesting to stress (Holt-­Lunstad, Smith, & Lay- that perceptual mechanisms of the central ton, 2010). Such effects are likely rooted nervous system are “inexpensive”1 when in human phylogeny. Indeed, Beckes and compared to more effortful, self-­regulatory, Coan (2011) and others (Berscheid, 2003; vigilant, “future-­thinking” processes associ- Brewer & Caporeal, 1990) have suggested ated with the operation of the PFC—a dif- that the dominant ecology to which humans ference that leads to both phylogenetic and are adapted is not any one terrain, diet, or ontogenetic pressures to achieve regulatory climate, but rather each other. One of the ends by perceptual and hence social means. defining features of human beings—­our We suggest that the ability to depend on oth- adaptability—i­s yoked to our ability to ers for as many costly processes as possible cooperate with others, an ability that was leads to (1) decreased overall cost in deal- likely shaped by periods of great environ- ing with an uncertain and potentially deadly mental instability (Richerson, Bettinger, environment, and (2) pressure—­throughout & Boyd, 2005). Cooperative behavior was the lifespan and over the course of evolu- likely selected as a means of managing the tion—­to form and maintain close social energy costs of our exceedingly expensive relationships. bodies, brains, and activities (Hill et al., 2011; McNally, Brown, & Jackson, 2012; Before going any further, we would like to Moll & Tomasello, 2007). So the first sense emphasize three important points. First, our in which SBT refers to a social baseline is view of social emotion regulation does not that social relatedness and its psychological imply that any and all effects an individual’s correlates constitute the normal, baseline behavior may have on the emotional behav- ecology of the functional human brain. That ior of another constitute regulatory effects. is, the human brain is designed to operate If one individual insults or compliments within a relatively predictable network of another, either behavior could cause the social relationships characterized by famil- receiver to activate an emotional response. This would not, however, constitute an instance of emotion regulation. An indi- vidual regulates another’s emotions when

Social Baseline Theory 223 iarity, shared intentionality, and interdepen- as a function of not only proximity to social dence (Tomasello, Carpenter, Call, Behne, resources but also the perceived quality of & Moll, 2005). those resources. The other sense in which SBT refers to From the perspective of SBT, the alone the social “baseline” results from work condition used in the hand-­holding para- using functional magnetic resonance imag- digm described earlier is not a “baseline” ing (fMRI; Coan, Schaefer, & Davidson, condition against which the experimental 2006), which suggests that social proximity hand-­holding conditions are compared. On and interaction regulate many aspects of the the contrary, the reverse is true—the hand-­ brain’s response to potential threats. Spe- holding conditions are closer to the environ- cifically, the authors subjected 16 married mental conditions to which the human brain women to the threat of mild electric shock is adapted, and the alone condition is far- under three conditions, counterbalanced ther away. This conclusion forms part of the across subjects: while alone in the scanner, basis of SBT—that social resources decrease while holding a stranger’s hand, and while the perceived cost of managing a threaten- holding her husband’s hand. The brain was ing environment, providing a primary and maximally active while facing the threat largely unconditional opportunity for econ- alone, exhibiting a set of responses repli- omizing both neural and behavioral activity. cating a large body of work on the brain’s response to threat, fear, stress, and pain Importantly, these effects are not depen- (Bishop, Duncan, Brett, & Lawrence, 2004; dent upon physical touch per se. For Brooks, Nurmikko, Bimson, Singh, & Rob- example, there is less hypothalamic activ- erts, 2002; Dedovic et al., 2005; Stark et al., ity during social rejection among individu- 2003). As levels of contact with perceived als asked to imagine an attachment figure social resources increased, however, many (Karremans, Heslenfeld, van Dillen, & Van of these activations either decreased in inten- Lange, 2011). Moreover, imagining a strong sity or vanished altogether. attachment figure seems to have resulted in not more but less activity in frontal regions For example, although both spouse and supporting self-­regulatory effort, suggesting stranger handholding caused decreased that even imagining the presence of a secure threat responding in regions associated attachment figure may regulate one’s emo- with the regulation of bodily arousal and tions in such a way as to conserve perceived the mobilization of behavioral action resources. plans—such as the ventral anterior cingu- late cortex, supramarginal and postcentral Social Systems Regulate gyri, and posterior cingulate—­only spouse the Emotions of Individuals hand-h­olding was specifically associated Critically, what we experience and catego- with additional attenuation in circuits asso- rize as emotion powerfully impacts the way ciated with effortful emotion regulation we perceive and engage with the world (Ste- and threat-­related homeostatic functions, fanucci, Gagnon, & Lessard, 2011). Emo- such as the right dlPFC and superior col- tional processes carry vital, often implicit, liculus. Most strikingly, women inhabiting and embodied information about the good- the highest quality relationships realized ness and badness of things (Clore & Tamir, dramatic attenuations in all the aforemen- 2002)—information that modifies percep- tioned threat-r­esponsive regions in addition tion and guides action (Barrett & Bar, 2009; to those supporting subjective suffering and Schwarz & Clore, 1983). It follows that the release of stress hormones, such as the emotions are indicators of, among other right anterior insula and hypothalamus. things, the tension between perceived per- Putting it all together, there appeared to be sonal resources and perceived environmen- a relatively linear, monotonic decrease in tal demands (Moore, Vine, Wilson, & Free- the degree of threat-­related neural process- man, 2012; Tomaka, Blascovich, Kelsey, ing as one progressed from being alone to & Leitten, 1993). In this way, emotion is being with anyone (stranger or spouse), to probably at or near the center of decisions being with a spouse, to being with a spouse that economize activity (Bechara, 2011; in a very high-q­ uality relationship. This sug- gests that the perception of threat decreases

224 SOCIAL ASPECTS Stefanucci et al., 2008). Accordingly, SBT season expectations, and early season per- suggests that the brain uses affective, expe- formance. Others have observed that one of riential, conceptual, and contextual knowl- the best predictors of collective IQ is not the edge—i­n effect, emotion (Barrett, 2006; IQs of individual group members but rather Lindquist & Barrett, 2008)—to bias percep- the degree to which each member is sensitive tion in ways that guide actions toward more to social cues expressed on the face (Wool- favorable outcomes (Beckes & Coan, 2011; ley, Chabris, Pentland, Hashmi, & Malone, Coan, 2008). 2010). From the perspective of SBT, each of these findings is attributable to the way Emotion has been identified by others social resources help to conserve costly vigi- as a key link between social relationships lance and self-­regulation efforts, either by and health. For example, the social buffer- imposing less interference with motor and ing hypothesis suggests that social relation- perceptual activity (in the case of players in ships can provide support that is tangible, the NBA), or by devoting those resources to such as material support when facing loss solving other kinds of problems (in the case of income, appraisal-­based support, such as of collective IQ). But the question of why when familiar others reduce the incidence social resources economize human neural or intensity of threat-r­elated appraisals, or and behavioral activity remains. We pro- emotional support, such as when famil- pose at least two broad, distal mechanisms iar others repair threats to self-e­steem or by which this is accomplished: risk distribu- increase a sense of felt belonging (Cohen & tion and load sharing. McKay, 1984). It has been suggested further that social relationships grow into part of Risk Distribution the brain’s understanding of itself and its In group settings, many species adjust the available resources (Beckes, Coan, & Has- level of their energy expenditure in accor- selmo, 2012). Moreover, as it makes predic- dance with the size of the group they inhabit tions about the potential cost of coping with (Krause & Ruxton, 2002). This benefit a given situation, the brain follows Bayes- of group living is called risk distribution ian rules of inference (Friston, 2010; Knill (Coan, 2008), in that environmental risk is & Pouget, 2004), in which judgments about probabilistically distributed across the group the level of personal resources to deploy are rather than being concentrated on one indi- based on (1) the current situation (particu- vidual. A clear example of this can be seen larly constraints, risks, and opportunities), in optimal foraging theory, which proposes (2) predicted possible future situations, (3) that animals forage in a way that maximizes situational goals, (4) current energy states, energy intake per unit of time (MacArthur and (5) expected future energy states (Sali- & Pianka, 1966). For example, animals nas, 2011). that forage must spend a certain amount of energy maintaining vigilance for preda- For example, hills appear to be less steep tors. Foraging in a larger group decreases when standing next to a friend (Schnall, the burden of vigilance on any one member, Harber, Stefanucci, & Proffitt, 2008), and and indeed, in many species, the number of this effect is correlated with friendship dura- individuals that are vigilant for predators tion: the longer the friendship, the less steep at any given time decreases as group size the hill appears to be. The perception that increases (Elgar, 1989; Roberts, 1996). With the hill is less steep serves as a marker that regard to the social regulation of emotion, one has, in effect, more bioenergetic capi- we can construe threat vigilance as a form tal in one’s budget—c­apital that can either of negative affect—­or at least as the result be spent more freely walking up the hill, or of a negative affect process that reflects the conserved over time. Diverse studies hint demand for vigilance. If true, then social at similar conclusions regarding the brain’s proximity can be viewed as decreasing vigi- Bayesian properties. For example, Kraus, lance behavior via the regulation of negative Huang, and Keltner (2010) have observed affect. Evidence for this can even be seen that more frequent physical touching among in the modulation by group size of physi- professional basketball players corresponds with increased late-­season performance, even after accounting for player status, pre-

Social Baseline Theory 225 ological indicators of stress in nonhuman on one’s behalf, thus further reducing the animals. For example, salivary cortisol con- need to expend one’s own energy, and com- centration in sheep is higher when the flock mensurably decreasing the need, for exam- size is smaller (Michelena et al., 2012). Risk ple, for negative affect. In other words, a distribution suggests further that the num- close companion can be trusted to share an ber of proximal conspecifics decreases nega- interest in one’s well-being. He or she may tive affect, because doing so conserves effort provide additional vigilance not only in some that would otherwise be spent either on vigi- general sense but also specifically, with one’s lance, self-r­ egulation, or both. This is exem- own welfare in mind (Davis, 1984). A close plified in humans in the hand-­holding study companion may share resources (e.g., Rog- discussed earlier, by the finding that holding ers & DeBoer, 2001), help care for young any hand—even that of a stranger—r­educes (e.g., Ehrenberg, Gearing-S­ mall, Hunter, & neural activity related to threat (Coan et Small, 2001), and help when one is sick or al., 2006), especially in regions supporting injured (e.g., Townsend & Franks, 1995). responses to acute threats (Mobbs et al., These additional benefits provide a much 2007). greater opportunity for economizing neural and behavioral activity, as can be seen in Interestingly, the individual benefits of the enhanced effect of holding hands with group living described earlier often benefit a close friend or a spouse (Coan, Beckes, & the group as well, such that as each individ- Allen, 2013; Coan et al., 2006). ual’s vigilance effort decreases, total group vigilance actually increases (Bertram, 1978; SBT maintains that load sharing impacts Pulliam, Pyke, & Caraco, 1982) and time decisions about personal resource budget- to detect predators decreases (Siegfried & ing in part by altering the way the brain Underhill, 1975). It is important to note, encodes what constitutes the “self,” with however, that group living introduces certain implications for the level of resources the risks and stressors as well, such as increased self is able to access. This is consistent with competition for resources, increased likeli- the observation that the neural representa- hood of disease transmission, and increased tions of threats directed at the self are highly conspicuousness (Alexander, 1974). Groups correlated with those directed at friends, therefore iteratively move toward an optimal implying a kind of “involuntary breach of size given the availability of food and other individual separateness” (Langer, 1974, resources (Higashi & Yamamura, 1993). p. 129) that does not seem to generalize to Note that risk distribution concerns only the strangers (Beckes et al., 2012). In this way, number of members in a group and makes an individual’s social support system can be no reference to the nature and quality of the construed as an extension of the self and, relationships between the group members. in turn, of how one perceives and interacts Coan (2008) has speculated that the brain with the world. is sensitive to risk distribution as a general purpose strategy that employs Bayesian-l­ike Capitalization calculations to assess the cost-e­ffectiveness Importantly, social proximity and interac- of affective behaviors at any given time. tion serve not only to down-­regulate negative emotions, but also to maintain and increase Load Sharing positive, approach-r­elated emotions. Admit- Load sharing builds on the benefits of risk tedly, our research has focused on the distribution by adding information that social regulation of response to threat, but increases certainty about the availability of we believe the basic ideas behind SBT are a given social resource—n­amely, familiar- generalizable to positive emotions as well. ity, reliability, and interdependence. The People frequently expend energy to induce, presence of an individual with whom one maintain, or even dampen experiences of has established a close relationship not only positive emotion, and this is reflected in probabilistically reduces one’s risk of envi- neural activity in prefrontal areas similar to ronmental threat but also signals access to a those seen in the self-r­egulation of negative person who will use his or her own resources emotion (Kim & Hamann, 2007). Social

226 SOCIAL ASPECTS relationships may act to conserve effortful propose potential mechanisms for the dis- positive self-­regulation, possibly by reducing tinction. reliance on prefrontal activity and increas- ing activity in reward circuits. Interestingly, Mediating Mechanisms of Social neural activity in the medial PFC decreases Emotion Regulation in mothers observing pictures of their chil- As mediators of emotion regulation (Baron dren, and in adults observing pictures of & Kenny, 1986), social resources serve as a their romantic partner, as compared to proximal mechanism through which emo- when they are observing pictures of a friend tion regulation effects are achieved. Attach- (Bartels & Zeki, 2000, 2004). Moreover, ment theory provides the quintessential positive responses by a relational partner to example of socially regulated emotion in one’s own positive news, a process known its description of mother–­child attachment as capitalization, may be a mechanism by interactions, in which infants seek attach- which social relationships enhance positive ment figures during periods of distress and emotions in ways that are less individually are soothed by their caregiver’s presence effortful. Capitalization increases daily posi- (Ainsworth, Blehar, Waters, & Wall, 1978; tive affect over and above the impact of the Bowlby, 1969/1982). A common example positive event itself (Gable, Reis, Impett, & might be a child’s vaccination procedure, Asher, 2004; Langston, 1994). Furthermore, where the child—d­ istressed at the prospect capitalization may strengthen trust and pro- of receiving a shot—calls for the mother’s social behavior, even when one is interacting support. In response, the mother may engage with a stranger (Reis et al., 2010). Following in soothing behaviors like hand-­holding and the SBT model, then, receiving supportive, offering reassuring words in a calm tone of positive feedback from a partner may sig- voice. Here, the mother’s behavior provides nal an increase in both personal and social an obvious mechanism through which the resources. It is interesting to note that the child’s distress is attenuated. Specifically, perceived responsiveness of one’s partner the child identifies a potential threat (the during positive events may be more strongly needle), and this threat causes two reactions, linked with the relationship’s well-being distress and the seeking of support from the and longevity than perceived responsiveness mother. In turn, the mother’s support behav- during negative events (Gable, Gonzaga, & ior exerts a down-r­egulatory influence on Strachman, 2006). the distress response. Importantly, SBT views all forms of The mother–­child dynamic is a par- social emotion regulation as systemic ticularly powerful example of how social and dynamic, challenging the widely held resources can mediate emotion regulation assumption that the basic unit of analysis in in part because the child in the example is human psychology is the single individual developmentally limited in his or her self-­ (Beckes & Coan, 2011). By extending the regulatory abilities, a limitation that is itself unit of analysis to the dyad or group, social rooted in the slow development of neural interaction and proximity emerge as adap- systems that will eventually serve the child’s tive strategies rooted in the sharing of efforts self-­regulatory needs (Coan, 2008). That that, left only to each individual member of the mother’s soothing works is obvious and a dyad or group, would be redundant, costly, well documented (Cassidy & Shaver, 2008). and inefficient. Ultimately, it may be that Less obvious is that the child is relatively dyads or groups benefit from social emo- incapable of emotion regulation without tion regulation at least as much as the indi- the mother. Thus, we can view the mother viduals who inhabit them (McComb, Moss, and child as an interactive system that is Durant, Baker, & Sayialel, 2001). In consid- intimately linked in an emotion activation–­ ering mechanisms for these dyadic or group regulation dynamic. Importantly, however, effects, social emotion regulation can be the mother–c­ hild behavioral and experiential mediational, reflecting a direct intervention dynamics described by attachment theorists on an emotional process by social resources, play out similarly within adult relationships or moderational, modifying the perception (Mikulincer & Shaver, 2007), where stress- of potentially emotional situations. In the ful situations motivate social proximity, and following sections, we briefly describe and

Social Baseline Theory 227 social proximity lowers autonomic arousal, the activation of neural circuits associated attenuates hypothalamic–­pituitary–­adrenal with reward processing (e.g., the caudate (HPA) axis activity, and improves immune nucleus, the nucleus accumbens) and effort- function (Baron, Cutrona, Hicklin, Russell, ful self-r­ egulation (e.g., the dlPFC). & Lubaroff, 1990; Heinrichs, Baumgart- ner, Kirschbaum, & Ehlert, 2003; Uchino, Thus, emerging evidence for proximal Cacioppo, & Kiecolt-G­laser, 1996). Of mechanisms supporting socially medi- interest with regard to adult relationships ated emotion regulation appears to include is that, unlike mother–­infant dyads, two or regions associated with both the automatic more adults each possess fully developed and and effortful self-regulation of emotion functioning brains. Thus, although children (vmPFC and dlPFC, respectively), as well require maternal support for emotion regu- as putative reward circuits (e.g., the nucleus lation, adults appear to utilize each other for accumbens). However, our laboratory has similar purposes more strategically. not been able to produce similar results, in our original study of hand-­holding by mar- A major focus of our work has been the ried couples (Coan et al., 2006), in a more identification of proximal neural mechanisms recent study of hand-h­olding by platonic linking social relationships to decreased friends (Coan et al., 2013), or in a study of threat responding and, in turn, increased the regulation of threat responses in chil- health and well-being. Although many can- dren by the presence of adult caregivers didate mechanisms have been postulated, (Conner et al., 2012). Indeed, in the latter few or none have been definitively identified. study, we observed decreased activation spe- Recently, Eisenberger and colleagues (2011) cifically in the vmPFC and ventrolateral PFC provided what they characterized as media- (vlPFC), where we used a region of interest tional evidence for the social regulation of (ROI) approach for the purpose of address- emotion by portions of the vmPFC. In this ing the question of prefrontal mediation of work, participants were shown pictures of social support. Ultimately, we have thus far a romantic partner, pictures of strangers, or failed to reliably identify any regions of the pictures of objects, while receiving painful brain that are normatively more active under stimulation. Putative pain-r­elated activa- threat conditions during supportive presence tions in regions such as the dorsal anterior or hand-h­olding than while alone. More- cingulate cortex (dACC) and anterior insula over, none of the regions identified by Eisen- were less active while participants viewed berger et al. (2011) and Younger et al. (2010) images of a romantic partner. Moreover, the are negatively correlated with any of the vmPFC was more active while participants threat-r­esponsive circuits apparently down-­ viewed romantic partners, even during pain regulated in our work by hand-­holding. stimuli, and vmPFC activation was positively associated with relationship duration and What may account for these apparently perceived partner support, while negatively inconsistent results? The answer, we think, associated with both subjective pain ratings has to do with the difference between touch and pain-­related neural activity. In a simi- and picture viewing. Specifically, there is lar study, Younger, Aron, Parke, Chatterjee, ample evidence that touch acts as an uncon- and Mackey (2010) collected brain images ditioned or primary stimulus in social ani- from 15 participants as they experienced mals, not least humans (Francis et al., 1999), moderate levels of thermal pain while they who may also use touch to exchange impor- viewed pictures of their romantic partner, tant social information (Gazzola et al., pictures of attractive strangers, or engaged 2012; Morrison, Löken, & Olausson, 2010). in a distraction task, hypothesizing that pic- By contrast, picture viewing likely involves tures of romantic partners would attenuate associative learning, particularly involv- pain processing via putative “reward sys- ing the expected value of the photographic tems.” Indeed, viewing pictures of roman- image—a task often mediated through tic partners both reduced subjective pain vmPFC activity (Kable & Glimcher, 2007). reports and attenuated circuits associated Thus, vmPFC activity may be necessary with pain processing. Moreover, subjective when utilizing photographs of loved ones to pain reports were inversely correlated dur- regulate emotion. But a careful reading of ing romantic partner picture viewing with Younger et al. (2010) reveals that, in addi- tion to showing their participants pictures

228 SOCIAL ASPECTS of loved ones, they instructed participants to ties can be modeled by SBT, such that past “focus on the picture and think about the social experiences or even prevailing social displayed person” (p. 2), effectively encour- conditions alter an individual’s perception of aging self-mediated rather than socially potential threats and rewards in his or her mediated emotion regulation. Ultimately, immediate environment even when he or she we would argue that both the Eisenberger et is alone at the occasion of measurement. al. (2011) and Younger et al. (2010) experi- ments manipulated self-r­egulatory systems A number of researchers have argued and are not best understood as examples of that early childhood experiences critically socially mediated emotion regulation at all. influence the ways in which individuals view social relationships later in life. For Moderating Mechanisms of Social humans and many other species, filial bond- Emotion Regulation ing—bonding with siblings and caregivers—­ Strictly speaking, the hand-­holding study occurs rapidly, unconditionally, and during a discussed earlier may also provide an period of neural development that may hold example of socially moderated emotion powerful consequences for subsequent func- regulation, in part because the provision tioning (Bowlby, 1969/1982; Coan, 2008). of social support preceded the presentation In infants, the PFC is particularly under- of threat cues, thus potentially altering the developed in comparison to that of adults, perception of those threat cues and obviat- and its development takes approximately ing the activation of portions of the threat two decades to complete. The human brain response. This illustrates one of the many in general grows exponentially during just ways in which social proximity and interac- the first 2 years of life (Franceschini et al., tion can be viewed as moderating emotional 2007)—growth reflected in the brain’s con- responses and self-­regulation needs, acting sumption of glucose, which continues to be as third variables that modify the conditions approximately double that of adults until 10 under which emotional responses and self-­ years of age (Chugani, 1998). Throughout regulation strategies are either called upon this early development, the brain is in effect or maximally effective. Taking the example adapting itself to environmental contingen- of the vaccination procedure discussed ear- cies through a process of axonal, dendritic, lier, it may be the case that, as in our hand-­ and synaptic “pruning”—a process that is holding experiments, the mother holds the itself dependent upon neural activity associ- child’s hand not in response to the child’s ated with environmental stimulation (Can- obvious distress but in anticipation of the cedda et al., 2004; Reichardt, 2006). Put child’s potential distress. In this case, mater- another way, the brain in early development nal hand-­holding may cause relaxation of adheres roughly to the colloquialism “use it the child’s vigilance processing, rendering or lose it”; environmental demands shape a the appearance of the needle less threaten- developing brain prepared for a wide range ing when it does arrive. of behavioral possibilities. By the process of pruning, the brain’s responses to prevail- Past social experiences also impact an ing contextual demands become more fixed individual’s self-­regulation requirements and hence more rapid—it becomes adapted and capabilities (Allen, Moore, Kuperminc, ontogenetically to its expected environment & Bell, 1998; Prinstein & La Greca, 2004; (Hebb, 1949; Posner & Rothbart, 2007). Weaver et al., 2004). For example, ongo- ing current daily social support may mod- During this development, interactions erate the perception of threats encountered between social experiences, self-r­egulatory while alone, which may in turn reduce the activity and socially bound activations in perceived need for engaging self-r­egulation medial prefrontal, temporoparietal, and capabilities (Eisenberger, Taylor, Gable, posterior temporal cortices, as well as in Hilmert, & Lieberman, 2007). Similarly, the amygdala, ventral tegmentum, nucleus mental representations of loved ones may be accumbens, periaqueductal gray and else- used as methods by which one can exert self-­ where, may shape expectations about the regulatory strategies related to distraction availability of social resources, opportuni- or reappraisal. Any or all of these possibili- ties for affiliation, expectations about the reliability of close relational partners, and so on (Lieberman, 2007). This in turn is likely

Social Baseline Theory 229 to shape individual differences in the way the existence of trait-like expectations—t­he social resources are approached and main- aforementioned attachment styles—regard- tained, perhaps especially in situations that ing the availability of social resources. blend social proximity with potential stress- Dominant views of attachment style can ors. Indeed, it is just this sort of experience-­ be expressed as two independent axes rep- based developmental divergence that early resenting tendencies toward social anxi- attachment researchers described as mani- ety and social avoidance. Individuals low festing in “working models” of attachment in both are said to be secure, in that their figures, a theoretical perspective expressed appraisals of potential threats and views of in more recent terms as attachment styles emotional disclosure tend to be less negative (Ainsworth et al., 1978; Main, 1996; Miku- (Mikulincer & Florian, 1998; Mikulincer & lincer & Shaver, 2007). Orbach, 1995). Critically, such individuals are likelier to seek out, utilize, and benefit A large and growing body of research has from emotional support from others (Fraley documented effects consistent with early psy- & Shaver, 2000; Larose, Bernier, Soucy, & chosocial brain development. For example, Duchesne, 1999; Rholes, Simpson, & Grich primates are highly sensitive to early proxim- Stevens, 1998). Secure individuals may even ity to and interaction with caregivers (Har- benefit more from capitalization attempts: low, 1958). In rats, maternal grooming may Shallcross, Howland, Bemis, Simpson, and influence the methylation of glucocorticoid Frazier (2011) found that insecurely attached receptor genes throughout the hippocampus individuals more often underestimated and possibly elsewhere, influencing in turn the responsiveness of their partners when the stress reactivity of the pups throughout sharing a positive event than did securely their lives and even into subsequent genera- attached individuals. Although neuroscien- tions (Weaver et al., 2004). Evidence sug- tific investigations of attachment style are gests a similar methylation process may rare, Gillath, Bunge, Shaver, Wendelken, occur in humans (McGowan et al., 2009; and Mikulincer (2005) have observed that van IJzendoorn, Bakermans-­Kranenburg, & greater trait attachment anxiety is associ- Ebstein, 2011). Hofer (2006) has proposed ated with more activity in the dACC and that early interactions with caregivers slowly temporal pole, areas associated with alarm evolve from the indirect regulation of sen- and negative affect—­and decreased activity sorimotor, thermal, and nutrient functions in the orbitofrontal cortex—­when partici- via responses to expressed emotion. At first pants are asked to think about a threatening the infant is totally dependent upon caregiv- relationship scenario. ers for access to primary reinforcers (food, water, warmth, touch). The only way for the Future Directions infant to access these resources is to get the We feel these perspectives on social relation- caregiver’s attention with expressed emo- ships and social emotion regulation have tion. A predictable pattern emerges where implications that extend both to basic and the infant cries out when a basic resource is applied domains of psychological science, needed, followed by the caregiver’s response, and that the ecological framework of SBT after which the infant stops crying. As the may be useful for understanding a number infant develops, his or her emotional reper- of important new findings. Clinically, we toire expands to include preferences beyond have recently noted that SBT holds implica- simple metabolic or thermal needs, but the tions for how we understand the frontolim- process of caregiver responsiveness contin- bic dysfunction and emotional dysregula- ues in much the same way, until it can be tion that characterize borderline personality said that the emotional feelings and expres- disorder (Hughes, Crowell, Uyeji, & Coan, sions are themselves the targets of the care- 2012), and we believe that SBT may be rel- giver’s responses. evant to the development and maintenance of other psychopathologies as well. For Attachment theorists have had perhaps example, although speculative, it is possible the most to say about how social experiences that the disruptions in social relationships contribute to individual differences in the social regulation of emotion (Mikulincer & Shaver, 2005; Mikulincer, Shaver, & Pereg, 2003). Specifically, attachment theory posits

230 SOCIAL ASPECTS that characterize autism spectrum disorders result in an overreliance on self-r­egulation, Note leading to chronic depletion and the use of self-r­egulatory strategies associated with 1. Accumulating evidence from a diverse col- small children. We suggest further that SBT lection of laboratories suggests the psycho- may be useful for understanding emerging logical functions supported by the cortex—­ approaches to relationship therapy that are including the PFC—are computationally, and proving to be efficacious for a number of hence bioenergetically, expensive, especially problems traditionally treated by individu- relative to more computationally basic per- ally oriented psychotherapies alone (Mon- ceptual processes (Dietrich, 2009; Halford, son et al., 2012; Naaman, Radwan, & John- Wilson, & Phillips, 1997, 1998). At the very son, 2009). least, psychological processes supported by the PFC are often subjectively experienced Ultimately, we emphasize that human life as effortful (Muraven & Baumeister, 2000; is in many ways defined by social relation- Sheppes, Catran, & Meiran, 2009). Sus- ships. According to SBT, a key function of tained prefrontally mediated behaviors (e.g., social relationships is the social regulation attention, self-­control, vigilance, behavioral of emotion. Emotions provide rapid, embod- inhibition) cause subjective exhaustion and a ied information about current states and steady decrease in the ability to perform those contextual demands (Clore & Tamir, 2002), very behaviors (Baumeister, Vohs, & Tice, guiding decision making and modifying per- 2007; Vohs & Heatherton, 2000). Because ception (Bechara, 2011; Stefanucci et al., the brain maintains a fairly constant meta- 2008). The modification of perception via bolic rate regardless of task (Sokoloff, Man- emotional responding is itself yoked to the gold, Wechsler, Kennedy, & Kety, 1955), an management of bioenergetic resources for the extended reliance on prefrontally mediated purpose of economizing action (Stefanucci processes may reduce resources (e.g., blood) et al., 2011). Because emotion is powerfully available to other regions. Indeed, there is regulated by social proximity (Coan et al., evidence of intrabrain competition for blood, 2013; Coan et al., 2006), all of the above is even within the cortex (Dietrich, 2009). controlled by proximity to social resources Although open to serious criticism (Kurzban, (Schnall et al., 2008). Thus, the impact of 2010), it may yet be possible that the PFC social relationships ripples through virtu- even places a higher metabolic demand on ally everything humans do, because social the brain’s resources. For example, the ratio relationships are tightly linked to emotional of glia to neurons—a­ ratio that covaries with processes, and emotional processes inform, metabolic demand—­is higher in the PFC of bias, and direct many, if not most, of the humans than in other primates (Sherwood brain’s activities. This may not always work et al., 2006). The PFC is among the last neu- in ways that are broadly socially desirable. ral systems to develop fully (Fuster, 2002) For example, the social regulatory pro- and among the first to show functional defi- cesses described throughout this chapter cits in the case of extreme emotional stress, may contribute to phenomena such as social malnourishment, intoxication, addiction, loafing (Karau & Williams, 1993), and the old age, and even exercise (Arnsten, 2009; “outsourcing” of health behaviors (Fitzsi- Del Giorno, Hall, O’Leary, Bixby, & Miller, mons & Finkel, 2011). The impact of social 2010; Goldstein & Volkow, 2011; Lipina & relationships—m­ anifest as perceived prox- Colombo, 2009; Paxton, Barch, Racine, & imity, interaction, and history—i­s thus likely Braver, 2008). All of this is despite a general to be felt within any specific subdomain of consensus that the human PFC is one of the psychological science, suggesting that any defining characteristics of human phylogeny such subdomain that does not account for (Ehrlich & Ehrlich, 2008)—a resource shaped social processes will be limited in impor- over evolutionary development that allows tant ways. On the other hand, the extent to humans a level of abstraction, creativity, vigi- which social processes impact psychological lance, self-­control, and anticipatory judgment phenomena opens wide the possibilities for that is unprecedented in the animal kingdom new testable hypotheses and exciting pro- and largely responsible for the globally domi- grams of research. nant status that humans enjoy.